University of Groningen Evaluating the social outcomes of ... · Evaluating the social outcomes of inclusive education a study on the social participation of pupils with special needs

University of Groningen

Evaluating the social outcomes of inclusive educationKoster, Marloes

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2008

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Koster, M. (2008). Evaluating the social outcomes of inclusive education: a study on the social participationof pupils with special needs in regular primary education and the development of a teacher questionnaire.s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 20-04-2020

https://www.rug.nl/research/portal/en/publications/evaluating-the-social-outcomes-of-inclusive-education(42c22bdf-2d59-477c-af58-22380ece7eb6).html



Evaluating the social outcomes of inclusive education

a study on the social participation of pupils with special needs in regular primary education

and the development of a teacher questionnaire

Marloes Koster

Colofon


a study on the social participation of pupils with special needs in regular primary

education and the development of a teacher questionnaire

Marloes Koster Proefschrift Rijksuniversiteit Groningen

ISBN 978-90-367-3581-0

Ontwerp en illustratie omslag: Hester Nijhoff

Drukwerk: Gildeprint Drukkerijen, Enschede

RIJKSUNIVERSITEIT GRONINGEN


a study on the social participation of pupils with special needs in regular primary education and the development of a teacher questionnaire

Proefschrift

ter verkrijging van het doctoraat in de

Gedrags- en Maatschappijwetenschappen

aan de Rijksuniversiteit Groningen

op gezag van de

Rector Magnificus, dr. F. Zwarts,

in het openbaar te verdedigen op

donderdag 20 november 2008

om 13.15 uur

door

Marloes Koster

geboren op 6 april 1980

te Rotterdam

Promotores: Prof. dr. H. Nakken

Prof. dr. S.J. Pijl

Copromotor: Dr. E.J. van Houten-van den Bosch

Beoordelingscommissie: Prof. dr. P.P.M. Leseman

Prof. dr. A.J.J.M. Ruijssenaars

Prof. dr. M.P.C. van der Werf

CONTENTS

5

Contents

CONTENTS ................................................................................................ 5

CHAPTER 1 GENERAL INTRODUCTION .......................................................... 9

1.1 Introduction .................................................................................... 10

1.2 Historical background of educating pupils with special needs: from segregation to inclusive education ........................................................... 12

1.3 Changes in educational policy in the Netherlands ................................. 13

1.4 The social dimension of inclusive education......................................... 15

1.5 Outline of the thesis......................................................................... 17

1.6 References...................................................................................... 18

CHAPTER 2 THE SOCIAL POSITION AND DEVELOPMENT OF PUPILS WITH A PUPIL-BOUND BUDGET IN REGULAR DUTCH PRIMARY SCHOOLS.................... 23

Abstract ............................................................................................... 24

2.1 Introduction .................................................................................... 25

2.2 Method........................................................................................... 28 2.2.1 Introduction.......................................................................................................28 2.2.2 Subjects ............................................................................................................28 2.2.3 Interviews and questionnaires..............................................................................29 2.2.4 Data analyses ....................................................................................................31

2.3 Results ........................................................................................... 32 2.3.1 Subjects ............................................................................................................32 2.3.2 Social position....................................................................................................33 2.3.3 Cognitive, social and social-emotional development ................................................38

2.4 Independent assessors’ judgments .................................................... 39 2.4.1 Reliability of assessments ....................................................................................39 2.4.2 Assessment of cognitive, social and social-emotional development ...........................41 2.4.3 Relation between rejected status and assessment of development............................41

2.5 Discussion ...................................................................................... 42

2.6 References...................................................................................... 45

CHAPTER 3 BEING PART OF THE PEER GROUP: A LITERATURE STUDY FOCUSING ON THE SOCIAL DIMENSION OF INCLUSION IN EDUCATION ......................... 49

Abstract ............................................................................................... 50

3.1 Introduction .................................................................................... 51

3.2 Research method............................................................................. 52

3.3 Selecting the literature ..................................................................... 53

3.4 Social integration............................................................................. 54 3.4.1 Explicit definitions of social integration ..................................................................54 3.4.2 Implicit definitions of social integration..................................................................55 3.4.3 Summary of the definitions of social integration .....................................................57

3.5 Social inclusion................................................................................ 61 3.5.1 Summary of the definitions of social inclusion ........................................................62

CONTENTS

6

3.6 Social participation........................................................................... 64 3.6.1 Summary of the definitions of social participation ...................................................65

3.7 Other related concepts ..................................................................... 67 3.7.1 Summary of the definitions of other related concepts ..............................................68

3.8 Conclusion ...................................................................................... 71

3.9 Discussion ...................................................................................... 73

3.10 References .................................................................................... 75

CHAPTER 4 ASSESSING SOCIAL PARTICIPATION OF PUPILS WITH SPECIAL NEEDS IN INCLUSIVE EDUCATION: THE CONSTRUCTION OF A TEACHER QUESTIONNAIRE...................................................................................... 83

Abstract ............................................................................................... 84

4.1 Introduction .................................................................................... 85

4.2 Construction of the social participation questionnaire ........................... 89 4.2.1 Method..............................................................................................................89 4.2.2 Selection of statements through consultation of a panel ..........................................89 4.2.3 Final selection of statements through assessment by a group of respondents.............90

4.3 Assessing the quality of the questionnaire........................................... 94 4.3.1 Method..............................................................................................................94 4.3.2 Results..............................................................................................................97

4.4 Discussion ...................................................................................... 99

4.5 References.....................................................................................103

CHAPTER 5 EXAMINATION OF THE PSYCHOMETRIC QUALITIES OF THE SOCIAL PARTICIPATION QUESTIONNAIRE..............................................................107

Abstract ..............................................................................................108

5.1 Introduction ...................................................................................109

5.2 Method..........................................................................................110 5.2.1 Participants......................................................................................................110 5.2.2 Instrument ......................................................................................................114 5.2.3 Analysis...........................................................................................................116

5.3 Results ..........................................................................................118 5.3.1 Mokken Scale Analysis ......................................................................................118 5.3.2 Separability of the subscales ..............................................................................121 5.3.3 Revised version of the Social Participation Questionnaire .......................................122 5.3.4 Discriminant validity .........................................................................................124

5.4 Discussion .....................................................................................126

5.5 References.....................................................................................128

CHAPTER 6 VALIDATION OF THE SOCIAL PARTICIPATION QUESTIONNAIRE ...133

Abstract ..............................................................................................134

6.1 Introduction ...................................................................................135

6.2 Method..........................................................................................137 6.2.1 Respondents ....................................................................................................138 6.2.2 Instruments.....................................................................................................140 6.2.3 Procedure ........................................................................................................144 6.2.4 Analysis...........................................................................................................147

6.3 Results ..........................................................................................149

CONTENTS

7

6.3.1 Convergent validity of the four subscales.............................................................149 6.3.2 Second order analysis .......................................................................................150

6.4 Discussion .....................................................................................153

6.5 References.....................................................................................156

CHAPTER 7 SOCIAL PARTICIPATION OF PUPILS WITH SPECIAL NEEDS IN REGULAR DUTCH PRIMARY EDUCATION.....................................................163

Abstract ..............................................................................................164

7.1 Introduction ...................................................................................165

7.2 Method..........................................................................................167 7.2.1 Introduction.....................................................................................................167 7.2.2 Respondents ....................................................................................................167 7.2.3 Instruments to assess key themes......................................................................169 7.2.4 Analysis...........................................................................................................171

7.3 Results ..........................................................................................172

7.4 Discussion .....................................................................................178

7.5 References.....................................................................................181

CHAPTER 8 GENERAL DISCUSSION ...........................................................187

8.1 Introduction ...................................................................................188

8.2 Main findings..................................................................................189 8.2.1 First experiences with the pupil-bound budget......................................................189 8.2.2 Construction of a model of social participation and developing a teacher questionnaire..............................................................................................................................190 8.2.3 Current situation regarding social participation of pupils with special needs..............191

8.3 Reflections on the study ..................................................................191

8.4 Implications of the study .................................................................195 8.4.1 Implications for educational policy ......................................................................195 8.4.2 Implications for teacher training .........................................................................196

8.5 Further research.............................................................................197

8.6 References.....................................................................................198

SUMMARY ..............................................................................................203

SAMENVATTING......................................................................................207

DANKWOORD .........................................................................................213

CONTENTS

8

Chapter 1 General introduction

CHAPTER 1

General introduction

CHAPTER 1

10

1.1 Introduction

For a long time it was assumed that pupils with special needs would not be able

to attend regular schools and that it was better to send them to special ones

(Pijl, 1997). Consequently, many countries had a system of special education

consisting of different types of special schools aiming at pupils with various types

of disabilities. Special education settings were viewed as possessing various

advantages, like lower teacher-pupil ratios, specially trained teachers, greater

individualisation of instruction in homogeneous classrooms and an increased

curricular emphasis on social and vocational goals (Johnson, 1962, in Kavale &

Forness, 2000).

However, the segregation of pupils with special needs came increasingly

into question as many of the presumed advantages could not be proved in

practice. For instance, Gartner and Lipsky (1987) concluded, on the basis of 50

studies concerning the academic performances of mainstreamed and segregated

pupils with disabilities, that there was no significant evidence that separate

special education programmes offered any significant benefits for these pupils.

On the contrary, academic achievement of segregated pupils with disabilities was

even lower compared to that of their counterparts in regular education settings.

Baker, Wang and Walberg (1994) came to similar conclusions after having

summarised the results of several studies on the effects of inclusion on pupils’

learning performances. They found that in the majority of cases, pupils with

special needs educated in regular classes performed better academically than

their counterparts in special settings (Baker et al., 1994). According to Rea,

McLaughin and Walther-Thomas (2002), findings regarding academic

achievement and social outcomes of pupils with learning disabilities are not

conclusive but suggest a positive trend when these pupils are integrated into

general education classrooms. In line with these findings, Kavale and Forness

(2000) state that empirical evidence about the efficacy of special education

continues to be equivocal.

It should be kept in mind that the above-mentioned results apply largely to

pupils with learning disabilities, as studies comparing performances in regular

and special education settings often aim at these pupils. Such comparative

studies are relatively rare for pupils with other categories of disabilities.

GENERAL INTRODUCTION

11

The lack of satisfactory academic performance by pupils in segregated

settings (in particular pupils with learning disabilities), combined with growing

demands for social equity and civil rights and increasing costs of special

education, prompted a drastic reconsideration of the special education delivery

system in the mid 1980s (Rea et al., 2002). Next to academic advantages, civil-

rights aspects of inclusion were emphasised. Segregating pupils with special

needs was increasingly considered as a violation of a pupil’s right to be educated

with typical peers in age-appropriate settings (Fisher, Roach & Frey, 2002; Rea

et al., 2002). Inclusive education was championed as a means to remove

barriers, improve outcomes and avoid discrimination (Lindsay, 2003).

As a result of this change in ideas about special-needs education, including

pupils with special needs into regular education became, and still is, an important

educational policy in many countries. Next to complying with children’s right to

be educated with their typical peers in public schools and improving academic

performance, increasing the social participation of pupils with special needs is a

major goal of inclusive education. Parents often report the latter as being their

first motive for sending their child to a regular school (Sloper & Tyler, 1992;

Strayhorn & Strain, 1986). However, successful social participation of pupils with

special needs turns out to be no matter of course. Research has repeatedly

shown that making friends and building positive relationships with classmates

can be difficult for these pupils. They often have a lower social position in the

classroom and report higher loneliness scores than their typical peers. Since

these negative social experiences can influence a child’s further development, it

is important to monitor the social participation of pupils with special needs.

Teachers can have an important role in the monitoring process.

This study tries to provide assistance for teachers to monitor the social

participation of pupils with special needs. Before proceeding to present the

design and aims of the study, this introductory chapter will briefly discuss the

historical background of educating pupils with special needs. The world-wide

trend from segregation towards inclusive education will be described. In addition,

attention will be paid to the Dutch educational policy with regard to pupils with

special needs and to the social dimension of inclusive education.

CHAPTER 1

12

1.2 Historical background of educating pupils with special needs: from

segregation to inclusive education

As described in the Introduction section, segregation of pupils with special needs

was increasingly regarded as undesirable, and policymakers grew more and more

convinced that these pupils should be educated alongside their typical peers in

regular schools to the greatest extent possible (Pijl & Meijer, 1994). In many

countries it was assumed that with extra effort pupils with all kinds of special

needs could attend regular schools (Nakken & Pijl, 2002). As a result, in almost

all countries with a system of separate special education, new ideas about the

educational care for these pupils arose, new regulations were introduced and

legislation was amended.

In the nineties, integration and mainstreaming were the terms typically

used to describe the provision allocated to pupils with special needs. Both terms

referred primarily to the physical placement of pupils with special needs in

regular schools (Farrell, 2004; Farrell, 2000; Gottlieb, 1981; Kauffman, 1995, in

Kavale & Forness, 2000). Several countries made an effort to implement policies

that fostered the integration of pupils with special needs (Avramidis & Norwich,

2002). Integration became one of the central themes in education. However,

there was a large variety of ways in which pupils could experience integration,

ranging from occasional visits of special-school pupils to a regular school to full-

time placement in a regular classroom of the local school (Farrell, 2000).

Gradually, the term integration became discredited, as it said nothing about the

quality of education but only about the setting in which a pupil was placed

(Farrell, 2004). Questions about how pupils should be best taught remained

unanswered (Gottlieb, 1981; Kauffman, 1995, in Kavale & Forness, 2000).

The term ‘inclusion’ was introduced as a more accurate way of describing

the quality of education offered to pupils with special needs within an integrated

school setting. Compared to integration, inclusion is a much broader concept that

implies that schools restructure themselves so as to be able to cater to all

children, irrespective of their disabilities or background (Frederickson & Cline,

2002). According to inclusive ideals, inclusive schools should meet the needs of

all pupils (Ferguson, 1996, in Kavale & Forness, 2000) and every pupil should

feel himself/herself as a full member of the school community (Ainscow, 1999, in

Freire & César, 2002). Farrell (2000, pp. 154) expresses these ideals clearly, by

stating that pupils in inclusive schools ‘take a full and active part in school-life,


13

are a valued member of the school community and are seen as an integral

member’. The movement towards inclusion has been strongly endorsed

internationally by the Salamanca Statement (UNESCO, 1994, in Ainscow & César,

2006). In recent decades many countries have made efforts to move educational

policy and practice in a more inclusive direction (Freire & César, 2002; Mittler,

2000, in Ainscow & César, 2006).

1.3 Changes in educational policy in the Netherlands

The Netherlands have a long history of separate special education for pupils with

various categories of disabilities (Tadema, 2007). In the early years of the 20th

century, a small-scale system of special schools for various groups of pupils

gradually arose alongside regular education (Den Boer, 1990). Special schools

were legally recognised in 1920, and in the following decades the number of

schools for pupils with special needs increased rapidly (Tadema, 2007). The

result was a wide-ranging system of special education, consisting of 15 different

types of special schools (Meijer, 1994). However, more and more policymakers,

educators and parents believed that segregation in education had gone too far

and became uneasy about the high proportion of pupils being educated in a

segregated system (Pijl & Van den Bos, 2001). Questions arose about the extent

to which pupils profited from separate education (Den Boer, 1990), and in

politics it was increasingly thought that pupils should have the right to be

educated with their typical peers in neighbourhood schools. Moreover, the

ballooning costs of special education were considered as alarming. As a result of

the increasing dissatisfaction about the growth of special education, attempts

were made to promote the integration of pupils with special needs into regular

schools. To put a stop to the growing number of pupils attending special schools,

the Special Education Interim Act (ISOVSO) was put into practice in 1985. The

purpose of the Act was to enable schools to develop themselves into

‘orthopedagogical-educational institutes’ (Den Boer, 1990). Within the context of

this Act, peripatetic supervision, which in fact is a visiting special-teacher model

(Pijl & Meijer, 1991), was made possible. The purpose of this type of supervision

was to stimulate replacing pupils from special schools into regular ones and to

prevent special-education referrals (Boerman & Hoogendoorn, 2002). In addition

to the arrangements for peripatetic supervision, schools were able to apply for

CHAPTER 1

14

extra teaching staff via the Aanvullend Formatie Beleid (Additional Staff Policy).

Pupils with problems who were beyond the scope of the usual regulations but

needed extra support, could make use of this policy (Scheepstra, 1998). For

instance, primary schools who educated pupils with Down Syndrome were

allowed to make an appeal for additional funding under this policy. In 1990 a

new government policy document, Weer Samen Naar School (WSNS, Together to

School Again), was intended to accommodate pupils with special needs in regular

education and to put a stop to the growth of special education (Karsten,

Roeleveld, Peetsma & Vergeer, 2001). Under this policy, all primary schools and

special schools for children who had learning and educational problems or who

had mild intellectual disabilities were grouped into regional clusters. As a result,

regular and special schools in the clusters started to collaborate (Pijl & Van den

Bos, 2001). The WSNS policy aims at pupils with relatively mild special needs.

With the introduction of the Wet op de Expertise Centra (Centres of Expertise

Act) in August 2003, the integration of pupils with more complex special needs

was stimulated.

Since this law came into practice, parents of children with special needs

have the right to choose between regular and special education for their child.

Pupils with auditory, speech/language, motor, intellectual or multiple disabilities

as well as severe behavioural, emotional and/or psychiatric problems can attend

a regular school. This is funded with a pupil-bound budget (financial ‘backpack’).

Only pupils who, on the basis of formal comprehensive assessment procedures,

have been labelled as having special needs, qualify for this budget. The budget

caters for educational personnel and teaching aids. Recent data show that an

increasing number of parents decided to send their child with special needs to a

regular school (De Greef & Van Rijswijk, 2006). Especially the number of pupils

diagnosed as having autistic spectrum and/or behavioural disorders attending

regular education (receiving a pupil-bound budget) grew rapidly, from 1549

pupils in October 2003 to 8055 in October 2005 – a growth of more than 400

percent (Grietens, Ghesquière & Pijl, 2006). However, not only the percentage of

pupils with special needs attending regular education has grown: the percentage

of pupils in special education settings has increased too. Similar to the situation

in regular settings, in special education settings particularly the number of pupils

diagnosed as having autistic spectrum and/or behavioural disorders has

increased drastically (De Greef & Van Rijswijk, 2006). In a period of five years


15

(2000 to 2005), the percentage of pupils diagnosed as having these types of

disabilities attending separate special education increased by 55 percent (De

Greef & Van Rijswijk, 2006). It seems that because of earlier and refined

diagnostics, more pupils receive an indication regarding autistic spectrum and/or

behavioural disorders. The same tendency might, to a lesser degree, apply to

pupils diagnosed as having other types of disabilities, as there seems to be an

overall increasing identification of pupils requiring extra services. In conclusion,

the introduction of the pupil-bound budget into the Dutch educational system

seems to have led to a new category of users of extra educational care: pupils

attending regular education, who in former years were deprived from extra

educational care, now qualify for this extra care thanks to the pupil-bound

budget (Grietens et al., 2006).

1.4 The social dimension of inclusive education

Relationships among pupils is a key issue of inclusive education (Pijl, 2007).

Many parents of children with special needs hope and expect the physical

presence of their children to lead to their social participation as well (Scheepstra,

1998; Sloper & Tyler, 1992; Strayhorn & Strain, 1986). They wish their child to

build positive relationships with mainstream pupils. However, reality turns out to

be less favourable. Research has repeatedly shown that inclusion of pupils with

special needs does not automatically lead to an increase of positive contacts and

friendships between these pupils and their typical counterparts (De Monchy, Pijl

& Zandberg, 2004; Guralnick, Connor, Hammond, Gottman & Kinnish, 1996;

Guralnick, Hammond, Connor & Neville, 2006; Guralnick, Neville, Hammond &

Connor, 2007; Lee, Yoo & Bak, 2003; Scheepstra, Nakken & Pijl, 1999). A study

by Frostad and Pijl (2007) of Norwegian inclusive classrooms, comprising pupils

with various categories of disabilities, suggests that nearly one-quarter of pupils

with special needs have serious difficulties in forming relationships in their peer

group, while for their typical peers this is only 8 percent. It is known from

several studies that within the group of pupils with special needs, pupils

diagnosed as having autistic spectrum disorders and pupils diagnosed as having

behavioural disorders find it particularly difficult to build relationships with typical

peers and are at risk of becoming isolated in the classroom (De Monchy et al.,

2004; Garrison-Harrell, Kamps & Kravits, 1997).

CHAPTER 1

16

This is worrisome, all the more so because having a low social position in

the class might negatively influence the functioning of the pupil in different

areas. For instance, Jackson and Bracken (1998) found that the self-concept

across various domains of rejected pupils was relatively low, as was the social

self-concept of neglected pupils. Ollendick, Weist, Borden and Greene (1992)

found that the academic achievements of pupils with a low social position,

especially the rejected ones, tend to be weaker. They also showed that these

pupils tend to have a higher failure rate in school as well as increased chances of

dropping out and of committing delinquent offences. A substantial number of

studies have shown that a low social position at school and peer relationship

difficulties might lead to maladjustment in later life (Bagwell, Newcomb &

Bukowski, 1998; Nelson, Rubin & Fox, 2005; Parker & Asher, 1987; Terry & Coie,

1991).

Because of the rather negative long-term effects of negative social

experiences at school, it is important to monitor the social participation of pupils

with special needs. This constitutes a vital task for teachers. There are several

instruments teachers can use to measure aspects of social participation, like

sociometric questionnaires and observation scales. However, the reliability

and/or validity of many of these instruments have not been proven. For instance,

Terry and Coie (1991) demonstrate that different methods to assess pupils’

sociometric status yield diverging results.

An instrument which encompasses the total concept of social participation

does not exist: most instruments only measure one aspect of social participation.

There is a need for a single teacher-friendly assessment instrument that

encompasses all important aspects of social participation and which is reliable

and valid. Such an instrument might help teachers notice problems in time, in

order to promote the social participation of pupils with special needs, for example

by encouraging contacts between these pupils and their classmates. In this

thesis, the development of an assessment instrument for teachers is central.

Taking into consideration the issues mentioned above, the aims of this

study are as follows:

1. To describe the first experiences with the pupil-bound budget in regular

Dutch primary education;

2. To elucidate the social dimension of inclusion in education;


17

3. To develop a teacher questionnaire to assess the social participation of

pupils with special needs and to subsequently assess the psychometric

qualities of the questionnaire;

4. To describe the current situation with regard to the social participation of

pupils with special needs in regular Dutch primary schools.

1.5 Outline of the thesis

This thesis is organised into eight chapters. Approaching the first aim, Chapter 2

describes the first experiences with the pupil-bound budget in regular Dutch

primary schools. The focus lies on the social position and development of 20

pupils with special needs who were placed in regular primary schools in the

2003-2004 school year. All pupils, on the basis of formal comprehensive

assessment procedures, have been labelled as having special needs and receive

a pupil-bound budget.

Chapter 3 concerns a literature review that aims at elucidating the social

dimension of inclusion, thereby meeting the second aim. The chapter focuses on

clarifying the often-used concepts of social integration, social participation and

social inclusion, and revealing their characteristic themes. An analysis of

literature was conducted to identify these concepts.

Chapters 4, 5 and 6 are connected with the third aim of the study. The

process of operationalising social participation and the subsequent development

of a teacher questionnaire is central in Chapter 4. The chapter is divided into two

parts. The first focuses on constructing a teacher questionnaire to assess social

participation. The second, empirical part of the study addresses whether that

questionnaire is a potentially reliable and applicable instrument to assess the

social participation of pupils with special needs in regular primary schools.

Chapter 5 addresses the psychometric qualities of the new teacher

questionnaire, named Social Participation Questionnaire, to assess the social

participation of pupils with special needs in regular primary education. The

questionnaire consists of four subscales representing four key themes of social

participation: ‘friendships/relationships’, ‘contacts/interactions’, ‘social self-

perception of the pupil’ and ‘acceptance by classmates’. This fifth chapter focuses

on the psychometric evaluation of the four subscales. In addition, the reliability

CHAPTER 1

18

and the discriminant validity of the entire questionnaire and its subscales are

described.

Chapter 6 aims at validating the Social Participation Questionnaire. More

specifically, the focus lies on the questionnaire’s convergent validity. In order to

examine the convergent validity, four instruments, each focusing on one of the

key themes of social participation, were used. Pupils’ scores on the total

questionnaire and on each of the four subscales were compared with their scores

on these instruments.

Chapter 7 returns to the situation of pupils with special needs in regular

Dutch primary schools, thereby elaborating on Chapter 2. Proceeding from the

fourth aim, Chapter 7 addresses the current state of affairs with regard to the

social participation of pupils with special needs in regular Dutch primary

education. It focuses on the four key themes of social participation:

friendships/relationships, contacts/interactions, social self-perception of the pupil

and acceptance by classmates.

In the final chapter, Chapter 8, conclusions are drawn and several critical

considerations that can be made on the study will be discussed. The chapter

ends with implications for educational policy and practice, and with suggestions

for future research.

1.6 References

Ainscow, M. & César, M. (2006). Inclusive education ten years after Salamanca:

setting the agenda. European Journal of Psychology of Education, 21(3),

231-238.

Avramidis, E. & Norwich, B. (2002). Teachers’ attitudes towards

integration/inclusion: a review of the literature. European Journal of

Special Needs Education, 17(2), 129-147.

Bagwell, C.L., Newcomb, A.F. & Bukowski, W.M. (1998). Preadolescent friendship

and peer rejection as predictors of adult adjustment. Child Development,

69(1), 140-153.

Baker, E.T., Wang, M.C. & Walberg, H.J. (1994). The effects of inclusion on

learning. Educational Leadership, 52(4), 33-36.

Boerman, R.A. & Hoogendoorn, G. (2002). Ambulante begeleiding. Tijdschrift

voor Orthopedagogiek, 41, 90-103.


19

De Greef, E.E.M. & Van Rijswijk, C.M. (2006). De groei van deelname aan cluster

4. Opvattingen over oorzaken en groeibeperkende maatregelen. Den

Haag: LCTI.

De Monchy, M., Pijl, S.J. & Zandberg, T.J. (2004). Discrepancies in judging social

inclusion and bullying of pupils with behaviour problems. European Journal

of Special Needs Education, 19(3), 317-330.

Den Boer, K. (1990). Special education in the Netherlands. European Journal of


Farrell, P. (2004). Making inclusion a reality for all. School Psychology

International, 25(1), 5-19.

Farrell, P. (2000). The impact of research on developments in inclusive

education. International Journal of Inclusive Education, 4(2), 153-162.

Fisher, D., Roach, V. & Frey, N. (2002). Examining the general programmatic

benefits of inclusive schools. International Journal of Inclusive Education,

6(1), 63-78.

Frederickson, N. & Cline, T. (2002). Special educational needs, inclusion and

diversity, a textbook. Buckingham/Philadelphia: Open University Press.

Freire, S. & César, M. (2002). Inclusive ideals/inclusive practices: How far is

dream from reality. Five comparative case studies. European Journal of


Frostad, P. & Pijl, S.J. (2007). Does being friendly help in making friends? The

relation between the social position and social skills of pupils with special

needs in mainstream education. European Journal of Special Needs

Education, 22(1), 15-30.

Garrison-Harrell, L., Kamps, D. & Kravits, T. (1997). The effects of peer networks

on social-communicative behaviors of students with autism. Focus on

Autism and other Developmental Disabilities, 12(4), 241-254.

Gartner, A. & Lipsky, D.K. (1987). Beyond special education: toward a quality

system for all students. Harvard Educational Review, 57(4), 367-395.

Gottlieb, J. (1981). Mainstreaming: fulfilling the promise? American Journal of

Mental Deficiency, 86(2), 115-126.

Grietens, H., Ghesquière, P. & Pijl, S.J. (2006). Toename leerlingen met

gedragsproblemen in primair en voortgezet onderwijs. Een Nederlands-

Vlaamse vergelijking. BOPO.

CHAPTER 1

20

Guralnick, M.J., Connor, R.T., Hammond, M.A., Gottman, J.M. & Kinnish, K.

(1996). The peer relations of preschool children with communication

disorders. Child Development, 67, 471-489.

Guralnick, M.J., Hammond, M.A., Connor, R.T. & Neville, B. (2006). Stability,

change, and correlates of the peer relationships of young children with

mild developmental delays. Child Development, 77(2), 312-324.

Guralnick, M.J., Neville, B., Hammond, M.A. & Connor, R.T. (2007). The

friendships of young children with developmental delays. A longitudinal

analysis. Journal of Applied Developmental Psychology, 28, 64-79.

Jackson, L.D. & Bracken, B.A. (1998). Relationship between students’ social

status and global and domain-specific self-concepts. Journal of School

Psychology, 36(2), 233-246.

Karsten, T., Peetsma, T., Roeleveld, J. & Vergeer, M. (2001). The Dutch policy of

integration put to the test: differences in academic and psychosocial

development of pupils in special and mainstream education. European

Journal of Special Needs Education, 16(3), 193-205.

Kavale, K.A. & Forness, S.R. (2000). History, rhetoric, and reality. Analysis of the

Inclusion Debate. Remedial and special education, 21(5), 279-296.

Lee, S.H., Yoo, S.Y. & Bak, S.H. (2003). Characteristics of friendships between

children with and without disabilities. Education and Training in

Developmental Disabilities, 38(2), 157-166.

Lindsay, G. (2003). Inclusive education: a critical perspective. British Journal of

Special Education, 30(1), 3-12.

Meijer, C.J.W. (1994). The Netherlands. In C.J.W. Meijer, S.J. Pijl & S. Hegarty

(Eds.), New perspectives in special education. A six-country study of

integration (pp. 95-112). London, New York: Routlegde.

Nakken, H. & Pijl S.J. (2002). Getting along with classmates in regular schools: a

review of the effects of integration on the development of social

relationships. International Journal of Inclusive Education, 6(1), 47-61.

Nelson, L.J., Rubin, K.H. & Fox, N.A. (2005). Social withdrawal, observed peer

acceptance, and the development of self-perceptions in children ages 4 to

7 years. Early Childhood Research Quarterly, 20, 185-200.

Ollendick, T.H., Weist, M.D., Borden, M.C. & Greene, R.W. (1992). Social status

and academic, behavioural, and psychological adjustment: a five-year


21

longitudinal study. Journal of Consulting and Clinical Psychology, 60(1),

80-87.

Parker, J.G. & Asher, S.R. (1987). Peer relations and later personal adjustment:

are low-accepted children at risk? Psychological Bulletin, 102(3), 357-389.

Pijl, S.J (2007). Introduction: the social position of pupils with special needs in

regular education. European Journal of Special Needs Education, 22(1), 1-

5.

Pijl, S.J. (Ed.) (1997). Integratie van regulier en speciaal onderwijs.

Onderwijskundig Lexicon. Alphen aan den Rijn: Samson.

Pijl, S.J. & Meijer, C.J.W. (1994). Introduction. In C.J.W. Meijer, S.J. Pijl. & S.

Hegarty (Eds.), New perspectives in special education. A six-country study

of integration (pp. xi-xiv). London, New York: Routlegde.

Pijl, S.J. & Meijer, C.J.W. (1991). Does integration count for much? An analysis

of the practices of integration in eight countries. European Journal of


Pijl, S.J. & Van den Bos, K.P. (2001). Redesigning regular education support in

the Netherlands. European Journal of Special Needs Education, 16(2),

111-119.

Rea, P.J., McLaughlin, V. & Walther-Thomas, C. (2002). Outcomes for students

with learning disabilities in inclusive and pullout programs. Council for

Exceptional Children, 68(2), 203-223.

Scheepstra, A.J.M. (1998). Leerlingen met Downs syndroom in de basisschool.

Groningen: Stichting Kinderstudies.

Scheepstra, A.J.M., Nakken, H. & Pijl, S.J. (1999). Contact with classmates: the

social position of pupils with Down’s Syndrome in Dutch mainstream

education. European Journal of Special Needs Education, 14(3), 212-220.

Sloper, T. & Tyler, S. (1992). Integration of children with severe learning

difficulties in mainstream schools: evaluation of a pilot study. Educational

and Child Psychology, 9(4), 35-45.

Strayhorn, J.M. & Strain, P.S. (1986). Social and language skills for preventive

mental health: What, how, who and when. In P.S. Strain, M.J. Guralnick &

H. Walker (Eds.), Children’s Social Behavior: Development, Assessment

and Modifications (pp. 287-330). New York: Academic Press.

CHAPTER 1

22

Tadema, A. (2007). From policy to practice. Developments in the education of

children with profound intellectual and multiple disabilities. Groningen:

Stichting Kinderstudies.

Terry, R. & Coie, J.D. (1991). A comparison of methods for defining sociometric

status among children. Developmental Psychology, 27(5), 867-880.

Chapter 2 The social position and development of pupils with a pupil-bound budget in regular Dutch primary schools1

CHAPTER 2 THE SOCIAL POSITION AND DEVELOPMENT OF PUPILS WITH A PUPIL-BOUND BUDGET IN REGULAR DUTCH PRIMARY SCHOOLS

This article is a slightly adapted version of: Koster, M., Pijl, S.J., Van Houten, E.J. & Nakken, H. (2007). The social position and development of pupils with SEN in mainstream Dutch primary schools. European Journal of Special Needs Education, 22(1), 31-46.

CHAPTER 2

24

Abstract

Since August 2003, pupils with auditory, speech/language, motor, intellectual or

multiple disabilities as well as severe behavioural, emotional and/or psychiatric

problems in the Netherlands have been entitled to receive a pupil-bound budget

when they attend a regular school. With this budget, educational personnel and

material aids can be funded. The first experiences with the pupil-bound budget in

regular Dutch primary schools are described in this article. The focus is on the

social position and development of 20 pupils with special needs who were placed

in regular primary schools in the 2003-2004 school year. The teacher, the

parent/parents and the peripatetic teacher of each of the pupils were

interviewed. The interviews aimed at the cognitive, social and social-emotional

development of the pupils. Besides, the teacher, the parent/parents, the

peripatetic teacher and the classmates of the pupils with special needs assessed

the social position of the pupil within the classroom via interviews and a

sociometric questionnaire. The results show that teachers and parents and, to a

lesser extent, the peripatetic teachers, had a more positive view of the social

position of the pupils with special needs than the classmates. The results of the

sociometric questionnaire show that the social position of the pupils with special

needs and that of their typical classmates do not differ significantly, however. In

addition, a panel of five independent assessors judged the development of the 20

pupils with special needs, by reading anonymous pupil dossiers. The

assessments show that the panel had some concerns about the development of

35 percent of these pupils. Nevertheless, the panel was (very) positive about

another 35 percent of the pupils with special needs. An expected relation

between the social position of the pupils with special needs and the panel’s

satisfaction about the development of these pupils was not found.

THE SOCIAL POSITION AND DEVELOPMENT OF PUPILS WITH A PUPIL-BOUND BUDGET

25

2.1 Introduction

Over the last decades, it is no longer taken for granted that pupils with special

needs are sent to schools for special education. For a long time it was assumed

that these pupils would not be able to attend regular schools (Pijl, 1997) and that

it was better to send them to special ones. The placement of these children at a

regular school was considered to be quite unrealistic. This point of view became

increasingly a matter for debate (Limpens, Nekkers & Ter Pelle, 2003). In many

countries it was increasingly assumed that with extra effort pupils with sensory,

motor and/or intellectual disabilities could attend regular schools (Nakken & Pijl,

2002). As a result, in almost all countries with a system of separate special

education, new ideas about the educational care for these pupils arose, new

regulations were introduced and legislation was amended. Integration of children

with special needs became, and still is, a current theme and an important aim.

In the Netherlands attempts are also being made to promote the integration of

children with special needs into regular schools. With the introduction of the

‘Centres of Expertise Act’ (Wet op de Expertise Centra, Ministerie van OCW) in

August 2003, the pupil-bound budget has made its entry into the Dutch

educational system. It is now law that parents of children with special needs

have the right to choose between regular and special education for their child.

Since the law became effective, pupils with auditory, speech/language, motor,

intellectual or multiple disabilities as well as severe behavioural, emotional

and/or psychiatric problems have been entitled to enrol in the so-called

‘Backpack’ policy. This means that when pupils with special needs attend a

regular school they receive a financial ‘backpack’. The money can be spent on

personnel and key teaching aids. It is to be expected that an increasing number

of parents will decide to send their child with special needs to a regular school.

Many parents of children with special needs hope and expect the physical

integration of their children, which is now made possible by the government, to

lead to their social integration as well. Parents often report the latter as being

their main motive for sending their child to a regular school (Scheepstra, 1998,

in De Monchy, Pijl & Zandberg, 2004; Sloper & Tyler, 1992; Strayhorn & Strain,

1986). In the opinion of parents, it is important for their child to go to the same

school as other children in the neighbourhood. They like their child to be

educated with other typically developing children and they often assume that

CHAPTER 2

26

contact with such children will have a positive effect on their child. Parents also

expect integration to lead to increased opportunities for contact with

neighbourhood children, thereby increasing the opportunity to learn how to

handle social situations, make friends and integrate into the local community (De

Monchy et al., 2004). Furthermore, some parents assume that as a result of

sending their child with special needs to a regular school, a change of attitude

will arise among other children, possibly leading to positive long-term effects on

attitudes towards special needs in wider society. Because of the emphasis

parents place on social integration, it is important to verify if this aspect of

integration is really achieved.

If social integration fails, some risks might develop. Research has shown that the

number of contacts and friendships with pupils without special needs does not

increase spontaneously when these pupils are sent to a regular school (Pijl,

2005). Physical integration is indeed a first important step, but it seems next

steps are also necessary, as it turns out that the acceptance of pupils with

special needs cannot be taken for granted. Research on social acceptance

consistently shows that pupils with special needs who are placed in a regular

school class are accepted to a lesser degree than their classmates without special

needs (Larrivee & Horne, 1991). In this context, a great number of researchers

have shown that pupils with special needs have a lower social status than the

other pupils (Larrivee & Horne, 1991). Terry and Coie (1991) state that the lack

of ‘status’ in childhood is a significant predictor of the possible development of

maladjustment in adulthood. Parker and Asher (1987) examined the

presupposition, widespread in social developmental literature, that peer-

relationship difficulties in childhood predict serious maladjustment in later life.

For this purpose they made a review of the literature on the subject. Their

results indicate that children with poor peer adjustment are at risk for later

difficulties in life, thereby confirming the existing presupposition (Parker & Asher,

1987).

The risks of having a low social position in class are discussed by various authors

in literature. The rejected position is especially connected with negative side

effects. According to Jackson and Bracken (1998), rejected children are more

likely to exhibit aggressive and ‘acting out’ behaviours than others. More often

than average they also lack the knowledge about how to join in group activities

effectively (Jackson & Bracken, 1998; Ollendick, Weist, Borden & Greene, 1992).


27

In addition, the academic achievements of rejected pupils tend to be weaker.

Furthermore, these pupils tend to have a higher failure rate in school, an

increased chance of dropping out and of committing delinquent offences

(Ollendick et al., 1992).

Because of the above-mentioned possible negative effects of integration, a

pessimistic view of integrating children with special needs within regular

education might emerge. Nevertheless, it might be expected that by taking

particular measures, the negative effects can decrease and more positive effects

can occur. Several researchers suggest that the attitude of typically developing

children can be influenced (Scheepstra, Nakken & Pijl, 1999). For instance,

research has shown that cooperative learning fosters cooperation between

children with special needs and their peers without special needs. Furthermore,

concern for peers is fostered by cooperative learning (Gartin, Murdick & Digby,

1992). Fox (1989, in Gartin et al., 1992) found that after pupils with special

needs and their chronological age peers without special needs participated in

cooperative learning situations, attitudinal changes in both groups of pupils were

noted. Rynders et al. (1993) are positive about the effects of cooperative

learning too. In their opinion, the use of cooperative learning strategies within

integrated recreational activities has proven to be a powerful combination in

facilitating the inclusion of children with and without special needs. Furthermore,

Hunt, Alwell, Farron-Davis and Goetz (1996, in Hall & McGregor, 2000) found

that peer relationships between pupils with special needs and pupils without

special needs were formed when researchers implemented intervention packages

which focused on adult strategies and formal peer friendship programmes.

This study focuses on the first experiences with the pupil-bound budget in

regular Dutch primary schools. Based on the results in international literature

(Larrivee & Horne, 1991; Ray, 1985; Sale & Carey, 1995), it is expected that

pupils with special needs in the Netherlands will have a less positive position in

regular school classes than their classmates without special needs. In addition,

rejected pupils with special needs when compared to the other pupils with special

needs are expected to develop in a less desirable way.

Therefore, the research question in this study is twofold: ‘What social position do

pupils with special needs have in regular Dutch primary school classes?’ and

‘How do pupils with special needs, particularly rejected ones, develop in a

CHAPTER 2

28

cognitive, social and social-emotional sense in regular Dutch primary school

classes?’

2.2 Method

2.2.1 Introduction

In this study, the focus is on the social position of 20 pupils with special needs in

regular schools. The social position was characterised by the pupils’ teacher,

their parent/parents, the peripatetic teacher and their classmates. This study

also addresses the cognitive, social and social-emotional development of the

pupils with special needs as described by their teacher, their parent/parents and

the peripatetic teacher. Furthermore, a panel consisting of five independent

assessors also assessed these three developmental areas.

2.2.2 Subjects

A sample of 20 pupils with special needs participating in full-time regular

education participated in this study. These pupils were chosen from a wider

sample of 115 pupils with a pupil-bound budget who were placed in regular

primary schools in the 2003-2004 school year. This wider sample was drawn

from the files of a national institute (CFI) directly linked to the Ministry of

Education. Among other things, this institute is responsible for making special

needs funding available to schools and it keeps records of regular schools with

recently admitted pupils with special needs. The wider sample can be regarded

as representative of the pupils with a pupil-bound budget in regular Dutch

primary schools with each having the Dutch version of a statement of special

needs. A sample of 20 pupils was taken in the study described in this article, four

of which had various communication disabilities (hard of hearing,

speech/language problems), seven had Down Syndrome, six were described as

having motor disabilities and/or intellectual disabilities while the remaining three

had severe behavioural disorders and/or PDD-NOS.

The sample consisted of the pupils with special needs aged from 4 to 11, plus

their peers (without special needs) in 19 classes in 18 schools. In one school two

pupils with special needs were in the same class. The total number of pupils

(including the 20 pupils with special needs and their classmates) in the 19


29

classes was 406. In the case of the 20 pupils with special needs, school leaders,

teachers and parents consented to the study.

2.2.3 Interviews and questionnaires

Three categories of informers were involved. The first category consisted of

classmates of the pupil with special needs. It was important to receive

information from fellow pupils as they associate with the pupil with special needs

on a daily basis. Secondly, the teachers and the parents are important informers

because they closely follow the development of the pupils with special needs.

Their opinions are valuable in evaluating the integration process, although it is

difficult for them to interpret their assessment in terms of which development

can be expected for a particular child, since most of them lack data on the

development of comparable children in other settings. The only professionals

with much experience in supporting pupils with special needs and their teachers

in regular education settings are the peripatetic teachers working in regular

settings. Most peripatetic teachers have wide experience in special education as

well. They are thus the best informed assessors available. For this reason, the

peripatetic teachers form the third category of informers.

Data collection took place in different ways. A questionnaire for pupils and three

interview protocols were drawn up. In addition, a questionnaire was drawn up for

members of a panel of independent assessors.

The questionnaire for all pupils (including the pupils with special needs and their

classmates) in the sample comprised a sociometric questionnaire. Based on Coie,

Dodge and Coppotelli (1982), in the sociometric questionnaire pupils were asked

which three pupils they liked to play with best and which three pupils they

disliked playing with.

The teacher interview addressed a number of general questions about all pupils

in the classroom; on the behaviour of the pupil with special needs in particular;

on the pupil’s social position and number of friends and on his or her cognitive,

social and social-emotional development. In the interviews teachers had to

characterise their pupil with special needs. They could choose between the

following replies (with the later coding categories in brackets): many classmates

like the pupil (popular); many think neutrally about him/her (average); many

CHAPTER 2

30

classmates think differently about the pupil (controversial); many ignore him/her

(ignored) and many dislike him/her (rejected) (Coie et al., 1982).

The parent interview also focused on the pupil’s cognitive, social and social-

emotional development and on his/her social position. Furthermore the interview

addressed the motives underlying the parents’ choice for regular education, and

the extent to which the education offered came up to their expectations.

The interview with the involved peripatetic teachers was aimed at the

advantages and disadvantages of placement at a regular school compared to

placement at a school for special education. This interview too was aimed at the

pupil’s cognitive, social and social-emotional development and his/her social

position.

Finally, it was considered necessary to compare the development and social

position of the pupils with special needs with that of a reference group.

Unfortunately, for these pupils with special needs a reference group was not

available. For instance, it is impossible to compare the integrated pupils with

special needs with pupils with the same kind of special needs in special education

because the settings are very different. It is also impossible to compare the

pupils with special needs with a pupil-bound budget with pupils with special

needs who were already educated at a regular school before the new integration

policy was put into effect. Because of the pupil-bound budget, pupils with severe

disabilities can attend regular education, whereas formerly attending special

education was their only option. The pupils with special needs who already

attended regular education are expected to have less severe disabilities than

those with a pupil-bound budget. For this reason these groups of pupils are not

comparable.

Since there was no reference group, the comparative assessment of the

development and social position of the pupils with special needs had to take

place in an alternative manner. It was decided to draw up a panel of experts for

this purpose. Five experienced peripatetic teachers were asked to be part of an

independent panel to evaluate pupils’ dossiers. The panel can be deemed

independent because each assessor judged the reports of eight pupils with

special needs they did not know. They assessed the pupils’ cognitive, social and

social-emotional development. The available data on pupils consisted of

information derived from interviews with teachers, parents and peripatetic


31

teachers as well as from the results of the sociometric questionnaire. Based on

these data pupil dossiers were made, typically comprising four to five pages of

text and a copy of the Individual Education Programme.

2.2.4 Data analyses

The sociometric data on the 20 pupils with special needs and 386 pupils without

special needs were analysed using UCINET software (Borgatti, Everett &

Freeman, 1999). The mean and standard deviation per class were calculated for

the number of times pupils were chosen as preferred to play with. For every

pupil the standard score of preferred to play with was the pupil’s score, minus

the mean of the class, divided by the standard deviation (following Cillessen &

Ten Brink, 1991; Coie et al., 1982). Accordingly, the standard score of being

disliked to play with was calculated.

The social preference and social influence scores are based on these two

standard scores. The social preference score is the standard score being

preferred to play with minus the standard score being disliked to play with. The

social influence score is the standard score being preferred to play with plus the

standard score being disliked to play with. Based on these scores the pupils were

divided into five groups: popular, average, controversial, ignored and rejected.

In the interview teachers had to characterise their pupil with special needs using

the words ‘popular’, ‘average’, ‘controversial’, ‘ignored’ and ‘rejected’. Cohen’s

kappa is calculated for agreement between the teachers’ responses and the

categories based on the sociometric data. Similarly, the parents and the

peripatetic teachers were asked to characterise the pupils with special needs.

The answers were analysed in the same way as for the teachers described

above.

In the interview teachers also had to indicate how many friends their pupil with

special needs had in class. In sociometry, friendship is defined as a reciprocal

choice, implying that two pupils nominate each other as their best friend (De

Monchy et al., 2004). The agreement between the teachers’ responses and the

actual number of friends (as appears from the sociometric questionnaire) was

calculated using the Gower measure of agreement (Gower, 1971). This measure

takes into account that answers that differ a little show more agreement

compared to answers that differ more (De Monchy et al., 2004).

CHAPTER 2

32

We transformed all the available raw data on each pupil into 20 dossiers which

were then sent to the independent panel of assessors. All of these were

peripatetic teachers with experience in supporting pupils with special needs and

their teachers in regular education. Five assessors were involved, two of which

had specialist knowledge and expertise on communication disabilities, two had

this knowledge and expertise in relation to Down Syndrome, while one

specialised in behavioural disorders. Each of the dossiers was assessed

independently by two assessors so that each assessor received eight pupil

dossiers.

The assessors were asked to read the dossiers of the pupils carefully and to

judge the pupil’s cognitive, social and social-emotional development on a 5–point

scale, ranging from 1), very negative and dissatisfying situation, to 5), very

positive and satisfying situation.

Gower’s coefficient (Gower, 1971) was used as an index for agreement between

the assessments. This index uses the absolute sum of differences between the

assessments and compared to Cohen’s kappa (Popping, 1983) is less influenced

by asymmetrical distributions. Gower’s coefficient is 0 if there is no agreement at

all and 1 if there is perfect agreement between the assessments. The assessors

were encouraged to give comments and to explain their assessments.

2.3 Results

2.3.1 Subjects

The definitive sample consisted of 20 pupils - ten boys and ten girls, of whom

two boys were in the same class. Their average age was 5.7 years. Thirteen

pupils were in Kindergarten, three were in Grade 1, two were in Grade 2 and the

remaining two were in Grade 4. The pupils had been in a regular class for seven

months on average and all pupils knew their peers well. The average class in the

study had 21.4 pupils. The total number of pupils in the 19 classes was 406.


33

2.3.2 Social position

Social position according to teachers, parents, peripatetic teachers and

classmates

In order to investigate whether the social position of the 20 pupils with special

needs deviated from that of their 386 peers in the 19 classes, their social

position was assessed using the social preference and social influence scores.

These were based on two questions for each of the pupils, i.e.: which three

pupils in the classroom do you like playing with best and which three pupils in

the classroom do you dislike playing with? Table 1 presents the findings of this

analysis.

Table 1. Sociometric categories in terms of being liked/disliked (n=406)

Category Pupils without special

needs

Pupils with special needs

Popular 81 (21.0%) 2 (10%)

Average 111 (28.8%) 8 (40%)

Controversial 41 (10.6%) 1 (5%)

Ignored 82 (21.2%) 5 (25%)

Rejected 71 (18.4%) 4 (20%)

Total 386 (100%) 20 (100%)

The data show that 10% of the pupils with special needs in our study were

popular, while in the peer group without special needs, 21% of the pupils were

regarded as popular. Of the pupils with special needs 40% belonged to the

‘average’ group. In the reference population almost 29% were regarded as

‘average’. There is also a difference in the percentage of pupils with special

needs and pupils without special needs regarded as ‘controversial’: one pupil

with special needs (5%) and 41 pupils without special needs (almost 11%). The

differences between the two groups were minimal with respect to the

percentages categorised as: ‘ignored’ or ‘rejected’. Generally, the differences

between the social position of the special needs group are limited compared to

that of their peers without special needs. By means of the Mann-Whitney U test,

the significance of the differences between the pupils with special needs and

their typical peers was calculated. The outcomes show that the differences

between these two groups are not significant (popular: U=3436000, Z=-1,19,

p>0.05; average: U=3426000, Z=-1,08, p>0.05; controversial: U=3643000,

CHAPTER 2

34

Z=-0,80, p>0.05; ignored: U=3715000, Z=-0,40, p>0.05; rejected:

U=3798000, Z=-0,18, p>0.05).

The social position of each of the pupils with special needs as rated by other

pupils in the classes was compared to teachers’ assessments. In Table 2 both

classmates’ and teachers’ assessments are presented (horizontally for

classmates and vertically for teachers).

Table 2. Social position of pupils with special needs according to teachers (vertical) and

classmates (horizontal) (n=20)

Teachers�

Classmates�

Popular

Average

Controversial

Ignored

Rejected

Total

Popular 2 0 0 0 0 2

Average 2 6 0 0 0 8

Controversial 1 0 0 0 0 1

Ignored 1 4 0 0 0 5

Rejected 0 4 0 0 0 4

Total 6 14 0 0 0 20

The data in Table 2 are again based on the social preference and social influence

scores. In analysing the data in the Table the data on the diagonal (from upper

left to lower right) are relevant as they represent the corresponding

assessments. In 8 of the 20 cases both classmates and teachers agreed on pupils

with special needs being assessed as ‘popular’ or ‘average’. The data under the

diagonal show that, compared to the classmates’ assessments, 12 teachers are

more positive regarding the position of pupils with special needs in their classes.

These teachers regard pupils with special needs not as ‘average’, ‘controversial’

or ‘ignored’, respectively, but as ‘popular’ (2, 1, 1), and not as ‘ignored’ or

‘rejected’ but as ‘average’ (4, 4). There are no scores above the diagonal,


35

meaning that no teacher was more negative about the social position of the

pupils with special needs compared to the classmates’ assessments.

Table 3. Social position of pupils with special needs according to parents (vertical) and

classmates (horizontal) (n=20)

Parents�

Classmates�

Popular

Average

Controversial

Ignored

Rejected

Total

Popular 0 1 1 0 0 2

Average 3 5 0 0 0 8


Ignored 2 3 0 0 0 5


Total 5 13 2 0 0 20

The data in Table 3 show parents assessments regarding the social position of

the pupils with special needs when compared to those of the classmates. In five

cases parents and classmates agree on pupils with special needs being assessed

as ‘average’. In the other 15 cases the parents and classmates disagree. The

data under the diagonal show that 13 parents are more positive about the social

position than the classmates. These parents regard pupils with special needs as

‘popular’ instead of ‘average’ or ‘ignored’ (3, 2), as ‘average’ instead of

‘controversial’, ‘ignored’ or ‘rejected’ (1, 3, 3) and as ‘controversial’ instead of

‘rejected’ (1). The data above the diagonal show that in the remaining two cases,

the parents are less positive than the classmates. The parents regard these two

pupils with special needs not as ‘popular’, but as ‘average’ and ‘controversial’ (1,

1).

CHAPTER 2

36

Table 4. Social position of pupils with special needs according to peripatetic teachers

(vertical) and classmates (horizontal) (n=18)

Peripatetic teachers�

Classmates�

Popular

Average

Controversial

Ignored

Rejected

Total

Popular 0 2 0 0 0 2

Average 1 4 2 0 0 7


Ignored 1 3 1 0 0 5


Total 2 12 4 0 0 18

Finally, the data in Table 4 show the differences and similarities in social position

assessments of the peripatetic teachers and the classmates. In five cases

peripatetic teachers agree with the classmates. Both the peripatetic teachers and

classmates assessed pupils with special needs as ‘average’ four times and in one

case they both assessed a pupil with special needs as ‘controversial’. The data

under the diagonal show that nine of the peripatetic teachers involved considered

the social position of the pupil with special needs as more positive than the

classmates. These peripatetic teachers regard pupils with special needs not as

‘average’ or ‘ignored’, respectively, but as ‘popular’ (1, 1), not as ‘ignored’ or

‘rejected’ but as ‘average’ (3, 3), not as ‘ignored’, but as ‘controversial’ (1).

The data above the diagonal show that four peripatetic teachers gave a less

positive characterisation of the social position of the pupils with special needs

than the classmates. These peripatetic teachers describe the position of the

pupils with special needs as ‘average’ instead of ‘popular’ (2), or ‘controversial’

instead of ‘average’ (2).

Summarising, it may be said that the social position of special needs pupils does

not differ a great deal from that of other classroom peers. The pupils with special


37

needs are a little less popular, less often belong to the controversial group and

are more often regarded as average. These differences are not significant.

Further, there is a discrepancy between classmates’ and teachers’ assessments,

between classmates’ and parents’ assessments and between classmates’ and

peripatetic teachers’ assessments. If the pupils’ assessments are used as

reference point, 40% of the teacher assessments are accurate, but 60% of the

teachers overestimate the actual social position of pupils with special needs.

Almost the same holds true for parents’ and peripatetic teachers’ assessments. If

the pupils’ assessments are used as a reference point, 25% of the parents and

28% of the peripatetic teachers give an accurate assessment, but 65% of the

parents and 50% of the peripatetic teachers have a too positive view on the

social position of the pupil with special needs. On the other hand, 10% of the

parents and 22% of the peripatetic teachers underestimate the social position of

the pupil with special needs, whereas none of the teachers underestimate.

Social position in terms of the number of friends

Another indication of the social position of the pupils with special needs in this

study is the number of friends they have in class. The first sociometric analysis,

the one on peers liking or disliking each other, was used to calculate the number

of friends as reported by the pupils. In sociometric analyses friends are defined

as two pupils selecting each other as peers they like or peers with whom they

like to play. There must be reciprocity in their choices (Cullinan, Sabornie &

Crossland, 1992; De Monchy et al., 2004; Juvonen & Bear, 1992).

CHAPTER 2

38

Table 5. Number of friends according to teachers (vertical) and classmates (horizontal)

(n=20)

Teachers�

Classmates�

0

1

2

3

Total

0 4 3 1 1 9

1 0 2 2 2 6

2 2 1 2 0 5

3 0 0 0 0 0

Total 6 6 5 3 20

Table 5 shows the number of friends according to teachers’ and pupils’ reports

(based on the sociometric questionnaire) for pupils with special needs in this

study. Again the upper left-lower right diagonal represents the corresponding

assessments: 40% of the assessments correspond, 45% of teachers

overestimate while 15% underestimate the number of friends the pupils with

special needs have in class. The data on the diagonal and also that close to it

show that teachers have a fairly accurate idea of the friendships of the pupils

involved. As an index of agreement for ratio-level data, Gower’s coefficient

(Gower, 1971) was calculated, which was 0.76. This indicates a reasonably high

agreement between the sociometric data and the teachers’ assessments.

2.3.3 Cognitive, social and social-emotional development

The learning outcomes of the 20 pupils varied from disappointing to beyond

expectations. Related to this, the satisfaction of teachers, parents and peripatetic

teachers with pupils’ cognitive development varied. Especially the cognitive

development of two pupils with Down Syndrome and of three pupils with motor

disabilities/intellectual disabilities were below the expectations of teachers and

peripatetic teachers. The parents were slightly more positive about the pupils’

cognitive development. All teachers, parents and peripatetic teachers were

mainly satisfied or very satisfied about the pupils’ social and social-emotional

development.


39

2.4 Independent assessors’ judgments

2.4.1 Reliability of assessments

In order to make sure that the independent assessors would give reliable

assessments on the cognitive, social and social-emotional development, each

dossier was assessed by two different assessors. A high level of agreement

between the assessments shows that different assessors came to similar

conclusions. Tables 6a-c present the data based on 20 pupil dossiers. The

findings in the Tables address the cognitive, social and social-emotional

development. Both the first column and the first row of each Table present the

values of the 5–point scale the assessors used. The remaining 25 figures show

the frequency of the various combinations of assessments. The frequency ‘2’ in

the matrix on social-emotional development (Table 6c) in row 2, column 4,

indicates that in two cases an assessor scores a ‘2’ (rather negative) while the

other assessor was much more positive (‘4’). The frequencies on the diagonal

show the concurring assessments. Gower’s coefficients on the data in the Tables

6a-c are 0.84, 0.75 and 0.75 respectively. This suggests that the assessors to a

reasonable degree agree on the cognitive, social and social-emotional

development. The conclusion is that different assessors come to comparable

assessments if they apply the 5-point scale on the pupil descriptions given. The

assessments can, therefore, be used in evaluating the effects on pupil level.

CHAPTER 2

40

Table 6a. Cognitive development (n=20)

1st assessor�

2nd assessor�

1

2

3

4

5

1 0 1 0 0 0

2 2 2 2 0 1

3 0 0 4 2 0

4 0 0 1 4 0

5 0 0 1 0 0

Table 6c. Social-emotional development

(n=20)

1st assessor�

2nd assessor�

1

2

3

4

5

1 0 0 0 1 0

2 1 4 3 2 0

3 0 1 0 2 0

4 0 0 3 2 0

5 0 1 0 0 0

Table 6b. Social development (n=20)

1st assessor�

2nd assessor�

1

2

3

4

5

1 0 0 0 0 0

2 0 3 3 3 0

3 0 2 0 2 0

4 0 3 1 3 0

5 0 0 0 0 0


41

2.4.2 Assessment of cognitive, social and social-emotional development

Tables 6a-c already give a first impression of the results of the evaluation. The

intersections of the first two columns and rows in each of the three matrices

show how many pupils were judged as being at risk. These represent

combinations of the low scores (1 and 2), i.e. ‘very negative and dissatisfying’

and ‘negative and dissatisfying’. Other combinations of scores, for example ‘1

and 3’, ‘1 and 4’ or ‘2 and 3’ are not taken as a clear indication that the

development of pupils was disappointing or very disappointing.

In the case of low scores, the assessors felt that the cognitive development of

five pupils was ‘worrying’, that the social-emotional development of another five

pupils was also ‘worrying’ as was the social development of three other pupils.

However, a further analysis of the data shows that there is an overlap between

these pupils. According to the assessors, a total of seven pupils were actually

involved: one pupil was developing negatively in all three areas, one in both the

social and social-emotional areas, two in both the cognitive and social-emotional

areas, one in the cognitive and social areas, one in the social-emotional area and

one in the cognitive area. These seven pupils represent 35% of the group. The

assessors regarded the development of the other thirteen pupils in this study as

being average or above average level. The intersections of the last two columns

and rows in each of the three matrices show how many pupils were assessed as

functioning above average. Corrected for overlap, the cognitive, social and

social-emotional development of, again, seven pupils (35%) were assessed as

above average.

2.4.3 Relation between rejected status and assessment of development

Because of the negative aspects linked to the rejected position, we investigated

the relation between the independent assessors’ judgments with regard to the

development of the pupils with special needs and their social position within the

classroom. We expected the cognitive, social and social-emotional development

of rejected pupils with special needs -as taken from the sociometric

questionnaire- to be assessed in a less positive way than the development of the

other pupils with special needs. As described in the method section, the panel of

independent assessors judged these three developmental areas of the pupils with

CHAPTER 2

42

special needs. The assessments consisted of scores on a 5-point scale, ranging

from 1, very negative and dissatisfying situation, to 5, very positive and

satisfying situation.

Table 7. Assessment of the development of rejected and other pupils with special needs

by the independent panel (n=20)

Developmental domain Mean score rejected

pupils (n=4)

Mean score other

pupils (n=16)

Cognitive development 3.50 2.85

Social development 2.75 3.00

Social-emotional development 2.50 2.95

As can be seen in Table 7, some small differences exist between the assessment

of rejected pupils with special needs and that of the other pupils with special

needs. The social development and social-emotional development of rejected

pupils with special needs are assessed less positively than those of the other

pupils with special needs. On the other hand, the cognitive development of the

rejected pupils with special needs is assessed more positively than that of the

other pupils with special needs.

The differences in assessment of the three developmental areas are not

significant for the rejected group and the other status groups when calculated

with the Mann-Whitney U test (Cognitive development, U=15.500, Z=-1.59,

p>0.05; Social development, U=25.500, Z=-0.63, p>0.05; Social-emotional

development, U=20.500, Z=-1.12, p>0.05).

2.5 Discussion

In this research, the focus was on the social position and development of 20

pupils with special needs with a pupil-bound budget who were placed in regular

primary schools in the 2003-2004 school year.

A comparison between the views of the teacher, the parent/parents, the

peripatetic teacher and the classmates about the social position of pupils with

special needs shows differences. The views of the teachers and parent/parents

about the social position were more positive than that of the classmates. When

compared to classmates’ views, the peripatetic teachers also had a more positive

view on the social position of pupils with special needs. Nevertheless, the


43

differences between the views of classmates and those of the peripatetic

teachers were smaller than those between the views of classmates and those of

teachers and parents.

Despite the discrepancies between teachers and classmates regarding their views

on the social position of the pupils with special needs, there is a reasonable

agreement between the teachers’ and classmates’ opinion about the number of

friends pupils with special needs have. The number of friends, as named by the

teacher and the number of friends as derived from the sociometric questionnaire

shows a certain similarity.

Most of the teachers, the parents and, to a lesser extent, the peripatetic teachers

are satisfied about the development of the pupils with special needs, particularly

their social and social-emotional development, although satisfaction with

cognitive development varied. For about one third of the pupils, the positive

vision of the teacher, parent/parents and peripatetic teacher contrasts with the

more concerned vision of the panel of independent assessors. For these pupils

the panel has ‘worries’ about the cognitive, social and/or social-emotional

development.

Our expectation that there would be a relationship between the rejected status

and the developmental assessment by the independent panel was unfounded. As

described in the introduction section, we expected the assessment of rejected

pupils with special needs to be less positive than the developmental assessment

of the other pupils with special needs. The social and social-emotional

development of the rejected pupils with special needs are indeed assessed less

positively than the development of the other pupils with special needs, but the

differences are not significant. In addition, the cognitive development of the

rejected pupils with special needs was even regarded in a slightly more positive

way than that of the other pupils with special needs.

We draw the conclusions mentioned above based on the research results.

However, a few critical considerations can be made on the research described in

this article. The first is the small number of pupils involved in the research.

Because only 20 pupils with disabilities were involved, it is difficult to draw

overall and general conclusions. Nonetheless, the sample can be regarded as

representative because it is extracted from a wider sample of 115 Dutch pupils

with special needs with a pupil-bound budget. For this reason the research

CHAPTER 2

44

results can be regarded as containing valuable information and as having

consequences which should be taken seriously.

The second consideration focuses on the age of the pupils with special needs in

this research. They were quite young (average age is 5;7). For the youngest

children especially, it is hard to decide which classmates they like and do not like

playing with. Coincidence could have influenced the sociometric research with

these young pupils. For example, it might be possible that a pupil chose a

classmate to play with because this classmate did something nice that day. At

another moment, the pupil might not have chosen this classmate as a playmate.

Therefore, interpretations in this area should be made cautiously.

A final consideration is the distribution of the types of disabilities of the pupils in

the research. Despite efforts to arrange a proportional distribution of types of

disabilities, quite a large group consisted of pupils with Down Syndrome.

Nevertheless, pupils with other kind of special needs are also represented in the

research sufficiently (see the method section). Only the pupils with behavioural

disorders and/or psychiatric disorders are slightly underrepresented.

Keeping in mind the critical considerations, the discrepancies in the views of the

people involved, regarding the development and social position of the pupils with

special needs, are worth noting. These discrepancies are in accordance with the

results of comparable studies. For instance, Scheepstra (1998, in De Monchy et

al., 2004) shows that teachers tend to overrate the social integration of pupils

with Down Syndrome in their class when compared to the views of the

classmates. Furthermore, the study of De Monchy et al. (2004) suggests that

teachers have a too positive view on the social position of the pupils with

behavioural disorders in their classes when compared to the results of a

sociometric test. In addition, our finding that pupils with special needs are less

popular than their typical counterparts needs is consistent with the results of

similar studies (Larrivee & Horne, 1991; Ray, 1985; Sale & Carey, 1995).

Our results emphasise the importance of research into the development and

social position of pupils with special needs in regular schools. It would be

desirable to follow pupils with special needs with a pupil-bound budget for a

longer period. Longitudinal research might make it possible to examine the

constancy of social positions and to examine the long-term development of social

relations, social-emotional functioning and learning performances. However, this

kind of research with pupils with special needs involves particular difficulties.


45

Pupils with special needs have a larger chance of changing school or of dropping

out than other pupils, resulting in a possible high drop-out rate of participants

during the research. For this reason it would be necessary to start with a large

number of participants.

In conclusion, considering the often found low social position of pupils with

special needs and the unfavourable side effects of a negative social position, it is

important to aim research at this topic. Further research into the assessment and

influencing of the social position of pupils with special needs in regular classes is

therefore recommended.

2.6 References

Borgatti, S.P., Everett, M.G. & Freeman, L.C. (1999). UCINET for Windows.

Natick: Analytic Technologies.

Cillessen, A.H.N. & Ten Brink, P.W.M. (1991). Vaststelling van relaties met

leeftijdgenoten. Pedagogische Studiën, 68(1), 1-14.

Coie, J.D., Dodge, K.A. & Coppotelli, H. (1982). Dimensions and types of social

status: A cross-age perspective. Developmental Psychology, 18(4), 557-

570.

Cullinan, D., Sabornie, E.J. & Crossland, C.L. (1992). Social mainstreaming of

mildly handicapped students. The Elementary School Journal, 92(3), 339-

351.




Gartin, B.C., Murdick, N.L. & Digby, A.D. (1992). Cooperative activities to assist

in the integration of students with disabilities. Journal of Instructional

Psychology, 19(4), 241-245.

Gower, J.C. (1971). A general coefficient of similarity and some of its properties.

Biometrics, 27, 857-871.

Hall, L.J. & McGregor, J.A. (2000). A follow-up study of the peer relationships of

children with disabilities in an inclusive school. The Journal of Special

Education, 34(3), 114-135.

CHAPTER 2

46

Jackson, L.D. & Bracken, B.A. (1998). Relationship between students’ social

status and global and domain-specific self-concepts. Journal of School

Psychology, 36(2), 233-246.

Juvonen, J. & Bear, G. (1992). Social adjustment of children with and without

learning disabilities in integrated classrooms. Journal of Educational

Psychology, 84(3), 322-330.

Larrivee, B. & Horne, M.D. (1991). Social status: a comparison of mainstreamed

students with peers of different ability levels. The Journal of Special

Education, 25(1), 90-101.

Limpens, M., Nekkers, J. & Ter Pelle, J. (2003). Samen of apart, anders of

minder. Onderwijs aan leerlingen met speciale leerbehoeften. Enschede:

Stichting leerplanontwikkeling.

Ministerie van OCW (2003). Wet op de Expertise Centra. Den Haag.

Nakken, H. & Pijl, S.J. (2002). Getting along with classmates in regular schools:

a review of the effects of integration on the development of social




longitudinal study. Journal of Consulting and Clinical Psychology, 60(1),

80-87.



Pijl, S.J. (2005). Interventies gericht op sociale integratie: training van sociale

vaardigheden voor leerlingen met beperkingen in het regulier onderwijs.

In B.F. van der Meulen, C. Vlaskamp & K.P. van den Bos (Eds.),

Interventies in de orthopedagogiek (pp. 122-135). Rotterdam:

Lemniscaat.

Pijl, Y.J. (1997). Twintig jaar groei van het speciaal onderwijs. De Lier:

Academisch Boeken Centrum.

Popping, R. (1983). Overeenstemmingsmaten voor nominale data. Groningen:

Rijksuniversiteit Groningen.

Ray, B.M. (1985). Measuring the social position of the mainstreamed

handicapped child. Exceptional children, 52(1), 57-62.

Rynders, J.E., Schleien, S.J., Meyer, L.H., Vandercook, T.L., Mustonen, T.,

Colond, J.S. & Olson, K. (1993). Improving integration outcomes for


47

children with and without severe disabilities through cooperatively

structured recreation activities: a synthesis of research. The Journal of


Sale, P. & Carey, D.M. (1995). The sociometric status of students with disabilities

in a full-inclusion school. Exceptional Children, 62(1), 6-19.

Scheepstra, A.J.M., Nakken, H. & Pijl, S.J. (1999). Contacts with classmates: the

social position of pupils with Down’s syndrome in Dutch mainstream






mental health: What, how, who and when. In P.S. Strain, M.J. Guralnick

and H. Walker (Eds.), Children’s Social Behavior: Development,

assessment and modifications (pp. 287-330). New York: Academic Press.


status among children. Developmental psychology, 27(5), 867-880.

CHAPTER 2

48

Chapter 3 Being part of the peer group: a literature study focusing on the social dimension of inclusion in education1

CHAPTER 3

BEING PART OF THE PEER GROUP: A

LITERATURE STUDY FOCUSING ON THE

SOCIAL DIMENSION OF INCLUSION IN

EDUCATION

This article is a slightly adapted version of: Koster, M., Nakken, H., Pijl, S.J. & Van Houten, E.J. (in press). Being part of the peer group: a literature study focusing on the social dimension of inclusion in education. International Journal of Inclusive Education.

CHAPTER 3

50

Abstract

Maximizing the interaction between pupils with and without special needs is

generally considered an important aspect of inclusion. However, it is frequently

questioned whether pupils with special needs in regular classrooms have

interactions and friendships with their peers. In order to be able to evaluate

these relationships, it is necessary to clarify concepts like social participation,

social integration and social inclusion. At the moment, there is much ambiguity

regarding these concepts. This article aims to elucidate these concepts and

reveal its characteristic themes.

An analysis of literature was carried out to identify these concepts. In the final

analysis 62 articles were included. This analysis showed that the concept social

integration and the related concepts of social inclusion and social participation

are often described inaccurately, with only a few researchers providing explicit

definitions or descriptions. In the majority of articles, implicit descriptions can be

derived from instruments used to measure social integration, social inclusion or

social participation: it is apparent there is much overlap among the use of

concepts by researchers.

Based on the analysis of the 62 articles, it can be concluded that the concepts

social integration, social inclusion and social participation are used as synonyms.

In our opinion, social participation is the most suitable concept. The analysis of

literature reveals four key themes central to all three concepts:

friendships/relationships, contacts/interactions, pupil’s social self-perception and

acceptance by classmates.

A LITERATURE STUDY FOCUSING ON THE SOCIAL DIMENSION OF INCLUSION IN EDUCATION

51

3.1 Introduction

The integration and, more recently, the inclusion of pupils with special needs into

regular education is a much-discussed topic worldwide. It is no longer taken for

granted that these pupils are sent to separate schools for special education.

Farrell (2000, pp. 154) describes inclusion as ‘taking a full and active part in

school-life, be a valued member of the school community and be seen as an

integral member’.

Research has shown that the opportunity for contacts and interactions with

typical peers is one of the main motives of parents for sending their child with

special needs to a regular school (Scheepstra, 1998, in Nakken & Pijl, 2002;

Sloper & Tyler, 1992; Strayhorn & Strain, 1986). In their opinion it is important

for their child to grow up as far as possible in a normal environment. Therefore,

they wish their child to attend a regular neighbourhood school. Parents often

assume that contacts with peers will have a positive effect on their child.

Furthermore, some parents believe that as a result of sending their child to an

ordinary school, a change of attitude will arise from other children, which can

possibly lead to positive long-term effects for the attitude of society as a whole

(Koster, Pijl, Van Houten & Nakken, 2007).

In addition, more intensive contacts with typical peers might have a positive

effect on the social-emotional development of the pupil with special needs (for

example on self-image). However, several researchers have shown there are also

risks in this area. Research shows that, compared to typical pupils, pupils with

special needs are more often teased, abused and ignored in mainstream settings,

which is, in fact, harmful to their self-image (Gottlieb, 1981, in Cole & Meyer,

1991). Wiener and Harris (1989) describe that pupils with special needs (in

particular pupils with learning problems) are rejected and neglected by their

peers. Research on social acceptance consistently shows that pupils with special

needs educated in a regular classroom are less accepted than their classmates

without special needs (Freeman & Alkin, 2000), generally have a social status

which is lower than that of their classmates (Coben & Zigmond, 1986; Goodman,

Gottlieb & Harrison, 1972; Kuhne & Wiener, 2000; Larrivee & Horne, 1991;

Manetti, Schneider & Siperstein, 2001; Pavri & Luftig, 2000; Sabornie, Marshall &

Ellis, 1990; Stone & La Greca, 1990) and are more socially isolated (Evans &

Labon, 1997, in Skårbrevik, 2005; Margalit & Efrati, 1996).

CHAPTER 3

52

From society’s point of view, it is important to aim research at the social

dimension of inclusion. However, there is uncertainty about what such concepts

as social participation, social integration, social status and social inclusion mean.

Many researchers use these in their studies, but it is sometimes unclear what

exactly they mean by them. These concepts are interpreted in numerous ways

(Storey & Smith, 1995), and a well-established criterion for determining the

degree to which pupils with special needs have contacts and friendships with

their peers does not yet exist (Guralnick, 1999, in Odom, 2000). Taking into

consideration the issues mentioned-above, the research question is as follows:

‘In what way may such concepts as social participation, social integration and

social inclusion be defined and what are the characteristic themes?’

3.2 Research method

To define the concepts mentioned above a literature analysis was carried out.

The literature used in this article was derived from two sources: the last six

volumes (2000 – 2005) of three prominent journals in the field of special needs

education (‘International Journal of Inclusive Education’, ‘European Journal of

Special Needs Education’ and ‘Scandinavian Journal of Educational Research’)

and a search of two electronic databases.

The three journals included in the literature search all meet international

publication and quality requirements and are often used as reference. Of the

2005 volume of the ‘International Journal of Inclusive Education’ only the first

two issues were examined, as the last two were not available. Articles with

abstracts and/or titles that included the following terms were selected: ‘social

integration’, ‘social inclusion’, ‘social participation’, ‘social position’ and ‘social

status’.

Subsequently, a search was made of the American Psychological Association

(psycINFO) and Education Resource Information Clearinghouse (ERIC) databases

from January 1995 until December 2005 and the same aforementioned search

terms used, combined with the additional search terms: ‘child(ren)’, ‘pupil(s)’,

‘student(s)’, ‘classmate(s)’, ‘peer(s)’, ‘disabled’, ‘disability/disabilities’,

‘impairment(s)’, ‘impaired’, ‘disorder(s)’, ‘handicap(s)’, ‘handicapped’, ‘special

needs’, ‘special educational needs/SEN’, ‘contact’, ‘relationship’, ‘communication’,

‘social network’, ‘school’, ‘classroom’ and ‘mainstream’.


53

There were five selection criteria for the articles:

• based on empirical research or a literature review, so editorials,

commentaries and prologues were excluded;

• published in international scientific journals;

• aimed at pupils with special needs;

• aimed at elementary school pupils or preschoolers;

• focused on the social dimension of inclusion in the school situation (rather

than in the wider community).

In analysing the literature the focus was on the following three themes:

• definition of the concepts used;

• relation between the various concepts used;

• any subdivision in themes within the concepts used.

3.3 Selecting the literature

The search within the last six volumes of the three international journals yielded

15 articles, of which 9 were from the ‘European Journal of Special Needs

Education’, 4 from the ‘International Journal of Inclusive Education’ and 2 from

the ‘Scandinavian Journal of Educational Research’. The search in the electronic

databases resulted in several hundred references.

Because of the large amount of hits, the references were immediately filtered,

based on the criteria formulated above. The first selection of the articles was

based on reading the titles and in cases of uncertainty about the suitability of an

article, also the abstract. Quite a number of articles were deleted from the file.

Many articles, for instance, looked promising but were not aimed at primary

school pupils and preschoolers or focused on the inclusion of other target groups.

After the first rough filtering process, 124 articles from the electronic databases

were selected for further analysis. Together with the 15 articles from the three

journals, the total number of articles was 139. Another 9 publications were

added after examining the references of the selected studies, resulting in 148

studies to be assessed. Of these, 62 were selected after carefully reading the

article. The other 86 articles were put aside, because they appeared not to focus

on including pupils with special needs in mainstream primary education. Most of

the deleted 86 articles focused on pupils with special needs in (post) secondary

education settings and/or on pupils without special needs.

CHAPTER 3

54

A first exploration of the selected literature clearly shows that various

terminologies are used to refer to the social dimension of inclusion. Most of these

are not defined explicitly. Authors often only describe the concepts in an implicit

way, for example by just describing which instruments were used to measure

them. In the following sections, the three most used concepts will be described:

social integration, social inclusion and social participation.

3.4 Social integration

The analysis of the 62 publications made clear that definitions of social

integration not only differ with respect to content, but also in the way they are

presented. Only nine articles presented an explicit definition of social integration.

These explicit definitions are now discussed followed by the implicit definitions of

social integration.

3.4.1 Explicit definitions of social integration

Boutot and Bryant (2005) give a precise definition of social integration. According

to the authors, successful social integration in a mainstream education classroom

means being visible to other pupils (social impact), being someone with whom

other pupils wish to spend time (social preference) and being a member of a

group of friends that spend time together (social network affiliation).

Cartledge and Talbert Johnson (1996) refer to a definition originally proposed by

Cullinan, Sabornie and Crossland (1992). They define social integration as: being

an accepted member of a group, having at least one mutual friendship and

participating actively and equivalently in group activities. Stinson and Antia

(1999) state that inclusive education aims to promote the social and academic

integration of pupils with disabilities. The authors emphasise the importance of

peer acceptance and friendship and consider peer interaction to be important for

social integration. They define the latter as the ability to interact with peers,

make friends with peers and be accepted by peers. This definition builds on work

done by Cullinan et al (1992) and Juvonen and Bear (1992). However, Juvonen

and Bear’s definition emphasizes not only the importance of friendships and peer

relations, but also of the pupil’s self-perception of social acceptance.


55

Guralnick (1999) evaluated the social integration of preschool-age pupils with

mild disabilities in inclusive early childhood settings by focusing on three

constructs: the connectedness (or extent) of peer interactions, the quality of

interpersonal relationships and the nature of adjustments that occur during social

exchanges. Social integration of such pupils is achieved when typically

developing pupils connect with and maintain the same quality of interpersonal

relationships with pupils with mild developmental delays as they do with other

pupils. Guralnick (1999) points out that this description departs from the

viewpoint of typical peers. Similar analyses can be made from the perspective of

pupils with special needs. Their ability to connect with their peers, with and

without special needs, constitutes an important index of social integration.

The definition of social integration by Schmidt (2000) is less specific than those

described above. Following Kobi (1983), Schmidt (2000) holds the view that

social integration refers to the frequency and intensity of social contacts between

pupils with and without special needs. This view is similar to Biklen’s (1985, in

Nietupski, 1995) that social integration is about developing positive social

interactions between children with and without special needs. Gable, Rucker and

Smith (1997) describe social integration as pupils’ perceptions of the extent to

which they participate in and are accepted as members of the school community.

The authors used a student survey to measure social integration, the items of

which show that acceptance by, and contact with, classmates as well as pupils

with special needs feeling accepted are important aspects of social integration.

3.4.2 Implicit definitions of social integration

In several articles focusing on social integration only implicit definitions are

presented. According to Cambra and Silvestre (2003), one of the factors which

plays an important role in social integration is ‘peer group socialisation’. They

consider social acceptance to be the essence of social integration. This is in

accordance with Stanovich, Jordan and Perot (1998), who assessed the peer

acceptance of pupils in inclusive classrooms. They consider social integration to

be synonymous with peer acceptance, which can be measured using a

sociometric rating scale. Pijl and Hamstra (2005) also hold the view that social

integration of pupils with special needs can be assessed by using a sociometric

questionnaire.

CHAPTER 3

56

Guralnick, Connor, Hammond, Gottman and Kinnish (1995) however, clearly

distinguish social integration from social acceptance. They examined the

immediate effects of mainstream settings on the social development of preschool

children by measuring acceptance using sociometric ratings, while social

integration was assessed by observing peer interaction. The authors used four

indices for social integration: active and passive interaction, and positive and

negative individual social behaviour. Just like Guralnick et al. (1995), Pearl et al.

(1998), Scheepstra, Nakken and Pijl (1999), Cava and Musitu (2000), Flem and

Keller (2000), Nakken and Pijl (2002) and Dyson (2005) do not consider social

acceptance as the only important aspect, but also highlight social relationships.

For example Pearl et al. (1998) state that focusing on social acceptance provides

an incomplete view of social integration and regard the relationships (affiliations)

between pupils with special needs and their peers as central to social integration.

Pearl et al. (1998) examined social integration of pupils in primary-school

classrooms by assessing social isolation, social behaviour and peer group

affiliations. Farmer and Farmer (1996) hold similar views as they gathered

information on affiliated patterns and social structural characteristics in

mainstream classrooms. They define social integration in terms of being isolated

or belonging to a peer network and distinguish types of peer networks (pro-

social, anti-social or shy cluster) (Farmer & Farmer, 1996). Storey and Smith

(1995) also emphasize peer networks. They consider ‘clique’ analysis of social

interactions to be a promising methodology for assessing social integration, since

it not only provides a framework for comparing frequency of interaction, but also

reveals the ‘who-with-whom’ dimension of interaction patterns (Storey & Smith,

1995). A focus on social interaction as an essential part of social integration can

also be found with Attwood (2000), Brown, Odom, Li and Zercher (1999),

Goldstein and English (1995), Kamps, Dugan, Potucek and Collins (1999) and

Luetke-Stahlman (1995).

Besides social interaction, Brown et al. (1999), in their assessment of social

integration / social inclusion (they use these terms interchangeably), focus on

the social initiations, social bids and social behaviours of pupils with and without

special needs. Mar and Sall (1995) and Nikolaraizi and De Reybekiel (2001) add

that next to social interaction (long term) relationships between pupils are also

important in this context.


57

Harper, Maheady, Mallette and Karnes (1999) only focus on this last aspect:

friendships between pupils, with and without special needs. They describe two

peer-tutoring programmes that have been used effectively to promote social

integration (and academic integration) of children with special needs in regular

classrooms. The benefits included improved friendship patterns within the

classroom (Harper et al., 1999).

Williams, Johnson and Sukhodolsky (2005) consider a combination of the

foregoing aspects to be important for social integration. In their descriptions of

strategies to facilitate the social integration of pupils with autism spectrum

disorders (ASD), they focus on the social interactions, (reciprocal) friendships,

peer acceptance and social involvement (affiliations) of pupils with ASD with their

peers (Williams et al., 2005). Yude, Goodman and McConachie (1998) also

consider friendship and acceptance to be essential and positive aspects of social

integration, while two other important aspects ‘being a victim’ and ‘being a bully’

are negative ones. Odom (2000) also distinguishes a positive and a negative

indicator of social integration: social acceptance and social rejection.

Finally, Hepler (1998) explores the social integration of pupils diagnosed as

having emotional difficulties by measuring four aspects: social status, knowledge

of specific cognitive and behavioural skills (like initiating and maintaining a

conversation, entering an ongoing activity), social interaction and pupils’

satisfaction with an integration programme.

3.4.3 Summary of the definitions of social integration

The analysis of articles focusing on social integration shows that in particular,

interaction, peer acceptance, friendship and relationships/affiliations are

regarded by many authors as essential aspects of social integration. Social

behaviour/skills, peer networks, pupils’ self-perception of acceptance, social

contacts, social initiations, social status and participation in group activities are

also named by one or more researchers as important for social integration. In

addition to these positive or neutral aspects, three negative aspects are

described: social isolation, social rejection and bullying. See Table 1 for an

overview of the outcomes.

Table 1. Systematic summary of analysis of social integration Author(s) Type of disability Grade/Age Items within key concept social integration

Boutot & Bryant (2005)

Mainly autism, PDD-NOS (few other disabilities)

Grades 2-5

Social impact, social preference, social network affiliation

Dyson (2005) Traumatic brain injury, orthopedic impairment (use of wheelchair), Down Syndrome, ADHD

Kindergarten (age 5-6)

Peer interactions, peer acceptance

Cambra & Silvestre (2003)

Special educational needs (hearing disabilities, cognitive, relational, learning, motor, visual problems)

Grades 4-6 of primary school, Grades 1 and 2 of secondary school

Peer group socialisation, social acceptance

Nakken & Pijl (2002) Sensory, motor and/or intellectual disabilities

Age 3-14 (review of 14 studies)

Social relationships

Nikolaraizi & De Reybekiel (2001)

Deafness, blindness, physical disability (use of wheelchair)

Grades 5-6 (age 10-12)

Interactions, friendship

Attwood (2000)

Asperger Syndrome No specification Social interactions, peer relationships

Cava & Musitu (2000) Socioaffective difficulties Grades 5 -8 Interpersonal relationships (as measured by sociometric questionnaire), peer acceptance

Flem & Keller (2000)

No specification No specification Social relationships

Odom (2000) Disabilities (no further specification)

Preschool Social acceptance and social rejection

Schmidt (2000)

Learning disabilities Grade 4 Frequency and intensity of social contacts

Brown, Odom, Li & Zercher (1999)

Mild to moderate, severe, and multiple disabilities

Preschool Social interactions, social initiations, social bids, social behaviours

Guralnick (1999) Mild disabilities Preschool Connectedness (or extent) of peer interactions, quality of interpersonal relationships, nature of adjustments during social exchanges

Author(s) Type of disability Grade/Age Items within key concept social integration

Harper, Maheady, Mallette & Karnes (1999)

Minority children with disabilities

No specification

Friendship

Kamps, Dugan, Potucek & Collins (1999)

Autism Grade 1 and Grade 4 Social interactions

Scheepstra, Nakken & Pijl (1999)

Down Syndrome Grade 1 Social relationships

Stinson & Antia (1999)

Hearing disabilities No specification Peer acceptance, friendship, (ability to have) interactions with peers

Hepler (1998) Emotional disabilities Grade 5 Social status, knowledge of specific cognitive and behavioural skills, social interactions, pupils’ satisfaction with integration programme

Pearl, Farmer, Van Acker, Rodkin, Bost, Coe & Henley (1998)

Mild disabilities Grades 4-6 Peer affiliations/relationships, social acceptance

Stanovich, Jordan & Perot (1998)

Special needs (no further specification)

Grades 2-8 Peer acceptance (as measured by a sociometric rating scale)

Yude, Goodman & McConachie (1998)

Hemiplegia Age 9-10 Acceptance, friendship, being a victim, being a bully

Gable, Rucker & Smith (1997)

Special educational needs (no further specification)

Grades 4, 6 and 8 Self-perception of the extent of participation in and being an accepted member of the school community

Cartledge & Talbert Johnson (1996)

Emotional and behavioural disorders

No specification Peer acceptance, mutual friendship, active and equivalent participation in group activities

Goldstein & English (1995)

Moderate developmental disabilities

Preschool Communicative interactions

Author(s) Type of disability Grade/Age Items within key concept social integration

Guralnick, Connor, Hammond, Gottman & Kinnish (1995)

Developmental (cognitive) delays

Preschool

Active and passive peer interactions, positive and negative individual social behaviours (distinction between peer acceptance and social integration)

Luetke-Stahlman (1995)

Hearing disabilities

No specification Social interactions

Mar & Sal (1995) Deaf-blindness Age 7-10 Socially integrated activities (for example visiting a friend at home, attending a party, contact with peers in the playground), interactions, relationships

Storey & Smith (1995)

Social delays, social isolation Preschool Interactions (as measured by sociometric questionnaires and clique analysis), friendship

Williams, Johnson & Sukhodolsky (1995)

Autistic spectrum disorders School-age (no specification)

Social interactions, playing together, social involvement with peers

Cullinan, Sabornie & Crossland (1992)

No specification No specification Peer acceptance, mutual friendship, active and equivalent participation in group activities

Juvonen & Bear (1992)

Learning disabilities Grade 3 Peer acceptance, friendship, self-perception of peer acceptance

Biklen (1985, in Nietupski, 1995)

Disabilities (no further specification)

No specification Social interactions


61

3.5 Social inclusion

In the former section a distinction was made between explicit and implicit

definitions of social integration. A similar distinction cannot be made for social

inclusion, since the analysis of articles only reveals implicit definitions.

De Monchy, Pijl and Zandberg (2004) examine the social inclusion of pupils with

behavioural problems. The authors do not define social inclusion, but the aspects

which they consider to be important for social inclusion may be derived from

their measuring instruments. De Monchy et al. (2004) assess the social inclusion

of pupils with behavioural problems by measuring their social position in the

classroom and by examining the degree to which they are bullied by classmates

or they actively bully other pupils themselves. The social position of the pupils is

described in terms of being liked, performing a task together and number of

friends. Similar aspects of social inclusion (acceptance, friendship, being a victim

and being a bully) are distinguished by Yude et al. (1998) and Savage (2005).

Ring and Travers (2005) describe the inclusion of a pupil with severe general

learning difficulties in a mainstream school. Data about the pupil’s level of social

inclusion are based on the pupil’s sociometric status, the frequency and nature of

interactions with classmates and the number of acknowledged and

unacknowledged initiatives to make contact with classmates.

Skårbrevik (2005) examines the social inclusion of pupils with special needs in

mainstream Norwegian classrooms using a teacher questionnaire. The

questionnaire comprises ten questions, covering a pupil’s participation in play

and activities, being alone during break times, being teased by other pupils,

invitations to parties and taking the lead in social activities - aspects seen as

essential for social inclusion. This is in line with Hunt, Alwell, Farron-Davis and

Goetz (1996), who investigated the effectiveness of an intervention designed to

facilitate the social inclusion of three pupils with severe multiple special needs by

promoting social relationships, friendships and interactive partnerships between

pupils. This implies that Hunt et al. (1996) see these three aspects as the core of

social inclusion.

Taylor, Asher and Williams (1987, in Frederickson & Furnham, 2004) are less

extensive in their description and measurement of social inclusion. They

examined the social inclusion of pupils with moderate learning difficulties by

using positive peer nominations and a play-rating scale. Frederickson and Turner

CHAPTER 3

62

(2003) also include a sociometric rating scale to assess social inclusion. They

evaluate the Circle of Friends intervention approach, aimed at improving the

social acceptance/social inclusion (the terms are used interchangeably) of pupils

with special needs. They state that the average play rating (acquired via a

sociometric rating scale) given by classmates provides a measure of social

acceptance/social inclusion during this activity. The evaluation also includes

teachers’ views on the social acceptance of pupils with special needs and the

self-perception of the pupils on their social acceptance. Just like Frederickson and

Turner (2003), Pavri and Luftig (2000) include pupils’ own perceptions in

assessing social inclusion. They examine the ‘social face’ of inclusive education

by investigating the perceived loneliness, social competence and social status of

pupils with learning disabilities in such a setting.

3.5.1 Summary of the definitions of social inclusion

The analysis of social inclusion reveals several aspects. Friendship, acceptance,

interaction, relationships, social status and bullying are aspects described most

frequently. Others include performing a task together, initiation, interactive

partnerships, pupils’ self-perception of acceptance, perceived loneliness and

perceived social competence. For an overview of the outcomes we refer to Table

2.

Table 2. Systematic summary of analysis of social inclusion

Author(s) Type of disability Grade/Age Items within key concept social inclusion

Ring & Travers (2005)

Severe general learning difficulty

Primary school age (multi-grade classroom)

Sociometric status, frequency and nature of peer interactions, acknowledged and unacknowledged initiations of pupil with special needs to get in touch with classmates

Savage (2005)

Speech/language difficulty Grade 7 Friendship, bullying, playing and working together

Skårbrevik (2005) Pupils who are found eligible for special education

Grades 1-9 Among other things: participation in play and in activities, joining other pupils on the way back and forth to school, being alone during breaks, being teased, receiving invitations to parties, taking the lead in social activities (summarized: interactions, relationship/friendship)

De Monchy, Pijl & Zandberg (2004)

Behavioural problems Grades 2-5 (age 9-12)

Social position in classroom, bullying, being bullied

Frederickson & Turner (2003)

Emotional and behavioural difficulties

Grades 1-5 (age 6-12)

Social acceptance (social inclusion and social acceptance are viewed as interchangeable concepts)

Hunt, Alwell, Farron-Davis & Goetz (1996)

Severe multiple disabilities Grade 1 and Grade 4

Social relationships, friendships, interactive partnerships

CHAPTER 3

64

3.6 Social participation

The analysis of articles featuring social participation also only reveals implicit

definitions. Of all the researchers describing social participation, Kennedy,

Cushing and Itkonen (1997) and Kennedy, Shukla and Fryxell (1997) are most

explicit in their descriptions. According to Kennedy, Cushing et al. (1997) social

participation comprises two aspects: social contacts between pupils with special

needs and their typical classmates and friendship networks. In another study

(Kennedy, Shukla et al. 1997) the concept of social participation is extended to

include a third dimension: social support behaviours (like greetings, material aids

and emotional support).

Hunt, Soto, Maier and Doering (2003) investigated the effectiveness of a

mainstream / special education collaborative process on the academic and social

participation of pupils with severe disabilities and pupils academically at risk in

mainstream classrooms. To assess the social participation of the pupils, the level

of their engagement and interaction patterns (e.g. initiating and responding to

interactions, participating in conversations, working collaboratively) with

classmates were systematically observed (Hunt et al., 2003). In another study

(Hunt, Soto, Maier, Müller & Goetz, 2002), the social participation of pupils with

augmentative and alternative communication (AAC) needs in mainstream

classrooms was studied. Again, the interaction patterns between pupils were

central in assessing social participation. Not only the frequency of the

interactions, but also the quality (i.e. positive, neutral or negative) and their

communicative function (i.e. a request, protest, comment, or assistance) were

identified. The research of Hunt et al. (2002) shows that social interaction

between pupils with special needs and their classmates seems to be at the core

of social participation. Just like Hunt et al. (2002), Bayliss (1995) emphasizes the

significance of the quality of interaction between pupils with special needs and

their typical peers and the importance of having familiar and equal relationships

between pupils.

Ialongo, Edelsohn, Werthamer-Larsson, Crockett and Kellam (1996) assess the

social participation of first-grade pupils with symptoms of anxiety and

depression. Their measuring instrument (the Peer Assessment) shows that social

participation, among other things, is about playing with other children and the

ability to make friends. Bokhorst, Goossens and De Ruyter (2001) also focus on


65

the social participation of anxious pupils. To measure social participation,

observations took place during breaktimes. The authors used an observation

scale which distinguished five levels of increasing social participation, ranging

from ‘unoccupied/not looking’ to ‘cooperative play’. Also Harper and McCluskey

(2002) distinguish levels of social participation by pupils with special needs. They

compare the free-play social behaviours of children with special needs in

inclusive preschool programmes with those of their typically developing

classmates. Four levels of social participation were distinguished, ranging from

‘solitary’ to ‘interaction’ (Harper & McClusky, 2002). In presenting their research

results the authors use the terms ‘social integration’ and ‘social participation’

interchangeably. Just like Harper and McCluskey (2002), Mirenda (1998, in

Power & Hyde, 2002) distinguishes four levels of social participation: none,

involved, active and competitive. The first level implies that the pupil with special

needs is not involved in social activities or in interaction with peers. A socially

competitive pupil is actively involved in the social dynamics of the group, exerts

influence within it and determines his/her own involvement in social activities

(Mirenda, 1998, in Power & Hyde, 2002).

In the ‘International Classification of Functioning, Disability and Health’ (ICF),

developed by the World Health Organization (WHO, 2001), ‘Activities and

Participation’ is an important component in the classification scheme. Domains

within this, which are relevant to the school situation, focus on interpersonal

interaction and relationships.

3.6.1 Summary of the definitions of social participation

The analysis of social participation makes clear that interactions comprise the

most important part of social participation. In addition, friendships, friendship

networks, (lasting) relationships, playing together and social contacts are

described by researchers as major aspects of social participation. Others include

social support behaviours, involvement in group activities and being solitary. See

Table 3 for an overview of the outcomes.

Table 3. Systematic summary of analysis of social participation

Author(s) Type of disability Grade/Age Items within key concept social participation

Hunt, Soto, Maier & Doering (2003)

Severe disabilities

Grade 4

Assessment by observation of levels of engagement and interaction patterns (e.g. initiating and responding to interactions, participating in conversations, providing and receiving assistance, working collaboratively)

Hunt, Soto, Maier, Müller & Goetz (2002)

Augmentive and alternative communication needs

Kindergarten, Grade 1, Grade 5

Frequency and quality of interactions, communicative function of interactions (i.e. request, protest, comment, assistance)

Bokhorst, Goossens & De Ruyter (2001)

Social anxiety Grades 1-2 Five levels of social participation, varying from ‘unoccupied/onlooking behaviour’ to ‘cooperative play’

Mirenda (1998, in Power & Hyde, 2002)

Hearing disabilities No specification Four levels of social participation, varying from ‘none’ to ‘competitive’

Kennedy, Cushing & Itkonen (1997)

Severe disabilities Age 11 and 18 Social contacts, friendship networks

Kennedy, Shukla & Fryxell (1997)

Severe disabilities Intermediate school age

Social interactions, friendship networks, providing and receiving social support behaviours

Ialongo, Edelsohn, Werthamer-Larsson, Crocket & Kellam (1996)

Anxious and depressive symptoms

Grade 1 Among other things: playing with other children, the ability to make friends

Bayliss (1995) (Severe physical) disabilities No specification (Quality of) interactions, familiar and equal relationships between pupils (which is the aim of inclusion)


67

3.7 Other related concepts

Several studies describe aspects of social integration, social inclusion or social

participation without mentioning these umbrella concepts. Peer acceptance or

social acceptance is a concept frequently used by researchers. Frederickson and

Furnham (1998), Cook and Semmel (1999), Manetti, Schneider and Siperstein

(2001), Davis, Howell and Cooke (2002), Kemp and Carter (2002), Scholtes,

Vermeer and Meek (2002), Doll, Murphy and Song (2003), and Kuhne and

Wiener (2000) consider peer acceptance (as measured by sociometric

questionnaires) to be of great importance in inclusive education. Arendse,

Vermeer and Severijnen (1998) and Scholtes et al. (2002) stress the importance

of perceived social acceptance. They measured this together with perceived

competence. Peers’ acceptance of children with cerebral palsy focused on aspects

like playing with friends at home, playing with friends in the playground,

receiving invitations to birthday parties, staying at a friend’s home and sitting

next to other children. Besides acceptance, Davis et al. (2002) emphasize the

importance of peer relations and friendships between pupils with special needs

and their typical peers.

Kanioglou, Tsorbatzoudis and Barkoukis (2005) and Conderman (1995) focus on

the social status of pupils with special needs in inclusive classrooms. Kanioglou et

al. (2005) examined the socialisation of primary school pupils described as

having developmental coordination disorders by measuring their social status in

the classroom using a sociogram while Conderman (1995) used a sociogram to

assess the social status of sixth and seventh graders with learning difficulties.

Contrary to these authors, Hall and McGregor (2000) believe that assessment

restricted to the social status of pupils with special needs might overlook

essential aspects. Children may have reciprocal relationships regardless of their

social status. Siperstein and Leffert (1997) as well as Cook and Semmel (1999)

also emphasize the need to go beyond sociometric assessments to examine

relationships of pupils with intellectual disabilities in greater depth, using

interviews focusing on qualitative features of relationships, as well as direct

observation of the behavioural manifestations of friendship.

Vaughn, Elbaum, Schumm and Hughes (1998) describe the social outcomes for

pupils with and without learning disabilities in inclusive classrooms more

extensively, focusing on peer acceptance, friendship and friendship quality, social

CHAPTER 3

68

skills and self-concept. Kemp and Carter (2002) also focus on more than one

aspect. Besides emphasising social acceptance in their study, they focus on the

social skills of pupils with intellectual disabilities and the social interactions

between these pupils and their classmates.

3.7.1 Summary of the definitions of other related concepts

The majority of researchers not using concepts like social integration, social

inclusion or social participation focus on peer- or social acceptance. Next to peer

acceptance, self-perceived social acceptance, (reciprocal) relationships,

friendships, social skills, social status, social interaction, self-concept and

perceived competence are also used as important aspects of inclusion in

education. Table 4 presents an overview of the outcomes.

Table 4. Systematic summary of related concepts

Key concept Author(s) Type of disability Grade/Age Items within key concept(s)

Peer acceptance

Frederickson & Furnham (2004)

Moderate learning difficulties

Middle school (age 8 -12)

Measurement by peer assessment questionnaire and sociometric questionnaire

Doll, Murphy & Song (2003)

Recess problems Grades 3-7 Measurement by sociometric nominations or sociometric rating scales

Davis, Howell & Cooke (2002)

Stuttering Age 8-14 Peer relations and friendships. Measurement of acceptance by using a sociometric questionnaire

Kemp & Carter (2002) Intellectual disabilities Preschool, follow-up in Grades 1-5

Social interactions, social status

Scholtes, Vermeer & Meek (2002)

Cerebral palsy Age 4-9 Among other things: playing with friends at home and in the playground, invitations to parties, staying at a friend’s home, sitting next to children

Manetti, Schneider & Siperstein (2001)

Intellectual disabilities Grades 3-5 Sociometric questionnaire, attitudes of typical pupils towards prospective pupils with special needs

Cook & Semmel (1999) Mild to severe disabilities Grades 2-6 Measurement by peer nominations with regard to playing and working

Frederickson & Furnham (1998)

Learning difficulties No specification (literature review)

Measurement by sociometric techniques

Social adjustment

Kuhne & Wiener (2000) Learning disabilities Grades 4-6 (age 9-12)

Peer acceptance, as measured by using sociometric questionnaires

Social face of inclusion

Pavri & Luftig (2000) Learning disabilities Grade 6 Perceived loneliness, perceived social competence, social status

Socialisation Kanioglou, Tsorbatzoudis & Barkoukis (2005)

Developmental Coordination Disorder

Grades 5-6 Social status (as measured by sociometric questionnaire)

Key concept Author(s) Type of disability Grade/Age Items within key concept(s)

Social status

Conderman (1995)

Learning disabilities

Grades 6-7

Use of sociogram, self-rating questionnaire, teacher rating scale

No specific concept

Hall & McGregor (2000) Multiple disabilities, Down Syndrome

Kindergarten/Grade 1, follow-up in Grades 4-6

Peer relationships, peer interactions

Vaugn, Elbaum, Schumm & Hughes (1998)

Learning disabilities Grades 3-6 Peer acceptance, friendship, friendship quality, social skills, self-concept

Siperstein & Leffert (1997)

Intellectual disabilities Grades 4-6 Social acceptance, social rejection, friendship, relationship

Farmer & Farmer (1996)

Learning disabilities, emotional and behavioural disorders

Grades 3-4 Social networks, social relationships, peer affiliations

Table 4. Scores of pupils with and without special needs on key themes of social participation

Number of

friends

Initiated

interactions

with

classmates

Received

interactions

with

classmates

Interactions

with teacher

Social self-

perception

score Grades

1/2

Social self-

perception

score Grade 3

Acceptance

score

Pupils without special needs

2.9 (1.4) n=202

14.7 (9.6) n=58

11.0 (6.4) n=58

3.1 (3.7) n=58

14.4 (2.3) n=99

17.3 (4.0) n=42

-0.06 (1.0) n=148


1.9 (1.3) n=137

10.7 (8.3) n=58

7.1 (5.2) n=58

8.8 (9.7) n=58

14.2 (3.2) n=67

17.5 (4.2) n=27

-0.71 (1.0) n=96

Category of

disabilities

Autistic spectrum disorder

2.0 (1.1) n=54

9.8 (8.3) n=26

6.9 (6.2) n=26

7.6 (10.5) n=26

13.1 (3.3) n=30

17.6 (4.3) n=11

-0.92 (0.9) n=42

Speech/ language disability

2.2 (1.4) n=27

10.3 (6.2) n=9

6.8 (3.8) n=9

7.7 (6.2) n=9

14.8 (2.8) n=14

17.0 (3.7) n=6

-0.25 (1.0) n=20

Motor disability

1.8 (1.6) n=25

8.0 (5.0) n=7

6.7 (4.1) n=7

5.7 (6.7) n=7

14.2 (3.5) n=5

17.4 (4.6) n=5

-0.25 (1.5) n=10

Behavioural disorder

1.6 (1.3) n=16

12.3 (6.2) n=8

6.1 (3.0) n=8

13.0 (12.7) n=8

15.0 (3.6) n=9

19.5 (4.2) n=4

-0.89 (1.1) n=13

Intellectual disability

1.5 (1.4) n=15

14.9 (12.2) n=8

9.3 (6.0) n=8

12.8 (9.0) n=8

15.8 (1.9) n=9

11.0 (no data) n=1

-0.99 (1.1) n=11


71

3.8 Conclusion

The analysis of literature (presented in Tables 1 - 4) confirms our assumption

that the social dimension of inclusion is described in numerous ways. This does

not imply that the various authors attach very different meanings to these

concepts. On the contrary, the analysis of the 62 articles shows that the concept

of social integration and its related concepts, social inclusion and social

participation, hardly seem to differ in practice with respect to content, if at all.

These three different concepts are often used in very similar ways. In daily

(research) practice the concepts social integration, social inclusion and social

participation are used almost synonymously. Much has been written about the

differences, especially between the concepts integration and inclusion (Booth,

2004; Hegarty, 1991), but in describing the social dimension of inclusion these

differences in meaning seem suddenly irrelevant to most authors. This fact is

visualized in Figure 1.

CHAPTER 3

72

Friendships/relationships

Contacts/interactions

Friendship network

Mutual friendship

Playing together

Working together on

tasks

Participation in group

activities

(Un) acknowledged initiations

Social isolation

Social self-perception of pupil

Self-perception of peer

acceptance

Satisfaction at school

Social self-concept

Self-perception of social

competence

Loneliness

Acceptance by classmates

Social preference

Social support (behaviours)

Bullying

Social rejection

Social inclusionSocial

integrationSocial

participation

Figure 1. Overview of key themes and their aspects within social integration (and the

related concepts social inclusion and social participation)

The top section of Figure 1 presents the concepts of social integration, social

inclusion and social participation as three expressions for basically the same

phenomenon. This is based on the literature analysis which pointed out that in

daily practice many researchers use these terms interchangeably and attach

similar meanings to social integration, social inclusion and social participation. It

is obvious that using different concepts for the same notion causes confusion and

ambiguity. Research would profit from using one concept for the social dimension

of inclusion. The concept ‘social integration’ refers too much to integration, which

is an outmoded term and nowadays almost has a negative connotation. The

concept ‘social inclusion’ might be regarded as a pleonasm as ‘inclusion’ is an


73

extensive concept which logically embraces the social dimension. That leaves us

with the concept ‘social participation’ as the most suitable concept for the social

dimension of inclusion. In the remainder of this article we will therefore use the

term ‘social participation’.

The analysis of literature further revealed four key themes central to social

participation. These are presented on the second line in Figure 1:

• friendships/relationships;

• contacts/interactions;

• social self-perception of the pupil;

• acceptance by classmates.

Based on these key themes, the following provisional description of social

participation is formulated:

‘Social participation of pupils with special needs in regular education is the

presence of positive social contact/interaction between these children and their

classmates; acceptance of them by their classmates; social

relationships/friendships between them and their classmates and the pupils’

perception they are accepted by their classmates’.

Finally Figure 1 presents an overview of the various ways in which the four key

themes are usually measured in the literature surveyed here.

3.9 Discussion

Inclusive education is often used as a concept referring to a broad process in

schools and their local communities (see Booth & Ainscow, 2002). At the same

time it is used to refer to an education system that meets the needs of a wide

diversity of pupils. It focuses not only on academic issues, but also on meeting

pupils’ physical, social and emotional needs. Illustrative is the OECD definition

(1995) which focuses on the “process which maximises the interaction between

disabled and non-disabled pupils”. Recent literature on inclusive education has

emphasized the importance of meeting social needs more emphatically.

It is therefore not surprising that several research studies have addressed this

topic. Researchers looked at whether pupils with special needs managed to have

contacts and friendships with their peers in inclusive settings and were interested

in the reactions of these peers. In studying this topic, however, a wide array of

barely defined concepts and research instruments were used. This resulted in a

CHAPTER 3

74

collection of studies reporting on very different aspects while using the same

concepts and in studies using very different concepts but actually describing the

same phenomena.

This study focuses on clarifying such concepts as social integration, social

inclusion and social participation. It is apparent that these concepts are often

defined inaccurately, with only a few authors defining what they mean. In most

cases, implicit definitions can be derived from instruments used to measure

social integration, social inclusion or social participation. Several articles even

lack implicit definitions.

The analysis of literature further shows that definitions of the concepts and their

operationalisation in research instruments are often mixed up. For example, in

several studies social acceptance is regarded as an important aspect of inclusive

education. However, in other studies social acceptance is viewed as an

operational transcription of umbrella concepts like ‘social integration’, ‘social

inclusion’ or ‘social participation’. Researchers also use concepts like ‘social

acceptance’, ‘social preference’, or ‘social nominations’ and suggest that these

are very different. In fact these are all associated with sociometry. By means of

sociometric questionnaires, social preference scores and social impact scores can

be calculated, mostly on the basis of received nominations. With these scores,

the degree of social acceptance or peer acceptance can be assessed. It is

important to realise that these terms all amount to the same idea.

Many studies use the concepts ‘social integration’, ‘social inclusion’ and ‘social

participation’ as if they mean exactly the same. The elaborations of these

concepts largely overlap and in practice hardly differ from each other. It is

therefore proposed to use only the term ‘social participation’ for the social

dimension of inclusion. It was further possible to distinguish four key themes

within the concept of social participation. The analysis of literature shows that

some studies focus on only one or two of these themes or on aspects of them.

We wish to emphasize that the importance of each of these themes might differ

for different pupils. Figure 1 enables researchers to select, in accordance with

their sample, their own criteria for social participation.

The provisional description of social participation and its key themes are based

on the outcomes of the literature analysis. It must be remembered that our

conclusions are limited to primary education. Possibly, an exploration of


75

literature which aims at secondary or post secondary education might reveal

other themes (or other accents within the themes) of social participation.

Keeping in mind this limitation, the analysis described in this article should be

considered as a first step in elucidating the concepts of social integration, social

inclusion and social participation. In further research, the translation of the key

themes into operational concepts should be central. When it is clear in which

manner the themes can be measured, the social participation of pupils with

special needs in regular primary schools can be monitored and problems can be

noticed in time. As the inclusion of pupils with special needs into mainstream

education is becoming increasingly common worldwide, the monitoring of this

major facet of inclusion is of great importance.

3.10 References

Arendse, A., Vermeer, A. & Severijnen, J. (1998). De Platenschaal voor zelf

waargenomen competentie en sociale acceptatie bij kinderen met

cerebrale parese: een factoranalytisch onderzoek. Bewegen &

Hulpverlening, 15, 224-234.

Attwood, T. (2000). Strategies for improving the social integration of children

with Asperger syndrome. Autism, 4(1), 85-100.

Bayliss, P. (1995). Integration and interpersonal relations: interactions between

disabled children and their non-disabled peers. British Journal of Special

Education, 22(3), 131-139.

Bokhorst, K., Goossens, F.A. & De Ruyter, P.A. (2001). Early detection of social

anxiety: reliability and validity of a teacher questionnaire for the

identification of social anxiety in young children. Social Behavior and

Personality, 29(8), 787-798.

Booth, T. (2004). Reflection. In P. Clough & J. Corbett (Eds.), Theories of

Inclusive Education (pp. 59-64). London: Sage.

Booth, T. & Ainscow, M. (2002). Index for inclusion. Developing learning and

participation in schools. Bristol: Centre for Studies on Inclusive Education.

Boutot, E.A. & Bryant, D.P. (2005). Social integration of students with autism in

inclusive settings. Education and Training in Developmental Disabilities,

40(1), 14-23.

CHAPTER 3

76

Brown, W.H., Odom, S.L., Li, S. & Zercher, C. (1999). Ecobehavioral assessment

in early childhood programs: a portrait of preschool inclusion. The Journal

of Special Education, 33(3), 138-153.

Cambra, C. & Silvestre, N. (2003). Students with special needs in the inclusive

classroom: social integration and self-concept. European Journal of Special

Needs Education, 18(2), 197-208.

Cartledge, G. & Talbert Johnson, C. (1996). Inclusive classrooms for students

with emotional and behavioral disorders: critical variables. Theory into

Practice, 35(1), 51-57.

Cava, M.J. & Musitu, G. (2000). Evaluation of an intervention programme for the

empowerment of self-esteem. Psychology in Spain, 4(1), 55-63.

Coben, S.S. & Zigmond, N. (1986). The social integration of learning disabled

students from self-contained to mainstream elementary school settings.

Journal of Learning Disabilities, 19(10), 614-618.

Cole, D.A. & Meyer, L.H. (1991). Social integration and severe disabilities: a

longitudinal analysis of child outcomes. The Journal of Special Education,

25(3), 340-351.

Conderman, G. (1995). Social status of sixth- and seventh-grade students with

learning disabilities. Learning Disability Quarterly, 18, 13-24.

Cook, B.G. & Semmel, M.I. (1999). Peer acceptance of included students with

disabilities as a function of severity of disability and classroom

composition. The Journal of Special Education, 33(1), 50-61.



351.

Davis, S., Howell, P. & Cooke, F. (2002). Sociodynamic relationships between

children who stutter and their non-stuttering classmates. Journal of Child

Psychology and Psychiatry, 43(7), 939-947.




Doll, B., Murphy, P. & Song, S.Y. (2003). The relationship between children’s

self-reported recess problems and peer acceptance and friendships.

Journal of School Psychology, 41, 113-130.


77

Dyson, L.L. (2005). Kindergarten children’s understanding of and attitudes

toward people with disabilities. Topics in Early Childhood Special

Education, 25(2), 95-105.

Farmer, T.W. & Farmer, E.M.Z. (1996). Social relationships of students with

exceptionalities in mainstream classrooms: social networks and

homophily. Exceptional Children, 62(5), 431-450.



Flem, A. & Keller, C. (2000). Inclusion in Norway: a study of ideology in practice.

European Journal of Special Needs Education, 15(2), 188-205.

Frederickson, N.L. & Furnham, A.F. (2004). Peer-assessed behavioural

characteristics and sociometric rejection: differences between pupils who

have moderate learning difficulties and their mainstream peers. British

Journal of Educational Psychology, 74, 391-410.

Frederickson, N.L. & Furnham, A.F. (1998). Use of sociometric techniques to

assess the social status of mainstreamed children with learning difficulties.

Genetic, Social, and General Psychology Monographs, 124(4), 381-433.

Frederickson, N. & Turner, J. (2003). Utilizing the classroom peer group to

address children’s social needs: an evaluation of the circle of friends

intervention approach. Journal of Special Education, 36(4), 234-245.

Freeman, S.F.N. & Alkin, M. (2000). Academic and social attainments of children

with mental retardation in general education and special education

settings. Remedial and Special Education, 21(1), 3-18.

Gable, R.K., Rucker, C.N. & Smith, E.V. jr. (1997). Assessing student perceptions

of affective outcomes of special education programs: instrument

development, validation, and comparisons to regular education students.

Educational and Psychological Measurement, 57(4), 685-697.

Goldstein, H. & English, K. (1995). Use of peers as change agents in

communicative interactions with preschoolers with disabilities. Preventing

School Failure, 39(4), 16-21.

Goodman, H., Gottlieb, J. & Harrison, R.H. (1972). Social acceptance of EMRS

integrated into a non-graded elementary school. American Journal of


CHAPTER 3

78

Guralnick, M.J. (1999). The nature and meaning of social integration for young

children with mild developmental delays in inclusive settings. Journal of

Early Intervention, 22(1), 70-86.

Guralnick, M.J., Connor, R.T., Hammond, M., Gottman, J.M. & Kinnish, K. (1995).

Immediate effects of mainstreamed settings on the social interactions and

social integration of preschool children. American Journal on Mental

Retardation, 100(4), 359-377



Education, 34(3), 114-126.

Harper, G.F., Maheady, L., Mallette, B. & Karnes, M. (1999). Peer tutoring and

the minority child with disabilities. Preventing School Failure, 43(2), 47-

52.

Harper, L.V. & McCluskey, K.S. (2002). Caregiver and peer responses to children

with language and motor disabilities in inclusive preschool programs. Early

Childhood Research Quarterly, 17, 148-166.

Hegarty, S. (1991). Towards an agenda for research in special education.

European Journal of Special Needs Education, 6, 87-99.

Hepler, J.B. (1998). Social integration of children with emotional disabilities and

nonhandicapped peers in a school setting. Early Child Development and

Care, 147, 99-115.

Hunt, P., Alwell, M., Farron-Davis, F. & Goetz, L. (1996). Creating socially

supportive environments for fully included students who experience

multiple disabilities. Journal of the Association for Persons with Severe

Handicaps, 21(2), 53-71.

Hunt, P., Soto, G., Maier, J. & Doering, K. (2003). Collaborative teaming to

support students at risk and students with severe disabilities in general

education classrooms. Exceptional Children, 69(3), 315-332.

Hunt, P., Soto, G., Maier, J., Müller, E. & Goetz, L. (2002). Collaborative teaming

to support students with augmentive and alternative communication needs

in general education classrooms. Augmentive and Alternative

Communication, 18, 20-35.

Ialongo, N., Edelsohn, G., Werthamer-Larsson, L., Crocket, L. & Kellam, S.

(1996). Social and cognitive impairment in first-grade children with


79

anxious and depressive symptoms. Journal of Clinical Child Psychology,

25(1), 15-24.

International Classification of Functioning, Disability and Health (2001). Geneva:

World Health Organization, WHO.



Psychology, 84(3), 322-330.

Kamps, D.M., Dugan, E., Potucek, J. & Collins, A. (1999). Effects of cross-age

peer tutoring networks among students with autism and general education

students. Journal of Behavioral Education, 9(2), 97-115.

Kanioglou, A., Tsorbatzoudis, H. & Barkoukis, V. (2005). Socialization and

behavioral problems of elementary school pupils with developmental

coordination disorder. Perceptual and Motor Skills, 101, 163-173.

Kemp, C. & Carter, M. (2002). The social skills and social status of mainstreamed

students with intellectual disabilities. Educational Psychology, 22(4), 391-

411.

Kennedy, C.H., Cushing, L.S. & Itkonen, T. (1997). General education

participation improves the social contacts and friendship networks of

students with severe disabilities. Journal of Behavioral Education, 7(2),

167-189.

Kennedy, C.H., Shukla, S. & Fryxell, D. (1997). Comparing the effects of

educational placement on the social relationships of intermediate school

students with severe disabilities. Exceptional Children, 64(1), 31-47.

Kobi, E.E. (1983). Praktizierte Integration: Eine Zwischenbilanz.

Vierteljahresschrift für Heilpädagogik und ihre Nachbargebiete, 52(2),

196-216.

Koster, M., Pijl, S.J., Van Houten, E.J. & Nakken, H. (2007). The social position

and development of pupils with SEN in mainstream Dutch primary schools.


Kuhne, M. & Wiener, J. (2000). Stability of social status of children with and

without learning disabilities. Learning Disability Quarterly, 23(1), 64-75.



Education, 25(1), 90-101.

CHAPTER 3

80

Luetke-Stahlman, B. (1995). Social interaction: Assessment and intervention

with regard to students who are deaf. American Annals of the Deaf,

140(3), 295-303.

Manetti, M., Schneider, B.H. & Siperstein, G. (2001). Social acceptance of

children with mental retardation: testing the contact hypothesis with an

Italian sample. International Journal of Behavioral Development, 25(3),

279-286.

Mar, H.H. & Sall, N. (1995). Enhancing social opportunities and relationships of

children who are deaf-blind. Journal of Visual Impairment & Blindness,

89(3), 280-287.

Margalit, M. & Efrati, M. (1996). Loneliness, coherence and companionship

among children with learning disorders. Educational Psychology, 16(1),

69-79.




Nietupski, J.A. (1995). The evolution of the lre concept for students with severe

disabilities. Preventing School Failure, 39(3), 40-47.

Nikolaraizi, M. & De Reybekiel, N. (2001). A comparative study of children’s

attitudes towards deaf children, children in wheelchairs and blind children

in Greece and in the UK. European Journal of Special Needs Education,

16(2), 167-182.

Odom, S.L. (2000). Preschool inclusion: what we know and where we go from

here. Topics in Early Childhood Special Education, 20(1), 20-27.

OECD (1995). Integrating pupils with special needs into mainstream schools.

Paris: OECD.

Pavri, S. & Luftig, R. (2000). The social face of inclusive education: are students

with learning disabilities really included in the classroom? Preventing

School Failure, 45(1), 8-14.

Pearl, R., Farmer, T.W., Van Acker, R., Rodkin, P.C., Bost, K.K., Coe, M. &

Henley, W. (1998). The social integration of students with mild disabilities

in general education classrooms: peer group membership and peer-

assessed social behaviour. The Elementary School Journal, 99(2), 167-

185.


81

Pijl, S.J. & Hamstra, D. (2005). Assessing pupil development and education in an

inclusive setting. International Journal of Inclusive Education, 9(2), 181-

192.

Power, D. & Hyde, M. (2002). The characteristics and extent of participation of

deaf and hard-of-hearing students in regular classes in Australian schools.

Journal of Deaf Studies and Deaf Education, 7(4), 302-311.

Ring, E. & Travers, J. (2005). Barriers to inclusion: a case study of a pupil with

severe learning difficulties in Ireland. European Journal of Special Needs

Education, 20(1), 41-56.

Sabornie, E.J., Marshall, K.J. & Ellis, E.S. (1990). Restructuring of mainstream

sociometry with learning disabled and nonhandicapped students.


Savage, R. (2005). Friendship and bullying patterns in children attending a

language base in a mainstream school, Educational Psychology in Practice,

21(1), 23-36.




Schmidt, M. (2000). Social integration of students with learning disabilities.

Developmental Disabilities Bulletin, 28(2), 19-26.

Scholtes, V., Vermeer, A. & Meek, G. (2002). Measuring perceived competence

and social acceptance in children with cerebral palsy. European Journal of


Siperstein, G.N. & Leffert, J.S. (1997). Comparison of socially accepted and

rejected children with mental retardation. American Journal on Mental

Retardation, 101(4), 339-351.

Skårbrevik, K.J. (2005). The quality of special education for students with special

needs in ordinary classes. European Journal of Special Needs Education,

20(4), 387-401.




Stanovich, P.J., Jordan, A. & Perot, J. (1998). Relative differences in academic

self-concept and peer acceptance among students in inclusive classrooms.

Remedial & Special Education, 19(2), 120-126.

CHAPTER 3

82

Stinson, M.S. & Antia, S.D. (1999). Considerations in educating deaf and hard-

of-hearing students in inclusive settings. Journal of Deaf Studies and Deaf

Education, 4(2), 163-175.

Stone, W.L. & La Greca, A.M. (1990). The social status of children with learning

disabilities: a re-examination. Journal of Learning Disabilities, 23, 32-37.

Storey, K. & Smith, D.J. (1995). Assessing integration in early childhood

education: clique analysis of social interactions. Education & Treatment of

Children, 18(2), 158-184.


mental health: What, how, who and when. In P.S. Strain, M.J. Guralnick

& H. Walker (Eds.), Children’s Social Behavior: Development, assessment

and modifications (pp. 287-330). New York: Academic Press.

Vaughn, S., Elbaum, B.E., Schumm, J.S. & Hughes, M.T. (1998). Social

outcomes for students with and without learning disabilities in inclusive

classrooms. Journal of Learning Disabilities, 31(5), 428-436.

Wiener, J. & Harris, P.J. (1989). Integration, rejection and neglect in learning-

disabled children. In J.J. Dumont, & H. Nakken, (Eds.), Cognitive, Social

and Remedial Aspects (pp. 109-115). Amsterdam/Lisse: Swets &

Zeitlinger.

Williams, S.K., Johnson, C. & Sukhodolsky, D.G. (2005). The role of the school

psychologist in the inclusive education of school-age children with autism

spectrum disorders. Journal of School Psychology, 43, 117-136.

Yude, C., Goodman, R. & McConachie, H. (1998). Peer problems of children with

hemiplegia in mainstream primary schools. Journal of Child Psychology

and Psychiatry, 39(4), 533-541.

Chapter 4 Assessing social participation of pupils with special needs in inclusive education: the construction of a teacher questionnaire1

CHAPTER 4

ASSESSING SOCIAL PARTICIPATION OF PUPILS WITH SPECIAL NEEDS IN INCLUSIVE

EDUCATION: THE CONSTRUCTION OF A TEACHER QUESTIONNAIRE

This article is a slightly adapted version of: Koster, M., Nakken, H., Pijl, S.J., Van Houten, E.J. & Lutje Spelberg, H.C. (in press). Assessing social participation of pupils with special needs in inclusive education: the construction of a teacher questionnaire. Educational Research and Evaluation.

CHAPTER 4

84

Abstract

Increasing the social participation of pupils with special needs is an important

goal of inclusive education. However, making friends and building positive

relations with classmates can be difficult for these pupils. They often have a

lower social position in the classroom than their typical peers. Research has

shown that teachers tend to overestimate their social participation. This study

addresses the development of a questionnaire aimed at helping teachers assess

the social participation of pupils with special needs and is divided into two parts.

In the first part of the study the construction of the teacher questionnaire is

described. The questionnaire consists of 30 statements related to four key

themes of social participation: ‘friendships/relationships’, ‘contacts/interactions’,

‘social self-perception of the pupil’ and ‘acceptance by classmates’, yielding four

subscales.

The second part of the study addresses an empirical study in which the quality of

the questionnaire was examined. A confirmatory factor analysis was conducted in

order to gather empirical evidence about our social participation model. The

outcomes did largely support the division of social participation into four key

themes. The analysis revealed that 22 out of 30 statements correlated strongest

with the subscale they were assigned to.

It turned out that the reliability of the questionnaire was high, whereas the

reliability of the four subscales varied. The construct validity was found to be

mainly acceptable. It is concluded that the questionnaire may become a valuable

tool for teachers to assess the social participation of pupils with special needs.

ASSESSING SOCIAL PARTICIPATION: THE CONSTRUCTION OF A TEACHER QUESTIONNAIRE

85

4.1 Introduction

The inclusion of pupils with special needs into regular education is an important

education policy in many countries. Research has shown that increased social

opportunities are parents’ primary motive for their child’s regular school

placement (Sloper & Tyler, 1992; Strayhorn & Strain, 1986). However, in some

cases positive contacts and friendships between pupils with special needs and

their peers without special needs do not occur spontaneously (Guralnick, Connor,

Hammond, Gottman & Kinnish, 1996; Guralnick, Hammond, Connor & Neville,

2006; Guralnick, Neville, Hammond & Connor, 2007; Lee, Yoo & Bak, 2003;

Scheepstra, Nakken & Pijl, 1999). A study by Frostad and Pijl (2007) of

Norwegian inclusive classrooms suggests that between 20 and 25 per cent of

pupils with special needs are not socially included in their peer group, while for

their typical peers this is only 8 per cent.

A review study by Nakken and Pijl (2002) shows that an evaluation of the effects

of mainstreaming on social contacts between these pupils and their typical peers

provides contradictory conclusions. Some studies report no effects, others report

positive effects (like increase of social contacts and friendships), while a few

report negative effects (like increased bullying and negative interaction). A

problem with evaluating the social aspect of inclusion is the ambiguity of the

concepts researchers use. As described by Koster, Nakken, Pijl and Van Houten

(in press), different concepts are adopted to describe the social dimension of

inclusion. Three umbrella concepts, namely ‘social participation’, ‘social

integration’ and ‘social inclusion’, are used frequently by researchers. However,

there is uncertainty about their meaning. Koster et al. (in press) made a review

of literature aimed at elucidating these concepts and revealing characteristic

themes. The literature used in the review was derived from two sources: the last

six volumes (2000 – 2005) of three prominent journals in the field of special

needs education (‘European Journal of Special Needs Education’, ‘International

Journal of Inclusive Education’ and ‘Scandinavian Journal of Educational

Research’) and a search of two electronic databases (ERIC and PsycINFO) from

January 1995 until December 2005. The articles had to be aimed at elementary

school pupils or preschoolers, while their focus had to be on the social aspect of

inclusion in the school situation (rather than in the wider community). In

reviewing the literature the focus was on the following three themes:

CHAPTER 4

86

• definition of the concepts used;

• relation between the various concepts used;

• any subdivision in themes within the concepts used.

The review, consisting of 62 articles in its final form, showed that the concepts of

‘social participation’, ‘social integration’ and ‘social inclusion’ are often described

loosely, with only a few researchers providing explicit definitions or descriptions.

In the majority of articles, implicit descriptions could be derived from

instruments used to measure the aforementioned concepts. In order to elucidate

the concepts, in each article it was tried to reveal which topics were considered

vital with regard to the social dimension of inclusion. It turned out that various

aspects are used by researchers to describe the same phenomena. Many also

use the three umbrella concepts synonymously. The elaborations of these

concepts largely overlap and in research practice hardly differ from each other.

Koster et al. (in press) recommend using the concept ‘social participation’.

Despite the large number of different aspects, from the review it could be

deduced that researchers agree on the importance of several core topics within

social participation. The review of literature resulted in a provisional description

of social participation and identified four key themes central to this concept:

‘friendships/relationships’, ‘contacts/interactions’, ‘pupil’s social self-perception’

and ‘acceptance by classmates’. These key themes should be seen as a

framework for constructing an instrument to measure social participation. Each

key theme is split into two or more sub-themes, which are also derived from the

review of literature (see Figure 1). The sub-themes are further specifications that

serve as a guide for translating each key theme into operational concepts.


87

Friendships/relationships


Friendship network

Mutual friendship

Playing together

Working together on

tasks

Participation in group

activities

(Un) acknowledged initiations

Social isolation

Social self-perception of pupil

Self-perception of peer

acceptance

Satisfaction at school

Social self-concept

Self-perception of social

competence

Loneliness


Social preference

Social support (behaviours)

Bullying

Social rejection

Social inclusionSocial

integrationSocial

participation

Figure 1. Overview of key themes and their sub-themes within social participation (and

the related concepts of social inclusion and social integration) (Koster et al., in press)

The model shows which topics are linked to the concept of social participation. In

order to measure this it has to be operationalised, so after constructing the

model, the sub-themes were elaborated upon in 38 more concrete and

measurable statements (for example: ‘After school the pupil with special needs

plays regularly with classmates’). These can be seen as the operational definition

of the four key themes. This process of formulating statements was intended for

a questionnaire addressing social participation.

Such a questionnaire is important, since the reliability and validity of many

instruments used to measure aspects of social participation have not been

CHAPTER 4

88

proven. For instance, according to Frederickson and Furnham (1998), available

data suggest that the test-retest reliability of the often used sociometric status

group classifications is questionable. This is in line with Bukowksi and Hoza

(1989) who conclude that the stability of many sociometric methods are not

overwhelmingly impressive. Furthermore, Terry and Coie (1991) demonstrate

that different methods to assess pupils’ sociometric status yield diverging results.

Moreover, an instrument which encompasses all aspects of social participation

does not exist: most instruments only measure one aspect of social participation.

Therefore, there is a need for one teacher friendly assessment instrument that

encompasses all important aspects of social participation and which is reliable

and valid. Such an instrument might help teachers monitoring the social

participation of pupils with special needs. This is of importance, all the more

because several studies (De Monchy, Pijl & Zandberg, 2004; Scheepstra et al.,

1999; Whitney, Smith & Thompson, 1994) have shown that teachers’

assessments of the social participation of pupils with special needs in their

classroom are not always appropriate. Teachers tend to overestimate the social

participation of these pupils and to overlook problems like bullying. An

assessment instrument which encompasses the total concept of social

participation might help teachers noticing problems in time. Subsequently,

teachers may be prompted to foster the social participation of pupils with special

needs, for example by encouraging contacts and relationships between these

pupils and their typical classmates.

In this study, the process of operationalising social participation and the

subsequent development of a teacher questionnaire is central. The study is

divided into two parts. The first focuses on constructing a teacher questionnaire

and answers the question of how social participation may be elaborated into a

questionnaire. The second, empirical, part of the study addresses whether such a

questionnaire is a potentially reliable and applicable instrument to assess the

social participation of pupils with special needs in regular primary schools.


89

4.2 Construction of the social participation questionnaire

4.2.1 Method

Two steps were taken to construct a social participation questionnaire: a panel

was consulted in order to select appropriate statements for this and then a final

selection of statements was made with the assistance of a large group of

respondents.

4.2.2 Selection of statements through consultation of a panel

Participants. The panel consisted of 14 people with much practical experience of

inclusive education. These included parents of a child with a disability attending

regular education, university students with disabilities who had attended a

regular primary school, regular primary school teachers teaching pupils with

special needs and peripatetic teachers.

Procedure. In the introduction section it was described how the four key themes

of social participation were translated into 38 statements. The topics addressed

in the statements are mainly based on literature and partly on the clinical and

research experience of the researchers who work for many years in the tradition

of research looking at the social dimension of inclusion in education. Since this

process might have been coloured by the researchers’ views (e.g. certain sub-

themes might have been overrepresented, while others might have been

undervalued), a panel consisting of 14 people with much practical experience in

inclusive education was asked to critically examine the statements. However,

before the list of statements was presented, the panel brainstormed in small

groups about topics they considered to be important for social participation. They

discussed social participation open-mindedly, without any influence of the results

of the literature review. Only after this brainstorming session was the list of

statements presented to the panel. The members were then asked if the

statements formulated by the researchers were in keeping with their views of

social participation and were invited to mention omissions from the list. Based on

the panel’s feedback, the researchers compiled a final list of statements giving a

complete picture of social participation. This detailed verification of the

statements was intended to increase the chance of including all essential topics

of social participation.

CHAPTER 4

90

Results. The panel agreed with the 38 statements formulated by the researchers,

but they also named supplementary topics which were then incorporated into an

additional 36 new statements. Of these, 22 were indicators of social participation

(for example: ‘The pupil has the feeling s/he is a full member of the group’) and

14 were indicators of social segregation (for example: ‘Classmates often laugh at

the pupil). All new statements were further specifications of the four key themes.

The majority of statements named by the panel were associated with contacts

and interactions between pupils with special needs and their classmates (for

example: ‘The pupil has fun with classmates’). In addition, many statements

were connected to friendships and relationships (for example: ‘Classmates invite

the pupil to play with them during school holidays’). Only a few statements

focused on the pupil’s perception (for example: ‘The pupil has the feeling s/he is

being teased by classmates’) and on the acceptance by classmates (for example:

‘Classmates step into the breach for the pupil when other children treat him/her

in an unpleasant manner’). However, according to the panel the topics addressed

in these statements are considered to be significant for successful social

participation, especially those related to the pupil’s feeling of belonging and

classmates viewing him or her as a full member of the group. A noticeable

finding from the panel session was the distinction between categories of

disabilities. The panel underlined the significance of distinguishing different

categories of disabilities when assessing which statements are important for

social participation. For instance, statements which address social participation of

pupils diagnosed as having behavioural problems might be less relevant in

describing the social participation of pupils with severe intellectual disabilities.

4.2.3 Final selection of statements through assessment by a group of

respondents

Participants. The 36 new statements were added to the original 38, thus making

a final list of 74 statements. Since a questionnaire consisting of 74 statements

would be unsuitable in educational practice, it was necessary to shorten the list.

We could have invited teachers, who are the questionnaire’s intended users, to

assess this large amount of statements for their pupils, in order to reveal the

statements’ psychometric characteristics. On the basis of these characteristics,


91

the list of statements could have been shortened. However, in stead of directly

presenting the list of statements to teachers, it was more appropriate to take an

extra step by asking a large group of respondents to assess the importance of

each statement. This extra step was considered necessary, as two non-

interchangeable themes are involved here, namely importance and clinical

applicability of the statements. We choose to make a first selection of statements

on the basis of importance (by asking the group of respondents to assess the

importance of each statement) and a second selection on the basis of clinical

applicability (by asking teachers to assess the applicability of the statements for

their pupils, which is the subject of the second part of this article).

The respondents, like the panel, had practical experience of inclusion and

consisted of parents, primary school teachers, students with disabilities who

study at university or attend a secondary school and peripatetic teachers but a

fifth group, consisting of educational advisors, was also added.

Procedure. The respondents were invited to look critically at our theoretical quest

to operationalise social participation. The group was drawn up by means of a

snowball sample. For each type of respondent a suitable sample procedure was

used. Parents were approached by writing letters to parent associations or

contacting their websites. Peripatetic teachers were selected by contacting their

coordinators at Regional Expertise Centres in the Netherlands. Parents and

peripatetic teachers then put us in contact with teachers who might be interested

in taking part in the research, while university students with disabilities were

approached via disability study groups at Dutch universities. The final group of

respondents, the educational advisors, were approached by emailing school

advisory services. Finally, at two conferences on inclusive education we displayed

posters informing participants about the study and inviting them to take part. As

a result of these varying selection procedures, a national distribution of

respondents was achieved.

These respondents were then mailed a questionnaire and asked to complete the

questionnaire for the category of disability with which they were familiar, i.e.

speech/language, motor, intellectual, behavioural or autistic problems. It was

decided to distinguish these various categories of disabilities, as a broad range of

pupils attend regular education and the questionnaire should be suitable for all of

them.

CHAPTER 4

92

All 273 respondents were asked to assess the degree to which the statements

represented social participation or social segregation. Reactions to each

statement could be given on a seven-point scale, ranging from ‘this is a very

important signal of social participation of the pupil’ (1) to ‘this is a very important

signal of social segregation of the pupil’ (7). In this way the most important

(positive or negative) statements for signalling social participation could be

determined.

Results. From the 273 intended respondents, 190 (70 per cent) completed the

questionnaire – a satisfactory percentage. Of the 190 respondents, more than

half were peripatetic teachers (n=106), while the other 84 respondents

comprised 56 parents, 16 teachers, 7 educational advisors and 5 students

(secondary education and university). A quarter of the respondents (n=48)

assessed statements for pupils with an autistic disorder; 45 respondents for

those with an intellectual disability, another 45 for those with a motor disability,

35 for those with speech/language disabilities and the final 17 assessed the

statements for pupils with behavioural disorders. Table 1 provides an overview of

the five groups of respondents and the type of disabilities they assessed.

Table 1. Overview of respondents and their assessments of categories of disabilities

(n=190)

Respondent� Category of disabilities � P

eripatetic

teacher

Parent

Teacher

Educational

advisor

Student

Total


21 22 4 1 0 48


13 2 2 0 0 17

Speech/language disability

19 11 3 1 1 35


28 9 4 4 0 45

Motor disability 25 12 3 1 4 45

Total 106 56 16 7 5 190


93

All respondents assessed the degree to which the 74 statements were important

signals of social participation. Contrary to the expectations of the panel, there

were no significant differences in the assessment of 71 of the 74 statements for

the different categories of disabilities (for 71 statements F varies between 0.22

and 3.67, p>0.05). For three statements the differences were significant (F

varies between 4.29 and 4.78, p<0.05). However, this can be attributed to

chance.

The 30 statements considered as important or very important by at least 75 per

cent of the respondents were selected for the Social Participation Questionnaire.

In total, six statements belonged to the ‘Friendships/relationships’ subscale,

fifteen belonged to the ‘Contacts/interactions’ subscale, four belonged to the

‘Pupil’s social self-perception’ subscale and five belonged to the ‘Acceptance by

classmates’ subscale. The statements were phrased in terms of social

participation or in terms of social segregation. Both the key theme

‘Friendships/relationships’ and the key theme ‘Acceptance by classmates’ are

only represented by statements phrased in terms of social participation. Table 2

gives some examples of the statements. Of the 30 final statements, 17 stem

from the original list of 38 statements developed by the researchers, while the

13 others resulted from the brainstorming session with the panel.

Table 2. Examples of statements categorised under the four key themes of social

participation

Key theme Number of

statements

Statement of social

participation

Statement of social

segregation

Friendships/

relationships

6 ‘the pupil is being invited to

birthday parties’

Contacts/interactions 15 ‘the pupil regularly has fun

with his/her classmates’

‘classmates regularly

exclude the pupil from

activities’

Pupil’s social self-

perception

4 ‘the pupil has the feeling

s/he belongs to the group’

‘the pupil feels lonely at

school’

Acceptance by

classmates

5 ‘classmates consider the

pupil as a full member of

the group’

CHAPTER 4

94

4.3 Assessing the quality of the questionnaire

4.3.1 Method

A confirmatory factor analysis was conducted in order to gather a first impression

of empirical evidence for the Questionnaire’s division into four subscales

representing the four key themes of social participation. The quality of the

Questionnaire was further addressed by examining its reliability, the item-total

correlations and its construct validity.

Participants. In order to select school classes for the empirical study, help was

sought from the respondents involved in constructing the questionnaire by

inviting them to put us in touch with schools educating one or more pupils with

special needs in Grades 1 to 3. In total 17 classes in 13 schools were involved in

the study. In five classes there were two or more pupils with special needs, i.e.

three classes had two pupils, one class had three and another four pupils. In

total 25 pupils with special needs participating in full-time regular primary

education took part in the empirical study. All 25 pupils had the Dutch equivalent

of a statement of special needs and were in Grades 1 to 3. The sample consisted

of pupils with the types of disabilities most common in regular Dutch primary

schools, who have the fastest growth in attending regular Dutch schools and

whose special needs are known to be linked to difficulties in social relationships

and contacts in the regular classroom. Nine pupils were assessed as having

(severe) behavioural disorders (including ADHD), three as having an autistic

spectrum disorder (PDD-NOS), three as having ADHD combined with PDD-NOS

symptoms, five as having a motor disability, two as having an intellectual

disability, two as having speech/language disabilities and one as a hard of

hearing and partially sighted pupil. Not only the pupils with special needs but

also their classmates (n=352) and the teachers (n=17) participated in the study.

The average age of the pupils with special needs was 8.4 years (SD=1.2), while

the average age of their typical peers was 8.0 years (SD=1.3). This small

difference in age was found to be non significant, t(54)=1.0, p=0.32.

The teachers of the participating classes completed the Social Participation

Questionnaire for the pupil(s) with special needs and for two other pupils without

special needs, the latter randomly chosen by using the numbers three and six on

the class register. If the pupil with special needs was number three or six on the


95

register than the following pupil on the list was chosen, while if pupils number

three and six were of the same sex, the next pupil of the opposite sex on the

register was chosen. In classrooms with two or more pupils with special needs,

the teacher was asked to complete the questionnaire for all pupils with special

needs and for two typical pupils. This meant the Social Participation

Questionnaire would be filled in for the 25 pupils with special needs and for 34

typical pupils (two pupils in each of the 17 classes). However, in one of the

classes, the teacher filled in the Questionnaire for three instead of two typically

developing pupils, so this brought the Questionnaire’s total sample to 25 pupils

with special needs (18 boys, 7 girls) and 35 typically developing pupils (17 boys,

18 girls).

Instruments. The teachers completed the Social Participation Questionnaire for

the pupil(s) with special needs and for two other pupils by answering each

question on a five-point scale, ranging from ‘this strongly applies to the pupil’ (1)

to ‘this does not apply at all to the pupil’ (5). The minimum score for the

questionnaire was 30 (30 x 1), while the maximum was 150 (30 x 5). The lower

the score, the more positive the social participation of the pupil was assessed.

In order to get a first impression of the construct validity of the Social

Participation Questionnaire, other procedures to measure social participation

were needed. However, there is no instrument which, like the Social Participation

Questionnaire, focuses on the total concept of social participation. Therefore, in

order to examine the Questionnaire’s construct validity, four instruments, each

focusing on one of the key themes of social participation, should be applied.

However, in this pilot study, gathering data using in total five instruments in

each classroom would be unfeasible. When administering this number of

instruments, the researchers would have to spend a long time in classes, which

would probably disturb the teachers’ schedule too much. Moreover, it would be

too large a burden for pupils and teachers. It was decided to assess both the key

themes ‘friendships/relationships’ and ‘acceptance by classmates’, as these

themes can be assessed by the same instrument. The sociometric nomination

method was chosen for this purpose. All pupils in the 17 classes completed a

sociometric questionnaire. Based on Coie, Dodge and Coppotelli (1982), this

asked pupils which three classmates they liked to play with best.

CHAPTER 4

96

Analysis. The Questionnaire’s division in four subscales was entirely based on a

literature review and the opinion of respondents with much experience of

inclusive education. The Oblique Multiple Group Method (Holzinger, 1944;

Nunnally, 1978) was used in order to verify the assignment of statements to

subscales. The correlation of each statement to each subscale was computed,

while correcting correlations between statements and subscales in which the

statements take part. If each statement has the highest correlation with the

subscale to which it is assigned, the proposed structure of the scales fits the

data. If a statement correlates higher with the subscales to which is its not

assigned, this indicates that such a statement is assigned wrongly (Stuive,

2007).

The reliability of the Questionnaire was assessed by calculating Cronbach’s

Alpha. The construct validity of the Questionnaire was assessed in two ways.

First, the scores of pupils with special needs on the Questionnaire were

compared with those of their typical classmates. Based on literature (e.g. De

Monchy et al., 2004; Pijl, Frostad & Flem, in press; Scheepstra et al. 1999) it

was expected that, compared to their classmates, the social participation of the

pupils with special needs would be assessed less positively. Second, the

outcomes were compared with those of another instrument commonly used to

assess social facets of inclusion, namely a sociometric questionnaire (see

Instruments). Pupils were asked to nominate three pupils with whom they liked

to play best. Based on their answers, an ‘indegree’ index could be calculated for

every pupil. Indegrees are measures of receptivity or popularity (Wasserman &

Faust, 1994), thus indicating how often a pupil is chosen by a classmate, and can

be considered as an index of acceptance. Next to the indegrees (as a measure of

acceptance), the number of reciprocal friendships of each pupil was calculated.

In sociometric analyses friends are defined as two pupils independently selecting

each other as peers they like or peers with whom they like to play (Cullinan,

Sabornie & Crossland, 1992; Juvonen & Bear, 1992).

The correlation between the score on the Questionnaire and the indegrees was

then calculated. In addition, a calculation was made of the correlation between

the score on the Questionnaire and the number of reciprocal friendships. It was

envisaged that the indegrees and reciprocal friendships would both correlate

negatively with the score on the Questionnaire, meaning that the more a pupil is

accepted by peers (having higher indegrees) and the higher the number of


97

his/her reciprocal friendships, the more positive the assessment of the pupil’s

social participation would be.

Finally, specific correlation patterns between the indegrees / the number of

reciprocal friendships and the four subscales were expected. It was supposed

that the indegrees would correlate stronger with the subscale ‘Acceptance by

classmates’ than with the other subscales, while pupils’ number of friendships

were expected to correlate stronger with the subscale ‘Friendships/relationships’

than with the other three subscales. Discovering this pattern was regarded as

another source of evidence for the Questionnaire’s construct validity.

4.3.2 Results

Factor analysis. The outcomes of the confirmatory factor analysis, which should

be interpreted cautiously because of the small sample (n=60), did largely

provide the expected empirical evidence for our theoretical model of social

participation. The outcomes of the analysis made clear that the division of the

Social Participation Questionnaire into four subscales, representing four key

themes of social participation, seems to be correct to a large extent. The analysis

revealed that 22 out of 30 statements correlated strongest with the subscale

they were assigned to. The remaining eight statements had higher correlations

with one or more subscales they were not part of (see Table 3). These first

preliminary findings are no basis to change the Social Participation

Questionnaire. Further analysis, with a larger sample, is needed to examine more

profoundly the Questionnaire’s division into distinct subscales.

Table 3. Outcomes Oblique Multiple Group Factor Analysis (n=60)

Subscale Number of statements

belonging to the

subscale

Number (and

numbers) of

exceeding statements

Friendships/relationships 6 0

Contacts/interactions 15 4 (7,10,13,24)

Pupil’s social self-perception 4 2 (12,27)

Acceptance by classmates 5 2 (1,20)

Reliability. The reliability of the Questionnaire was assessed by calculating

Cronbach’s Alpha. Alpha is 0.93, which implies a high reliability (Field, 2005).

CHAPTER 4

98

Also the reliabilities of the subscales ‘Friendships/relationships’ and

‘Contacts/interactions’ are high (respectively 0.94 and 0.88). However, the

subscales ‘Pupil’s social self-perception’ and ‘Acceptance by classmates’ show

reliabilities which are insufficient (respectively 0.69 and 0.64), assuming that

values of alpha around 0.70 are good for ability and related tests (Field, 2005).

The item-total correlations ranged from -0.03 to 0.86. One of the 30 statements

was incongruous and correlated negatively with the total score on the

Questionnaire (with an item-total correlation of -0.03). This implies that this

statement, aimed at pupils daring to tell the pupil with special needs that they

sometimes do not want to play with him/her, does not meet the criteria and

should be removed from the list. Alpha of the questionnaire as a whole remained

0.93 after this statement was removed, while alpha of the

‘Contacts/interactions’ subscale slightly improved (from 0.88 to 0.89). The item-

total correlations of the other 29 statements ranged from 0.12 to 0.86.

Validity. A comparison was made between the scores on the Questionnaire of

pupils with special needs and those of their classmates without special needs in

the 17 classes (no distinction was made between boys and girls, since the mean

score of boys and girls does not differ significantly, t(58)=0.49, p=0.63). On

average, pupils with special needs received a higher mean score (M=72.8,

SD=20.3) than their typical classmates (M=52.8, SD=13.4). This difference is

significant t(53)=-4.4, p<0.05 and represents a large effect size (ES=1.2)

(Cohen, 1992). The outcomes imply that, on average, the social participation of

pupils with special needs is assessed less positively than the social participation

of pupils without special needs. This is in accordance with our expectation that

pupils with special needs would experience more problems in social participation

than their typically developing classmates.

Secondly, correlations were calculated between the score of a pupil on the

Questionnaire and the indegrees s/he had as well as the number of reciprocal

friendships. It turned out that pupils with special needs have significantly lower

indegrees on average than their peers without special needs (2.1 versus 3.2,

t(58)=2.4, p<0.05), which means they are less accepted. The number of

reciprocal friendships is nearly the same for both groups of pupils (pupils with

special needs: 1.4, pupils without special needs: 1.5). The difference is not

significant, t(58)=0.6, p=0.52. The expected negative correlation between the


99

number of positive nominations and the total score on the questionnaire was

found: the higher the score on the questionnaire, the lower indegrees the pupil

had (-0.49, p<0.05). This means that the less positive the social participation is

assessed, the lower the number of times a pupil with special needs is chosen by

classmates to play with. The expected correlation pattern was not found.

Contrary to expectation, the correlation between the indegrees and the

‘Acceptance by classmates’ subscale (-0.24, p=0.07) was less strong than the

correlations between the indegrees and the other subscales (ranging from -0.35

to -0.55, p<0.05). This means the external measuring of this key theme

(sociometric indegrees) shows the weakest correlation with the statements on

the Questionnaire which focus on acceptance. Therefore, either the statements

on the Questionnaire are not representing acceptance properly, or the external

measuring is unsuitable to assess acceptance.

In addition, and as expected, the number of friendships correlated negatively

with the score on the Questionnaire. The higher the score on the Questionnaire,

the lower the number of reciprocal friendships and vice versa (-0.41, p<0.05),

which implies that the less positive the social participation is assessed, the lower

the number of friendships. Besides, the expected correlation pattern was found:

the number of reciprocal friendships showed a stronger correlation with the

‘Friendships/relationships’ subscale (-0.43, p<0.05) than with the other three

(correlations ranged from -0.28 to -0.34, p<0.05). That is, the external

measuring of friendships/relationships (the reciprocal friendships as assessed by

the sociometric questionnaire) correlates best with the statements on the

Questionnaire related to ‘friendships/relationships’. This implies that this key

theme is operationalised correctly in the Questionnaire.

4.4 Discussion

It is known from several studies (De Monchy et al., 2004; Scheepstra et al.,

1999; Whitney et al., 1994) that teachers’ assessments of the social participation

of pupils with special needs are not always appropriate. Pupils with special needs

are often less popular than their classmates, have fewer friends and are more

often neglected and rejected. However, teachers tend to overlook or

underestimate these problems. One possible explanation for this phenomenon is

related to the Cognitive Dissonance Theory of Festinger and Carlsmith (1959):

CHAPTER 4

100

teachers might be inclined to overlook problems and to overestimate the (social)

development of the pupil with special needs, because they made a well-

considered choice to educate the pupil. A negative social development of the

pupil would be incongruous with their hopes and expectations. The core of the

problem is that teachers’ assessments are often more positive than those of

fellow classmates (see also De Monchy et al., 2004). Since pupils’ assessment is

based on a much larger group of assessors who have no professional interest in

the social participation of pupils with special needs, it is justifiable to use this as

a reference point (De Monchy et al., 2004). However, it would be too limited to

only use sociometric data provided by pupils, since social participation is a much

broader concept. It is vital that teachers make accurate assessments of the total

social participation of pupils, because only when they notice that a pupil becomes

isolated, feels lonely or is being teased, can they take appropriate measures.

Taking measures to resolve or lessen problems with regard to the social situation

of a pupil is important, since negative social experiences at school (like isolation,

a low social status) might lead to externalising and internalising problems

(Durrant, Cunningham & Voelker, 1990) and to maladjustment in later life

(Nelson, Rubin & Fox, 2005; Parker & Asher, 1987; Terry & Coie, 1991).

Thus another way of assessing the social participation of pupils with special

needs and their interaction with fellow pupils is much needed. As mentioned

above, if teachers accurately estimate the social participation of pupils with

special needs, they are more likely to take proper action to stimulate

participation.

This study is aimed at contributing to the development of an instrument which

aims at helping teachers assess pupils’ social participation. At first, the concept

social participation was translated into concrete statements. Next, a

questionnaire was developed to measure the social participation of pupils with

special needs in regular primary schools. Since the Questionnaire consists of

concrete statements focusing on daily practice, this was expected to foster the

accuracy of the teachers’ judgements. In addition, by filling in the Questionnaire

for a pupil, the teacher is forced to think critically about the pupil’s situation in

the classroom, which might lead to renewed and refined insights.

The empirical data showed that the Questionnaire might become usable in

practice. A confirmatory factor analysis did largely provide the expected empirical

evidence for our theoretical model of social participation. The analysis revealed


101

that 22 out of 30 statements correlated strongest with the subscale they were

assigned to. In future research, including a large sample, the quality of the

separate statements for each subscale and the separability of the four subscales

should be assessed more thoroughly.

The reliability of the entire Questionnaire is high. However, the reliability of the

Questionnaire’s four subscales varied per scale. The subscales

‘Friendships/relationships’ and ‘Contacts/interactions’ showed a high reliability,

whereas the reliabilities of the subscales ‘Pupil’s social self-perception’ and

‘Acceptance by classmates’ were fairly low. In order to strengthen both

subscales, several statements will be added. One statement, belonging to the

‘Contacts/interactions’ subscale will be removed from the list.

If we focus on the outcomes of the sociometric questionnaire in relation to pupils’

scores on the Social Participation Questionnaire, the construct validity of the

questionnaire turns out to be acceptable. The expected negative correlation

between both the indegrees and the number of reciprocal friendships and the

score on the Questionnaire could be confirmed. In addition, the expected

correlation pattern between the number of reciprocal friendships and the

Questionnaire’s subscales was found: the number of friendships showed the

strongest correlation with the ‘Friendships/relationships’ subscale. However, for

the positive nominations (indegrees) the expected pattern was not

demonstrated: the indegrees correlated less strong with the ‘Acceptance by

classmates’ subscale than with the other subscales. It will be necessary to

further examine the construct ‘acceptance by classmates’ in order to improve the

validity of this subscale.

Finally, the expectation that pupils with special needs would experience more

problems in social participation than their typical classmates was confirmed. In

the current study, we did not differ between pupils with different categories of

disabilities. Within the group of 25 pupils with special needs, seven different

categories of disabilities were included. It was decided to include so many

categories as the Questionnaire was developed for all pupils. In order to examine

the applicability of the Questionnaire, it was desirable to involve pupils with

various categories of disabilities in the study. In a future study, it would be

advisable to involve larger homogeneous subgroups of pupils, that way it might

be possible to discover if correlations between the scores on the Social

CHAPTER 4

102

Participation Questionnaire and the scores on external instruments differ for

different subgroups.

Keeping these outcomes in mind, one may draw the conclusion that the Social

Participation Questionnaire may become a valuable tool for teachers to assess

the social participation of pupils with special needs. At the moment, only the first

step in the development of the Questionnaire has been taken. In follow-up

research, the qualities of the subscales and its statements could be examined. In

addition, further examination of the construct validity could take place by

comparing the Questionnaire with other external criteria (like observation scales,

sociometric rating scales or social self-perception questionnaires). A next step

should consist of examining the meaning of the scores. At the moment, we do

not know which scores on the Questionnaire are alarming. Future research

should aim at establishing a cut-off point, beyond which a score is cause for

concern and additional measures need to be taken.

Future research should also aim at the effectiveness of intervention methods.

Several methods have been developed to stimulate or improve the relations of

pupils with special needs and their classmates, some of which have been

reported successful. An often used method is social skills training. However, the

long-term effects of this kind of training are often disappointing. One possible

explanation is that classmates’ image of the pupils does not change and they do

not alter their attitude and behaviour. As a result the trained pupils do not have

many possibilities to practice their new skills: they will not see the benefit of

their new skills and after some time the acquired skills will fade away (Frostad &

Pijl, 2007). Therefore, measures to stimulate social participation of pupils with

special needs should not merely focus on those pupils but also on classmates.

Peer network approaches, in which both the pupil with special needs and typical

peers are involved, seem to be promising. Peer buddy programmes and peer

networks (for example the Circle of Friends approach) have demonstrated

increases in peer acceptance and in establishing friendships between pupils with

special needs and their typical peers (Campbell Miller, Cooke, Test & White,

2003; Frederickson & Turner, 2003). Further examination of the effects of peer

network approaches on the social participation of pupils with special needs is

needed.


103

4.5 References

Bukowski, W.M. & Hoza, B. (1989). Towards a further understanding of peer

relationships and their contributions to child development. In T.J. Berndt &

G.W. Ladd (Eds.), Peer relationships in child development (pp.15-45). New

York: John Wiley.

Campbell Miller, M., Cooke, M.L., Test, D.W. & White, R. (2003). Effects of

friendship circles on the social interactions of elementary age students

with mild disabilities. Journal of Behavioral Education, 12(3), 167-184.

Cohen, J. (1992). A power primer. Quantitative methods in psychology, 112(1),

155-159.

Coie, J.D., Dodge, K.A. & Coppotelli, H. (1982). Dimensions and types of social

status: A cross-age perspective. Developmental Psychology, 18(4), 557-

570.



351.




Durrant, J.E., Cunningham, C.E. & Voelker, S. (1990). Academic, social and

general self-concepts of behavioural subgroups of learning disabled

children. Journal of Educational Psychology, 82(4), 657-663.

Festinger, L. & Carlsmith, J.M. (1959). Cognitive consequences of forced

compliance. Journal of Abnormal and Social Psychology, 58, 203-210.

Field, A. (2005). Discovering Statistics Using SPSS (2nd ed.). London: SAGE

publications.

Frederickson, N.L. & Furnham, A.F. (1998). Sociometric classification methods in

school peer groups: a comparative investigation. Journal of Child

Psychology and Psychiatry, 39(6), 921-933.

Frederickson, N. & Turner, J. (2003). Utilizing the classroom peer group to

address children’s social needs: and evaluation of the Circle of Friends

Intervention Approach. The Journal of Special Education, 36(4), 234-245.



CHAPTER 4

104


Education, 22(1), 15-30.

Guralnick, M.J., Connor, R.T., Hammond, M.A., Gottman, J.M. & Kinnish, K.

(1996). The peer relations of preschool children with communication

disorders. Child Development, 67, 471-489.

Guralnick, M.J., Hammond, M.A., Connor, R.T. & Neville, B. (2006). Stability,

change, and correlates of the peer relationships of young children with

mild developmental delays. Child Development, 77(2), 312-324.

Guralnick, M.J., Neville, B., Hammond, M.A. & Connor, R.T. (2007). The

friendships of young children with developmental delays. A longitudinal

analysis. Journal of Applied Developmental Psychology, 28, 64-79.

Holzinger, K.J. (1944). A simple method of factor analysis. Psychometrika, 9(4),

257-262.



Psychology, 84(3), 322-330.

Koster, M., Nakken, H., Pijl, S.J. & Van Houten, E.J. (in press). Being part of the

peer group: a literature study focussing on the social dimension of

inclusion in education. International Journal of Inclusive Education.

Lee, S.H., Yoo, S.Y. & Bak, S.H. (2003). Characteristics of friendships between

children with and without disabilities. Education and Training in

Developmental Disabilities, 38(2), 157-166.







Nunnally, J.C. (1978). Psychometric Theory (2nd ed.). New York: McGraw-Hill.



Pijl, S.J., Frostad, P. & Flem, A. (in press). The social position of pupils with

special needs in regular schools. Scandinavian Journal of Educational

Research.


105









H. Walker (Eds.), Children’s Social Behavior: Development, Assessment

and Modifications (pp. 287-330). New York: Academic Press.

Stuive, I. (2007). A comparison of confirmatory factor analysis methods. Oblique

Multiple Group Method versus Confirmatory Common Factor Method.

Groningen: Facilitair Bedrijf GrafiMedia RuG.



Wasserman, S. & Faust, K. (1994). Social Network Analysis: Methods and

Applications. Cambridge: Cambridge University Press.

Whitney, I., Smith, P.K., & Thompson, D. (1994). Bullying and children with

special educational needs. In P.K. Smith & S. Sharp (Eds.), School

Bullying: insights and perspectives (pp. 213-240). New York: Routledge.

CHAPTER 4

106

Chapter 5 Examination of the psychometric qualities of the Social Participation Questionnaire1

CHAPTER 5

EXAMINATION OF THE PSYCHOMETRIC QUALITIES OF THE SOCIAL PARTICIPATION QUESTIONNAIRE

A slightly adapted and shortened version of this article has been re-submitted for publication and was co-authored by Marieke E. Timmerman, Han Nakken, Sip Jan Pijl and Els J. van Houten

CHAPTER 5

108

Abstract

Enhancing social participation of pupils with special needs is an important goal of

inclusive education. This study addresses the psychometric qualities of a new

teacher questionnaire to assess the social participation of pupils with special

needs in regular primary education. This Social Participation Questionnaire

initially consisted of 34 statements related to four key themes of social

participation: ‘friendships/relationships’, ‘contacts/interactions’, ‘pupil’s social

self-perception’ and ‘acceptance by classmates’, yielding four subscales.

A non-parametric item response analysis, a Mokken Scale Analysis, was used to

examine the quality of the Questionnaire. Based on the analysis results, ten

statements were removed. The resulting four subscales appeared intermediate to

strong. As Mokken’s double monotonicity model turned out to fit well for each

subscale, the subscale scores are on an ordinal scale and the separate

statements are invariantly ordered. The subscale scores are comparable across

pupils with and without special needs, because differential item functioning

appeared to be absent. Subsequent analyses supported the division of

statements into the four subscales.

The Questionnaire as a whole and its subscales were found to be reliable. Finally,

evidence has been obtained for the Questionnaire’s discriminant validity, as the

expected score differences between pupils with and without special needs were

clearly present.


109

5.1 Introduction

Peer relationships are important in the live of children. Its is widely

acknowledged that childrens’ peer relationships and friendships contribute to

their social and emotional development (Bagwell, 2004; Male, 2007) and also

peer acceptance and social status are seen as vital to the well-being of children.

Most primary school pupils find no difficulty in making friends and building

positive relations with their peers (Powless & Elliott, 1993). However, this does

not apply to pupils with special needs, as research has shown that these pupils

experience considerably more difficulty building friendships in inclusive

classrooms (Frostad & Pijl, 2007). This is striking, as within inclusive education

social participation is considered to be vital. Parents often report increased social

opportunities as their main motive for sending their child with special needs to a

regular school (Sloper & Tyler, 1992). They expect their child to build positive

relationships with typical peers. However, research has repeatedly shown that

inclusion of pupils with special needs does not automatically lead to an increase

of friendships between these pupils and their typical counterparts (De Monchy,

Pijl & Zandberg, 2004). A study by Frostad and Pijl (2007) of Norwegian inclusive

classrooms suggests that nearly a quarter of pupils with special needs have

serious difficulties forming relationships in their peer group, while for their typical

peers this is only 8 percent. This is worrisome, because the consequences of

negative social experiences in school can be far-reaching. For instance, a low

social status in childhood and peer relationship difficulties might lead to

externalising (e.g. aggression) and internalising problems (e.g. anxiety)

(Durrant, Cunningham & Voelker, 1990), and to maladjustment later in life

(Bagwell, Newcomb & Bukowski, 1998; Parker & Asher, 1987). Because of the

rather harmful long-term effects of negative social experiences at school, it is

important to monitor the social participation of pupils. This entails a vital task for

teachers.

Koster, Nakken, Pijl, Van Houten and Lutje Spelberg (in press) developed

a questionnaire for teachers to assess the social participation of pupils with

special needs. First, an analysis of literature was conducted in order to define the

concept of social participation. This analysis revealed four key themes in social

participation: friendships/relationships, contacts/interactions, pupil’s social self-

perception and acceptance by classmates (Koster, Nakken, Pijl & Van Houten, in

CHAPTER 5

110

press). The key themes laid the foundation for the so-called Social Participation

Questionnaire, which aims at helping teachers make more accurate assessments

of the social participation of their pupils, and aids in noticing problems in time.

In the present study, the quality of the Social Participation Questionnaire is

central. Using empirical data from both pupils with and without special needs,

the quality of the separate statements for each subscale, the separability of the

four subscales and their reliability will be assessed. The Questionnaire’s

discriminant validity will be examined by comparing the scores of pupils with

special needs with those of pupils without special needs.

5.2 Method

5.2.1 Participants

For practical reasons, data collection took place in two periods. In each period,

300 regular primary schools were invited to participate in the study. According to

files from the Ministry of Education, the invited schools have at least one pupil in

Grades 1 to 3 who receive a so-called pupil-bound budget. Such a budget is

allocated to pupils who, according to independent committees, meet the Dutch

national criteria for a pupil-bound budget. Herewith, several categories of

disabilities are distinguished, each with their own criteria. Those involve, among

other things, categories of the Diagnostic and Statistical Manual of Mental

Disorders (DSM IV) and intellectual ability, which are assessed by qualified

psychiatrists or psychologists who operate independently from the committees.

The invited schools were randomly drawn from the population of 2074 Dutch

regular primary schools with a pupil with a pupil-bound budget in Grade 1, 2

and/or 3 within 2.5 hours of travelling time from the city of Groningen. This area,

covering nearly two-third of the Netherlands, was chosen to make the data

collection feasible.

First sub-sample. Of the 300 invited schools, 53 were involved in the study. One

school was too busy to participate and 22 schools were unable to participate

because they had no pupils with special needs in Grades 1, 2 or 3. The remaining

224 schools did not respond to our invitation. To examine possible bias in results

from the cooperating schools, random selections of about 20 percent of the


111

schools who had not responded at all to our invitation to cooperate in the study

(45 schools) and 35 percent of the schools who did cooperate (20 schools) were

invited to complete a short questionnaire by phone.

This questionnaire is largely based on a questionnaire developed by Frostad and

Pijl (2007). In total, 41 non-cooperating schools (91%) and 19 (95%)

cooperating schools answered the questions. For the non-cooperating schools,

the first question aimed at the reason for not participating in the study. It turned

out that 12 schools were not able to participate, because they had no pupils with

special needs in Grade 1, 2 or 3. These 12 schools were not asked to complete

the remainder of the questionnaire. The remaining 29 non-cooperating schools

and 19 cooperating schools completed the questionnaire. The questionnaire

consisted of eight questions addressing attitudes to inclusion, teacher

characteristics, satisfaction about the budget for materials and schooling,

satisfaction about collaboration with parents and the degree to which teachers

felt supported by the director, colleagues and peripatetic teachers at educating

pupils with special needs. The questions had to be answered on a 5-point scale,

ranging from 0 (yes, very much) to 4 (no, not at all). Comparison of the mean

scores of cooperating and non-cooperating schools showed only slight

differences, which appeared to be nonsignificant (see Table 1).

CHAPTER 5

112

Table 1. Outcomes non-response survey first sub-sample

Statements non-response

questionnaire

Mean cooperating

schools (n=19)

Mean non-cooperating

schools (n=29)

95% CI dif

Attitude towards inclusion 1.05 (SD=0.68) 1.03 (SD=0.62) -0.41;0.37

Support by director 0.89 (SD=0.74) 0.55 (SD=0.57) -0.72;0.06

Support by colleagues 0.53 (SD=0.70) 0.66 (SD=0.61) -0.26;0.51

Support by peripatetic

teachers

0.42 (SD=0.51) 0.48 (SD=0.63) -0.29;0.41

Satisfaction about

collaboration with parents

0.53 (SD=1.12) 0.55 (SD=1.06) -0.62;0.67

Satisfaction about budget for

special materials

0.68 (SD=0.58) 0.72 (SD=0.59) -0.31;0.39

Satisfaction about

opportunities for schooling

0.89 (SD=0.32) 0.83 (SD=0.38) -0.28;0.15

Degree to which extra

schooling has taken place

1.79 (SD=0.86) 1.86 (SD=0.83) -0.44;0.57

In addition, the outcomes showed that the mean total number of pupils on the

schools, the mean number of pupils with special needs on the schools and the

mean degree of teaching experience of teachers did not significantly differ for

both groups of schools (total number of pupils: 95% CI of mean difference [95%

CI dif]=-1.34;1.18; number of pupils with special needs: 95% CI dif=-2.1;3.78;

degree of teaching experience: 95% CI dif=-1.56;1.18). The comparison

suggests that the cooperating schools do not over-represent schools with the

most positive view on and most positive experiences with inclusive education in

the Netherlands. These outcomes suggest that there is no reason to assume any

systematic bias in the sample involved in the study.

A final remark with regard to the sample concerns the noteworthy fact that

almost 30 percent of the invited schools were unable to participate because they

had no pupils with special needs in Grades 1 to 3. The names of the schools were

distracted from files of the Ministry of Education. The data in the files applied to

the school year 2005-2006, since there were no such files available for the

school year 2006-2007 in which the study took place. Due to the fact that the


113

data in the files were somewhat outdated, probably several schools were invited

wrongfully for the study.

As described above, 53 schools participated in the study. Since several schools

had a pupil with special needs in more than one class and since 15 classes

contained two or more pupils with special needs, 75 classes with in total 96

pupils with special needs participated in the study. Next to the 96 pupils with

special needs, the classes comprised 1652 pupils without special needs.

Second sub-sample. Of the 300 invited schools, 66 took part. Contrary to the

first sub-sample, the schools of the second sub-sample were asked to respond to

our invitation to participate in the study not only when they were willing to

participate, but also when they did not want to participate. If schools did not

want to take part in the study, they were asked to write down the reason for not

participating. In total, more than 70 percent of the schools (n=211) responded

to our invitation to participate. Of these 211 schools, 66 schools were willing to

take part in the study, while 145 schools were not able or not willing to

participate. Several reasons for not participating were mentioned. Almost 30

percent of the schools had no pupils with special needs in Grades 1 to 3, about

20 percent of the schools were too busy, and another 14 percent of the schools

considered the study as being a too large burden for the pupils with special

needs and their classmates. About 7 percent of the schools already participated

in another study and almost 5 percent of the schools were not interested in the

study. Several other schools (about 3 percent) were occupied with organisational

changes, or parents did not gave permission to participate in the study (about 3

percent). The remainder of the schools (about 18 percent) did not mention a

reason for not participating. As in the former sample the participating schools did

not differ from the non-participating ones, there was no reason to assume any

systematic bias in the sample involved in the study. Therefore, it was decided to

refrain from repeating the non-response survey in the second sub-sample.

As named above, 66 schools took part in the study. In total, 27 schools had a

pupil with special needs in more than one class and 34 classes contained two or

more pupils with special needs. Consequently, 105 classes with in total 141

pupils with special needs participated in the study. The total number of pupils

(including the 141 pupils with special needs and their classmates) in the 105

classes was 2426.

CHAPTER 5

114

Total sample (sub-sample 1 and sub-sample 2). In total, 119 schools with 180

participating classes were involved. Each teacher was asked to fill in the Social

Participation Questionnaire for the pupil(s) with special needs and for two pupils

without special needs. As the Questionnaire was expected to be filled in for a

diversity of children, this resulted in a broad range of scores. In seven classes,

the teacher filled in the Questionnaire for two pupils with special needs and for

only one pupil without special needs. The final sample thus consisted of 590

pupils, 237 of them with special needs and 353 without special needs. An

overview of categories of disabilities and its distribution is presented in the

second column of Table 2.

Table 2. Distribution of pupils with special needs into categories of disabilities and gender

Category of disabilities Number of pupils in absolute

numbers and percentages

Boy Girl

Behavioural disorder 29 (12.2%) 25 (86.2%) 4 (13.8%)

Autistic spectrum disorder 97 (40.9%) 83 (85.6%) 14 (14.4%)

Motor disability 35 (14.8%) 27 (77.1%) 8 (22.9%)

Intellectual disability 26 (11.0%) 10 (38.5%) 16 (61.5%)

Speech/language disabilities 47 (19.8%) 32 (68.1%) 15 (31.9%)

Learning disabilities 2 (0.8%) 2 (100%) 0

Chronic illness 1 (0.4%) 1 (100%) 0

Total 237 (100%) 180 (75.9%) 57 (24.1%)

5.2.2 Instrument

The Social Participation Questionnaire was constructed during a step by step

process. An exploration of literature (Koster, Nakken, Pijl & Van Houten, in

press) revealed four key themes of social participation: friendships/relationships,

contacts/interactions, pupil’s social self-perception and classmates’ acceptance.

In addition, on the basis of literature, tens of statements concerning social

participation were described. The content of the statements is mainly based on

literature and partly on the clinical and research experience of the researchers

who work for many years in the tradition of research looking at the social

dimension of inclusion in education. Since this process might have been coloured

by the researchers’ views, a panel consisting of 14 people with much practical

experience in inclusive education was asked to critically examine the statements.

The panel members were asked whether the formulated statements were in line


115

with their views of social participation, and to mention omissions. This detailed

verification of the statements was intended to increase the chance of including all

essential topics of social participation. Based on the panel’s comments, a list of

74 statements was compiled, yielding a picture of social participation (Koster,

Nakken, Pijl, Van Houten & Lutje Spelberg, in press). A second check of the

relevance of each statement was made using the ratings of a large group of

respondents (n=190) who all had experience with inclusive education. The

respondents were asked to rate the importance of each statement. Those

statements rated as important or very important by at least 75 percent of the

respondents were selected for the Questionnaire. In a pilot study the 30 resulting

statements were psychometrically examined on the basis of scores obtained on

60 pupils (25 with special needs and 35 without). The results of a confirmatory

factor analysis (Oblique Multiple Group Method, Holzinger, 1944) largely

supported the Questionnaire’s division into four subscales. The analysis revealed

that 22 out of 30 statements correlated strongest with the subscale they were

assigned to. The remaining eight statements had higher correlations with one or

more subscales they were not part of. The reliabilities of the

‘Friendships/relationships’ and ‘Contacts/interactions’ subscales were high

(respectively 0.94 and 0.88), whereas the reliabilities of the ‘Social self-

perception of pupil’ and ‘Acceptance by classmates’ subscales were fairly low

(respectively 0.69 and 0.64). In order to further improve the reliability of the

latter subscales, respectively three and two statements were added. One

statement, belonging to the ‘Contacts/interactions’ subscale was removed from

the list, as this statement was incongruous and correlated negatively with the

total score on the Questionnaire. Thus, the revised Questionnaire under study

here consisted of 34 statements representing the four key themes of social

participation in 6 to 14 statements, which have to be scored on a 5-point scale

(ranging from ‘this does not apply at all’ to ‘this strongly applies’). The majority

of the statements (n=24) are phrased in terms of social participation, while the

remainder (10 statements) is formulated in terms of social segregation. The key

themes ‘friendships/relationships’ and ‘acceptance by classmates’ are

represented only by statements phrased in terms of social participation. Table 3

gives some examples of both types of statements.

CHAPTER 5

116

Table 3. Examples of statements of the Social Participation Questionnaire

Key theme Statement representing

social participation

Statement representing

social segregation

Friendships/relationships ‘the pupil is a member of a

group of friends’

Contacts/interactions ‘in free time (e.g. during

recess) the pupil plays with

classmates’

‘classmates regularly exclude

the pupil from activities’


perception

‘the pupil has the feeling s/he

belongs to the group’

‘the pupil feels lonely at school’

Acceptance by classmates ‘classmates consider the pupil

as a full member of the group’

5.2.3 Analysis

Item Response modelling (see e.g. Embretson & Reise, 2000; Hambleton,

Swaminathan & Rogers, 1991; Van der Linden & Hambleton, 1997) was used to

assess the quality of the individual statements of each of the four subscales, and

to examine their mutual separability. An Item Response Model (IRM) expresses

the links between item responses and a latent trait, which pertains to the

measured concept. For polytomous items, per item each response category is

related to the latent trait via an item response function (IRF). An IRF may follow

a specific function, yielding parametric IRMs, or meet certain assumptions,

yielding nonparametric IRMs. We considered both nonparametric and parametric

IRMs (the normal-ogive multidimensional model [McDonald, 1997]), but as the

latter failed to fit the data, we only discuss the nonparametric IRMs.

The two nonparametric models considered here are based on three

assumptions: unidimensionality (i.e., all items in the subscale measure the same

latent trait), local independence (i.e., a pupil’s score on an item is not influenced

by the scores on the other items of the scale) and monotonicity (i.e., all IRFs are

monotone, non-decreasing). These three assumptions comprise the Monotone

Homogeneity Model (MHM; Junker & Sijtsma, 2001; Mokken, 1971; Sijtsma &

Molenaar, 2002). When those assumptions are met for a set of items, the

important implication is that the individuals’ sum scores of the items provide the

ordering of the individuals on the latent trait. When the three assumptions of the

MHM model hold, and the IRFs do not intersect, the more restrictive Double

Monotonicity Model (DMM) holds. The nonintersection of the IRFs implies


117

invariant item ordering, which means that the items have the same ordering in

terms of probabilities of positive responses to those items (or item category), for

all values of the latent trait.

For each of the subscales, the fit of both the MHM and DMM is assessed using the

programme Mokken Scale analysis for Polytomous items (MSP; Molenaar &

Sijtsma, 2000). For each item, the Item Scalability Coefficient Hi indicates

whether item i fits the MHM, because under the MHM 0 ≤ Hi ≤ 1. Moreover, it

indicates to what extent the item aids in discriminating across individuals. To

ensure items meet the MHM assumptions and to have sufficient discrimination

power, it is generally accepted that Hi should exceed 0.3 (Mokken, 1971).

Coefficient H is a weighted mean of the Hi-values of a set of items making up a

scale, and indicates the degree to which pupils can be ordered by means of the

sum score on the items. With 0.3 ≤ H < 0.4 a subscale is regarded as weak, with

0.4 ≤ H ≤ 0.5 as moderate and with H > 0.5 as strong. Apart from the Item

Scalability Coefficients, other diagnostics are used to assess whether the

monotonicity assumption and the non-intersection assumption of the DMM hold

for each item. Those diagnostics are summarised into criteria values, where a

value ≥ 80 strongly suggests an assumption violation, values between 40 and 80

are questionable, and a value ≤ 40 is satisfying (Molenaar & Sijtsma, 2000). To

assess possible differential item functioning across the groups of pupils with and

without special needs, the ordering of the response categories of all items is

compared across groups. A different ordering suggests differential item

functioning, which would imply that the scales could not be used to compare

individuals from both groups.

To assess the separability of the subscales, the automated item selection

procedure (AISP) in MSP will be used, following the guidelines of Sijtsma and

Molenaar (2002; pp. 80-82). The AISP aims at finding clusters of items, such

that within each cluster the Item Scalability Coefficients exceed a (user-

specified) boundary, and between clusters the Item Scalability Coefficients are

lower than the boundary. To this end, a stepwise selection procedure is applied.

The AISP will be performed on all items using various boundary values, which will

indicate whether the items pertain to one scale only or whether they are

separable into more than one subscale. When more subscales are distinguished

in the AISP, it will be evaluated whether the items are actually assigned to the

prespecified subscales.

CHAPTER 5

118

To sum up, the following analysis strategy will be used: First, for each

subscale the separate statements will be assessed using the Scalability

Coefficients, criteria values to assess the assumptions of the MHM and the DMM,

and the diagnostics for differential item functioning across the two groups. If

necessary, and after considering their content, statements may be deleted from

the subscales. Second, the scalability of each resulting subscale is assessed.

Third, the separability of the subscales will be examined.

When the subscales are in their definite composition, the reliability of both the

subscales and the total Questionnaire will estimated using Rho. Rho suffers from

bias to a lesser extent than classical reliability estimates (Sijtsma & Molenaar,

2002). According to Nunnally and Bernstein (1994, pp. 265), reliabilities of 0.80

and more are satisfactory in basic research.

Finally, discriminant validity of the entire Questionnaire and the four subscales

will be examined by comparing the scores of pupils with special needs with those

of typical pupils. The former are expected to receive lower scores on the

Questionnaire and on each of the four subscales. As there is no overlap of

content between the criteria used to classify the pupils and the Questionnaire’s

statements, a confirmation of this expectation implies evidence for the

Questionnaire’s discriminant validity.

5.3 Results

5.3.1 Mokken Scale Analysis

The four subscales were analysed separately to identify deviant statements, and

to assess possible differential item functioning across pupils with and without

special needs. In the following, only the examination of the properties of the

‘Friendships/relationships’ subscale will be described in detail. Examination of the

other subscales was done similarly. The qualities of the statements of those

subscales will be described briefly; a detailed description can be obtained from

the first author.

‘Friendships/relationships’ subscale. For the six statements of this subscale it was

found that 0.40 ≤ Hi ≤ 0.55. For each of the statements, the criteria values did

not indicate any violation of the monotonicity assumption, but suggested a


119

violation of the non-intersection assumption for one statement (‘after school, the

pupil plays with one or several classmates’). Because this statement showed

much overlap with another statement (‘the pupil makes engagements with one

or more classmates to play after school’) in terms of content, it was removed

from the scale. The scalability of both the remaining five statements and the

resulting subscale appear intermediately strong, at 0.42 ≤ Hi ≤ 0.54, and

H=0.49 for the entire subscale.

Comparison of the ordering of the response categories for all statements

across the two groups revealed a slight indication of differential item functioning

of one statement (‘the pupil is being invited to birthday parties’). Inspection of

the order revealed that for pupils with relatively low scores on the Friendship

scale, pupils without special needs are invited more often than pupils with special

needs, but the difference appeared to be small. On the basis of its content and

its sufficient Hi (0.42), it was decided to retain this statement in the subscale.

Table 4 presents an overview of the quality of the statements of this subscale.

Table 4. Hi Coefficients and criteria values of statements ‘Friendships/relationships’

subscale

Statement Hi Criteria values

Selected statements

‘the pupil is a member of a group of friends’ (8) 0.53 38/-

‘the pupil has one or more friends in the classroom’ (11) 0.54 40/-

‘the pupil makes engagements with one or more classmates to play

after school’ (19)

0.49 23/-

‘one or more classmates invite the pupil to play during holidays’ (25) 0.43 34/-

‘the pupil is being invited to birthday parties’ (28) 0.42 54/-

Removed statement

‘after school, the pupil plays with one or more classmates’ (21) 0.50 66/44

‘Contacts/interactions’ subscale. For 11 out of 14 statements, we found

that 0.42 ≤ Hi ≤ 0.65, while for the remaining three statements it was revealed

that 0.35 ≤ Hi ≤ 0.39. It was decided to remove these three statements from the

subscale. For the remaining 11 statements, the criteria values did not indicate

any violation of the monotonicity assumption. However, for two statements, the

criteria values indicated a violation of the non-intersection assumption (relevant

criteria values of 70 and 112 for the one statement; relevant criteria values of

CHAPTER 5

120

100 and 175 for the other). Both statements were removed from the subscale.

The five removed statements had weak discrimination power, since almost all

pupils received a high score on these statements (mean scores range from 4.4 to

4.8). The discrimination power of the remaining nine statements is stronger, as

the distribution of scores on these statements is more diverse (mean scores

range from 3.5 to 4.1).

The scalability of the nine statements appears strong, 0.62 ≤ Hi ≤ 0.74. The

same applies to the resulting subscale, H=0.70. Differential item functioning

appeared to be absent, which implies that the scale scores are directly

comparable across pupils with and without special needs.

‘Social self-perception of pupil’ subscale. For five out of seven statements

it was found that 0.43 ≤ Hi ≤ 0.53, while for the remaining two statements 0.30

≤ Hi ≤ 0.34. For one of these two statements, the non-intersection assumption

was violated (relevant criteria values of 107 and 104). Both statements aim at

the pupil daring to tell or to ask something in the classroom. These statements

are somewhat deviating from the other statements in the subscale, which focus

on feelings of the pupil (like happiness, loneliness, sense of belonging) in stead

of daring. Both statements were removed from the subscale, after which the

scalability of the remaining five items slightly increased (0.47 ≤ Hi ≤ 0.56). For

none of the remaining five statements any violation of monotonicity assumption

or non-intersection assumption was suggested and there is no indication of

differential item functioning. Overall scalability coefficient H is 0.51, which

implies a strong subscale.

‘Acceptance by classmates’ subscale. For five out of seven statements of

this subscale it was found that 0.43 ≤ Hi ≤ 0.53, while for the other two

statements Hi was 0.36 and 0.33. For each of the remaining five statements, the

criteria values did not suggest any violation of both the monotonicity assumption

and the non-intersection assumption. The scale scores are directly comparable

across pupils with and without special needs, as differential item functioning

appeared to be absent. The scalability of the individual five statements and the

resulting subscale appears strong, with 0.48 ≤ Hi ≤ 0.61 and H=0.54 for the

entire subscale.


121

To summarise, ten out of 34 statements were removed because of their

low Hi value and/or high criteria values. For each of the four resulting subscales,

based on 24 statements, no violations of the four assumptions of the DMM were

observed. In addition, for each subscale differential item functioning appeared to

be absent, implying that the scale scores are directly comparable across pupils

with and without special needs.

5.3.2 Separability of the subscales

The separability of the four subscales was examined with the automated item

selection procedure (AISP). First, the AISP was applied to all 24 statements at

different boundary levels. The results indicated a multidimensional item set, as

with increasing boundary values four subscales were distinguished. With a

boundary value as high as c=0.60, the first selected scale contained all nine

statements of the ‘Contacts/interactions’ subscale. The subsequent three scales

consisted of two statements each, neatly belonging to the

‘Friendships/relationships’, ‘Pupil’s social self-perception’ and ‘Acceptance by

classmates’ subscales, respectively. The remaining nine statements were

indicated as non-scalable, which is not surprising given the high boundary value.

With boundary values c lower than 0.60, the first selected scale consisted of the

‘Contacts/interactions’ statements, mixed with certain statements from the other

three subscales. This is not surprising, because the ‘Contacts/interactions’

subscale itself was found to be much stronger (H=0.70) than the remaining three

(0.49<H<0.54). Presumably, statements of the other three subscales not only

contain aspects related to their own key theme, but also aspects connected to

contacts and interactions.

Repeating the AISP at different boundary levels with all statements except

for the nine statements of the ‘Contacts/interactions’ subscale resulted in a clear

distinction of the three remaining subscales. This indicates that the statements

which initially were attracted to the ‘Contacts/interactions’ subscale also contain

unique aspects that belong to the key themes they are representing. Those

results suggest that the four subscales are distinguished to a reasonable extent.

CHAPTER 5

122

5.3.3 Revised version of the Social Participation Questionnaire

The raw scores on the ‘Contacts/interactions’ subscale range from 9 (9x1) to 45

(9x5), the raw scores on the three other subscales from 5 (5x1) to 25 (5x5). In

order to facilitate interpretation of the subscale scores, they were linearly

rescaled so that the minimum score on all subscales is 0 and the maximum 25.

For example, the raw scores on the ‘Acceptance by classmates’ subscale were

rescaled by subtracting 5 and then multiplying by 25/20. The total score on the

Questionnaire ranges between 0 and 100.

Reliability

The reliability of the final Questionnaire was estimated by Rho. Rho of the total

Questionnaire is 0.95, Rho of the subscales varies between 0.80 and 0.95 (see

Table 5), meaning a reasonable reliability (Nunnally & Bernstein, 1994).

Table 5. Scalability (H) and reliability (Rho) of final subscales

Subscale Number of

statements

Subscale H Rho

Friendships/relationships 5 0.49 0.80

Contacts/interactions 9 0.70 0.95

Pupil’s social self-perception 5 0.51 0.82

Acceptance by classmates 5 0.54 0.83

Correlation between subscales

As suggested by the AISP, there appears a clear distinction between the four

subscales. In order to examine the relationships between the subscales,

Spearman’s Rho was calculated. The correlations between all four subscales

appeared to be strong (see Table 6), implying that the scales are closely

connected. The correlation between the subscales ‘Friendships/relationships’ and

‘Contacts/interactions’ is strongest. This seems to be understandable, since

friends generally spend a lot of time together. As a result, pupils with many

friends presumably will have a large amount of interactions. The strong

correlations between the other subscales are explainable too. For instance,

having friends is associated with enhanced opportunities to exercise behaviours

and acquire skills related to social, emotional, and cognitive growth (Ladd, 1990;

Newcomb & Bagwell, 1996). Acquiring these skills may facilitate the acceptance


123

by peers, which subsequently may have a positive influence on the pupil’s social

self-perception. Besides, having more mutual friends is related to a more positive

social self-concept (Vandell & Hembree, 1994, in Bagwell, 2004). On the other

hand, the lack of friends may negatively influence a pupil’s self-perception

(Newcomb & Bagwell, 1996; Parker & Asher, 1993). Parker and Seal (1996)

revealed that ‘chronically friendless’ children are perceived by their peers as shy,

timid, withdrawn and easily angered. These characteristics might prevent peers

from interacting with these pupils, which probably negatively influences the self-

concept of the pupils and their acceptance by peers.

Although the subscales strongly correlate, the correlations are not perfect. In

addition, it turned out that for pupils with special needs, the correlations between

the subscales are less strong, compared to pupils without special needs (see

Table 6). The differences between these two groups of pupils are significant with

regard to the correlations of the subscales ‘Pupil’s social self-perception’ &

‘Friendships/relationships’, ‘Acceptance by classmates’ &

‘Friendships/relationships’ and ‘Contacts/interactions’ & ‘Pupil’s social self-

perception’ (p<0.05). The lower connection between subscales for pupils with

special needs gives extra cause for distinguishing the four subscales. If only the

total score on the Questionnaire would be viewed, valuable differences in the key

themes might remain unnoticed. If also the subscale scores are taken into

consideration, this is expected to result in a more proper and balanced view of

the social participation of the pupil.

Table 6. Correlations between subscales

Friendships/

relationships

Contacts/

interactions

Pupil’s social

self-perception

Acceptance by

classmates

Contacts/

interactions

0.77

(0.70, 0.73)


perception

0.61

(0.42, 0.64)

0.71

(0.62, 0.72)

Acceptance by

classmates

0.54

(0.39, 0.63)

0.60

(0.59, 0.64)

0.55

(0.47, 0.59)

(first number, without brackets, is correlation for all pupils, first number between brackets is correlation for

pupils with special needs, second number between brackets is correlation for pupils without special needs)

All correlations are significant at the 0.01 level

CHAPTER 5

124

5.3.4 Discriminant validity

To assess the Questionnaire’s discriminant validity, a comparison was made

between the scores of pupils with special needs and those of their typical

classmates. Based on literature (e.g., De Monchy et al., 2004; Pijl, Frostad &

Flem, in press; Scheepstra, Nakken & Pijl, 1999), it was expected that the social

participation of pupils with special needs would be assessed less positively than

that of their classmates. A distinction was also made between different

categories of disabilities. Pupils diagnosed as having autistic spectrum disorders

and pupils diagnosed as having behavioural disorders were expected to receive

the lowest mean scores on the Questionnaire, as research has shown that these

pupils find it particularly difficult to build relationships with typical peers and are

at risk of becoming isolated in the classroom (De Monchy et al., 2004; Garrison-

Harrell, Kamps & Kravits, 1997). Both groups of pupils were expected to receive

low scores on each of the four subscales. Pupils diagnosed as having motor

disabilities were expected to receive the highest scores on the four subscales, as

their type of disability presumably has the least impact on social functioning in

the classroom. In addition, motor disabilities are visible and understandable for

classmates, which fosters acceptance (Lewis, 1995, in Laws & Kelly, 2005).

Comparisons were made between subgroups of pupils with different categories of

disabilities for the scores on the total Questionnaire and for the scores on the

four subscales.

Total Questionnaire. On average, pupils with special needs got a

substantially lower score (M=67.5, SD=16.6) than their typical counterparts

(M=80.2, SD=15.1), and this difference appeared significant (95% CI dif =

10.1;15.3). This implies that the social participation of pupils with special needs

is assessed lower compared to the social participation of their classmates without

special needs. As this confirms the expectation that the social participation of

pupils with special needs would be assessed less highly, it can be regarded as

evidence for the Questionnaire’s discriminant validity.

A one-way analysis of variance (ANOVA) revealed significant group

differences between pupils with different categories of disabilities: F(4,

225)=4.88, p<0.05 (where the learning disabilities and chronic illness categories

were excluded since the group sizes were too small). Pupils diagnosed as having

behavioural disorders (M=62.9, SD=18.3) and autistic spectrum disorders

(M=64.1, SD=15.6) got the lowest scores, which confirms our expectation. The


125

mean score of pupils diagnosed as having intellectual disabilities is slightly higher

(M=67.6, SD=16.0). Pupils diagnosed as having speech/language disabilities

(M=74.3, SD=14.1) and pupils diagnosed as having motor disabilities (M=73.2,

SD=16.5) have the highest mean scores. Scheffé post hoc comparison shows

that the social participation of pupils diagnosed as having autistic spectrum

disorders is assessed significantly lower than that of pupils diagnosed as having

speech/language disabilities (p<0.05). There were no significant differences

between other categories of disabilities.

Subscales. Closer inspection revealed that for the four subscales the same

pattern arises as for the total Questionnaire. Corresponding to the scores on the

total Questionnaire, for each of the subscales the mean scores of pupils with

special needs are significantly lower than those of their typical classmates (see

Table 7).

Table 7. Mean scores on subscales of pupils with (n=233) and without special needs

(n=346)

Subscale Mean score, pupils

with special needs

Mean score, pupils

without special needs 95% CI dif

Friendships/relationships 13.9 (SD=6.5) 19.1 (SD=6.5) 4.2;6.2

Contacts/interactions 18.0 (SD=4.4) 21.8 (SD=3.5) 3.2;4.4

Pupil’s social self-perception 18.7 (SD=4.9) 21.1 (SD=4.1) 1.7;3.2

Acceptance by classmates 16.8 (SD=4.8) 18.1 (SD=4.8) 0.5;2.1

An ANOVA on the ‘Acceptance by classmates’ subscale showed no

significant differences between pupils with different categories of disabilities: F

(4, 227)=1.70, p=0.15. ANOVAs on the ‘Friendships/relationships’,

‘Contacts/interactions’ and ‘Pupil’s social self-perception’ subscales revealed

significant differences between pupils with different categories of disabilities

(respectively, F(4, 227)=3.50, p<0.05; F(4, 229)=6.55, p<0.05; F(4,

228)=2.98, p<0.05). Scheffé post hoc comparison revealed that the scores on

the ‘Friendships/relationships’ subscale significantly differed between pupils

diagnosed as having speech/language disabilities and pupils diagnosed as having

autistic spectrum disorders, whereby the latter received the lower scores, as

hypothesised. In addition, the scores on the ‘Contacts/interactions’ subscale

significantly differed between pupils diagnosed as having speech/language

disabilities and pupils diagnosed as having motor disabilities on the one hand,

CHAPTER 5

126

and pupils diagnosed as having autistic spectrum disorders on the other hand,

where as hypothesised the latter showed lower scores. For an overview of the

results, we refer to Table 8.

Table 8. Mean scores on subscales of pupils with different categories of disabilities



Motor disability


Speech/language disabilities

# pupils (% total)*

29 (12.2)

97 (40.9)

35 (14.8)

26 (11.0)

47 (19.8)

Friendships/ relationships

13.2 (SD=7.4)

12.6 (SD=5.9)

15.6 (SD=7.0)

13.0 (SD=7.5)

16.3 (SD=5.5)

Contacts/ interactions

17.0 (SD=4.7)

16.9 (SD=4.2)

19.9 (SD=3.9)

17.5 (SD=4.3)

19.9 (SD=3.8)

Pupil’s social self-perception

17.5 (SD=4.9)

17.7 (SD=5.2)

19.8 (SD=4.7)

19.8 (SD=4.6)

20.0 (SD=3.7)


15.1 (SD=5.2)

16.7 (SD=4.6)

17.9 (SD=4.6)

17.3 (SD=4.6)

17.4 (SD=4.9)

*Due to the small number of pupils with learning disabilities (n=2) and pupils with

chronic illness (n=1), no data about these pupils are included in the Table.

5.4 Discussion

In this paper, the psychometric qualities of a new teacher questionnaire to

assess the social participation of pupils with special needs in regular primary

schools were examined. A Mokken Scale Analysis was conducted for this

purpose. Based on the results of various analyses, ten out of 34 statements were

removed from the Questionnaire.

As for each of the resulting subscales the Mokken’s double monotonicity

model appeared to fit well, the subscale scores are on an ordinal scale, and the

separate statements per subscale are invariantly ordered along the subscale. The

scale scores are directly comparable across pupils with and without special

needs, because differential item functioning appeared to be absent. Subsequent

analyses did support the division of social participation into the four distinct key

themes. The Questionnaire as a whole and its subscales turned out to be

sufficiently reliable. Evidence for the Questionnaire’s discriminant validity was

also obtained as the expectation that the social participation of pupils with

special needs would be assessed as less high using the Questionnaire could be

confirmed: both on the total Questionnaire and on each of the four subscales,


127

the mean scores of pupils with special needs were significantly lower than those

of their typical counterparts. More aspects of the Questionnaire’s construct

validity will be examined in further research.

Scheffé post hoc comparisons of the scores on the Social Participation

Questionnaire made clear that there are significant differences between pupils

with different categories of disabilities. It was revealed that the scores on two

out of four subscales significantly differed between pupils diagnosed as having

autistic spectrum disorders and pupils diagnosed as having speech/language

disorders (and for one subscale also pupils diagnosed as having motor

disabilities), whereby the latter received the highest scores. The other

differences in subscale scores between pupils with different categories of

disabilities were not significant. This might be related to the fact that the

subgroups of disabilities contained rather small numbers of pupils. We could have

decided to involve less categories of disabilities in the study, in order to compose

larger subgroups of specific categories of disabilities. However, we consciously

included such a variety of disabilities in the sample, as the Questionnaire has

been developed for all pupils. This fits with the fact that in the Netherlands, the

group of pupils with special needs attending regular education is rather

heterogeneous. In order to examine the applicability of the Questionnaire, it was

desirable to involve pupils with various categories of disabilities in the study. In a

future study, when the Questionnaire has its definitive form, it would be

advisable to involve larger homogeneous subgroups of pupils. That way it might

be possible to aim analyses on possible differences in scores on the

Questionnaire of pupils with different categories of disabilities. Among other

things, one could assess possible differential item functioning between subgroups

of pupils with different categories of disabilities. In the current study, we do not

know if the scale scores are directly comparable across pupils with different

categories of disabilities, as the subgroups were too small to conduct analyses.

However, on theoretical grounds, we assume that the Questionnaire is applicable

for pupils with various categories of disabilities, as none of its statements are

particularly focused on specific categories of disabilities.

The low mean scores of the pupils diagnosed as having autistic spectrum

disorders (whether or not in combination with behavioural disorders) were in

accordance with other studies, which revealed that in particular these pupils

experience difficulty building relationships with typical peers and are at risk of

CHAPTER 5

128

becoming isolated in the classroom (De Monchy et al., 2004; Garrison-Harrell et

al., 1997). Further research into differences between categories of disabilities is

recommended, since treating pupils with special needs as one homogenous

group might reveal a too negative picture of the social participation of pupils with

specific types of disabilities.

In conclusion, the Social Participation Questionnaire has the potential to

become a useful tool for teachers, as it helps them obtain clarity about the social

participation of their pupils. The Questionnaire takes into consideration the

comprehensiveness of the concept of social participation by providing insight into

four distinguished themes of social participation, while other instruments only

focus on parts of this concept (for instance sociometric questionnaires mainly

focus on pupils’ friendships and/or their acceptance by classmates, and

observation scales mainly focus on interactions). The Social Participation

Questionnaire is a teacher-friendly instrument because it can be completed in a

short time and provides a reliable and wide picture of the social participation of

pupils. Further research into the Questionnaire’s psychometric properties is

needed. The outcomes of the present study provided evidence for the

Questionnaire’s discriminating power, because of the presence of the expected

score differences between pupils with and without special needs. It should be

kept in mind that this difference in scores is a necessary, but not sufficient

prerequisite for establishing the discriminant validity of our Questionnaire. More

aspects of the Questionnaire’s validity will be examined in further research.

After having obtained evidence for the validity of the Social Participation

Questionnaire, its usability in practice will be examined. The Questionnaire is

expected to help teachers make more appropriate assessments of the social

participation of their pupils. As a result, teachers might be encouraged to take

measures in time, which will contribute to optimise the situation of pupils with

special needs in inclusive classrooms.

5.5 References

Bagwell, C.L. (2004). Friendships, peer networks, and antisocial behavior. In J.

B. Kupersmidt & K.A. Dodge (Eds.), Children's peer relations: From

development to intervention (pp. 37-57). Washington, D.C.: American

Psychological Association.


129



69(1), 140-153.







Embretson, S.E. & Reise, S.P. (2000). Item response theory for psychologists.

Mahwah, New Jersey: Lawrence Erlbaum.




Education, 22(1), 15-30.




Hambleton, R.K., Swaminathan, H. & Rogers, H.J. (1991). Fundamentals of Item

Response Theory. Volume 2. Newbury Park: Sage Publications.

Holzinger, K.J. (1944). A simple method of factor analysis. Psychometrika, 9(4),

257-262.

Junker, B.W. & Sijtsma, K. (2001). Nonparametric Item Response Theory in

action: an overview of the special issue. Applied Psychological

Measurement, 25(3), 211-220.


peer group: a literature study focusing on the social dimension of inclusion

in education. International Journal of Inclusive Education.

Koster, M., Nakken, H., Pijl, S.J., Van Houten, E.J. & Lutje Spelberg, H.C. (in

press). Assessing social participation of pupils with special needs in

inclusive education: the construction of a teacher questionnaire.

Educational Research and Evaluation.

Ladd, G.W. (1990). Having friends, keeping friends, making friends, and being

liked by peers in the classroom: predictors of children’s early school

adjustment? Child Development, 61, 1081-1100.

CHAPTER 5

130

Laws, G. & Kelly, E. (2005). The Attitudes and Friendship Intentions of Children

in United Kingdom Mainstream Schools towards Peers with Physical or

Intellectual Disabilities. International Journal of Disability, Development

and Education, 52(2), 79-99.

Male, D.B. (2007). The friendships and peer relationships of children and young

people who experience difficulties in learning. In L. Florian (Ed.), The

SAGE handbook of Special Education (pp. 460-474). London: SAGE

Publications.

McDonald, R.P. (1997). Normal-ogive multidimensional model. In W.J. Van der

Linden & R.K. Hambleton (Eds.), Handbook of modern item response

theory (pp. 258-269). New York: Springer.

Mokken, R.J. (1971). A theory and procedure of scale analysis with applications

in political research. The Hague: Mouton.

Molenaar, I.W. & Sijtsma, K. (2000). User’s manual MSP5 for Windows.

Groningen: iecProGAMMA.

Newcomb, A.F. & Bagwell, C.L. (1996). The developmental significance of

children’s friendship relations. In W.M. Bukowski, A.F. Newcomb & W.W.

Hartup (Eds.), The company they keep: friendship in childhood and

adolescence (pp. 289-321). Cambridge: Cambridge University Press.

Nunnally, J.C. (1978). Psychometric Theory (2nd ed.). New York: McGraw-Hill.

Nunnally, J.C. & Bernstein, I.H. (1994). Psychometric Theory (3rd ed.). New

York: Mc.Graw-Hill, Inc.

Parker, J.G. & Asher, S.R. (1993). Friendship and friendship quality in middle

childhood: links with peer group acceptance and feelings of loneliness and

social dissatisfaction. Developmental Psychology, 29(4), 611-621.



Parker, J.G., & Seal, J. (1996). Forming, losing, renewing, and replacing

friendships: Applying temporal parameters to the assessment of children's

friendship experiences. Child Development, 67, 2248-2268.

Pijl, S.J., Frostad, P. & Flem, A. (in press). The social position of pupils with

special needs in regular schools. Scandinavian Journal of Educational

Research.


131

Powless, D.L. & Elliott, S.N. (1993). Assessment of social skills of native

American preschoolers: teachers’ and parents’ ratings. Journal of Social

Psychology, 31, 293-307.




Sijtsma, K. & Molenaar, I.W. (2002). Introduction to Nonparametric Item

Response Theory. Volume 5. Thousand Oaks: Sage Publications.




Van der Linden, W.J. & Hambleton, R.K. (1997). Handbook of modern item

response theory. New York: Springer.

CHAPTER 5

132

Chapter 6 Validation of the Social Participation Questionnaire1

CHAPTER 6

VALIDATION OF THE SOCIAL PARTICIPATION QUESTIONNAIRE

A slightly adapted version of this article has been submitted for publication and was co-authored by Alexander E.M.G. Minnaert, Han Nakken, Sip Jan Pijl and Els J. van Houten

CHAPTER 6

134

Abstract

This study addresses the convergent validity of a new teacher questionnaire to

assess the social participation of pupils with special needs in regular primary

schools. This Social Participation Questionnaire consists of four subscales

representing four key themes of social participation: friendships/relationships,

contacts/interactions, pupil’s social self-perception and acceptance by

classmates.

Inspection of the correlations between the Questionnaire’s subscales and other

instruments to assess the four key themes revealed that evidence for the

Questionnaire’s convergent validity was rather spurious. A second order analysis,

a confirmatory factor analysis using the LISREL program, removed this lack of

clarity and provided evidence for the Questionnaire’s convergent validity. The

outcomes of this analysis support the model of social participation, distinguishing

four key themes.


135

6.1 Introduction

Educating pupils with special needs in inclusive classrooms is an important

objective of many countries. Inclusion in education focuses not only on academic

issues, but also on meeting pupils’ physical, social and emotional needs (Koster,

Nakken, Pijl & Van Houten, in press). In this respect, maximizing social

participation of pupils with special needs is generally considered as a main aim of

inclusion. As described by Koster, Nakken, Pijl & Van Houten (in press), social

participation is about the presence of positive social contacts/interactions

between pupils with special needs and their classmates, acceptance of them by

their classmates, social relationships/friendships between them and their

classmates, and the pupils’ own perception of their social position.

The different aspects of social participation are closely connected. For

instance, having interactions with classmates seems to be a prerequisite for

making friends and, at the same time, pupils with many friends presumably have

many interactions, as friendships in pre-adolescence are associated with shared

activities (LaGreca, 1997, in Male, 2007). Having friends is associated with

enhanced opportunities to exercise behaviours and acquire skills related to

social, emotional and cognitive growth (Ladd, 1990; Newcomb & Bagwell, 1996).

Acquiring these skills may facilitate acceptance by peers, which subsequently

may have a positive influence on the pupil’s social self-perception. Conversely, a

lack of friends may negatively influence a pupil’s social self-perception. Pupils’

social self-perception may also influence their relationships. For instance, pupils

with a low social self-perception may avoid social situations, which prevents

them from building relationships with peers (Pijl & Frostad, 2008). Because of

this strong interrelatedness of the aspects of social participation, pupils may

become involved in a vicious circle, either positive or negative. This latter

scenario is alarming, as it is known from international research that peer

relationship difficulties, a low social status and a negative self-concept might lead

to psychological maladjustment later in life (Bagwell, Newcomb & Bukowski,

1998; Nelson, Rubin & Fox, 2005; Parker & Asher, 1987; Terry & Coie, 1991).

In this respect, pupils with special needs are at risk, since international

studies have frequently shown that inclusion of these pupils does not

automatically lead to an increase of friendships between them and their typical

peers (De Monchy, Pijl & Zandberg, 2004). Whereas the majority of pupils

CHAPTER 6

136

without special needs find no difficulty in making friends and building positive

relations with their peers (Powless & Elliott, 1993), pupils with special needs

experience considerably more difficulty building friendships in inclusive

classrooms (Frostad & Pijl, 2007). Several studies have shown that pupils with

(mild) disabilities — compared to their typical counterparts — report significantly

higher loneliness scores (Heiman & Margalit, 1998; Lackaye & Margalit, 2006;

Luftig, 1988; Pavri & Monda-Amaya, 2000; Williams & Asher, 1992) and a higher

degree of social dissatisfaction with their peer relationships (Taylor, Asher &

Williams, 1987, in Gresham & McMillan, 1997). In addition, the self-concept of

pupils with special needs is found to be significantly lower (Bender & Wall, 1994;

Cambra & Silvestre, 2003; Pijl, Skaalvik & Skaalvik, 2008), which might lead to

externalizing problems (e.g. aggression) and internalizing problems (e.g.

anxiety) (Durrant, Cunningham & Voelker, 1990).

Because of the rather harmful long-term effects of negative social

experiences at school, it is important to monitor the social participation of pupils

with special needs, an increasing percentage of whom attend regular education.

This concerns a vital task for teachers, since they should be the first ones to

notice if a pupil is excluded by peers and becomes isolated (De Monchy et al.,

2004). However, several studies (De Monchy et al., 2004; Scheepstra, Nakken &

Pijl, 1999; Whitney, Smith & Thompson, 1994) have shown that teachers’

assessments of the social participation of pupils with special needs in their

classroom are not always appropriate. They seem to have too positive a view of

the social participation of these pupils, and negative occurrences (like bullying or

neglecting) seem to be noticed insufficiently. It is vital that teachers make

appropriate and complete assessments of the social participation of pupils,

because only when they notice that a pupil becomes isolated or is being teased

can they take appropriate measures. Koster, Nakken, Pijl, Van Houten and Lutje

Spelberg (in press) developed a questionnaire for teachers to assess the social

participation of pupils with special needs. This Social Participation Questionnaire

aims at pupils in Grades 1 to 3, a crucial age period in the lives of children. The

friendships and contacts of pupils in lower grades (pre-schoolers) are still of a

transitory nature, as is their social position in class. In higher grades, children’s

friendships and contacts are more stable. However, in the highest grades of

primary education the social position of pupils tends to be so stable that it is very

hard to change it. Pupils’ views of each other are often persistent, so changing


137

these views is very difficult. Therefore, it is desirable to start intervening at an

early age instead of postponing interventions to the higher grades.

The Social Participation Questionnaire is based on a model of social

participation developed by Koster, Nakken, Pijl and Van Houten (in press). This

model is derived from a review of the literature and distinguishes four key

themes within the concept of social participation: friendships/relationships,

contacts/interactions, pupil’s social self-perception and acceptance by

classmates. The Social Participation Questionnaire covers the total concept of

social participation, with four key themes represented in four subscales. Unlike

other instruments, which aim at certain aspects of social participation (like

sociometric questionnaires or observation scales), the Social Participation

Questionnaire takes into consideration the comprehensiveness of the total

concept of social participation. This study addresses the question of whether the

Social Participation Questionnaire is valid to assess the social participation of

pupils with special needs in regular primary schools.

6.2 Method

In this study, the focus lies on one aspect of the construct validity, namely the

convergent validity of the Social Participation Questionnaire. Convergence means

that evidence from different sources gathered in different ways all indicates the

same or similar meaning of the construct (Kerlinger & Lee, 2000). Convergent

validity is therefore the degree to which two or more attempts to measure the

same concept are in agreement (Bagozzi & Philips, 1982, in Ariño, 2003).

Examination of this type of validity is important, since the correspondence of

results when a concept is measured in different ways is one of the most

convincing evidences of construct validity (Singleton, Straits & Miller Straits,

1993). Examination of the convergent validity of the Social Participation

Questionnaire will take place by comparing the outcomes of the Questionnaire

with the outcomes of other instruments that are also meant to measure social

participation. As there is no instrument which, like the Social Participation

Questionnaire, focuses on the total concept of social participation, several

instruments that aim at certain aspects of social participation were selected (see

instruments section). The Social Participation Questionnaire consists of four

subscales, representing the four important themes within the concept of social

CHAPTER 6

138

participation mentioned above: friendships/relationships, contacts/interactions,

social self-perception of the pupil and acceptance by classmates. In order to

examine the convergent validity of the Questionnaire, four instruments, each

focusing on one of these themes, were applied. Pupils’ scores on the

Questionnaire and on each of the four subscales were compared with their scores

on these instruments.

6.2.1 Respondents

The study took place in Grades 1 to 3 of regular Dutch primary schools that have

at least one pupil with an official special-needs indication. In order to examine

the Questionnaire’s convergent validity it was expected that each key theme

would have to be assessed by one questionnaire, which would result in four

questionnaires per pupil. The researchers considered filling in more than two

questionnaires too large a burden for pupils. Furthermore, when administering

four questionnaires in each class, the researchers would have to spend a long

time in classes. This would probably disturb the teachers’ schedules too much,

and may deter them from participating. It was therefore decided to divide the

study into two. By splitting the study, pupils had to fill in two questionnaires at

most, and time spent in the classroom was expected to be less than one hour. As

a result, disturbance of the daily routine would be minimized.

In total, 600 schools were invited to participate. The first sub-sample of 300

schools was used to compare the Questionnaire with instruments to measure

both the social self-perception of pupils and their acceptance by peers. The

second sub-sample of 300 schools was applied to compare the outcomes of the

Questionnaire with instruments to assess the friendships of pupils as well as their

contacts and interactions. The invited schools are a randomly drawn sample from

the population of 2074 Dutch regular primary schools within 2.5 hours of

traveling time from the city of Groningen. This area, covering nearly two-third of

the Netherlands, was chosen to make the data collection feasible. According to

files from the Ministry of Education, the schools have at least one pupil in Grades

1 to 3 who receives a pupil-bound budget. Such a budget is allocated to pupils

who, according to independent committees, meet certain Dutch national criteria.

Several categories of disabilities are distinguished, each with their own criteria.

These involve, among other things, categories of the Diagnostic and Statistical


139

Manual of Mental Disorders (DSM IV) and intellectual ability, assessed by

qualified psychiatrists or psychologists who operate independently from the

committees.

First sub-sample. Of the 300 invited schools, 53 were involved in the study. A

non-response survey (Koster, Timmerman, Nakken, Pijl & Van Houten, 2008)

showed that there are no significant differences (at α=5%) between cooperating

and non-cooperating schools concerning their attitudes towards and experiences

with inclusive education. Hence there is no reason to assume any systematic bias

in the sample involved in the study. Since several of the 53 participating schools

had a pupil with special needs in more than one class, and since 15 classes

included two or more pupils with special needs, 75 classes with a total of 96

pupils with special needs were involved in the study. The analyses in this article

aimed at social self-perception and acceptance by classmates are based on the

96 pupils with special needs and 148 pupils without special needs. For these 244

pupils, the teacher filled in the Social Participation Questionnaire. An overview of

the categories of disabilities and their distribution in the first sub-sample is

presented in the second and third columns of Table 1.

Second sub-sample. Of the 300 schools that were invited to participate in the

study, 66 took part. As no evidence of systematic bias was found in the former

sample, we refrained from repeating the non-response survey. In total, 27

schools had a pupil with special needs in more than one class and 34 classes

included two or more pupils with special needs. Consequently, 105 classes with a

total of 141 pupils with special needs participated in the study. The analyses in

this article that aim at friendships as well as contacts/interactions are based on

(a selection of) these 141 pupils with special needs and 205 pupils without

special needs. For these 346 pupils, the teacher filled in the Social Participation

Questionnaire. An overview of the categories of disabilities and its distribution in

the second sub-sample is presented in the fourth and fifth columns of Table 1.

CHAPTER 6

140


in both sub-samples

Category of

disabilities

# pupils

(% total)

first sub-

sample

# boys (%

disabilities

category)

first sub-

sample

# pupils

(% total)

second

sub-sample

# boys (%

disabilities

category)

second sub-

sample

Behavioural disorder 13 (13.5%) 11 (84.6%) 16 (11.3%) 14 (87.5%)

Autistic spectrum

disorder

42 (43.8%) 38 (90.5%) 55 (39.0%) 45 (81.8%)

Motor disability 10 (10.4%) 9 (90.0%) 25 (17.7%) 18 (72.0%)

Intellectual disability 11 (11.5%) 5 (45.5%) 15 (10.6%) 5 (33.3%)

Speech/language

disabilities

20 (20.8%) 13 (65.0%) 27 (19.1%) 19 (70.4%)

Learning disabilities 0 0 2 (1.4%) 2 (100%)

Chronic illness 0 0 1 (0.7%) 1 (100%)

Total 96 (100%) 76 (79.2%) 141 (100%) 104 (73.8%)

6.2.2 Instruments

Social Participation Questionnaire

The Social Participation Questionnaire consists of 24 statements, five of which

belong to the ‘Friendships/relationships’ subscale, nine belong to the

‘Contacts/interactions’ subscale, five to the ‘Pupil’s social self-perception’

subscale and five to the ‘Acceptance by classmates’ subscale. In total, 19

statements are phrased in terms of social participation and five are formulated in

terms of social segregation. Table 2 gives some examples of both types of

statements.


141

Table 2. Examples of statements of the Social Participation Questionnaire

Key theme Statement representing social

participation

Statement representing

social segregation

Friendships/relationships ‘the pupil is a member of a group of

friends’

Contacts/interactions ‘in free time (e.g. during recess) the

pupil plays with classmates’

‘classmates regularly

exclude the pupil from

activities’


perception

‘the pupil has the feeling s/he

belongs to the group’

‘the pupil feels lonely at

school’

Acceptance by classmates ‘classmates stand up for the pupil

when pupils from other classes or

schools (would) treat the pupil

unpleasantly’

The statements have to be assessed on a 5-point scale, ranging from ‘this does

not apply at all to the pupil’ (1) to ‘this strongly applies to the pupil’ (5). In order

to facilitate interpretation of the subscale scores, they were linearly rescaled so

that the minimum score on all subscales is 0 and the maximum 25. The total

score on the Questionnaire ranges between 0 and 100. The higher the score, the

higher the social participation of the pupil is assessed. In a previous study

(Koster et al., 2008) the quality of the subscales was examined and approved by

means of a Mokken Scale Analysis, which uses non-parametric item response.

For each of the subscales, the Mokken’s double monotonicity model appeared to

fit well, which implies that the resulting subscale scores are on an ordinal scale

and that the individual statements per subscale are invariantly ordered along the

subscale. The scale scores are directly comparable across pupils with and without

special needs, as differential item functioning appeared to be absent. Moreover,

the data supported the division of social participation into the four distinct key

themes. Although there is a clear distinction between the Questionnaire’s four

subscales, the correlations between them appeared to be strong (Spearman’s

Rho ranged from 0.54 to 0.77). Therefore, the concept of social participation

should be regarded as a macro construct consisting of four closely connected

sub-dimensions. Even though the subscales are strongly related, it is still

important to distinguish them, as the correlations are not perfect. If only the

score on the total Questionnaire were viewed, valuable differences in the key

CHAPTER 6

142

themes might remain unnoticed. If the subscale scores are also taken into

consideration, the expected result is a more appropriate and balanced view of

the social participation of pupils.

The reliability of the Questionnaire turned out to be high (Rho is 0.95), and the

four subscales were found to be reliable (Rho varies between 0.80 and 0.95). In

addition, one aspect of discriminant validity of the Questionnaire was

demonstrated. As pupils with special needs are generally known to have lower

levels of social functioning in regular classrooms, they were expected to obtain

on average lower scores on the four subscales and on the total Questionnaire

than pupils without special needs. Because these differences were clearly

present, evidence was obtained for the Questionnaire’s discriminant validity

(Koster et al., 2008).

Instruments to assess key themes

• Assessment of friendships

Many researchers (e.g. Asher & Hymel, 1981, in Asher, Parker & Walker, 1996;

Larrivee & Horne, 1991; Parker & Asher, 1993) regard reciprocal friendship

nomination as the best method to assess friendship. This method, which requires

children to name classmates who fit a particular sociometric criterion (Larrivee &

Horne, 1991), has been used to assess the friendship of children of various ages.

It has been adopted in the literature as the primary method for assessing

friendship (Yugar & Shapiro, 2001), and is likely to be valid for the whole range

of childhood and adolescent years (Bukowski & Hoza, 1989). As the nomination

method is generally regarded as the most suitable to assess friendship, it was

selected for our study.

• Assessment of contacts/interactions

Observation schemes are often used when assessing contacts and interactions

between children (e.g. Blatchford, Bassett & Brown, 2005; Cushing, Horner &

Barrier, 2003; Hall & McGregor, 2000; Murphy, 2004; Odom et al., 1999; Ohna,

2005; Roberts, Pratt & Leach, 1990). For our study, observation schemes

focusing on the nature and number of classroom interactions and which make

use of the time-sampling method were selected.

An exploration of the literature revealed three observation schemes

(Gresham, 1982; Scheepstra et al., 1999; Wood; 1972, in McCauley, Bruininks &


143

Kennedy, 1976) suitable for pupils in Grades 1 to 3, which aim at nature and

number of interactions and which do not take up a lot of time (maximum 30

minutes per pupil). It was chosen to mainly use Gresham’s Observation

Categories (Gresham, 1982), since the applied categories of social interaction are

very clear and the interobserver agreement is high (0.93 to 1.00 in a study of

Montague and Rinaldi, 2001; 0.95 to 1.00 in a study of Lago-Delello, 1998). In

addition, the categories provide an overview of the nature of interactions

(positive and negative), initiated and received interactions, and number of

interactions. Following the observation schemes of Scheepstra et al. (1999) and

Wood (1972, in McCauley et al., 1976), not only interactions between pupils but

also interactions between pupils and teachers are added, since research has

shown that pupils with special needs have many interactions with their teachers

(Scheepstra et al., 1999). Pupils are observed during lessons and free time,

given that when observing social behaviours it is important to collect

observations across a variety of settings and situations (Gresham, 2001;

Hamilton, 2005).

• Assessment of social self-perception

In order to select instruments to measure pupils’ social self-perception, several

instruments were compared for content, psychometric qualities, size, applicability

to pupils in Grades 1-3 and availability of a Dutch version. After a first selection

on the basis of these aspects, three instruments remained: the Self-Perception

Profile for Children (SPP-C, Harter, 1985, in Berndt & Burgy, 1996), the Self-

Description Questionnaire I (SDQI, Marsh, Parker & Smith, 1983, in Berndt &

Burgy, 1996) and the Pictorial Scale of Perceived Competence and Social

Acceptance for Young Children (PSPCSA, Harter & Pike, 1984). A comparison of

the social subscales of these instruments was made. Since the content, size and

reliability of the scales were rather similar (Berndt & Burgy, 1996), the

applicability for first-to-third graders and the availability of Dutch versions were

decisive. It turned out that none of the three instruments aimed at pupils in all

three grades. The SDQI focused on pupils in Grades 2-9, the SPP-C focused on

pupils in Grades 3-8 and the PSPCSA on first- and second-graders. Therefore, it

was necessary to select two instruments for our study. Because of the very

similar content, the same response format and the availability of Dutch versions,

it was decided to use the SPP-C and the PSPCSA. As a result, in Grade 3 the

CHAPTER 6

144

social subscale of the Dutch version of the SPP-C (in Dutch this profile is

abbreviated as CBSK) (Veerman, Straathof, Treffers, Van den Bergh & Ten Brink,

1997) was used. The reliability (Alpha) of the social subscale is 0.74 (Veerman et

al., 1997), which is sufficient when used for research purposes (Nunnally, 1967).

The social subscale of the Dutch version of the PSPCSA (Harter & Pike,

1984; Van Rossum & Vermeer, 1992) was selected for pupils in Grades 1 and 2.

In this scale, verbal items are supplemented by pictures. The reliability of the

social subscale is sufficient (α=0.78) (Van Rossum & Vermeer, 1992).

• Assessment of acceptance by classmates

Peer acceptance is often assessed with sociometric techniques (Berndt & Burgy,

1996). Both procedures of nomination (see Assessment of friendships) and

procedures of peer rating are frequently used. In the procedure of peer ratings,

all classmates rate each of their peers in terms of how much they like them or

would like to play with them on a Likert-type scale. Usually, a 3- or 5-point scale

is used (Hymel, Vaillancourt, McDougal & Renshaw, 2002; Jiang & Cillessen,

2005; Terry & Coie, 1991). There seems to be general agreement that using a

rating scale is the best method to assess acceptance, while the nomination

method is most suitable for the assessment of friendship (Asher & Hymel, 1981,

in Asher et al., 1996; Larrivee & Horne, 1991; Parker & Asher, 1993). According

to Asher et al. (1996), research suggests the utility of distinguishing acceptance,

based on the average ratings children receive from friendship, based on

reciprocal friendship choices. The evidence consistently suggests that acceptance

and friendship are non-overlapping yet not fully independent dimensions of

individual differences (Asher et al., 1996). In accordance with the above-

mentioned results, in our study the peer rating method was selected as

instrument to assess acceptance.

6.2.3 Procedure

Social Participation Questionnaire. The teachers of the 180 classes (75 classes in

the first sub-sample, 105 in the second) filled in the Social Participation

Questionnaire for the pupils with special needs and for two pupils with typical

development. The latter were chosen at random by using the 3rd and 16th pupil

in the class register. If the pupil with special needs was number 3 or 16 on the


145

register, then the following pupil on the list was chosen. Also, if pupils number 3

and 16 were of the same sex, the first following pupil of the opposite sex on the

list was chosen. In classrooms with two or more pupils with special needs, the

teacher completed the Questionnaire for all pupils with special needs and for two

typical pupils.

Assessment of friendships (nomination). All pupils in the 105 classes of the

second sub-sample were asked to fill in a sociometric questionnaire. They were

asked which classmates they considered to be their best friends. Following

Frostad and Pijl (2007), the pupils were allowed to give a maximum of five

nominations.

Assessment of contacts/interactions (observation schedule). For practical

reasons, observations took place only in some of the classes of the second sub-

sample. In 59 classes, observations were carried out during lessons and during

free time. In each class, both the pupil with special needs and a pupil of the

same sex without special needs (for whom the teacher filled in the Social

Participation Questionnaire) were observed for 20 minutes, divided into periods

of five minutes. Fifteen minutes of the observation took place during lessons, the

remaining five minutes took place during free time. Each five-minute period was

divided into 30 ten-second intervals. If a interaction occurred during that period,

a tick was noted in the correct category (positive/neutral/negative initiated

interaction with classmate; positive/neutral/negative received interaction with

classmate; initiated interaction with teacher/received interaction with teacher). If

more than one interaction occurred in a period, only the first one was noted. The

observers were five university students and the first author, who initially

received three hours’ training using videotaped recordings of a classroom

situation. During the training sessions, it turned out that almost all interactions

were coded as neutral: both positive and negative interactions were rarely

coded. Therefore, it was decided to leave the nature of interactions aside. After

training, the agreement between observers was determined by calculating

Cohen’s kappa for three major aspects: ‘interaction/no interaction between pupil

and classmates’, ‘initiated/received interaction of pupil with classmates’ and

‘interaction/no interaction between pupil and teacher’. For these aspects, Cohen’s

kappas were respectively 0.84, 0.76 and 0.72, suggesting reasonable

agreement.

CHAPTER 6

146

Assessment of social self-perception (CBSK / Pictorial Scale). Social self-

perception of pupils was assessed in the 75 classes of the first sub-sample. In

Grade 3, the social subscale of the CBSK was administered as a group test, while

the social subscale of the Pictorial Scale in Grades 1 and 2 was administered

individually. On the basis of the scores on the social subscale of the

CBSK/Pictorial Scale, a self-perception score was calculated for each pupil. The

minimum score of the CBSK (consisting of six questions) was 6, the maximum

score 24. The score on the Pictorial Scale (consisting of five questions) could

vary between 5 and 20. For both scales it applies that, the higher the score, the

higher the social self-perception.

Assessment of acceptance by classmates (peer rating). All pupils in the 75

classes of the first sub-sample were asked to fill in a rating scale containing the

names of all their classmates. They were asked to indicate on a 3-point scale to

what degree they would like to play with each classmate. They could choose

between the following three answering categories, each visually supported by

smileys: 1) yes, I would like to ☺, 2) I don’t care �, and 3) no, I would not like

to �.

In Grades 2 and 3 the rating scale was administered as a group test, while

in Grade 1 the scale was administered individually: the researcher read aloud the

names of all classmates and the pupil mentioned how much s/he liked to play

with each of them. Table 3 shows which instruments to assess the four key

themes were applied to what number of pupils (with and without special needs).

Table 3. Instruments for assessment of key themes related to number of pupils

Subscale Instrument # pupils with

special needs

# pupils

without special

needs

Friendships/relationships Sociometric nomination

method

140 202


Observation schedule 59 59


perception

Social subscale of

CBSK/Pictorial Scale

27, 67 42, 99

Acceptance by

classmates

Sociometric Rating scale 96 148


147

6.2.4 Analysis

The Questionnaire’s convergent validity was assessed by comparing the scores of

pupils on the four subscales with their scores on the four instruments, which

successively assess friendships, contacts/interactions, social self-perception and

acceptance. Two levels of expectations on convergent validity can be

distinguished. First, specific correlation patterns were expected. It was supposed

that pupils’ scores on the instrument to assess one of the key themes would

correlate the strongest with the subscale representing that key theme. For

instance, pupils’ number of friendships were expected to correlate more strongly

with the ‘Friendships/relationships’ subscale than with the other three subscales.

Secondly, correlation coefficients which meet or exceed .35 were regarded as

another source of evidence of convergent validity (Hammill, Brown & Bryant,

1989, in Trout, Ryan, La Vigne, & Epstein, 2003).

Friendships. The sociometric data resulting from the reciprocal friendship-

nomination method were analysed using the UCINET software (Borgatti, Everett

& Freeman, 1999) to identify friendships. Friendship was defined as a reciprocal

choice, implying that two pupils choose each other as each other’s best friend

(Frostad & Pijl, 2007). The number of friendships was expected to correlate

positively with the score on the ‘Friendships/relationships’ subscale and with the

score on the total Questionnaire. It was decided to use the absolute number of

friendships (instead of standardized Z-scores) in calculating correlations with the

Social Participation Questionnaire, as class size had only minimal influence on a

pupil’s number of friendships.

Interactions. For each pupil the total number of (both initiated and received)

interactions with fellow classmates and the total number of interactions with the

teacher were calculated. The number of interactions with fellow classmates was

expected to show a positive correlation with the ‘Contacts/interactions’ subscale

and with the total Questionnaire. Conversely, a pupil’s number of interactions

with the teacher was expected to correlate negatively with his/her scores on the

‘Contacts/interactions’ subscale and on the total Questionnaire, since having

interactions with the teacher might be at the expense of interactions with

classmates.

Social self-perception. For each pupil a social self-perception score was

calculated on the basis of the outcomes of the CBSK (Grade 3)/Pictorial scale

(Grades 1 and 2). The raw scores of both scales were used for this purpose. For

CHAPTER 6

148

pupils in Grade 3 the score could range between 6 and 24, for pupils in Grades 1

and 2 the score could vary from 5 to 20. The social self-perception score of

pupils was expected to correlate positively with their score on the ‘Pupil’s social

self-perception’ subscale and with their score on the total Questionnaire.

Acceptance. Each pupil received points from all classmates, since all classmates

had indicated on a 3-point scale (see Procedure section) to what degree they

would like to play with the pupil (☺= 1 point, �= 0 points, �= -1 point).

Counting up all points resulted in a raw score for each pupil. As the number of

points a pupil could receive was strongly related to class size, Z-scores were

calculated. This was done by subtracting the mean score of the class from the

pupil’s raw score, then dividing the difference by the standard deviation of the

class. By means of these Z-scores, a correction for class size was made. As a

result, scores of pupils could not only be compared with their classmates’ scores,

but also with scores of pupils from other classes (comprising various numbers of

pupils). The Z-score of acceptance was expected to correlate positively with the

‘Acceptance by classmates’ subscale and with the total Questionnaire.

Finally, confirmatory factor analyses using LISREL software (LISREL 8.80,

Jöreskog & Sörbom, 2007) were conducted in order to further evaluate our

model of social participation. These analyses are a more comprehensive

approach towards assessing the Questionnaire’s convergent validity, as it is a

multi-trait, multi-method assessment. Moreover, these analyses were expected

to provide insight into the degree to which the data fit our model of social

participation (see Introduction section). The Maximum Likelihood (ML) estimation

algorithm was used to verify the conceptual structure of the data.

Four indexes were used to evaluate model fit: the ratio of chi-square to degrees

of freedom is given as a first indication of model fit. This ratio (χ2/df) is, in

comparison with chi-square or the P-statistic, less sensitive to group sizes and

departures from normality (Rozendaal, Minnaert & Boekaerts, 2003). Following

Byrne (1989) and Browne and Cudeck (1993, in Vermunt & Minnaert, 2003), a

χ2/df ratio equaling or below 2 is considered as a good fit. Next to the χ2/df ratio,

the following three indexes were used to evaluate model fit: the root mean

square error of approximation (RMSEA), the goodness of fit index (GFI) and the

incremental fit index (IFI). The latter is based on comparison of the fit of a target

model with that of a null model, and aims at quantifying the proportional

improvement in fit for a target model relative to a null model (Schmukle & Hardt,


149

2005). According to Jaccard and Wan (1996, in Vellutino, Tunmer, Jaccard &

Chen, 2007), statistically non-significant RMSEAs that are no greater than 0.08

suggest acceptable model fit, as do GFIs ≥0.90. Finally, IFIs ≥0.90 indicate an

acceptable fit with the data (Bollen & Curran, 2006).

6.3 Results

6.3.1 Convergent validity of the four subscales

As anticipated, the number of friendships, the number of interactions with

classmates, the self-perception score and the acceptance score correlated

positively with the score on the total Questionnaire and with the score on the

subscales, whereas the number of interactions with the teacher showed a

negative correlation with the scores on the total Questionnaire and the subscales.

As described in the Analysis section, specific correlation patterns were regarded

as evidence for the Questionnaire’s convergent validity. It was supposed that

pupils’ scores on the instrument to assess one of the key themes would correlate

the strongest with the subscale representing that key theme. For the

‘Friendships/relationships’ subscale the expected correlation pattern was

demonstrated, as the number of friendships showed a stronger correlation with

the ‘Friendships/relationships’ subscale than with the other three subscales. For

the ‘Contacts/interactions’ subscale the expected correlation patterns were

mainly demonstrated: the number of both initiated and received interactions with

classmates showed a stronger correlation with the ‘Contacts/interactions’

subscale than with the other subscales. However, the number of interactions with

the teacher showed a slightly stronger correlation with the

‘Friendships/relationships’ subscale than with the ‘Contacts/interactions’

subscale. For the other two subscales, the correlation patterns deviated from the

expectations. The expectation that the self-perception score would show a

stronger correlation with the ‘Pupil’s social self-perception’ subscale than with the

three other subscales could not be confirmed. The expected pattern could not be

demonstrated, neither for pupils in Grade 3 nor for those in Grades 1 and 2. The

same applies for the ‘Acceptance by classmates’ subscale: contrary to the

expectations, the acceptance score showed a stronger correlation with the

CHAPTER 6

150

‘Friendships/relationships’ and ‘Contacts/interactions’ subscales than with the

‘Acceptance by classmates’ subscale.

After having examined the correlation patterns, the strength of correlations was

inspected. As explained in the Analysis section, correlation coefficients should

meet or exceed 0.35 in order to be regarded as evidence for convergent validity.

It turned out that only for the ‘Friendships/relationships’ and ‘Acceptance by

classmates’ subscales were the correlation coefficients equal to or above 0.35.

For the other two subscales, the correlations coefficients were below this value.

Table 4 provides a detailed overview of the correlations for each of the four

subscales.

Table 4. Correlation between (subscales of) Social Participation Questionnaire and scores

on instruments to assess the four key themes

Interactions Subscale Number of friend-ships

Initiated inter-actions with classmates

Received inter-actions with classmates

Inter-actions with teacher

Self-perception score

Accep- tance score

Friendships / relationships

0.50** 0.16 0.22* -0.29** Grade 1/2: 0.33** Grade 3: 0.29*

0.50**

Contacts / interactions

0.45** 0.20* 0.27** -0.28** Grade 1/2: 0.27** Grade 3: 0.29*

0.58**

Pupil’s social self-perception

0.32** 0.11 0.20* -0.14 Grade 1/2: 0.22** Grade 3: 0.25*

0.35**


0.26** 0.04 0.09 -0.14 Grade 1/2: 0.17* Grade 3: 0.12

0.35**

Total Questionnaire

0.46** 0.15 0.23* -0.25** Grade 1/2: 0.29** Grade 3: 0.30*

0.54**

* Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level

6.3.2 Second order analysis

Introduction. Inspection of the correlation matrix (Table 4) reveals that evidence

for the Questionnaire’s convergent validity is rather spurious. A confirmatory

factor analysis using the LISREL program was conducted in order to attempt

removing the lack of clarity. The data from the correlation matrix were used as

input for this second order analysis.


151

Model specification. The sociometric nomination method and the

‘Friendships/relationships’ subscale are supposed to be indicators of the latent

factor ‘Friendship’ (ξ1), the observation schedule and the ‘Contacts/interactions’

subscale are indicators of the latent factor ‘Contact’ (ξ2), the social subscale of

the CBSK/Pictorial Scale and the ‘Pupil’s social-self-perception’ subscale are

expected to be indicators of the latent factor ‘Social-self-perception’ (ξ3) and,

finally, the sociometric rating scale and the ‘Acceptance by classmates’ subscale

are supposed to be indicators of the latent factor ‘Acceptance’ (ξ4).

Two separate analyses were run for both sub-samples (see Participants section).

In order to run the analyses, the sub-samples were split into Grades 1/2 and

Grade 3, as the instrument to assess pupils’ social self-perception differed for the

pupils from these different grade groups.

Results. Inspection of the fit indexes concerning ξ1 (Friendship) and ξ2 (Contact)

called up the necessity to release the method factor regarding initiated and

received interactions with classmates. As both methods are observation scales

focusing on interactions between pupils, the scales have some common variance.

Hence it is obvious to release this method factor. As presented in the upper part

of Table 5, there is a significant gain due to releasing the method factor. A

significant change in ∆χ2/df is realised and the fit indexes (GFI, RMSEA, P of

close fit) meet the aforementioned criteria (see Analysis section). Also, the IFI is

acceptable after release of the method factor (0.96 for Grades 1/2, 0.99 for

Grade 3). Because of this major improvement of fit, the release of the method

factor can be regarded as valid. After having released the method factor for both

Grades 1/2 and Grade 3, the confirmatory factor analysis to verify the conceptual

structure of the data on item level indicated a good fit.

The results on ξ3 (Social-self-perception) and ξ4 (Acceptance) are

presented in the lower part of Table 5. As can be seen, for pupils in Grades 1/2

the confirmatory factor analysis indicates a good fit for ξ3 and ξ4. However, for

pupils in Grade 3 the model fits the data less well.

CHAPTER 6

152

Table 5. Goodness-of-fit indices for competing models for ξ1 ‘Friendship’, ξ2 ‘Contact’, ξ3

‘Social self-perception’ and ξ4 ‘Acceptance’

Competing models χ2 df χ2/df GFI RMSEA P of close fit

ξ1 ξ2 Grades 1/2 Basic model

527.48

14

37.68

0.35

1.37

0.00

Release of method factor

15.54 10 1.55 0.94 0.08 0.22

Grade 3 Basic model

320.89

14

22.92

0.26

1.75

0.00

Release of method factor

11.30 10 1.13 0.91 0.06 0.42

ξ3 ξ4 Grades 1/2 Basic model

6.44

4

1.61

0.98

0.06

0.36

Grade 3 Basic model

10.28

5

2.06

0.93

0.13

0.11

Table 6 shows the results of the maximum likelihood estimations of ξ1 and ξ2 for

pupils in Grades 1/2 and Grade 3. With regard to ξ2 the correlation patterns are

notable, as in Grade 3 the maximum likelihood estimation shows that the

interactions with classmates present no significant correlations with this latent

factor; by contrast, in Grades 1/2 the correlations between interactions with

classmates and ξ2 are evidently significant. For the interaction with the teachers,

the opposite is true: in Grade 3 the interaction with the teacher shows a

significant correlation with ξ2 and in Grades 1/2 interaction with the teacher only

minimally correlates (significant at the 0.10 level) with this latent factor.

Table 6. ML solutions of the measurement model for ξ1 and ξ2

λ

Grades 1/2 Grade 3

ξ1

Friendship

ξ2

Contact

ξ1

Friendship

ξ2

Contact

Subscale ‘Friendships/relationships’ 0.95** 0.95**

Number of friends 0.48** 0.64**

Subscale ‘Contacts/interactions’ 0.95** 0.95**

Initiated interactions with classmates 0.37** -0.11

Received interactions with classmates 0.40** 0.11

Interactions with teacher -0.22o -0.44**

o Correlation is significant at the 0.10 level * Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level


153

The results of the maximum likelihood estimations of ξ3 and ξ4 for both groups

of pupils are presented in Table 7. As can be seen in the Table, the correlation

patterns of Grades 1/2 and Grade 3 are very similar.

Table 7. ML solutions of the measurement model for ξ3 and ξ4

λ

Grades 1/2 Grade 3

ξ3

Social

self-

perception

ξ4

Acceptance

ξ3

Social

self-

perception

ξ4

Acceptance

Subscale ‘Pupil’s social self-

perception’

0.95** 0.95**

Social self-perception score 0.24** 0.27*

Subscale ‘Acceptance by classmates’ 0.95** 0.89**

Acceptance score 0.41** 0.36**

* Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level

The correlations between the latent factors are strong, implying that the

constructs are closely connected. For Grades 1/2, the disattenuated correlation

between ξ1 (Friendship) and ξ2 (Contact) is 0.83, for Grade 3 it is 0.84. The

correlations between ξ3 (Social-self-perception) and ξ4 (Acceptance) are slightly

weaker, as for Grades 1/2 the disattenuated correlation is 0.60 and for Grade 3

this correlation is 0.63. The disattenuated correlations are not perfect, indicating

that the constructs are not wholly overlapping. However, the intercorrelatedness

is high.

6.4 Discussion

In this paper, the convergent validity of a new teacher questionnaire to assess

the social participation of pupils with special needs in regular primary schools

was examined. Pupils’ scores on the total Social Participation Questionnaire and

on each of the four subscales were compared with their scores on four other

instruments, each focusing on one of the key themes of social participation.

Inspection of a correlation matrix, consisting of correlations between the

Questionnaire’s subscales and other instruments to assess the four key themes,

revealed that evidence for the Questionnaire’s convergent validity was rather

CHAPTER 6

154

spurious. Only the convergent validity of the ‘Friendships/relationships’ subscale

was satisfactory, as both the strength of the correlation (>0.35) and the

correlation pattern were in accordance with expectations. For the

‘Contacts/interactions’ and ‘Acceptance by classmates’ subscales, the convergent

validity was only proven partly. For the former subscale the expected correlation

patterns were mainly found, but the correlation coefficients were below 0.35,

while for the latter subscale the opposite was the case. For the ‘Pupil’s social self-

perception’ subscale, no evidence for its convergent validity was demonstrated,

as the correlation coefficients were below 0.35 and the expected patterns were

not demonstrated.

The relatively low correlations between the Social Participation Questionnaire and

the instruments to assess the four key themes of social participation are not

surprising, as several categories of respondents were involved. Whereas the

Social Participation Questionnaire was filled in by teachers, pupils and observers

were involved in the assessment of the four key themes. It is known from

various studies that assessments of different informants often provide discrepant

outcomes (Achenbach, 2006; Bartels et al., 2003). For instance, a meta-analysis

by Achenbach, McConaughy and Howell (1987) showed only limited correlation

between maternal and paternal ratings of problem behaviour of the same child.

As in our study teachers, pupils and observers were involved as respondents,

their varying perspectives might have negatively influenced the correlations

between their assessments. Moreover, the various assessment instruments

applied by the different types of respondents might have reduced the

correlations.

Because of the partly low correlations between the different assessment

instruments and the Social Participation Questionnaire, and due to the correlation

patterns which partly deviated from the expectations, the outcomes on the

Questionnaire’s convergent validity were rather inconclusive. However, a second

order analysis, a confirmatory factor analysis using the LISREL program, did

provide evidence for the Questionnaire’s convergent validity. The results from

this second order analysis provide support for the model of social participation,

distinguishing four key themes. The fit indexes for ξ1 (Friendship) and ξ2

(Contact) were satisfactory in all grades. The fit indexes for ξ3 (Social-self

perception) and ξ4 (Acceptance) were acceptable for Grades 1/2, but for Grade 3


155

the model fitted the data less well. The fit indexes suggest that our model is

theoretically viable for Grades 1/2 and only partly viable for Grade 3.

With regard to ξ2 (Contact), an unknown factor seems to play a role. In Grade 3

the maximum likelihood estimation shows that the interactions with classmates

show no significant correlation with the latent factor Contact, whereas in Grades

1/2 the interaction with the teacher shows only a minimal correlation (significant

at the 0.10 level) with this latent factor. One possible explanation for this

correlation pattern might have to do with the fact that in the higher grades the

focus lies increasingly on academic learning and less on play. Pupils with special

needs may be ever more dependent on the teacher, as the gap between them

and their typical classmates becomes larger. As a result, pupils with special

needs in Grade 3 might be more focused on the teacher instead of associating

with classmates. Another possible explanation concerns the different focus of the

‘Contacts/interactions’ subscale versus the observation schedule. Whereas the

subscale encompasses both the quality and the quantity of interactions, the

observation schedule aims solely at quantity of interactions. This difference

might have influenced the correlation between the observations and the latent

factor ‘Contact’.

In further examining the Social Participation Questionnaire, investigating its

utility should be a next step — for instance, examining the meaning of the scores

on the Questionnaire is advisable, as it would be valuable for teachers to know

which scores on the Questionnaire are cause for concern. Future research should

aim at establishing a cut-off point beyond which a score is alarming and

additional measures need to be taken. In addition, further research into the

convergent validity of the subscales ‘Social self-perception of pupil’ and

‘Acceptance by classmates’ for pupils in Grade 3 is recommended.

Taking into consideration the outcomes of this study, it can be concluded that the

Social Participation Questionnaire is partly proven to be a tool of good quality for

teachers. The Questionnaire seems to have the potential to become valuable for

them. Since the Questionnaire consists of concrete statements focusing on daily

practice, this is expected to foster the accuracy of teachers’ judgments. Besides,

by filling in the Questionnaire for a pupil, the teacher is forced to think critically

about the pupil’s situation in the class, which might lead to renewed and refined

insights. As shown in this study and in a former study (Koster et al., 2008), the

Questionnaire as a whole is reliable and valid. In addition, the distinction into

CHAPTER 6

156

four subscales was proven to be correct, indicating that, although overlapping to

a reasonable extent, within social participation four key themes can be

distinguished. Aside from the total score, it seems important to take into

consideration the scores on the four subscales in order to gain further insight into

the different themes of social participation. Such a perspective is expected to

result in a more appropriate and balanced view of the social participation of

pupils, offering starting points for differentiated interventions.

6.5 References

Achenbach, T.M. (2006). As others see us. Clinical and research implications of

cross-informant correlations for psychopathology. Current Directions in

Psychological Science, 15(2), 94-98.

Achenbach, T.M., McConaughy, S.H. & Howell, C.T. (1987). Child/Adolescent

behavioral and emotional problems: implications of cross-informant

correlations for situational specificity. Psychological Bulletin, 101(2), 213-

232.

Ariño, A. (2003). Measures of strategic alliance performance: an analysis of

construct validity. Journal of International Business Studies, 34, 66-79.

Asher, S.R., Parker, J.G. & Walker, D.L. (1996). Distinguishing friendship from

acceptance: implications for intervention and assessment. In W.M.

Bukowski, A.F. Newcomb & W.W. Hartup. (Eds.), The company they keep.

Friendship in childhood and adolescence (pp. 366-405). Cambridge, New

York, Melbourne: Cambridge University Press.



69(1), 140-153.

Bartels, M., Hudziak, J.J., Van den Oord, E.J.C.G., Van Beijsterveldt, C.E.M.,

Rietveld, M.J.H. & Boomsma, D.I. (2003). Co-occurence of aggressive

behavior and rule-breaking behavior at age 12: Multi-rater analyses.

Behavior Genetics, 33(5), 607-621.

Bender, W.N. & Wall, M.E. (1994). Social-emotional development of students

with learning disabilities. Learning Disability Quarterly, 17, 323-341.


157

Berndt, T.J. & Burgy, L. (1996). Social Self-Concept. In B.A. Bracken (Ed.),

Handbook of Self-Concept (pp. 171-209). New York: John Wiley & Sons,

Inc.

Blatchford, P., Bassett, P. & Brown, P. (2005). Teachers’ and pupils’ behavior in

large and small classes: a systematic observation study of pupils aged 10

and 11 years. Journal of Educational Psychology, 97(3), 454-467.

Bollen, K.A. & Curran, P.J. (2006). Latent Curve Models. A structural equation

perspective. New Jersey: John Wiley & Sons, Inc.



Bukowski, W.M. & Hoza, B. (1989). Popularity and friendship: issues in theory,

measurement, and outcome. In T.J. Berndt & G.W. Ladd (Eds.), Peer

relationships in child development (pp. 15-45). New York: John Wiley &

Sons.

Byrne, B.M. (1989). Multigroup comparisons and the assumption of equivalent

construct validity across groups: methodological and substantive issues.

Multivariate Behavioral Research, 24(4), 503-523.

Cambra, C. & Silvestre, N. (2003). Students with special educational needs in the

inclusive classroom: social integration and self-concept. European Journal


Cushing, L.S., Horner, R.H. & Barrier, H. (2003). Validation and congruent

validity of a direct observation tool to assess student social climate.

Journal of Positive Behavior Interventions, 5(4), 225-237.










Education, 22(1), 15-30.

Gresham, F.M. (2001). Social skills instruction for exceptional children. Theory

into Practice, 11(2), 129-133.

CHAPTER 6

158

Gresham, F.M. (1982). Social interactions as predictors of children’s likeability

and friendship patterns: a multiple regression analysis. Journal of

Behavioral Assessment, 4(1), 39-54.

Gresham, F.M. & McMillan, D.L. (1997). Social competence and affective

characteristics of students with mild disabilities. Review of Educational

Research, 67(4), 377-415.



Education, 34(3), 114-126.

Hamilton, D. (2005). An ecobehavioural analysis of interactive engagement of

children with developmental disabilities with their peers in inclusive

preschools. International Journal of Disability, Development and

Education, 52(2), 121-137.

Harter, S. & Pike, R. (1984). The Pictorial Scale of Perceived Competence and

Social Acceptance for Young Children. Child Development, 55, 1969-1982.

Heiman, M. & Margalit, M. (1998). Loneliness, depression, and social skills

among students with mild mental retardation in different educational

settings. The Journal of Special Education, 32(3), 154-163.

Hymel, S., Vaillancourt, T., McDougal, P. & Renshaw, P.D. (2002). Peer

acceptance and rejection in childhood. In P.K. Smith & C.H. Hart (Eds.),

Blackwell Handbook of Childhood Social Development (pp. 265-284).

Malden, Oxford, Carlton: Blackwell Publishing.

Jiang, X.L., & Cillessen, A.H.N. (2005). Stability of continuous measures of

sociometric status: a meta-analysis. Developmental Review, 25(1), 1-25.

Jöreskog, K.G. & Sörbom, D. (2007). Lisrel 8.80 for Windows (Computer

software) Lincolnwood, IL: Scientific Software International, Inc.

Kerlinger, F.N. & Lee, H.B. (2000). Foundations of behavioral research (4th ed.).

Orlando: Harcourt College Publishers.


peer group: a literature study focusing on the social dimension of inclusion

in education. International Journal of Inclusive Education.

Koster, M., Nakken, H., Pijl, S.J, Van Houten, E.J. & Lutje Spelberg, H.C. (in

press). Assessing social participation of pupils with special needs in

inclusive education: the construction of a teacher questionnaire.

Educational Research and Evaluation.


159

Koster, M., Timmerman, M.E., Nakken, H., Pijl, S.J., & Van Houten, E.J. (2008).

Evaluating social participation of pupils with special needs in regular

primary schools: examination of a teacher questionnaire. Manuscript

submitted for publication.

Lackaye, T.D. & Margalit, M. (2006). Comparisons of achievement, effort and

self-perceptions among students with learning disabilities and their peers

from different achievement groups. Journal of Learning Disabilities, 39(5),

432-446.




Lago-Delello, E. (1998). Classroom dynamics and development of serious

emotional disturbance. Exceptional Children, 64(4), 479-492.


students with peers of different ability levels, The Journal of Special

Education, 25(1), 90-101.

Luftig, R.L. (1988). Assessment of the perceived school loneliness and isolation

of mentally retarded and nonretarded students. American Journal of



people who experience difficulties in learning. In L. Florian (Ed.), The

SAGE handbook of Special Education (pp. 460-474). London: SAGE

Publications.

McCauley, R.W., Bruininks, R.H. & Kennedy, P. (1976). Behavioral interactions of

hearing impaired children in regular classrooms. The Journal of Special

Education, 10(3), 277-284.

Montague, M. & Rinaldi, C. (2001). Classroom dynamics and children at risk: a

followup. Learning Disability Quarterly, 24(2), 75-83.

Murphy, B. (2004). Social interaction and language use in Irish infant classrooms

in the context of the revised Irish Primary School Curriculum (1999).

Literacy, 38(3), 149-155.




CHAPTER 6

160





Nunnally, J.C. (1967). Psychometric Theory. New York: McGraw-Hill Book

Company.

Odom, S.L., McConnell, S.R., McEvoy, M.A., Peterson, C., Ostrosky, M.,

Chandler, L.K., Spicuzza, R.J., Skellenger, A., Creighton, M. & Favazza,

P.C. (1999). Relative effects of interventions supporting the social

competence of young children with disabilities. Topics in Early Childhood


Ohna, S.E. (2005). Researching classroom processes of inclusion and exclusion.







Pavri, S. & Monda-Amaya, L. (2000). Loneliness and students with learning

disabilities in inclusive classrooms: self-perceptions, coping strategies, and

preferred interventions. Learning Disabilities Research & Practice, 15(1),

22-33.

Pijl, S.J. & Frostad, P. (2008). Self-concept and social position of students with

special needs in regular education. Manuscript in preparation.

Pijl, S.J., Skaalvik. E.M. & Skaalvik, S. (2008). Students with special needs and

the composition of their peer group. Manuscript in preparation.

Powless, D.L. & Elliott, S.N. (1993). Assessment of social skills of native

American preschoolers: teachers’ and parents’ ratings. Journal of Social


Roberts, C., Pratt, C. & Leach, D. (1990). Classroom and playground interaction

of students with and without disabilities. Exceptional Children, 57(3), 220-

233.

Rozendaal, J.S., Minnaert, A. & Boekaerts, M. (2003). Motivation and self-

regulated learning in secondary vocational education: information-


161

processing type and gender differences. Learning and Individual

Differences, 13, 273-289.

Scheepstra, A.J.M., Nakken, H. & Pijl, S.J. (1999). Contacts with classmates: the

social position of pupils with Down’s syndrome in Dutch mainstream


Schmukle, S.C. & Hardt, J. (2005). A cautionary note on incremental fit indices

reported by LISREL. Methodology, 1(2), 81-85.

Singleton, R.A., Straits, B.C., & Miller Straits, M. (1993). Approaches to social

research (2nd ed.). New York: Oxford University Press.



Trout, A.L., Ryan, J.B., La Vigne, S.P. & Epstein, M.H. (2003). Behavioral and

emotional rating scale: two studies of convergent validity. Journal of Child

and Family Studies, 12(4), 339-410.

Van Rossum, J.H.A. & Vermeer, A. (1992). Het meten van waargenomen

competentie bij basisschoolkinderen door middel van een platentest.

Bewegen & Hulpverlening, 9(3), 198-212.

Veerman, J.W., Straathof, M.A.E., Treffers, P.D.A., Van den Bergh, B.R.H. & Ten

Brink, L.T. (1997). Handleiding Competentie Belevingsschaal voor

Kinderen. Lisse: Swets & Zeitlinger.

Vellutino, F.R., Tunmer, W.E., Jaccard, J.J. & Chen, R. (2007). Component of

reading ability: multivariate evidence for a convergent skills model of

reading development. Scientific Studies of Reading, 11(1), 3-32.

Vermunt, J. & Minnaert, A. (2003). Dissonance in student learning patterns:

when to revise theory? Studies in Higher Education, 28(1), 49-61.

Whitney, I., Smith, P.K. & Thompson, D. (1994). Bullying and children with

special educational needs. In P.K Smith & S. Sharp (Eds.), School Bullying:

insights and perspectives (pp. 213-240). New York: Routledge.

Williams, G.A. & Asher, S.R. (1992). Assessment of loneliness at school among

children with mild mental retardation. American Journal on Mental

Retardation, 96(4), 373-385.

Yugar, J.M. & Shapiro, E.S. (2001). Elementary children’s school friendships: a

comparison of peer assessment methodologies. School Psychology Review,

30(4), 568-585.

CHAPTER 6

162

Chapter 7 Social participation of pupils with special needs in regular Dutch primary education

CHAPTER 7

SOCIAL PARTICIPATION OF PUPILS WITH SPECIAL NEEDS IN REGULAR DUTCH PRIMARY EDUCATION

CHAPTER 7

164

Abstract

Consistently with the international tendency towards inclusive education,

inclusion of pupils with special needs is increasingly promoted in the Netherlands.

This study addresses the state of affairs with regard to the social participation of

pupils with special needs in regular Dutch primary classrooms. More specifically,

the focus lies on four key themes within social participation: pupils with special

needs’ friendships/relationships, contacts/interactions, social self-perception, and


The outcomes of the study revealed that on average, pupils with special needs

have a significantly lower number of friends and are less often members of a

cohesive subgroup compared to their typical peers. In addition, pupils with

special needs have less interactions with classmates, have more interactions with

the teacher and are less accepted than pupils without special needs. The social

self-perception of both groups of pupils does not differ. A comparison between

pupils with different categories of disabilities revealed no significant differences.


165

7.1 Introduction

Educating pupils with special needs in inclusive classrooms is an important

objective of many countries. In the Netherlands too, attempts are being made to

include children with special needs in regular schools. In 1990, a new

government policy document, Weer Samen Naar School (WSNS, Together to

School Again), was intended to stimulate inclusion of pupils with special needs in

regular education (Pijl & Van den Bos, 2001). Under this policy, all primary

schools and special schools for children who had been diagnosed as having

learning and educational problems or as having mild intellectual disabilities were

grouped into regional clusters. As a result, regular and special schools in the

clusters started to collaborate (Pijl & Van den Bos, 2001). The WSNS policy aims

at pupils with relatively mild special needs. With the introduction of the Wet op

de Expertise Centra (Centres of Expertise Act) in August 2003, the inclusion of

pupils with more complex special needs was stimulated as since then, parents of

children with special needs have the right to choose between regular and special

education for their child. Pupils with auditory, speech-language, motor,


and/or psychiatric problems can attend a regular school. This is funded with a

pupil-bound budget (financial ‘backpack’). Pupils who receive such a budget have

been officially labelled as having special needs on the basis of formal

comprehensive assessment procedures. When such pupils attend a regular

school they receive the budget, which caters for additional educational staff and

teaching aids.

Recent data show that a growing number of parents have opted for regular

education for their child with special needs, as the number of pupils with a pupil-

bound budget attending regular education is increasing rapidly (De Greef & Van

Rijswijk, 2006). The main motive of parents for sending their child with special

needs to a regular school are their child’s increased social opportunities

(Scheepstra, 1998, in Nakken & Pijl, 2002; Sloper & Tyler, 1992; Strayhorn &

Strain, 1986). They hope their child can build positive relationships with typically

developing peers.

However, international studies have repeatedly shown that including pupils with

special needs does not automatically lead to an increase of friendships between

pupils with special needs and their typical counterparts (Buysse, Davis Goldman

CHAPTER 7

166

& Skinner, 2002). A study by Frostad and Pijl (2007) of Norwegian inclusive

classrooms suggests that nearly one quarter of pupils with special needs have

serious difficulties forming relationships in their peer group, while for their typical

peers this is only 8 percent. It is known from several studies that within the

group of pupils with special needs, pupils diagnosed as having autistic spectrum

disorders and pupils diagnosed as having serious behavioural disorders find it

particularly difficult to build relationships with typical peers and are at risk of

becoming isolated in the classroom (Chamberlain, Kasari & Rotheram-Fuller,

2007; De Monchy, Pijl & Zandberg, 2004; Garrison-Harrell, Kamps & Kravits,

1997). Pupils diagnosed as having intellectual disabilities are also at risk of

isolation in the classroom (Nowicki, 2006).

This is worrisome, all the more because being isolated in the class might

negatively influence the functioning of a pupil in different areas. Several studies

have shown that pupils with (mild) disabilities, compared to their typical

counterparts, report a higher degree of social dissatisfaction with their peer

relationships (Taylor, Asher & Williams, 1987, in Gresham & McMillan, 1997) and

show significantly higher loneliness scores (Heiman & Margalit, 1998; Lackaye &

Margalit, 2006; Luftig, 1988; Pavri & Monda-Amaya, 2000; Williams & Asher,

1992). In addition, the self-concept of pupils with special needs is found to be

significantly lower (Bender & Wall, 1994; Cambra & Silvestre, 2003; Pijl, Skaalvik

& Skaalvik, 2008), which might lead to externalising problems (e.g. aggression)

and internalising problems (e.g. anxiety) (Durrant, Cunningham & Voelker,

1990).

Because of the rather harmful long-term effects of negative social experiences at

school (Bagwell, Newcomb & Bukowski, 1998; Nelson, Rubin & Fox, 2005; Parker

& Asher, 1987; Terry & Coie, 1991), it is important to monitor the social

participation of pupils. Social participation has been described as the presence of

positive social contacts/interactions between pupils with special needs and their

classmates; acceptance of them by their classmates; social

relationships/friendships between them and their classmates; and the pupils’

own perception of having a positive relation with peers (Koster, Nakken, Pijl &

Van Houten, in press).

Since it is known from other countries with a long tradition of inclusive education

that pupils with special needs experience difficulties in their social participation, it

is important to examine the state of affairs concerning the social participation of


167

pupils with special needs in the Netherlands. Resulting from the above, the

following research questions are formulated:

1. Does the social participation of pupils with special needs in regular Dutch

primary schools differ from the social participation of pupils without special

needs?

2. Does the social participation of pupils with various categories of disabilities

differ?

In answering the research questions, the focus will lie on the four key themes of

social participation mentioned above: friendships/relationships,

contacts/interactions, the pupil’s social self-perception and acceptance by

classmates.

7.2 Method

7.2.1 Introduction

This study addresses the state of affairs of the social participation of pupils with

special needs in regular Dutch primary schools. The data were gathered by seven

graduate students and the first author. As the method, including participants and

instruments, is already described in detail in previous articles (Koster, Minnaert,

Nakken, Pijl & Van Houten, 2008; Koster, Timmerman, Nakken, Pijl & Van

Houten, 2008), this will only be addressed briefly in the current article. For more

detailed information we refer to these previous articles.

7.2.2 Respondents

The study took place in Grades 1 to 3 of regular Dutch primary schools with at

least one pupil with special needs receiving a pupil-bound budget. Such a budget

is allocated to pupils who, according to independent committees, meet the Dutch

national criteria for a pupil-bound budget. Herewith, several categories of

disabilities are distinguished, each with their own criteria. These involve, among

other things, categories of the Diagnostic and Statistical Manual of Mental

Disorders (DSM IV) and intellectual ability, which are assessed by qualified

psychiatrists or psychologists who operate independently from the committees.

CHAPTER 7

168

Data collection took place in two periods. In each period, 300 regular primary

schools were invited to participate in the study. In the first sub-sample, 75

classes of 53 schools were involved, comprising 244 pupils – 96 with special

needs and 148 without. An overview of the categories of disabilities and their

distribution in the first sub-sample is presented in the second and third columns

of Table 1. Data collection focused on both the social self-perception of pupils

and their acceptance by peers.

The second sub-sample, in which 105 classes of 66 schools took part, was

applied to assess the friendships of pupils as well as their contacts and

interactions in the classroom. The respondents were 346 pupils, 141 with special

needs and 205 without. An overview of the types of disabilities and their

distribution in the second sub-sample is presented in the fourth and fifth columns

of Table 1. More detailed information about the sub-samples can be found in

Koster, Timmerman et al. (2008).


in both sub-samples

Category of

disabilities

First sub-sample Second sub-sample

# pupils

(% total)

# boys (%

disabilities

category)

# pupils

(% total)

# boys (%

disabilities

category)

Behavioural disorder 13 (13.5%) 11 (84.6%) 16 (11.3%) 14 (87.5%)

Autistic spectrum

disorder

42 (43.8%) 38 (90.5%) 55 (39.0%) 45 (81.8%)

Motor disability 10 (10.4%) 9 (90.0%) 25 (17.7%) 18 (72.0%)

Intellectual disability 11 (11.5%) 5 (45.5%) 15 (10.6%) 5 (33.3%)

Speech-language

disabilities

20 (20.8%) 13 (65.0%) 27 (19.1%) 19 (70.4%)

Total 96 (100%) 76 (79.2%) 138 (97.8%) 101 (73.2%)

Two pupils with learning disabilities and one pupil with chronic illness were involved in the second sub-

sample, but because of the small numbers, these pupils were excluded from the analyses in this

article.


169

7.2.3 Instruments to assess key themes

The manner in which each key theme is assessed is described briefly below. An

extensive description of the selection procedure for each instrument can be

found in Koster, Minnaert et al. (2008).

Assessment of friendships/relationships. The reciprocal friendship-

nomination method was used to assess friendships. All pupils in the 105 classes

of the second sub-sample were asked which classmates they considered to be

their best friends. Following Pijl, Frostad and Flem (2008), pupils were allowed to

give a maximum of five nominations. Pupils’ number of friendships could be

calculated on the basis of their answers: friends were defined as two pupils

independently selecting each other as best friend. As next to dyadic friendships

several researchers stress the importance of examining membership of peer

networks when studying inclusion (Cullinan, Sabornie & Crossland, 1992; Farmer

& Farmer, 1996; Kindermann, 1993), it was decided to include assessment of

cohesive subgroups in the study. The reciprocal nomination method to identify

dyadic friendships was also used for the identification of cohesive subgroups.

Subgroups of pupils have been identified through statistical analyses (see

Analysis).

Assessment of contacts/interactions. Observations during lessons and free

time took place in order to assess pupils’ contacts and interactions. Mainly based

on Gresham’s observation categories (Gresham, 1982), an observation schedule

was constructed in which initiated and received interactions with both classmates

and the teacher were coded. The observers were five graduate students and the

first author, who initially received three hours’ training using videotaped

recordings of a classroom situation. After training, the agreement between

observers was determined by calculating Cohen’s kappa for three major aspects:

‘interaction/no interaction between pupil and classmates’, ‘initiated/received

interaction of pupil with classmates’ and ‘interaction/no interaction between pupil

and teacher’. For these aspects, Cohen’s kappas were respectively 0.84, 0.76

and 0.72, suggesting reasonable agreement.

For practical reasons, observations took place only in some of the classes

of the second sub-sample. In 58 classes observations were carried out during

lessons and during free time. In each class, both the pupil with special needs and

a randomly selected pupil of the same sex without special needs were observed

for 20 minutes, divided into 5-minute periods. Each period of five minutes was

CHAPTER 7

170

divided into 30 ten-second intervals. If an interaction occurred during that

period, a tick was noted in the correct category (initiated interaction with

classmate; received interaction with classmate; interaction with teacher). When

more than one interaction occurred in a period, only the first one was noted.

Assessment of social self-perception. As there is no Dutch self-perception

scale which is suitable for pupils in Grades 1, 2 and 3 (Koster, Minnaert et al.,

2008), different instruments were applied for different age groups in the first

sub-sample. In Grade 3 the social subscale of the Dutch version of the Self-

Perception Profile for Children (SPP-C, Harter, 1985, in Berndt & Burgy, 1996)

was used. In Dutch this profile is abbreviated as CBSK (Veerman, Straathof,

Treffers, Van den Bergh & Ten Brink, 1997). The social subscale of the Dutch

version of the Pictorial Scale of Perceived Competence and Social Acceptance for

Young Children (PSPCSA, Harter & Pike, 1984; Van Rossum & Vermeer, 1992)

was selected for pupils in Grades 1 and 2. In this scale, verbal items are

supplemented by pictures. The reliability (Alpha) of the social subscale of the

CBSK is 0.74 (Veerman et al., 1997) and that of the Pictorial Scale is 0.78 (Van

Rossum & Vermeer, 1992), which is sufficient for research purposes (Nunnally,

1967).

In Grade 3 the social subscale of the CBSK was administered as a group test,

while the social subscale of the Pictorial Scale in Grades 1 and 2 was

administered individually. On the basis of the scores on the social subscale of the

CBSK/Pictorial Scale, a self-perception score was calculated for each pupil. The

minimum score of the CBSK (consisting of six questions) was 6, the maximum

score 24. The score on the Pictorial Scale (consisting of five questions) could

vary between 5 and 20. For both scales it applies that the higher the score, the

higher the social self-perception.

Assessment of acceptance by classmates. The sociometric peer rating

method was selected as instrument to assess acceptance of pupils in the first

sub-sample. All pupils in the 75 classes were asked to fill in a rating scale

containing the names of all classmates. They were asked to indicate on a 3-point

scale to what degree they would like to play with each classmate. They could

choose between the following three answering categories, each visually

supported by smileys: 1) yes, I would like to ☺, 2) I don’t care �, and 3) no, I

would not like to �. In Grades 2 and 3 the rating scale was administered as a

group test, in Grade 1 it was administered individually: the researcher read out


171

loud the names of all classmates and the pupil mentioned how much s/he liked

to play with each of them.

Table 2 shows the instruments used to assess the four key themes and the

numbers of responding pupils (with and without special needs).

Table 2. Instruments to assess key themes related to number of pupils

Subscale Instrument # pupils with

special needs

# pupils

without

special needs

Friendships /relationships Reciprocal nomination

method

137 202

Contacts /interactions Observation schedule 58 58


perception

Social subscale of CBSK

“ ” Pictorial Scale

27

67

42

99

Acceptance by classmates Sociometric Rating scale 96 148

7.2.4 Analysis

Friendships/relationships. The data resulting from the reciprocal friendship

nomination method were analysed using UCINET software (Borgatti, Everett &

Freeman, 1999) to identify friendships and NEGOPY 4.30 software (Richards,

1995) to identify subgroups and social roles in the classroom. Friendship was

defined as a reciprocal choice, implying that two pupils choose each other as

their best friend (Frostad & Pijl, 2007). A cohesive subgroup in the classroom

was defined as a set of at least three pupils who have more links with members

of the group than with non-members, are connected by some path to each of the

group members, and remain connected when up to 10% of the group is removed

(Richards, 1995). Aside from being a member of a cohesive subgroup, pupils

could occupy other roles in the classroom. These roles can be subclassified into

the following isolated roles (a-d) and participant roles (e-g) (Pearson & Michell,

2000; Richards, 1995):

a. isolate type 1: pupil with no reciprocated links;

b. isolate type 2: pupil with one reciprocated link, but not a dyad;

c. isolated dyad: a pair of pupils linked only to each other;

d. tree node: an isolate type 2 has only one link. If one or more of these

isolates are attached to the first one, the first one is called a tree node;

CHAPTER 7

172

e. liaison 1 (or direct liaison): a pupil who has more than 50% of his linkage

with members of groups in general, but not with members of any single

group. This pupil links groups directly;

f. liaison 2 (or indirect liaison): a pupil who has less than 50% of his linkage

with members of groups. Most links will be with other liaisons. This pupil

connects groups indirectly;

g. group member: a pupil who has more than 50% of his linkage with other

members in the same group. The pupil must have at least two links with

other members.

Roles a to d are called isolated, since pupils who occupy these roles are

minimally connected to others in the group. The other roles (e-g) are categorised

as participant roles, since pupils who occupy these roles have at least two links

with other participants (Richards, 1995).

Contacts/interactions. The total number of (both initiated and received)

interactions with fellow classmates and the total number of interactions with the

teacher were calculated for each pupil.

Social self-perception. For each pupil a social self-perception score was

calculated on the basis of the outcomes of the CBSK (Grade 3) /Pictorial scale

(Grades 1 and 2). The raw scores of both scales were used for this purpose. For

pupils in Grade 3 the score could range between 6 and 24, for pupils in Grades 1

and 2 the score could vary from 5 to 20.

Acceptance by classmates. Each pupil received scores from all classmates

on a 3-point scale (see Instruments). Counting up all scores resulted in a raw

score for each pupil. As the score a pupil could receive was strongly related to

class size, Z-scores were calculated. This was done by subtracting the mean

score of the class from the pupil’s raw score, then dividing the difference by the

standard deviation of the class. As a result, scores of pupils could not only be

compared with their classmates’ scores, but also with scores of pupils from other

classes (comprising various numbers of pupils).

7.3 Results

Friendships/relationships. Based on the friendship nominations and allowing only

reciprocal links, each pupil was categorised as a group member, tree node, dyad,


173

or one of the isolate or liaison types. The results for both the pupils with special

needs and their typical classmates are presented in the top part of Table 3. The

data show that pupils with special needs occupy more often an isolated (48.9%)

and less often a participant role (51.1%) compared to their typical classmates (of

whom 21.3% have an isolated and 78.7% a participant role). The difference

between both groups of pupils is significant (t(337)=-5.55, p<0.05). Next to

comparing the group of pupils with special needs with the group of typical pupils,

a distinction was made between different categories of disabilities, since treating

pupils with special needs as one homogenous group might reveal too negative a

picture of the social participation of pupils with specific types of disabilities. We

examined the differences in social roles between the five main categories of

disabilities (see Respondents).

Table 3. Social roles of pupils with and without special needs

Isolated roles Participant roles

N Isolate type 1

Isolate type 2

Isolated dyad

Tree node

Liaison type 1

Liaison type 2

Group member

Pupils without special needs

202 100%

11 5.4%

22 10.9%

3 1.5%

7 3.5%

1 0.5%

24 11.9%

134 66.3%


137 100%

19 13.9%

34 24.8%

7 5.1%

7 5.1%

4 2.9%

8 5.8%

58 42.3%

Category of

disability


54 100%

2 3.7%

15 27.8%

3 5.6%

2 3.7%

1 1.9%

4 7.4%

27 50.0%

Speech/ language disability

27 100%

2 7.4%

6 22.2%

2 7.4%

0 0%

1 3.7%

1 3.7%

15 55.6%

Motor disability

25 100%

6 24.0%

7 28.0%

1 4.0%

4 16.0%

0 0%

0 0%

7 28.0%


16 100%

4 25.0%

4 25.0%

0 0%

0 0%

1 6.3%

2 12.5%

5 31.3%


15 100%

5 33.3%

2 13.3%

1 6.7%

1 6.7%

1 6.7%

1 6.7%

4 26.7%

As presented in the bottom part of Table 3, more than half (55.6%) of the pupils

diagnosed as having speech/language disabilities and half of the pupils diagnosed

as having an autistic spectrum disorder are a group member, whereas only

CHAPTER 7

174

26.7% of the pupils with an intellectual disability belong to a group of friends. Of

these latter pupils, one third are totally isolated, which means they have no

friends. About a quarter of the pupils with motor disabilities and the pupils with

behavioural disorders also have no friends at all. ANOVA reveals that the

differences between categories of disabilities are not significant (F(4, 132)=2.37,

p=0.06).

In 34 of the 105 participating classes there were two or more pupils with special

needs in the classroom. Since research has shown that pupils tend to associate

with similar peers (Farmer & Farmer, 1996; McPherson, Smith-Lovin & Cook,

2001), we assumed that pupils with special needs who were in the same

classroom would often be connected to each other (for instance, being a member

of the same cohesive subgroup or being a dyad). However, it turned out that

pupils with special needs were connected to each other in only eight of the 34

classes. More specifically, in six classes the pupils with special needs belonged to

the same cohesive subgroup. In one class, two out of three pupils with special

needs were a dyad and in another class one of the pupils with special needs was

an isolate type 2, who was connected to the other pupil with special needs who

was a group member. There were no connections between the pupils with special

needs in the other 26 groups. The low number of pupils with special needs who

were connected to each other might be related to the fact that in almost two-

thirds (22) of the 34 classes, the categories of disabilities of the pupils differed.

Besides, gender might have played a role. In ten classes, a boy with special

needs and a girl with special needs were involved. None of these pupils were

connected to each other. In elementary education (especially in the higher

grades), boys’ and girls’ networks are often strictly separated, as children prefer

to associate with peers of their own gender (Gest, 2006; Sippola et al., 1997, in

Nowicki, 2006). There appeared to be thus little similarity between many of the

pupils with special needs.

Next to examining the social role of pupils, their number of friendships was

calculated. Analyses revealed that the average number of friends of pupils with

special needs (M=1.9, SD=1.3) is significantly lower compared to that of typical

pupils (M=2.9, SD=1.4, t(340)=6.48, p<0.05) (see second column in Table 4).

This difference represents a moderate effect size, ES=0.71 (Cohen, 1992). In

order to take into account possible differences between categories of disabilities,

a further analysis was done. As presented in the second column of Table 4, the


175

mean number of friends is highest for pupils diagnosed as having speech/

language disabilities (2.2), while this number is lowest for pupils diagnosed as

having an intellectual disability (1.5). However, a one-way analysis of variance

(ANOVA) revealed that the differences between categories of disabilities are not

significant, at F(4, 132)=1.2, p=0.32.

Contacts/interactions. A comparison was made between pupils with and

without special needs with regard to their initiated and received interactions with

classmates and their interactions with the teacher. It turned out that pupils with

special needs initiated fewer interactions with classmates (M=10.7, SD=8.3)

than pupils without special needs (M=14.7, SD=9.6). This difference is

significant at t(114)=2.39, p<0.05, and represents a moderate effect size

(ES=0.44). The number of received interactions also differed significantly. Pupils

with special needs received fewer interactions (M=7.1, SD=5.2) than their

typical classmates (M=11.0, SD=6.4, t(107.7)=3.58, p<0.05), representing a

moderate effect size (ES=0.67). In addition, it turned out that pupils with special

needs (M=8.8, SD=9.7) had significantly more interactions with the teacher than

pupils without special needs (M=3.1, SD=3.7, t(116)=-4.2, p<0.05). The effect

size is large (ES=-0.86).

As presented in the third and fourth columns of Table 4, pupils diagnosed as

having intellectual disabilities had the most interactions with classmates,

whereas pupils diagnosed as having behavioural disorders had the most

interactions with the teacher. The latter received the least interactions from

classmates. Pupils diagnosed as having motor disabilities initiated the least

interactions with classmates and had the least interactions with the teacher.

However, the differences between the various groups of pupils are not

significant: ANOVA revealed no significant group differences between pupils with

different categories of disabilities in terms of initiated interactions with

classmates (F(4, 53)=0.84, p=0.50), received interactions with classmates (F(4,

53)=0.43, p=0.79) and interactions with the teacher (F(4, 53)=1.00, p=0.41).

For an overview of the outcomes on contacts/interactions, see third to fifth

columns of Table 4.

Social self-perception. In all grades, the social self-perception of pupils

with and without special needs turned out to be comparable. In Grade 3, the

mean score of pupils with special needs (M=17.5, SD=4.2) did not differ

significantly from the mean score of their typical counterparts (M=17.3, SD=4.0,

CHAPTER 7

176

t(67)=-0.17, p=0.87; ES=-0.05). In Grades 1/2 the mean score of both groups

of pupils did not differ significantly either (pupils with special needs, M=14.2,

SD=3.2; pupils without special needs, M=14.4, SD=2.3; t(164)=0.66, p=0.51;

ES=0.08).

As can be seen in the sixth column of Table 4, in Grades 1/2 pupils diagnosed as

having an intellectually disability had the highest mean social self-perception

score, whereas pupils diagnosed as having autistic spectrum disorders had the

lowest. In Grade 3, pupils diagnosed as having behavioural disorders had the

highest mean social self-perception score. Contrary to the outcomes of Grades

1/2, in Grade 3 pupils diagnosed as having intellectual disabilities had the lowest

mean score.

ANOVA revealed that the group differences between pupils with various

categories of disabilities are not significant: Grade 1/2: F(4, 62)=1.72, p=0.16;

Grade 3: F(4, 22)=0.85, p=0.51. For an overview of the outcomes on pupils’

social self-perception, see sixth and seventh columns of Table 4.

Acceptance by classmates. A comparison between the acceptance score of

pupils with and without special needs revealed that the former are significantly

less accepted (pupils with special needs, M=-0.71, SD=1.0; pupils without

special needs, M=-0.06, SD=1.0; t(242)=4.9, p<0.05). The effect size is

moderate (ES=0.64).

As shown in the last column of Table 4, pupils diagnosed as having

speech/language disabilities and pupils diagnosed as having motor disabilities are

the most accepted by classmates, whereas pupils diagnosed as having

intellectual disabilities are the least accepted. However, similarly to the

assessment of the other key themes, ANOVA revealed no significant group

differences between pupils with various categories of disabilities: F(4, 91)=2.34,

p=0.06.


177

CHAPTER 7

178

7.4 Discussion

This study addressed the state of affairs concerning the social participation of

pupils with special needs in regular Dutch primary schools. A comparison was

made between the social participation of pupils with special needs and their

classmates without special needs in Grades 1 to 3 and between pupils with

different categories of disabilities. Social participation was subdivided into

friendships/relationships, contacts/interactions, pupil’s social self-perception and


It turned out that with regard to three key themes of social participation, the

situation of pupils with special needs is less favourable compared to that of their

typical counterparts. Concerning the key theme friendships/relationships, the

outcomes revealed significant differences between these two groups of pupils.

Pupils with special needs had on average fewer friends and belonged to a group

of friends less often. Looking at the key theme contacts/interactions, pupils with

special needs turned out to have fewer interactions with their classmates but

more interactions with the teacher. Pupils with special needs are at a

disadvantage too with regard to the key theme acceptance by classmates, as the

degree to which they were accepted by their classmates turned out to be

significantly lower than the degree of acceptance of typical pupils. The situation

concerning the key theme pupil’s social self-perception is different: the social

self-perception of pupils with special needs did not differ from that of pupils

without special needs. It was expected that pupils with special needs would have

a lower social self-perception than their typical peers, because of their lower

number of friendships, their lower acceptance and their lower number of

interactions with classmates. However, other studies also reveal that pupils with

special needs tend to have a relatively high self-perception (Elias & Van

Nieuwenhuijzen, 2001). According to Elias and Van Nieuwenhuijzen (2001) and

Verhagen and Vermeer (1997), typically developing children do have a realistic

picture of their social competence, while children with special needs might lack

such a realistic picture. Research has revealed that some children, especially

aggressive boys, tend to be positively biased in their self-perceptions (Asher,

Parkhurst, Hymel & Williams, 1990; Boivin & Begin, 1989; in Sletta, Valås &

Skaalvik, 1996). Elias and Van Nieuwenhuijzen (2001) and Gresham and

MacMillan (1997) state that the self-examination of competence of pupils with


179

special needs might be influenced by their need to protect themselves from

negative evaluations. This flattering self-perception might also apply to (some of)

the pupils with special needs in this study. Contrary to expectations, the analyses

revealed no significant differences between pupils with different categories of

disabilities on any of the four distinguished areas of social participation.

The outcomes are fairly worrisome, as pupils with special needs perform

significantly less well than pupils without special needs on three out of four

important areas of social participation. However, the outcomes should be slightly

nuanced. In the first place, taking into consideration the social participation of

pupils with special needs on its own merits without making comparisons with

typical pupils might provide some nuance. For instance, although pupils with

special needs have fewer friends compared to their typical peers, the vast

majority of the pupils with special needs do have one or more friends. This is a

beneficial outcome, since having at least one friend in the classroom may be a

source of companionship and emotional support (Ladd, 1990), and may protect

against the negative effects of low acceptance (Asher et al., 1990; Newcomb &

Bagwell, 1996). Similarly, most pupils with special needs have a positive social

self-perception, are accepted and have a reasonable number of interactions with

peers.

Second, it is unknown how the pupils with special needs involved in the study

would have functioned in special-education settings. The social participation of

pupils with specific categories of disabilities in special schools might have been

disappointing too. For instance, Mand (2007) found that not only in regular

classes but also in special-education settings, a large proportion of pupils with

behavioural disorders have a negative social position in the classroom. These

pupils are to a comparable degree rejected in both educational systems. Hence in

special education too there is a real chance of pupils’ social participation not

working out that favourably.

Third, instead of treating the group of pupils with special needs as one

homogenous group, a distinction between categories of disabilities seems to

provide nuances for pupils with specific disabilities. Beforehand, pupils diagnosed

as having autistic spectrum disorders, serious behavioural disorders and/or

intellectual disabilities were expected to experience the most social participation

problems, as research has revealed that these pupils find it particularly difficult

to build relationships with typical peers and are at risk of becoming isolated in

CHAPTER 7

180

the classroom (Chamberlain et al., 2007; De Monchy et al., 2004; Garrison-

Harrell et al., 1997; Nowicki, 2006). Pupils diagnosed as having motor disabilities

were expected to perform best on each of the four key themes of social

participation, as their type of disability is expected to have the least impact on

social functioning in the classroom. In addition, motor disabilities are visible and

understandable for classmates, which fosters acceptance (Lewis, 1995, in Laws &

Kelly, 2005). Contrary to expectations, there were no significant differences

between pupils with various categories of disabilities. In order to reveal

significant differences, the number of pupils should be larger. Further research,

including larger subgroups of pupils with specific categories of disabilities, is

recommended. A second factor which might have played a role concerns the

severity and complexity of the disabilities involved in the study. This might be an

explanation especially for the relatively high scores of pupils diagnosed as having

autistic spectrum disorders. Most of the pupils with an autistic spectrum disorder

were categorised as having a Pervasive Developmental Disorder - Not otherwise

Specified (PDD-NOS). Relatively few pupils were categorised with other, more

severe types of autistic spectrum disorders.

A final remark concerns the societal meaning of the findings of this study. In

accordance with several other studies, the current study revealed that, compared

to their typical peers, the social participation of pupils with special needs is

assessed as lower. It follows that, although social participation is considered one

of the most important outcomes of inclusion, for a substantial part of the pupils

with special needs optimal social participation is not fully realised in practice.

This is a striking situation: inclusion is promoted because it is assumed to be

positive for pupils with special needs, but on the other hand we know that

inclusion of these pupils might result in negative outcomes (e.g. loneliness,

rejection) for some of them. Ethically, it is inadmissible to just notice that pupils

with special needs have a larger chance of being socially excluded than their

typically developing peers, and do nothing about it. Therefore, schools should

take measures. One possible solution might be to rearrange the composition of

the classroom (Pijl et al., 2008): placing small groups of pupils with special

needs in a regular classroom, rather than single pupils, to ensure social cohesion

among the pupils with special needs. This would respect the tendency of pupils

to associate with similar peers (Farmer & Farmer, 1996; Male, 2007; McPherson

et al., 2001). In that manner, both the stimulating environment of the regular


181

classroom and the presence of peers with comparable special needs are

combined. This might apply particularly to pupils with hearing disabilities, who

because of limited communication possibilities seem to prefer to associate with

other peers with hearing disabilities in integrated settings (Minnett, Clark &

Wilson, 1995). Further research into rearranging the classroom and into other

interventions to improve social participation of pupils with special needs is highly

recommended.

Acknowledgements

Special thanks to Per Frostad (Norwegian University of Science and Technology)

and Erik Logtenberg for their contribution to the study described in this article.

7.5 References

Asher, S.R., Parkhurst, T., Hymel, S. & Williams, G.A. (1990). Peer rejection and

loneliness in childhood. In S.R. Asher. & J.D. Coie, (Eds.), Peer rejection in

childhood (pp.253-273). Cambridge: Cambridge University Press.



69(1), 140-153.

Bender, W.N. & Wall, M.E. (1994). Social-emotional development of students

with learning disabilities. Learning Disability Quarterly, 17, 323-341.

Berndt, T.J. & Burgy, L. (1996). Social Self-Concept. In B.A. Bracken (Ed.),

Handbook of Self-Concept (pp. 171-209). New York: John Wiley & Sons,

Inc.



Buysse, V., Davis Goldman, B. & Skinner, M.L. (2002). Setting effects on

friendship formation among young children with and without disabilities.


Cambra, C. & Silvestre, N. (2003). Students with special educational needs in the

inclusive classroom: social integration and self-concept. European Journal


CHAPTER 7

182

Chamberlain, B., Kasari, C. & Rotheram-Fuller, E. (2007). Involvement or

isolation. The social networks of children with autism in regular

classrooms. Journal of Autism and Developmental Disorders, 37, 230-242.

Cohen, J. (1992). A Power Primer. Psychological Bulletin, 112(1), 155-159.


mildly handicapped students, The Elementary School Journal, 92(3), 339-

351.



Haag: LCTI.







Elias, C. & Van Nieuwenhuijzen, M. (2001). Competentiebeleving en sociale

informatieverwerking bij licht verstandelijk gehandicapte kinderen.

Tijdschrift voor Orthopedagogiek, 40, 611-619.

Farmer, T.W. & Farmer, E.M.Z. (1996). Social relationships of students with

exceptionalities in mainstream classrooms: social networks and

homophily. Exceptional Children, 62(5), 431-450.




Education, 22(1), 15-30.




Gest, S.D. (2006). Teacher reports of children’s friendships and social groups:

agreement with peer reports and implications for studying similarity.

Social Development, 15(2), 248-259.

Gresham, F.M. (1982). Social interactions as predictors of children’s likeability

and friendship patterns: a multiple regression analysis. Journal of

Behavioral Assessment, 4(1), 39-54.


183

Gresham, F.M. & McMillan, D.L. (1997). Social competence and affective

characteristics of students with mild disabilities. Review of Educational

Research, 67(4), 377-415.

Harter, S. & Pike, R. (1984). The Pictorial Scale of Perceived Competence and

Social Acceptance for Young Children. Child Development, 55, 1969-1982.

Heiman, M. & Margalit, M. (1998). Loneliness, depression, and social skills

among students with mild mental retardation in different educational

settings. The Journal of Special Education, 32(3), 154-163.

Kindermann, T.A. (1993). Natural Peer Groups as Contexts for Individual

Development: The Case of Children's Motivation in School. Developmental

Psychology, 29(6), 970-977

Koster, M., Minnaert, A.E.M.G., Nakken, H., Pijl, S.J., & Van Houten, E.J. (2008).

Assessing social participation of pupils with special needs in inclusive

education: validation of the Social Participation Questionnaire. Manuscript



peer group: a literature study focussing on the social dimension of

inclusion in education. International Journal of Inclusive Education.

Koster, M., Timmerman, M.E., Nakken, H., Pijl, S.J., & Van Houten, E.J. (2008).

Evaluating social participation of pupils with special needs in regular

primary schools: examination of a teacher questionnaire. Manuscript


Lackaye, T.D. & Margalit, M. (2006). Comparisons of achievement, effort and

self-perceptions among students with learning disabilities and their peers

from different achievement groups. Journal of Learning Disabilities, 39(5),

432-446.








CHAPTER 7

184

Luftig, R.L. (1988). Assessment of the perceived school loneliness and isolation

of mentally retarded and nonretarded students. American Journal of


Male, D.B. (2007). Friendships and peer relationships. In L. Florian (Ed.), The

SAGE Handbook of Special Education (pp. 460-471). London: Sage.

Mand, J. (2007). Social position of special needs pupils in the classroom – a

comparison between German special schools for pupils with learning

difficulties and integrated primary school classes. European Journal of


McPherson, M., Smith-Lovin, L., & Cook, J.M. (2001). Birds of a Feather:

homophily in social networks. Annual review of sociology, 27, 415-444.

Minnett, A., Clark, K. & Wilson, G. (1995). Play behaviour and communication

between deaf and heard of hearing children and their hearing peers in an

integrated pre-school. American Annals of the Deaf, 139, 420-429.



relationships, International Journal of Inclusive Education, 6(1), 47-61.








Nowicki, E.A. (2006). A cross-sectional multivariate analysis of children’s

attitudes towards disabilities. Journal of Intellectual Disability Research,

50(5), 335-348.

Nunnally, J.C. (1967). Psychometric Theory. New York: McGraw-Hill Book

Company.



Pavri, S. & Monda-Amaya, L. (2000). Loneliness and students with learning

disabilities in inclusive classrooms: self-perceptions, coping strategies, and

preferred interventions. Learning Disabilities Research & Practice, 15(1),

22-33.


185

Pearson, M. & Michell, L. (2000). Smoke Rings: social network analysis of

friendship groups, smoking and drug-taking. Drugs: Education, Prevention

and Policy, 7(1), 21-37.

Pijl, S.J., Frostad, P. & Flem, A. (2008). The social position of pupils with special

needs in regular schools. Scandinavian Journal of Educational Research,

52(4), 387-405.

Pijl, S.J., Skaalvik. E.M. & Skaalvik, S. (2008). Students with special needs and

the composition of their peer group. Manuscript in preparation.

Pijl, S.J. & Van den Bos, K.P. (2001). Redesigning regular education support in

the Netherlands. European Journal of Special Needs Education, 16(2),

111-119.

Richards, W.D. (1995). Negopy 4.30. Manual and User’s Guide. Burnaby: Simon

Fraser University.

Sletta, O., Valås, H., & Skaalvik, E.M. (1996). Peer relations, loneliness, and self-

perceptions in school-aged children. British Journal of Educational







H. Walker (Eds.), Children’s social behavior: Development, assessment

and modifications (pp. 287-330). New York: Academic Press.



Van Rossum, J.H.A. & Vermeer, A. (1992). Het meten van waargenomen

competentie bij basisschoolkinderen door middel van een platentest.

Bewegen & Hulpverlening, 9(3), 198-212.

Veerman, J.W., Straathof, M.A.E., Treffers, P.D.A., Van den Bergh, B.R.H. & Ten

Brink, L.T. (1997). Handleiding Competentie Belevingsschaal voor

Kinderen. Lisse: Swets & Zeitlinger.

Williams, G.A. & Asher, S.R. (1992). Assessment of loneliness at school among

children with mild mental retardation. American Journal on Mental

Retardation, 96(4), 373-385.

CHAPTER 7

186

Chapter 8 General discussion

CHAPTER 8

GENERAL DISCUSSION

CHAPTER 8

188

8.1 Introduction

During the last decades, the inclusion of pupils with special needs has been

increasingly promoted. In the Netherlands, the government supported inclusion

by amending legislation and regulations on special needs education. Since the

pupil-bound budget was introduced into the Dutch educational system in August

2003, the number of pupils with special needs in regular education has increased

(De Greef & Van Rijswijk, 2006). Many parents have socially motivated

arguments in choosing regular education for their child. They hope and expect

their child mixing with pupils without special needs will lead to social

participation, thus implying their child will build friendships with typical peers and

becomes a full member of the regular class. Furthermore, some parents assume

that as a result of sending their child with special needs to a regular school, a

change of attitude will arise among other children, possibly leading to positive

long-term effects on attitudes towards disability in wider society. This focus on

the social outcomes of including pupils with special needs fits with the reasoning

behind inclusion. As described by Farrell (2000), inclusion implies that all pupils

take a full and active part in school-life, are valued members of the school

community and are seen as an integral member. Since this social dimension can

be regarded as the core of inclusion, and because it is known from numerous

studies (e.g. Asher & Coie, 1990; Parker & Asher, 1987; Terry & Coie, 1991) that

difficulties in this area (like being rejected or feeling lonely) can have negative

consequences on pupils’ development, it is important to verify if this aspect of

inclusion is achieved in practice.

This study looks at the social dimension of inclusion and consists of three phases.

In the first phase (8.2.1), the initial experiences with the pupil-bound budget in

regular Dutch primary education are central. The second phase (8.2.2) focuses

on helping teachers accurately assess pupils’ social participation. To this end, the

construction of a model of social participation, the development of a teacher

questionnaire to assess the social participation of pupils with special needs and

the subsequent assessment of the psychometric qualities of this questionnaire

were addressed. The third phase (8.2.3) focuses on the consequences of

inclusion on the social development of pupils with special needs by describing

their current situation in this respect in regular Dutch primary schools.

GENERAL DISCUSSION

189

8.2 Main findings

8.2.1 First experiences with the pupil-bound budget

The preparatory study described in Chapter 2, about the first experiences with

the pupil-bound budget in regular Dutch primary education, revealed that the

social position of pupils with special needs did not significantly differ from the

social position of typical classmates. This outcome deviated from other studies

(e.g. De Monchy, Pijl & Zandberg, 2004; Larrivee & Horne, 1991; Sale & Carey,

1995), in which it was revealed that pupils with special needs occupy a lower

social position in class on average than their typically developing peers. Despite

the fact that this outcome seems unexpected at first glance, it might be

explained by two factors. First, the young age of the children (average age 5;7)

could be a factor that played a role. Whereas during preschool, friendships are

largely based on proximity and formed through change encounters, in later age

periods children of like minds increasingly seek each other out and become

friends because of shared interests and similarity (Schaffer, 1996). As the pupils

in our sample were quite young (mainly preschoolers), this explains why the

pupils with special needs were accepted to the same degree as pupils without

special needs. Secondly, the overrepresentation of pupils with Down Syndrome,

who in literature have traditionally been described as cheerful, happy and

sociable (Walz & Benson, 2002), might partly explain why the social position of

the pupils with special needs was comparable to that of typical pupils. Next to

the pro-social behaviours of pupils with Down Syndrome, the visibility of their

disability seems to foster acceptance (Lewis, 1995, in Laws & Kelly, 2005).

Further examination of the outcomes of the preparatory study revealed a

discrepancy between the teachers’ views on the social position of the pupil with

special needs and the classmates’ views (as measured with a sociometric

questionnaire). Teachers’ views turned out to be far more positive, which is in

accordance with the results of comparable studies in which teachers were more

positive about the social position of pupils with special needs compared to the

results of sociometric questionnaires completed by pupils (De Monchy et al.,

2004; Scheepstra, 1998). In addition to this discrepancy between teachers’ and

pupils’ views on the social position of pupils with special needs, another one was

revealed. Whereas the teachers, parents and to a lesser degree the peripatetic

teachers of the pupils with special needs were generally satisfied with the pupils’

CHAPTER 8

190

cognitive, social and social-emotional development, a panel of independent

assessors had some concerns about the development of one third of the pupils

with special needs. Teachers and parents were especially more positive

compared to the panel concerning pupils’ social and social-emotional

development.

In concluding, the preparatory study indicated that, although the social position

of pupils with special needs did not differ from that of their typical counterparts,

it is important to study the social outcomes of inclusion, as teachers seem to

overestimate pupils’ social(-emotional) development.

8.2.2 Construction of a model of social participation and developing a

teacher questionnaire

The discrepancy in views on the social and social-emotional development of

pupils with special needs, as shown in the preparatory study, led to the next

phase of the study. The tendency of teachers to be too optimistic about the

social and social-emotional development of the pupils with special needs is

worrying seeing that they are the obvious adults to monitor pupils’ social and

emotional development in school. It is only when teachers notice a pupil

becoming isolated or teased that appropriate measures can be taken. This

underlines the importance of developing an instrument for teachers which helps

them assess the social participation of pupils with special needs.

Thus, it was decided to construct a model of social participation and then develop

an instrument for teachers aimed at helping them assess the social participation

of pupils with special needs. This comprised the core of the second phase of the

study. Based on an analysis of literature, a model of social participation,

distinguishing four key themes (‘friendships/relationships’,

‘contacts/interactions’, ‘social self-perception of the pupil’ and ‘acceptance by

classmates’), was constructed. This formed the basis for the teacher

questionnaire, named Social Participation Questionnaire. Examination of the

psychometric qualities of the Questionnaire revealed that both the reliability and

the construct validity (enclosing discriminant and convergent validity) were

satisfactory.

GENERAL DISCUSSION

191

8.2.3 Current situation regarding social participation of pupils with

special needs

The outcomes with respect to the social participation of pupils with special needs

in regular Dutch primary education are in line with those of many international

studies, i.e. causes for concern. With regard to three key themes of social

participation, the situation of pupils with special needs turned out to be less

favourable compared to that of their typical counterparts. Concerning

friendships/relationships the outcomes showed significant differences between

these two groups of pupils. Pupils with special needs had less friends, on

average, and were less often a member of a group of friends. Looking at

contacts/interactions, pupils with special needs turned out to have less

interaction with their classmates, whereas they had more with the teacher.

Regarding acceptance by classmates, pupils with special needs were again at a

disadvantage, as the degree to which they were accepted by their classmates

was significantly lower compared to that of typical pupils. The situation deviated,

however, for the key theme pupil’s social self-perception in that pupils with

special needs did not differ from pupils without special needs in this respect.

Both groups of pupils had on average a relatively positive social self-perception.

8.3 Reflections on the study

Some critical considerations need to be taken into account regarding this study.

With regard to the first phase, in which the preparatory study took place, two

points can be mentioned concerning the sample. The sample size was small

(n=20) and pupils with Down Syndrome were overrepresented. However, it was

not possible to involve a larger sample as the study was conducted at the behest

of the Ministry of Education and had to be completed in a short time. The fact

that more than one third of the pupils involved in the study were diagnosed as

having Down Syndrome is related to the fact that at the time the study took

place (the 2003-2004 school year), it was this group of pupils with special needs

in particular who attended regular education. The increase in the number of

pupils with other categories of disabilities attending regular education started

later.

Another critical consideration is related to the second phase of the study and

concerns the assessment of pupils’ friendships/relationships and

CHAPTER 8

192

contacts/interactions. In order to validate the ‘Friendships/relationships’

subscale, pupils’ scores for this were compared with their number of friendships

as derived from the reciprocal nomination method. This implies that the

assessment of friendship was limited to quantitative data, namely their number

of friendships. Also in assessing pupils’ interaction, only quantitative data were

gathered, by counting the number of interactions. It might have been better

when not only these data, but also qualitative data would have been gathered.

These qualitative data could have led to a different weighting of the quantitative

data, which would possibly have had a positive influence on the validation of the

‘Friendships/relationships’ and ‘Contacts/interactions’ subscales, as these include

both the quality and the quantity of friendships and interactions.

Another reason why it could have been better to choose a more qualitative

approach in the assessment of pupils’ friendships, is related to the fact that

several researchers (e.g. Aboud & Mendelson, 1996; Kutnick & Kington 2005;

Newcomb & Bagwell, 1996; Parker & Asher, 1993; Schneider, Wiener & Murphy,

1994) emphasize the importance of taking into consideration the quality of

friendships in stead of merely focusing on quantity, since having one close friend

might be more important for a child than having a larger number of less close

friends. Although this might have been valuable for the study, it would also have

been complex. Among other things, the quality of friendship presumably should

have been assessed in different ways for the different age groups, as the

characteristics of peer relationships vary for different age periods (Schaffer,

1996). In addition, compared to quantitative friendships measures, relating

qualitative friendship measures to pupils’ scores on the Social Participation

Questionnaire would have been far more difficult. This would probably have

negatively influenced the validation process and as our aim was to validate the

Social Participation Questionnaire, the quantitative approach was justifiable.

Regarding the reciprocal nomination method to assess friendships, a second

critical consideration can be made. The answers of pupils to the question who

they consider their best friends in the group, are subject to the issues of the day.

This might especially apply to the youngest children. For instance, it might be

possible a pupil nominated a classmate as friend because (s)he did something

nice that day. At another moment, the pupil might not have chosen this

classmate as a friend. Teachers’ ratings of the Social Participation Questionnaire

statements are less dependent on such issues. The fact that the scores on the

GENERAL DISCUSSION

193

Social Participation Questionnaire, which provide a balanced picture of pupils’

social participation, are compared with data that are dependent on the issues of

the day, causes friction and might have negatively influenced the Questionnaire’s

validation.

Another critical consideration, referring to both the second and the third phase,

is the distribution of categories of disabilities. Although attempts were made to

obtain a proportional distribution of categories of disabilities, some subgroups

were overrepresented. In the studies described in Chapters 5, 6 and 7, pupils

diagnosed as having an autistic spectrum disorder comprised over 40 percent of

the total sample. The overrepresentation of these pupils might have had a

negative influence on the score of the group of pupils with special needs as a

whole, since such pupils in particular, together with pupils diagnosed as having

severe behavioural disorders, have difficulty in building relationships with peers

and are at risk of becoming neglected and rejected (Chamberlain, Kasari &

Rotheram-Fuller, 2007; De Monchy et al., 2004; Garrison-Harrell, Kamps &

Kravits, 1997). However, this negative influence might be enervated a little, as

most of the pupils with an autistic spectrum disorder were categorized as having

a Pervasive Developmental Disorder - Not otherwise Specified (PDD-NOS).

Relatively few pupils were categorized with other, more severe, types of autistic

spectrum disorders. Nonetheless, it would have been better if there had been a

proportional distribution of disabilities. Among other things, this would have

enabled comparisons between pupils with different categories of disabilities and

to discover significant differences between them.

Irrespective of these critical considerations with regard to the development and

examination of the Social Participation Questionnaire, one might conclude that

the Questionnaire is an instrument of good quality, which has the potential to

become a valuable tool for teachers.

Regarding the third phase of the study, in which the current situation concerning

the social participation of pupils with special needs in regular primary schools

was central, several critical considerations can be made. Firstly with regard to

the instruments used to assess pupils’ social participation (described in Chapter

7). This study was largely aimed at developing an instrument to help teachers

assess the social participation of pupils with special needs. In addition, the study

aimed to describe the current state of social participation of pupils with special

needs in regular Dutch primary schools. Whereas in Chapters 4 to 6, the

CHAPTER 8

194

development of the instrument for teachers (the Social Participation

Questionnaire) and the examination of its psychometric qualities were central, in

Chapter 7 the current situation was described. This latter was done without using

the Social Participation Questionnaire. The Questionnaire was only in its

definitive form after the analyses described in Chapter 6 were conducted and a

new round of data collection was not possible. Nonetheless, the data gathered

during the validation process were regarded as a suitable alternative to describe

the current state of pupils’ social participation in regular education, as it provides

interesting and relevant outcomes regarding each of the key themes. It would

have been a lost opportunity if these would only have been used as input to

validate the Questionnaire. In future research, the Social Participation

Questionnaire is intended to be used on a large scale in order to provide an

overview of the social participation of pupils with special needs.

Another consideration with regard to the third phase of the study concerns the

interpretation of the outcomes. Those concerning the state of the social

participation of pupils with special needs give rise for concern as in three out of

four important areas of social participation, pupils with special needs perform

significantly less well than typical pupils. However, the outcomes should be

slightly nuanced.

First, taking into consideration the social participation of pupils with special needs

on its own merits without making comparisons with typical pupils provides some

nuance. For instance, although pupils with special needs have less friends

compared to their typical peers, the vast majority of them (85.7%) have one or

more friends. This is a beneficial outcome, since having at least one friend in the

classroom may be a source of companionship and emotional support (Ladd,

1990) and may protect against the negative effects of low acceptance (Asher,

Parkhurst, Hymel & Williams, 1990; Newcomb & Bagwell, 1996). Similarly, most

pupils with special needs have a positive social self-perception, are accepted and

have a reasonable number of interactions with peers.

Second, distinguishing different categories of disabilities, in stead of considering

the pupils with special needs as one homogeneous group, provides some nuance

for certain subgroups. In our study, the social participation of pupils diagnosed

as having speech/language disabilities was assessed highest, followed by those

diagnosed as having motor disabilities. The social participation of pupils

diagnosed as having autistic spectrum disorders, behavioural disorders and

GENERAL DISCUSSION

195

intellectual disabilities was assessed less positively. In stead of treating pupils

with special needs as one homogenous group, it might be better to distinguish

different categories of disabilities.

Although the worrisome results regarding the social participation of pupils with

special needs in regular education can be slightly ameliorated, they should be

taken seriously. Some people might consider it unfair to compare pupils with

special needs with their typical counterparts, as the former will probably perform

less well in several areas. One could consider laying down specific standards for

pupils with special needs, in stead of comparing them with the norms of typical

pupils. However, this would be an artificial solution, as although teachers might

take account of those norms, pupils and peers will not. Irrespective of having

specific standards for pupils with special needs, the consequences of being lonely

or of being neglected will remain the same. Thereby, although on first thoughts it

might seem unfair to compare pupils with special needs with typical peers, it is

justifiable to make the comparison.

8.4 Implications of the study

8.4.1 Implications for educational policy

Inclusive education mainly arose from an ideological perspective. It was thought

that all pupils with special needs should be educated alongside their typical

peers, which fitted with the wider societal tendency towards inclusion, implying

no one should be excluded. Despite the fact that ideological reasons laid the

foundation for the inclusion of pupils with special needs, it is crucial to evaluate

these pupils’ development in regular education. In Chapters 2 and 7 of this

study, the social participation of pupils with special needs in regular education is

evaluated. The outcomes described in Chapter 7 in particular have important

implications for educational policy, as they again show that the physical presence

of pupils with special needs in regular education does not automatically lead to

optimal social participation. Policymakers should therefore take into

consideration that additional measures need to be taken in order to make

inclusion successful.

Based on the outcomes of the third phase of the study, in stead of concluding

that additional measures need to be taken in regular education, one might

CHAPTER 8

196

conclude that it is better to send pupils with special needs to special schools.

After all, on three out of four important areas of social participation, these pupils

perform significantly less well than their typical counterparts. However, this is a

too simplified conclusion, as it is unknown how the pupils with special need in the

study would have functioned in special education settings. Possibly, the social

participation of pupils with specific categories of disabilities in special schools

might have also fallen short of expectations. For instance, Mand (2007) found

that not only in regular classes, but also in special education settings, a large

proportion of pupils with behavioural disorders have a negative social position in

the classroom. These pupils are unpopular in both education systems to a similar

degree. Related to this, Karsten, Peetsma, Roeleveld and Vergeer (2001) found

little evidence to support the idea that at-risk pupils make less progress, in

academic or psychosocial development, in regular education settings than those

in special schools. The outcomes of the studies of Karsten et al. (2001) and Mand

(2007) indicate that the social participation of pupils with special needs does not

automatically improve when they attend a special school, as in both regular and

special education settings some pupils are seen as outsiders and are at risk of

becoming isolated.

8.4.2 Implications for teacher training

As the number of pupils with special needs attending regular education is

growing, teacher training should increasingly aim at taking this group into

account. It would be advisable to distinguish different categories of disabilities in

teacher training, since this study has shown that the social participation of pupils

with different categories of disabilities differs.

In school curricula, the focus is mainly on achieving academic skills, like reading

and writing and, in keeping with a modern competitive society, the importance

attached to examining these subjects is increasing. As a result, in teacher

training, academic skills have priority, with too little attention seemingly paid to

social aspects of education. It is vital to emphasize these social aspects more, as

a pupil’s wellbeing in the classroom seems to be a prerequisite for acquiring

academic skills. Several studies (e.g. Asher & Coie, 1990; Ollendick, Weist,

Borden & Greene, 1992) have shown that when a child becomes isolated or is

rejected in the classroom, this negatively affects school performance. If teacher

GENERAL DISCUSSION

197

training were to focus more on pupils’ social participation, teachers might be able

to notice problems earlier in this areas and take measures. It is anticipated this

would then positively influence pupils’ performances in several areas.

8.5 Further research

The Social Participation Questionnaire helps teachers assess the social

participation of pupils with special needs. When teachers notice problems in this

area, they need to know which interventions can be arranged. Therefore, it

would be advisable to study possible interventions.

In former years, interventions were usually solely aimed at the pupil with special

needs, of which the often used social skills training is an example. The long-term

effects of this kind of training are often disappointing (Pijl, 2005). One possible

explanation is that classmates’ image of the pupils who participate in the training

does not change and they do not alter their attitude and behaviour. As a result

the trained pupils do not have many possibilities to practice their new skills: they

will not see the benefit of their new skills and after some time the acquired skills

will fade away (Frostad & Pijl, 2007). From an educational (orthopedagogical)

perspective, interventions involving both pupils with special needs as well as

classmates and/or teachers seem more appropriate. In stead of focusing on the

pupil with special needs, the functioning of pupils within their environment and

the characteristics of this environment should be pivotal. Analysing

environmental factors and arranging changes to the environment should be

central to any interventions.

Several factors, which might influence a pupil’s social participation could be

directive for the nature of the interventions. A first factor concerns the

characteristics of pupils, with and without special needs, in inclusive classrooms,

as these characteristics might well influence pupils’ social position in the

classroom. Recent studies suggest that the ‘similarity hypothesis’ (Male, 2007)

might be one of the possible causes of social problems of pupils with special

needs in inclusive classrooms. This hypothesis is based on the idea that children

have the tendency to associate with similar peers. This tendency seems to be a

disadvantage for pupils with special needs in inclusive classrooms, as their

special needs make them different from their typical peers. In view of the

CHAPTER 8

198

effectiveness of interventions, it is important to discover the relation between

pupils’ characteristics and their social participation in the classroom.

Teacher characteristics are a second factor, as teachers’ attitudes towards

inclusive education might influence their acting in the classroom. It is widely

acknowledged that the view of teachers is a key element in successfully

implementing inclusive education (Avramidis & Norwich, 2002; Monsen &

Frederickson, 2004).

Third, the organisation of the class and school are factors which should be taken

into consideration within the scope of interventions. For instance, the attitude of

school staff in schools with much experience in inclusive education might differ

from that of a school staff who lack experience, which might influence a pupil’s

social participation. In addition, preconditions (like teaching aids, support staff)

could be relevant aspects.

A final factor concerns parent characteristics, as the outcome of inclusion seems

to depend to a large extent on the attitudes of parents of children with and

without special needs (Smith Myles & Simpson, 1990).

In concluding, when more research is aimed at studying the factors which

influence the social participation of pupils with special needs, and when these

factors are taken into consideration when drawing up interventions, this is

expected to contribute to optimising the situation of pupils with special needs in

inclusive classrooms.

8.6 References

Aboud, F.E. & Mendelson, M.J. (1996). Determinants of friendship selection and

quality: Developmental perspectives. In W.M.Bukowski, A.F. Newcomb &

W.W. Hartup (Eds.), The company they keep: Friendship in childhood and


Asher, S.R. & Coie, J.D. (1990). Peer rejection in childhood. Cambridge:

Cambridge University Press.

Asher, S.R., Parkhurst, T., Hymel, S. & Williams, G.A. (1990). Peer rejection and

loneliness in childhood. In S.R. Asher. & J.D. Coie, (Eds.), Peer rejection

in childhood (pp.253-273). Cambridge: Cambridge University Press.

GENERAL DISCUSSION

199

Avramidis, E. & Norwich, B. (2002). Teachers’ attitudes towards

integration/inclusion: a review of the literature. European Journal of


Chamberlain, B., Kasari, C. & Rotheram-Fuller, E. (2007). Involvement or

isolation. The social networks of children with autism in regular

classrooms. Journal of Autism and Developmental Disorders, 37, 230-242.



Haag: LCTI.









Education, 22(1), 15-30.




Karsten, T., Peetsma, T., Roeleveld, J. & Vergeer, M. (2001). The Dutch policy of

integration put to the test: differences in academic and psychosocial

development of pupils in special and mainstream education. European

Journal of Special Needs Education, 16(3), 193-205.

Kutnick, P. & Kington, A. (2005). Children’s friendships and learning in school:

cognitive enhancement through social interaction? British Journal of

Educational Psychology, 75, 521-538.






Education, 25(1), 90-101.

CHAPTER 8

200






people who experience difficulties in learning. In L. Florian, (Ed.), The

SAGE handbook of Special Education (460-474). London: SAGE

Publications.

Mand, J. (2007). Social position of special needs pupils in the classroom – a

comparison between German special schools for pupils with learning

difficulties and integrated primary school classes. European Journal of


Monsen, J.J. & Frederickson, N. (2004). Teachers’ attitudes towards

mainstreaming and their pupils’ perceptions of their classroom learning

environment. Learning Environments Research, 7, 129-142.







longitudinal study, Journal of Consulting and Clinical Psychology, 60(1),

80-87.






Pijl, S.J. (2005). Interventies gericht op sociale integratie: de training van sociale

vaardigheden voor leerlingen met beperkingen in het regulier onderwijs.

In B.F. van der Meulen, C. Vlaskamp & K.P. van den Bos (Eds.),

Interventies in de orthopedagogiek: bijdragen aan de wetenschap der

orthopedagogiek (pp. 122-135). Rotterdam: Lemniscaat.

Sale, P. & Carey, D.M. (1995). The sociometric status of students with disabilities

in a full-inclusion school. Exceptional Children, 62(1), 6-19.

GENERAL DISCUSSION

201

Schaffer, H.R. (1996). Social Development. Oxford: Blackwell Publishers.

Scheepstra, A.J.M. (1998). Leerlingen met Downs syndroom in de basisschool.

Groningen: Stichting Kinderstudies.

Schneider, B.H., Wiener, J. & Murphy, K. (1994). Children’s friendships: the giant

step beyond peer acceptance. Journal of Social and Personal Relationships,

11, 323-340.

Smith Myles, B. & Simpson, R.L. (1990). Mainstreaming modification preferences

of parents of elementary-age children with learning disabilities. Journal of

Learning Disabilities, 23(4), 234-239.



Walz, N.C. & Benson, B.A. (2002). Behavioral Phenotypes in Children with Down

Syndrome, Prader-Willi Syndrome, or Angelman Syndrome. Journal of

Developmental and Physical Disabilities, 14(4), 307-321.

CHAPTER 8

202

Summary

SUMMARY

SUMMARY

204

This thesis is aimed at the social participation of pupils with special needs in

regular primary education, in particular at developing an instrument for teachers

to assess the social participation of these pupils.

Chapter 1 describes the background and rationale of the study. Over the

past century, the Netherlands has built up a comprehensive system of special

education. During the last decades, however, the policy has been to promote the

inclusion of pupils with special needs in regular education. A large step was made

in August 2003 with the so-called Wet op de Expertise Centra (Centres of

Expertise Act), which enables pupils with auditory, speech/language, motor,


and/or psychiatric problems to qualify for a pupil-bound budget (the so-called

‘backpack’) and gives parents the opportunity to choose between regular and

special education for their child. An increasing number of parents have decided

to send their child with special needs to a regular school because they envisage

more opportunities for their offspring’s social participation.

Chapter 2 focuses on the first experiences with the pupil-bound budget in

regular primary education. A first impression is given of the cognitive, social and

social-emotional development of twenty, mainly young, pupils (average age 5;7

years), with a pupil-bound budget several months after their placement at a

regular primary school. It looks at the assessments of teachers, parents and

peripatetic teachers regarding the children’s development. Teachers’ satisfaction

about pupils’ cognitive development varied, but in the main they were satisfied

about pupils’ social and social-emotional development. A notable fact in all

assessments is that the parents were somewhat more positive, while the

peripatetic teachers were slightly more cautious when assessing pupils’

development. The outcome of a sociometric questionnaire, completed by all

pupils of groups with one or more pupils with special needs, provides a more

negative picture than that of parents, teachers and peripatetic teachers. This

showed that nine out of twenty pupils with special needs were typified as ignored

or rejected, although this was not mentioned by teachers, parents or peripatetic

teachers. In fact the results also showed that some of the pupils without special

needs were ignored or rejected and the percentage did not differ significantly

from the percentage of pupils with special needs who were.

Since the study was aimed at the initial experiences with pupil-bound

funding, it was impossible to compare the development of pupils with such a

SUMMARY

205

budget with the development of a control group. As an alternative, a panel of

independent assessors, consisting of five experienced peripatetic teachers, were

asked to judge the cognitive, social and social-emotional development of the

twenty pupils with special needs based on anonymous pupil dossiers. The

outcomes show that the panel had concerns about the development of 35 per

cent of the pupils, which deviates from the views of teachers, parents and, to a

lesser degree, the peripatetic teachers. On the other hand, the panel was

positive about an equal percentage of other pupils with special needs.

Chapter 3 concerns a literature review aimed at elucidating the social

dimension of inclusion in education. The review shows that frequently used

concepts like social integration, social inclusion and social participation are

considered synonymous by many researchers. In this study, social participation

was chosen as a central concept. Based on the literature review, it was

concluded that four key themes can be distinguished within the concept of social

participation: friendships/relationships, contacts/interactions, social self-

perception of the pupil and acceptance by classmates. Subsequently, on the

basis of these themes a model of social participation was constructed.

Chapter 4 describes how the model of social participation was used to

develop a teacher questionnaire intended to help teachers accurately assess the

social participation of pupils with special needs. The first experiences with the

quality of the Questionnaire, which comprised four subscales representing the

four key themes of social participation, are also described. The outcomes

concerning the psychometric qualities turned out to be promising, but should be

interpreted with caution because of the small sample. The results of a

confirmatory factor analysis (Multiple Group Method) largely supported the

division of social participation into four key themes. The reliability of the

Questionnaire turned out to be high, whereas the subscales’ reliability varied

from moderate to high. Based on these results, a few adaptations were made. It

was concluded that a large-scale study should be undertaken to examine the

quality of the Questionnaire more in-depth. Chapters 5 to 7 report on this.

Chapter 5 addresses the psychometric qualities of the Social Participation

Questionnaire. By means of a non-parametric item response analysis,

(Mokken Scale Analysis), the quality of the Questionnaire and its subscales were

examined. Based on the analysis results, ten statements were removed from the

Questionnaire, resulting in a definite version consisting of 24 statements. The

SUMMARY

206

outcomes of the analyses showed that the four subscales were intermediate to

strong. Mokken’s double monotonicity model (DMM) turned out to fit well for

each subscale. The subscale scores were on an ordinal scale and the separate

statements were invariantly ordered. Furthermore, the subscale scores are

comparable across pupils with and without special needs. In addition, the

outcomes of the analyses supported the model of social participation, in which

four key themes are distinguished. The Questionnaire as a whole and its

subscales were found to be reliable. Finally, evidence was obtained for the

Questionnaire’s discriminant validity.

Chapter 6 focuses on the convergent validity of the Social Participation

Questionnaire. The correlations between the Questionnaire’s four subscales and

external criteria turned out to be disappointing: some correlations were relatively

weak and some correlation patterns deviated from expectations. Consequently,

there was no conclusive evidence for the Questionnaire’s convergent validity. A

second order analysis, a confirmatory factor analysis using the LISREL program,

did provide evidence for the convergent validity of the Questionnaire. Based on

the outcomes of the Questionnaire’s psychometric qualities, it can be concluded

that the Questionnaire can become a valuable tool for teachers in helping them

notice problems regarding the social participation of pupils with special needs in

time. Further research into the utility of the Questionnaire and into the meaning

of scores is advisable.

Chapter 7 makes use of the research data employed for developing and

assessing the quality of the Social Participation Questionnaire. On the basis of

this research data the current state of affairs regarding the social participation of

pupils with special needs in regular primary education is described. The

outcomes are quite worrisome, as in three out of four important areas of social

participation, pupils with special needs perform less well than their fellow pupils

without special needs. However, the outcomes should be seen in context. The

vast majority of the pupils with special needs involved the study are being

accepted, have one or more friends in the classroom, have a reasonable number

of interactions with classmates and have a positive social self-perception.

In Chapter 8 the main findings of this study are summarized and the

methodological difficulties considered. In the light of this the implications for

educational policy and for educational practice are discussed and suggestions for

future research made.

Samenvatting

SAMENVATTING

SAMENVATTING

208

Dit proefschrift richt zich op de sociale participatie van leerlingen met

beperkingen binnen het regulier basisonderwijs, in het bijzonder op de

ontwikkeling van een instrument waarmee leerkrachten de sociale participatie

van deze leerlingen in kaart kunnen brengen.

Hoofdstuk 1 beschrijft de achtergrond van en aanleiding tot het

onderzoek. Nederland heeft in de afgelopen eeuw een omvangrijk systeem van

speciaal onderwijs opgebouwd. In de laatste decennia is echter beleid in gang

gezet om de inclusie van kinderen met beperkingen in het regulier onderwijs te

bevorderen. Een grote stap werd gezet met de Wet op de Expertise Centra

(WEC). Sinds de introductie van deze wet op 1 augustus 2003 komen kinderen

met auditieve, communicatieve, motorische, verstandelijke of meervoudige

beperkingen, als ook ernstige gedrags- emotionele en/of psychiatrische

problemen in aanmerking voor leerlinggebonden financiering (LGF), het

zogenaamde Rugzakje. Toekenning van een leerlinggebonden budget geeft

ouders de mogelijkheid zelf te kiezen voor deelname van hun kind aan regulier of

speciaal onderwijs. Een toenemend aantal ouders kiest voor plaatsing op een

reguliere school, waarbij met name de vergrote mogelijkheden voor sociale

participatie van hun kind de reden vormen voor deze keuze.

Hoofdstuk 2 is gericht op de eerste ervaringen met leerlinggebonden

financiering in het regulier basisonderwijs. Er wordt een indruk gegeven van de

cognitieve, sociale en sociaal-emotionele ontwikkeling van twintig overwegend

jonge leerlingen (gemiddelde leeftijd is 5;7 jaar) met een Rugzakje een aantal

maanden na de plaatsing op een reguliere basisschool. Het gaat daarbij om de

oordelen die leerkrachten, ouders en ambulant begeleiders hebben over de

ontwikkeling van de leerlingen. De tevredenheid van de leerkrachten over de

cognitieve ontwikkeling van de leerlingen is wisselend. Over de sociale en

sociaal-emotionele ontwikkeling van de leerlingen zijn de leerkrachten over het

geheel genomen tevreden. Opvallend in al deze beoordelingen is dat de ouders

vaak nog wat positiever oordelen dan de leerkrachten, terwijl de ambulant

begeleiders wat meer reserves laten blijken. De uitkomsten van een

sociometrische vragenlijst, ingevuld door alle leerlingen van groepen met één of

meer leerlingen met een beperking, leveren een negatiever beeld op dan dat van

ouders, leerkrachten en ambulant begeleiders. Uit de uitkomsten van deze lijst

blijkt dat negen van de twintig leerlingen met beperkingen als genegeerd of

verworpen worden getypeerd. Deze typeringen worden door geen van de

SAMENVATTING

209

leerkrachten, ouders of ambulant begeleiders genoemd. Overigens blijkt ook een

deel van de leerlingen zonder beperkingen te worden genegeerd of verworpen

volgens de uitkomsten van de sociometrische vragenlijst: de percentages

genegeerde en afgewezen leerlingen verschillen niet noemenswaardig tussen de

groep leerlingen met beperkingen en die zonder. Aangezien het onderzoek

gericht was op de eerste ervaringen met LGF, was het niet mogelijk om de

ontwikkeling van de leerlingen met een Rugzakje te vergelijken met een

normgroep. Als alternatief is een panel van externe beoordelaars, bestaande uit

vijf ervaren ambulant begeleiders, gevraagd op basis van geanonimiseerde

leerling-dossiers een oordeel te geven over de ontwikkeling van de twintig

leerlingen op cognitief, sociaal en sociaal-emotioneel gebied. De uitkomsten laten

zien dat het panel zich in enige mate zorgen maakte over 35% van de leerlingen,

wat in contrast staat met het oordeel van leerkrachten, ouders en in mindere

mate ambulant begeleiders. Anderzijds was het panel (zeer) positief gestemd

over eenzelfde percentage andere leerlingen met een beperking.

Hoofdstuk 3 betreft een literatuurverkenning waarbij het verhelderen van

de sociale dimensie van inclusie in het onderwijs centraal staat. Uit de

literatuurverkenning komt naar voren dat de vaak gebruikte termen ‘sociale

integratie’, ‘sociale inclusie’ en ‘sociale participatie’ door veel onderzoekers als

synoniemen worden beschouwd en als zodanig worden gehanteerd. In dit

onderzoek is sociale participatie gekozen als centraal begrip. Op basis van de

literatuurverkenning werd geconcludeerd dat vier thema’s kunnen worden

onderscheiden binnen sociale participatie: vriendschappen/relaties,

contacten/interacties, sociale zelfperceptie van de leerling en acceptatie door

klasgenoten. Aan de hand van deze thema’s is vervolgens een model van sociale

participatie ontworpen.

In hoofdstuk 4 wordt beschreven hoe het model van sociale participatie

wordt gebruikt voor het ontwikkelen van een vragenlijst voor leerkrachten. De

‘Vragenlijst Sociale Participatie’ is bedoeld om leerkrachten te ondersteunen bij

het maken van accurate beoordelingen van de sociale participatie van leerlingen

met beperkingen. In dit hoofdstuk worden de eerste bevindingen met betrekking

tot de kwaliteit van de vragenlijst beschreven. De vragenlijst bestaat uit vier

subschalen waarin de vier hoofdthema’s van sociale participatie worden

geoperationaliseerd. De uitkomsten met betrekking tot de psychometrische

eigenschappen van de vragenlijst bleken veelbelovend te zijn, maar dienen met

SAMENVATTING

210

enige voorzichtigheid geïnterpreteerd te worden in verband met het kleine aantal

respondenten. De uitkomsten van een confirmerende factoranalyse (Multiple

Groep Methode) ondersteunden grotendeels de verdeling van sociale participatie

in vier hoofdthema’s. De betrouwbaarheid van de vragenlijst als geheel was

hoog, terwijl de betrouwbaarheid van de vier subschalen uiteenliep van matig tot

hoog. Op basis van deze uitkomsten werden enkele wijzigingen in de vragenlijst

aangebracht en werd geconcludeerd dat door middel van grootschaliger

onderzoek de kwaliteit van de vragenlijst verder onderzocht zou moeten worden.

In de hoofdstukken 5 tot en met 7 wordt hiervan verslag gedaan.

Hoofdstuk 5 richt zich op de psychometrische kwaliteiten van de

Vragenlijst Sociale Participatie. Door middel van een non-parametrische item

respons analyse (Mokken-analyse) is de kwaliteit van de vragenlijst als geheel en

van de vier afzonderlijke subschalen onderzocht. Op basis van de uitkomsten van

de analyses is een tiental items uit de vragenlijst verwijderd, resulterend in een

definitieve vragenlijst bestaande uit 24 items. De uitkomsten van de analyses

laten zien dat de vier subschalen gematigd sterk tot sterk zijn. Het dubbele

monotonie model (DMM) van Mokken bleek van toepassing te zijn op iedere

subschaal. De subschaalscores liggen op een ordinale schaal en de afzonderlijke

items zijn invariant geordend. Daarnaast zijn de subschaalscores vergelijkbaar

voor leerlingen met en zonder beperkingen. Bovendien wordt het model van

sociale participatie, waarin de vier hoofdthema’s worden onderscheiden,

ondersteund door de uitkomsten van de analyses. De betrouwbaarheid van de

vragenlijst als geheel bleek hoog te zijn en ook de vier subschalen zijn

betrouwbaar bevonden. Tenslotte bleken er sterke aanwijzingen te zijn voor de

discriminante validiteit van de vragenlijst.

In hoofdstuk 6 staat de convergente validiteit van de Vragenlijst Sociale

Participatie centraal. De correlaties tussen de vier subschalen van de vragenlijst

en externe criteria bleken tegen te vallen: een aantal correlaties was relatief laag

en enkele correlatiepatronen weken af van de verwachtingen. Hierdoor was er

geen afdoende steun voor de convergente validiteit van de vragenlijst. Een

secundaire analyse, te weten een confirmerende factoranalyse met behulp van

LISREL, verschafte echter wel aanwijzingen voor de convergente validiteit van de

Vragenlijst Sociale Participatie. Op basis van de uitkomsten betreffende de

psychometrische kwaliteiten van de Vragenlijst Sociale Participatie, kan

geconcludeerd worden dat de vragenlijst voor leerkrachten een waardevol

SAMENVATTING

211

hulpmiddel kan worden bij het tijdig signaleren van problemen op het gebied van

sociale participatie van leerlingen met beperkingen. Nader onderzoek naar onder

andere de bruikbaarheid van de vragenlijst en naar de betekenis van scores is

wenselijk.

Hoofdstuk 7 maakt gebruik van onderzoeksmateriaal dat gebruikt is bij

de ontwikkeling en kwaliteitstoetsing van de Vragenlijst Sociale Participatie. Aan

de hand van dit onderzoeksmateriaal wordt de actuele stand van zaken met

betrekking tot de sociale participatie van leerlingen met beperkingen binnen het

regulier basisonderwijs beschreven. De uitkomsten zijn in enige mate

zorgwekkend, aangezien leerlingen met beperkingen op drie van de vier

hoofdthema’s van sociale participatie lager scoren dan hun medeleerlingen

zonder beperkingen. Enige nuancering is echter op zijn plaats. Veruit het

merendeel van de leerlingen met beperkingen wordt geaccepteerd en heeft één

of meer vrienden in de groep, een acceptabel aantal interacties met klasgenoten

en een positief sociaal zelfbeeld.

In hoofdstuk 8 worden de belangrijkste bevindingen uit dit proefschrift

samengevat en de methodologische problemen van het onderzoek besproken. In

het licht hiervan wordt een discussie gevoerd over de implicaties van dit

onderzoek voor het onderwijsbeleid en voor de onderwijspraktijk en worden

aanbevelingen voor vervolgonderzoek gedaan.

SAMENVATTING

212

Dankwoord

DANKWOORD

DANKWOORD

214

Het is zover: het proefschrift is af. Een groot aantal mensen heeft direct of

indirect bijgedragen aan de totstandkoming van dit proefschrift. Op deze plek wil

ik hen allen hiervoor hartelijk bedanken. Een aantal personen wil ik in het

bijzonder noemen.

In de eerste plaats gaat mijn dank uit naar de drie personen die mij de afgelopen

jaren hebben begeleid: prof. dr. Han Nakken, prof. dr. Sip Jan Pijl en dr. Els van

Houten.

Han, ik prijs mij gelukkig met jou als 1e promotor. Na je emeritaat had je het

rustiger aan kunnen gaan doen, maar voor mijn gevoel is het tegenovergestelde

gebeurd. Je hebt niet alleen de grote lijn van mijn onderzoek in de gaten

gehouden, maar ook al mijn teksten (van de eerste pogingen tot de definitieve

versies) grondig gelezen en van commentaar voorzien. Jouw betrokkenheid en

inzet heb ik zeer gewaardeerd, evenals de vele kopjes thee waar jij mij vrijwel

ieder overleg van voorzag. Ik bewonder jouw enorme hoeveelheid vakkennis,

jouw scherpe, doch opbouwende manier van feedback geven en jouw passie voor

de wetenschap. Dat laatste werkt aanstekelijk!

Sip Jan, als 2e promotor was je sterk betrokken bij mijn onderzoek. Ik heb grote

bewondering gekregen voor jouw vakkennis en deskundigheid op het gebied van

inclusief onderwijs. Jouw vermogen om (in mijn ogen) tegenvallende resultaten

te relativeren en de zonnige zijde ervan te belichten heb ik als zeer waardevol

ervaren. Daarnaast heb ik erg veel plezier beleefd aan het samen met jou

deelnemen aan conferenties in Glasgow, Genève, Gent en Göteborg.

Els, jij was als co-promotor aan mijn onderzoek verbonden. Ik wil jou allereerst

hartelijk bedanken voor het doorsturen van de vacature in het najaar van 2003.

Als jij niet aan mij had gedacht als potentiële Promovendus, had ik dit

proefschrift niet kunnen schrijven. Daarnaast ben ik je niet alleen dank

verschuldigd voor de betrokkenheid bij mijn onderzoek, maar ook voor je

ondersteuning bij het begeleiden van studenten en het voorbereiden van

colleges. Ik heb veel van je geleerd.

Daarnaast wil ik mijn collega’s uit het AIO-huis bedanken voor de fijne tijd. Ik

heb het mede dankzij jullie de afgelopen jaren enorm naar mijn zin gehad. Van

het vieren van verjaardagen, het samen lunchen en het met elkaar uiteten gaan

heb ik erg genoten. In het bijzonder wil ik Marianne Punter bedanken. Marianne,

gedurende drie jaar waren wij kamergenootjes. Bedankt voor je hulp, je

DANKWOORD

215

luisterend oor, je boekentips en je gezelligheid. Ik ben zeer blij dat je mijn

paranimf wilt zijn.

De scholen die hebben deelgenomen aan het onderzoek mogen niet onvermeld

blijven. Zij waren onmisbaar voor het onderzoek Alle leerlingen, leerkrachten,

intern begeleiders, ambulant begeleiders en ouders die hebben deelgenomen aan

het onderzoek wil ik heel hartelijk bedanken voor hun medewerking.

De vele schoolbezoeken zou ik nooit alleen hebben kunnen behappen. Met hulp

van studenten is het gelukt de scholen die door heel Nederland verspreid lagen

te bezoeken. In chronologische volgorde wil ik de studenten bedanken die in het

kader van hun doctoraalscriptie of masterthesis een bijdrage hebben geleverd

aan het onderzoek: Christel Koldeweij, Annet Hessels, Evelien Klei, Sandra

Meijer, Maaike Mollema en Maria van Gosliga. Bedankt voor jullie bijdrage aan

mijn onderzoek. Speciale dank gaat uit naar Ysanne Drijber. Gedurende het half

jaar dat jij onderzoeksstagiaire was heb je veel en goed werk verricht. Bedankt

voor je inzet. Paula Dijk en Ellen Hendriks hebben in de functie van

studentassistent een grote bijdrage geleverd aan het verzamelen van data. Het

vroege opstaan en de lange reistijden (inclusief vertragingen) hebben jullie prima

doorstaan. Ik heb jullie inzet enorm gewaardeerd.

Collega’s prof. dr. Alexander Minnaert, dr. Henk Lutje Spelberg en dr. Marieke

Timmerman wil ik bedanken voor hun adviezen met betrekking tot statistische

bewerkingen, het bieden van ondersteuning bij analyses en het meewerken aan

artikelen.

Dr. Per Frostad wil ik bedanken voor zijn hulp bij het leren werken met het

statistische programma Negopy. Per, thanks for your help and for showing me

the beautiful Norwegian mountains.

Mijn dank gaat uit naar Lynn George en Ruth Rose voor het redigeren van de

artikelen die zijn opgenomen in dit proefschrift. Dankzij hun kritische feedback

zijn de teksten naar een hoger niveau getild.

Hester Nijhoff wil ik bedanken voor het maken van de illustratie op de omslag die

de finishing touch vormt van mijn proefschrift.

Mijn laatste woorden van dank zijn gericht aan het thuisfront. Lieve papa, mama,

San, Remco, Ben en Gerdien, bedankt voor jullie interesse in mijn onderzoek.

Mama, wanneer ik zenuwachtig was voor conferenties, heb ik me erg gesteund

gevoeld door jouw lieve sms-jes. Ook mijn andere familie en vrienden en

vriendinnen wil ik bedanken voor de belangstelling voor mijn onderzoek. In het

DANKWOORD

216

bijzonder wil ik Inge bedanken. Ik heb genoten van onze vele thee-, eet-, skate-,

squash-, en hardloopdates de afgelopen jaren. Zij vormden een uitstekende

afwisseling van het werk.

Erik, mijn slotwoorden zijn voor jou. Straks ga ik met de doctorstitel aan de haal,

maar eigenlijk verdien jij ‘em ook een beetje. Als ik ingewikkelde vragen had

over de verwerking van mijn data en dacht dat ik onoverkomelijke problemen

had, bedacht jij vaak ingenieuze oplossingen. Het was fijn om erop te kunnen

vertrouwen dat het uiteindelijk altijd wel goed zou komen. Het is een

geruststellende gedachte dat jij mij ook in de laatste fase van mijn promotie, in

de functie van paranimf, ter zijde zal staan. Bedankt daarvoor, maar bovenal

bedankt voor jouw vertrouwen in mij.

Documents

University of Groningen Evaluating the social outcomes of ... · Evaluating the social outcomes of inclusive education a study on the social participation of pupils with special needs