Do children do what they say? Responses to hypothetical and real-life social problems in children with mild intellectual disabilities and behaviour problems

Journal of Intellectual Disability Research doi: 10.1111/j.1365-2788.2005.00674.x

pp

–

©

Blackwell Publishing Ltd

419

Blackwell Science, LtdOxford, UKJIRJournal of Intellectual Disability Research

-

Blackwell Publishing Ltd,

6419433

Original Article

Do children do what they say?M. van Nieuwenhuijzen

et al.

Correspondence: Maroesjka van Nieuwenhuijzen, Department of Special Education, Utrecht University, PO Box

,

TC Utrecht, the Netherlands (e-mail: [email protected]).

Do children do what they say? Responses to hypothetical and real-life social problems in children with mild intellectual disabilities and behaviour problems

M. van Nieuwenhuijzen,

1

E. R. Bijman,

1

I. C. W. Lamberix,

1

L. Wijnroks,

1

B. Orobio de Castro,

2

A. Vermeer

1

& W. Matthys

1,3

1

Department of Special Education, Utrecht University, Utrecht, the Netherlands

2

Department of Developmental Psychology, Utrecht University, Utrecht, the Netherlands

3

Department of Child and Adolescent Psychiatry, and Rudolf Magnus Institute of Neuroscience, University Medical Center, Utrecht, the Netherlands

Abstract

Background

Most research on children’s social problem-solving skills is based on responses to hypo-thetical vignettes. Just how these responses relate to actual behaviour in real-life social situations is, how-ever, unclear, particularly for children with mild intellectual disabilities (MID).

Method

In the present study, the spontaneous and selected responses of

children with MID to hypo-thetical situations from the Social Problem-Solving Test for children with MID (SPT-MID) were com-pared to their actual behaviour in comparable staged standardized real-life conflict situations. Correlations to externalizing behaviour problems were assessed using the Teacher’s Report Form (TRF).

Results

The results show children with MID and accompanying externalizing behaviour problems to behave more aggressively in the staged real-life con-flicts and provide more spontaneous aggressive responses to the hypothetical vignettes than children

with MID and no accompanying externalizing beha-viour problems; they did not, however, select more aggressive responses from the hypothetical options provided. A moderate correlation was found between the aggressiveness of the spontaneous responses in the hypothetical situations and actual behaviour in the staged real-life situations. In addition, both the spontaneous aggressive responses under hypothetical circumstances and the actual aggressive behaviour under staged real-life circumstances were related to teacher-rated aggressive behaviour in the classroom.

Conclusions

It is concluded that the hypothetical vignettes from the SPT-MID do provide information on both the actual behaviour and knowledge of social problem-solving skills of children with MID.

Keywords

aggressive behaviour, mild intellectual disabilities, social information processing, social problem solving

Introduction

Many studies have shown children with aggressive behaviour problems to respond differently to hypo-

Journal of Intellectual Disability Research

M. van Nieuwenhuijzen

et al

. •

Do children do what they say?420

©



,

–

thetical social problems comparing with other chil-dren. More specifically, children with aggressive behaviour problems have poor social problem-solving skills: they are more likely to respond aggressively than other children in social problem situations and also believe that aggressive behaviour is more effec-tive than other forms of responses (Crick & Dodge

). There is also some evidence that children with mild intellectual disabilities (MID) and concomitant behaviour problems are more likely to respond aggressively to hypothetical social problems than chil-dren with MID and no accompanying behaviour problems (Leffert & Siperstein

). Very few stu-dies have been conducted with this specific popu-lation although children with MID generally show more behaviour problems than children with normal intellectual abilities (Linna

et al.

; Dekker

et al.

).Social problem-solving skills are generally assessed

by posing questions about hypothetical situations, such as ‘What would you do if . . . happened to you?’ The problematic situations are typically described out loud, presented via an audiocassette (Dodge & Tomlin

; Quiggle

et al.

; Orobio de Castro

et al.

), presented via video vignettes (Dodge

et al.

; Lochman & Dodge

; Leffert & Si-perstein

; Matthys

et al.

), or presented using pictures of social problems (Healy & Master-pasqua

; Dodge

et al.

). These procedures provide information on the social problem-solving skills of children by requiring them to reflect on their behaviour. However, it is not at all clear whether this research method measures the social problem-solving skills actually used by children during daily life (Crick & Dodge

).One can ask whether the responses of children to

hypothetical vignettes are consistent with the beha-viour they may display during real-life problem situ-ations. A long research tradition concerned with the relationships between intended and actual behaviour has shown only moderate correlations for adults (Ajzen & Fishbein

). Very little is known about this relationship in children and even less about chil-dren – or adults – with MID.

The aim of the present study is therefore to inves-tigate the relationships between responses provided in hypothetical problematic situations, so-called hypothetical responses, and responses occurring in similar real-life situations, so-called actual behaviour.

Many researchers have commented that the findings obtained under hypothetical circumstances need to be supplemented with findings from real-life circum-stances (Shantz & Shantz

; Vitaro & Pelletier

; Cuperus

; Underwood & Galen

; Orobio de Castro

et al.

b). However, such stud-ies are rare. To our knowledge no such studies con-cerned children with MID, therefore the studies reviewed below all concern children from the normal population.

In some studies children have been asked to role-play how they would respond to hypothetical prob-lematic situations using glove puppets and toys (Mize & Cox

; Zahn-Waxler

et al.

; Mize & Ladd

). The results of these studies have proved incon-sistent, however. In the study by Mize & Ladd (

), a clear relationship was found between the children’s role-play behaviour and both behavioural observa-tions and teacher ratings of their behaviour but not their hypothetical responses to problematic situa-tions. In contrast, in the study by Mize & Cox (

), a relationship was found between hypothetical responses and both behavioural observations and evaluations by others. It may be that role-playing is too artificial and unrealistic to be used as a measure of real-life behaviour, as both Mize & Cox (

) and Vitaro & Pelletier (

) have suggested.An alternative method of assessment is to observe

actual behaviour during everyday situations. This, however, is a very time-consuming method of data collection and, perhaps more importantly, yields non-standardized data. That is, children with behaviour problems are often rejected, associate with deviant peer groups, and are frequently involved in conflicts with peers (Coie & Dodge

). The daily social interactions of such children therefore differ systematically from the daily social interactions of their peers, which makes comparison of their behaviour in nonstandardized social interactions problematic.

A more standardized procedure involves the con-trived playgroup. In this procedure, four to six chil-dren are placed together in a room containing age-appropriate toys. The children’s behaviour is then observed and recorded on video for several sessions on a number of different days (Coie & Kuperschmidt

; Dodge

; Dodge

et al.

; Schwartz

et al.

; Schwartz

et al.

; Coie

et al.

). Using this procedure, it has proved possible to distinguish


M. van Nieuwenhuijzen

et al

. •

Do children do what they say?421

©



,

–

the different patterns of social interaction associated with: different types of social status (Coie & Kuper-schmidt

; Dodge

; Schwartz

et al.

), different dyadic relations (Dodge

et al.

; Coie

et al.

), and different forms of aggression inclu-ding proactive vs. reactive aggression (Schwartz

et al.

). While the contrived playgroup procedure enables observation of spontaneous play behaviour, a drawback is no control over the play situation and total dependence on the interactions that sponta-neously arise between the children. In order to compare behaviour and responses, however, the same or at least highly similar situations must occur and further standardization appears to be necessary.

For the assessment of social problem-solving skills, the most informative situations are obviously those in which actual social problems arise. Situations of provocation and situations in which a child is being slighted or excluded by other children have been found to be most problematic for children, followed by coping with competition and coping with the social expectations of peers and teachers (Dodge

et al.

; Matthys

et al.

). Thus, ideally, research should involve actual responses to standar-dized real-life situations involving provocation, disad-vantage, or exclusion.

Very little research has been conducted to date using standardized real-life problematic social situa-tions. Lochman (

) and Lochman & Dodge (

) once used real-life situations in which coping with competition played an important role and asked pairs of children to interpret and respond to a hypo-thetical situation with each child presented a different perspective. The interpretations and responses of each child were then repeated back to him or her so that the child clearly knew what his or her opinions were. Thereafter, the two children were asked to dis-cuss the problem with each other and each child was told that he or she should try to win by convincing the other of the rightness of his or her ideas. In such a manner, a standardized competitive real-life situa-tion was created. The behaviour of the children during the discussion was then scored by the investigators and further evaluated by the children themselves. Both aggressive and nonaggressive chil-dren were found to have poor insight into their own behaviour in conflict situations. Both groups under-estimated the degree of aggression by the aggressive child and overestimated the degree of aggression by

the nonaggressive child. In addition, Lochman & Dodge (

) used a maze game requiring a marble to be rolled into a hole at the end of a maze without falling into other holes. The maze board was con-structed in such a manner that it had to be held by two handles and cooperatively moved; each child was asked to hold one handle and the two children had to cooperate to be successful. The researchers deli-berately devised a difficult situation to also allow the exhibition of competitive behaviour. The findings proved similar to those for the discussion task. In other words, the children showed distorted percep-tions of their own and the other children’s behaviour under both competitive and cooperative circum-stances. The results also show reflection on one’s own behaviour to be difficult for children.

Two studies by Underwood and colleagues (Underwood

et al.

; Underwood & Bjornstad

) used real-life provocation situations. Children were first placed at a disadvantage and then provoked by another child, who was a trained actor. More specifically, the children were asked to play a com-puter game that was programmed to give the actor an advantage and make the participant lose. In addi-tion, the actor was trained to make provocative com-ments. The investigators scored the verbal responses, facial expressions, and gestures of the participants. Only low to moderate agreement (

r

s of

.

to

.

) was found for the emotional behaviour demonstrated by the children and their self-reports of emotions (Underwood & Bjornstad

).The above studies using standardized real-life

problem situations show children’s perceptions of their own behaviour and emotions to not be

% consistent with their actual behaviour. However, no direct comparison was made between the actual behaviour of the children and hypothetical responses, which lie somewhere between self-reflection and actual behaviour. Therefore it is not clear whether the hypothetical responses would be consistent with actual behaviour of the children. Only three studies have – to our knowledge – assessed both hypothetical responses and actual behaviour in comparable con-flict situations (Dodge

; Vitaro et al. ; Vitaro & Pelletier ). Real-life situations were found to better differentiate between children with and with-out behaviour problems than hypothetical situations. However, the correlations between hypothetical responses and actual behaviour were not reported.


M. van Nieuwenhuijzen et al. • Do children do what they say?422

© Blackwell Publishing Ltd, Journal of Intellectual Disability Research , –

Only Dodge () reported moderate correlations between the number of aggressive responses and behaviour in a real-life situation.

In sum, children’s behaviour in real-life social si-tuations has been studied, but little effort has been made to standardize the situations presented. In those studies using more or less standardized situa-tions, moreover, direct comparison of the children’s hypothetical responses to their real-life responses or behaviour has not been undertaken. Given the amount of research conducted on the social problem-solving skills of children and the prominence of social problem-solving skills in theory, assessment, and intervention, the lack of insight into the relationship between hypothetical responses and actual behaviour is quite problematic. In addition, research along these lines has yet to be conducted with children with MID, which leaves the validity of such research methods for use with children with MID completely open.

In order to address these issues, the relationship between the hypothetical responses and actual beha-viour of children with MID was compared using a number of staged real-life conflict situations made as similar as possible to a number of hypothetical con-flict situations. The hypothetical situations were pre-sented in the form of video vignettes from the Social Problem-Solving Test (SPT) (Cuperus ; Matthys et al. ), which has been shown to distinguish between children with externalizing behaviour disor-ders (oppositional defiant disorder/conduct disorder and attention deficit hyperactivity disorder; ODD/CD and ADHD) and children without such disorders (Matthys et al. ). Video presentation of the hypo-thetical situations was chosen because children with MID may have particular difficulties with the com-prehension of verbal information. Information was also collected on both the spontaneous responses of the children and the responses selected from the three alternatives presented on video.

In the studies reviewed above, the relationship between the hypothetical responses and actual beha-viour in real-life situations was determined by testing whether groups differing in behaviour problems also differed on these measures (Mize & Ladd ; Mize & Cox ; Vitaro & Pelletier ). A more direct test of this relationship would be to correlate hypo-thetical responses and actual behaviour. In the present study, it was decided to use both approaches. We first examined the capacity of both hypothetical

and real-life situations to discriminate between chil-dren with MID and accompanying or no accompa-nying behaviour problems. We then assessed the correlations between the aggressiveness of the spon-taneous responses and the selected responses of the children to the hypothetical situations, on the one hand, and the actual behaviour of the children in the staged real-life conflict situations, on the other hand. In addition, the correlations between these different variables and teacher-ratings of aggressive behaviour within the classroom were also examined. We expected group differences to appear for both the hypothetical and real-life situations. We also expected positive correlations between the spontaneous responses and the selected responses of the children to the hypothetical situations, on the one hand, and the actual behaviour of the children in the staged real-life situations, on the other hand. Finally, we expected positive correlations to appear between the responses of the children under both the hypothetical and real-life circumstances and the teacher ratings of aggressive classroom behaviour.

Method

Participants

A total of children attending seven special-education schools in different regions of the Nether-lands participated in the study. The children came from different classes of six elementary schools for special education and one secondary school for spe-cial education. In , as part of the educational policy ‘Together to school again’ (Weer samen naar school), schools for children with MID and schools for children with learning disabilities were combined to form schools for special education. The children in the present study came from the earlier schools for children with MID. For attendance, the schools require below-average intellectual functioning (i.e. IQ range of –).

The children had a mean age of . years (SD = .) and a mean intelligence score of (SD = .). Sixty-four per cent were boys, which is comparable to the total population of children in special education in the Netherlands (% boys; CBS ).

After parental permission was obtained for parti-cipation in the study and video recording, the chil-




dren were assigned to one of three groups depending on their scores on the Teacher’s Report Form (TRF) (Achenbach ; for the Dutch version, see Verhulst et al. ) (also see Measures below). A total of children with scores falling in the clinical range (T-score > ) on the broadband scale for Externalizing behaviour problems formed the group of children with externalizing behaviour problems. A total of children with scores falling outside the clinical range (T-score £ ) formed the group of children without externalizing behaviour problems. In addition, a se-parate ‘game partner’ or neutral group of children was formed. This group included only children who showed no extreme aggressive and no withdrawn scores on the TRF, that is, scores on the scales Aggression and Withdrawn not falling in the clinical and borderline range (T-scores < ). These ‘neutral’ children were then assigned randomly to the partici-pants in the provocative situations but not included further in the data analyses.

The children with MID and accompanying exter-nalizing behaviour problems obviously differed with regard to total problem behaviour and externalizing problem behaviour from the children with MID and no externalizing behaviour problems, as can be seen from Table . The groups did not differ with regard to internalizing problem behaviour, age, sex, or intelligence.

Measures

Behaviour problems

As already mentioned, teachers completed the Dutch version of the TRF (Achenbach ; for the Dutch

version, see Verhulst et al. ). The Dutch version has been shown to have good reliability and validity for both the general population (Verhulst et al. ) and for children with MID (Dekker et al. ). Using the Dutch norms, T-scores were obtained for the syndrome scale Externalizing problem behaviour and for the narrow band scales measuring Aggressive behaviour and Withdrawn behaviour.

Intelligence

An estimate of the intelligence of the participants was obtained using the Vocabulary and Block Design sub-tests from the Dutch WISC-RN (van Haassen et al. ). These two subtests were selected because, together, they strongly correlate with the complete WISC-RN (r = .) (Silverstein ). In schools where the WISC-RN had been administered in the previous year, the scores from the files were used and the test was not re-administered.

Social problem solving in hypothetical situations

The SPT (Cuperus ; Matthys et al. ) was designed to assess social problem-solving skills. The original SPT consists of video vignettes. Each vignette is followed by a structured interview, which proceeds according to the steps of Dodge’s () social information processing model. For purposes of the present study, the original version of the SPT was revised for children with MID (SPT-MID; van Nieuw-enhuijzen et al. ). The number of video vignettes was reduced and the questions were somewhat sim-plified. The SPT-MID consists of five video vignettes

Table 1 Teacher’s Rating Form (TRF) scores, age, and IQ for groups of children with MID and externalizing behaviour problems vs. nosuch problems

Externalizing problems

No externalizing problems

Mean SD Mean SD F p

TRF total T-score 70.25 6.93 53.84 7.03 75.54 0.000TRF externalizing T-score 73.04 7.41 48.03 6.57 178.03 0.000TRF internalizing T-score 61.16 7.53 58.06 8.93 1.88 0.18Age 11.33 1.20 11.65 1.15 1.04 0.31IQ 73.83 14.14 67.27 11.82 3.38 0.07




in which a group of children enact problematic social situations that fall within the category of ‘being disadvantaged’.

Before each video vignette was shown, the partici-pant was asked to imagine that he or she was the protagonist in the vignette. Directly after the showing of a vignette, the video was stopped and the partici-pant was asked: ‘What would you do if this happened to you?’ The response was then scored as pro-social/assertive, antisocial/aggressive, or passive/submissive. Next, a set of videotaped pro-social/assertive, antiso-cial/aggressive, and passive/submissive responses enacted by child actors were presented in a rando-mized order. The participant was asked to select the response that he or she judged best for the circum-stances depicted in the video. Completion of the SPT-MID took approximately min.

The answers provided to the open question by of the participants were scored independently by both the second and third authors, who were trained graduate students at the time. This rendered an inter-rater reliability of . (Cohen’s kappa). The number of vignettes receiving a spontaneous aggressive response was counted to obtain a total score for the number of spontaneous aggressive responses. Similarly, the number of vignettes for which the aggressive response was selected from the set of alternatives was counted to obtain a total score for the number of aggressive response selections. Both variables had a min-imum of (= never given) and a maximum of (= always given).

Social problem solving in real-life situations

The actual behaviour of the participants was observed in two provocative situations: the Fishing Game and Dominoes. We selected situations in which the child was placed at a disadvantage and situations that were very comparable to the hypothetical situa-tions posed in the SPT-MID.

The Fishing Game situation concerned peer-group entry and involved a game of skill involving a moto-rized red disk containing fish that alternatively open and close their mouths. The aim of the game is to catch as many fish as possible with a fishing rod, of which four were provided. In the present study, the game was played by one participant and two class-mates or the so-called game partners. The game part-ners began the game minutes before the participant

and were given the following instructions: ‘In a minute or so, we are going to play a game, a fishing game. You will both be given a fishing rod, and you must try to catch as many fish as possible. You will play the game a few times. And on each occasion, another child will join you. The first child will soon be coming to play with you, but I want to make a secret plan with you. Pretend that you cannot hear him. Just keep on playing with each other and don’t allow the other child to join in. DO not tell the participant that we have planned this together. Keep playing for a while and then, when I give you a sign, let him join in.’ In the meantime, the participant was brought from his or her class into the testing room. The participant was told: ‘There are two children in there playing a game, a fishing game. It is very excit-ing. There is a spare fishing rod and a chair, so maybe you can join in.’ When the participant entered the testing room, however, the game partners continued to play the game and did not allow the participant to join in (until the agreed-upon signal was given to the game partners). In order to end the interaction in a positive manner, the three children were allowed to play the game together for some time. Finally, the participant was informed of how the situation had been staged and it was made clear that the other children had been instructed to not let the participant join in. The entire procedure lasted approximately minutes.

The behaviour of the participants during the Fish-ing Game was scored using the video recordings by the second and third authors. Both the verbal and nonverbal aspects of the behaviour of the participants were scored.

The verbal behaviour of a participant was scored as verbal active when the participant said five or more words and thus talked ‘a lot’ and verbal passive when the participant said less than five words and thus said almost nothing. The content of the verbal behaviour was scored in terms of orientation. When the partic-ipant made comments about joining into the game or addressed one or both of the game partners, this was scored as direct verbal behaviour. Conversely, indirect verbal behaviour consisted of comments not related to joining into the game. Finally, the verbal behaviour of the participants during the Fishing Game was scored for positive or negative valence. Comments judged as not being detrimental to the well-being of the game partners were scored as positive verbal. Con-




versely, comments judged as detrimental to the well-being of the game partners were scored as negative verbal.

The nonverbal behaviour of the participants during the Fishing Game was scored using the same criteria. Behaviour was scored as nonverbal active when, for example, the participant entered the room and immediately sat down at the table. Behaviour was scored as nonverbal passive when the participant entered the room and remained at a distance – standing, for example, by the door. The content of the nonverbal behaviour of the participants was si-milarly scored as having a direct or indirect orientation, and the valence of the nonverbal behaviour of the participants was also scored as having a positive or negative valence.

Based on the different possible combinations of verbal and nonverbal behaviour, eight categories of behaviour ranging from very submissive to very aggressive were distinguished, with assertive beha-viour falling in between (see Table ). Submissive

behaviour involved the child adopting a subordinate role, not standing up for him/herself, more or less withdrawing from the situation, or simply not responding to the situation. Assertive behaviour involved the child maintaining contact with the game partner in a positive manner and standing up for him/herself. Aggressive behaviour involved the child somehow placing the other child or children at a psychological, material, or other disadvantage. Each child was then assigned a Fishing game behaviour score, which could range from (= very submissive) to (= very aggressive) and thereby reflect the rele-vant category of behaviour (see Table ). The beha-viour of children was independently rated by both the second and third authors and found to show sufficient inter-rater reliability of . (Cohen’s kappa).

The Dominoes Game situation involved setting up rows of dominoes, and the aim of the task was to investigate how a child responds when prevented by the actions of another child from winning a prize. In

Table 2 Response scoring system for Fishing Game situation

Type of behaviour Description: examples of participant behaviour Category Score

VB passive/NV passive Says nothing but stays at door Very submissive 1VB passive/NV active indirect Says nothing and does something not related to

the gameVB passive/NV active direct Says nothing, sits down at table, and picks up fishing

rodSubmissive 2

VB active indirect/NV passive Makes comments not related to the game and staysat door

VB active indirect/NV active indirect Makes indirect comments not related to the game, sitsdown at table, but does not pick up fishing rod

Mildly assertive 3

VB active indirect/NV active direct Makes indirect comments not related to the game,sits down, and picks up fishing rod

VB active direct/NV active indirect Asks: ‘Can I join in?’ Moderatelyassertive

4

VB active direct/NV active direct Asks: ‘Can I join in?’ and picks up fishing rod Very assertive 5VB active indirect negative/NV active

indirect negativeComplains/swears but not directly about participation,

sits down at table, and messes around with fishingrod

Mildly aggressive 6

VB active direct negative/NV activeindirect negative

Calls game partners names and messes around withfishing rod

Moderatelyaggressive

7

VB active indirect negative/NV activedirect negative

Complains/swears but not directly about participationand pushes game partners

VB active direct negative/NV activedirect negative

Calls game partners names and shoves game partners Very aggressive 8

VB, verbal; NV, nonverbal.




this situation, two children – one participant and one game partner – were positioned next to each other at a table with a piece of paper on it. Two spiral paths were drawn on the paper – a red path and a green path, and the paths were connected at one end. Each child had a box with domino pieces. The following instructions were then given. ‘You are going to play a game, Dominoes. As you can see, we are filming it on video. This is because we want to see how you played later on. Nobody else will see the video. Here is an example of the paths (point at it). The red side is for you (participant) and the green side is for you (game partner). What you have to do is, both at the same time, place all of the pieces upright along your path until you get to the finish. You must use only your own colour, and you are not allowed to help each other. If the pieces fall over, then you have to begin again. Do you understand? What is very impor-tant is that you can win a big prize together! You should both try to build the whole path within the time, both of you at the same time. I’m going to set the stopwatch for seven minutes. If you both reach the end within that time, you can win euro. If both of you do not reach the end, then you do not win the prize. If one of you reaches the end and the other does not, then neither of you wins a prize. Okay, I’m going to set the time and then we can begin. Good luck!’ After the instructions were read out loud, the children started at the beginning where the two spi-rals intersected each other. This meant that if one child knocked his or her dominoes over, the other child’s would also fall.

The aim of the Dominoes Game was to reach the finish within minutes. If both children succeeded, then they both won a prize. But both of the children had to be successful in order to win. The reward was included to emotionally involve the children in the game and motivate them to play. The participant was placed at a disadvantage when the game partner experienced difficulties with his domino pieces. That is, a third of the game partner’s pieces were manipu-lated to be uneven on the short side, which meant that they were unstable and easily fell over. In addi-tion, the path on the game partner’s side of the paper was raised with pieces of paper in five places. This manipulation was not visible to the children but made the surface unstable and reaching the end impossible. The participant was thus placed at disadvantage because his/her pieces continually fell over as a result

of the actions of the game partner. For the game to nevertheless end positively, the children were given a consolation prize at the end.

A participant’s behaviour after the domino pieces had fallen over twice was scored using the video recordings and the same coders as for the Fishing Game. The verbal and nonverbal aspects of the behaviour of the participants in the Dominoes Game were rated similarly to the verbal and nonverbal aspects of their behaviour in the Fishing Game. On the basis of the different possible combinations of verbal and nonverbal behaviour, six categories of behaviour ranging from very submissive to very aggressive, with assertive behaviour falling in between, could again be distinguished (see Table ). Each child was then assigned a Dominoes behaviour score, which could range from (= very submissive) to (= very aggressive). There were fewer categories of behaviour for the Dominoes situation than for the Fishing Game situation because the nonverbal aspects of the behaviour of the participants played a smaller role in the Dominoes situation, in which the children were seated next to each other, than in the Fishing Game situation, in which the children’s approach behaviour constituted an important aspect of their social problem-solving behaviour. The inter-rater reliability for the Dominoes game was again calculated for of the children rated by the second and third authors and found to be . (Cohen’s kappa).

Procedure

The children were tested at their schools on two different days. On the first day, the staged real-life situations were presented: the Fishing Game in the morning and Dominoes in the afternoon. On the second day, approximately weeks after the first day, the SPT-MID was administered in the morning and the two subtests from the WISC-RN in the afternoon. The testing was consistently done in this order so that the children would be inclined to respond as sponta-neously as possible to the staged real-life situations and not be influenced by the questions constituting the SPT-MID. All of the tests were administered in a quiet room at the children’s schools. For the Fishing Game, the game partners were first collected from the class and given their instructions and thereafter the participant. For Dominoes, the participant and




game partner were collected from their classes and brought into the test room at the same time. Video recordings were made of the children’s behaviour in both situations.

Prior to the initiation of this study, we seriously considered the ethical implications of the procedures and discussed these at length. Was it ethical to pur-posefully place children in a situation of conflict? In the end, we decided that it was permissible when several important aspects were attended to: provision – when judged necessary – of a clear explanation of what was done after conclusion of the experimental situation and provision of a positive conclusion to the interactions.

In the Fishing Game situation, the game partners were involved in the manipulation beforehand. In order to prevent any harm to the relation between the participant and the game partners, it was therefore made very clear to the participant that the research assistant had assigned the game partners the task of excluding the participant and that the game partners thus had no choice with regard to their behaviour. The Fishing Game situation was also concluded with the participant actually playing with the other two children. And it was seen that – in each case – the participant played an enjoyable game and thus com-pleted the task with a positive feeling.

For the Dominoes situation, in contrast to the Fishing Game situation and the situations in other studies (e.g. Underwood et al. ; Underwood & Bjornstad ), we did not tell the children at the

end of the testing session about the manipulations that had been made for three reasons. First, both the game partners and participants were unaware of how the pieces of the game partner had been manipulated, which means no question of unequal treatment. Se-cond, a major feature of the Dominoes game in gen-eral is that the pieces fall down. This means that those children familiar with the game are clearly aware of this risk and that the frustration experienced by the participants is more or less a normal part of the game and not caused by the game partner per se. In addi-tion, both players could – in principle – cause the domino pieces to fall over and not just the game partner (although only the pieces of the game partner were manipulated). In other words, the game partner could also experience considerable frustration when the participant prevented the players from winning the prize. Finally and perhaps most importantly, we believed that telling the children that we had ma-nipulated the materials in the Dominoes situation could affect their confidence in adults. Given that children with MID have often had negative experi-ences with adults, we did not want to add to this. That is, by not informing the children about the re-search set-up, we could prevent them from having yet another negative experience with adults. In order to reduce the children’s disappointment about not win-ning the prize and allow the Dominoes interaction to end positively, the children all received a consolation prize. And the children were also given time to talk about the game.

Table 3 Response scoring system for Dominoes situation

Type of behaviour Description: examples of participant behaviour Category Score

VB passive Says less than five words Very submissive 1VB active indirect Makes comments not related to the game Submissive 2VB active direct Says something about winning, provides encouragement,

gives directions when pieces fall overAssertive 3

VB active indirect negative/NV active positive

Makes negative comments not related to the game(swearing)

Mildly aggressive 4

VB passive/NV activeindirect negative

Says nothing but throws pieces away

VB active indirect negative/NV active indirect negative

Makes negative comments (swearing) and throws piecesaway

Moderately aggressive 5

VB active direct negative/NV active direct negative

Calls game partner names and stamps feet or hitssomething or game partner

Very aggressive 6

VB, verbal; NV, nonverbal.




Data analysis

To answer the question of whether the responses of the children with MID and concomitant externali-zing behaviour problems differed from the responses of children with MID and no such externalizing behaviour problems, a number of analyses were undertaken. First, the hypothetical responses of the children to conflict situations were compared in a multivariate analysis of variance () with Group as the independent variable and Spontaneous aggressive responses and Aggressive response selec-tion as the dependent variables. Second, to compare aggressive behaviour in the real-life situations, inde-pendent sample t-tests with Group as the indepen-dent variable and Aggressive Dominoes behaviour and Aggressive Fishing game behaviour as the depen-dent variables were performed. Third, to assess the relationship between hypothetical responses and actual behaviour, Pearson correlations were calcu-lated between the Spontaneous aggressive response scores and Aggressive response selection scores for the children, on the one hand, and their Fishing game behaviour scores, Dominoes behaviour scores, and Aggressive behaviour in the class as reported by the teacher, on the other hand. Inspection of distribu-tions of the variables indicated that skewness was smaller than for all included variables and thus parametric statistics were appropriate.

Results

Group differences in hypothetical situations

A multivariate main effect of Group was found for the children’s hypothetical responses to conflict situ-ations, F (,) = ., p = .. A univariate effect

of Group was found for Spontaneous aggressive responses, F (,) = ., p = .. As can be seen from Table , children with MID and concomitant externalizing behaviour problems scored higher on Spontaneous aggressive responses than children with MID and no such externalizing behaviour problems. No univariate effect was found for Aggressive response selection.

Group differences in real-life situations

The mean rating for the behaviour of the participants in the Dominoes situation along a scale of – was . (SD = .). This indicates mainly assertive behaviour, with a normal range of behaviour from submissive to aggressive. In the Fishing Game situation, none of the children displayed any aggres-sive behaviour whatsoever. It was therefore decided to use only the Dominoes situation in subsequent analyses.

For the Dominoes situation, clear differences in the amount of aggressive behaviour displayed by the two groups were found, t (, ) = ., p = .. The mean rating for the children with MID and concom-itant externalizing behaviour problems was mildly aggressive (M = ., SD = .) while the mean rat-ing for the children with MID and no such external-izing behaviour problems was mainly submissive (M = ., SD = .).

Relationship between responses to hypothetical and real-life situations of conflict

As can be seen from Table , the children’s Sponta-neous aggressive responses to hypothetical conflict situations correlated moderately with observed aggressive behaviour during the Dominoes situation

Table 4 Means (M) for hypothetical responses to conflict situations according to group

Externalizingproblems

No externalizingproblems

Mean SD Mean SD F p

Spontaneous aggressive response 1.38 1.06 0.41 0.50 20.84 0.000Aggressive response selection 1.46 1.18 0.97 1.23 2.25 0.14




and strongly with teacher ratings of aggressive beha-viour in class. Aggressive response selection did not correlate significantly with any of these measures. Actual aggressive behaviour also correlated strongly with teacher ratings of aggressive behaviour.

Discussion

In this study, the relationship between the aggressive-ness of the hypothetical responses and actual beha-viour of children with MID when confronted with social conflict situations was examined. Our first question was whether both clearly hypothetical and staged real-life situations could differentiate between children with MID and accompanying externalizing behaviour problems vs. children with MID and no such accompanying externalizing behaviour prob-lems. The second question was whether a relationship exists between the hypothetical responses and actual behaviour of the MID children studied here when confronted with comparable situations.

We have found both methods to indeed discrimi-nate between children with MID with and without externalizing behaviour problems. With respect to the responses of the children to the hypothetical conflict vignettes, those with MID and externalizing beha-viour problems generated more spontaneously aggressive responses than those with MID and no such behaviour problems while the groups did not differ with regard to aggressive response selection. It appears that children with MID and externalizing behaviour problems recognize nonaggressive responses to social problems when presented as one of a number of response options but do not sponta-neously provide such responses when confronted with such problems. In the real-life Dominoes game

situation, the children with MID and externalizing behaviour problems displayed more aggressive beha-viour than the children with MID and no such pro-blems. In contrast to research among children with normal intellectual abilities, which has found only one of these two methods to differentiate between chil-dren with behaviour problems and those without (Mize & Ladd ; Mize & Cox ; Vitaro & Pelletier ), both the hypothetical and real-life situations differentiated between the two groups in the present study. These contrasting findings may be attributed to both characteristics of the sample and the methods used. In the present study children were older than in the other studies. Older children may have better verbal skills than younger children, there-fore showing fewer problems with responding to hypothetical situations. The finding of no effects of hypothetical methods may be attributed by the young age of the sample. Second, the method of staged real-life situations used in the present study appears to be valid to measure real-life behaviour; the situations were engaging enough to evoke aggression. In con-trast, other observational methods may not have been that engaging, as is indicated by the results of a study by Underwood & Bjornstad (), in which children did not report aggressive emotions during a provoca-tion situation.

With regard to the second question, as expected, a correlation was found between the spontaneous aggressive responses of the children to hypothetical conflict situations and the ratings of aggressive beha-viour during a staged real-life conflict situation although the correlation was not a very strong one. Nevertheless, the more often an aggressive hypothe-tical response occurred, the higher the level of aggres-sion observed in the real-life situation. If a participant said that he or she would respond aggressively in a

Table 5 Correlations between hypothetical aggressive responses, observed behaviour and behaviour in class (TRF)

Observed aggressive behaviour TRF aggressive behaviour

Hypothetical aggressive responsesSpontaneous 0.30* 0.49**Selected 0.16 0.24

Observed aggressive behaviour 0.63**

TRF, Teacher’s Report Form.*p < .; **p < ..




hypothetical conflict situation, then he or she was indeed more likely to do so when actually confronted with such a situation. Moreover, the spontaneous aggressive responses of the children to the hypothe-tical conflict situations strongly correlated with teacher ratings of aggressive behaviour in class.

Unexpectedly, no significant correlations were found to occur between aggressive response selection and actual behaviour. In addition, response selection did not correlate with teacher ratings of aggressive behaviour in class. Children’s spontaneous responses thus reflect their actual behaviour more closely than their selected responses. This can be explained in terms of a discrepancy between spontaneously pro-viding a response and recognizing a response that one would not necessarily think of oneself. For spontane-ous behaviour, the individual can and may simply react. For response selection, the individual must first evaluate the alternatives, which include an adequate (assertive) response, and then select the alternative that he or she judges most suitable.

The responses of children with MID to hypothe-tical vignettes are frequently taken to be valid indica-tors of their actual behaviour. Now that a significant correlation has been demonstrated, we are somewhat justified in drawing inferences about the actual behaviour of children with MID on the basis of their responses to the hypothetical situations presented as part of the SPT-MID. This should nevertheless be done with utmost care as we did not find – or expect – a perfect correlation. As known from social psycho-logical research in general, only a moderate correla-tion exists between the intention to perform a particular behaviour and actual performance of that behaviour (Ajzen & Fishbein ). Dodge () has also found moderate correlations between response generation and actual behaviour in children without disabilities. In addition, the verbal report of mental processes underlying behavioural decisions is known to be difficult (Nisbett & DeCamp Wilson ).

The explanation of the moderate correlation found to occur between the hypothetical responses and actual behaviour of the children studied here perhaps lies in some basic differences between the measures used. Although we attempted to devise hypothetical and real-life situations that equally reflected ‘being disadvantaged’, it was impossible to prevent any dif-ferences in the content of the tasks whatsoever. ‘Being

disadvantaged’ covers a broad range of situations, including exclusion from an activity and destruction of one’s property. In addition, research has shown certain types of situations to be much more proble-matic for children than other types (Matthys et al. ).

Another difference between the two measures was the degree of involvement in the situation. The amount of involvement and emotion present in real-life situations is obviously greater than in hypothetical situations. The participation of children in hypothe-tical situations also tends to be rather unnatural (Oro-bio de Castro et al. ). When confronted with a hypothetical situation, the child must not only try to imagine him/herself within the situation but also as the disadvantaged child in the case of the present research. The child in a – albeit staged – real-life situation is actually part of the situation and therefore has real-life concerns at stake. Along these lines, research on children’s social information processing has shown the differences to be more marked in si-tuations where the children feel tense or threatened (Dodge & Somberg ; Orobio de Castro et al. b). It is likely that different amounts of involve-ment and emotion were elicited in our research as the children only received a reward (i.e. prize) in the Dominoes situation and not in the hypothetical situ-ations or in the real-life Fishing Game situation. The relatively lower degree of involvement and/or emotion in the Fishing Game situation when compared to the Dominoes situation with its reward may also explain the absence of aggressive behaviour in particularly the Fishing Game situation.

A third difference between the two methods used to elicit responses to social conflict in the present research was the amount of time available for reflec-tion. In real-life situations, there is often little or no time to process the available information and reflect on the possible response options (Orobio de Castro et al. a). Hypothetical situations thus allow more time for reflection than real-life situations and the mental steps involved in the process of behavioural decision-making to occur one at a time, which makes the response to a hypothetical problem situation less spontaneous and the connection to behaviour in real life possibly less close.

The responses of the children to the hypothetical conflict situations may also have been relatively more open to socially desirable response tendencies than




the actual behavioural responses of the children under more or less real-life circumstances. Neverthe-less, social desirability may also have played a role in the behaviour of the children in the real-life situations because the circumstances were not completely na-tural. Researchers were present during the observa-tions and a video camera was also clearly visible and operating. Conversely, the game partners were class-mates and thus familiar to the participants, which presumably made the situation more realistic but less threatening than if the game was played with com-plete strangers, as lack of familiarity has been shown to inhibit aggressive behaviour (Matthys et al. a,b).

Yet another consideration is that answering ques-tions about hypothetical problematic situations defi-nitely requires adequate verbal skills while the verbal skills of children in real-life conflict situations may be less important. It is known that children with MID experience difficulties with verbal tasks (Hunt & Marshall ).

Finally, the definition of the concept of aggression in the real-life situations may simply have differed from its definition in the hypothetical situations. Although we tried to operationalize the concept of aggression as consistently as possible across the dif-ferent situations, the scoring systems nevertheless dif-fered because nonverbal responses could only be scored for the real-life situations. In the hypothetical situation, moreover, the answers of the children were simply scored as submissive, assertive, or aggressive while both the verbal behaviour and nonverbal beha-viour of the children in the real-life situations were scored separately and then combined into six (in the case of the Dominoes game) categories of behaviour ranging from very submissive to very aggressive.

In sum, both hypothetical and real-life situations were found to differentiate between two groups of children with MID differing with regard to externa-lizing behaviour problems. In addition, both the spon-taneous responses of the children to hypothetical conflict situations and their actual behaviour in staged real-life situations of conflict were found to correlate with teacher ratings of aggressive behaviour in class. The two methods can be seen as yielding information on different aspects of social problem solving. The hypothetical situation provides informa-tion on a child’s knowledge of social problem-solving skills while the real-life situation provides information

on a child’s actual problem-solving skills. In the hypo-thetical situation, a distinction can be made between spontaneous responses and recognition of an ade-quate response. Given the moderate associations found between particularly the children’s spontane-ous responses in the hypothetical situations and their behaviour in real-life situations, however, the hypo-thetical method can also be said to provide informa-tion on both children’s knowledge of behaviour and the likelihood of such. In this sense, the hypothetical method appears to be more informative than the real-life method. In addition, the hypothetical method is less taxing for participants and easier to administer than the real-life method. It should nevertheless be remembered that the assessment of spontaneous responses using the hypothetical method is still only a proxy for actual behaviour. The choice of measures should depend on careful evaluation of the advan-tages and disadvantages of the two approaches in light of the specific research question.

We believe the present findings have important practical implications for understanding and treating children with MID and behaviour problems. An ade-quate approach to an individual child’s behaviour problems requires a thorough understanding of the specific social perceptions, evaluations, and problem solving that preceded the behaviour. Assessment of social information processing may serve this end and may be most efficiently conducted using hypothetical vignettes. However, it has long remained unclear whether such assessment provided valid information for children with MID. The present study indicates that for these children responses to hypothetical vignettes are valid indicators of social information processing, as they are related with actual aggressive behaviour. This finding paves the way for valid assess-ment of social information processing and for the application of knowledge concerning social informa-tion processing in response to hypothetical vignettes reviewed in the introduction to treatment of aggres-sive behaviour problems.

Acknowledgements

The authors would like to thank the participants and their teachers. We are grateful to Harm Jongerius for his creative contribution at an early stage of this study.




References

Achenbach T. M. () Manual for the Teacher’s Report Form and Profile. University of Vermont, Department of Psychiatry, Burlington.

Ajzen I. & Fishbein M. () Attitude–behavior relations: a theoretical analysis and review of empirical research. Psychological Bulletin , –.

CBS, Centraal Bureau voor de Statistiek (Central Bureau for Statistics) () http://www.cbs.nl

Coie J. D., Cillessen A. H. N., Dodge K. A., Hubbard J. A., Schwartz D., Lemerise E. A. & Bateman H. () It takes two to fight: a test of relational factors and a method for assessing aggressive dyads. Developmental Psy-chology , –.

Coie J. D. & Dodge K. A. () Aggression and antisocial behavior. In: Handbook of Child Psychology, Vol. , th Edn. (Series ed. W. Damon; Volume ed. N. Eisenberg), pp. –. Wiley, Toronto.

Coie J. D. & Kuperschmidt J. B. () A behavioral anal-ysis of emerging social status in boys’ groups. Child Devel-opment , –.

Crick N. R. & Dodge K. A. () A review and reformu-lation of social information processing mechanisms in children’s social adjustment. Psychological Bulletin , –.

Cuperus J. M. () Sociale probleemoplossing bij kinderen met gedragsstoornissen [Social problem solving in children with problem behavior]. Dissertation. Universiteit Utrecht, Utrecht, the Netherlands.

Dekker M. C., Koot H. M., van der Ende J. & Verhulst F. C. () Emotional and behavioral problems in chil-dren and adolescents with and without intellectual dis-ability. Journal of Child Psychology and Psychiatry , –.

Dodge K. A. () Behavioral antecedents of peer social status. Child Development , –.

Dodge K. A. () A social information processing model of social competence in children. In: Minnesota Sympo-sium on Child Psychology. Cognitive Perspectives on Chil-dren’s Social and Behavioral Development, Vol. (ed. M. Perlmutter), pp. –. Erlbaum, Hillsdale, NJ.

Dodge K. A., Lochman J. E., Harnish J. D., Bates J. E. & Pettit G. S. () Reactive and proactive aggression in school children and psychiatrically impaired chronically assaultive youth. Journal of Abnormal Psychology , –.

Dodge K. A., McClaskey C. L. & Feldman E. () Situ-ational approach to the assessment of social competence in children. Journal of Consulting and Clinical Psychology , –.

Dodge K. A., Pettit G. S., McClaskey C. L. & Brown M. M. () Social competence in children. Monographs of the Society for Research in Child Development .

Dodge K. A., Price J. M., Coie J. D. & Christopoulos C. () On the development of aggressive dyadic relation-ships in boys’ peer groups. Human Development , –.

Dodge K. A. & Somberg D. R. () Hostile attributional biases among aggressive boys are exacerbated under con-ditions of threats to the self. Child Development , –.

Dodge K. A. & Tomlin A. M. () Utilization of self-schemas as a mechanism of interpretational bias in aggressive children. Social Cognition , –.

van Haassen P. P., de Bruyn E. E. J., Pijl Y. J., Poortinga Y. H., Lutje Spelberg H. C., Van der Steene G., Coetsier P., Spoelders-Claes R. & Stinissen J. () WISC-R, Nederlandstalige Uitgave [WISC-Royal, Dutch Version]. Swets & Zeitlinger, Lisse.

Healy K. N. & Masterpasqua F. () Interpersonal cog-nitive problemsolving among children with mental retar-dation. American Journal on Mental Retardation , –.

Hunt N. & Marshall K. () Exceptional Children and Youth: An Introduction to Special Education. Houghton Mifflin Company, Boston/Toronto.

Leffert J. S. & Siperstein G. N. () Assessment of social-cognitive processes in children with mental retardation. American Journal on Mental Retardation , –.

Linna S. L., Moilanen I., Ebeling H., Piha J., Kumpulainen K., Tamminen T. & Almqvist F. () Psychiatric symp-toms in children with intellectual disability. European Child and Adolescent Psychiatry , –.

Lochman J. E. () Self- and peer perceptions and attri-butional biases of aggressive and nonaggressive boys in dyadic interactions. Journal of Consulting and Clinical Psy-chology , –.

Lochman J. E. & Dodge K. A. () Social-cognitive pro-cesses on severely violent, moderately aggressive, and non-aggressive boys. Journal of Consulting and Clinical Psychology , –.

Lochman J. E. & Dodge K. A. () Distorted perceptions in dyadic interactions of aggressive and nonaggressive boys: effects of prior expectations, context, and boys’ age. Development and Psychopathology , –.

Matthys W., Cuperus J. M. & van Engeland H. () Deficient social problem-solving in boys with ODD/CD, with ADHD, and with both disorders. Journal of the American Academy of Child and Adolescent Psychiatry , –.

Matthys W., De Vries H., Hectors A., Veerbeek M., Heide-mann W., Goud M., Van Hooff J. & Van Engeland H. (b) Differences between conduct disordered and normal control children in their tendencies to escalate and neutralize conflicts when interacting with normal peers. Child Psychiatry and Human Development , –.

http://www.cbs.nl




Matthys W., Maassen G. H., Cuperus J. M. & van Engeland H. () The assessment of the situational specificity of children’s problem behavior in peer-peer context. Journal of Child Psychology and Psychiatry , –.

Matthys W., Van Loo P., Pachen V., De Vries H., Van Hooff J. & Van Engeland H. (a) Behavior of conduct dis-ordered children in interaction with each other and with normal peers. Child Psychiatry and Human Development , –.

Mize J. & Cox R. A. () Social knowledge and social competence: number and quality of strategies as predic-tors of peer behavior. Journal of Genetic Psychology , –.

Mize J. & Ladd G. W. () Predicting preschoolers’ peer behavior and status from their interpersonal strategies: a comparison of hypothetical-reflective and enactive assess-ments. Developmental Psychology , –.

van Nieuwenhuijzen M., Bijman E. R., Lamberix I. C. W., Wijnroks L. & Matthys W. () Handleiding voor de SPT-MLK [Manual for the SPT-MID]. Utrecht University, Department of Special Education, Utrecht, the Netherlands.

Nisbett R. E. & DeCamp Wilson T. () Telling more than we can know: verbal reports on mental processes. Psychological Review , –.

Orobio de Castro B., Bosch J. D., Veerman J. W. & Koops W. (a) The effects of emotion regulation, attribution prompts, and delay, on response aggressiveness in antiso-cial boys. Cognitive Therapy and Research , –.

Orobio de Castro B., Slot N. W., Bosch J. D., Koops W. & Veerman J. W. (b) Negative affect exacerbates hostile attributions of intent in highly aggressive boys. Journal of Clinical Child and Adolescent Psychology , –.

Orobio de Castro B., Veerman J. W., Koops W., Bosch J. D. & Monshouwer H. J. () Hostile attribution of intent and aggressive behavior: a meta analysis. Child Development , –.

Quiggle N. L., Garber J., Panak W. F. & Dodge K. A. () Social information processing in aggressive and depressed children. Child Development , –.

Schwartz D., Dodge K. A. & Coie J. D. () The emer-gence of chronic peer victimization in boys’ play groups. Child Development , –.

Schwartz D., Dodge K. A., Coie J. D., Hubbard J. A., Cillessen A. H. N., Lemerise E. A. & Bateman H. ()

Social-cognitive and behavioral correlates of aggression and victimization in boys’ play groups. Journal of Abnor-mal Child Psychology , –.

Shantz C. U. & Shantz D. W. () Conflict between children: social-cognitive and sociometric correlates. In: Peer Conflict and Psychological Growth (ed. M. W. Berkow-itz), pp. –. Jossey-Bass, San Francisco, CA.

Silverstein A. B. () Reappraisal of the validity of WAIS, WISC, and WPPSI short forms. Journal of Consulting and Clinical Psychology , –.

Underwood M. K. & Bjornstad G. J. () Children’s emotional experience of peer provocation: the relation between observed behavior and self-reports of emotions, expressions, and social goals. International Journal of Behavioral Development , –.

Underwood M. K. & Galen B. R. () Laboratory meth-ods for investigating anger and aggression in peer inter-actions. In: Creating Naturalistic Laboratory Methods to Observe Peer Relations and Emotional Expression (Chair M. K. Underwood). Symposium Presented at the Bien-nial Meeting of the Society for Research in Child Devel-opment, Washington, DC.

Underwood M. K., Hurley J. C., Johanson C. A. & Mosley J. E. () An experimental, observational investigation of children’s responses to peer provocation: developmen-tal and gender differences in middle childhood. Child Development , –.

Verhulst F. C., van der Ende J. & Koot M. H. () Handleiding voor de Teacher’s Report Form (TRF) [Manual for the TRF]. Erasmus University Rotterdam/Sophia Chil-dren’s Hospital/Academic Hospital Rotterdam, Rotter-dam, the Netherlands.

Vitaro F. & Pelletier D. () Assessment of children’s social problem-solving skills in hypothetical and actual conflict situations. Journal of Abnormal Child Psychology , –.

Vitaro F., Pelletier D. & Coutu S. () Impact of a negative social experience on the social reasoning process of aggressive-rejected children. Perceptual and Motor Skills , –.

Zahn-Waxler C., Cole P. M., Richardson D. T., Friedman R. J., Michel M. K. & Belouad F. () Social problem solving in disruptive preschool children: reactions to hypothetical situations of conflict and distress. Merrill-Palmer Quarterly , –.

Accepted August

Documents

Do children do what they say? Responses to hypothetical and real-life social problems in children with mild intellectual disabilities and behaviour problems