Differential influences of affective factors and contextual factors By Prof. Shek Kam Tse

Science of Learning Strategic Research Theme

Meeting of Minds Series :Interrogating Hong Kong

students’ superior performance in PIRLS

(Progress in International Reading Literacy Study)

A multilevel analysis of PIRLS2011 data from Hong Kong:

Differential influences of affective factors and contextual factors on high-

proficiency readers and low-proficiency readers

Prof. Shek-kam Tse and Dr. Xiao-yun Xiao

Differential influences of affective factors and contextual factors on high-proficiency readers and low-proficiency readers: a multilevel analysis of PIRLS data from Hong Kong

Prof. Shek-kam Tse and Dr. Xiao-yun Xiao

Prof. TSE Shek Kam Dr. LAM Wai Ip Dr. CHEUNG Wai Ming Dr. LOH Ka Yee

PIRLS Research Team Members

5

Progress in International Reading Literacy Study (PIRLS)

(PIRLS) is organized by the International Association for Evaluation of Educational Achievement (IEA)

Conducting regular international research on assessments of reading literacy and the factors associated with its acquisition globally.

Focuses on the achievement of fourth grade young children

PIRLS has been conducted on a regular 5-year cycle since 2001.

The Hong Kong PIRLS study was funded by the Education Bureau of Hong Kong Government

6

PIRLS conducted in the last ten years

PIRLS 2001

PIRLS 2006

PIRLS 2011

PIRLS 2016

7

Countries 2001

2006

2011

Australia Azerbaijan Chinese Taipei Hong Kong SAR Indonesia Iran Israel New Zealand Oman Qatar Saudi Arabia Singapore United Arab Emirates

Asia & OceaniaNo. of Participants: 45New comers: 17Second time: 10Third time: 18

45 countries or regions participating in PIRLS 2011

8

PIRLS 2011 ParticipantsCountries 200

12006

2011

Austria Belgium (French)

Bulgaria Croatia Czech Republic Denmark England Finland France Georgia Germany Hungary Ireland

Europe Countries 2001

2006

2011

Italy Lithuania Malta Netherlands Northern Ireland Norway Poland Portugal Romania Russia Slovak Republic Slovenia Spain Sweden

9

Countries 2001

2006

2011

Canada Colombia Trinidad and Tobago

United States

Americas

Countries 2001

2006

2011

Morocco

Africa

PIRLS 2011 Participants

10

Major focuses of PIRLS

Reading purposes and processes of reading comprehension

Student reading behaviors and attitudes

Factors affecting student reading achievement and behaviors(Home, teachers, schools, curriculum and community)

11

Contexts of Developing Children’s reading Literacy

(PIRLS 2011 Assessment Framework, Page 35)

Reading assessment framework

12

Purpose for ReadingLevelsof Reading Comprehension

Acquire and Use Information Literary Experience

Level 4: Examine and evaluate content, language, and textual elements

Level 3: Interpret and integrate ideas and informationLevel 2: Make straightforward inferences

Level 1: Focus on and retrieving explicitly stated information

Retrieving and Straightforward Inferencing

(Low level)

Interpreting, Integrating, and Evaluating

(High level)

13

Research design and methodology

Instruments Test Booklet Questionnaires

School Teacher Learn to Read (Parents) Students

14

School Questionnaire

School Questionnaire

Completed by: School principal of participating schools

It investigates: school characteristics Instructional time resources and technology parental involvement school climate for learning teaching staff the role of the principal students’ reading readiness

15

Teachers Questionnaire

Teacher Questionnaire

Completed by: Chinese language or reading teacher of the sampled P4 class

It investigates: classroom contexts for developing reading

literacy characteristics of the class tested, activities for teaching reading and

promoting the development of students' reading literacy;

classroom resources assessment practices and home-school connections teachers collaboration with others professional development teacher education and training

16

Parents Questionnaire

Learn to Read Survey (Parents)

Completed by: parents or primary caregivers of each student

It investigates: child-parent literacy interactions home literacy resources parents’ reading habits and

attitudes home-school connections basic demographic and

socioeconomic information of the students’ family

17

Students Questionnaire

Student Questionnaire

Completed by: student who takes the PIRLS reading test.

It investigates: Student readiness to learn Student motivation Student self-concept Student reading literacy

behaviors Home resources for learning Language(s) spoken in the home

18

Ranking Country/Region Avg. Scale Score1 Russian Federation 5652 Hong Kong SAR 5644 Singapore 558

10 Sweden 54918 United States 54019 England 53922 Chinese Taipei 535

500PIRLS Scale Centerpoint

Ranking Country/Region Avg. Scale Score1 Sweden 5613 England 5539 United States 542

14 Russian Federation 52814 Hong Kong SAR 52814 Singapore 528


PIRLS 2001, 2006 & 2011Trend Study

PIRLS 2001 PIRLS 2006+36

19

Ranking Country/Region Avg. Scale Score1 Hong Kong SAR 5712 Russian Federation 5683 Singapore 5675 United States 5567 Chinese Taipei 5538 England 552

12 Sweden 542PIRLS Scale Centerpoint 500

Ranking Country/Region Avg. Scale Score1 Russian Federation 5652 Hong Kong SAR 5644 Singapore 558

10 Sweden 54918 United States 54019 England 53922 Chinese Taipei 535


PIRLS 2006 PIRLS 2011

PIRLS 2001, 2006 & 2011Trend Study

+7

20

Reading achievement of 45 participating

countries in PIRLS 2011

1. Our primary goal in conducting the current study was to examine the relationship between affective factors, home environment, school context, and reading attainment among Chinese good and poor readers in a subsample of the large-scale dataset gathered during the PIRLS 2011 survey.

2. Affective factors include student’s reading attitude, student’s reading motivation and student’s reading-concept.

3. We also particularly wanted to compare the strength of this relationship between good and poor readers directly by testing, through use of a multilevel logit regression model, the extent to which these variables predicted the likelihood of the students in the sample being good or poor readers.

Aim of Study

In PIRLSs’ reports of 2001, 2006 and 2011, students’ reading attitudes, motivations, and self-concepts were found to be associated with their reading attainment across countries globally (Mullis et al. 2012; Mullis et al. 2003; Mullis et al. 2007), and might be varied with cultures.

The socioeconomic status (SES) of the family is another strong predictor of academic achievement (Bradley & Corwyn, 2002; McCulloch & Joshi, 2001; Shonkoff & Phillips, 2000; Sirin, 2005).

School SES is another variable strongly associated with academic performance in many studies (Ogle et al. ,2003).

Factors and Reading Attainment (1)

Peer bullying at school and school safety and order are key factors associated with academic achievement and progress (Glew et al., 2005).

Factors for analysis:1. Student’s reading attitude2. Student’s reading motivation3. Student’s reading-concept4. Home SES5. School SES6. Peer bullying7. School bullying8. School Safety and order

Factors and Reading Attainment (2)

Methods (1)Measurements on Concepts





Data from 3,875 Hong Kong SAR Grade 4 students participating in an international comparative assessment were analyzed. Students are classified as: High proficiency: at or above 550 Average: at or above 475 and below 550 Low proficiency: : below 475 (Defined by international benchmarks of PIRLS)

Multilevel logit regression analysis was used to model the relationship between affective factors (i.e., reading attitude, reading motivation, and reading self-concept) peer bullying, family context (i.e., home socioeconomic status/SES), and school context (i.e., school SES, school bullying, and school safety and order).

Methods (6)Samples and analysis

Methods (7)

Methods (8)

Results & Conclusions (1)Fixed Effects on Factors: Low Proficiency Readers

Fixed effects Coefficient SE t-ratioApproximate

df

For readers with low proficiency

Intercept (γ00(1)) -1.93 0.12 -15.66 *** 128

School SES (γ01(1)) -1.22 0.77 -1.59 13

School bullying (γ02(1)) 0.06 0.09 0.67 64

School safety and order slope (γ03(1)) -0.06 0.05 -1.08 53

Students’ reading attitudes Intercept (γ10(1)) 0.01 0.02 0.36 102

Students’ reading motivation Intercept (γ20(1)) 0.02 0.03 0.81 59

Students’ reading self-concept Intercept (γ30(1)) -0.09 0.03 -2.96 ** 21

Home SES Intercept (γ40(1)) -0.17 0.26 -0.64 21

Peer bullying Intercept (γ50(1)) 0.06 0.02 3.03 ** 131

Results & Conclusions (2)Fixed Effects on Factors: High Proficiency

Readers

Fixed effects Coefficient SE t-ratioApproximate

df

For readers with high proficiency

Intercept (γ00(1)) 0.91 0.08 11.02 *** 77

School SES (γ01(1)) 1.21 0.38 3.19 ** 28

School bullying (γ02(1)) -0.17 0.06 -2.84 ** 128

School safety and order slope (γ03(1)) 0.00 0.03 0.12 128

Students’ reading attitudes Intercept (γ10(1)) 0.05 0.02 2.44 * 84

Students’ reading motivation Intercept (γ20(1)) -0.05 0.02 -2.68 ** 99

Students’ reading self-concept Intercept (γ30(1)) 0.16 0.02 8.74 *** 53

Home SES Intercept (γ40(1)) -0.03 0.13 -0.23 90

Peer bullying Intercept (γ50(1)) -0.03 0.02 -1.96 * 68

The likelihood of being a reader with high proficiency was found to be associated with:1. reading attitude, 2. reading motivation, 3. reading self-concept, 4. peer bullying, 5. school bullying, and 6. school SES.

Whereas the likelihood of being a reader with low proficiency was associated with reading self-concept and peer bullying only.

Results & Conclusions (3)

Individual Effect

School Effect

A particularly interesting finding in this study was the negative relationship between reading motivation and the probability of students being high-proficiency readers.

In our study, school SES was a significant predictor of students being high-proficiency readers, but NOT low-proficiency readers.

Our findings suggest that peer bullying behaviors had a role to play in the reading development of readers with high proficiency and readers with low proficiency.


Many researchers have found a positive association between family SES and academic achievement in general. However, in our study neither the probability of readers with low proficiency nor the probability of readers with high proficiency was predicted by family SES when the effects of affective factors and school factors were taken into account.

These findings suggest that the influence of family context on reading attainment is not as strong as that of school context or student characteristics (such as reading-related attitude, motivation, and self-concept examined in this study).


Thank you!