UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The WISC-III comprehension and picture arrangement subtests as measures of social functioning : fact… Arvanitakis, Maria Alexia 2001

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2001-0147.pdf [ 3.3MB ]
Metadata
JSON: 831-1.0089849.json
JSON-LD: 831-1.0089849-ld.json
RDF/XML (Pretty): 831-1.0089849-rdf.xml
RDF/JSON: 831-1.0089849-rdf.json
Turtle: 831-1.0089849-turtle.txt
N-Triples: 831-1.0089849-rdf-ntriples.txt
Original Record: 831-1.0089849-source.json
Full Text
831-1.0089849-fulltext.txt
Citation
831-1.0089849.ris

Full Text

THE WISC-III COMPREHENSION AN D PICTURE ARRANGEMENT SUBTESTS AS MEASURES OF SOCIAL FUNCTIONING: FACT OR FICTION? by MARIA ALEXIA ARVANITAKIS B.A., McGill University, 1995 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENT FOR T H E D E G R E E O F MASTER O F A R T S in T H E FACULTY O F G R A D U A T E STUDIES (Department of Educational and Counseling Psychology and Special Education School Psychology Program) We accept this thesis as conforming to the required standard THE UNIVERSITY O F BRITISH COLUMBIA August 15, 2000. © Maria Alekia Arvanitakis, 2000 UBC Special Collections - Thesis Authorisation Form Page 1 of 1 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the Univ e r s i t y of B r i t i s h Columbia, I agree that the Li b r a r y s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis for sc h o l a r l y purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or p u b l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date http://www.library.ubc.ca/spcoll/thesauth.html 4/2/01 ABSTRACT The traditional assumption that the Wechsler Intelligence scales Picture Arrangement (PA) and Comprehension (COMP) subtests are interpretable as measures of social competence was tested. Although the assumption has existed for over half a century, there is a lack of evidence to support this contention. Performance on the PA and COMP subtests of the WISC-III was correlated with various indices of social functioning, using multiple sources of information (i.e., participant self-report, teacher report and peer nominations within a normal school-aged population) in a sample of 74, eight to twelve year old children. After general intelligence was partialled out, performance on the COMP subtest related to children's perception of social self-efficacy, and performance on the PA subtest was related to assertive problem solving strategies. No other social measure correlated significantly with performance on the PA and COMP subtests. Results of regression analyses reiterated the correlational analyses suggesting that although there was some modest predictive power for the PA subtest to predict assertive problem solving strategies and for performance on the COMP subtest to predict a child's perception of social self-efficacy, performance on the PA and COMP subtests do not predict the majority of the social skills that were included in the present investigation. Table of Contents ABSTRACT ._ ii Table of Contents iii List of Tables v Acknowledgments vi CHAPTER I 1 Introduction 1 CHAPTER II 4 Literature Review ' 4 Historical Overview 4 Measures of Social Intelligence 7 Social Sensitivity 8 Social Participation 9 Social Maturity 9 Personality 10 Psychopathology 11 Social Skills 15 Problem Solving 16 Social Status 16 Summary 17 Statement of the Purpose 19 CHAPTER III 21 METHOD 21 Participants 21 iii Measures - 22 Picture Arrangement and Comprehension subtest of the WiSC-lll 22 The Kaufman Brief Intelligence Test 23 Social Skills Rating System 24 Sociometric Status 25 The Children's Conflict Resolution Measure 26 Loneliness and Social Dissatisfaction Questionnaire 27 Relational Provisions Loneliness Questionnaire 28 The Children's Social Self-Efficacy Scale for Peer Interactions 29 Procedure 31 CHAPTER IV 32 RESULTS 32 Data Analyses Overview 32 CHAPTER V 48 DISCUSSION 48 REFERENCES 55 APPENDIX A: Parental Consent Form 62 APPENDIX B: The Children's Conflict Resolution Measure 64 APPENDIX C: Loneliness and Social Dissatisfaction Questionnaire 74 APPENDIX D: Relational Provisions Loneliness Questionnaire 76 APPENDIX E: Social Self-Efficacy Questionnaire 77 iv List of Tables Table 1. Descriptive Information 33 Table 2. Intercorrelations of Social Measures 35 Table 3. Intercorrelations among Intelligence Measures 39 Table 4. List of Measures 40 Table 5. Correlations between Overall Intelligence, PA, and COMP with Various Indices of Social Functioning 42 Table 6. Hierarchical Regression Analyses: Comprehension 45 Table 7. Hierachical Regression Analyses: Picture Arrangement 46 Acknowledgments I would like to express sincere appreciation to the members of my Master's committee Drs. Shelley Hymel, William McKee and Kimberly Schonert-Reichl for their assistance, encouragement, and wise counsel during the preparation of this thesis and throughout the master's program. In particular I am very grateful to Dr. Hymel as chair of the committee for her guidance and erudite review of various drafts of this document. The friendship, participation, and constant encouragement of my colleagues and friends; Kelly Lemon, Tanya McCreith, and Faye Kan/at was key to completing this project, and I thank them for their assistance. I would also like to thank my classmates Stacey Bablitz and Louise Mercer for their continuous support during the more challenging periods of this project and throughout my studies at UBC. I owe a special thank you to my dear friend Louise LeBlanc, who while undergoing similar challenges herself was a constant source of strength. I am very grateful to the administration, teachers and, most importantly, the students at Our Lady of Sorrow Elementary School who offered a glimpse into their lives and relationships. Ultimately, I thank my parents and family for their constant love and support, for helping to make this endeavor possible, and for helping me believe in myself and to always shoot for the stars. vi CHAPTER I Introduction The Wechsler Intelligence Scale for Children (WISC, WISC-R, WISC-lll) and its adult counterpart, the Wechsler Adult Intelligence Scales (WAIS, WAIS-R, WAIS-III) are the most widely used measures of intellectual ability, with excellent psychometric qualities (Wechsler, 1991). Although the thirteen subtests that comprise the WISC-lll were originally designed as components of an overall intelligence scale, individual subtests of the Wechsler Scales traditionally have been used to describe specific individual abilities (e.g., Blatt & Allison, 1981; Rapaport, Gill & Schafer, 1968; Saftler 1992). Of particular interest to the present thesis are the Comprehension (COMP) and Picture Arrangement (PA) subtests. Because of their socially relevant content, these two subtests are often assumed to measure "social intelligence" (Lipsitz, Dworkin, & Erlenmeyer-Kimling 1993; Sipps, Berry & Lynch, 1987 as cited in Campbell & McCord, 1996) or intelligence related to social competence (Zimmerman & Woo-Sam, 1973). Blatt and Allison (1981, p. 201) argued: "The PA subtest also reflects, as does the Comprehension subtest, the response to stimuli that are concerned with social interactions. The two subtests allow comparison of well-learned social conventionalities (Comprehension), with the capacity to anticipate and plan in social context (PA)." Allison, Blatt and Zimet (1968) further described the PA subtest as measuring the child's ability "to grasp the essential message of a social interaction" (p. 29). Although it is widely assumed that the PA and COMP subtests are related to social intelligence, there exists little empirical evidence to support the contention (Campbell & McCord, 1996), with suggestions that supportive evidence rests largely on the "face validity" of these subtests (Lipsitz, Dworking & Erlenmeyer-Kimling, 1993). Despite the scant evidence supporting these assumptions, clinicians continue to 1 interpret PA and COMP subtest scores as indicators of social competence. Indeed, practitioners in the field are taught that the PA subtest measures a child's ability to "anticipate the consequences of initial acts or situations, as well as the ability to interpret social situations" (Sattler, 1992, p. 1095). They are also taught that the COMP subtest measures: "the child's knowledge of conventional standards of behaviour, extensiveness of cultural opportunities, and level of development of conscience or moral sense. Success suggests that the child has social judgment, common sense, and a grasp of social conventionality" (Sattler 1992, p.1089). The assumptions that prevail surrounding these two subtests as measures of social ability are the focus of the present investigation. As the following review will demonstrate, past research regarding the links between social functioning and performance on the COMP and PA subtests have been limited and results have been mixed. Although some studies have demonstrated significant associations between COMP and PA performance and various indices of social competence (e.g., Kippner, 1964, Schill, 1966, Schill, Kahn & Meuleman, 1968, Searight, Dunn, Grisso, Margolis, & Gibbon, 1989, Sipps et al., 1987), others have not (e.g., Dickstein & McEvitt, 1971, Ramos & Die, 1986; Simon & Evans, 1980; Nobo & Evans, 1986; Lipsitz et al., 1993). Moreover, as discussed in the following literature review, research in this area is fraught with methodological difficulties, making firm conclusions regarding the utility of these subtests as indices of social functioning difficult, if not impossible. In an attempt to extend the scientific knowledge base concerning the relationship between performance on the COMP and PA subtests and social functioning, the present study addresses some of the limitations of previous literature in four ways. First, links between social functioning, COMP and PA performance are examined using the most 2 recent edition of the child Wechsler scale (WISC-lll), which is used in current student evaluations. Given suggestions that the "social" content of these two subtests has diminished over the revisions of the WISC (Kaufman, 1995), evaluation of hypothesized links with social functioning must be made in terms of the most recent revision. Second, a variety of indices of social functioning derived from recent research on children's peer relations are considered (e.g., behavioral, social cognitive and affective measures), using multiple sources of information (i.e., participant self-report, teacher report and peer nominations). This allows for a broad and more thorough evaluation of social competence and social intelligence from a number of perspectives. Third, an alternative measure of intelligence is also included as a co-variate in order to statistically eliminate overall cognitive functioning as an explanatory variable. When a separate evaluation of overall intellectual ability is not considered, it is difficult to determine whether observed associations reflect poor performance on the PA and COMP per se or more limited overall intellectual functioning. Finally, although most studies in this area have considered adult samples and/or clinical samples, the present study examines links between social functioning and performance on the PA and COMP subtests within a normal, school-aged population, enhancing the generalizability of results. 3 CHAPTER II Literature Review Historical Overview In 1939, David Wechsler realized the limitations of the Stanford-Binet and designed an instrument that would measure intellect by tapping a diverse sample of capabilities. In a major departure from earlier methods of assessment, Wechsler devised the format that we currently know as the Wechsler scales, consisting of a collection of subtests. Wechsler believed in the unitary nature of intelligence (Wechsler, 1958), but subscribed to the idea that intellect can be best measured through interdependent, qualitatively different abilities (Wechsler, 1969). Although the subtests of the Weschler scales were developed to tap discrete talents that may represent facets of intelligence, the instrument was designed to be interpreted as a whole, not as measuring separate aspects of intellect (Wechsler, 1958 as cited in Ramos & Die, 1985). Guilford (1967) has found that the capabilities measured on various subtests of the WISC are interrelated, but despite statistical analysis, there is no clear understanding of how particular subtest scores relate to behaviour. Clinicians have a strong tendency to use the Wechsler scales in a diagnostic manner, although research and statistical analysis do not justify the practice (Anastasi, 1982, as cited in Ramos & Die, 1985). Historically, the PA and COMP subtests of the Wechsler scales have been characterized repeatedly as being nonverbal and verbal measures of social competence, respectively (Blatt & Allison, 1968; Rapaport, 1946; Wechsler, 1944). As a result, clinicians interpret performance on the PA and COMP subtests as measures of social ability. Is such an interpretation warranted? The COMP subtest of the WISC-III scale is "a series of orally presented questions that require the child's solving of everyday problems or understanding of 4 social rules and concepts" (Wechsler, 1991, p. 6). The questions require the child to explain situations, actions, or activities that relate to events familiar to most children. The questions cover the content areas of knowledge of one's body, interpersonal relations, and social mores (Sattler, 1992). According to Sattler, success on the COMP subtest depends on the child's possession of practical information plus an ability to draw on previous experiences. Responses are thought to reflect the child's knowledge of conventional standards of behaviour, extensiveness of cultural opportunities, and level of development of conscience or moral sense. The PA subtest includes "a set of colorful pictures, presented in a mixed up order, which the child rearranges into a logical story sequence" (Wechsler, 1991). This subtest is thought to measure the child's ability to anticipate the consequences of initial acts or situations, as well as the ability to interpret social situations (Sattler, 1992). It is widely assumed that the PA subtest measures a person's ability to evaluate and comprehend a situation using pictorial cues that have been visually organized (Blatt, 1965; Glasser & Zimmerman, 1967; Kahn & Meuhleman, 1968a, 1968b; Rapaport et al., 1979; Schill, 1966; Woo-Sam & Zimmerman, 1972). When performance is poor, it is suspected that the client may have an impaired capacity to reflect, anticipate, and plan a course of action (Blatt & Allison, 1986), and to understand antecedent and consequent events (Blatt, 1965). Because the PA items typically involve human or practical situations, it is also thought that a person with a high PA score is adept at sizing up and comprehending social situations. For example, Shafer (1948) attributed the high PA score often found for adults diagnostically classified as narcissistic to their "characteristically facile social anticipation" (p.54). He also noted that adults with psychopathic character disorders frequently have a very high PA score, especially the "shrewd schemer" who can quickly evaluate a situation and manipulate it for his/her own 5 end. Although PA and COMP have been viewed as predictive of social intelligence, there has been considerable difficulty defining social intelligence, social functioning and/or social competence. The concept of social intelligence is vague, and Ramos and Die (1985) argued that "social intelligence itself is a concept that may not exist" (p. 254). E.L. Thorndike (1920) described social intelligence as the ability to understand others and "act wisely in human relations". He proposed that social intelligence was itself an aspect of a person's IQ. Factor analysis of an early tool designed to measure such skill revealed that the instrument simply quantified abstract reasoning ability. Wechsler (1958, as cited in Ramos & Die, 1985) believed that general intelligence could be applied to all situations. Thus, "social" intelligence might best be conceptualized as one aspect of general intelligence. More recent authors within the intelligence literature have also attempted to define social intelligence but have not succeeded in clarifying the concept. For example, in his theory of multiple intelligences, Gardener (1985) specifically distinguished interpersonal or social functioning as a separate intelligence. He defined social intelligence as an understanding of others' moods, feelings, behaviors, motivations and intentions and the ability to act on that understanding. With regard to the school-aged child, the focus of the present thesis, Gardner (1983) suggested that interpersonal intelligence included understanding of the different roles others play, and an understanding of reciprocity, perspective taking, and fairness. According to Gardner, social intelligence at this age is particularly concerned with peer relations and friendships, and with understanding one's place in peer groups, cliques and dominance hierarchies. Difficulties in developing interpersonal relationships with others lead the child to feelings of failure and aloneness. 6 In his book Emotional Intelligence Daniel Goleman (1995), shows that data from a wide range of disciplines and various parts of the life span converge to suggest that one's social and affective skills are at least as important to one's future life success as what are traditionally seen as "intellectual" skills. Robert Sternberg (1995) concluded that social intelligence is both distinct from academic abilities and a key part of what makes people do well in the practicalities of life. In their review article, Rubin, Bukowski and Parker (1997) provide an historical review of the various definitions of social competence. These include "an organism's capacity to interact effectively with its environment." (White, 1959, p.297); "the effectiveness or adequacy with which an individual is capable of responding to various problematic situations which confront him." (Goldfried & D'Zurilla, 1969, p. 161); "an individual's everyday effectiveness in dealing with his environment." (Zigler, 1973); "a judgement by another that an individual has behaved effectively." (McFall, 1982, p.1); "attainment of relevant social goals in specified social contexts, using appropriate means and resulting in positive developmental outcomes." (Ford, 1982, p.324); the ability 'lo make use of environmental and personal resources to achieve a good developmental outcome" (Waters & Sroufe, 1983; p.81); and 'Ihe ability to engage effectively in complex interpersonal interaction and to use and understand people effectively."(Oppenheimer, 1989, p.45). Rubin and Rose-Krasnor (1992) have defined social competence as the "ability to achieve personal goals in social interaction while simultaneously maintaining positive relationships with others over time and across situations." Measures of Social Intelligence Given the extensive yet vague definitions of the construct of social intelligence, it is not surprising that studies examining the links between social functioning and 7 performance on the PA and COMP subtests have considered a wide array of indices of social functioning. In reviewing the literature, it became evident that there is little consensus among researchers regarding definitions of social functioning or whether the PA and COMP subtests measure "social intelligence" (Lipsitz et al., 1993; Sipps et al., 1987) or intelligence related to "social competency" (Zimmerman & Woo-Sam, 1973). For the present review, seven indices of social competence that have been related to PA and COMP performance are distinguished: social sensitivity, social participation, social maturity, personality, psychopathology, problem solving and social status. Social Sensitivity. Early studies, conducted with adult populations, provided some support for the notion that performance on the PA and COMP subtests is associated with social sensitivity. Schill et al. (1968) compared the performance of high and low PA participants on Greenspoon's (1955) classic "plural nouns" verbal conditioning task, reflecting the degree to which a person's behavior (emitting plural nouns) could be modified by subtle social cues (nodding and saying "mm-hmm"). Results indicated that high scorers on PA were more likely to modify their behaviour in response to the reinforcement than were low PA scorers, a response that was interpreted as greater sensitivity to subtle social cues. One major criticism of Schill et al.'s study is that the researchers made no attempt to control for overall intelligence (g) (Ramos & Die, 1986, Lipsitz etal., 1993). It remains unclear whether the participant's performance was a result of "social sensitivity" per se or general intelligence. More recently, Campbell and McCord (1996) found that PA scores predicted undergraduate student's ability to interpret the nonverbal behaviour of others, but not significantly better than Full Scale IQ scores. It is also not clear that social sensitivity as defined in this study provides an adequate index of social intelligence, social functioning or social competence. 8 Social Participation. One study conducted with adults examined links between performance on PA and social participation. Schill, Kahn, and Muehleman (1967) showed that college sophomores with high PA scores tended to report greater participation in extracurricular activities in high school and college than did the low PA participants. As with previous studies, no effort was made to control for overall intelligence, making it difficult to determine whether greater social participation is attributable to more sophisticated picture arrangement skills per se or greater overall intelligence. Also, one may question whether greater social participation reflects greater social competence or social intelligence. Social Maturity. Two studies have examined performance on the PA and COMP subtests of the WISC and social maturity using different indicators of social maturity. Krippner (1964) compared scores on the PA and COMP subtests to scores on the Vineland ocial Maturity Scale in a sample of 40 eight- to twelve-year-old boys. Krippner found that high COMP scores, but not PA scores, were associated with higher social age on the Vineland Social Maturity Scale. The magnitude of the correlation between social maturity and COMP scores was modest at best (r (51) = .27, p< .05). Similarly, Brannigan (1975) investigated the relationship between the PA and COMP subtests and social maturity in children as reflected by teacher ratings of emotional and interpersonal maturity. He found that both boys and girls who scored high on the COMP subtest of the WISC were also rated as more interpersonally and emotionally mature by their teachers (rs range from .22 to.39). Brannigan found no significant relationship between the PA subtest and teacher-rated social maturity. Both studies, then, show that understanding social norms and mores (COMP subtest) is moderately related to greater social maturity in children, but the ability to sequence events (PA subtest) is not. As Krippner noted, "there are no indications from this study that WISC Picture Arrangement can be utilized 9 to measure, even roughly, a child's social competence" (p. 367). Personality. With adult samples, several researchers have investigated the relationship between the PA and COMP subtests of the WAIS and various personality traits or characteristics. The results have been mixed. Schill (1966) demonstrated that adults whose MMPI scores showed them to be socially introverted scored significantly lower on the PA subtest in comparison to adults who were assessed as social extroverts. However, Johnson (1969) failed to replicate this relation using a psychiatric population. In a sample of older participants, Turner and Horn (1976) were also unable to replicate this finding, concluding that, "sociability or extroversion does not seem to be related to PA performance" (p.592). The relationship between performance on the PA subtest of the WAIS and social introversion may be more complicated, however. In a sample of psychiatric patients, Terry and Berg (1984) found that individuals who were socially introverted and high in "psychopathic deviance" scored higher on PA and so did individuals who were socially extroverted but low in "psychopathic deviance". Terry and Berg suggest that high psychopathic extroverts are more "socially intelligent" in terms of their ability to evaluate social situations in order to exploit them for selfish ends. In contrast, low psychopathic introverts display less social intelligence consistent with their tendency "to be reclusive, secretive, and socially retarded" (p. 971). More recently, Sipps, Berry and Lynch (1987) predicted WAIS-R PA and COMP scores from 18 subscales of the California Psychological Inventory (CPI) in a sample of adults. Results of a multiple regression analysis indicated that 36% of the variance in COMP scores could be predicted by three of the 18 CPI subscales: capacity for status (ability to be resourceful and versatile in new social situations), flexibility (in thinking and social behavior) and communality (honesty, conscientiousness, common sense, good 10 judgment). Similarly, 28% of the variance of PA scores could be predicted by three of the 18 subscales: capacity for status, flexibility and femininity. Importantly, Sipps et al. further demonstrated that some of these relationships remained significant even after partialling out performance on the Vocabulary subtest of the WAIS; considered an index of general intelligence. However, many of the other socially relevant indices on the CPI (e.g., social presence, responsibility, good impression, sociability and dominance) did not contribute significantly to predicting PA and COMP performance. Thus, PA and COMP scores were related to some, but certainly not all aspects of "social intelligence" reflected in CPI scores. Most recently, Nobo and Evans (1996) found no significant relationship between PA or COMP scores on the WAIS-R and various indices of social functioning in adults (university undergraduates), including social desirability, social introversion, social adroitness (skill at social manipulation and persuasion), conformity, or social confidence. On the basis of their results, Nobo and Evans suggest "caution in drawing personality inferences from these WAIS-R subtests" (p. 90). Taken together, these studies, all conducted with adult samples, provide little clear evidence of a link between social aspects of personality and either PA or COMP subtest scores. It is also noteworthy that the weak to nonsignificant relations have been evident for both WAIS and WAIS-R versions of the PA and COMP subtests, as the content of the items on these subtests have become less socially focused in more recent versions of the test (Kaufman, 1995). To my knowledge, no studies relating PA and COMP performance to personality indicators have been conducted with children. Psvchopathology. Several researchers (Beebe & McBurnett, 2000; Lypsitz, Dwoekin & Erlenmeyer-Kimling, 1993; Toomey, Wallace, Corrigan, Schuldberg, & Green, 1997; Tymchuck, Simmons, & Neafsey (1977); and Reiff & Gerber(1990) have 11 examined the relationship between various types of psychopathology and performance on the Wechsler scales. Autism and Schizophrenia and most recently Attention Deficit Hyperactivity Disorder have been the three major disorders researched, as individuals with these disorders are thought to exhibit significant social deficits. Tymchuck et al. (1977) found that children and adolescents (age not specified) with high functioning Autism scored significantly lower on the COMP, but not the PA subtest, relative to a matched group of nonautistic children. On the basis of these results, and despite the fact that they did not include a measure of social problem solving, Tymchuck et al. suggested that these individuals have difficulty interpreting social nuances. These results should be interpreted with caution; it may be that these results reflect poor verbal expression; a characteristic of the Autism syndrome, rather than the notion that poor performance on the COMP subtest is the result of difficulty with the interpretation of social nuances. Similarly, Toomey et al. (1997) found that Schizophrenic adults, who are typically characterized by extreme social skills deficits, performed significantly worse on both PA and COMP subtests than did controls. However, the results also suggested that there is no correlation between social perception on the Profile of Nonverbal Sensitivity (PONS) and performance on the PA and COMP subtests within the normal control group. Thus, a pattern of poor performance on the PA and COMP was only supported by research with particular special populations and not control populations. Several researchers (Lyons & Levine, 1978; Toomey, Wallace, Corrigan, Schuldberg, & Green, 1997) have examined the relationship between various types of psychopathology and performance on the Wechsler scales. Autism and Schizophrenia have been the two major disorders researched, as individuals with these two disorders are thought to exhibit significant social deficits. Tymchuck et al. found that high functioning Autistic children and 12 adolescents (age not specified) scored significantly lower on the COMP, but not PA subtest, and significantly higher on the Block Design subtest relative to a matched group of nonautistic children. On the basis of these results, Tymchuck et al. suggest that these individuals have difficulty interpreting social nuances. In a small N study, Toomey et al. (1997) found that Schizophrenic adults who are typically characterized by extreme social skills deficits, performed significantly more poorly on both PA and COMP subtests than did controls. Several researchers (Lyons & Levine, 1978; Toomey, Wallace, Corrigan, Schuldberg, & Green, 1997) have examined the relationship between various types of psychopathology and performance on the Wechsler scales. Autism and Schizophrenia have been the two major disorders researched, as individuals with these two disorders are thought to exhibit significant social deficits. Tymchuck et al. found that high functioning Autistic children and adolescents (age not specified) scored significantly lower on the COMP, but not PA subtest, and significantly higher on the Block Design subtest relative to a matched group of nonautistic children. On the basis of these results, Tymchuck et al. suggest that these individuals have difficulty interpreting social nuances. In a small N study, Toomey et al. (1997) found that Schizophrenic adults who are typically characterized by extreme social skills deficits, performed significantly more poorly on both PA and COMP subtests than did controls. Several researchers (Lyons & Levine, 1978; Toomey, Wallace, Corrigan, Schuldberg, & Green, 1997) have examined the relationship between various types of psychopathology and performance on the Wechsler scales. Autism and Schizophrenia have been the two major disorders researched, as individuals with these two disorders are thought to exhibit significant social deficits. Tymchuck et al. found that high functioning Autistic children and adolescents (age not specified) scored significantly 13 lower on the COMP, but not PA subtest, and significantly higher on the Block Design subtest relative to a matched group of nonautistic children. On the basis of these results, Tymchuck et al. suggest that these individuals have difficulty interpreting social nuances. In a small N study, Toomey et al. (1997) found that Schizophrenic adults who are typically characterized by extreme social skills deficits, performed significantly more poorly on both PA and COMP subtests than did controls. In a study of 32 learning disabled children, Reiff and Gerber (1990) predicted scores on a measure of nonverbal social perception (the Profile of Nonverbal Sensitivity, PONS, Rosenthal et al., 1979) from age and performance on PA, COMP and the Digit Span subtests of the WISC-R. Significant correlations were observed between both the PA and COMP subtest scores and PONS scores (r = .51 and .42, respectively). Results of regression analyses indicated that PA was the strongest predictor of nonverbal social perception, accounting for 25% of the variance in PONS scores, but inclusion of the other three predictors added significantly to the prediction for a total of 34% of the variance accounted for. Given that performance on the Digit Span subtest, assessing auditory memory, was also a significant predictor of PONS performance, it is not clear that nonverbal social perception was predicted by PA and COMP performance per se, general intelligence, or by memory. Lipsitz, Dworkin and Erlenmeyer-Kimling (1993) found little association between two measures of social adjustment and PA and COMP scores in children and adolescents at risk for developing psychopathology. Most recently Beebe and McBurnett (2000) compared the PA and COMP subtest scores with mother- and teacher-reported social functioning in 142 children with Attention Deficit Hyperactivity Disorder (ADHD) and 30 control children, after general intelligence had been accounted for. This disorder was chosen because it has been found to be associated with social dysfunction in 14 children (e.g., Barkley, 1998 cited in Beebe & McBurnett, 2000). The Beebe and McBurnett study is particularly noteworthy because it is one of the few studies that address some of the limitations of past studies. They used the most recent version of the Wechsler scale, accounted for overall "g", and used data collected from multiple sources, making the results more generalizable than previous studies. Their results indicated that the COMP subtest related to some aspects of social functioning whereas the PA subtest was unrelated to social functioning once general intelligence was controlled. Thus, results of studies with special populations have been mixed. Schizophrenics did poorly on both PA and COMP, children with Attention Deficit Hyperactivity Disorder did poorly on COMP, and children with Autism had poor performance on COMP and not PA. Although these results may have some implications for the above-mentioned populations, they are not necessarily generalizable to a normal population of children. Social Skills. Campbell and McCord (1999) examined the relationship between performance on the Personality Inventory for Children (PIC) FACTOR II: Social Incompetence and Social Skills Scale scores and the PA and COMP subtests of the WAIS-R and WISC-R among 200 individuals from 6-20 years of age. Correlations between the PA and COMP scaled scores and the PIC measures of social deficiencies were not significant for either the WISC-R or the WAIS -R. They did find, however, that ipsative scatter analyses revealed that participants with relatively stronger COMP performance were rated as experiencing fewer social problems than other participants. Campbell and McCord (1999) caution the use of these subtests as measure of social competence. Relatively strong COMP scores only corresponded to average T scores on each PIC scale (i.e., T= 50) and did not represent "strong" social competence 15 ratings by mothers. Furthermore, although weak COMP scorers were rated as having more social difficulties than strong COMP scorers, relatively poor COMP performance did not translate into clinically significant social incompetence nor peer relationship problems as would be expected. Unlike the present study, Campbell and McCord's study included single-informant ratings (i.e., maternal report) to assess social competence in children and adolescents and this may not accurately reflect participants' social incompetencies and problems with peer relations across settings. Campbell and McCord failed to use the most recent version of the Wechsler Scale, and their sample population consisted of children who had been referred for emotional, behavioral and academic problems, thus further limiting the generalizability of their results. Problem Solving. Several authors (Ault, 1973; Blatt, 1965; Meichenbaum & Goodman, 1971) suggest that problem solving ability contributes to performance on the PA task (Ramos & Die, 1986). However, to my knowledge, no studies to date have directly examined the links between social problem solving abilities and performance on either the PA or COMP subtests. Social Status. Only one study to date has examined the association between performance on the WISC-R PA subtest and social status or acceptance/rejection within the peer group. In a study of 217 third through fifth grade students, Simon and Evans (1980) compared students who were accepted versus rejected by their classroom peers on a sociometric measure in terms of their performance on the PA subtest. Specifically, they hypothesized that "assuming that Picture Arrangement measures social reasoning and that higher levels of social reasoning lead to greater peer acceptance, the mean score of the accepted group was predicted to exceed that of the rejected group." (p. 558). Contrary to predictions, they found no significant difference between the PA scores of accepted versus rejected children. 16 Summary. A review of the extant literature indicates some, albeit minimal, evidence that performance on the PA subtest is related to social sensitivity and social participation in adults. Social maturity has been associated with higher performance on the COMP but not PA subtests in children. Studies of adults also failed to demonstrate clear or consistent links between PA and COMP performance and various social aspects of personality. Recent research using social indices of personality indicated fewer social problems related to higher COMP performance. However, studies linking performance on the COMP and PA subtests with various forms of psychopathology have yielded mixed results. Children and adolescents with Autism performed more poorly on the COMP but not PA subtests, and schizophrenic adults perform more poorly on both COMP and PA relative to controls. Among children with learning disabilities, performance on the PA and COMP subtests is associated with nonverbal social perception, but so is performance on the Digit Span subtest, assessing auditory memory. Despite arguments that problem solving is a component of PA performance, no studies to date have investigated this relationship with adults or children. Finally, performance on the PA subtest does not appear to vary as a function of children's social status, despite hypotheses to the contrary. Taken together, the present review of literature indicates little support for the argument that performance on the PA and COMP subtests actually reflects social intelligence, social functioning or social competence. There are several reasons why the relationship between the PA and COMP subtests and social abilities may have not been demonstrated empirically. First, as Lipsitz et al. (1993) and Campbell and McCord (1996) suggest, past studies, with the exception of Beebe et al. (2000), have failed to account for the relationship between general intelligence, "g", and social criterion measures (e.g., Kippner, 1964), making it 17 difficult to determine whether poor performance on the PA and COMP subtests is associated with general intelligence or poor social functioning per se. Second, only Beebe et al. (2000) have examined the association between social abilities and PA/COMP performance using the most recent version of the Wechsler intelligence scale and since the content of the PA and COMP subscales have changed somewhat in the most recent version of the WISC-lll, the predictive validity of the current scale is not well known. Third, many of the previous studies have examined single and sometimes obscure indices of social functioning, rather than evaluating associations between performance on the PA and COMP subtests and a broad range of social indices derived from current literature. For the purpose of this study, social functioning is defined as the ability to achieve personal goals in social interaction while simultaneously maintaining positive relationships with others over time and across situations (Rubin & Rose-Krasnor, 1992, as cited in Rubin, Bukowski & Parker, 1997). Accordingly, social functioning was measured from a variety of sources (self-report, teacher and peer ratings), tapping different aspects of social competence and social skills. Social skills are defined as acceptable, learned behaviors that enable a person to interact effectively with others and to avoid socially unacceptable responses (Gresham & Elliott, 1984,1990). Sharing, helping and initiating relationships requesting help, giving complements, and saying "please" and "thank you" are examples of social skills. Developing such skills to enable successful relationships is one of the most important accomplishments of childhood. Accordingly, social functioning as assessed in the present study is reflected in a) peer evaluations of social acceptance/ rejection, b) teacher and self evaluations of actual social behavior, c) social problem solving abilities, and d) self reports of social self-efficacy as well as loneliness and social dissatisfaction, 18 tapping behavioral, cognitive and affective aspects of social competence. Statement of the Purpose The present thesis is designed to investigate whether or not the assumed relation between "social intelligence" (social functioning and/or social competence) and performance on the PA and COMP subtests of the Wechsler Intelligence Scale for Children (Third Edition) can be empirically supported in a sample of normal elementary-aged children. To this end, children in grades four through six completed both the PA and COMP subtests of the WISC-lll in a single individual testing session with an examiner trained in the administration of level C tests. Participants also completed several different measures of social functioning, including peer assessments of attraction (sociometric ratings), self-report measures of social problem solving, social self-efficacy and loneliness and social dissatisfaction. Teachers were also asked to evaluate participants in terms of social behavior. The measures used to assess social functioning have been taken from the most recent literature on children's interpersonal relations (see Rubin, Bukowski, & Parker, 1998), have established psychometric quality, and cover a broad range of social competencies. Thus, in contrast to previous studies, numerous aspects of social functioning are evaluated in relation to PA and COMP performance. In addition, during individual testing, children completed a standardized test of overall intelligence. Inclusion of this measure provides an independent assessment of intelligence, which could be covaried out of the relations between "social intelligence" measures and the PA and COMP subtests. In this way, one can examine whether PA and COMP performance predicts social functioning over and above general intelligence. Based on the literature reviewed previously, relations between performance on the PA and COMP subtests and "social intelligence" is expected to show little or no 19 relationship to indices of social functioning within this normal population. However, based on clinical practice and text book suggestions, one might expect that students who perform less well on the PA and COMP subtests to be more rejected and less accepted by their peers, demonstrate less positive social problem solving and report less social self-efficacy and greater loneliness and social dissatisfaction. Poor performance on PA and COMP subtests may also be associated with less positive social behavior as rated by students and teachers. In order to truly demonstrate that PA and COMP are valid indicators of social functioning, these relationships must be maintained even after independent assessments of general intellectual ability are partialled out. 20 CHAPTER III METHOD Participants Participants included 74 children1 (38 boys and 36 girls) and their classroom teachers from Grade 4 to 6 classrooms in a single parochial school in Lower Mainland Vancouver, British Columbia. Participants represented a diverse range of ethnic backgrounds and socioeconomic levels, but all were English speaking. All children were between the ages of 8-12 years at the time of testing. The mean age was 10.53 years. This age group was selected for several reasons. First, with regard to assessments of intellectual ability, Zigler, Balla, and Hodapp (1984) found that, for most children, measured intelligence remains relatively stable after 5 years of age. Second, with regard to social competence, research suggests that the development of oppositional, antisocial behavior problems begins early in life and these problems are stable overtime (Kazdin, 1987; Oleweus, 1979 in Greshem, 1999). Olweus, for example, found that aggressive, antisocial behavior in middle elementary school boys was as stable as measures of intelligence over one-year (r = .76) and five-year [r= .69) intervals. To recruit participants, a letter describing the general purpose of the study and a consent form was sent home to parents from each of the classrooms (see Appendix A). Parents were requested to read the letter, sign and return the accompanying consent form to the school. Students who received parental permission and who themselves agreed to be a part of the study were included in the study. A total of 94.7 percent of the possible sample participated in the study. 1 Two children's overall intelligence scores fell two standard deviations outside the normal range and were therefore omitted from the present study. All children represented fell within the average range. All analyses were conducted with the same sample of children. 21 Measures Several measures of intellectual and social functioning were administered to participants. Each measure is described below. Picture Arrangement and Comprehension subtests of the Weschler Intelligence Scale for Children -Third Edition (WISC-III, Wechsler, 1991). Following established arguments regarding the use of the PA and COMP subtests of the WISC-III as indices of social functioning (e.g., Ramos & Die, 1986; Schill et al., 1966, 1967, 1968), the PA and COMP subtests of the WISC-III were individually administered by the author, following standardized administration procedures. The PA subtest is a set of colorful pictures, presented in a mixed up order, which the child rearranges into a logical story sequence (Wechsler, 1991). The PA subtest assesses visual sequencing ability, but given the social content of the pictures, some have suggested that the measure also taps the ability to "anticipate consequences and interpret social situations" (Sattler, 1992, p.157). The COMP subtest is a series of orally presented questions about societal norms and every-day situations (Wechsler, 1991), but also taps understanding of behavioral and "social conventions, conscience and social judgment" (Sattler, 1992, p. 153). Previous research has shown the COMP and PA subtests to be reliable (reliability coefficients range from .76 to.77), and stable (stability coefficients range from .60 to. 75) (Wechsler, 1991). Standardized procedures were adhered to in the administration and the scoring of the WISC-III subtests (see Wechsler, 1991 for guidelines). Following Wechsler (1991), responses to the PA and COMP subtest items were summed to yield a single scaled score for each subtest. Scaled scores range from 1 -20, with higher scores reflecting better nonverbal sequencing ability (PA) and greater ability to reason about 22 various aspects of interpersonal relations, and social mores (COMP). Average scaled scores range from 8 to 12. The mean score on the PA subtest in the present study was 10.16, and scores ranged from 2-19. On the COMP subtest, the mean score was 8.75, and scores ranged from 3-13. The Kaufman Brief Intelligence Test (K-BIT; Kaufman, 1990). In order to statistically eliminate overall cognitive functioning as an explanatory variable for performance on the PA and COMP subtests of the WISC-lll, the K-BIT was administered. The K-BIT is 130-item verbal and nonverbal intelligence test that measures two different types of reasoning ability. The first area tapped is Verbal functioning, assessing Vocabulary or word knowledge across 82 items included in two subtests. On the Expressive Vocabulary subtest, participants were required to provide the name of a picture or object such as "lamp" or "calendar". The second subtest, Definitions, required that participants provide the word that best fits two clues (a phrase description and a partial spelling of the word). Nonverbal cognitive functioning is assessed using the Matrices subtest, a 48-item nonverbal measure composed of several types of items involving visual stimuli, both meaningful (people and objects) and abstract (designs and symbols). The child is presented with a series of pictures or patterns, and asked to complete a visual analogy (i.e., a picture of a hat is to a picture of a person's head as a sock is to (choice of "foot", "boot", "umbrella", etc.) or asked to choose form a series of patterns which of five items completes a visual matrix. All items require understanding of relationships among the stimuli, and all are multiple choice, requiring the person either to point to the correct response or to say its letter, thereby minimizing verbal expression demands. The K-BIT has sound psychometric properties, with internal consistency estimates of .78 to .89 for the age range considered in this study, and test-retest 23 reliability coefficients ranging from .73 to .91 (Kaufman & Kaufman, 1990). In terms of validity, as a measure of overall intelligence, the K-BIT has been found to correlate .43 with the full scale IQ score on the WISC-R and .68 with the Performance IQ Score of the WISC-R (Kaufman & Kaufman, 1990). Following standardized procedures, the examiner recorded student responses to the K-BIT items on the test record form during testing. Responses across both verbal and nonverbal tasks and subtests were later summed to yield a single, overall index of intelligence for each participant. Standard scores range from 40 to 160, with average scores ranging from 90 to 110. Higher scores signify greater intelligence as measured by the K-BIT. Total scores ranged from 55 to 126, with the mean IQ score being 101 on the K-BIT. Social Skills Rating System (SSRS; Gresham & Elliot, 1990). The teacher form of the SSRS was used to obtain teacher evaluations of the children's social competence and adaptive behaviors at school (Gresham & Elliott, 1990). The SSRS - Teacher Form (SSRS: TF; Elementary Level) is a 57-item scale that asks the teacher to indicate "how often" a behavior occurs for an individual child as well as "how important" that behavior is in order for the child to be successful in that teacher's classroom. The Teacher scale provides for an assessment of three different social skills (Cooperation, 10 items; Assertion, 10 items; Self-Control, 10 items), three problem behaviors (Externalizing, 6 items; Internalizing, 6 items, Hyperactivity, 6 items), as well as Academic Competence (9 items). Psychometrically, the SSRS:TF has been shown to have good internal consistency across subscales, with alpha coefficients ranging from .78 for Internalizing Problems to .95 for Academic Competence (Gresham & Elliott, 1990). In terms of validity, the SSRS:TF has been shown to correlate significantly with other indices of social behavior, including the Social Behavior Assessment (Stephens, 1978) (r = .68), 24 the Harter Teacher Rating Scale (TRS; Harter,1978) (r= .70), and the CBCL-Teacher Scores (r=.81) (Achenbach & Elerbeck, 1991 as cited in Gresham & Elliott, 1990). Following established procedures, teacher ratings on items relevant to each subscale were summed to yield two different measures of social behavior: Social Skills and Problem Behaviors, as well as a third score measuring Academic Competence. Across all three indices, higher scores were indicative of higher levels of the identified behavior in each case. Average scores ranged from 90 to 109. The mean score on the Social Skills scale for the present sample was 104 and standard scores ranged from 78 to 130. The mean Problem Behaviors score for the present sample was 97 and scores ranged from 85 to 131. Sociometric Status. To determine the degree to which each child was accepted versus rejected within the classroom, a sociometric rating scale measure was administered. For this measure participants were asked to rate on a 1 -5 scale how much they "liked to be with" each student in their class, with higher numbers reflecting greater liking. The average (mean) rating received from the same sex classmates was computed as an index of overall acceptance or likeability among peers, with higher scores indicating greater peer acceptance. Rating scale sociometric measures have been shown to be a valid and reliable index of social status or likeability, with evidence of good reliability and validity with children (see Rubin, Bukowski, & Parker, 1998 for a review). As well, several researchers have demonstrated that sociometric acceptance and rejection are stable over time (e.g., Asher & Dodge, 1986; Coie & Dodge, 1983; Roff, Sells & Golden, 1972; see Rubin et al., 1998 for a review). Asher and Dodge (1986), for example, reported stability correlations of .55 to .69 when measured by peer ratings over a over a six-month period and Hymel, Rubin, Rowden and Le Mare (1991) reported a stability 25 correlation of .56 from grades 2 to 5. Hymel and Rubin (1985) have noted a number of advantages of turning to peer informants for assessment of social competence. First, as "insiders", peers can identify characteristics of children and of relationships which are considered relevant from the perspectives of those who determine a child's social status and integration within a peer group. Second, the judgments of peers are based on experiences that may not be known to adults and other non-peer "outsiders". Third, because peer evaluations are completed by a number of reporters (i.e., the peers), the likelihood that error occurs due to an idiosyncratic aspect of a single reporter's experience with the child is reduced. The Children's Conflict Resolution Measure (Chung & Asher. 1992). Children's ability to solve typical social problems with peers was assessed using a self-report; multiple-choice questionnaire designed for use with elementary school age children by Chung and Asher (See Appendix B). Chung and Asher developed this measure to asses children's goals and strategies in conflict situations. On this measure, children are presented with a series of six hypothetical peer conflict situations and must choose one of five possible solutions to each problem, each reflecting a different response style: prosocial, hostile/coercive, assertive, passive and adult seeking strategies. Following procedures developed by Chung and Asher, responses to each of the six problem solving scenarios were summed to yield the number or percentage of vignettes for which the child selected each of the five types of strategies or responses. Each strategy score thus indicates the child's tendency to engage in that type of strategy in response to peer social problems. Chung and Asher found Cronbach's alpha coefficients for prosocial, hostile/coercive, assertive, passive and adult seeking strategies were .79, .90, .40, .55, and .57, respectively. Chung and Asher also evaluated the validity of their measure, demonstrating that prosocial, aggressive, withdrawn and average students 26 differed significantly and in expected ways in terms of their use of prosocial, hostile/coercive and passive strategies in response to peer conflict situations. Loneliness and Social Dissatisfaction Questionnaire (Asher & Wheeler, 1985). The Loneliness and Social Dissatisfaction Questionnaire, a 16-item self-report measure, was used to assess children's affective appraisal of their current social situation (Appendix C). The measure has been Identified as being internally consistent and stable in numerous studies (see Asher, Parkhurst, Hymel & Williams, 1990) for a review. All of the items focus on the peer context. The questionnaire consists of 16 primary items and 8 filler items that ask about hobbies, interests, and school subject preferences. The primary items assess four related constructs: (a) children's feelings of loneliness (e.g., "I'm lonely"); (b) children's perceptions of the degree to which certain important relationship provisions are being met (e.g., "There's nobody I can go to when I need help"); (c) children's perceptions of their social competence (e.g., "I'm good at working with other children"); and (d) children's appraisals of their current peer relationships (e.g., "I have lots of friends"). Children are asked to respond to the items using a 5-point Likert scale ranging from "that's always true about me" to "that's never true about me". The original version of the Illinois Loneliness and Social Dissatisfaction Questionnaire was used with third- through sixth-grade children. Recently, varziations also have been developed for use with middle-school children (Parkhurst & Asher, 1992), kindergarten and first-grade children (Barth & Parke, 1993; Cassidy & Asher, 1992), and mid-elementary-school students with mild retardation (Williams & Asher, 1992). Following procedures recommended by Asher Hymel and Renshaw (1984), student responses to each of the 16 items were summed (with some items reverse scored) to compute a single index of loneliness and social dissatisfaction, ranging from 16-80, with higher scores indicative of greater loneliness and social dissatisfaction. 27 There is considerable evidence that the original measure and its variations are psychometrically sound (e.g., Asher et al., 1984; Asher & Wheeler, 1985; Barth & Parke, 1993; Cassidy & Asher, 1992; Parkhurst & Asher, 1992; Williams & Asher, 1992). In several elementary and middle school samples, results of factor analyses consistently indicate a single factor composed of the 16 primary items, with the "pure" loneliness items (i.e., items 9,17, and 21) typically being the highest loading items. Previous research has also demonstrated both short-term and long-term stability in children's reports of loneliness. For example, Renshaw and Brown (1993) found that initial loneliness scores correlated .66 with loneliness scores assessed 10 weeks later, and .56 with loneliness scores assessed one year later. Hymel et al. (1983) reported similar stability over a one-year period. The internal reliability of the 16-item loneliness scale (measured using coefficient alpha) has been consistently about .90 in studies of both middle school students and regular education students in mid-elementary school. The reliability is lower (e.g., .79) but still acceptable with younger children and with older elementary school children with mild retardation (Williams and Asher, 1992). In the present sample, the internal consistency (Chronbach Alpha) of the loneliness measure was .64, somewhat lower than in the original sample, but nevertheless acceptable for research purposes. Relational Provisions Loneliness Questionnaire (RPLQ; Hayden-Thomson, 1989). Another measure focusing on children's perceptions of their relationships within both peer and family contexts is the Relational Provisions Loneliness Questionnaire (RPLQ, Appendix D) questionnaire constructed by Hayden-Thomson (1989). The measure was designed to assess four constructs: (a) a child's sense of group integration in the family (e.g., "In my family, I feel part of a group of people who do things together"), (b) a sense of group integration in the peer group (e.g., "I am part of a 28 group of friends that does things together"), (c) a sense of personal intimacy in the family (e.g., "There is someone in my family who I can turn to"), and (d) a sense of personal intimacy in the peer group (e.g., "There is someone my age I could go to if I were feeling down"). On the RPLQ, children are asked to respond to each of the 7 items included in each subscale using a 5-point, Likert scale ranging from "always true" to "not at all true." Previous research has verified these four distinct constructs through factor analyses, and has demonstrated the internal consistency of each, using coefficient alpha (group integration in peer context r = .80; personal intimacy in peer context r= .86; group integration in family context r= .89; personal intimacy in family context r = .90). Note that none of the items on the Hayden-Thomson (1989) questionnaire directly ask about feelings of loneliness. Hayden-Thomson demonstrated the construct validity of the measure by correlating each subscale with children's responses to separate items that explicitly focused on loneliness, reporting significant but moderate relations between reported feelings of loneliness and each of the four RPLQ scores (i.e., r=.48 for peer group-integration; r=.40 for peer intimacy; r = .55 for family integration; and r = .60 for family intimacy). Although significant, the magnitude of these correlations is modest suggesting that affective responses such as loneliness are not identical to the reported availability of family and peer social support as tapped by the RPLQ. For the purposes of the present study, only peer items of the RPLQ were used, given the focus on interpersonal relations within the peer context. Both the Peer Integration and Peer Intimacy subscales were found to be internally consistent within the present sample (r = .87 for Peer Integration, r = .84 for Peer Intimacy). The Children's Self-Efficacv for Peer Interactions (CSPI; Wheeler & Ladd, 1982). Children's feelings of social self-efficacy was assessed using the CSPI, a 22-item questionnaire designed to measure children's perceptions of their ability to use prosocial 29 verbal persuasive skills in specific peers situations (Appendix E). Each item on the CSPI consists of a statement describing a social situation (e.g., "You want to start a game.") followed by an incomplete sentence requiring the child to evaluate his or her ability to perform a specified social skill (e.g., "Asking other kids to play is for you"). For each situation, children were asked to circle one of four response choices: HARD! (1), hard (2), easy (3) or EASY! (4). Twelve of the items depict conflict situations and ten items depict nonconflict situations. Responses for all items are summed for a total self-efficacy score for each child (range = 22 to 88), with higher scores indicating greater social self-efficacy. Wheeler and Ladd (1982) found that the CSPI has demonstrated adequate internal consistency, reporting Alpha coefficients of .85 for the conflict items and .73 for the nonconflict component. The correlation between the conflict component and nonconflict total scores was .46, suggesting that these two item clusters comprise distinct but related components of the scale, although these two subcales have been combined in most studies using the CSPI. The stability of CSPI scores has also been demonstrated, with test-retest correlation coefficients of .90 for boys and .80 for girls observed over a two-week period. Wheeler and Ladd (1982) also demonstrate some evidence for the convergent validity of the CSPI, based on correlations between the CSPI correlates and measures of self-concept (e.g., correlations ranging from -.36 to -.49 between CSPI scores and self-reported anxiety on the Piers-Harris Self-Concept scale) as well as teacher ratings of social self-efficacy (correlations ranging from .21 to .40) and peer ratings of status (correlations ranging from .26-.40). In the present study, children's responses to the CSPI were summed to compute a single index of social self-efficacy, following suggestions made by Ladd and Wheeler 30 (1982). Within the present sample, overall CSPI scores ranged from 32 to 88 with a mean score of 66.3. Alpha coefficients with the present sample indicated an internal consistency of .94 for all the items of the scale. Procedure Children participated in a single individual testing session with the author and/or an equally qualified graduate student and a single group testing session, both conducted within the school. During the individual session, the PA and COMP subtests were administered along with the K-BIT (approximately 45 minutes). During group testing (approx. 25-30 minutes), all of the measures of social functioning were administered. Order of group and individual testing was counterbalanced across classrooms. During group testing, each item was read aloud in order to reduce the potential confound of reading ability. Participating teachers were asked to complete the SSRS- Teacher Form on each participant within 2-3 weeks of group and individual testing. 31 CHAPTER IV RESULTS Data Analyses Overview The purpose of this study is to determine if any aspect of social functioning can be predicted by performance on the PA or COMP subtests of the WISC-lll, after the effects of overall intelligence have been controlled. Accordingly, preliminary analyses examined the internal consistency of each of the measures used with the present sample. As well, an analysis of variance was undertaken in order to determine if there was a significant difference between male and female participants' scores and to determine whether further analyses should be undertaken with boys and girls separately. Once the psychometric adequacy of the measures was established, initial analyses considered the interrelations among the various indices of social functioning as well as among the IQ subtest and test scores, in order to determine the overlap and redundancy among conceptually similar measures. Following this, primary analyses began with consideration of the zero-order relations between indices of social functioning and scores on the IQ, PA and COMP measures, followed by regression analyses examining the prediction of social functioning from PA and COMP performance, after partialling out overall IQ (K-BIT). Table 1 contains descriptive information listing all the measures included in the present study. 32 Table 1 Descriptive Information Variable N Mean (SD) R a n g e Age 74 10.53 .91 8.08-12.07 Intelligence Measures Wechsler Intelligence Scale (Third-Edition) Picture Arrangement Subtest 73 10.27 3.94 2-19 Comprehension Subtest 73 8.82 3.66 3-13 Kaufman Brief Intelligence Test Composite Score 74 103.16 11.76 77-126 Social Measures Social Problem Solving Prosocial Strategies 74 3.07 1.76 0-6 Hostile/ Coercive Strategies 74 .45 1.16 0-5 Assertive Strategies 74 1.38 1.16 0-4 Passive Strategies 74 .43 .88 0-5 Adult Seeking Strategies 74 .66 1.00 0-5 Loneliness and Social Dissatisfaction Total Score 74 42.24 5.36 33-59 Social Support Integration and Belongingness 76 24.23 4.59 13-30 Intimacy 75 30.08 4.97 15-35 Social Self-Efficacy Total Social Self Efficacy 72 66.08 14.02 32-88 Sociometric Status Average Peer Liking 72 3.66 .57 1.64-4.88 Teacher Rating Social Skil ls Rating System* Social Skills Standard Score 50 104.58 14.63 78-130 Problem Behavior Standard Score 50 96.36 13.41 85-131 Academic Performance Standard Score 50 99.64 11.54 74-115 * SSRS forms were not completed by one of the classroom teachers, and therefore the sample size on this measure is reduced. In order to determine whether there was any significant difference in performance between the boys and girls of the present study, an Analysis of Variance (ANOVA) was undertaken. Results indicated there were statistically significant sex 33 differences for scores on the K-BIT Composite score (F (1, 72)=4.32, p < .04) as well as on the Hostility subscale of the Social Problem Solving questionnaire (F (1,72) = 5.78, p_< .02). On no other measure was there a significant difference of performance for boys and girls. Given this finding, subsequent analyses were conducted with the entire sample, boys and girls together. Furthermore, given the small N of the present study, boys and girls were considered together. Pearson Product Moment correlations were computed in order to examine the interrelations among the various indices of social competence and social functioning. Of particular interest was whether the many different indices of social functioning used in the present study provided unique and non-redundant measures. Results of the correlational analyses are presented in Table 2 below. 34 52 b « • m w CM o> o — CM O CM CO o C D o 00 o 03 o o 00 o O T - N. CO o o C O o CD CM o CO CO CO o CO o o •tf O OJ O y- r-en CO o co r~ r- CM o o CO 00 LO co "tf CO o L O o CO , _ q o LO LO CM o o o "tf CM CO Jo T CO C O o CO o ^ m T-O O CM CM * -CO CO O T-CM CO o CO CO LO • t f o CO LO CM 0) J O CO I— CO o 1— =J CO CO CD CO 'o o cn CO c g JS 0 o a c 5 O) c J E o> I ^ o k. 0-• t f ii o o co o •= 0. • t f II CD > 2 CD O o i2 to o I • t f II z CD > •e CD co CO < •tf r~-II Z CD > CO CO CO CL CM II c o u co CO 10 w * Q _ II a Z o « ° c CD CD CO 3 TJ < i - CM LO CAeS co CO to c o c o CM II CO CO CD c Ol c 'oi c o CD m •tf n o 03 u ~ UJ II "5 z co • t f ii z o> c CD CD Q L CD O 2 > I £ 1 O 2 « 4 o B >, » a a § g 3 I 1 -o s ai £ o (0 CD to .5 — cs E o u o CO CO J= O CO OC to CO CO 2 "o o CO o LO II S2 g •> CO x: CD CO E oj .£>. O .2 * - CM l q v V Ql 18 V 35 Results of the correlational analyses for the Social Problem Solving Questionnaire (Chung & Asher, 1992) indicated that prosocial strategies are negatively related to each of the other four approaches of social problem solving. The more a child uses a prosocial response style to solve problems, the less likely they are to use hostile, assertive, or passive strategies. Use of the remaining response styles were not significantly interrelated, suggesting that each of the indices of social problem solving taps a unique style of problem solving. Accordingly, in the present investigation, it was considered informative to evaluate each of these indices of problem solving style separately, in order to determine whether any of these response styles are predictive of performance on PA and/or COMP scores. Students who responded more prosocially in response to social problems were significantly more likely to be well liked by classmates (r (72) = .26, p_ < .05) on the peer liking measure, whereas students who responded more assertively were less likely to be well liked (r (72) = -.36, p_< .01). Children who responded more passively to social problems had a lower sense of integration and belongingness with their peer group (r (73) = -.24, Q < .05). The degree to which students engaged in more passive, adult-seeking or hostile/coercive strategies was surprisingly unrelated to overall peer liking. Thus, although there were some associations observed between social problem solving and overall peer status or liking, as would be expected, these indices of social functioning were not considered overlapping. As may be expected, children who have a hostile response style on the Chung and Asher social problem solving measure tended to receive high teacher ratings on the Problem Behavior's subscale of the Social Skills Rating System (r (48) = .30, p <.05). The significant, but modest correlation observed here suggests some, but not complete overlap in self versus teacher perceptions of negative problem 36 solving behavior. There was a significant yet modest correlation between children's self reported loneliness and children's sense of integration and belongingness to a peer group (r(70)= -.36, p<.01) as well as perceived social self-efficacy (r(72)= -.30, p_<.05). Children who feel that they are part of a cohesive group and who perceived themselves as socially effective reported significantly less loneliness, although the magnitude of these relations is modest, suggesting that each measure taps a somewhat unique aspect of perceived social functioning. There was no significant relationship observed between children's loneliness and style of problem solving, peer liking, social skills or problem behaviors as rated by teachers. Interestingly, a significant relationship was observed between children's sense of social support (both a sense of Integration and Belongingness as well as a sense of Intimacy) and their perceived sense of self-efficacy (r (70) = .58, p <.01, and r (69) = .43, p<.01 respectively). Children who felt that they had a strong level of social support and intimacy with peers also felt a sense of social self-efficacy. Perceived social support was found to be unrelated to children's problem solving style, peer liking, and teacher rating of social skills and problem behaviors. As mentioned earlier, social self-efficacy related to loneliness modestly, but was not significantly related to social problem solving, peer liking, or teacher ratings. Thus, it appears that both perceived social support and perceived social self efficacy provide unique indices of social functioning. Teacher rating of children's social skills as well as teacher's ratings of problem behavior on the SSRS were significantly and fairly highly related (r (49)= -.72, p<.01) to one another. As mentioned previously, higher scores on the Problem Behaviors subscale was related to a hostile/coercive response style on the Social Problem Solving 37 Questionnaire. Sociometric status did not correlate with many of the social measures administered. It was related to a prosocial problem solving strategy (r (72) = 26, p<.05) and correlated negatively with assertiveness (r (72) = -.36, p<.01) suggesting that the less assertive you are the more well liked you are. Results of the preliminary correlational analyses suggested that that there were some significant intercorrelations observed among the Social Problem Solving Questionnaire scores. Specifically, students that use prosocial problem solving strategies as less likely to employ hostile, coercive ones or assertive ones. There were some modest associations observed among the Social Support and Social Self-Efficacy scores, as well as the Loneliness and Social Self-Efficacy scores. The overlap among measures is generally moderate (.3 to .5), suggesting that there is a fair degree of unshared variance. Taken together, these finings suggest that each of the social measures reflects different aspects of social competence. Correlational analyses were also conducted to examine interrelationships among the various indices of academic and intellectual functioning considered in the present study. Results are presented in Table 3 below. 38 Table 3 Intercorrelations among Intelligence Measures Intelligence Measures K-BIT WISC-III SSRS:TF K-BIT (N=73) Vocabulary Matrices Composit e PA COMP Academic Achievement Vocabulary Subtest .22 . 7 6 " .32** 48** .53** Matrices Subtest ~ .76** .35** .19 .46** Composite Subtest - .45** .44** .62** WISC-III (N=73; Picture Arrangement Subtest - .32** ."46** Comprehension Subtest - .41** SSRS:TF (N=50) Academic Achievement _____ As can be seen in Table 3, rather strong and consistent correlations were observed among the various indices of cognitive functioning, as might be expected. The highly positive correlations between the K-BIT subscale and total scores supports the use of only the overall IQ score in subsequent analyses. Also as expected, slightly higher correlations were observed between K-BIT Vocabulary subtest score and performance on the COMP subtest of the WISC-III, also tapping verbal components of cognitive functioning, as well as between K-BIT Matrices subtest scores and performance on the PA subtest of the WISC-III, tapping nonverbal components of intellectual functioning. The consistently positive and significant correlations observed between the various IQ measures and teacher ratings of academic functioning on the SSRS provides some validity for these IQ measures as tapping processes relevant to educational performance in the classroom. Table 4 indicates the measures included in the present study. 39 Tab le 4 . T a b l e of M e a s u r e s Intelligence Measure Social Measures Social Problem Solving 1. Kaufman Brief Intelligence Test K-BIT Composite IQ 1. Prosocial (N=74) score.(N=74) 2. WISC-III Picture Arrangement Subtest (PA) (N= 73) 2. Hostile/Coercive (N=74) 3. WISC-III Comprehension Subtest (COMP) (N=73) 3. Assertive (N=74) 4. Teacher Rating of Academic Performance (Social Skills 4. Passive (N=74) Rating System (SSRS:TF) (N=50) 5. Adult Seeking (N=74) Loneliness & Social Dissatisfaction 6. Loneliness (N=74) Social Support 7. Peer Integration and Belonging (N=74) 8. Peer Intimacy (N=73) Social Self-Efficacy 9. Total Social Self Efficacy (N=72) Sociometric Status 10. Average Peer Liking (N=72) Social Skills Rating System 11. Social Skills (N=50) 12. Problem Behaviors (N=50) Of primary interest in the present study , however, is whether these indices of IQ can predict assessments of social functioning. The next set of analyses examined the correlations between performance on the COMP and PA subtests, overall IQ and each of the indices of social functioning examined in the present study. In Table 5, zero-order correlations of the K-BIT, PA and COMP scores with various indices of social 40 performance are presented in the first three columns, followed by partial correlations, examining the relationships between social functioning and PA/COMP scores once overall IQ (K-BIT scores) is partialled out. 41 Table 5 Correlations between Overall Intelligence, the PA and COMP subtests with Indices of Social Functioning Zero-order Correlations Social Problem Solving Partial Correlations K-BIT COMPOSITE WISC- WISC-PA COMP PA COMP Prosocial Strategies (N = 74) .09 .12 .16 .09 .14 Hostile/ Coercive Strategies (N = 74) -.09 -.00 -.16 -.06 -.20 Assertive Strategies (N = 74) .08 -.21 .07 -.31* -.09 Passive Strategies (N = 74) -.21 -.20 -.15 -.12 -.16 Adult Seeking Strategies (N = 74) .20 .20 -.07 .09 -.18 Loneliness Loneliness Total (N = 74) - .33" -.22 -.21 -.07 -.13 Social Support Integration/ Belongingness (N=74) .18 .12 .19 .02 .07 Intimacy (N=74) .03 .00 .21 .00 .24 Social Self-Efficacy Total Social Self-Efficacy (N=73) -.13 .07 . 3 1 " .08 .33* Sociometric Status Average Peer Liking (N = 73) -.04 .00 .03 .21 .03 Teacher Rating S S R S T F Social Skills Standard Score (N= 49) .24 .08 .32* -.02 .26 S S R S T F Problem Behaviors Standard Score (N=49) -.32* -.18 .33* -.04 -.24 SSRS:TF Academic Achievement (N=49) .62" .46 . 4 1 " .24 .24 *E < .05 (2-tailed). ** e < .01 (2-tailed). Pearson Product correlations revealed that there was little relationship between 42 the various measures of social functioning and overall intelligence as measured by the K-BIT composite score. Overall intelligence correlated modestly at best with loneliness (r (73) = -.33, p < .01), indicating that children with higher overall intelligence rated themselves less lonely. Overall intelligence as measured by the K-BIT was also significantly correlated with a low score on problem behavior as rated by the classroom teachers (r (47)= -.32, p<.05), and teacher ratings of academic achievement (r (47) = .62, p < .01). More intelligent children were reported to have fewer behavioral problems according to their teacher and to be achieving well at school. No other social measure had a significant relationship to overall intelligence. The absence of a significant relationship between the IQ measure used and the social indices questions the use of intelligence tests as means of measuring anything other than academic intelligence and supports the modern day theorists (e.g., Goleman, 1995, Gardner,1983) who argue that the social aspect of intelligence is separate from academic intelligence. Despite arguments that performance on PA reflects social functioning, results of the correlational analyses indicated that only the assertive response style related to PA scores (r (73) = -.31, p <.05) in the present sample once intelligence was accounted for. Thus, these limited findings call into question previous assumptions that social competence/intelligence is measured by the PA subtest. The research to date has demonstrated some support for the COMP subtest of the WISC-III as a measure of "social intelligence/competence". Results of the present correlational analyses revealed that there is a relationship between certain measures of social competence and scores on the COMP subtest in several areas before intelligence is factored out as a contributing factor, and that only one area of social competence remains significantly related to COMP scores when intelligence was accounted for. Students who performed well on the COMP subtest were rated by their teachers as 43 having better social skills, and rated themselves to have higher social-efficacy. Once overall intelligence is accounted for as a contributing factor, COMP scores only related to students impressions of their social efficacy (r (71) = .33, p<.05). Given the modest but significant correlations (ranging from .31 to .33) observed, a series of regression analyses were undertaken in order to determine whether performance on the PA or COMP subtests could predict any of the social indices considered in the present study once overall IQ was accounted for. Specifically, a series of two hierarchical multiple regression analyses were conducted for each of the social indices evaluated in the present sample one predicting PA and the other predicting COMP, with overall IQ entered initially (Step 1), followed by the relevant index of social functioning (Step 2). In essence, these analyses examined the ability of performance on the PA and COMP subtests to predict performance on various measures of social functioning once the variance due to overall intelligence is factored out. Given that, across the intelligence measures, scores were standardized and based an age norms, and because there were few statistically significant sex differences across the variables of the present study, boys and girls were considered together. Results of the multiple regression analyses are presented in Table 6. 44 Table 6. Hierarchical Regression Analyses: Comprehension Steps and Criteria F Change Sig. F Change Social Problem Solving Analysis 1: Prosocial Strateaies Step 1: K-BIT Step 2: COMP .01 .03 .40 1.47 NS NS Analysis 2: Hostile/ Coercive Strateaies Step 1: K-BIT Step 2: COMP .01 .03 .70 1.15 NS NS Analysis 3: Assertive Strateaies Step 1: K-BIT Step 2: COMP .01 .01 .66 .06 NS NS Analysis 4: Passive Strateaies Step 1: K-BIT Step 2: COMP .05 .05 3.69 .27 NS NS Analysis 5: Adult Seekina Strateaies Step 1: K-BIT Step 2: COMP .04 .08 3.28 2.44 NS NS Loneliness Analysis 6: Loneliness Total Score Step 1: K-BIT Step 2: COMP .09 .10 7.23 .68 .01* NS Social Support Analysis 7: Intearation/Belonaina Step 1: K-BIT Step 2: COMP .03 .05 2.34 1.07 NS NS Analysis 8: Intimacy Stepl: K-BIT Step 2: COMP .00 .05 .07 3.36 NS NS Social Self-Efficacv Analysis 9: Social Self-Efficacv Total Score Step 1: K-BIT Step 2: COMP .01 .10 .87 6.44 NS .01* Sociometric Status Analysis 10: Averaae Peer Likina Stepl: K-BIT Step 2: K-BIT, COMP .00 .00 .01 .09 NS NS Teacher Ftatina (SSRS) Analysis 11: Social Skills Step 1: K-BIT Step 2: K-BIT, COMP .04 .11 1.99 3.46 NS NS Ananvsis 12: Problem Behavior Step 1: K-BIT Step 2: COMP .09 .14 4.39 3.14 .04* NS *fi<.05 45 Table 7. Hierarchical Regression Analyses: Picture Arrangement Step and Criteria R Square F Change Sig. F Change Social Problem Solving Analvsis 1: Prosocial Strateaies Step 1: K-BIT Step 2: PA .01 .02 .40 .64 NS NS Analvsis 2: Hostile/ Coercive Strateaies Step 1: K-BIT Step 2: PA .01 .01 .70 .13 NS NS Analvsis 3: Assertive Strateaies Step 1: K-BIT Step 2: PA .01 .09 .66 5.81 NS .02 Analvsis 4: Passive Strateaies Step 1: K-BIT Step 2: PA .05 .06 3.67 .8U NS NS Analvsis 5: Adult Seekina Strateaies Step 1: K-BIT Step 2: PA .04 .06 3.28 1.19 NS NS Loneliness Analvsis 6: Loneliness Total Score Step 1: K-BIT Step 2: PA .09 .10 7.23 .84 .01* NS Social Support Analvsis 7: Intearation/Belonaina Step 1: K-BIT Step 2: PA .03 .04 2.34 .20 NS NS Analvsis 8: Intimacv Step 1: K-BIT Step 2: PA .00 .00 .07 .02 NS NS Social Self-Efficacy Analvsis 9: Social Self-Efficacv Total Score Step 1: K-BIT Step 2: PA .01 .01 .87 .05 NS NS Sociometric Status Analvsis 10: Averaae Peer Likina Step 1: K-BIT Step 2: PA .00 .00 .01 .00 NS NS Teacher Ratina (SSRS) Problem Behavior Step 1: K-BIT Step 2: PA .09 .09 4.39 .12 .04* NS Analvsis 11: Social Skills Step 1: K-BIT Step 2: PA .04 .04 1.99 00 NS NS *E<.05 46 Results of the regression analyses replicate findings reported in correlational analyses and demonstrate that performance on both the PA and COMP subtests contribute minimally to the prediction of various indices of social functioning, even after overall IQ is partialled out. Specifically, overall IQ accounted for a small but significant amount of variance in the prediction of self reported loneliness (R2 = .10, rj<.01), teacher ratings of problem behaviors (R2 = .09, Q< .01), and teacher rating of problem behavior (R2 = .09, Q< .04). When PA and COMP scores were added on Step 2 of the regression analyses results indicated that PA scores contribute significantly to the prediction of the use of assertive problem solving strategies, and that the COMP subtest predicted a greater sense of social self-efficacy (R2 = .10, JD<.01) after overall IQ was partialled out. However, neither PA nor COMP performance was predictive of prosocial, hostile/coercive, passive or adult-seeking problem solving strategies, loneliness, social support, sociometric status nor teacher rating of social behavior. Thus, in sum, present data provided little support for the concurrent validity of the PA and COMP subtests as predictors of "social competence/intelligence". Only one of the eleven social indices was predicted from performance on the PA subtest, and one other of the social indices was predicted by performance on the COMP subtest. 47 CHAPTER V DISCUSSION The present thesis tested the traditional assumption that the WISC-lll PA and COMP subtests are interpretable as measures of social intelligence or competence. A review of the research to date regarding the links between social functioning and performance on the PA and COMP subtests indicates that results have been mixed and fraught with methodological problems, making interpretations difficult, if not impossible. Despite limited empirical evidence supporting this contention, however, clinicians continue to interpret the PA and COMP subtests as indices of social intelligence or competence. Some of the limitations of previous literature were addressed in the present study in four ways. First, the WISC-lll; the most recent version of the Wechsler scale was used to examine links between social functioning, COMP and PA performance. Second, various indices of social functioning were considered (e.g., behavioral, social, cognitive and affective measures), derived from various sources of information (i.e., participant self-report, teacher report and peer nominations), allowing for a broad and more thorough evaluation of social competence and social intelligence. Third, overall intelligence was statistically eliminated as an explanatory variable of performance on PA and COMP. Finally, although most studies in this area have considered adult samples and/or clinical samples, the present study examined links between social functioning and performance on the PA and COMP subtests within a normal, school-aged population, enhancing the generalizability of the findings. Several preliminary analyses were undertaken in order to determine the psychometric adequacy of the measures chosen and to determine whether subsequent analyses should consider examination of boys and girls separately or 48 together. Results indicated that boys and girls did not differ in performance across measures, with the exception of boys having more hostile responses than girls did on the Children's Conflict Resolution Measure. Following this, subsequent analyses did not differentiate between the two sexes. Furthermore, the internal consistencies of the measures utilized in the present study were usually quite comparable to those reported in prior research and appeared to be adequate for use in the present study. Preliminary correlational analyses also considered the interrelations among the various indices of social functioning as well as among the measures of intellectual functioning in order to determine the overlap and redundancy among conceptually similar measures. Results indicated significant relations among the various measures of social functioning, consistent with prior research, although the magnitude of these correlations was modest at best across all measures with the exception of the social skills and problem behaviors subtests of the SSRS, that indicated a strong negative relationship to one another. Both subtests were retained due to the fact that they are thought to measure opposite skills. Thus, although related, each of the measures of social functioning appeared to tap a somewhat unique aspect of social competence, and were therefore retained for subsequent analyses. Results of Pearson Product Moment Correlations indicated moderate and consistent relationships among the various indices of cognitive functioning, as might be expected. The positive correlations between the K-BIT subscales and total scores supported the use of only the overall IQ score in subsequent analyses. The significant but modest (.2 -.3) correlations observed between overall IQ scores on the K-BIT and scores on both the COMP and PA subtests of the WISC-III lend support to the assumption that both PA and COMP performance is part of overall intellectual functioning, but that the overall K-BIT score provided a somewhat distinct measure of 49 more general intelligence for the present study. The primary focus of the thesis was on whether or not PA and COMP performance predicted social competence or social intelligence. Results of the correlational analyses indicated that performance on the PA subtest was generally unrelated to any index of social functioning. Although self-reported loneliness was significantly related to PA performance in zero-order correlations, this relationship did not remain significant in partial correlations after the influence of overall IQ was partialled out. The Assertiveness subscale of the Social Problem Solving Questionnaire was not significantly related to PA performance in zero-order correlations but was significantly related to PA performance once IQ was partialled out. However, the amount of variance accounted for was modest at best. Results of hierarchical regression analyses confirmed these findings. Once overall IQ was considered, performance on the PA subtest was a significant predictor of only one of the twelve indices of social functioning - the use of more assertive problem solving strategies accounting for only about 9% of the variance for this measure. Overall, however, contrary to arguments that performance on the Picture Arrangement subtest reflects social functioning, results of both correlational and regression analyses provided little, if any, evidence to support this contention within a normal sample of elementary school age children, despite consideration of a variety of indices of social competence. For the most part, the predictive power of the PA subtest to predict social functioning was minimal and the present results are consistent with other recent studies (e.g., Beebe et al., 2000; Campbell & McCord, 1996; Lipsitz et al., 1993) that do not support the use of PA as a measure of social intelligence and questions Sattler's (1992) suggestion that PA measures "a child's ability to comprehend and evaluate a social situation" (p.1123). 50 Previous research has demonstrated some support for the COMP subtest of the WISC-III as a measure of "social intelligence/competence". Results of the present zero-order correlational analyses revealed that in the present sample, stronger performance on the COMP subtest was significantly associated with more positive teacher ratings of social skills, and a greater sense of social self-efficacy, teacher ratings of academic achievement, but was not related to social problem solving, loneliness, peer intimacy and integration, nor sociometric status. Once overall IQ was partialled out, significant relationships remained between COMP performance and greater self-reported social self-efficacy. Moreover, the relationships between COMP performance and social problem solving strategies, self-reported loneliness, perceived peer integration and peer intimacy support, sociometric status, teacher rating of social problem solving skills and problem behaviors remained nonsignificant. These correlational results were only partially confirmed by results of regression analyses. Specifically, results of regression analyses revealed that once overall intelligence was controlled for, COMP performance accounted for 10% of the variance self-reported social self-efficacy. Thus, children who performed better on the COMP subtest perceived themselves as more socially efficacious although the amount of variance accounted for in each case was minimal. Moreover, no other predictions reached significance. Thus, performance on the COMP subtest was not predictive of social problem solving style, self-reported loneliness, perceived peer integration and peer intimacy, actual peer acceptance or teacher rating of students' social skills and problem behaviors. Results of the present study make it difficult to believe that performance on COMP is a valid predictor of social functioning in elementary school aged children. The present study provides us with a glimpse of how the school-aged child 51 performs on the PA and COMP subtests of the WISC-lll and how their performance relates to their social functioning as reflected in indices of affective, cognitive and behavioral social competence scales. Studies such as this are imperative in order to improve the theory of intelligence, the research on the assessment of various aspects of intelligence and, most importantly, the clinical practice of psychological assessment. The results of the present study most profoundly affect practitioners who administer and interpret the WISC-lll, their trainers, and the children who are affected by the results. It is important that valid interpretation of test scores be adhered to in order to ensure sound practice. The present thesis does not strongly support the assumption that has existed for half a century that the PA and COMP subtest tap social intelligence. Furthermore, overall intelligence was not observed to correlate with social functioning. Thus, academic or cognitive "intelligence" as measured by the present IQ tests assess something other than social competencies (e.g., academic abilities). This lack of correlation suggests the need to expand the breath of "overall intelligence tests" to cover social functioning. Contemporary theorists (e.g., Gardner, 1983, Goleman, 1995) acknowledge that there are many classes of intelligence, but an encompassing measure has yet to be developed. The fact that this study used various raters to measure social competence (peer, teacher and self-evaluations) is a first in this literature to date and it is felt that given this multi-rater approach of data collection, that the results are more sound and more generalizable. Unlike most other studies of the validity of the PA and COMP subtests, the present study used the most recent version of the WISC-lll making the findings more pertinent to clinicians practicing today. Having used the most recent measures, the findings do not support the use of the PA and COMP subtests as indicators of 52 social competence, and therefore the use of these subtests as a means of tapping social skills and awareness is even more strongly discouraged. Furthermore, if, as Kaufman (1994) has stated, the Picture Arrangement subtest has changed considerably and is less "socially laden", then it should follow that interpretations of the subtest should also be revised. Whereas most of the previous studies in this area have been correlational, the present study used a more robust statistic - regression analysis to further explore the relationship between the WISC-III PA and COMP subtests and social functioning. There are several limitations to the present study. First, participants all came from a parochial school in Greater Vancouver, which may lead to questions regarding the representativeness of the sample and raise concerns regarding generalizability. The present sample size was also limited and there, but certainly adequate for the analyses conducted. Replication of these findings with larger and more diverse clinical as well as non-clinical samples would be valuable. The present study failed to demonstrate clear or consistent links between performance on the COMP subtest of the WISC-III and various indices of social functioning. However, stronger relationships may be observed if efforts were made to differentiate socially relevant items on the COMP from those that are less socially relevant. Specifically, future research may wish to examine the relationship of specific comprehension items that require social judgement over ones that do not, as suggested by Beebe et al. (2000). For example, questions that ask "What is the thing to do if a boy (girl) much smaller than you starts to fight with you?" (Wechsler, 1991, p. 135) involves more social awareness than questions that ask "Tell me some reasons why you should turn off lights when nobody is using them." (Wechsler, 1991, p. 134). Results may indicate that certain questions of the COMP 53 subtest are linked to social functioning more than others. However, the usefulness of this finding is questionable in that practitioners use a scaled score comprised of the sum of all the items when interpreting results on the COMP subtest. Therefore individual item analysis would not be useful other than for anecdotal purposes. 54 REFERENCES Allison, J . , Blatt, S.J., & Zimet, C. N. (1968). The interpretation of psychological tests. New York: Harper and Row. Asher, S.R., & Wheeler, V.A., (1985). Children's loneliness: A comparison of rejected and neglected peer status. Journal of Consulting and Clinical Psychology. 53, 500-505. Asher, S.R., & Dodge, K.A., (1986). Identifying children who are rejected from their peers. Developmental Psychology, 22, 444-449. Asher, S.R., & Hopmeyer, A. (1999). Loneliness in childhood. In Children's Needs II. Development Problems and Alternatives. G.G. Bear, K.M. Minke, and A. Thomas (Eds.). National Association of School Psychologists. Asher, S.R., Hymel, S., & Renshaw, P.D., (1984). Loneliness in children. Child Development, 55, 1456-1464. Asher, S.R. Parkhurst, J.T. Hymel, S., & Williams, G.A. (1990). Peer rejection and loneliness in childhood. In Asher, S.R. (Ed); Coie, J.D. (Ed); et-al. (1990). Peer rejection in childhood. New York, NY, USA: Cambridge University Press (pp. 253-273). Beebe, D.W., Pfiffner, L.J., & McBurnett, K. (2000). Evaluation of the validity of the Wechlser Intelligence Scale for Children - Third Edition Comprehension and Picture Arrangement subtests as measures of social intelligence. Psychological Assessment, 12 (1), 97-101. Blatt, S.J., & Allison, J . (1981). The intelligence test in personality assessment. In A.l. Rabin (Ed.) Assessment with Projective Techniques: A Concise Introduction (pp.187-231). New York: Springer. Blatt, S.J., & Allison, J . (1968). The Intelligence Test in Personality 55 Assessment. In A.l. Rabin (Ed.), Projective Techniques in personality assessment. New York, Spinger. Brannigan, G.G., (1975a). Wechsler picture arrangement scores and children's problem solving. Journal of Psychology. 91(197-200). Brannigan, G.G., (1975b). Wechsler picture arrangement and comprehension scores as measures of social maturity. Journal of Psychology. 89 (1). 133-135. Campbell, J.M., McCord, D.M. (1996). The WAIS-R comprehension and picture arrangement subtests as measures of social intelligence: Testing traditional interpretations. Journal of Psvchoeducational Assessment. 14_(3), 240-249. Campbell, J.M., & McCord, D.M. (1999). Measuring social competence with the Wechsler Picture Arrangement and Comprehension subtests. Assessment 6(3) 215-223. Chung, T.T., & Asher, S.R. (1996). Children's goals and strategies in peer conflict situations. Merrill-Palmer Quarterly. 42 (1). 125-147. Cronbach, L.J. (1951). Coefficient alpha and the internal structure of tests. Psvchometrika. 16, 297-334. Culberton, F.M., Feral, C.H., & Gabby, S. (1985). Pattern Analysis of Wechsler Intelligence Scale for Children-Revised profiles of delinquent boys. Journal of Clinical Psychology 45 (4) 651 -660. Ford, M.E. (1982). Social cognition and social competence in adolescence. Developmental Psychology. 18, 323-340. Gardner, H. (1985). The Mind's New Science. New York: Basic Books. Gardner, H. (1983). Frames of mind. The theory of multiple intellences. New York: Basic Books. 56 Glasser, A.J., & Zimmerman, I.L (1967). Clinical interpretation of the Wechsler Intelligence Scale for Children. New York: Grune & Stratton. Goldfried, M.R., & D'Zurilla, T.J. (1969). A behavior-analytic model for assessing competence. In C D . Spielberger (Ed.), Current issues in clinical community psychology (pp.151 -198). New York: Academic Press. Goleman, D. (1995). Emotional Intelligence. New York: Bantam. Gresham, F.M., & Elliott, S.N., (1990). Manual for The Social Skills Rating System. Teacher Form. Minnesota: American Guidance Service, Inc. Gresham, F.M., & Elliott, S.N., (1990). The Social Skills Rating System. Teacher Form. Minnesota: American Guidance Service, Inc. Guilford, J.P. (1967). The nature of human intelligence. New York. McGraw-Hill. Hayden-Thomson, L. (1989). The development of the Relational Provision Loneliness Questionnaire for children. Unpublished doctoral dissertation, University of Waterloo, Waterloo, Ontario. Hymel, S., & Rubin, K.H., (1985). Children with peer relationship and social skills problems: Conceptual, methodological, and developmental issues. In G.J. Whitehurst (Ed.) Annals of Child Development (Vol 2), Greenwich, CT: JAI Press. Hymel, S., Rubin, K.H., Rowden, L, & Le Mare, L. (1990). Children's peer relationships: Longitudinal predictions of internalizing and externalizing problems from middle to late childhood. Child Development. 61, 2004-2021. Johnson, D.T., (1969). Introversion, extroversion and social intelligence: A replication. Journal of Clinical Psychology. 25 (2), 181-183. Kaufman, A.S. (1994). Intelligent testing with the WISC-III. New York: John Wiley & Sons. 57 Kaufman, A.S., & Kaufman, N.L, (1990). Manual for the Kaufman Brief Intelligence Test. Minnesota: American Guidance Service. Krippner, S. (1964).WISC Comprehension and Picture Arrangement subtests as measures of social competence. Journal of Clinical Psvcholoov, 20, 366-376. Lipsitz, J.D., Dworkin, R.H., & Erlenmeyer-Kimling, L. (1993). Wechsler comprehension and picture arrangement subtests and social adjustment. Psychological Assessment, 5, 430-437. Lyons, V.M., & Levine, S.M., (1978). Correlation between responsiveness to information and vicarious learning. Perceptual and Motor Skills. 47, 467-474. McFall, R.M., (1982). A review and reformulation of the concept of social skills. Behavioral Assessment, 4,1-33. Nobo, J . , & Evans, R.G. (1986). The WAIS-R PA and COMP subtests as measures of social behaviour characteristics. Journal of Personality Assessment. §0(1), 90-92. Oppenheimer, L. (1989). The nature of social action: Social competence versus social conformism. In B. Schneide, G. Attili, J . , Nadel & R. Weissberg (Eds.), Social competence in developmental perspective (pp. 40-70). Dordecht, Netherlands: Kluwar International Publishers. Ramos, M., Die, A.H. (1986). The WAIS-R Picture Arrangement Subtest: What do scores indicate? The Journal of General Psychology, 113, 251-261. Rapaport, D. (1946). Diagnostic Psychological Testing (Vol.ll). Chicago: Year Book Publishers. Rapaport, D., Gill, M.M., & Shafer, R. (1968). Diagnostic psychological testing (Rev. Ed.) New York: International Universities Press. Reiff, H.B., & Gerber, P.J. (1990). Cognitive correlates of social perception in 58 students with learning disabilities. Journal of Learning Diabilities. 23_(4), 260-262. Renshaw, P.D., & Brown, P.J. (1993). Loneliness in middle childhood: Concurrent and longitudinal predictors. Child Development. 64,1271-1284. Rubin, K.H., & Rose-Krasnor, L (1992). Interpersonal Problem Solving. In V.B. Van Hassett & M. Hersen (Eds.), Handbook of Social Development (pp.283-323). New York: Plenum. Rubin, K.H., Bukowski, W., & Parker, J.G. (1997). Peer interactions relationships and groups. In W. Damon (Series Ed.) and N. Eisenberg (Vol.Ed.), Handbook of Child Psychology: Vol. 3. Social emotional and personality development (5 t h edition, pp. 619-700). New York: Wiley. Sattler. J.M.. (1992). Assessment of Children - (Third Edition). Boston: Jerome M. Sattler, Publisher. Schill, T (1966). The effects of MMPI social introversion on WAIS PA performance. Journal of Clinical Psychology. 22, 72-74. Schill, T., Kahn, & M., Muehleman, T. (1968a). Verbal conditionability and Wechsler PA Scores. Journal of Consulting and Clinical Psychology. 32(6), 718-721. Schill, T., Kahn, M., & Muehleman, T. (1968b). WAIS PA performance and participation in extra curricular activities. Journal of Clinical Psychology. 24, 95-96. Schafer, (1948). The Clinical Application of Psychological Tests. New York: International Universities Press. Simon, J.M., & Evans, J.R. (1980). WISC-R PA subtest scores and peer acceptance/rejection. Perceptual and Motor Skills. 5J. (2), 558. Sipps, G.J., Berry, W., & Lynch, E.M., (1987). WAIS-R and social intelligence: A test of established assumptions that uses the CPI. Journal of Clinical Psychology. (43) 5, 499-504. 59 Stabler, B., Whitt, J.K., Moreault, D.M., D'Ercole, A.J. , & Underwood, L.E. (1980). Psychological Reports. 46. 743-746. Sternberg, R. (1985). Practical intelligence and people skills: Beyond I.Q. New York: Cambridge University Press. Terry, R.L., & Berg, A.J. (1984). The relationship between the WAIS PA and MMPI S i is mediated by the MMPI P D . Journal of Clinical Psychology. 40(4), 970-971. Thorndike, E.L. (1920). Intelligence and its uses, Harper's Magazine, pp.227-235. Toomey, R., Wallace, J , Corrigan, P.W., Schuldberg, D., & Green, M.F. (1997). Social processing correlates of nonverbal social perception in schizophrenia. Psychiatry Interpersonal and Biological Processes. 60(4), 292-300. Turner, R.G & Horn, J.M. (1976). MMPI correlates of WAIS subtest performance. Journal of Clinical Psychology, 32, 583-594. Tymchuk, A.J., Simmons, J.Q., Neafsey, S., (1977). Intellectual characteristics of adolescent childhood psychotics with high verbal ability. Journal of Mental Deficiency Research. 21.133-138. Waters, E. & Sroufe, L.A. (1983). Social competence as a developmental construct. Developmental Review, 3, 79-97. Wechsler, D. (1944). The Measurement of Adult Intelligence. Baltimore: Williams & Wilkins. Wechsler, D. (1969). The Range of Human Capacities. New York, Hafner. Wechsler, D. (1991). Manual for the Wechsler intelligence Scale for Children. Third Edition, New York, Psychological Corporation. White, S. (1959). Motivation reconsidered: The concept of competence. Psychological Review. 66, 297-333. 60 Wheeler, V.A., & Ladd, G.W. (1982). Assessment of children's self-efficacy for social interactions with peers. Developmental Psychology. 18_(6), 795-805. Williams, G.A., & Asher, S.R. (1992). Assessment of loneliness at school among children with mental retardation. American Journal of Mental Retardation, 96, 373-385. Woo-Sam, J.M. & Zimmerman, I.L. (1972). Speed as a variable on three WISC subtests. Perceptual and Motor Skills. 34, 451-455. Zigler, E. (1973). Project Head Start: Success or Failure? Learning, 1^  43-47. Zigler, E., Balla, D., & Hodapp, R (1984). On the definition and classification of mental retardation. American Journal of Mental Deficiency, 89, 215-230. 61 APPENDIX B The Children's Conflict Resolution Measure (Chung & Asher, 1992) WHAT ARE YOU GOING TO DO? THIS IS A QUESTIONNAIRE ABOUT HOW STUDENTS SLOVE PROBLEMS THAT HAPPEN AT SCHOOL. ON EACH AGE THERE IS A STORY, AND IN EACH STORY, YOU NEED TO PRETEND THAT THESE THINGS ARE HAPPENING TO YOU AT SCHOOL. THEN YOU HAVE TO DECIDE WHAT YOU WOULD DO IF THAT HAPPENED TO YOU. WE HAVE WRITTEN DOWN FIVE DIFFERENT THINGS THAT SOME STUDEN6S DO TO SOLVE EACH PROBLEM. AFTER YOU READ ALL FIVE OF THEM, YOU CAN PICK THE ONE YOU ARE MOST LIKELY TO DO (IF ALL THIS WAS REALLY HAPPENING TO YOU). THERE ARE NO RIGHT OR WRONG ANSWERS, JUST DIFFERENT WAYS DIFFERENT STUDENTS TRY TO SOLVE PROBLEMS AT SCHOOL. 64 STORY A You are putting together a puzzle during the lunch break. You have spent quite a lot of time on the puzzle and it is almost finished. A classmates comes over and grabs one of the remaining pieces. You ask him or her to give it back so that you can finish the puzzle, but he or she refuses to do that. 1. What are you going to do or say when this classmate refuses to give the puzzle piece back to you? Circle one: (a) I would ask him or her to play with me and finish the puzzle together. (b) I would just quit working on the puzzle. (c) I would grab the puzzle piece back. (d) I would go ask the teacher for help. (e) I would tell him or her that I need the puzzle piece and to give it back. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 65 STORY B You and your classmates are working on art assignments. The teacher gave each of you the job of cutting some pictures from a newspaper and gluing them on a paper to make up a story. A classmate in your group is cutting a picture from the newspaper. All of a sudden, you see him or her accidentally cut a finger and the finger is bleeding. What are you going to do or say when this classmate cuts his/her finger and the finger is bleeding? Circle one: (a) I would keep on doing my assignment. (b) I would ask if he or she needs the teacher. (c) I would ask another kid to take this classmate to the office for a band aid. (d) I would take him or her to the office for a band aid. (e) I would tell the teacher. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 66 STORY C The next hour is an art class. The teacher says everyone can choose a place around school and draw a picture of the school. After a while, you find a spot where you want to sit down and draw your picture. As you place your crayon and art pad on the ground, a classmate runs over and says to you, "Hey you can't sit here because I just decided I want this spot." You say, "Yes, I can." But this classmate says, "No you can't," and puts his/her crayon and art pad where you are planning to sit. 1. What are you going to do or say when this classmate again says you cannot sit there and puts his/her crayon and art pad where you are planning to sit? Circle one: (a) I would go ask the teacher to decide who should sit there. (b) I would tell the classmate to go somewhere to else or I would push his or her crayon and art pad away. (c) I would tell the classmate that I was there first and I want to sit there. (d) I would suggest sharing the place so that both of us can draw in the same place. (e) I would go find another spot. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 67 STORY D You have been sick and the doctor told you not to play outside. However, during the lunch hour a classmate comes up to you and asks if you could play dodge ball because they need one more person. What are you going to do or say when this classmate invites you to play dodge ball? Circle one: (a) I would tell him or her that I cannot play because I have a cold. (b) I would say "Yes" and play dodge ball with them. (c) I would just walk away. (d) I would ask this classmate to find someone else to play with them. (e) I would ask the teacher if I can go out to play. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 68 STORY E Your class is playing the last inning of a kickball game. You kick a home run, but as you score, a classmate on the opposite team says that you are out because you didn't touch second base and that his/her team wins. You tell this classmate you did touch second base, but he or she keeps saying that you didn't. What are you going to do or say when this classmate says again that you are out? Circle one: (a) I would ask the teacher to help out. (b) I would suggest that I take my turn again and do it over. (c) I would give up arguing even though I knew he or she is wrong. (d) I would tell this classmate once again that I did touch second base and that my team wins. (e) I would tell this classmate that he or she is just a bad sport and to shut up. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 69 STORY F Your teacher tells the class it is now free reading time. Everyone chooses a book to read from the bookshelf. You go to the bookshelf and choose a book that you really want to read. Then a classmate comes over and tells you to give him/her the book because he/she wants to read it first. As you are going back to your desk, this classmate blocks your way and does not let you go. 1. What are you going to do or say when this classmate blocks you way and does not let you go? Circle one: (a) I would tell this classmate to let me get to my desk because I took the book first and I want to read it first. (b) I would give the book to him or her although I don't want to. (c) I would tell this classmate that we can both read the book at the same time. (d) I would ask the teacher to decide who should read the book. (e) I would tell him/her to get out of my way or I would push him/her away. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 70 STORY G Someone in your class brings a computer game to the class. The game is really exciting and everyone likes it. During the lunch hour, the teacher says that everyone can take turns to play the computer game. You are the last person and it is now you turn, but a classmate comes over and says that he/she wants to play the game once again. You tell this classmate that it is your turn, but this classmate says, "I want to play one more time." 1. What are you going to do or say when this classmate again says he/she wants to play the computer game? Circle one: (a) I would tell him/ her to shut up and go away. (b) I would still say that it is my turn and I want to play it. (c) I would ask the teacher to help out. (d) I would just quit the game and let the classmate play. (e) I would tell this classmate that he/she can have another turn when I am done. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 71 STORY H It is lunch time. You place your lunch box on a table where several of your friends are sitting, because you like to have lunch with them. You walk away for a while to get something to drink. Later when you come back, a classmate is sitting at your seat and your lunch box has been put aside. What are you going to do or say when this classmate is sitting in your seat and your lunch box has been put aside? Circle one: (a) I would tell or yell at him/her to get out of my seat. (b) I would just pick another seat somewhere else. (c) I would ask this classmate to move over a little bit so that we both can sit together. (d) I would ask a lunch room person to help out. (e) I would tell him/ her that I was sitting there and ask him/her to take another seat. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 72 STORY I You and several classmates are chatting with each other in the hallway. A classmate runs by and accidentally trips on something on the ground. The other classmates begin to lauygh at this classmate and this classmate seems to be very upset. What are you going to do or say when you see this classmate become upset? Circle one: (a) I would say something nice to make this classmate feel better. (b) I would tell the other classmates to stop laughing. (c) I would give this classmate a hand and help them get up. (d) I would ask the teacher to come for a look. (e) I would walk away. REMEMBER THERE ARE NO RIGHT OR WRONG ANSWERS DO NOT TURN THE PAGE 73 APPENDIX C Loneliness and Social Dissatisfaction Questionnaire (Asher & Wheeler, 1985). HOW I FEEL On the next few pages are some saying which may be true or not about you. Read each sentence and decide whether or not you agree with the saying. C I R C L E T H E BIG " Y E S " IF Y O U F E E L T H A T T H E S E N T E N C E IS R E A L L Y T R U E F O R Y O U . C I R C L E T H E L ITTLE "yes" IF Y O U F E E L T H A T T H E S E N T E N C E IS S O R T O F T R U E F O R Y O U . C I R C L E T H E "sometimes" IF Y O U F E E L IN B E T W E E N . T H A T IS S E N T E N C E IS P A R T L Y T R U E F O R Y O U B U T P A R T L Y N O T T R U E F O R Y O U . C I R C L E T H E L ITTLE "no" IF Y O U F E E L T H A T T H E S E N T E N C E IS S O R T O F N O T T R U E F O R Y O U . C I R C L E T H E BIG " N O " IF Y O U F E E L T H A T T H E S E N T E N C E IS N O T A T A L L T R U E F O R Y O U . H E R E A R E A F E W E X A M P L E S T O TRY: A. I like roller blading. Y E S yes sometimes no N O B. I don't like going to the movies. Y E S yes sometimes no N O C. I like tO do homework. Y E S yes sometimes no N O D. I don't like to ride bikes. Y E S yes sometimes no N O THERE ARE NO RIGHT OR WRONG ANSWERS, JUST WHAT YOU THINK. PLEASE BE HONEST. YOUR NAME WILL NEVER APPEAR ON THIS FORM. 1. It's easy for me to make new friends at school. Y E S yes sometimes no N O 2. I have nobody to talk to. Y E S yes sometimes no N O 3. I'm good at working with other students. Y E S yes sometimes no N O 4. It's hard for me to make friends. Y E S yes sometimes no N O 5. I have lots of friends. Y E S yes sometimes no N O 6. I feel alone. Y E S yes sometimes no N O 7. I can find a friend when I need one. Y E S yes sometimes no N O 8. It's hard to get other students to like me. Y E S yes sometimes no N O 9. I don't have anyone to hang around with. Y E S yes sometimes no N O 10. I get along with other students. Y E S yes sometimes no N O 11 . I feel left out of things. Y E S yes sometimes no N O 74 12. There's nobody I can go to when I need help. YES yes sometimes no NO 13. I don't get along well with other students. YES yes sometimes no NO 14. I'm lonely. YES yes sometimes no NO 15. I am well liked by the students in my class. YES yes sometimes no NO 16. I don't have any friends. YES yes sometimes no NO 75 APPENDIX D Relational Provisions Loneliness Questionnaire (Hayden, 1989) F O R T H E FOLLOWING SAYINGS, THINK A B O U T Y O U R S E L F A N D P E O P L E Y O U R O W N A G E W H E N Y O U A N S W E R . 1. I feel part of a group of friends that do things together. Y E S 2. There is someone my age I can turn to. 3. I have a lot in common with other students. I. There is someone my age I could go to if I were feeling down. 5. I feel in tune with other students. 6. I have at least one really good friend I can talk to when something is bothering me. 7. I feel other students want to be with me. 8. I have a friend who is really interested in hearing about my private thoughts and feelings. 9. I feel that I usually fit in with other students around me. 10. I have a friend I can tell everything to. I I . When I want something to do for fun, I can usually find friends to join me. 12. There is somebody my age who really understands me. 13. When I am with other students, I feel like I belong. 14. There is a friend I feel close to. yes sometimes no N O Y E S yes sometimes no NO Y E S yes sometimes no NO Y E S yes sometimes no NO Y E S yes sometimes no NO Y E S yes sometimes no NO Y E S yes sometimes no N O Y E S yes sometimes no NO Y E S yes somet imes no NO Y E S yes sometimes no NO Y E S yes sometimes no NO Y E S yes somet imes no N O Y E S yes sometimes no NO Y E S yes sometimes no NO 76 APPENDIX E The Children's Self-Efficacy for Peer Interactions (Wheeler & Ladd, 1982). IS IT HARD OR EASY? In this questionnaire, we want to know how easy or hard it is for you to do things with your classmates. Each question tells you a situation where you have to do something with other kids. You can tell us how easy or hard it would be for you to do each one by circling on of the four words after each sentence. Circle the big word "HARD!" If it would be VERY difficult or hard for you to do. Circle the little word, "hard" if it would be sort of hard for you to do. Circle the little word, "easy" if it would be sort of hard for you to do. Circle the little word, "EASY!" if it would be sort of hard for you to do. Remember that there are no RIGHT or WRONG answers, just what is true for you. You can be honest because your answers are confidential and will not be shown to anyone. 1. Rnmfi kids want to play a game. Asking them if you can play is for you. HARD! Hard easy EASY! 2. Some kids are arguing about how to play a game. Telling them the rules is for you. HARD! Hard easy EASY! 3. Some kids are teasing your friend. Telling them to stop is for you. HARD! Hard easy EASY! 4. You want to start a game. Asking the other kids to play the game is for you. HARD! Hard easy EASY! 77 5. A kid tries to take your turn during a game. Telling the kid it's your turn is for you. HARD! Hard easy EASY! 6. Some kids are going to lunch. Asking if you can sit with them is for you. HARD! Hard easy EASY! 7. A kid cuts in front of you in line. Telling the kid not to cut is for you. HARD! Hard easy EASY! 8. A kid wants to do something tat will get you into trouble. Asking the kid to do something else is for you. HARD! Hard easy EASY! 9. Some kids are making fun of someone in your classroom. Telling them to stop is for you. HARD! Hard easy EASY! 10. Some kids need more people to be on their teams. Asking if you can be on a team is for you. HARD! Hard easy EASY! 11. You have to carry some things home after school. Asking another kid to help you is for you. HARD! Hard easy EASY! 12. A kid always wants to be first when you play a game. Telling the kid you are going first is for you. HARD! Hard easy EASY! 13. Your class is going on a trip and everyone needs a partner. Asking someone to be your partner is for you. HARD! Hard easy EASY! 14. A kid does not like your friend. Asking the kid to be nice to your friend is for you. 78 HARD! Hard easy EASY! 15. Some kids are deciding what game to play. Telling them about a game you like is for you. HARD! Hard easy EASY! 16. You are having fun playing a game but the other kids want to stop. Asking them to keep playing is for you. HARD! Hard easy EASY! 17. You are working on a project. Asking another kid to help is for you. HARD! Hard easy EASY! 18. Some kids are using your play area. Asking them to move is for you. HARD! Hard easy EASY! 19. Some kids are deciding what to do after school. Telling them what you want to do is for you. HARD! Hard easy EASY! 20. A group of kids wants to play a game that you like. Asking them to play a game that you like is for you. HARD! Hard easy EASY! 21. Some kids are planning a party. Asking them to invite your friend is for you. HARD! Hard easy EASY! 22. A kid is yelling at you. Telling the kid to stop is for you. HARD! Hard easy EASY! 79 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0089849/manifest

Comment

Related Items