n w The copyright of this thesis vests in the author. No o T quotation from it or information derived from it is to be published without full acknowledgeement of the source. p The thesis is to be used for private study or non- a C commercial research purposes only. f o Published by the Universit y of Cape Town (UCT) in terms y t of the non-exclusive license granted to UCT by the author. i s r e v i n U 1 Wechsler Abbreviated Scale of Intelligence: Preliminary normative data for 12-15-year-old English- and Afrikaans-speaking Coloured learners in the Western Cape Kaylee van Wyhe – VWYKAY001 n A dissertation submitted in fulfilment of the requirements for the award of the degree of w Master of Social Science in Psychology o T Faculty of Humanities e p University of Cape Town a 2012 C f o y t i s r e v i n COMPULSORY DEUCLARATION This work has not been previously submitted in whole, or in part, for the award of any degree. It is my own work. Each significant contribution to, and quotation in, this dissertation from the work, or works, of other people has been attributed, and has been cited and referenced. Signature: __________________________________ Date: __________________ 2 ACKNOWLEDGEMENTS Dr Kevin Thomas, University of Cape Town: The supervisor of this project. I would like to thank him for his guidance, constructive criticism, time, expertise and effort on this project. Dr Helen Ferrett, University of Stellenbosch: I would like to thank her for all her expert advice and encouragement throughout this project. I would like to thank the principals, teachers and administration staff of the participating schools for their hospitality and helpfulness during my time spent there. n w The participants and their parents: A heartfelt thanks for their willingness to participate in the o project. T e p I would also like to thank Ms Berenice Alexander and Ms Patsy Thompson for their much a appreciated assistance with data collection and recrCuitment. f o I thank the National Research Foundation for making this research initiative possible. y t i s My parents, family and friends: A sincere thanks to them for their unconditional support, r e encouragement and patience throughout this project. v i n U 3 TABLE OF CONTENTS ACKNOWLEDGEMENTS 2 LIST OF TABLES 6 ABSTRACT 8 n INTRODUCTION w 9 o Theories of intelligence T 10 e Spearman’s two-factor theory 11 p a Vernon’s hierarchical model of abilities 11 C Thurstone’s multiple factor theory 12 f o Guilford’s Structure of the Intellec t theory 12 y t Gardner’s theory of multiplei intelligences 13 s r e Sternberg’s triarchic theory of intelligence 13 v i Cattell-Horn-Canrroll theory 14 U Wechsler Intelligence Scales 16 WASI: Description and Utility 17 Intelligence Testing in South Africa 19 Importance of Normative Data 19 Normative data in South Africa 20 Cross-Cultural Aspects of Intelligence Testing 21 Cultural bias in intelligence testing 21 Language bias in intelligence testing 23 Understanding the Impact of Quality of Education on Intelligence Test Performance 24 The Impact of Socioeconomic Status on General Intellectual Functioning 27 4 Specific Aims and Hypotheses 30 METHODS 31 Research Design and Setting 31 Participants 32 Exclusion criteria 32 Materials 32 Demographic and socioeconomic questionnaire 32 Marin Acculturation Scale (MAS) 32 n Wechsler Abbreviated Scale of Intelligence (WASI) 33 w WASI-SA adaptation procedure 34 o Procedure T 34 Ethical Considerations e 35 p Data Management and Statistical Analyses 36 a C RESULTS 39 f Sample Characteristics o 40 Aim 1: Cross-lingual comparisons y 41 t Aim 2: Cross-cultural comparisons i 43 s r Aim 3: Demographic influences on WASI test performance 49 e v Hierarchical Regression Model: VIQ 49 i n Hierarchical Regression Model: PIQ 50 U Hierarchical Regression Model: FSIQ 52 Aim 4: Normative data, appropriately stratified 53 Normative Data: Case illustrations 61 DISCUSSION 65 Aim 1: Cross-lingual comparisons 66 Aim 2: Cross-cultural comparisons 69 5 Aim 3: Demographic influences on WASI test performance 74 Aim 4: Real-world significance and practical implications of the preliminary normative data 76 The association between race, language, and quality of education 77 Limitations and Recommendations for Future Research 79 Summary and Conclusion 80 REFERENCES 82 APPENDIX A: Demographic questionnaire and asset index 100 APPENDIX B: Marin Acculturation Scale 106 APPENDIX C: English and Afrikaans adapted versions of the WASI manuals 110 n APPENDIX D: Child and parent consent form 159 w APPENDIX E: o 161 T Table E1: Hierarchical Regression Model I: Diagnostic data – redundancy e coefficients p Figure E1. Normal probability plot of residualas for final regression model C APPENDIX F: f 162 o Table F1: Hierarchical Regression Model I: Diagnostic data – redundancy y coefficients t i s Figure F1. Normal probarbility plot of residuals for final regression model e v i APPENDIX G: n 163 U Table G1: Hierarchical Regression Model I: Diagnostic data – redundancy coefficients Figure G1. Normal probability plot of residuals for final regression model 6 LIST OF TABLES Table 1. Demographic Characteristics of the Current Sample 40 Table 2. Cross-Lingual Comparisons of WASI Performance 42 Table 3. Cross-Cultural Comparison I: Entire South African sample versus US normative sample 44 Table 4. Cross-Cultural Comparison II: First-language English n w Coloured South African adolescents with advantaged quality of education versus US o normative sample 45 T e Table 5. Cross-Cultural Comparison III: First-languagep English Coloured South African a adolescents with disadvantaged quality of education C versus US normative sample 46 f o Table 6. Cross-Cultural Comparison IVy: First-language Afrikaans Coloured South African t adolescents with advantaged qualityi of education s r versus US normative sample 47 e v i n Table 7. Cross-Cultural Comparison V: First-language Afrikaans Coloured South African U adolescents with disadvantaged quality of education versus US normative sample 48 Table 8. Hierarchical Regression Model I: Predictors of WASI VIQ test performance in the current sample 50 Table. 9. Hierarchical Regression Model II: Predictors of WASI PIQ test performance in the current sample 51 Table 10. Hierarchical Regression Model II: Predictors of WASI FSIQ test performance in the current sample 53 7 Table 11. WASI Normative Data for 12-year-old First-language English Coloured South African Adolescents with Advantaged and Disadvantaged Quality of Education 54 Table 12. WASI Normative Data for 12-year-old First-language Afrikaans Coloured South African Adolescents with Advantaged and Disadvantaged Quality of Education 55 Table 13. WASI Normative Data for 13-year-old First-language English Coloured South African Adolescents with Advantaged and Disadvantaged Quality of Education 56 n Table 14. WASI Normative Data for 13-year-old First-language Afrikaans Coloured South w African Adolescents with Advantaged and Disadvantaged Quality of Education 57 o T Table 15. WASI Normative Data for 14-year-old First-langueage English Coloured South p African Adolescents with Advantaged and Disadvantaged Quality of Education 58 a C Table 16. WASI Normative Data for 14-year-ol d First-language Afrikaans Coloured South f o African Adolescents with Advantaged and Disadvantaged Quality of Education 59 y t i s Table 17. WASI Normative Data for 15-year-old First-language English Coloured South r e African Adolescents with Advantaged and Disadvantaged Quality of Education 60 v i n Table 18. WASI NormUative Data for 15-year-old First-language Afrikaans Coloured South African Adolescents with Advantaged and Disadvantaged Quality of Education 61 Table 19. Score Conversions and Qualitative Classifications 62 Table 20. Case A: Interpretation of WASI test performance using local and non-local norms 63 Table 21. Case B: Interpretation of WASI test performance using local and non-local norms 64 8 ABSTRACT The Wechsler Abbreviated Scale of Intelligence (WASI) is an internationally recognized brief measure of general intellectual functioning. The WASI was developed, standardized, and normed in United States on a predominantly White sample. Despite this fact, and the fact that there are no South African norms for the instrument, it is used in a variety of clinical and research settings in this country. This situation is problematic, particularly because South Africa is an exceptionally diverse and multicultural country, and because the accurate interpretation of intelligence test performance relies on the use of appropriate normative data. The current study had two primary objectives. First, I set out to describe the development of a n culturally and linguistically adapted version of the WASI suitable for use in English and w Afrikaans first-language speakers. Second, I set out to establish, for that adapted instrument o and for English and Afrikaans first-language speakers aged 12 toT15 years and resident in the Western Cape, a preliminary set of locally appropriate normes, stratified by age, language, and p quality of education. This study provides valuable guidelines for collection and use of a normative data for research and clinical purposes inC South Africa. Keywords: Wechsler tests; intelligence tests; cu lture-fair; adolescents; quality of education; f o normative data. y t i s r e v i n U 9 INTRODUCTION The effects of socioeconomic status, education, culture, and ethnicity on intelligence test performance have been widely documented, and continue to present much controversy and debate (Cohen & Swerdlik, 2005; Owen, 1998; Pretorius et al., 2009). The administration of intelligence tests in a country like South Africa, where issues like cultural differences, educational disparities, and sociopolitical disadvantages are particularly relevant, presents numerous challenges to clinicians and researchers. During the apartheid era, the White minority imposed extreme racial division on the Black (African, Coloured, and Indian) majority.1 Apartheid policies meant that Black n individuals had to live and receive schooling separately from Whites; they were discriminated w against and were refused equal opportunities. Despite the end of apartheid, the many years of o discrimination and deprivation experienced by Black individualsT has left a complex legacy that is currently evident in almost all spheres of South Africean society (van der Berg, 2009). p Performance on intelligence tests is one of the spheres in which that legacy is evident. a Black people in South Africa routinely obtain signiCficantly lower test scores than European and American normative samples on intelligenc e tests (Anderson, 2001; Lynn & Owen, f o 1994; Shuttleworth-Edwards et al., 2004; Viljoen, 1992). For instance, Skuy, Schutte, y Fridjhon, and Carroll (2001) measured test performance on an English administration of the t i s Wechsler Intelligence Scale for Children-Revised (WISC-R; Wechsler, 1974) and the r e Individual Scales for African Language Speaking children (IS-A; Landman, 1991) in a v sample of healthy, neuroliogically intact South African Black African students with varying n home languages (N =U 152, age range 8-15 years). Their results demonstrated that, compared to published American norms, the South African students scored poorly (many in what might be described as the “impaired” range) on many of the subtests. This study illustrated the point that only locally-appropriate normative data should be used in interpreting performance on intelligence tests. Globally, the practice of psychological assessment (in clinical neuropsychology, for instance) is, to an important degree, dependent upon the development of appropriate normative data. The validity of test results is compromised when inappropriate norms are applied; assessment is, in this case, effectively rendered meaningless. In South Africa, almost all popular and commonly-used psychological tests were developed in English-speaking 1Throughout this thesis, I will use the term “Black” in the generic sense (i.e., to refer collectively to Black African, Coloured, and Indian individuals).