Journal of Autism and Developmental Disorders, Vol. 31, No. 4, 2001 Siblings Relationships of Children with Autism Laura Kaminsky1 and Deborah Dewey2,3 This study investigated sibling relationships of children with autism compared to children with Down syndrome and siblings of normally developing children. Ninety siblings (30 per group) between the ages of 8 and 18 participated in this study. Results indicated that sibling relationships in families of children with autism were characterized by less intimacy, prosocial behavior, and nurturance than those of the two comparison groups. Both siblings of children with autism and siblings of children with Down syndrome reported greater admiration of their sibling and less quarreling and competition in their relationships relative to normally developing comparison children. KEY WORDS: Autism; sibling relationships; Down syndrome. INTRODUCTION Knott et al., 1995); however, the results of these studies suggest that children with developmental disabilities and their siblings show marked similarities in their interactions compared to normally developing sibling pairs. Children with disabilities such as Down syndrome have been found to spend similar amounts of time interacting with their sibling and engage in similar types of interactions as normally developing sibling pairs (Abramovitch et al., 1987; Knott et al., 1995; Stoneman, Brody, Davis, & Crapps, 1987). Observational studies, however, have reported some differences in interactions. Specifically, children with disabilities, regardless of their birth order were more likely to show interactions with their sibling that were more typical of younger children, such as imitative behaviors (Abramovitch et al., 1987). Children with disabilities were also less likely to initiate interactions with their sibling than nondisabled children (Abramovitch et al., 1987). Further, siblings of children with disabilities tended to be more prosocial and nurturing towards their affected sibling than siblings of normally developing children (Abramovich et al., 1987; Lobato, 1991). Because impairment in social abilities is one of the deficits associated with autism, it seems likely that interactions of children with autism and their siblings may differ from other children with developmental disabilities. One study that specifically investigated the inter- Research in child development has tended to emphasize parent–child relationships and relationships with peers (Stocker, 1994; Stocker & Dunn, 1990). Siblings, however, share a unique and important relationship with one another within the family unit (Knott, Lewis, & Williams, 1995). These relationships have been found to be important in the development of social skills in early childhood (Dunn, 1988; Dunn & Munn, 1986). Positive relationships among siblings can also be an important source of social support for children and are associated with lower levels of conduct disorder and loneliness in children and higher selfworth (Jenkins & Smith, 1990; Stocker, 1994). Few studies have investigated how developmental disability in children influences interactions with siblings (Abramovitch, Stanhope, Pepler, & Corter, 1987; 1 Department of Psychology, University of Calgary, Calgary, Alberta, Canada. 2 Departments of Psychology and Paediatrics, University of Calgary and Behavioural Research Unit, Alberta Children’s Hospital, Alberta, Canada. 3 Address all correspondence to Deborah Dewey, Behavioural Research Unit, Alberta Children’s Hospital, 1820 Richmond Road S.W., Calgary, Alberta, Canada, T2T 5C7; e-mail: dmdewey@ucalgary.ca 399 0162-3257/01/0800-0399$19.50/0 © 2001 Plenum Publishing Corporation 400 actions of children with autism and their siblings compared to children with Down syndrome found that children with autism spent less time with their sibling, used a smaller number and less variety of both prosocial and antagonistic initiations towards their sibling, and imitated their sibling less (Knott et al., 1995). Despite these differences, however, Knott et al. noted that children with autism did demonstrate some social skills in their play and did spontaneously initiate some of the interactions with their sibling. Further, although children with autism interacted less often than children with Down syndrome, they showed similar patterns of interactions and similar levels of initiations and responding compared to pairs of normally developing siblings. Very little research has investigated nondisabled siblings’ perspectives on their relationship with their sibling with autism. Because of the social deficits demonstrated by children with autism, one might hypothesize that these siblings would report differences in their relationships compared to siblings of children with developmental delays and siblings of normally developing children. In a study by McHale, Sloan, and Simeonsson (1986), siblings of children with autism, siblings of children with mental retardation, and siblings of normally developing children were interviewed about their relationship with their sibling. In addition, the mothers of these children were interviewed about the relationships of their children. Results indicated that both children and mothers in all three sibling groups reported positive relationships between siblings. No group differences were found on the children’s reports of their sibling relationships. Mothers of children with disabilities, however, rated their children’s sibling relationships more positively than mothers of normally developing sibling pairs. Despite the similarity between groups, the authors noted that siblings of children with autism and Down syndrome tended to show a much wider range of responses than siblings of normally developing children. Some of the siblings of children with disabilities gave very positive reports of their relationship with their sibling, whereas others gave very negative reports. In contrast, siblings of normally developing children tended to have less varied responses. McHale et al. (1986) also reported that siblings viewed their relationship with their sibling with autism or Down syndrome more positively when they accepted the child’s role as a member of the family, perceived minimal parental favoritism, and were not worried about the future of the child with the disability. Welldeveloped coping abilities, understanding of their sibling’s disability, and positive responses from parents and peers towards the child with autism or mental disability also encouraged positive sibling relationships. Kaminsky and Dewey Several demographic variables were also found to have an influence on sibling perceptions. Children who were younger than the child with the disability tended to feel more rejecting towards their sibling than older siblings did. Siblings also reported more concerns about the future and a less positive role of their sibling in the family when the child was male. In addition, larger family size was related to fewer feelings of embarrassment about the child with the disability and fewer feelings of burden. Bagenholm and Gillberg (1991) conducted a study with a sample of siblings of children with autism, mental disabilities, and normally developing children in Sweden. In contrast to the results of McHale et al. (1986), differences were found between groups in this study. Although all groups reported quite good relationships with siblings, siblings of children with autism reported significantly less positive attitudes towards their sibling and significantly less positive descriptions of their sibling’s role in the family relative to comparison siblings. Although McHale et al. (1986) and Bagenholm and Gillberg (1991) investigated siblings perceptions of their relationship with their sibling with autism, these two studies examined only some of the many aspects involved in sibling relationships. Important factors that may play a significant role in the relationship of normally developing children with their sibling with autism such as dominance, affection, companionship, intimacy, admiration, and competitiveness were not investigated. In the present study, we used the Sibling Relationships Questionnaire (Buhrmester & Furman, 1990), a questionnaire that has not been used in previous research with children with autism, to investigate these qualities of sibling relationships in families of children with autism, families of children with Down syndrome, and families of normally developing children. METHOD Participants Participants were 90 children and adolescents between the ages of 8 and 18 years of age and one of their parents. Three groups of 30 children and adolescents were included in the study: (a) individuals who had a sibling who had been diagnosed with autism, (b) those who had a sibling who had Down syndrome, and (c) individuals who had a sibling with no known disability. For the sake of clarity the sibling who participated in the study is referred to as the “participating sibling.” The children with autism, Down syndrome, and the compar- Sibling Relationships of Children with Autism ison normally developing siblings are called “reference siblings.” Participating siblings in the three sibling groups were matched for gender and approximate age of the participating sibling, and birth order in relation to the reference sibling (younger or older). Each group of participating siblings was composed of an equal number of male and female children, with the average age of participating siblings being 11 years for each group. Most participating siblings in each of the three groups (80%) were older than the reference siblings. Only families of children with autism or Down syndrome who had been diagnosed with these disorders by a physician were included in this study. Parents provided this information before the child participated in this research. Families of children with other Pervasive Developmental disorders such as Asperger Disorder and Pervasive Developmental Disorders Not Otherwise Specified were not included as participants in this study. Participating families were recruited from parent support groups, special schools for children with disabilities, and the Calgary Catholic School District. In families where there was more than one participating sibling who was willing to volunteer for the study, the child closest in age to the reference sibling was selected as a participant. In cases where there were two possible participating siblings who were equally close in age to the reference sibling, both children completed questionnaires, and the sibling who provided the best match with the other sibling groups was included as a participant. All participating siblings were living in the same household as the reference sibling. Twenty-one siblings completed the questionnaires but were excluded from the study due to inadequate matching. Measures Sibling Relationship Questionnaire–Revised The Sibling Relationship Questionnaire (SRQ; brief version, Buhrmester & Furman, 1990), consists of 15 scales each containing 2–3 items, for a total of 39 items that assess children’s perceptions of their siblings. These scales are also grouped into four SRQ factors: Closeness/Intimacy, Power, Conflict, and Rivalry. The Closeness/Intimacy factor is made up of seven individual scales: Prosocial Behavior, Companionship, Similarity, Admiration of Sibling, Admiration by Sibling, and Affection. The Power factor is calculated by adding the Nurturance of Sibling and Dominance over Sibling scales minus the Nurturance by Sibling and Dominance by Sibling scales. The Conflict Factor con- 401 sists of Quarreling, Antagonism, and Competition. Finally the Rivalry Factor consists of the individual scales of Maternal Partiality and Paternal Partiality. On the SRQ, the child was asked to rate how well a particular characteristic described their relationship with their sibling on a five-point Likert scale which ranged from hardly ever true to extremely much. Higher scores on each of the scales or factors indicated greater levels of the specified quality in the sibling relationship. For example, higher scores on the Affection scale indicate greater perceived affection between siblings. Buhrmester and Furman (1990) reported internal consistency coefficients for the SRQ that range from .71–.81 for children in the 3rd, 6th, 9th, and 12th grades. Of the 60 alpha coefficients calculated for the 15 scales in each age group of children, 55 were greater than .60 (Buhrmester & Furman, 1990). The Gilliam Autism Rating Scale The Gilliam Autism Rating Scale (GARS; Gilliam, 1995), is a behavioral checklist that helps to identify the presence and severity of autism in individuals between the ages of 3 and 22 years. The GARS was included in this study as a measure of severity of autism. It was normed on a sample of 1,092 children and young adults who had been diagnosed with autism. The checklist includes three subtests, each of which have 42 items. The subtests measure Stereotyped Behaviors, Communication, and Social Interaction. A fourth subtest, Developmental Disturbances is optional for parents to complete. The total of the subtest scores yields an Autism Quotient. Items in the first three subtests are presented on a four-point Likert scale (0–3). The fourth subtest, Developmental Disturbance uses yes/no questions. Alpha coefficients for the GARS subtests and overall quotient are as follows: Stereotyped Behaviors .90, Communication .89, Social Interaction .93, Developmental Disturbances .88, and Autism Quotient .96. Test–retest reliability for a 2-week interval for each subtest ranged from .81 to .88. Interrater reliabilities have been calculated between teachers, parents, and between a teacher and a parent. Reliabilities ranged from .55 for parent-to-parent ratings of Social Interaction to .99 for teacher-to-parent ratings of Stereotyped Behaviors and the Autism Quotient. Adaptive Behaviors Questionnaire Parents of children who had autism and parents of children who had Down syndrome were asked a series of 15 items partially adapted from the Vineland Adaptive Behavioral Scales (Sparrow, Balla, & Cicchetti, 402 1984). These items were administered to parents in order to gain information about the adaptive level of functioning of the child (Appendix A). The first items ask the parent questions about the child’s degree of disability, performance in school, and special services provided to the child. The remaining nine questions ask about a variety of adaptive behaviors. Scores on this portion ranged from 0–10, with higher scores indicating greater adaptive functioning. Demographics Questionnaire Parents were asked a series of 13 questions about family demographics (Appendix B). Specifically, this questionnaire asks about family size, presence of children with disabilities in the home, family income, marital status, maternal and paternal education, and maternal and paternal occupations. Procedure After parental and child consent was obtained, participants completed questionnaires administered by the experimenter. The Sibling Relationship Questionnaire was carefully explained to the participating sibling by the researcher. The first two items of the questionnaire were completed with the researcher’s assistance. Children below the age of 12 years were asked to read the first few questions aloud, in order to evaluate the child’s reading skills. Children who required additional help with reading items on the questionnaires were provided with assistance. Only two children needed help reading numerous items on the questionnaire. Parents were asked to complete a family demographics questionnaire. Parents of children with autism were also administered the GARS. In addition, parents of children with autism or Down syndrome were administered a questionnaire on their child’s adaptive behaviors. All participants were encouraged to ask the experimenter any questions that arose. The majority of participants (92%) took part in the study in their homes. Three participants in the group of siblings with Down syndrome and one in the normally developing group completed the questionnaires at the researcher’s office. In addition, for two families who resided outside of Calgary the parent’s demographic questionnaire was completed and mailed to the researcher, and the sibling was interviewed by telephone. The two siblings interviewed by telephone were 11 and 14 years of age and appeared to have no difficulty answering the questions over the telephone. Children were not informed before participating in the study whether or not their parents would have access to their responses on the SRQ. However, when Kaminsky and Dewey parents expressed a desire to see their children’s answers, they were shared with the parent only when the children gave consent. RESULTS Family Characteristics Group differences were examined on a variety of family demographic variables using a series of analyses of variance (ANOVA) and chi-square tests. For all of the analyses, the autism, Down syndrome, and normally developing groups were compared with each other. ANOVAs indicated no group differences on the following demographic variables: participating siblings’ age, F(2, 87) 5 0.14, p . .05, reference siblings age, F(2, 87) 5 0.07, p . .05, and number of children per household, F(2, 87) 5 1.33, p . .05. Chi-square analyses indicated that the gender of the participating siblings was not significantly different among groups, x2(2, N 5 90), p . .05. However, group differences were found in the gender of the reference siblings, x2(2, N 5 90) 5 8.90, p , .05. To control for experimentwise error rate, a Bonferonni adjustment of .05/3 5 .017 was utilized. The Mann-Whitney test was used for follow-up testing. A significant difference was found between the gender of the reference siblings with autism and the reference siblings in the normally developing group, U 5 285, p , .01. There was a substantially larger portion of boys with autism, with a 6:1 ratio of male to female. Thus, participating siblings of children with autism were more likely to be reporting about their relationships with brothers rather than sisters, compared to participating siblings of normally developing children. This finding was anticipated, given that the ratio of male to female with autism in the general population is reported to be 4–5:1 (American Psychological Association, 1994). The Down syndrome group was not found to differ from the autism or the normally developing group in terms of the gender of the reference sibling. No significant group differences were identified on a variety of family constellation variables including: the birth order of participating siblings in relation to the reference sibling, x2(4, N 5 90) 5 2.12, p . .05, type of family (two parent, single, or blended two parent), x2(4, N 5 90) 5 7.09, p . .05, and the presence of an adopted child in the family, x2(2, N 5 88) 5 2.03, p . 05. Two children with Down syndrome were adopted and were not biologically related to their sibling who participated in the study. Otherwise all sibling pairs were biological siblings. No significant difference was found in the proportion of reference siblings in each group who had ad- Sibling Relationships of Children with Autism 403 ditional illnesses or disabilities, x2(2, N 5 90) 5 4.28, p . .05. Similarly, no significant group differences were found for the presence of disability or illness in participating siblings, as well as other children in the family, x2(2, N 5 90) 5 4.32, p . .05. Sociodemographic factors were assessed using a number of variables. No significant differences were found between groups for fathers’ years of education, x2(10, N 5 90) 5 11.50, p . .05, or annual family income, x2(16, N 5 89) 5 13.34, p . .05. Years of mothers’ education significantly differed by group, x2(10, N 5 90) 5 22.50, p , .05, with mothers of participating siblings of normally developing children having greater years of education than mothers of children with autism or mothers of children with Down syndrome. Follow-up analyses using the Mann-Whitney U test, and a Bonferroni correction of .05/3 5 .017, however, failed to reach significance. No significant correlations were found between SRQ scores and mothers’ years of education, thus this variable was not included as a covariate in any of the analyses. Socioeconomic status (SES) of the groups was also compared by rating mothers’ and fathers’ occupations on the Blishen Index of Canadian Occupations (Blishen, Carroll, & Moore, 1986). This index provides a numerical rating of occupations based on generated income, prestige and education requirements for specific occupations. Occupations can be divided into one of six class values based on the numerical ratings, with higher class values being associated with greater SES. No significant group differences were found on either mothers’ SES, x2(10, N 5 89) 5 8.22, p . .05, or fathers’ SES, x2(10, N 5 89) 5 8.00, p . .05 (Table I). Severity of Disability An ANOVA was used to examine whether there were differences in overall severity of disability in the autism and Down syndrome groups, as measured by total scores on the adaptation questionnaire. No effect for group on overall level of disability was found, F(1, 58) 5 2.60, p , .05. The average adaptive ability scores were 6.4 (SD 5 2.3) for the autism group and 7.2 (SD 5 1.63) for the Down syndrome group. Group differences in ability on each of the nine basic adaptation skills were also assessed. Only the ability to communicate through language (either spoken or sign) was found to be significantly different, x2(1, N 5 60) 5 6.67, p , .05. Results indicated that children with autism were less likely to communicate through language compared to children with Down syndrome. Ten percent of children with autism did not communicate through either spoke or sign language, whereas all of the children with Down syndrome used language to communicate with others. Interestingly, however, the ability of the child with autism to communicate was not significantly correlated with any of the four SRQ factors. The GARS also provided information as to the severity of disability in the group of children with autism. Children with autism in this study had a mean overall Autism Quotient of 93.2. Thus, the mean score on autism severity as measured by the GARS was in the average range for the reference siblings with autism. The reference siblings’ Autism Quotient scores, however, ranged from very low (i.e., 67) to very high (i.e., 132), with the majority of children (72%) scoring in the average range (i.e., 80–120). Sibling Relationships Sibling Relationship Factor Scores To identify how autism influenced aspects of sibling relationships; SRQ factor scores were assessed in relation to participating sibling group and gender using a multivariate analysis of variance (MANOVA). Four SRQ factor scores were included in the analysis: Closeness/ Intimacy, Power, Conflict, and Rivalry. Results indicated Table I. Family Characteristics Group Autism Variables Sibling’s age Child’s age No. of children Father’s SES value Mother’s SES value Down syndrome Normally developing M SD M SD M SD 11.67 9.79 2.47 3.77 2.80 2.95 3.86 0.68 1.57 1.56 11.88 9.44 2.80 4.00 2.97 2.14 3.62 0.85 1.41 1.76 11.54 9.43 2.60 4.00 2.77 2.33 4.13 0.86 1.39 1.77 404 a significant overall effect for participating sibling group, Wilks’s lambda, F(8, 83) 5 4.62, p , .001, and a significant overall effect for participating sibling gender, Wilks’s lambda, F(4, 87) 5 3.49, p , .05. A Bonferroni adjustment of .05/4 5 .013, was utilized to control for experimentwise error rate in subsequent analyses. All post hoc comparisons were performed using Scheffé’s method. In all analyses of group differences all three sibling groups were compared with one another. Univariate Fs indicated a significant group difference in the Closeness/Intimacy factor, F(2, 87) 5 5.31, p , .01. Post hoc comparisons showed that participating siblings of children with Down syndrome report significantly higher levels of closeness and intimacy than the other two groups of participating siblings. A significant effect for group was also found on the Conflict factor, F(2, 87) 5 8.72, p , .001. Participating siblings of children with autism and Down syndrome reported significantly less conflict with their sibling compared to participating siblings of normally developing children. No significant group differences were found for the SRQ Power, F(2, 87) 5 3.04, p . .02, and Rivalry factors, F(2, 87) 5 1.95, p . .02. Univariate Fs indicated no significant effects of sibling gender on the following SRQ factors: Closeness/Intimacy, F(1, 88) 5 0.26, p . .02, Power, F(1, 88) 5 2.04, p . .02, and Rivalry, F(1, 88) 5 6.12, p . .02. A trend towards a significant effect for sibling gender was found on the Conflict factor, F(1, 88) 5 6.12, p 5 .02, with males reporting more conflict with their siblings than females, irrespective of group. Sibling Relationship Individual Scales To investigate group and gender differences on the 15 individuals subscales of the SRQ, a series of ANOVAs were conducted. Due to the large number of scales analyzed, only alpha values of .01 and less were considered significant results. As these analyses were exploratory in nature, group differences with an alpha level of p , .10 were noted as trends. Significant effects of group were found on the following scales: Nurturance of sibling, F(2, 87) 5 6.28, p , .01; Nurturance by sibling, F(2, 87) 5 5.37, p , .01; Affection, F(2, 87) 5 5.67, p , .01; Intimacy, F(2, 87) 5 7.79, p 5 .001; Competition, F(2, 87) 5 9.20, p , .001; Admiration of sibling, F(2, 87) 5 7.60, p 5 .001; Admiration by sibling, F(2, 87) 5 4.80, p 5 .01; Quarreling, F(2, 87) 5 8.82, p , .001. A trend for a group effect was found for the scales measuring Prosocial behavior, F(2, 87) 5 3.96, p , .05; and Companionship, F(2, 87) 5 2.45, p 5 .09. No significant Kaminsky and Dewey group effects were identified for the Maternal Partiality, Dominance of Sibling, Dominance by Sibling, Paternal Partiality, and Antagonism scales. Post hoc analyses using Scheffé’s method indicated a number of differences between groups on the SRQ scales. The Prosocial, Intimacy and Nurturance by Sibling scales differentiated participating siblings of children with autism from the other groups of siblings. Participating siblings of children with autism report less prosocial behavior compared to participating siblings of children with Down syndrome. No differences were found between the participating siblings of children with autism and the participating siblings of children normally developing children on prosocial behavior. Further, the participating siblings of children with Down syndrome were not significantly different from the participating siblings of normally developing children on prosocial behaviour. Participating siblings of children with autism also report less intimacy with their reference sibling and less nurturance by their reference sibling than both participating siblings of children with Down syndrome and participating siblings of normally developing children. No differences were found between participating siblings of Down syndrome and participating siblings of normally developing children on this measure. Post hoc comparisons on the Nurturance of Sibling, Admiration of Sibling, and Affection scales, identified differences in the relationships of siblings of children with Down syndrome compared to siblings of children with autism and siblings of normally developing children. Participating siblings of children with Down syndrome reported that they were more nurturing of their reference sibling than the other groups of siblings. They also reported that their sibling with Down syndrome admired them more than participating siblings of children with autism or normally developing children. Participating siblings of children with Down syndrome also reported more affection in their sibling relationships than participating siblings of normally developing children, but not participating siblings of children with autism. On the Competition, Admiration by Sibling, and Quarrel subscales post hoc comparisons identified differences between siblings of children with autism and Down syndrome, and siblings of normally developing children. Participating siblings of children with autism and participating siblings of children with Down syndrome reported greater admiration of their reference sibling than participating siblings of normally developing children. Participating siblings of children with autism and Down syndrome also reported that they Sibling Relationships of Children with Autism 405 were less competitive with their sibling and quarreled less than participating siblings of normally developing children. In general, siblings in all three groups reported quite positive relationships with their siblings. Table II presents the means and standard deviations of SRQ factor and scale scores for each group. Significant effects for gender were found on the scales measuring Prosocial behavior, F(1, 88) 5 7.50, p , .01; Nurturance by sibling, F(1, 88) 5 7.78 p , .01; Nurturance of sibling, F(1, 88) 5 6.30, p 5 .003; Affection, F(1, 88) 5 6.23, p 5 .01; and Competition, F(1, 88) 5 10.16, p 5 .002. Table III presents the means and standard deviations of SRQ factor and scale scores for male and female. Examination of means for each gender indicated that participating males, irrespective of group, reported less prosocial behavior and affection in their sibling relationships, less nurturance by their siblings, and greater competition with their siblings as compared to participating females. A trend towards a gender effect was found for the Maternal Partiality scale, with males regardless of group indi- cating greater maternal partiality, F(1, 88) 5 2.86, p 5 .09. No significant differences due to gender were found on the remaining scales. DISCUSSION Very few studies have investigated siblings’ perceptions of their relationship with their sibling with autism. Using the SRQ, this study investigates aspects of sibling relationships that have not been investigated previously in siblings of children with disabilities such as dominance, affection, companionship, intimacy, admiration, and competitiveness. Results indicate that participating siblings of children with autism report less intimacy with their reference sibling and less nurturance by their reference sibling than participating siblings of children with Down syndrome and participating siblings of normally developing children. Participating siblings of children with autism also reported less prosocial behavior in their sibling relationship than participating siblings of children with Down syndrome. Table II. Group Means on Sibling Relationship Questionnaire (SRQ) Factors and Scales Group Autistic Variables SRQ factors Closeness Power Conflict Rivalry SRQ Closeness factor scales Intimacy Prosocial Companion Similarity Admiration of sibling Admiration by sibling Affection SRQ Power factor scales Nurturance of sibling Nurturance by sibling Dominance of sibling Dominance by sibling SRQ Conflict factor scales Quarreling Antagonism Competition SRQ Rivalry factor scales Maternal partiality Paternal partiality Down syndrome Normal develop M SD M SD M SD 3.0 2.6 2.3 2.9 0.6 1.5 0.8 0.3 3.5 1.9 2.4 2.8 0.7 1.7 0.9 0.4 3.1 1.4 3.1 3.0 0.6 2.2 0.9 0.3 1.4 2.8 3.2 2.6 3.7 3.3 4.4 0.6 0.9 1.1 0.9 0.9 1.1 0.8 2.1 3.4 3.7 3.1 3.9 3.9 4.5 0.9 0.8 0.9 0.8 1.0 1.0 0.8 2.1 3.1 3.4 2.8 3.1 3.2 3.9 0.9 0.7 0.8 1.2 0.8 0.7 0.8 3.3 1.5 2.5 1.9 0.8 0.7 1.0 0.9 3.7 2.1 2.6 2.4 0.8 1.0 0.9 1.1 3.1 2.1 2.7 2.2 0.7 1.0 0.9 0.9 2.6 2.4 1.8 0.9 0.9 1.0 2.6 2.7 1.9 1.2 1.1 1.0 3.5 2.9 2.8 0.9 0.7 1.3 2.8 2.8 0.3 0.5 2.7 2.8 0.5 0.5 2.9 2.9 0.5 0.3 406 Kaminsky and Dewey Table III. Gender Means on Sibling Relationship Questionnaire (SRQ) Factors and Scales Gender Male Variables SRQ factors Closeness Power Conflict Rivalry SRQ Closeness factor scales Intimacy Prosocial Companion Similarity Admiration of sibling Admiration by sibling Affection SRQ Power factor scales Nurturance of sibling Nurturance by sibling Dominance of sibling Dominance by sibling SRQ Conflict factor scales Quarreling Antagonism Competition SRQ Rivalry factor scales Maternal partiality Paternal partiality Female M SD M SD 3.1 2.2 2.8 2.9 0.7 1.6 0.9 0.4 3.3 1.7 2.4 2.8 0.6 2.0 0.9 0.4 1.9 2.9 3.5 2.7 3.5 3.5 4.0 0.8 0.8 1.0 1.1 1.0 1.1 0.9 1.8 3.3 3.4 2.9 3.6 3.5 4.4 0.9 0.8 0.9 0.8 0.9 0.9 0.7 3.3 1.7 2.7 2.2 0.8 0.8 1.0 1.0 3.4 2.2 2.5 2.1 0.9 1.0 0.8 1.0 3.0 2.8 2.5 1.0 0.9 1.2 2.7 2.9 1.8 1.1 0.8 1.0 2.9 2.9 0.4 0.4 2.7 2.9 0.5 0.5 Although no previous literature has addressed how autism influences these aspects of sibling relationships, discussion of some of the particular features of autism provides insight into these findings. Siblings of children with autism may be less likely to report high levels of prosocial behavior, intimacy, and nurturance by their sibling, because of the variety of social deficits related to autism. Children with autism may be rated as less nurturant by their siblings, because they often have deficits in understanding other’s perspectives, and tend to be unresponsive to other’s emotions (Rutter & Schopler, 1987). Limitations in cognitive functioning may further impede children with autism from assisting and attending to their normally developing siblings. Similarly, intimacy and prosocial behavior in sibling relationships with children with autism may be inhibited by social and communicative deficits, which are characteristic of autism. Children with autism have deficits such as a lack of reciprocity in conversation, or an overall lack of speech, which likely inhibit intimate interchanges with siblings such as sharing thoughts and feelings (Rutter & Schopler, 1987). Of the children with autism in the present study, 10% could not communicate through either spoken or sign language. Furthermore, many children with autism rarely seek out others for comfort, affection, or to initiate play. Sibling relationships with children with autism may be characterized by less prosocial behavior because children with autism are less likely to initiate interactions with their siblings. This finding is consistent with that of Knott et al. (1995), who reported that children with autism tend to initiate both less prosocial and antagonistic initiations with their siblings. Knott et al. also reported that both children with autism and their siblings were less likely to respond to each other’s initiations. Participating siblings of children with both autism and Down syndrome reported differences in their relationships compared to siblings of normally developing children. Specifically, participating siblings of children with disabilities reported greater admiration of their reference sibling than participating siblings of normally developing children. They also reported that they were less competitive and quarreled less with their sibling than normally developing siblings. These results confirm previous studies, which have indicated that siblings of children with disabilities tend to feel quite positive about their sibling (Bagenholm & Gillberg, 1991; McHale et al., 1986). Previous research has not addressed how having a sibling with a disability influences antagonistic interaction such as quarreling and competitiveness. However, it can be speculated that the cognitive limitations of the child with the disability may have resulted in the normally developing siblings feeling less inclined to compete with the child with the disability. Furthermore, a portion of the children with disabilities, particularly children with autism, had considerable deficits in communication abilities that may have resulted in a reduction in the level of negative verbal interactions among siblings. Particular aspects of sibling relationships also differentiated siblings of children with Down syndrome from siblings of children with autism and siblings of normally developing children. Participating siblings of children with Down syndrome reported that they were more nurturant of their sibling and had higher levels of overall closeness and intimacy than the other groups of participating siblings. They also indicated that their reference sibling admired them more than did participating siblings of children with autism or participating siblings of normally developing children. Participating siblings of children with Down syndrome reported greater affection in their sibling relationships than participating siblings of normally developing children, but not more than participating siblings of children with autism. Sibling Relationships of Children with Autism These results are consistent with previous observational studies reporting that siblings of children with Down syndrome tend to be more prosocial towards their sibling than siblings of normally developing children (Abramovitch et al., 1987). The present results also correspond with research on preschool siblings of children with disabilities, which has found that siblings of children with disabilities tend to be more nurturing of their sibling than normally developing controls (Lobato, 1981). An additional interesting result was that participating siblings of children with autism and Down syndrome did not report greater parental partiality towards their sibling than normally developing children. This finding contradicts anecdotal reports that have stated that siblings of children with intellectual disabilities tend to resent the attention and time that the child with a disability requires from parents (Howlin, 1988; Lobato, 1983). One limitation of the present study is the moderate sample size in relation to the number of variables included in this study. Since autism is a relatively rare disorder in the population, the number of possible participants was restricted. It should be noted, however, that the sample size in this study is similar to or exceeds the sample sizes of published studies involving siblings of children with disabilities (Bagenholm & Gillberg, 1991; McHale et al., 1986). A second limitation of this study is that the majority of participating siblings in this sample (80%) were older rather than younger than their reference sibling in the autism, Down syndrome, or normally developing group. This limits the generalizability of the results of this study to siblings who are younger than the child with autism. Furthermore, because of the small number of younger participating siblings, it was not possible to carry out analyses that compared the younger and older participating siblings. Additional research is needed on the sibling relationships of younger siblings of children with autism. A third limitation of this research relates to the gender distribution of children with autism in this study. As was expected, given the higher rate of autism in boys in the general population, the sample of reference siblings with autism had a larger number of boys than the sample of reference siblings with Down syndrome and the sample of normally developing reference siblings. This resulted in many more participating siblings in the autism group reporting about their relationship with a brother rather than a sister. Due to the lack of reference girls in the autism group, investigation of how samesex versus opposite-sex sibling dyads differ in terms of sibling relationships was not possible. 407 A fourth limitation relates to the applicability of this study’s findings to families of children with autism in the general population. A large portion of families of children with autism and Down syndrome in this sample were recruited from parent support groups. Several of these families were very active in support groups. As only a small number of families in the study did not attend support groups, no formal analyses were conducted to determine whether families who were support group attendees differed from the other families. In addition, this study largely included families who were middle class or upper middle class in terms of their SES. Although there were no differences between groups in SES, the fact that most of the participants were middle to upper class in terms of SES may also limit the applicability of results to siblings of children with autism in the general population. In conclusion, the results of this research provide new information on how autism influences relationships between siblings. Siblings of children with autism reported lower levels of intimacy, prosocial behavior, and nurturance by their siblings compared to children with Down syndrome and normally developing children. This study also highlights some of the positive influences of having a sibling with a disability on sibling relationships. Both siblings of children with autism and Down syndrome reported greater admiration by their siblings and less competitiveness and quarreling in their relationship with their sibling than normally developing sibling pairs. APPENDIX A Adaptive Behavior Questionnaire Form for Parents of Children Who Have Autism or Down Syndrome Please answer the following questions about your child’s abilities and level of independence. 1. What is the name and age of your child with autism or Down’s syndrome? Name Age 2. What is this child’s I.Q., mental age, and/or degree of disability (if known), please specify. 408 Kaminsky and Dewey 3. What grade is your child in at school? 4. At what grade level or age level does your child currently perform at in school? 16. Can your child feed him/herself with a fork or spoon? Yes No APPENDIX B 5. Have you received any special services from the school system for this child such as placement in special classes, teachers aid. Please specify. General Questionnaire for Parents First we would like to ask you some questions about your family. 1. Please complete the following information about all members of your household. 6. Have you received any other types of support services for this child such as respite services, assistance of a professional caregiver in the home etc. Please specify. 7. Can your child communicate with others through spoken language (or sign language)? Yes No 8. Does your child use sentences of four or more words? Yes No 9. Can your child communicate his/her wants or needs to others? Yes No Sex No 11. If your child can read, at what grade level are they currently reading at? Date of Birth Father Mother Child Child Child Child Child 2. Have any children in your family been diagnosed with autism, Down’s syndrome, a language disorder, learning disability, developmental problem, attentional problem or a chronic illness? Yes 10. Can your child read? Yes Name No 3. If YES, what specific illness (es) or type of problem(s) and which child (ren) is/are affected? Name Type of Problem 12. Can your child print/write his or her first name? Yes No 13. Can your child follow simple instructions given to them by a parent? Yes No 14. Can your child dress his/herself completely? Yes No 15. Can your child tie shoelaces into a bow without assistance? Yes No 4. If you have a child with Down’s syndrome or autism please complete questions 1–16 on form E (attached). 5. Are any of your children adopted? Yes No 6. If YES, specify which child (ren) is/are adopted. Name Sibling Relationships of Children with Autism 7. What is your annual family income? below 10,000 51,000–60,000 10,000–20,000 61,000–70,000 21,000–30,000 71,000–80,000 31,000–40,000 81,000–90,000 41,000–50,000 91,000–100,000 FOR MOTHER: 8. What is your present marital status? (please check all that apply) Married Separated Living with someone Never married and not living with someone Divorced Widowed 9. From the list below, please indicate the highest level of education that you completed. a) b) c) d) e) f) No high school Some high school High school diploma Some post-secondary, but no diploma or degree Post-secondary diploma (e.g., technical) University degree 10. What is your occupation? FOR FATHER: 11. What is your present marital status? (please check all that apply) Married Separated Living with someone Never married and not living with someone Divorced Widowed 12. From the list below, please indicate the highest level of education that you completed. a) b) c) d) No high school Some high school High school diploma Some post-secondary, but no diploma or degree 409 e) Post-secondary diploma (e.g., technical) f) University degree 13. What is your occupation? ACKNOWLEDGMENTS Support for this research was provided by grants from the Alberta Heritage Foundation for Medical Research, the Surrey Place Centre Foundation and the Alberta Children’s Hospital Foundation. This study was presented at a meeting of the Society for Research in Child Development, Albuquerque, NM, April 1999. We thank the parents and children who participated in this study. REFERENCES Abramovitch, R., Stanhope, L., Pepler, D., & Corter, C. (1987). The influence of Down syndrome on sibling interaction. Journal of Child Psychology and Psychiatry, 28, 865–879. American Psychological Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Bagenholm, A., & Gillberg, C. (1991). Psychosocial effects on siblings of children with autism and mental retardation: A population based study. Journal of Mental Deficiency Research, 35, 291–307. Blishen, B. R., Carroll, W. K., & Moore, C. (1986). The 1981 socioeconomic index for occupations in Canada. Canadian Review of Sociology and Anthropology, 24, 465–488. Buhrmester, D., & Furman, W. (1990). Perceptions of sibling relationships during middle childhood and adolescence. Child Development, 61, 1387–1398. Dunn, J. (1988). Sibling influences on childhood development. Journal of Child Psychology and Psychiatry, 29, 119–127. Dunn, J., & Munn, P. (1986). Sibling quarrels and maternal intervention: Individual differences in understanding and aggression. Journal of Child Psychology and Psychiatry, 27, 583–595. Gilliam, J. E. (1995). Gilliam Autism Rating Scale Manual. Austin, TX: Pro-Ed. Howlin, P. (1988). Living with impairment: The effects on children of having an autistic sibling. Child: Care, Health and Development, 14, 395–408. Jenkins, J. M., & Smith, M. A. (1990). Factors protecting children living in disharmonious homes: Maternal reports. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 60–69. Knott, F., Lewis, C., & Williams, T. (1995). Sibling interaction of children with learning disabilities: A comparison of autism and Down’s syndrome. Journal of Child Psychology and Psychiatry, 6, 965–976. Lobato, D. (1985). Brief report: Preschool siblings of handicapped children: Impact of peer support and training. Journal of Autism and Developmental Disorders, 15, 345–350. McHale, S. M., Sloan, J., & Simeonsson, R. J. (1986). Sibling relationships of children with autistic, mentally retarded, and nonhandicapped brothers and sisters. Journal of Autism and Developmental Disorders, 16, 399–413. Rutter, M., & Schopler, E. (1987). Autism and pervasive developmental disorders: Concepts and diagnostic issues. Journal of Autism and Developmental Disorders, 17, 159–186. 410 Sparrow, S. S., Balla, D. A., & Cicchetti, D. V. (1984). Vineland Adaptive Behavior Scales Manual. Circle Pines, MN: American Guidance Services. Stocker, C. M. (1994). Children’s perceptions of relationships with siblings, friends, and mothers: Compensatory processes and links with adjustment. Journal of Child Psychology and Psychiatry, 35, 1447–1459. Kaminsky and Dewey Stocker, C., & Dunn, J. (1990). Sibling relationships in childhood: Links with friendships and peer relationships. British Journal of Developmental Psychology, 8, 227–244. Stoneman, Z., Brody, G. H., Davis, C. H., & Crapps, J. M. (1987). Mentally retarded children and their older same-sex siblings: Naturalistic in-home observations. American Journal on Mental Retardation, 92, 290–298.