Br. J . educ. Psychol., 61,355-372
CHILDREN WITH MILD LEARNING DIFFICULTIES IN AN INTEGRATED AND IN A SPECIAL SCHOOL: COMPARISONS OF BEHAVIOUR, TEASING AND TEACHERS’ ATTITUDES BY MARGARET MARTLEW AND JEAN HODSON (Department of Psychology, University of Shefield) SWARY. Comparisons were made between children and teachers in two schools, one a mainstream school with an integratedresourceunit for children with mild learning difficulties,the other a special school. Seventeen children with mild learning difficulties from the special school and ten children with mild learning difficulties from the mainstream school, matched with ten mainstream children for age and sex, took part in the study. Observations were made on children in the playground and their social proximity and positive and negative behaviours were categorised. Self-reports on teasing, bullying and friendships were also collected from children in the mainstream school. Teachers from both schools completed a questionnaire on their attitudes to integration. There was no difference in amount of social contact children with mild learning difficultieshad in the mainstream and special school nor were there significantdifferences in total positive and negative behaviours. Mainstream children, however, played significantly less frequently with children with mild learning difficulties and this was more marked in the older than the younger children.Self-reportsfromchildren in the integratedschool indicated that the children with mild learning difficulties were teased/bullied more and made fewer friends than the mainstream children. Teachers in the mainstream were more positive and teachers in the special school were more negative towards integration. The special school teachers’ expectations fitted more with the findings of the study than did those of the mainstream teachers.
INTRODUCTION THEintegration of children with special educational needs into mainstream schools has opened many forms of debate. These debates, ethical, social and educational, concern the rights of children, their opportunitiesto interact with and learn from peers and the best ways of teaching them. To date, most research interest comes from the USA where the series of federal laws passed between 1966and1974culminatedin the Educationfor All Handicapped Children Act in 1975. Britain has responded with less interest both in terms of integrating children (Swann, 1985, 1988) and in how integralion is best effected (Merry, 1986). Developmental and educational arguments have been put forward supporting the beneficialaspects of integration. For example, less advanced children through observational learning can acquire social skills and age-appropriate behaviours which can be modelled on the behaviour of normally developing peers. Mainstream schools can be seen to have educationaladvantagescompared with special schools because they have a broader curriculum and aim for higher achievements. Alternatively there is support for the view that some children with disabilities need protection from the pressures of mainstream education. Socially, special needs children may suffer as they may be rejected by mainstream peers. There are levels of opinion between these views and local policy plans for services in many respects represent a compromise. The aims of integration therefore are to enhance the development of special needs children and to increase their acceptance among ordinary peer groups through the reduction of negative stereotypes.Wamock (1978)defined three levels of integration; locational,social and functional, the three being interlinked when functional integration is achieved as there 355
356
Social Interactions of MLD Children
is both locational and social association as well as “joint participation in educational programmes”. Social and functionalintegrationare important as peer interaction can play an important role in development.Hartup (1978), for example,proposed that children are likely to show linguistic,cognitive and socialdelays in developmentif they do not have successful interactionswith peers. In terms of observational learning in social situations,Vaughan and Waters(l981) showedthatmorecompetentchildrenareobservedandimitatedmorethanless competent children. Goldman (1981) found that younger children in mixed age classes played in a more mature manner than children in same age groups. The presence of the older children had effects on the behaviour of the others in the group. Extrapolations have been made from such studies that special needs children would benefit from exposure to normal play and learning experiences and so acquire more mature and socially appropriateforms of behaviour. Models of more advanced learning would give children with special needs the opportunity to observe, imitate and adopt normal patterns of interaction (Guralnick, 1981a; 1986a). There is evidence showing that positive effects can result from integrating children (Beckman and Kohl, 1984; Cooke et al., 1977; Guralnick and Groom, 1986). For instance, Guralnick (1981b) did find that the presence of mainstream peers can reduce the level of inappropriate play in mildly impaired children, play could become more appropriate (Guralnick, 1976)and thechildrencould model socially appropriatebehaviour (Cookeet al., 1977). Benefits have also been demonstrated for non-disabled children, Sass0 and Rude (1986) for example showing that they gained in status after taking part in a peer initiation intervention programme with severely disabled children. Some form of intervention programme however is advocated to make integration effective: mere physical placement is insufficient (Stobart, 1986). Groups form on the basis of many dimensions, including age and sex, physical characteristics, socio-economic status and popularity (Rubenstein, 1984). A number of behaviour patterns commonly found among children,particularly those with special needs, can make them vulnerable to not having positive socialrelationships. Strain and Kerr (1981) suggested that many special needs children do not generally employ reinforcing behaviours (e.g., verbal compliments)and as a result are rarely sought out for social interactions by more advanced chitdren. Also they often misread subtle social cues or are slow to respond and therefore seem to behave in ways that are viewed as bizarre or unpredictable. Irrespectiveof whether a child was disabled or not, White (1980) found certain behaviours, such as producing non-communicative sounds, wandering aimlessly, conversing more than an average amount with teachers and being negative to other children encouraged negative behaviours from others. He still maintained however that the special child’s self esteem was enhanced by being in an integrated setting. Negative effects following integration have been found in several studies. Preschool and junior school children with a range of mild to moderate impairments have been found to interact less frequently with them than their mainstream peers (Cavallaro and Porter, 1980; Guralnick, 1986b; Martlew and Cooksey, 1989) and mainstream children have been observed to stay in closer proximity to other mainstream children than to children with disabilities(Porter et al., 1978).In the case of severely disabled children,Guralnick (1986b) found the interactions to be minimal. Severalfactorscan accountfor differencesin research findings.Lindsay and Desforges (1986),for instance,pointout thedifferentialeffectsthat buildingdesign andclassroomethos can have on integration. In their studies of three preschool integrated nurseries in Sheffield, they found that the special needs children interacted more with ordinary children than with their special peers but the nursery with the structured environment encouraged more interactionsthan the one with a freeplay ethos. Also the special children initiatedmost of the interactionsand the social integrationwas less than had been hoped for. In the case of severely disabled children, Armfield (1985) indicated that moderately, severely and profoundly
MARGARET MARTLEW
AND
JEAN HODSON
357
disabled children tended to be found in special classes in mainstream schools (locational integration) and their membership of the school was limited to that of physical presence. Studiescarried out in the seventies showed that integration did not necessarily change attitudes. Gottlieb et al. (1978) found mainstream children held negative attitudes towards their special needs peers both before and after integration. Bruinicks (1978) also found that children with moderate impairments were viewed in negative and prejudiced ways whether or not the child was in the same or separate classrooms. A number of other studies have documented similar findings of the social non-acceptance of disabled children using sociometric analyses (Bryan, 1974; Gottlieb and Budoff, 1973). This effect seems to be greater for children who are moderately and severely impaired while children classified as mildly impaired could be well integrated with their mainstreampeers (Guralnick,1980).The review commissioned by the Warnock Committee (Cave and Madison, 1978) failed to provide clear evidence of educationalbenefits or of social outcomes of integrated education. Children who are not fully integrated into peer groups may become the victims of bullying and teasing. Research in mainstream schools shows that teachers tend to underestimate the extent of teaching and bullying that exists (Olweus, 1978), a point reinforced by Besag (1989) who reported that adults saw little bullying face on. Few investigationshave consideredthis issue in relation to children with special needs although O’Moore and Hillery (1989) as a result of their investigation into bullying in Dublin schools point out that this group seemed to be particularly vulnerable. The physically handicapped children Anderson (1973) studied in mainstream education were teased occasionally and teachers underestimated the amount of teasing that went on. The ability and attitudes of teachers are important factors in ensuring the success or failure of integrating the child with special needs into mainstream schools. A few years after the enactmentof The Educationfor All Handicapped ChildrenAct (1975) in America, some negativeattitudestowards mainstreamingwereapparent(Palmer, 1979) although Guerin and Szatlocky (1974) did find some teachers favourably inclined to integration practices. Many mainstream teachers indicated their lack of support for the integration of the child with special needs (MacMillanet al., 1978).The general attitudeswere that specialneeds children were best catered for by providing specialprovision, while teachers in mainstream preferred to hand over children with problems particularly where there was pressure to maintain high academic standards (Gartner and Lipsky, 1987). Johnson (1987) and Johnson and Johnson (1986) proposed that these attitudes could be changed with in-service training. This study examines the issues of social integration, bullyingheasing and teacher attitudes within the scope of one study by comparing children with special needs in a mainstream and in aspecial school. Comparisonswerealso made with mainstreampeers. The mainstream school had incorporated all three of Warnock’s recommended levels of integration so we were interested in the level of interaction in the playground. We were also interested in whether observed behaviour in the playground related to children’sself-reports of being teasedbullied. By including a questionnaire to teachers in these schools on their attitudes to integration we were also able to assess their view of the benefits and disadvantages of the system they worked in and their commitment to it. METHOD Children The children with special educational needs had mild learning difficulties (MLD), classificationbeing in accordancewith the child’s individual statementof special needs. The MLD children came from two schools,a mainstream school and a special school. Each MLD child in the mainstream school was matched individually with a mainstream peer of the same age and sex.
Social Interactions of MLD Children
358
Schools The special school (SS) enrolled 128pupils between the ages of 3-16 years who came from a range of socio-economicbackgrounds. Their special educational needs ranged from generalised and specific, mild to moderate learning difficulties. The younger children had some limited contactwith mainstream schools,for examplethey took part with other schools in cross-country runs. The older children had contact with the local college of further educationand with the localindustrialcommunity.The schoolwas situated in a suburban area of a town and the children were drawn mostly from the southern area of the city. The mainstream school (MSES)was situated in a suburbanarea of the city and enrolled 154 children from the local area, mainly from middle socio-economic backgrounds. The MLD children were bussed in and came from a range of socio-economic backgrounds. The school catered for children in the 7-11+ years age range and had an integrated resource unit for 20 children with mild or specific learning difficulties. These children were provided with individual work programmes within two small classes of younger (7-9 years) and older children (9-11 years). The MLD children joined mainstream children for playtimes, dinnertimes,assemblies and were fully integrated in the same class as MS children for such lessonsasgames,artandcraftandmusic.PlacementofMLDchildrenwas forup to two years, followed by a case conference at which a placement panel considered the school’s recommendations for either the child’s entry to a special school at 11+ years or to the child’s local mainstream comprehensive school. Observational study: social proximity and play behaviour Special school Observations were made on 18MLD children,ten boys and eight girls. Their mean age was 9 years 8 months (range 7-11+ years). All had mild learning difficulties, some had additionalproblems such as behaviouraldifficulties,one child had epilepsy,one had mild to moderate learnin difficulties. All the children in the school who fitted the criteria of having mild learningdif iculties and fell within the stipulated age range were included in the study.
B
Mainstream school Ten mainstream controls, eight boys and two girls, mean age 9 years 6 months were individually matched for age and sex with ten MLD children,eight boys and two girls, mean age 9 years 6 months (range 7-11+ years). The MLD children had mild learning difficulties, some had additional problems such as asthma, epilepsy or behavioural problems. All the children in the school with mild learning difficulties within this age range were included in the study. Observations were made in the school playgrounds during free playtime when all the children in the school were in the playground. An interval sampling procedure was adopted using a category checklist, behaviours being recorded that occurred during each 10-second period. Each child was observed for four two-minuteperiods, two sets of observationsbeing made during the same play period with an interval of approximately 10minutes between the two. The second pair of observations were completed approximately two weeks later. Each child was observed for eight minutes therefore,yielding48 sets of observationsper child. The order in which the children were observed was randomised but the first pair of observations were completed before the second set was collected. Categories The categories covered social proximity and appropriate/inappropriateplay behaviours. The social proximity categorieswere solitary play; interaction with an adult; with one other child; interaction at the periphery of a group and interaction at the centre of a group. Appropriate play/positive social interaction was categorised as organised games; rough and tumble play; chasing; play with schoolyardequipment and other play assessed as
MARGARET MARTLEWAND JEAN HODSON
359
positive but occurring too infrequentlyto merit separate categorisation. Inappropriateplay/ negative interaction categories were verbal and physical aggression; wandering aimlessly; standingand staring; stereotypicbehaviour and other negative behaviours such as snatching, refusing to share, etc. All observationswere carried out by the same observer. An inter-observeragreement was estimatedby the principal observer and a second observer categorisingand recording the behaviour of the same target children over three one-hour free play sessions.The number of agreements was calculated and statistically transformed to an agreement percentage. No category agreement fell below 89 per cent and the mean percentage agreement for the three categories was 98 per cent social proximity; 96 per cent appropriate play; 96 per cent inappropriateplay. Because of the requirements to identify who was playing with whom it was not possible to make the observations“blind”;both observersknew which children were in the MLD and MS groups. There is no suggestion from the recordings that this resulted in any bias in the observers’judgments. The observer spent one week at each site before data were collected so that she could familiarise herself with the children and they could adapt to and ignore her presence. Any attemptsby children to interactwith the observer were ignored. Adaptation was indicated by the marked decline in attempts to interactand movements/glancestowards her. This practice was described by Connolly and Smith (1972) who showed that children’s behaviour was undisturbed by the observer’s presence following a period of adaptation. Child interview; teasing and friendships Nine MLD children and eight mainstream controls from the original sample in the mainstream school took part in this study. Three children fell ill and were absent for the remainder of the term so were unable to take part in the interview. The children were asked questionsrelating to their perceptions of friendship, teasing and fear of attendingschool. Ten of the questions were scored and the interviewer marked the appropriate score for the children’s responses on the form during the interview (see Appendix 1).
Teasing was defined as verbal aggression,threats, name calling, etc., and bullying was defined as physical aggression, hitting, kicking, tripping, etc. Children were asked to say what they thought teasing and bullying was if they reported themselves as the recipients of this type of behaviour to ensure their responses were correctly categorised. We have no information of a behavioural kind against which to evaluate the children’s perceptions of teasing/bullying and friendship. Teacher questionnaire All the teachers in both schools completed the questionnaire; eight teachers were from the mainstream school and 10 were from the special school. The questionnaireconsisted of 22 items relating to theeffectsof integratingchildrenwith MLD into mainstream schools(see Appendix 2). Teacherswereasked to indicate their views by marking their levelof agreement with statementson a five point scale which gave a score from 1-5for each question. Eighteen items focusedon thechild in the integrated setting, lobeingphrasedasnegative to integration (items 3,6,7,10,12,15,17,19,20,21) and eightpositive(items 1,2,4,5,8,9.11,18), while the remaining four (items 13, 14, 16,22) focused on teacher/educational issues.
This questionnairewas designed so that items covering similar topics were put in both negative and positive form from which it was possible to obtain an estimate of internal consistencyand reliability.There was a correlation of r = -0.897 between teachers’ responses to the positive and negative questions, indicating a high level of consistency in their judgments.
Social Interactions of MLD Children
360
RESULTS
Observational study: social proximity The effects of integration on social interactions were analysed using the MannWhimey U-test to examine differences between the MS and MLD children in the MSES school and between the two groups of MLD children in both schools.Therewas nodifference in the number of occasions the MS and MLD children in the mainstream school were observed to be alone, with an adult or with a mainstream child (Table 1). There were significantdifferences between the MS and MLD children in time spent with an MLD child (U = 14.5, Pc0.03); in a group of MS children (U = 7, PcO.001); a group of MLD children (U = 13, PcO.01). Differences were also found when the mainstream group was divided to examinewhether the MS and MLD children were in the centre of the group (U= 14,Pe0.03) or at the periphery (U = 21, Pc0.05). TABLE 1 MWNSAND STANDARD DEVIATIONS FORMSAND MLD CHILDREN ON MWSURES OF S O ~ APROXIMITY L
Mainstream MS (N=lO)
Special School MLD (N=10)
MLD (N17)
Mean
SD
Mean
SD
Mean
SD
Alone
4.90
5.31
6.60
7.76
9.65
9.03
Adult
1.90
2.64
3.90
4.31
7.94
7.11
One child (MS)
6.60
5.19 9.59
6.60
One child (MLD)
0.40
0.97
Group (MS) Group (MLD)
32.30 0.40
9.38 0.70
11.80 12.10
11.60 8.931
20.53
11.45
Group centre (MS)
14.60
5.70
4.50
6.62
Group periphery (MS)
17.70
10.86
7.30
9.06
Comparisonsbetween the MLD children in the mainstream and special school showed nodifferencesin timeschildrenwereobserved to bealone, withanadult,anotherchildorwith a group. The analyses suggest that mainstream children interacted with each other both in pairs and in groups rather than with children with MLD. When MLD children were with a group they tended to remain on the periphery. There were no differences in the patterns of interaction between MLD children in the mainstream and in the special school. The data for children in the integrated school were regrouped to investigate developmental differencesin peer interaction in younger and older MS and MLD children (Table 2). There was no significant difference between older MS and older MLD children in the time spent with one other MS peer but there were significant differences between the younger
MARGARET MARTLEW AND JEAN HODSON
361
children. Younger mainstreamcontrols spent significantly more time with one MS peer than did the younger children with MLD (U = 1, Pc0.01).Older and younger MS children spent significantlymore time with MS groups than the older and younger MLD children (younger: U = 2, Pe0.02,older: U = 3, Pc0.03).The developmental trend for the mainstream children was to move into group play rather than stay with just one peer. The older MLD children showed a modest increase in both singlepeer and group play in comparison with the younger group. TABLE 2 OF TIMES A N MS OR MLD CHILD WAS RATEDTO BE WITH ONE@HER CHILD OR IN A GROUP BY AGEOF NUMBER TARGET CHILD
Younger
MS (N=5)
Older
MLD (N=5)
MS (N=5)
MLD (N=5)
Mean
SD
Mean
SD
Mean
SD
Mean
One child (MS)
10.0
2.55
2.6
3.29
3.2
5.02
6.2
8.56
Group (MS)
28.2
2.17
10.4
13.2
11.70
12.7
12.3
36.4
SD
TABLE 3
NUMBER OF TIMES MS AND MLD CHILDREN WERE RATEDAS BEINGENGAGED IN APPROPRIATE OR INAPPROPRIATE INTERACTIONS
Mainstream MS (N=10) MLD (N=10)
Special School MLD (N17)
Mean
SD
19.41
14.5
12.03
3.15
5.77
9.14
34.1
10.44
29.77
10.14
Mean
SD
Mean
SD
Organised games
17.4
Total appropriate
40.7
Aggression
1.6
2.0 1
3.6
5.66
1.41
1.87
Negative
8.4
8.57
16.8
11.59
16.78
9.8
10.0
8.46
20.4
15.31
17.65
8.75
Appropriate
Inappropriate
Total inappropriate
Social Interactions of M L B Children
362
Play and behaviour The play and socialbehaviour categorieswere combinedinto appropriateand inappropriate play as only a few categories had sufficiententries to perform statisticalcomparisons. Inappropriate and appropriate play showed practically no differences between the groups (Table 3). Thecombinedappropriateplay categoriesshowedno significantdifferencethough there were differences in the sub-categoriesof organised games (U= 28.5, PcO.01) and use of the equipment (U = 27, PcO.01) between the MLD children in the mainstream and those in the special school. This is not surprising as there was virtually no equipmentin the MSES school but it is interesting that there was more organisedplay as a result in the MSES school. Inappropriate behaviour was divided into aggressive behaviour and all other combined negative behaviours. There were no differences between any of the groups in observed aggression,whether physical or verbal, given or received or in combined scores for negative behaviours. Interview:friendships and teasing Independent group t-tests were used to examine differences between the responses of mainstream and MLD children in the mainstream school to the questions on teasing and friendship(Table4).The MLD children reported being teased significantly more frequently than MS children (Pc0.05,t = 2.407, df 15) and they formed significantly fewer friendships with schoolpeers bothinandoutofschool(P
CHILDREN WITH MILD LEARNING DIFFICULTIES IN AN INTEGRATED AND IN A SPECIAL SCHOOL: COMPARISONS OF BEHAVIOUR, TEASING AND TEACHERS’ ATTITUDES BY MARGARET MARTLEW AND JEAN HODSON (Department of Psychology, University of Shefield) SWARY. Comparisons were made between children and teachers in two schools, one a mainstream school with an integratedresourceunit for children with mild learning difficulties,the other a special school. Seventeen children with mild learning difficulties from the special school and ten children with mild learning difficulties from the mainstream school, matched with ten mainstream children for age and sex, took part in the study. Observations were made on children in the playground and their social proximity and positive and negative behaviours were categorised. Self-reports on teasing, bullying and friendships were also collected from children in the mainstream school. Teachers from both schools completed a questionnaire on their attitudes to integration. There was no difference in amount of social contact children with mild learning difficultieshad in the mainstream and special school nor were there significantdifferences in total positive and negative behaviours. Mainstream children, however, played significantly less frequently with children with mild learning difficulties and this was more marked in the older than the younger children.Self-reportsfromchildren in the integratedschool indicated that the children with mild learning difficulties were teased/bullied more and made fewer friends than the mainstream children. Teachers in the mainstream were more positive and teachers in the special school were more negative towards integration. The special school teachers’ expectations fitted more with the findings of the study than did those of the mainstream teachers.
INTRODUCTION THEintegration of children with special educational needs into mainstream schools has opened many forms of debate. These debates, ethical, social and educational, concern the rights of children, their opportunitiesto interact with and learn from peers and the best ways of teaching them. To date, most research interest comes from the USA where the series of federal laws passed between 1966and1974culminatedin the Educationfor All Handicapped Children Act in 1975. Britain has responded with less interest both in terms of integrating children (Swann, 1985, 1988) and in how integralion is best effected (Merry, 1986). Developmental and educational arguments have been put forward supporting the beneficialaspects of integration. For example, less advanced children through observational learning can acquire social skills and age-appropriate behaviours which can be modelled on the behaviour of normally developing peers. Mainstream schools can be seen to have educationaladvantagescompared with special schools because they have a broader curriculum and aim for higher achievements. Alternatively there is support for the view that some children with disabilities need protection from the pressures of mainstream education. Socially, special needs children may suffer as they may be rejected by mainstream peers. There are levels of opinion between these views and local policy plans for services in many respects represent a compromise. The aims of integration therefore are to enhance the development of special needs children and to increase their acceptance among ordinary peer groups through the reduction of negative stereotypes.Wamock (1978)defined three levels of integration; locational,social and functional, the three being interlinked when functional integration is achieved as there 355
356
Social Interactions of MLD Children
is both locational and social association as well as “joint participation in educational programmes”. Social and functionalintegrationare important as peer interaction can play an important role in development.Hartup (1978), for example,proposed that children are likely to show linguistic,cognitive and socialdelays in developmentif they do not have successful interactionswith peers. In terms of observational learning in social situations,Vaughan and Waters(l981) showedthatmorecompetentchildrenareobservedandimitatedmorethanless competent children. Goldman (1981) found that younger children in mixed age classes played in a more mature manner than children in same age groups. The presence of the older children had effects on the behaviour of the others in the group. Extrapolations have been made from such studies that special needs children would benefit from exposure to normal play and learning experiences and so acquire more mature and socially appropriateforms of behaviour. Models of more advanced learning would give children with special needs the opportunity to observe, imitate and adopt normal patterns of interaction (Guralnick, 1981a; 1986a). There is evidence showing that positive effects can result from integrating children (Beckman and Kohl, 1984; Cooke et al., 1977; Guralnick and Groom, 1986). For instance, Guralnick (1981b) did find that the presence of mainstream peers can reduce the level of inappropriate play in mildly impaired children, play could become more appropriate (Guralnick, 1976)and thechildrencould model socially appropriatebehaviour (Cookeet al., 1977). Benefits have also been demonstrated for non-disabled children, Sass0 and Rude (1986) for example showing that they gained in status after taking part in a peer initiation intervention programme with severely disabled children. Some form of intervention programme however is advocated to make integration effective: mere physical placement is insufficient (Stobart, 1986). Groups form on the basis of many dimensions, including age and sex, physical characteristics, socio-economic status and popularity (Rubenstein, 1984). A number of behaviour patterns commonly found among children,particularly those with special needs, can make them vulnerable to not having positive socialrelationships. Strain and Kerr (1981) suggested that many special needs children do not generally employ reinforcing behaviours (e.g., verbal compliments)and as a result are rarely sought out for social interactions by more advanced chitdren. Also they often misread subtle social cues or are slow to respond and therefore seem to behave in ways that are viewed as bizarre or unpredictable. Irrespectiveof whether a child was disabled or not, White (1980) found certain behaviours, such as producing non-communicative sounds, wandering aimlessly, conversing more than an average amount with teachers and being negative to other children encouraged negative behaviours from others. He still maintained however that the special child’s self esteem was enhanced by being in an integrated setting. Negative effects following integration have been found in several studies. Preschool and junior school children with a range of mild to moderate impairments have been found to interact less frequently with them than their mainstream peers (Cavallaro and Porter, 1980; Guralnick, 1986b; Martlew and Cooksey, 1989) and mainstream children have been observed to stay in closer proximity to other mainstream children than to children with disabilities(Porter et al., 1978).In the case of severely disabled children,Guralnick (1986b) found the interactions to be minimal. Severalfactorscan accountfor differencesin research findings.Lindsay and Desforges (1986),for instance,pointout thedifferentialeffectsthat buildingdesign andclassroomethos can have on integration. In their studies of three preschool integrated nurseries in Sheffield, they found that the special needs children interacted more with ordinary children than with their special peers but the nursery with the structured environment encouraged more interactionsthan the one with a freeplay ethos. Also the special children initiatedmost of the interactionsand the social integrationwas less than had been hoped for. In the case of severely disabled children, Armfield (1985) indicated that moderately, severely and profoundly
MARGARET MARTLEW
AND
JEAN HODSON
357
disabled children tended to be found in special classes in mainstream schools (locational integration) and their membership of the school was limited to that of physical presence. Studiescarried out in the seventies showed that integration did not necessarily change attitudes. Gottlieb et al. (1978) found mainstream children held negative attitudes towards their special needs peers both before and after integration. Bruinicks (1978) also found that children with moderate impairments were viewed in negative and prejudiced ways whether or not the child was in the same or separate classrooms. A number of other studies have documented similar findings of the social non-acceptance of disabled children using sociometric analyses (Bryan, 1974; Gottlieb and Budoff, 1973). This effect seems to be greater for children who are moderately and severely impaired while children classified as mildly impaired could be well integrated with their mainstreampeers (Guralnick,1980).The review commissioned by the Warnock Committee (Cave and Madison, 1978) failed to provide clear evidence of educationalbenefits or of social outcomes of integrated education. Children who are not fully integrated into peer groups may become the victims of bullying and teasing. Research in mainstream schools shows that teachers tend to underestimate the extent of teaching and bullying that exists (Olweus, 1978), a point reinforced by Besag (1989) who reported that adults saw little bullying face on. Few investigationshave consideredthis issue in relation to children with special needs although O’Moore and Hillery (1989) as a result of their investigation into bullying in Dublin schools point out that this group seemed to be particularly vulnerable. The physically handicapped children Anderson (1973) studied in mainstream education were teased occasionally and teachers underestimated the amount of teasing that went on. The ability and attitudes of teachers are important factors in ensuring the success or failure of integrating the child with special needs into mainstream schools. A few years after the enactmentof The Educationfor All Handicapped ChildrenAct (1975) in America, some negativeattitudestowards mainstreamingwereapparent(Palmer, 1979) although Guerin and Szatlocky (1974) did find some teachers favourably inclined to integration practices. Many mainstream teachers indicated their lack of support for the integration of the child with special needs (MacMillanet al., 1978).The general attitudeswere that specialneeds children were best catered for by providing specialprovision, while teachers in mainstream preferred to hand over children with problems particularly where there was pressure to maintain high academic standards (Gartner and Lipsky, 1987). Johnson (1987) and Johnson and Johnson (1986) proposed that these attitudes could be changed with in-service training. This study examines the issues of social integration, bullyingheasing and teacher attitudes within the scope of one study by comparing children with special needs in a mainstream and in aspecial school. Comparisonswerealso made with mainstreampeers. The mainstream school had incorporated all three of Warnock’s recommended levels of integration so we were interested in the level of interaction in the playground. We were also interested in whether observed behaviour in the playground related to children’sself-reports of being teasedbullied. By including a questionnaire to teachers in these schools on their attitudes to integration we were also able to assess their view of the benefits and disadvantages of the system they worked in and their commitment to it. METHOD Children The children with special educational needs had mild learning difficulties (MLD), classificationbeing in accordancewith the child’s individual statementof special needs. The MLD children came from two schools,a mainstream school and a special school. Each MLD child in the mainstream school was matched individually with a mainstream peer of the same age and sex.
Social Interactions of MLD Children
358
Schools The special school (SS) enrolled 128pupils between the ages of 3-16 years who came from a range of socio-economicbackgrounds. Their special educational needs ranged from generalised and specific, mild to moderate learning difficulties. The younger children had some limited contactwith mainstream schools,for examplethey took part with other schools in cross-country runs. The older children had contact with the local college of further educationand with the localindustrialcommunity.The schoolwas situated in a suburban area of a town and the children were drawn mostly from the southern area of the city. The mainstream school (MSES)was situated in a suburbanarea of the city and enrolled 154 children from the local area, mainly from middle socio-economic backgrounds. The MLD children were bussed in and came from a range of socio-economic backgrounds. The school catered for children in the 7-11+ years age range and had an integrated resource unit for 20 children with mild or specific learning difficulties. These children were provided with individual work programmes within two small classes of younger (7-9 years) and older children (9-11 years). The MLD children joined mainstream children for playtimes, dinnertimes,assemblies and were fully integrated in the same class as MS children for such lessonsasgames,artandcraftandmusic.PlacementofMLDchildrenwas forup to two years, followed by a case conference at which a placement panel considered the school’s recommendations for either the child’s entry to a special school at 11+ years or to the child’s local mainstream comprehensive school. Observational study: social proximity and play behaviour Special school Observations were made on 18MLD children,ten boys and eight girls. Their mean age was 9 years 8 months (range 7-11+ years). All had mild learning difficulties, some had additionalproblems such as behaviouraldifficulties,one child had epilepsy,one had mild to moderate learnin difficulties. All the children in the school who fitted the criteria of having mild learningdif iculties and fell within the stipulated age range were included in the study.
B
Mainstream school Ten mainstream controls, eight boys and two girls, mean age 9 years 6 months were individually matched for age and sex with ten MLD children,eight boys and two girls, mean age 9 years 6 months (range 7-11+ years). The MLD children had mild learning difficulties, some had additional problems such as asthma, epilepsy or behavioural problems. All the children in the school with mild learning difficulties within this age range were included in the study. Observations were made in the school playgrounds during free playtime when all the children in the school were in the playground. An interval sampling procedure was adopted using a category checklist, behaviours being recorded that occurred during each 10-second period. Each child was observed for four two-minuteperiods, two sets of observationsbeing made during the same play period with an interval of approximately 10minutes between the two. The second pair of observations were completed approximately two weeks later. Each child was observed for eight minutes therefore,yielding48 sets of observationsper child. The order in which the children were observed was randomised but the first pair of observations were completed before the second set was collected. Categories The categories covered social proximity and appropriate/inappropriateplay behaviours. The social proximity categorieswere solitary play; interaction with an adult; with one other child; interaction at the periphery of a group and interaction at the centre of a group. Appropriate play/positive social interaction was categorised as organised games; rough and tumble play; chasing; play with schoolyardequipment and other play assessed as
MARGARET MARTLEWAND JEAN HODSON
359
positive but occurring too infrequentlyto merit separate categorisation. Inappropriateplay/ negative interaction categories were verbal and physical aggression; wandering aimlessly; standingand staring; stereotypicbehaviour and other negative behaviours such as snatching, refusing to share, etc. All observationswere carried out by the same observer. An inter-observeragreement was estimatedby the principal observer and a second observer categorisingand recording the behaviour of the same target children over three one-hour free play sessions.The number of agreements was calculated and statistically transformed to an agreement percentage. No category agreement fell below 89 per cent and the mean percentage agreement for the three categories was 98 per cent social proximity; 96 per cent appropriate play; 96 per cent inappropriateplay. Because of the requirements to identify who was playing with whom it was not possible to make the observations“blind”;both observersknew which children were in the MLD and MS groups. There is no suggestion from the recordings that this resulted in any bias in the observers’judgments. The observer spent one week at each site before data were collected so that she could familiarise herself with the children and they could adapt to and ignore her presence. Any attemptsby children to interactwith the observer were ignored. Adaptation was indicated by the marked decline in attempts to interactand movements/glancestowards her. This practice was described by Connolly and Smith (1972) who showed that children’s behaviour was undisturbed by the observer’s presence following a period of adaptation. Child interview; teasing and friendships Nine MLD children and eight mainstream controls from the original sample in the mainstream school took part in this study. Three children fell ill and were absent for the remainder of the term so were unable to take part in the interview. The children were asked questionsrelating to their perceptions of friendship, teasing and fear of attendingschool. Ten of the questions were scored and the interviewer marked the appropriate score for the children’s responses on the form during the interview (see Appendix 1).
Teasing was defined as verbal aggression,threats, name calling, etc., and bullying was defined as physical aggression, hitting, kicking, tripping, etc. Children were asked to say what they thought teasing and bullying was if they reported themselves as the recipients of this type of behaviour to ensure their responses were correctly categorised. We have no information of a behavioural kind against which to evaluate the children’s perceptions of teasing/bullying and friendship. Teacher questionnaire All the teachers in both schools completed the questionnaire; eight teachers were from the mainstream school and 10 were from the special school. The questionnaireconsisted of 22 items relating to theeffectsof integratingchildrenwith MLD into mainstream schools(see Appendix 2). Teacherswereasked to indicate their views by marking their levelof agreement with statementson a five point scale which gave a score from 1-5for each question. Eighteen items focusedon thechild in the integrated setting, lobeingphrasedasnegative to integration (items 3,6,7,10,12,15,17,19,20,21) and eightpositive(items 1,2,4,5,8,9.11,18), while the remaining four (items 13, 14, 16,22) focused on teacher/educational issues.
This questionnairewas designed so that items covering similar topics were put in both negative and positive form from which it was possible to obtain an estimate of internal consistencyand reliability.There was a correlation of r = -0.897 between teachers’ responses to the positive and negative questions, indicating a high level of consistency in their judgments.
Social Interactions of MLD Children
360
RESULTS
Observational study: social proximity The effects of integration on social interactions were analysed using the MannWhimey U-test to examine differences between the MS and MLD children in the MSES school and between the two groups of MLD children in both schools.Therewas nodifference in the number of occasions the MS and MLD children in the mainstream school were observed to be alone, with an adult or with a mainstream child (Table 1). There were significantdifferences between the MS and MLD children in time spent with an MLD child (U = 14.5, Pc0.03); in a group of MS children (U = 7, PcO.001); a group of MLD children (U = 13, PcO.01). Differences were also found when the mainstream group was divided to examinewhether the MS and MLD children were in the centre of the group (U= 14,Pe0.03) or at the periphery (U = 21, Pc0.05). TABLE 1 MWNSAND STANDARD DEVIATIONS FORMSAND MLD CHILDREN ON MWSURES OF S O ~ APROXIMITY L
Mainstream MS (N=lO)
Special School MLD (N=10)
MLD (N17)
Mean
SD
Mean
SD
Mean
SD
Alone
4.90
5.31
6.60
7.76
9.65
9.03
Adult
1.90
2.64
3.90
4.31
7.94
7.11
One child (MS)
6.60
5.19 9.59
6.60
One child (MLD)
0.40
0.97
Group (MS) Group (MLD)
32.30 0.40
9.38 0.70
11.80 12.10
11.60 8.931
20.53
11.45
Group centre (MS)
14.60
5.70
4.50
6.62
Group periphery (MS)
17.70
10.86
7.30
9.06
Comparisonsbetween the MLD children in the mainstream and special school showed nodifferencesin timeschildrenwereobserved to bealone, withanadult,anotherchildorwith a group. The analyses suggest that mainstream children interacted with each other both in pairs and in groups rather than with children with MLD. When MLD children were with a group they tended to remain on the periphery. There were no differences in the patterns of interaction between MLD children in the mainstream and in the special school. The data for children in the integrated school were regrouped to investigate developmental differencesin peer interaction in younger and older MS and MLD children (Table 2). There was no significant difference between older MS and older MLD children in the time spent with one other MS peer but there were significant differences between the younger
MARGARET MARTLEW AND JEAN HODSON
361
children. Younger mainstreamcontrols spent significantly more time with one MS peer than did the younger children with MLD (U = 1, Pc0.01).Older and younger MS children spent significantlymore time with MS groups than the older and younger MLD children (younger: U = 2, Pe0.02,older: U = 3, Pc0.03).The developmental trend for the mainstream children was to move into group play rather than stay with just one peer. The older MLD children showed a modest increase in both singlepeer and group play in comparison with the younger group. TABLE 2 OF TIMES A N MS OR MLD CHILD WAS RATEDTO BE WITH ONE@HER CHILD OR IN A GROUP BY AGEOF NUMBER TARGET CHILD
Younger
MS (N=5)
Older
MLD (N=5)
MS (N=5)
MLD (N=5)
Mean
SD
Mean
SD
Mean
SD
Mean
One child (MS)
10.0
2.55
2.6
3.29
3.2
5.02
6.2
8.56
Group (MS)
28.2
2.17
10.4
13.2
11.70
12.7
12.3
36.4
SD
TABLE 3
NUMBER OF TIMES MS AND MLD CHILDREN WERE RATEDAS BEINGENGAGED IN APPROPRIATE OR INAPPROPRIATE INTERACTIONS
Mainstream MS (N=10) MLD (N=10)
Special School MLD (N17)
Mean
SD
19.41
14.5
12.03
3.15
5.77
9.14
34.1
10.44
29.77
10.14
Mean
SD
Mean
SD
Organised games
17.4
Total appropriate
40.7
Aggression
1.6
2.0 1
3.6
5.66
1.41
1.87
Negative
8.4
8.57
16.8
11.59
16.78
9.8
10.0
8.46
20.4
15.31
17.65
8.75
Appropriate
Inappropriate
Total inappropriate
Social Interactions of M L B Children
362
Play and behaviour The play and socialbehaviour categorieswere combinedinto appropriateand inappropriate play as only a few categories had sufficiententries to perform statisticalcomparisons. Inappropriate and appropriate play showed practically no differences between the groups (Table 3). Thecombinedappropriateplay categoriesshowedno significantdifferencethough there were differences in the sub-categoriesof organised games (U= 28.5, PcO.01) and use of the equipment (U = 27, PcO.01) between the MLD children in the mainstream and those in the special school. This is not surprising as there was virtually no equipmentin the MSES school but it is interesting that there was more organisedplay as a result in the MSES school. Inappropriate behaviour was divided into aggressive behaviour and all other combined negative behaviours. There were no differences between any of the groups in observed aggression,whether physical or verbal, given or received or in combined scores for negative behaviours. Interview:friendships and teasing Independent group t-tests were used to examine differences between the responses of mainstream and MLD children in the mainstream school to the questions on teasing and friendship(Table4).The MLD children reported being teased significantly more frequently than MS children (Pc0.05,t = 2.407, df 15) and they formed significantly fewer friendships with schoolpeers bothinandoutofschool(P
