american journal on mental retardation, 1995, vol. 99, no. 5, … › chdd › guralnick › pdfs...

American Journal on Mental Retardation, 1995, vol. 99, No. 5, 457–476

Parent Perspectives of PeerRelationships and Friendships inIntegrated and Specialized ProgramsMichael J. Guralnick, Robert T. Connor, and Mary HammondUniversity of Washington

In-depth interviews of a large community-based sample of mothers ofpreschool-age children with established disabilities and those classifiedas at-risk were conducted focusing on issues of children's peer relationsand friendships. Results revealed that mothers of children enrolled inboth integrated and specialized programs perceived their respectivesettings as valuable for the development of their child's peer relationsand friendships. Mothers of children in integrated settings reported thattheir child played better and was more social due to the participation ofchildren without special needs. A similar pattern was obtained formothers of children enrolled in specialized programs in response to ahypothetical situation. Nevertheless, both groups of mothers wereclearly concerned about peer rejection and noted the importance ofhaving other children with special needs in the program available as ameans of promoting tolerance and acceptance.

The emergence of community-based pro-grams that integrate preschool-age chil-dren with and without disabilities has beenaccompanied by an extensive examinationof critical issues related to the process ofservice provision and to the general impactof inclusive programs on children andfamilies (Guralnick, 1990a; Lamorey &Bricker, 1993; Odom & McEvoy, 1988).One issue that is central to inclusive pro-grams concerns young children's socialinteractions with their peers and their abil-ity to establish friendships. In part, theprominence of this issue stems from thefact that peer relations and friendshipsrepresent, perhaps more than any otheraspect of development, an expression ofthe philosophical and ideological basis ofinclusive programs. The concepts of ac-

ceptance, rejection, social integration, andsocial isolation are all linked to children'sinteractions with their peers.

In addition to its central role relatedto the goals of inclusive programs, interesthas occurred because establishing suc-cessful relationships with peers andforming friendships constitute vital devel-opmental tasks during the preschool yearsand carry important implications forchildren's cognitive, communicative, andgeneral prosocial development as well asfor an emerging sense of self (Bates, 1975;Garvey, 1986; Hartup, 1983; Howes, 1988;Rubin & Lollis, 1988). Moreover, parents ofpreschool-age children with and withoutdisabilities recognize the significance ofpeer relations and friendships, highly valu-ing social development (Quirk, Sexton,

©American Association on Mental Retardation 457

Ciottone, Minami, & Wapner, 1984).Observational and sociometric re-

search have contributed to our under-standing of the impact of inclusive programson young children's peer relations andfriendships. One general finding is that thepeer-related social interactions of childrenwith disabilities occur at a higher level inintegrated in comparison to specializedprograms (Buysse & Bailey, 1993; Guralnick& Groom, 1988a; Miller et al., in press).Direct observations of social interactionpatterns in integrated settings also suggestthe value of inclusive settings for childrenwith disabilities. Specifically, in integratedsettings, children with disabilities interactextensively with their typically developingpeers; numerous opportunities exist forobservational learning, such as that whichoccurs during parallel play; and the peer-related social/communicative environmentappears to be appropriate and responsivebecause children without disabilities areable to make necessary adjustments inaccordance with the developmental sta-tus of their peers (Guralnick & Groom,1987; Guralnick & Paul-Brown, 1980, 1984;but see Vandell & George, 1981, for anexception).

Yet it is also evident that childrenwith disabilities experience considerabledifficulties with their peers in integratedsettings. In general, despite the occurrenceof more frequent and positive social ex-changes on the part of children with dis-abilities in integrated as opposed tospecialized settings, typically developingchildren actually interact with children withdisabilities less frequently than would beexpected in terms of their availability, chil-dren with disabilities are less accepted andmore rejected based on peer sociometricand discourse analyses, and have far moredifficulty establishing reciprocal friendshipsthan do developmentally matched typi-cally developing children (Guralnick &Groom, 1987, 1988b; Guralnick & Paul-Brown, 1984, 1989; Hadley & Rice, 1991).Moreover, despite the fact that extensiveindividual variation exists, problems in

peer-related social competence are never-theless characteristic of a disproportion-ately large number of children withdisabilities (Guralnick, 1990b), with corre-sponding difficulties in interacting withpeers and establishing friendships expectedto occur in any setting.

The complexity and ambivalence as-sociated with this issue are also reflected inparents' perceptions. Interview, survey, andquestionnaire data addressing the benefitsand drawbacks of integrated programs haveoften included questions related to socialinteractions or social acceptance. In gen-eral, the responses of parents of childrenwith disabilities have indicated that theybelieve integrated programs promote ac-ceptance or facilitate the social skills oftheir child (Bailey & Winton, 1987; Blacher& Turnbull, 1982; Guralnick, 1994; Reichartet al., 1989). However, when questionedabout possible drawbacks of integratedsettings, considerable concerns are reportedin connection with possible rejection orignoring of their child by other children(Bailey & Winton, 1987; Guralnick, 1994;Turnbull, Winton, Blacher, & Salkind, 1982).

It is important to point out that onlya few questions on these general surveysof the benefits and drawbacks of integra-tion have focused on peer relations, friend-ships, or social acceptance. Beyond thefact that these questions have not beenphrased in any uniform way, little or noopportunity has been given to parents toexpress the rationales for their opinions. Ingeneral, investigators have not probed thedetails of the actual or anticipated effectsof inclusive or noninclusive settings nor ofparental beliefs in relation to the possiblecontributions of children with and withoutdisabilities to specific aspects of social playwith peers and to the development offriendships. Also absent are comparisonsacross children with different disabilities, afactor that may influence the perspectivesof parents on these issues (see Stoneman,1993). Yet parents have historically playeda central role in fostering change in servicesystems and provide unique perspectives

458 AJMR , Volume 99, No. 5

on critical issues (Turnbull & Turnbull,1990). Accordingly, in view of the ambiva-lent results from existing surveys of par-ents and observational and sociometricresearch on children's peer interactions, itis vital that a thorough understanding beobtained of parental perceptions of thei mpact of integrated and specialized set-tings on the development of their child'speer relations and friendships.

To obtain information about parentalperceptions in this study, we administereda detailed interview focusing directly onissues related to peer relations and friend-ships to a large community-based sampleconsisting of families whose children wereenrolled in integrated as well as special-ized programs. A unique feature of thesein-depth interviews was the use of probequestions to elicit the rationales of parentsfollowing their initial responses to ques-tions. For children enrolled in integratedprograms, descriptive information wasobtained from parents examining how valu-able they perceived this setting to be inrelation to fostering their child's peer rela-tions and friendships, the contributions ofchildren with and without special needs totheir child's social interactions with peers,concerns regarding rejection, and the im-portance of having other children withspecial needs available in the program. Asimilar set of issues was examined forfamilies of children enrolled in specializedprograms, although a hypothetical situa-tion was established to evaluate the rela-tive contributions of children with andwithout disabilities. In addition to theseanalyses, which emphasized comparisonsamong groups of children with differentdisabilities in integrated or specialized set-tings, comparisons were also made be-tween matched groups whose childrenwere enrolled in integrated and special-ized programs in relation to perceptions ofthe value of their respective setting infacilitating their child's peer relations andfriendships. Parental satisfaction with non-social aspects of the setting and associa-tions between parental perspectives, family

demographics, and child characteristicswere also examined.

MethodSampleChildren meeting criteria for classificationas having either cognitive delays, commu-nication disorders, or physical disabilitiesor being at-risk (see below for definitions)were recruited through formal contact withlocal school districts and community agen-cies providing services to young childrenwith disabilities in a large metropolitancommunity in the northwestern UnitedStates. Because we were interested in par-ents' perspectives in relation to preschoolprograms, only children enrolled in someform of regular integrated (see later discus-sion), specialized, or combination of pro-grams were selected. To be included in thesample, children were required to be be-tween 48 and 71 months of age and meetinclusion criteria for one of the disability orat-risk categories. Children who were le-gally blind, had major uncorrected hearingloss, lived with the primary caregiver forless than 6 months, or currently lived in ahome without a female caregiver wereexcluded from the study. No attempt wasmade to recruit children enrolled in pro-grams serving those with severe or mul-tiple disabilities. Non-English-speakingfamilies were also excluded.

Records of children whose parentsconsented to participate were carefullyreviewed. In addition, for classificationpurposes only, the revised version of theWechsler Preschool and Primary Scale ofIntelligence—WPPSI-R (Wechsler, 1989)was administered to children individually .Full-Scale IQs as well as performance (PIQ)and verbal (VIQ) scores were obtained.The revised version of the Test for Audi-tory Comprehension of Language (Carrow-Woolfolk, 1985) also was individuallyadministered. This test consists of scalesfor word classes and relations, grammaticalmorphemes, and elaborated sentences, butwe only used the total score (standard

Peer Relations 459

score) for categorization purposes in thepresent study.

On the basis of this information, chil-dren were placed into one of four groupsrepresenting specific special needs. (Theterm special needs was used to character-ize children for parents participating in theinterviews. This generic term is generallyunderstood by families, is commonly usedby service providers, and represents thebroad developmental range of children inthis sample.)

The first group contained childrenwith cognitive delays, those who obtainedWPPSI-R Full Scale IQs of 80 or below andwho did not meet criteria for children withphysical disabilities. Children were ex-cluded from this category if they obtaineda PIQ or Test for Auditory Comprehensionof Language-Revised score greater than 90.Classified here as well were children whoreceived scores of 0 on at least threesubtests for either the performance orverbal scales of the WPPSI-R. The Commu-nication disorders group contained chil-dren who obtained a PIQ of 90 or greateror Full-Scale IQ greater than 85 and com-pleted a comprehensive speech and hear-ing assessment administered by qualifiedprofessionals resulting in a categorical di-agnosis of communication disorder and arecommendation for regular therapy. Inaddition, to be classified with communica-tion disorders a child must have met one ofthe following criteria: (a) PIQ > VIQ differ-ential of at least 15 points, (b) a Test forAuditory Comprehension of Language-Re-vised total score of 90 or less, or (c) adiagnosed articulation disorder. The physi-cal disability group was composed of chil-dren with a clearly diagnosed physicalimpairment, such as cerebral palsy or spinabifida. The at-risk group was composed ofchildren who obtained a Full-Scale IQbetween 81 and 90 and did not meetcriteria for any other classification. In addi-tion, to be included in the at-risk category,children must have been receiving earlyintervention services by their local schooldistricts. Most common areas of concernwere related to fine- and gross-motor de-

lays or relatively minor speech and lan-guage difficulties.

Recruitment continued until a mini-mum of 280 families agreed to participate,including at least 25 families from thephysical disability group (due to its lowincidence) and the at-risk group (due tothe limited number of at-risk children re-ceiving services by school districts). Fromthis larger sample, mothers of 262 subjectsin the four special needs groups completedthe interview.

Child Characteristics andFamily Demographics

Other child characteristic measures wereobtained in addition to the WPPSI-R andthe Test for Auditory Comprehension ofLanguage-Revised scores. First, to supple-ment the receptive language assessmentof the Test for Auditory Comprehensionof language-Revised, the expressive com-ponents of the Preschool Language Scalewere administered (Zimmerman, Steiner,& Pond, 1979). Because of the lack ofstandardization, only raw scores wereused (range = 0 to 48 for verbal abilityand 0 to 23 for articulation). Second, theVineland Adaptive Behavior Scales Sur-vey Form (Sparrow, Balla, & Cicchetti,1984) was administered to mothers bytrained interviewers. Standard scoreswere obtained for each of the four do-mains (Communication, Daily LivingSkills, Socialization, and Motor Skills) aswell as for the total adaptive behaviorscore. Third, the mother's assessment ofher child's behavior problems was ob-tained from the Child Behavior Checklist(Achenbach & Edelbrock, 1983). Moth-ers rated the frequency of different be-havior problems from a 118-itemquestionnaire using a 3-point scale. Onlythe broad band internalizing and exter-nalizing scales (Tscores), in conjunctionwith a total behavior problem score,were used. Higher scores indicate greaterperceived behavior problems.

Finally, responses to a parent ques-

460 AJMR, Volume 99, No. 5

tionnaire provided basic demographic in-formation on marital status, child's genderand ethnicity, maternal age, and the par-ents' employment status, occupation, andeducation. The Hollingshead Four FactorIndex of Social Status (Hollingshead, 1975)was used to calculate a measure of familystatus (range = 8 to 66). Table I presentsthe number of subjects with completedinterviews from each of the four groups, aswell as individual child characteristic anddemographic measures.

ProcedureFollowing agreement by the family to par-ticipate in the study, we mailed them apacket containing questionnaires and con-sent forms. The questionnaires includedrequests to provide demographic informa-tion and to complete the Child BehaviorChecklist. (This mailing also included otherscales related to their child's peer relationsthat were part of a larger investigation ofparental perspectives on children's peer-related social development as well as as-

Table 1Sample Characteristics by Group

Group

Cognitive delay'(n = 116)

Communicationdisorder(n = 84)

Physical disability(n = 30)

At-risk(n = 32)

Sample characteristic Mean or % SD Mean or % SD Mean or % SD Mean or % SD

Program typeIntegrated (%) 16.4 34.5 20.0 15.6Specialized (%) 83.6 65.5 80.0 84.4

DemographicsChild's age° 58.65 7.36 58.15 5.61 60.00 6.17 58.59 6.69Child's gender° (%) 69.0 75.0 53.3 75.0Child's ethnicity° (%) 82.9 88.0 89.7 74.2Mother's age' 34.63 6.07 33.80 4.64 34.10 5.44 35.48 7.04Mother's education. 13.99 1.90 13.90 1.67 14.28 1.49 13.52 2.34Marital status' 92.8 82.9 75.9 71.0Family social status 43.12 13.70 43.91 13.69 39.55 12.69 38.68 14.05

Child characteristicsWPPSI-RFull-Scale IQ 63.39 11.29 96.68 11.01 85.87 22.23 85.69 2.48Performance IQ 65.29 12.60 104.14 10.82 84.30 22.51 86.03 5.11Verbal IQ 67.14 10.93 91.11 12.87 89.77 19.09 88.09 4.47

TACL-R'Total scale 63.29 18.48 91.69 15.16 85.97 19.80 86.00 10.01Word class & relations 62.59 32.29 92.42 15.66 86.17 24.64 89.06 9.90Grammatical

morphemes 68.51 18.53 91.49 15.29 85.43 20.24 87.41 11.62Elaborated sentences 74.84 10.90 96.42 14.74 93.57 16.78 89.00 10.34PLSVerbal ability 17.65 8.14 28.54 6.71 30.90 9.40 29.06 6.45Articulation 10.92 6.16 14.18 4.19 15.70 6.82 15.00 4.46VABSTotal adaptive behavior 65.97 11.16 84.61 14.41 72.60 15.95 75.72 10.44Communication 69.08 11.93 82.45 11.48 87.33 17.40 78.19 10.65Daily living skills 69.58 13.25 89.02 15.53 73.27 17.30 81.03 13.68Socialization 79.59 11.73 92.23 14.97 92.07 16.94 85.63 13.57Motor skills 66.25 16.17 89.17 17.24 58.03 18.50 78.56 16.32

CBCL'Total behavior problems 56.86 10.28 55.76 10.80 54.90 7.18 58.32 8.17Externalizing 53.04 11.13 53.77 11.84 51.00 9.32 55.32 8.13Internalizing 48.31 10.07 52.12 10.28 50.62 8.72 53.77 8.21

'Number of subjects may vary slightly for each measure due to incomplete data sets, particularly on the WPPSI-R forchildren with significant cognitive delays. °Calculated in months. Percentage male. Percentage Caucasian. 'Calculatedin years. 'Percentage partnered. "Hollingshead Four-Factor Index of Social Status. °Wechsler Preschool and Primary Scaleof Intelligence-Revised; VIQ = verbal IQ; PIQ = performance IQ; FSIQ = Full-Scale IQ. 'Test of Auditory Comprehension-Revised. 'Preschool Language Scale. "Vineland Adaptive Behavior Scales. 'Child Behavior Checklist.

Peer Relations 461

sessments of social support, stress, and thechild's social network.) Two separate ap-pointments with the mothers were thenarranged to administer the Vineland Adap-tive Behavior Scales and to complete theinterviews. Questionnaires and consentswere also collected during these visits, andadditional information was gathered re-lated to the larger study.

During this period, individual testingof children was carried out primarily attheir preschool program by research assis-tants trained and supervised on the variousassessment instruments by licensed psy-chologists and a communication disordersspecialist. Training staff made periodicobservations of assessment sessions dur-ing the course of the study, and eachmeasure was rechecked for accuracy andcorrect use of tables before final entry foranalysis.

Placements of Children inPreschool Programs

Mothers were first interviewed to obtaininformation with respect to their child'scurrent preschool placement. Informationwas obtained as to the nature and extent ofthe involvement of typically developingchildren in their child's preschool programto permit a classification of the programinto either an integrated (standard involve-ment or planned interactions with typicalchildren) or a specialized program (pro-gram contains only children with specialneeds and no planned exchanges withtypical children exist). Programs consid-ered as mainstreamed, reverse main-streamed, and those containing plannedintegration experiences with typical chil-dren were all placed in the integrationcategory for subsequent analyses (seeGuralnick & Groom, 1988a; Odom &McEvoy, 1988, for definitions). From thetotal of 262 children with special needs, 59were classified as participating in inte-grated programs, the remaining 203 inspecialized programs. Children enrolledconcurrently in both specialized and inte-grated programs were placed in the inte-

462 AJMR , Volume 99, No. 5

grated category. Information was also ob-tained from the mother on the program'sstaffing patterns (including resource staffavailable), location, extent of child's par-ticipation, developmental approach, goalsof the program, reasons why the currentprogram was selected, and the frequencywith which the mother observed her childin the program.

Interviews on Peer Relationsand FriendshipsIn-depth interviews of mothers of childrenin the four special needs groups werecarried out to examine the relation amongdisability group, type of educational place-ment, and perceptions of the developmentof their child's peer relations and friend-ships. Interviewers presented a series ofindividual questions that could typically beresponded to with brief answers indicatingagreement, disagreement, uncertainty, orlack of information. If an answer was notspontaneously offered, each question wasthen followed by "probes," which askedmothers to provide rationales for theirresponses. Mothers' responses were re-corded verbatim and then coded (see fol-lowing discussion).

Separate but related interviews weredeveloped for mothers of children partici-pating in integrated or specialized settings.In the first section mothers of children inintegrated programs were asked a series ofquestions examining their perceptions ofthe value of that setting in facilitating theirchild's peer relations. (Occasionally, ques-tions on similar topics were placed indifferent sections due to the length of aparticular section or to minimize the pos-sibility that mothers would fall into a gen-eral pattern of responding.)

Specific questions focused on thevalue of the setting (and correspondingrationales) in relation to sharing, resolvingconflicts, cooperating with peers, encour-aging play, and learning from observingother children in the program. Additionalquestions addressed the behavior of theirchild's peers in relation to social play

interactions and whether gains in peer-related social play occurred, possibilitiesfor rejection by peers, and the impact ofthe setting on their child's self-esteem.

The second section addressed theissue of friendships. Mothers were askedwhether the setting in which their childparticipated facilitated the development offriendships (a specific definition was pro-vided). A list of best friends was alsoobtained, including information regardingthe friend's age, gender, special needs,duration of the friendship, and locationwhen first met. In a related issue, motherswere asked about the potential value ofplaying with younger or same-age peerswithout special needs.

In the first two sections, when probed,mothers could have indicated that thesetting was valuable because of the charac-teristics of the peers. However, teacher-related and other rationales, including theimpact of classroom structure, were alsoexpected to occur frequently. Accordingly,the next two sections of this interviewwere designed to focus the mothers moredirectly on the impact of the characteristicsof the peers in the setting. Specifically, thethird section probed differences that mightexist as a consequence of opportunities tointeract with children with and withoutspecial needs. Mothers were asked theextent to which their child played withpeers representing special needs ornonspecial needs groups and whether theirchild's peer-related social interactionsbenefited more from playing with chil-dren from one or the other group.

The fourth section focused on moth-ers' perceptions of the value of havingchildren with special needs available intheir child's preschool program. The pri-mary issues addressed were the impor-tance of having other children with specialneeds in the program, possible uniquebenefits to their child in terms of socialplay, effects on the child participating in aspecialized program (containing childrenwith similar special needs), and any con-cerns should children with special needs

form separate playgroups in integratedprograms.

The final section of the interviewexamined the general educational/ devel-opmental environment that is not directlyconnected to peer relations and friend-ships. An initial set of questions focused onthe extent to which appropriate demandswere being placed on the child to learnskills and information not related to socialplay, whether the pace and complexity ofthe program were appropriate, how theeducational program was adapted to theirchild's special needs, and whether anycharacteristics of the program preventedfull participation of their child, includingphysical barriers. These questions werepresented without reference to the pres-ence of children without special needs.Mothers were also asked about the relatedservices their child received, whether anIndividualized Educational Plan (IEP) wascompleted, and whether and how teachersshould discuss or describe their child'sspecial needs to other children in theprogram.

The interview protocol for childrenenrolled in specialized programs followeda similar structure. The first section fo-cused on the value of the specializedsetting with regard to children's peer rela-tions; the second, on friendships. Onemajor difference, however, was the thirdsection, which focused on mothers' per-ceptions of the relative contributions ofchildren with or without disabilities. In thisinstance, respondents were asked to envi-sion a situation in which their child wasenrolled in an integrated program. Oncethat was established, a set of questionssimilar to those of mothers of childrenenrolled in integrated settings was pre-sented. Emphasis was placed on the poten-tial impact of the presence of childrenwithout special needs on the social skills,conflict resolution, friendship formation,and rejection of their child with specialneeds. Questions also were similar to theintegration interview regarding age prefer-ences, the possibility of different expecta-

Peer Relations 463

tions placed on the children with specialneeds by teachers, issues related to thepace and complexity of the setting, andimpact on the child's self-esteem or self-image. The remaining two sections of theintegration interview focused on the valueof the availability of children with specialneeds and an assessment of the educa-tional/developmental environment.

CodingAs noted, each question typically gener-ated a brief response indicating overallagreement, disagreement, or uncertainty.A series of codes reflecting the range ofpossible initial answers (yes, no, alright,fair, not matter, not know/not sure) as wellas other codes specific to individual ques-tions were established.

For probes, we conducted a contentanalysis by using an iterative process foreach separate question. By maintaining abroad set of response categories (schemas),we were able to apply the same schema toa series of related questions. A total of 35separate coding schemas were needed tobe responsive to all questions on the twointerviews. Initial categories and their defi-nitions were established on a randomsample of 30% of the responses to a ques-tion. For certain questions for the integra-tion interview, we added additionalcategories subsequent to the initial 30%sample. However, no new categories withineach schema were needed for the special-ized interview.

The detailed coding manual that wasdeveloped (available from the first author)contained an extensive definition of eachcategory as well as sample responses. Forexample, one schema was applied to theseries of questions concerning the value ofthe setting (integrated or specialized) infacilitating the peer relations of the childwith special needs. Sixteen categories weredeveloped, including major areas relatedto program quality (e.g., the importance ofthe teacher's role or the structure of theprogram) and the presence of peers (spe-cific interactions with children with orwithout special needs). Separate catego-

ries were formed to capture the positive ornegative impact of these factors. Whenapplicable, multiple categories were coded.

ReliabilityFive coders were trained for a 6- to 8-weekperiod. Following a thorough study of thecoding manual and discussions to clarifyissues, we scored a series of practice inter-view protocols and provided feedback.We selected 10 interviews to qualify forcoding the main set of interviews to serveas standards. Each interview served as theunit of analysis during the training phase,and coders were required to achieve over-all agreement of 85% with the previouslyestablished codes as determined by thecoding coordinator.

Interobserver agreement was alsoobtained throughout the course of thecoding for 25% of the interviews selectedrandomly for each coder. In this instance,both percentage agreement and Cohen's(1960) Kappa were applied to each ques-tion across interviews. For the integrationinterview, mean percentage agreement was86.7% (range = 50 to 100%); the mean forCohen's Kappa was .79 (range = .38 to1.00). The specialized interview yielded amean percentage agreement of 91.7% (range= 67 to 100%) and a mean Cohen's Kappaof .86 (range = .40 to 1.00). Individualquestions were eliminated if they failed toachieve 65% agreement and a Kappa coef-ficient of .60. Those individual questionsaddressing common issues that were com-bined for analysis were included if theoverall reliability met the criteria. Applyingthese criteria, we eliminated five stem ques-tions and nine probe questions from theintegration interview, but only one stemand one probe question from the special-ized interview.

ResultsThree separate analyses of the data werecarried out: (a) children enrolled in inte-grated programs, (b) children enrolled inspecialized programs, and (c) matchedgroups comparing children enrolled in


integrated and specialized programs. Thefirst two sets of analyses were primarilyconcerned with mothers' descriptions oftheir children's peer relations and friend-ships in relation to the educational settingand their associations with child and familycharacteristics. The comparisons betweenthe integrated and specialized settings pro-vided an opportunity to examine moredirectly any differences in mothers' per-ceptions with regard to the value of eachsetting in connection with the develop-ment of their child's peer relations andfriendships. Demographic factors listed inTable 1 did not differ across the fourspecial needs groups with regard to type ofprogram, p > .05.

Integrated ProgramsIn this section, we provide a descriptiveanalysis of the overall responses of moth-ers whose children were enrolled in inte-grated programs. Mean proportions forinitial responses to the questions are pre-sented as are responses to probe questionswhen informative. For the first section ofthe interview, responses were obtainedand averaged for seven interrelated ques-tions on the value of the setting in relationto their child's learning to share, deal withconflicts, and cooperate as well as to ques-tions related to the appropriate respon-siveness of their child's peers during socialplay. As indicated in Table 2, mothers wereoverwhelmingly favorable in their views ofthe value of this setting, responding posi-tively to a high proportion of items (mean_ .81, SD=.21). In fact, virtually all mothersperceived their child to have made recentgains in learning to share, resolve conflicts,and play cooperatively with others (83%).(Except for the initial question in the firstsection in which items were summed, allremaining data are presented as percent-ages of positive responses [usually yes] inrelation to all other possibilities [e.g., no oruncertain].

Probes of initial responses to obtainmothers' rationales related to the value ofthe setting for their child's peer relations

revealed an interesting pattern. As indi-cated in Table 2 (only categories that wereselected at least 10% of the time wereincluded), three (positive) categories pre-dominated. First, mothers believed that theintegrated setting promoted the develop-ment of their child's peer relations due tothe effectiveness of the adults (usuallyteachers) in the setting. This responseencompassed both their technical skillsand personal characteristics. Second, moth-ers noted the importance of the quality ofthe program with an emphasis on struc-tural aspects. This included the availabilityof appropriate materials, organization ofthe curriculum, and the physical arrange-ment of the room as they might affect peerrelations. Third, mothers attributed thesetting's value to the presence of peers ingeneral. Opportunities for social learningand positive participation with peers ingeneral characterized this category. It isimportant to note that other categoriesemphasizing the availability of typicallydeveloping children or children with spe-cial needs were possible but were rarelycoded. (Section III of the interview ad-dressed this issue more directly.)

It is interesting that when asked di-rectly about the possible impact of thepresence of children without special needson their child's self-esteem, only 15% ofthe mothers were concerned. In fact, 27%thought the effect was a positive one.However, 40% of the mothers were con-cerned about possible rejection of theirchild by peers in the integrated setting.

For Section II, in addition to perceiv-ing the integrated setting to be of value interms of general peer relations, mothersalso found that this setting supported thedevelopment of individual friendships. Infact, 75% of the mothers perceived theintegrated program to be a good setting fortheir child to make friends, and most (76%)were able to identify one or more bestfriends in their child's class (see Table 2).Most children identified as friends werewithin one year of their own child's chro-nological age—CA (91%), were the samegender (91%), and met at the preschool

Peer Relations 465

Table 2Initial Responses and Responses to ProbeQuestions for the Integrated Programs Interview

Proportion ofrespondents

Interview (n = 59)

Section l: Value of Setting for PeerRelations

Positive: sharing, resolve conflicts,cooperate, etc.'

Probes (positive)Effectiveness of adults

Effectiveness of structure

Effectiveness of peers in general

Gains in peer relationsEffects on self-esteem

PositiveNegativeNeutral

Rejection by peersSection It: Friendships

Positive value of settingOne or more best friendsFriends with nonspecial needsGet together outside of preschoolEnough friends (general)

Enough friends (nonspecial needs)Enough friends (special needs)

Prefer same age for nonspecial needschildren

ProbesModel appropriate skillsSimilar interests

Prefer younger for nonspecial needschildren

ProbesSimilar developmental levels

Section III: Comparison Between Specialand Nonspecial Needs Children

Play separatelyPlay with special needsPlay with nonspecial needsPlay better and more social due tononspecial needs childrenProbes

Competence model (positive)Not related to childrenNot notice differences

Learns more social skills from nonspecialneeds children

ProbesCompetence model (positive)

Not related to characteristics ofpeers

Learns to share, resolve conflicts andcooperate from nonspecial needs

ProbesCompetence model (positive)

Not related to characteristics ofpeers

Section IV: Value of Special Needs ChildrenImportant to have special needs peers in

programProbes

Similar skills and abilities



Interview (n = 59)

Encourages tolerance by peers .25Better program structure/services .13

Unique benefits from playing withspecial needs peers .80

ProbesEncourages tolerance andacceptance toward peers .43

Increased self-efficacy .22Improved social skills .13

Comfortable if child only one withspecial needs .46

Section V: Educational/DevelopmentalEnvironment

Appropriate demands for skills andinformation .78

Program adapted to special needs .53Concerns regarding pace/complexity

demands .07Program limits full participation .14Completed an IEP .63Changes needed to improve program .39

ProbesImprove program structure .24Improve adult effectiveness .14

Obtaining special services needed .56

'Data for this question are proportions of mothersresponding positively averaged across the seven items,with SDs in parentheses. Values in the table for otherquestions (and probes) are proportions of mothers codedpositive for the initial response and proportion selectingeach category for probe items. Only probe responses of.10 or greater are presented in the tables.

.81(. 21)

.27(. 23).28

(. 20).32

(.22).83

.27

.',5

.58

.40

.75

.76

.65

.48

.48

.55

.63

.71

.16

.30

.24 (93%). Of interest (see Table 2), 65% of the

13 children identified as best friends did nothave any special needs, and nearly half ofthe friends played together outside of the

15.80 school program as well. Nevertheless, over.90 50% of the mothers in the sample still did

.50not feel their children had a sufficientnumber of friends (with or without special

.35 needs).

.06

.28Finally, the issue of the CAs of their

child's peers was examined. Mothers were.66 asked whether the contributions of chil-

.60 dren without special needs would be morevaluable if they were same-age or younger

.20 than their child. Seventy-one percent be-

26 lieved that their child's peers should be thesame CA, primarily because they model

.18 appropriate skills and have similar inter-31 ests. Twenty-four percent of the mothers

believed that a younger nonspecial needs

.69peer would contribute more because ofsimilarities in developmental level.

.29 Section III provided an opportunity

for mothers to compare the relative ben-efits their child might derive from interact-ing with special or children without specialneeds. As noted in Table 2, relatively fewmothers (15%) believed that separate sub-groups of children formed during play thatwere based on developmental status, andmost thought that their child played withchildren with and without special needs.Overall, half the mothers thought that theirchild played better and was more socialdue to the participation of the childrenwithout special needs in the program. Whenprobed, 35% of the mothers believed thatthis was a consequence of the fact that thechildren without special needs provided apositive model of competence (stimula-tion, motivation, responsiveness). Motherswho believed that their child's social playdid not benefit from the participation ofchildren without special needs usuallynoted that other program factors (e.g.,class size) were more important or thatchildren simply do not notice the differ-ences in other children (28%).

A more specific question focusing onwhether their child learned more socialskills from children without special needsresulted in over two thirds providing apositive response, which, upon probing,was again attributed to the competence ofthe children without special needs. How-ever, when the question was even morespecific, namely, addressing the child'sability to take the perspective of others inmore complex ways (i.e., share, resolveconflicts, cooperate), a much smaller pro-portion of mothers felt that children with-out special needs provided any benefits(26%). As revealed by the probes, appar-ently a competence ►model is not judgedsufficient for these more complex aspectsof peer-related social competence. In fact,31% of the mothers believed that theseforms of social interaction were not associ-ated with the characteristics of their child'speers, either with or without special needs.

In Section IV, mothers were inter-viewed with regard to the value of thechildren with special needs from the per-spective of peer relations. We note that

overall, approximately one quarter of thechildren in integrated programs had spe-cial needs. As shown in Table 2, 69% of themothers believed that it is important tohave other children with special needs inthe program. Probes revealed that mothersfelt that having other children with similarskills and abilities was important and thatby having a number of children with spe-cial needs in the program, all childrenwould be more likely to develop tolerancefor and an understanding of differences. Itis interesting that 13% of mothers observedthat by having more children with specialneeds in the program, special servicescould be more easily and effectively adaptedto the special needs of their child. Whenasked about unique benefits that resultfrom playing with children who have spe-cial needs (80% agreed such benefits ex-ist), the themes of tolerance, acceptance,and promoting more positive attitudes to-wards others were prominent (see Table2). Others noted the possibility of in-creased social skills through modeling ofchildren with special needs and the ben-efits to the child's sense of self throughacceptance of his or her own level of skillsand ability. Only 46 0

/6 of the mothers wouldbe comfortable if their child were the onlyone with special needs in the program.

The final section of the interview(Section V) addressed mothers' percep-tions of the integrated setting as a valuabledevelopmental/educational environment.As seen in Table 2, the overall responsewas positive, with most parents believingthat the program places appropriate de-mands on their children to learn non-play-related skills and information and that theprogram is specifically adapted to theirchild's special needs. Few concerns werenoted with regard to the pace and com-plexity of the program or any limits to fullparticipation. In addition, 63% of the chil-dren had a completed IEP. Many mothers(39%) did suggest that changes were neededto improve the program, particularly theprogram's structure. A somewhat largerpercentage thought that additional special(related) services would be beneficial.

Peer Relations 467

Comparison Between Children WithCognitive Delays and Communication Dis-orders. In this analysis, comparisons weremade between mothers of children withcognitive delays (n = 19) and those withcommunication disorders only (n = 29).The samples from children classified as at-risk and those with physical disabilitieswere too small to analyze meaningfully. Itis important to note that the cognitivedelay and communication disorder groups,although a subset of the larger sampledescribed in Table 1, still maintained equiva-lent demographic factors; no significantdifferences were obtained for any measurebetween the two groups, separate t tests: p> .05. Due to the large number of ques-tions, a more conservative significancelevel, p < .01, was set for all comparisonsbetween children with cognitive delaysand communication disorders based onmothers' responses to the interview.

Despite certain trends, mothers fromboth groups held similar views for bothinitial responses and probe questions. Theonly reliable difference that did occur wasrelated to the number of friends withoutspecial needs. Mothers reported that 80%of the children with communication disor-ders had enough friends without specialneeds in comparison to only 27% for chil-dren with cognitive delays, x 2 (1, N= 30) =656, p < .01. However, this result may wellbe attributed to the fact that fewer childrenwithout special needs were enrolled inintegrated programs for children with cog-nitive delays than those with communica-tion disorders, t(44) = 3.08, p < .01. In fact,children with cognitive delays were in-volved less frequently in fully mainstreamedprograms (i.e., more reverse mainstreamedor integrated programs, x 2 (2, N = 48) =13.35, p < .001, were more likely to belocated in public schools, x 2 (1, N = 48) =6.89, p < .01, and have more resource staffavailable, t(44) = 2.76, p < .01. It is interest-ing that despite these differences in pro-gram type, mothers held similar viewsregarding the program's impact on theirchild's peer relations, friendships, and re-lated issues.


Specialized Programs

A total of 203 children from the largersample were enrolled in specialized pro-grams. As indicated in Sections I and II ofTable 3, a substantial proportion of moth-ers of children enrolled in specializedprograms perceived the setting to be of valuein promoting their child's peer relationsand friendships. For children's peer rela-tions, the rationale (see probes) providedby mothers reflected a pattern similar tothat described for mothers of childrenenrolled in integrated programs. For friend-ships, 76% of the mothers believed thespecialized setting to be a good one fortheir child to make friends, primarily attrib-uting to this setting the availability of peersin general. Moreover, nearly 80% of themothers reported that their child had oneor more best friends in the program. Friendswere typically similar in CA (96% withinone year of age), of the same gender(89%), and met at the preschool (97%). Allof the children's best friends (100%) alsohad special needs and about one third alsoplayed together outside of school.

Table 3Initial Responses and Responses to ProbeQuestions for the Specialized Groups Interview


Interview In = 203)

Section I: Value of Setting for PeerRelations

Positive: sharing, resolve conflicts, .79cooperate, etc.' (.19)Probes (positive)

Effectiveness of adults .20(20)

Effectiveness of structure .25(. 20)

Effectiveness of peers in general .30(.22)

Effectiveness of own child ' s .18behaviors (.18)

Gains made in peer relations .81Probes

Less negative behavior .23More social skill .63More assertive .07

Section II: FriendshipsPositive value of setting .76

ProbeAvailability of peers in general .45

One or more best friends in class .79(continued next page)

(Table 2 continued)


Interview (n = 203)

Section III: Comparison of Peers With andWithout Special Needs°

Play better and more social if nonspecialneeds children in program .35

ProbesProvide competence model .26Not related to child characteristics .22Children don't notice differences .19

Learn social skills more from nonspecialneeds children .51Probes

Provide competence model .49Not related to child .11Not notice differences .13

Learn to share, resolve conflicts, andcooperate more from nonspecial needschildren .08Probes

Not related to characteristics of peers .39Not notice differences .28

Enough friends if nonspecial needschildren in program .90

Enough nonspecial needs friends .84Enough special needs friends .88

Effect on self-esteem .11Probes

Nonspecial needs children providecompetence model .10

Not related to children .20Not notice differences .21

Would be concerned about rejection .51Probes

Negative behavior of own child .11Negative peer attitude .24Positive behavior of own child .25Positive peer attitude .12

Concerns about pace or demands .36Probes

Negative regarding child's ability .30Negative regarding program .12Positive regarding child's ability .27Positive regarding program .13

Teacher expectationsNot any different .40Higher .31Lower .09

Age nonspecial needs children wouldbe more valuableSame age (within a year) .70

ProbesModel appropriate skills .37Similar interests .21

Younger .23Probe

Similar developmental level .17Section IV: Value of Having Special NeedsChildren Available

Important to have other special needschild in program .79

ProbesSimilar skills important .38Encourages tolerance by peers .25

Better program structure tu re/sery ices .18Benefits from playing with special needschild .78


Interview (n = 203)

ProbesEncourages tolerance andacceptance toward peers .60

Increased self-efficacy .16Improved social skills .12

Section V: Educational/DevelopmentalEnvironment

Appropriate demands for skills andinformation .85

Program adapted to special needs .86Getting special services needed .81Completed an IEP .86Concerns regarding pace/complexity .02Program limits full participation .06Changes needed to improve program .30

ProbesImprove program structure .19Improve adult effectiveness .04

'Data for this question are proportions of mothers re-sponding positively averaged across the seven items, withSDs in parentheses. Values in the table forother questions(and probes) are proportions of mothers coded positive forthe initial response and proportion selecting each categoryfor probe items. Only probe response of .10 or greater arepresented in the tables. 'Items in Section III are speculativebased on supposition that the child is enrolled in programincluding children with and without special needs.

It is interesting to note that only 28%of families reported that they had an optionto enroll their child in a program contain-ing children without any special needs. Ofthose who did not have an option, 44%stated that they would have preferred tohave a choice of program that includedchildren without special needs. In SectionIII this issue was probed further and wasunique in that mothers were provided withthe opportunity to speculate about theirchild's peer relations and friendships if heor she were to be enrolled in an integratedprogram.

Specifically, mothers were asked toimagine that their child was enrolled in anintegrated program in which half of thechildren did not have any special needs. Asindicated in Table 3, only 35% of themothers thought their child would playbetter and be more social, primarily attrib-uting those benefits to the competencemodels presented by children without spe-cial needs. For the same reason, 51% of themothers believed that their child wouldlearn more social skills from children with-out special needs. However, as was the

Peer Relations 469

case for mothers of children enrolled inintegrated programs, only 8% of the moth-ers felt that their child would benefit fromchildren without special needs for moredemanding social tasks (i.e., sharing, re-solving conflicts, playing cooperatively).For the most part, mothers attributed thisbelief to their contention that the charac-teristics of peers are not particularly impor-tant for these more complex social skills.Nevertheless, most mothers believed thattheir child would have enough friends inthis program (90%), with a sufficient num-ber of friends who did (88%) and did not(84%) have special needs. Mothers alsogenerally did not believe that participationof children without special needs wouldhave any impact on their child's self-es-teem (11%).

However, slightly over 50% of themothers were concerned about their childbeing rejected in the integrated program.Probes revealed that concerns about rejec-tion revolved around the attitudes of thepeers in the setting and their own child'sbehavior (see Table 3). Those not con-cerned emphasized the positive behaviorof their own child as well as positive peerattitudes. Because more general questionsregarding rejection have suggested thatmothers are more concerned about peerrejection in integrated than in specializedsettings (Guralnick, 1994), further analyseswere conducted. Specifically, the groupwas divided into mothers concerned aboutrejection of their child in integrated set-tings (51%) and those indicating an ab-sence of concern (49%). Demographicsand the distributions of children in the fourspecial needs groups were examined firstand revealed no differences on any mea-sures between the two groups, p > .05.However, analyses of child characteristicmeasures indicated that mothers concernedabout rejection had children who obtainedlower scores for Full-Scale IQ, 1(184) _1.99, p < .05, VIQ, 1(184) = 2.52, p <.05, andthe Vineland motor scale, 1(194) = 2.25, p<.05, than children of mothers not con-cerned about rejection. In addition, chil-dren whose mothers were concerned about


rejection had significantly more total be-havior problems as measured by the ChildBehavior Checklist, t(194) = 2.92, p < .01.

Mothers were also asked about re-lated aspects of the hypothetical integratedprogram. As Table 3 indicates, about onethird were worried about the pace ordemands of the program as a consequenceof the involvement of children withoutspecial needs, and about the same propor-tion felt that teacher expectations of theirchild would be higher. Only 9% felt thatexpectations would be lower. Most moth-ers (70%) also believed that children with-out special needs of the same age ratherthan younger children would be morevaluable primarily due to the existence ofsimilar interests and modelling of appro-priate skills.

In Section IV, the questions focused onthe value of being with peers who also havespecial needs. Nearly 80% of the mothersthough that it was important that other chil-dren with special needs participate in theirchild's program. They felt that it wouldbenefit their child to have children availablewith similar skills, that tolerance and accep-tance by peers would be enhanced, and thatthere would be an overall positive effect onthe program's structure and services. Moth-ers also reported that their child wouldbenefit or has benefitted in specific waysfrom playing with children with special needs.A high percentage (60%) believed that play-ing with other children with special needsencourages more tolerance and acceptanceof their peers. Increased self-efficacy andimproved social skills were also noted asbenefits, but to a lesser extent (see Table 3).

In Section V, mothers were asked toevaluate then child's current specialized pro-gram in terms of the quality of the generaleducational/developmental environment.Mothers in this sample were overwhelminglypositive, noting that (a) demands for nonso-cial-related skills and information were ap-propriate, (b) the program was adaptedspecifically to meet their child's needs, (c)needed special services were available, and(d) their child had a completed IEP. Mothersalso had virtually no concerns about the pace

and complexity of the program or that limitswere placed on their child's full participation.On average, 30% of the mothers believed theprogram could be improved, primarilythrough changes in the program's structure.

Comparisons Among Different Groupsof Children and Correlations WithinGroups. Separate analyses were conductedcomparing responses of mothers of chil-dren falling into each of the four specialneeds groups. No significant differences, p> .01, were obtained, suggesting that irre-spective of the type of their child's disabil-ity or at-risk status, mothers hold similarviews across all dimensions investigated(i.e., peer relations, friendships, hypotheti-cal comparisons for children with andwithout special needs, value of havingchildren with special needs available, andthe educational/developmental environ-ment). Moreover, there was no associationbetween child's type of disability and anystructural aspects of the program such asthe number of children with special needsin the program, the number of teachers oraides, or the availability of resource staff.Finally, separate matrices were developedcorrelating demographic and individualchild measures (e.g., IQ, language scores)with each of the major responses to theinterview questions. No meaningful pat-terns were obtained.

Comparisons Between Specializedand Integrated Programs. Sections I and Vof both the integrated and specializedprogram interviews contained related ques-tions. To obtain comparisons not con-founded by family or child characteristics,we formed equivalent groups by matchingeach of the 59 children enrolled in anintegrated program with children who hadsimilar family and child characteristics fromthe specialized groups. The following pro-cess was followed to accomplish this com-parison. First, children from the same specialneeds group (cognitive delay, communica-tion disorder, physical disability, at-risk) asthe to-be-matched (integrated) child wereidentified from among those enrolled inspecialized programs. A small subgroupwas then formed based on key demo-

graphic and child characteristics withinpredetermined ranges. Specifically, a sub-group was formed on the basis of CA (±6months), family social status (±10), Full-Scale IQ score (±10), and Vineland Adap-tive Behavior Scales total score (t10). Fromthis subgroup, one child was randomlyselected. Occasionally, the range had to beexpanded to achieve a match. Neverthe-less, this process resulted in matched groupsconfirmed by two-tailed t tests, p >.01, andchi-square tests (for dichotomous mea-sures) carried out on all child characteristicand demographic variables.

Separate t tests for the proportion ofmothers reporting on the value of theirrespective setting with regard to improvingtheir child's peer relations (Section I) re-vealed no differences between the twogroups except for one aspect of the proberelated to their child's own behaviors,t(116) = 2.63, p <.01. In particular, moremothers of children in the specialized thanintegrated program were positive becauseof confidence in their own child's abilitiesand behaviors. However, a pattern of reli-able differences was obtained when com-paring the educational/developmentalenvironments of the programs (Section II).Specifically, mothers of children in special-ized programs were much more satisfiedthan mothers of children in integratedprograms with regard to the program beingadapted specifically to the special needs oftheir child,

x1(1, N= 118) = 9.80, p < .01,

and for obtaining special services,xZ

(1, N= 118) = 6.56, p <.01. Moreover, a greaterproportion of children in specialized thanintegrated programs had a completed IEP,iC 2 (1, N= 118) = 11.97, p <.001.

Additional comparisons were ob-tained with regard to the characteristics ofthe program itself (number of childrenwith special needs, hours per week childattends, length of time enrolled, number ofteachers/aides, number of resource staffmembers, frequency with which parentobserves program, location of program).Mothers were also asked to rate the goalsof their child's program (e.g., social inter-action, communication/speech, individual

Peer Relations 471

development, motor skills) and the reasonthat specific program was selected. Asexpected, more children with special needs,t(116) = 9.32, p < .001, and more resourcestaff members, t(114) = 4.45, p < .001, werefound in specialized rather than integratedprograms, and a greater proportion ofspecialized programs were operated by thepublic school, X 2 (1, N= 118) = 18.95, p <.001. No other differences were obtained,although a strong trend was observed formothers in integrated programs to ranksocial interaction more often as a goal oftheir child's program than did mothers inspecialized programs, x Z(1, N= 118) = 4.89,p < .05. Nevertheless, social interactionwas highly ranked by mothers of childrenin both integrated programs (ranked first)and specialized programs (ranked behindcommunication/speech but part of a clus-ter of similar rankings including schoolreadiness and motor skills).

Discussion

In-depth interviews with mothers of pre-school-age children with established dis-abilities and those at-risk addressed therelation between their child's peer rela-tionships and friendships and the type ofeducational placement. Based on mothers'rankings of the goals of their child's pro-gram, it was evident that irrespective ofplacement in an integrated or specializedprogram, the development of their child'sability to interact socially with peers wasgiven high priority. Moreover, it was equallyevident that, based both on absolute per-centages as well as comparisons betweenmatched groups, mothers of children inboth integrated and specialized settingsperceived their respective setting to be ofvalue for the development of their child'speer relationships and friendships and pro-vided similar rationales.

Mothers of children enrolled in inte-grated programs were asked to comparethe contributions of the children with andwithout special needs in the setting to thedevelopment of their own child's peerrelations. Well over half thought their child


benefitted specifically from relationshipswith children without special needs inconnection with watching and learningsocial skills, playing better, and becomingmore social. Probe questions revealed thatthe primary rationale for these beliefs wasthat children without special needs servedas more competent models who are emu-lated by their child. In contrast, mothersdid not believe that children without spe-cial needs contributed to their child's abil-ity to resolve conflicts, share, or cooperatein play. It is interesting that this distinctionbetween the value of children withoutspecial needs in promoting the overallsocial interactions of children with specialneeds and the ability to carry out importantsocial tasks (i.e., improved peer-relatedsocial competence) closely corresponds toresearch findings on the relation betweeneducational setting and children's peerrelationships. Specifically, evidence indi-cates that children with disabilities in inte-grated as compared to specialized programsmanifest increased levels of social interac-tions (e.g., positive leads, using others asresources, observe more) but that theirability to carry out important social taskssuch as resolving conflicts remains essen-tially unaffected (Guralnick, 1990b, in press;Guralnick & Groom, 1987, 1988a).

Similarly, when mothers of childrenenrolled in specialized programs were pre-sented with a hypothetical situation inwhich their child was enrolled in an inte-grated setting, they also recognized certainbenefits of having children without specialneeds available. Specifically, similar tomothers of children in integrated pro-grams, approximately one half felt thattheir child would watch and learn moresocial skills from children without specialneeds, primarily as a consequence of thecompetence model provided by childrenwithout special needs. However, virtuallynone felt that their child's ability to solvesocial tasks would benefit.

Despite the benefits attributed tochildren without special needs, the vastmajority of the mothers of children inintegrated programs saw value in having

other children with special needs availablein their child's program. In part, this wasbased on the stated importance of havingother children with similar skills and abili-ties available for their child. However,greater emphasis was given to the beliefthat the presence of children with specialneeds encourages tolerance and accep-tance of others. A similar response patternwas obtained for mothers of children en-rolled in specialized programs. Clearly, ascommunities design integrated options forfamilies, they must consider these appar-ently widely held beliefs in order to maxi-mize participation.

In a related matter, the possible rejec-tion of children with special needs bychildren without special needs is a perva-sive and troubling issue. On the one hand,parents want their children to "learn aboutthe real world" (e.g., Bailey & Winton,1987; Guralnick, 1994), yet they appear tobe equally concerned that rejection of theirchild is more likely to occur in integratedthan specialized settings (e.g., Guralnick,1994). Indeed, 40% of the mothers ofchildren currently enrolled in integratedprograms were concerned about rejection,primarily as a consequence of negativebehaviors on the part of their own child.Projecting into the hypothetical integratedsituation for children enrolled in special-ized programs, mothers were slightly moreconcerned about rejection (51%). It is in-teresting, however, that the reasons fortheir concern were not related primarily totheir child's behavior but rather to negativepeer attitudes. In contrast, mothers notconcerned about rejection cited their child'spositive behavior as the primary reason.Moreover, children of mothers who wereconcerned about rejection had lower scoreson measures of cognitive and motor devel-opment but were rated higher in behaviorproblems in comparison to children ofmothers not indicating rejection as a con-cern. To the extent to which these negativeattitudes actually exist in integrated set-tings and whether they manifest them-selves in rejection of children withdisabilities is not known.

Evidence is available, however, indi-cating that in integrated settings childrenwith special needs are, as a group, lessaccepted and more often rejected from asociometric perspective and tend to haveprogressively greater difficulties gainingaccess to peer groups (e.g., Guralnick &Groom, 1987). In view of the dispropor-tionately large number of children withspecial needs exhibiting problems in peer-related social competence (Guralnick,1990b), it is not surprising that rejection bypeers would be more common. However,no data are available to suggest that rejec-tion is more common in integrated asopposed to specialized programs. Researchrelated to this issue and to the specificforms rejection might take would helpclarify these concerns and provide a basisfor developing strategies to minimize prob-lems from the perspectives of both parentsand professionals.

Concerns regarding rejection notwith-standing, it is interesting to note that theoverwhelming majority of mothers in bothsettings reported that their child had atleast one best friend in the program andthat they met at the preschool. Of interest,young children with special needs havenearly twice as many nonspecial needsfriends than friends with special needs inintegrated programs, suggesting a reason-able degree of social integration. More-over, nearly half of school-based friendshipsextended to the community, suggestingfurther that friendships formed in the pre-school have important implications forcommunity integration as well. Clearly, theabsence of children without special needsin specialized programs has implicationsfor both preschool and community inter-actions with children without special needs.

As revealed by the mmttched groupanalyses, mothers of children in both inte-grated and specialized programs favorablyevaluated their child's educational/develop-mental program. They believed that the cur-ricular demands were appropriate and hadfew concerns about the program's pace orcomplexity or about their child's ability tofully participate. However, this analysis also

Peer Relations 473

revealed what appears to be a persistentconcern of fanulies whose children are en-rolled in integrated settings. In particular,significantly fewer mothers of children inintegrated programs believed the programwas adapted specifically to their child's spe-cial needs and were more concerned aboutobtaining needed special services. More-over, in comparison to mothers of children inspecialized settings, those in integrated set-tings reported that they did not have acompleted IEP for their child. This concernregarding the availability of special servicesand adaptations to children with specialneeds appears repeatedly both for this sample(Guralnick, 1994) and for other samples(e.g., Bailey & Winton, 1987). Accordingly,in order to establish quality inclusive set-tings, these issues must be addressed at alllevels, including resource allocation,teacher training programs, and the effec-tive and efficient uses of consultation anditinerant service models.

An additional finding of this studywas that the perspectives of mothers on awide range of issues related to peer rela-tionships and friendships did not differ inaccordance with their child's disability sta-tus within the range of the sample investi-gated. The absence of differences amongchildren with different types or severitiesof disabilities is consistent with previousresearch addressing more general aspectsof the benefits and drawbacks ofmainstreaming (see Guralnick, 1994, andStoneman, 1993) and extends these find-ings to maternal perceptions of the value ofpeer relationships and friendships. Al-though children with behavior problemsmay appear to generate a unique set ofresponses (Green & Stoneman, 1989;Guralnick, 1994), in general mothers ofyoung children at-risk and those with es-tablished but diverse disabilities appear toshare similar opinions regarding both gen-eral (broadly based benefits and draw-backs) and specific (peer relationships andfriendships) characteristics of children'seducational settings.

Interview research, even that carriedout with large samples, cannot avoid many


of the problems typically associated withthis type of investigation. For example,there may be a natural tendency for fami-lies to overstate their satisfaction with theirchild's current placement and to minimizethe benefits of other options. Related con-cerns exist with regard to the accuracy ofmothers' responses, although mothers'perceptions are nevertheless important.Moreover, these data relied on familieswilling to participate, and confidentialityrelated to the recruitment p-ocedures didnot permit an assessment of families whoelected not to consent. Despite the fact thatour sample was large and generally re-flected the proportion of children enrolledin integrated and specialized programs inour community, concerns remain aboutthe representativeness of this or any sample.Nevertheless, data obtained from in-depthinterviews described in this study appearto provide a logical and consistent patternof outcomes. Focused replications andconvergent approaches related to parentalconcerns, peer rejection in particular, wouldhelp clarify issues identified in this study.

ReferencesAchenbach, T. M., & Edelbrock, C. (1983).

Manual for the Child Behavior Checklist andRevised Child Behavior Profile. Burlington:University of Vermont, Department of Psy-chiatry.

Bailey, D. B., Jr., & Winton, P. J. (1987).Stability and change in parents' expectationsabout mainstreaming. Topics in Early Child-hood Special Education, 7, 738.

Bates, E. (1975). Peer relations and the acqui-sition of language. In M. Lewis & L. A.Rosenblum (Eds.), The origins of behavior.Vol. 4. Friendship and peer relations (pp.259-292). New York: Wiley.

Blacher, J., & Turnbull, A. P. (1982). Teacherand parent perspectives on selected socialaspects of preschool mainstreaming. TheExceptional Child, 29, 191-199.

Buysse, V., & Bailey, D. B., Jr. 0993). Behav-ioral and developmental outcomes in youngchildren with disabilities in integrated andsegregated settings: A review of comparativestudies. The Journal of Special Education,26, 434-461.

Carrow-Woolfolk, E. (1985)• Manual for theTest for Auditory Comprehension of Lan-guage-Revised. Allen, TX: DLM TeachingResources.

Cohen, J. (1960). A coefficient of agreementfor nominal scales. Educational and Psycbo-logical Measurement, 20, 37-46.

Garvey, C. (1986). Peer relations and thegrowth of communication. In E. C. Mueller &C. R. Cooper (Eds.), Process and outcome inpeer relationships (pp. 329-345). Orlando,FL: Academic Press.

Green, A. L., & Stoneman, Z. (1989). Atti-tudes of mothers and fathers of nonhandi-capped children. Journal of Early Interven-tion, 13, 292-304.

Guralnick, M. J. (1990a). Major accomplish-ments and future directions in early child-hood mainstreaming. Topics in EarlyChildhood Special Education, 10(2), 1-17.

Guralnick, M. J. (1990b). Peer interactionsand the development of handicapped chil-dren's social and communicative compe-tence. In H. Foot, M. Morgan, & R. Shute(Eds.), Children helping children (pp.275-305). Sussex, England: Wiley.

Guralnick, M. J. (1994). Mothers' perceptionsof the benefits and drawbacks of early child-hood mainstreaming. Journal of Early Inter-vention, 18, 168-183.

Gurainick, M. J. (in press). Social compe-tence with peers: Outcome and process inearly childhood special education. In P. L.Safford (Ed.), Yearbook in early childhoodeducation: Early childhood special educa-tion (Vol. 5). New York: Teacher's CollegePress.

Guralnick, M. J., & Groom, J. M. (1987). Thepeer relations of mildly delayed andnonhandicapped preschool children inmainstreamed playgroups. Child Develop-ment, 58, 1556-1572.

Guralnick, M. J., & Groom, J. M. (1988a).Peer interactions in mainstreamed and spe-cialized classrooms: A comparative analysis.Exceptional Children, 54, 415-425.

Guralnick, M. J., & Groom, J. M. (1988b).Friendships of preschool children inmainstreamed playgroups. DevelopmentalPsychology, 24, 595-604.

Guralnick, M. J., & Paul-Brown, D. (1980).Functional and discourse analyses ofnonhandicapped preschool children's speechto handicapped children. American Journalof Mental Deficiency, 84, 444-454.

Guratnick, M. J., & Paul-Brown, D. (1984).

Communicative adjustments during behavior-request episodes among children at differentdevelopmental levels. Child Development, 55,911-919.

Guralnick, M. J., & Paul-Brown, D. (1989).Peer-related communicative competence ofpreschool children: Developmental and adap-tive characteristics. Journal of Speech andHearing Research, 32, 930-943.

Hadley, P. A., & Rice, M. L. (1991). Conversa-tional responsiveness of speech- and lan-guage-impaired preschoolers. Journal ofSpeech and Hearing Research, 34, 1308-1317.

Hartup, W. W. (1983). Peer relations. In E. M.Hetherington (Ed.) (P. H. Mussen, SeriesEd.), Handbook of child psychology: Vol. 4.Socialization, personality, and social devel-opment (pp. 103-196). New York: Wiley.

Hollingshead, A. B. (1975). Fourfactor indexof social status. Unpublished manuscript,Yale University, Department of Sociology,New Haven.

Howes, C. (1988). Peer interaction of youngchildren. Monographs of the Society for Re-search in Child Development, 53(1, SerialNo. 217).

Lamorey, S., & Bricker, D. D. (1993). Inte-grated programs: Effects on young childrenand their parents. In C. A. Peck, S. L. Odom,& D. D. Bricker (Eds.), Integrating youngchildren with disabilities into communityprograms (pp. 249-270). Baltimore: Brookes.

Miller, L.J., Strain, P. S., Boyd, K., McKinley,J., Hunsicker, S., & Wu, A. (in press).Preschool mainstreaming: Outcomes for chil-dren with disabilities and normally develop-ing children. Journal of Early Intervention.

Odom, S. L., & McEvoy, M. A. (1988). Integra-tion of young children with handicaps andnormally developing children. In S. L. Odom& M. B. Karnes (Eds.), Early intervention forinfarcts and children with handicaps (pp.241-267). Baltimore: Brookes.

Quirk, M., Sexton, M., Ciottone, R., Minami,H., & Wapner, S. (1984). Values held bymothers for handicapped and nonhandi-capped preschoolers. Merrill-Palmer Quar-terly, 30, 403-418.

Reichart, D. C., Lynch, E. C., Anderson, B.C., Svobodny, L. A., DiCola, J. M., & Mer-cury, M. G. (1989). Parental perspectives onintegrated preschool opportunities for chil-dren with handicaps and children withouthandicaps. Journal of Early Intervention, 13,6-13.

Rubin, K. H., & Lollis, S. P. (1988). Origins

Peer Relations 475

and consequences of social withdrawal. In J.Belsky & T. Nezworski (Eds.), Clinical im-plications of attachment (pp. 219-252).Hillsdale, NJ: Erlbaum.

Sparrow, S. S., Balla, D. A., & Cicchetti, D. V.(1984). Vineland Adaptive Behavior Scales.Circle Pines, MN: American Guidance Ser-vice.

Stoneman, Z. (1993). The effects of attitudeon preschool integration. In C. A. Peck, S. L.Odom, & D. D. Bricker (Eds.), Integratingyoung children with disabilities into com-munity programs (pp. 223-248). Baltimore:Brookes.

Turnbull, A. P., & Turnbull, H. R. (1990).Families, professionals, and exceptionality.A special partnership (2nd ed.). Columbus,OH: Merrill.

Turnbull, A. P., Winton, P. J., Blacher, J., &Salkind, N. (1982). Mainstreaming in thekindergarten classroom: Perspectives of par-ents of handicapped and nonhandicappedchildren. Journal of the Division of EarlyChildhood, 6, 14-20.

Vandell, D. L., & George, L. B. (1981). Socialinteraction in hearing and deaf preschoolers:Successes and failures in initiations. ChildDevelopment, 52, 627-635.

Wechsler, D. (1989). Wechsler Preschool andPrimary Scale of Intelligence-Revised. SanAntonio: Psychological Corp.

Zimmerman, I. L., Steiner, V. G., & Pond, R.E. (1979). Preschool language scale (rev.ed.). Columbus, OH: Charles E. Merrill.

Received 1110194, first decision 4116194, accepted512 7194.

This research was supported by Grant No.H024K90002 from the Department of Education,Office of Special Education Programs. Apprecia-tion is expressed to J. Nielsen and L. Feshbach forassistance in various phases of this project and tothe families who participated in this study. Re-quests for reprints should be sent to Michael J.Guralnick, Child Development and Mental Retar-dation Center, University of Washington, WJ-10,Seattle, WA 98195.


american journal on mental retardation, 1995, vol. 99, no. 5, … › chdd › guralnick › pdfs...

Documents