PERFORMANCE OF GHANAIAN SCHOOL CHILDREN ON THE KAUFMAN ASSESSMENT BATTERY FOR CHILDREN – SECOND EDITION (KABC-II) – AN EXPLORATORY VALIDATION STUDY BY DEIRDRE SALLY QUARTEY 10175263 THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON, IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MASTER OF PHILOSOPHY IN PSYCHOLOGY DEGREE JULY, 2014 University of Ghana http://ugspace.ug.edu.gh i . DECLARATION This is to certify that this thesis is the result of research carried out by Deirdre Sally Quartey under supervision, towards the award of the Master of Philosophy in Clinical Psychology Degree in the University of Ghana, Legon. ………………………… DEIRDRE SALLY QUARTEY (STUDENT) Date………………………… ………………………… PROF. C. C. MATE-KOLE (PRINCIPAL SUPERVISOR) Date……………………….. ………………………… DR. A. ANUM (CO-SUPERVISOR) Date………………………… University of Ghana http://ugspace.ug.edu.gh ii . ABSTRACT The Kaufman Assessment Battery for Children-second Edition (KABC-II) is a foreign psychological test which is increasingly being employed by practitioners in Ghana in the assessment of intelligence and cognitive abilities in children although it has not been locally normed as yet. Thus, the performance of 90 school children in Ghana aged between 6 and 12 years, whose sociocultural backgrounds may differ considerably from those for whom the KABC-II were normed, was investigated in this study. Children aged between 10 years and 12 years 11months performed significantly better than those aged between 6 years and 7 years 11months on all five subscales but only on two subscales when the former age group was compared to those aged between 8 to 9 years 11 months. Children from private schools performed significantly better than those from public schools on all five subscales. On the global indexes, those from private schools performed better on the FCI than the MPI whereas the converse was true for children in public schools. The findings of the study suggest a suitability of the KABC-II as an assessment tool in Ghana. However, the socioeconomic background of the child being examined must be taken into consideration. University of Ghana http://ugspace.ug.edu.gh iii . DEDICATION This thesis is dedicated to my family. Thank you for loving me. University of Ghana http://ugspace.ug.edu.gh iv . ACKNOWLEDGEMENTS To God be all the glory for how far He has brought us. My utmost gratitude goes to the Almighty for the His neverending grace and faithfulness towards me. My immense gratitude to my supervisors, Professor C.C. Mate-Kole for being with me every step of the way, and Dr. A. Anum for his unbelievable patience for me. I could never have made it without their help. I also thank all my lecturers at the Department of Psychology for the knowledge they have imparted to me. Next, I acknowledge the Ethical Clearance Board of the Noguchi Memorial Institute for Medical Research (NMIMR), the Greater Accra Regional Dierctoe for the Ghana Education Service, the various heads of schools, consenting parents and guardians and the wonderful school-children who so willingly participated in the study. Of course, I cannot forget ―my friends‖ who assisted me in diverse ways to get this research done. God richly bless you all. Finally, I would like to thank my family for their love, support, encouragement and immeasurable sacrifice, just to see me through. I love you. And God bless you for being the ―wind beneath my wings‖. University of Ghana http://ugspace.ug.edu.gh v . TABLE OF CONTENTS Content Page DECLARATION ............................................................................................................................. i ABSTRACT .................................................................................................................................... ii DEDICATION ............................................................................................................................... iii ACKNOWLEDGEMENTS ........................................................................................................... iv TABLE OF CONTENTS ................................................................................................................ v LIST OF TABLES ....................................................................................................................... viii LIST OF ABBREVIATIONS ......................................................................................................... x CHAPTER ONE ............................................................................................................................. 1 INTRODUCTION .......................................................................................................................... 1 1.1 Definition of Intelligence .......................................................................................................... 1 1.2 Concepts of Intelligence ........................................................................................................... 1 1.3 History of Psychological Testing .............................................................................................. 5 1.4 Uses of Psychological tests ..................................................................................................... 10 1.5 Statement of the Problem ........................................................................................................ 11 1.6 Aims/Objectives of the study: ................................................................................................. 12 CHAPTER TWO .......................................................................................................................... 13 LITERATURE REVIEW ............................................................................................................. 13 2.1 General issues in cognitive testing .......................................................................................... 14 2.2 Cultural issues in cognitive testing ......................................................................................... 18 2.3 Language and intelligence ...................................................................................................... 21 University of Ghana http://ugspace.ug.edu.gh vi . 2.4 Sex differences in intelligence ................................................................................................ 23 2.5 Socioeconomic status, parental education and intelligence .................................................... 24 2.6 The Kaufman Assessment Battery for Children- Second Edition (KABC-II) ....................... 29 2.6.1 Standardization Issues of the KABC-II ............................................................................... 32 2.6.2 Reliability and Validity Issues of the KABC-II ................................................................... 33 2.7 Rationale for the study ............................................................................................................ 35 2.8 Statement of hypotheses ......................................................................................................... 36 2.9 Operational Definitions:.......................................................................................................... 37 CHAPTER THREE ...................................................................................................................... 39 METHODOLOGY ....................................................................................................................... 39 3.1 Study Population ..................................................................................................................... 39 3.2 Study design ............................................................................................................................ 39 3.3 School selection ...................................................................................................................... 39 3.4 School type.............................................................................................................................. 40 3.5 Sample size and sampling technique ...................................................................................... 40 3.6 Exclusion criteria .................................................................................................................... 42 3.7 Instruments .............................................................................................................................. 42 3.7.1 The Kaufman‘s Assessment Battery for Children, Second Edition (KABC-II) .................. 42 3.8 Procedure ................................................................................................................................ 46 3.9 Data Management and Analysis ............................................................................................. 47 University of Ghana http://ugspace.ug.edu.gh vii . CHAPTER FOUR ......................................................................................................................... 48 RESULTS ..................................................................................................................................... 48 4.1 Summary of Key Findings ...................................................................................................... 78 CHAPTER FIVE .......................................................................................................................... 79 DISCUSSION ............................................................................................................................... 79 5.1 Age Differences and Performance on KABC-II Scales .......................................................... 79 5.2 Type of School and Performance on the KABC-II Scales ..................................................... 81 5.3 Sex Differences in performance on KABC-II Scales ............................................................. 84 5. 4 The interaction effects of the independent variables (Age, Type of School and Sex) on the KABC-II scales ................................................................................................................. 86 5.5 Implications of the study ......................................................................................................... 88 5.6 Limitations and Recommendations......................................................................................... 89 5.7 Conclusions ............................................................................................................................. 90 REFERENCES ............................................................................................................................. 91 APPENDICES ............................................................................................................................ 109 Appendix I: Ethical Clearance from Noguchi Memorial Institute for Medical Research (NMIMR) ........................................................................................................................ 109 Appendix II: Consent Form ........................................................................................................ 110 Appendix III: Departmental Introductory Letter ........................................................................ 111 APPENDIX IV: KABC-II .......................................................................................................... 112 University of Ghana http://ugspace.ug.edu.gh viii . LIST OF TABLES Table 1: Frequency and Percentages of School- children in Private and Public Schools .......... 41 Table 2: Descriptive Statistics and Internal Consistencies of KABC-II Scales and Subtests for 6-7 years 11 months .................................................................................................... 49 Table 3: Descriptive Statistics and Internal Consistencies of KABC-II Scales and Subtests for 8-9 years 11 months .................................................................................................... 51 Table 4: Descriptive Statistics and Internal Consistencies of KABC-II Scales and Subtests for 10-12 years 11 months ................................................................................................ 54 Table 5: Descriptive Statistics of Age Groups, School Type and Sex on Learning/Glr ............ 57 Table 6: Three-Way ANOVA of Age Groups, School Type and Sex on Learning/Glr ............ 58 Table 7: Multiple Comparisons of Age Groups on their Performance in Learning/Glr ............ 60 Table 8: Descriptive Statistics of Age Groups, School Type and Sex on Sequential/Gsm ....... 61 Table 9: Three-Way ANOVA of Age Groups, School Type and Sex on Sequential /Gsm ...... 62 Table 10: Multiple Comparisons of Age Groups on their Performance in Sequential/Gsm ...... 64 Table 11: Descriptive Statistics of Age Groups, School Type and Sex on Simultaneous/Gv ... 65 Table 12: Three-Way ANOVA of Age Groups, School Type and Sex on Simultaneous Processing ................................................................................................................... 66 Table 13: Multiple Comparisons of Age Groups on their Performance in Simultaneous/Gv ... 68 Table 14: Descriptive Statistics of Age Groups, School Type and Sex on Planning ................. 69 Table 15: Three-Way ANOVA of Age Groups, School Type and Sex on Planning/Gf ........... 70 Table 16: Multiple Comparisons of Age Groups on their Performance in Planning/Gf ............ 72 Table 17: Descriptive Statistics of Age Groups, School Type and Sex on Knowledge/Gc ...... 73 Table 18: Three-Way ANOVA of Age Groups, School Type and Sex on Knowledge/Gc ....... 74 University of Ghana http://ugspace.ug.edu.gh ix . Table 19: Multiple Comparisons of Age Groups on their Performance in Knowledge/Gc ....... 76 Table 20: Paired sampled t-test for differences in performance on MPI and FCI among private school children ............................................................................................................ 77 Table 21: Paired sampled t-test for differences in performance on MPI and FCI among public school children ............................................................................................................ 77 University of Ghana http://ugspace.ug.edu.gh x . LIST OF ABBREVIATIONS APA - American Psychological Association CHC - Cattell-Horn-Carroll FCI -Fluid Crystallized Index IQ -Intelligence Quotient KABC-II - Kaufman Assessment Battery for Children, Second edition MOESS -Ministry of Education, Science And Sports, Ghana MPI -Mental Processing Index NVI -Non Verbal Index PASS - Planning, Attention-Arousal, Simultaneous and Successive theory RCPM - Raven‘s Coloured Progressive Matrices SES - Socioeconomic status WISC IV -Wechsler Intelligence Scale for Children, Fourth edition WISC IV FSIQ – Wechsler Intelligence Scale for Children, Fourth edition, Full-Scale Intelligence Quotient University of Ghana http://ugspace.ug.edu.gh 1 . CHAPTER ONE INTRODUCTION ―...You cannot take a person who for years has been hobbled by chains, bring him up to the starting line of a race and say – you are free to compete with us – and truly believe that you are treating him fairly.‖ Lyndon Johnson (as cited in De Beer, 2000, p.1) 1.1 Definition of Intelligence Although ―intelligence‖ is one of the most talked about subjects within Psychology, despite a long history of research and debate, there is still no standard definition of what exactly constitutes ―intelligence‖ (Legg & Hutter, 2007). According to Sternberg (Gregory, 1998), ―Viewed narrowly, there seem to be almost as many definitions of intelligence as there were experts asked to define it.‖ Indeed, intelligence has been defined differently in different epochs, ranging from Pythagoras‘ depiction of it as ―winds‖, to Descartes‘ definition that it is the ability to distinguish true from false (Salovey & Mayer, 1989). The most often cited definition then, perhaps, is Wechsler‘s description that it is ―the global capacity of a person to act purposefully, to think rationally, and to deal effectively with his environment‖ (Wechsler, 1944). 1.2 Concepts of Intelligence Following the intelligence debate in America, sparked by the publication of Herrnstein and Murray‘s book, The Bell Curve, in 1994, the Task Force established by the Board of Scientific Affairs (BSA) of the American Psychological Association (APA) to review the matter observed that although individuals differ from one another in their ability to understand complex ideas, adapt effectively to the environment, reason and learn from experience, these differences, substantial at times, are never entirely consistent (Plucker, 1996). Indeed, it is even unlikely that University of Ghana http://ugspace.ug.edu.gh 2 . an individual‘s intellectual performance will be the same under different circumstances or in different settings. This complex and inexplicable phenomenon is what has been aptly described as concepts of ―Intelligence‖. Some theorists have argued that there are multiple forms of intelligence, some of which cannot be accurately measured by psychometric means. Intelligence, others have noted, is influenced by an individual‘s biological and neural make-up, their developmental progressions, the environment and culture (Plucker, 2013). This has led to the postulation of various theories on the concepts of intelligence. Notable among these are Spearman‘s Two-Factor theory, Sternberg‘s Triarchic theory, Catell and Horn‘s Fluid and Crystallized Intelligence and Carroll‘s Three Stratum Theory. Spearman concluded that intellectual abilities are made up of two components, namely, the general intelligence or g-factor and the collection of specific cognitive intellectual skills known as the s-factor. According to Spearman, the g-factor is the ―leading part of intelligence‖ and consists of three processes. The first process, ―the apprehension of experience‖ consists of an individual‘s cache of experiences which he draws upon to solve problems. The second process, which he called the ―eduction of relationships‖, is the ability to draw out a logical relationship between two stimuli. The ―eduction of correlates,‖ which is the ability to note the similarities between two stimuli, forms the third process (Kane & Brand, 2003). Cattell‘s theory proposes that human intelligence is composed of two separate factors – fluid intelligence and crystallized intelligence. The former reflects an individual‘s basic reasoning abilities, and the latter, their knowledge from experiences. The acquisition and expansion of crystallized intelligence depends, in part, on fluid intelligence in that one needs that innate ability University of Ghana http://ugspace.ug.edu.gh 3 . of comprehension to assimilate what we acquire through education, experience and the environment (Kane & Brand, 2003). Another theory of intelligence, propounded by Carroll (Carroll, 1997) is one of a hierarchical nature comprising three strata of abilities. The first stratum is made up of narrow abilities and reflects the individual‘s specific experiences, learning, and strategies. These also include abilities such as length estimation and meaningful memory, among others. The second stratum of broad abilities represents some specialization of abilities and established traits. It consists of eight factors, including fluid intelligence, crystallized intelligence, general memory and learning, broad visual perception, broad auditory perception, broad retrieval ability, broad cognitive speediness, and processing speed. The third stratum, a single ―general‖ intellectual ability, is similar to Spearman‘s ‗g‘ (Kane & Brand, 2003; Carroll, 1997). In contrast to this, the Planning, Attention-Arousal, Simultaneous and Successive (PASS) theory of intelligence, first proposed by Das, Kirby and Jarman (1975) and later elaborated by Das, Naglieri and Kirby (1994), and Das, Kar and Parrila (1996), opposes the g-factor theory on the basis that the brain is made up of interdependent but separate functional systems. Following largely from the neuropsychological work of Luria (1973) on the modularization of brain function, and supported by decades of neuroimaging research, the PASS Theory of Intelligence proposes that cognition is organized in three systems and four basic psychological processes. The first process, planning, involves executive functions which are responsible for the control and organization of behaviour, the selection and construction of strategies, and performance monitoring. The second, the attention process, is concerned with the maintenance of arousal levels and alertness, and ensuring focus on relevant stimuli. The third comprise two processes, University of Ghana http://ugspace.ug.edu.gh 4 . simultaneous and successive processing which encode, transform, and retain information. The former is employed when the relationship between items and their integration into whole units of information is required and the latter, when organizing separate items in a sequence such as remembering a specific order of words or actions (Das et al., 1994). In spite of, or probably, because of all the difficulties in defining it clearly, interest in intelligence dates back thousands of years, and the classification and categorization of individuals on the basis of their intelligence is by no means a novel concept (Perlman & Kaufman, 2000). Perhaps efforts to address the variations in the concept of intelligence and to measure it accurately, have led to the development and employment of thousands of psychological tests. Research has revealed that performance in these psychological tests is influenced by myriads of factors including subject characteristics such as age, gender, race, ethnicity, culture, socioeconomic status, language, education and test taking attitudes, among others. Thus, when assessing individuals from socioeconomic, educational, cultural and language backgrounds that are different from those of the sample for which a test was normed, great care must be taken in order to avoid making diagnoses and claims that are biased or incorrect (Van der Merwe, 2008); a classic example being the case of the new immigrants as they entered the United States of America through Ellis Island at the end of World War II (Richardson, 2011). Another concept of intelligence, worthy of note is emotional intelligence. This is essentially a type of social intelligence that is concerned with the ability to evaluate one's own as well as other people‘s emotions. Awareness of these emotions and the interplay among them, according to Salovey and Mayer (1990), plays an important role in the determination of one‘s thoughts or University of Ghana http://ugspace.ug.edu.gh 5 . actions when solving problems. Despite its usefulness, several criticisms about emotional intelligence have been put forth; some authorities deem it as a form of interpersonal cordiality rather than ability and therefore find it inappropriate that it be labeled as a kind of intelligence. Others see no relation between intelligence and emotion, perceiving them as two different entities; the former, controversial and the latter, not. (Mayer & Salovey, 1993). Lending support to the existence of emotional intelligence and its correlation to IQ, Barbey, Colom and Grafman (2012) found that IQ and conscientiousness significantly predicted emotional intelligence. They were also able to identify and map out the neural basis of general intelligence and emotional intelligence. A significant overlap between general intelligence and emotional intelligence, both in brain activity and behaviour was also noted. Higher scores on general intelligence tests were found to reliably predict higher performance on measures of emotional intelligence; with similar regions of the brain responsible for both. Another phenomenon of intelligence worthy of note is the Flynn Effect, the significant and persistent increase in both fluid and crystallized intelligence test scores noted globally from one generation to the next in the past century or so. These increases have however been higher for tests of fluid intelligence than for crystallized intelligence (Flynn, 1994). Plausible explanations that have been advanced for the Flynn Effect include better nutrition (Storfer, 1990), societal changes (Brand, 1987a, 1987b), more and better education, enhanced socioeconomic status, urbanization and the advent of television, among others (Flynn, 1998c). 1.3 History of Psychological Testing The roots of testing, as observed by Anastasi (1976), are lost in antiquity but perhaps, the first recordings of the use of an assessment procedure for selection are found in the Bible in Judges University of Ghana http://ugspace.ug.edu.gh 6 . chapter 7 versus 1 to 8, whereby Gideon observed his soldiers drinking from a river so he could select those who remained on the alert (Bowman, 1989). Aside that, some scholars believe that much credit can also be given to the Chinese for their more sophisticated testing programmes dating as far back as 4000 years to 2200 B.C., when the emperor Yushun had his officials orally examined every third year to determine their fitness for office and promotion (Gregory, 1992). Unfortunately, there is little archeological evidence to support these claims since despite the development of elaborate writing systems in 1115 BCE, no inscriptions or writings have yet been found to suggest the existence of such an examination process. Modern scholars of ancient China however agree that royal examinations most probably began around 200 years to 100 BCE, during the late Qin and early Han Dynasties (Franke, 1960; Rodzinski, 1979) with the emperor administering the first ever written examinations to nominees for royal office in 165 BCE ( Hucker, 1978). They also admit that there may have been some assessment procedures for selecting officials prior to this period. This form of testing was further modified and refined during the Han Dynasty (206 B. C. E. - 220 C. E.), with the use of test batteries, comprising diverse topics such as agriculture, civil law, geography, military affairs and revenue becoming quite common. Subsequent development and modification of these tests during the Ming Dynasty (1368 – 1644 CE) led to national multistage testing programmes involving specially- equipped local and regional testing centres. Passing each level of testing accorded the subject more power and increased eligibility for office in the civil service (Kaplan & Saccuzzo, 2005). These examinations though tough and distressful, seemed to work well and ensured that people became part of the national government on account of their knowledge and talent, and not their descent or bloodlines. The successes of this system University of Ghana http://ugspace.ug.edu.gh 7 . encouraged the Europeans and the Western world to copy and adopt it in the 19th century (Kaplan & Saccuzzo, 2005). Kaplan and Saccuzzo (2005) believe that although humans realized long ago that individuals differ in so many ways, developing the appropriate tools for measuring these differences has never been an easy feat. This prompted the pioneering experiments and proposition of theories by Charles Darwin, Sir Francis Galton and James McKeen Catell, among others. According to Darwin‘s theory, higher forms of life evolved partially because of the differences among individuals within the same species; that is, some characteristics are more adaptive than others in a particular environment. He further argued that those individuals with the most adaptive characteristics tend to survive at the expense of those who are less fit. These so- called superior adaptive characteristics are purportedly passed down from one generation to the next, leading to the evolution of currently complex and intelligent levels. To further validate this concept of survival of the fittest, Galton went on to demonstrate that individuals actually do differ in sensory and motor functioning, such as reaction time, visual acuity and physical strength. Galton‘s work was further extended by Catell who set in motion the processes that eventually led to the development of modern psychological tests. Indeed, Galton is attributed with being the first scientist to discover that psychological variables were normally distributed much in the same way as physical variables (Kaplan & Saccuzzo, 2005) A second line of inquiry which sought to demystify human consciousness using scientific methods also developed, based on the work of German psychophysicists J. E. Herbart, E. H. Weber, G.T. Fechner and Wilhelm Wundt. Their results of their work proffered the notion that psychological testing, like all other experiments require rigorous experimental control which can University of Ghana http://ugspace.ug.edu.gh 8 . only be achieved by ensuring that tests are administered under highly standardized conditions. Subsequently, Wundt who founded the first formal laboratory for psychological research at the University of Leipzig in Germany in 1879, became widely known as the ―Father of Experimental Psychology‖ (Kaplan & Saccuzzo, 2005). At the turn of the twentieth century however, a significant breakthrough in the development of modern psychological tests occurred. In 1904, the French minister of public instruction appointed a commission to identify, evaluate and make appropriate recommendations for the educational placement of intellectually subnormal children. Alfred Binet, a psychologist and member of that commission, working in collaboration with the French physician, Theodore Simon, developed the first major general intelligent test, the Binet - Simon scale. The needs of society drive innovation and innovations in psychological or intelligence test development throughout the past century reflect the social milieu in which they occurred (Wechsler, 2003). Following several modifications to the Simon-Binet Scale in the early 1900s, Terman and his colleagues at Stanford University developed the well-standardized Stanford- Binet scale for the United States. Terman also coined the term I.Q. (intelligence quotient) which was calculated by dividing the mental age by the chronological age and multiplying the quotient by 100. It is of worthy of note that although the methodology for test development was improving all the time, identification of intellectually subnormal people, remained the primary focus of intelligence testing. During World War I, the Army Alpha, a largely verbal test developed by Robert Yerkes, Lewis Terman and Henry Goddard among others, was used to screen recruits for the United States Army. Based on verbal and numerical abilities, ability to follow directions and knowledge University of Ghana http://ugspace.ug.edu.gh 9 . information, the Army Alpha determined an individual‘s capability of serving, job classification and potential for leadership positions. The biases of this test against recruits with limited literacy led to the development of the Army Beta which could give a non-verbal measure of intelligence and was thus deemed a fairer assessment of recruits. The development of the Wonderlic Personnel Test by E. F. Wonderlic in 1936 as the first short- term cognitive abilities test was another landmark in the history of intelligence testing. Known to measure the general cognitive abilities in vocabulary, mathematics and reasoning, it was initially used as vocational and intelligence test in the field of Industrial and Organizational Psychology in employee selection, assessing their job potential, educational potential and training potential. It served as an important tool in the selection of candidates within the US navy for training as pilots and navigators during World War II. With the expansion of the special education system in America in the 1950s, there arose a need for the identification and diagnosis of the nature of learning disabilities in children. It was at this stage that intelligence testing began to focus on measuring more discrete aspects of an individual‘s cognitive functioning (Kaplan & Saccuzzo, 2005). The next major development in the history of intelligence testing was the creation of Wechsler‘s Intelligence Scales, different versions of which are used in testing adults and children. These tests were the first intelligence scales to base their scores on a standardized normal distribution rather than an age-based quotient, as was the case with the original Stanford-Binet test. In recent years, however, the Cattell-Horn-Carroll (CHC) theory, a psychological theory of human cognitive abilities derived from the integrated works of Raymond Cattell, John L. Horn and John Bissell Carroll, has emerged as the most comprehensive and empirically supported University of Ghana http://ugspace.ug.edu.gh 10 . psychometric theory structure of cognitive and academic abilities to date (McGrew, 2005). The Cattell-Horn-Carroll (CHC) theory which evolved from extensive research on human cognitive abilities over a number of decades using factor analysis model, basically conceptualizes intelligence as a collection of distinct cognitive abilities, in which intellectual growth is dependent on innate abilities, experiences, environment and emerging abilities (Lynch & Warner, 2012). The theory proposes that the relationships among these cognitive abilities can be deduced by classifying them into three different strata: stratum I, "narrow" abilities; stratum II, "broad abilities"; and stratum III, a single "general" ability (or g). The nine broad stratum abilities are divided to include more than 70 narrow abilities (Flanagan, Ortiz & Alfonso, 2007). The Cattell-Horn-Carroll (CHC) model is increasingly being employed as the basis for the selection, organization and interpretation of many tests of intelligence and cognitive abilities (Flanagan, Alfonso, & Ortiz, 2012). 1.4 Uses of Psychological tests The fundamental function of psychological tests, according to Anastasi (1990) is to measure differences between individuals or between the reactions of the same individual on different occasions. In other words, they are used in the evaluation of behaviour, estimation of cognitive abilities and prediction of personality traits. In clinical practice, these tests enhance the process of psychological assessment by providing supplementary information which helps in the diagnoses and management of clients. In education, psychological tests are used in the identification, classification, evaluation and placement of the intellectually gifted, the intellectually deficient and increasingly, in the diagnosis of academic failures. They are also used in the evaluation of teaching methods and the University of Ghana http://ugspace.ug.edu.gh 11 . appraisal of curricula. In the colleges, high schools and universities, such tests help in vocational and educational counseling and in the selection of applicants for special professional programmes (Anastasi, 1990). In many industries and organizations, psychological testing serves as an adjunct to interviewing in the selection of personnel for employment, job assignment, retention, promotion, demotion, transfer or termination of appointments. Some psychological tests also serve as vital tools in an employer‘s arsenal for determining employees‘ honesty and integrity (Seagull & Caputo, 2006). Another important area which can simply not be overlooked is the use of psychological tests in research. With the aid of these tests, studies on the nature and extent of individual and group differences, the identification and influence of psychological traits on lifestyle, and the investigation of biological as well as sociocultural factors associated with behavioural differences have all been made possible (Anastasi, 1990; Hanes & Norli, 2011; Schooler, 1996). 1.5 Statement of the Problem Perusal of existing literature and a search of the internet did not yield any documented psychological test indigenously developed in Ghana. As such, most psychological tests in use in Ghana have to be necessarily imported from other foreign countries where such as the United States of America, United Kingdom, and Germany among others. In Ghana, some of the psychological tests which were developed and standardized in other foreign countries have already been adapted and standardized for Ghanaians. These include the Raven‘s Coloured Progressive Matrices (RCPM) for Children (Anum, 1996), the Wechsler Intelligence Scale for Children-Third edition (WISC-IIIUK) by Edwin (2001) and the Multidimensional Aptitude Battery (MAB) by Debra (2002), among others. These previous standardizations however, have University of Ghana http://ugspace.ug.edu.gh 12 . revealed marked differences in performance of older versus younger children, urban versus rural children, high socioeconomic status (SES) versus low SES background and public versus private school children. In Anum‘s work with the RCPM in Ghana for instance, he observed highly significant differences between the performance of urban and rural children with the former doing much better than their rural counterparts. This finding confirmed the relevance of socioeconomic status (SES) variables in determining performance on psychological tests. Further evidence proved that children from different SES and cultural backgrounds were likely to perform differently on the test with the economically disadvantaged group performing more poorly. Edwin (2001) obtained similar results in her work with the WISC –IV in Ghana, with children from lower SES backgrounds in both urban and rural areas scoring lower on both arithmetic and spelling components of the WISC-IV. As such aside establishing local norms for any psychological test in Ghana, it is important to also test for differences in performances between these two populations. This study is thus an attempt to not only establish local norms for the KABC-II in Ghana, but to investigate the differences in performance of children in private and public schools on the test as well. 1.6 Aims/Objectives of the study: The aim of this study is to investigate the performance of Ghanaian school-children on the KABC-II. The various objectives to be considered are:  To determine whether there are significant age differences on test performance  To determine the influence of type of school attended on test performance.  To find out the differences in performance between boys and girls on the KABC-II. University of Ghana http://ugspace.ug.edu.gh 13 . CHAPTER TWO LITERATURE REVIEW Research addressing controversial topics related to the performance of different groups in cognitive tests has a long history (Manly & Echemendia, 2007). Klineberg‘s study of the test data from World War I recruits in the early 20th century (Klineberg, 1928) demonstrated that contrary to popular belief that African Americans and immigrants were intellectually inferior, a group of African American recruits educated in Northern states performed better on the Army Alpha and Beta tests than their Caucasian counterparts who had been born and bred in Southern states. Although the question regarding the appropriateness of the use of tests of intelligence and cognitive abilities developed in the United States to assess people from other countries has been a topic for debate in many circles, versions of these tests which are adapted to measure intelligence in these non- Western cultural contexts still have the potential to provide useful information about the test taker and should not be written off (Sánchez-Escobedo & Hollingworth, 2009; Holding, Taylor, Kazungu, Mkala, Gona, Mwanye, Mbonani & Stevenson, 2004). Ideally, psychological assessments ought to be based on emic testing which doubtlessly has to do with issues pertaining to a particular culture. However, the development of psychological tests requires much time and finances. In lieu of that, etic tests can still be used after rendering them less unfair by norming the test results on a particular culture or redesigning the test so that it equally applies to all cultures. Many however believe that no etic test can be that widely applicable and yet, fully fair (Mitrushina, Boone, Razani, & D'Elia, 2005; Van der Merwe, 2008; Bangirana, Seggane-Musisi, Allebeck, Giordani, John, Opoka, Byarugaba, Ehnvall & Boivine, 2009). University of Ghana http://ugspace.ug.edu.gh 14 . In Ghana, some of these psychological tests which have been imported into the country but have not yet been validated on the Ghanaian population are being used indiscriminately by practitioners in the assessment and management of clients (Edwin, 2001). This means that clients‘ performances are scored based on norms which may be socioculturally inappropriate for Ghanaians, thus yielding inaccurate results. According to Strauss, Sherman & Spreen (2006), using the appropriate norms for a test is as important as the selection of the test itself. This is because they form the bane upon which diagnoses are made and decisions for management taken. When inadequate norms are used, healthy individuals may be misdiagnosed as cognitively impaired, leading to needless therapy or neglect. In line with this, the objective of this study is to investigate the performance of Ghanaian school-children on the second edition of the Kaufman Assessment Battery for Children (KABC- II) in order to determine its suitability for the country and, in so doing, provide preliminary local normative data upon which subsequent clients‘ performance can be based. 2.1 General issues in cognitive testing Cognitive abilities are brain-based skills that are needed to carry out any task, ranging from the simplest to the most complex. Opoku (in press) defines them simply as all the activities generated from within the conscious mind which influence our behaviour. He further asserts that cognitive abilities are more concerned with the mechanisms of learning, remembering, problem- solving, paying attention, reasoning and even physical dexterity, than with any actual knowledge. Although the terms cognitive ability and intelligence are sometimes used interchangeably, Opoku (in press), vehemently argues that while intelligence as a general construct may include many cognitive abilities, it is debatable whether the two constructs are actually synonymous. University of Ghana http://ugspace.ug.edu.gh 15 . Most measurements of cognitive ability are based on performances of individuals on neuropsychological or educational tests (Carroll, 1976). These tests are objective and reliable neuropsychological instruments that have been designed to measure the abilities of interest (such as intelligence, scholastic aptitude and cognitive and mental abilities) having been first standardized based on a sample from a given population. The norms thus obtained serve as a standard against which an individual‘s performance can be evaluated and interpreted (Lezak, 2005; Mitrushina, Boone, Razani, & D'Elia, 2005). However, bearing in mind that the standardization sample is only a representation of the general population and that individuals do not necessarily come from a homogenous population, it is important to be cautious when selecting tests and interpreting norms as it cannot be assumed, even within a certain race, ethnic, culture or language group that individuals would have acquired the same knowledge and developed the same characteristics (Van der Merwe, 2008). To buttress this point, Mitrushina, et al. (2005) argues that "all normative data are of limited use" because it is a specific group‘s performance in a psychological test that forms the norm for that group, and as such, each group represents its own norm (Van der Merwe, 2008). This does not only imply that specific norms are only useful for those groups with characteristics that are similar to the normative sample but that psychological tests have to be necessarily validated and normed on any population for which it is to be used. Most psychological tests developed and imported from the West however, have not been validated in the African region, making it impossible to know whether they actually measure what they were intended to in the African population (Bangirana, Seggane-Musisi, Allebeck, Giordani, John, Opoka, Byarugaba, Ehnvall & Boivine, 2009). This certainly raises concerns regarding the generalizability of tests developed University of Ghana http://ugspace.ug.edu.gh 16 . in the Western world and the applicability of imported normative data to other parts of the world such as ours. Although there has been much controversy and debate over this issue till date, some authors maintain that versions of these tests which are adapted to measure intelligence in these non- Western cultural contexts still have the potential to provide useful information about the test taker and should not be disregarded (Sánchez-Escobedo & Hollingworth, 2009). Others also attest that while the raw scores obtained on a test can vary for groups that differ according to certain characteristics, the standard normalized scores provide a fair platform for comparison among different groups (Ardila, 1996; Bangirana et al., 2009). In contrast to this, Anderson (2001) opines that despite the availability of tests with acceptable psychometric properties, the use of inadequate normative data as well as the lack of clear guidelines on what constitutes cognitive impairment, is likely to undermine the value of neuropsychological assessment. This could probably cause misclassification or diagnostic errors, which would inevitably lead to deserving people being denied the needed attention while those who do not need it are overindulged. Ostrosky-Solı´s, Gomez-Per`ez, Matute, Rosselli, Ardila & Pineda (2007) share a similar view in their research to develop and standardize the neuropsychological battery Neuropsi Attention and Memory for Spanish- speaking populations in America. The study noted that when tests developed in other populations are used within Spanish speakers, they are frequently just translated and the norms of other populations used. Simple translation, use of inappropriate visual stimuli and use of norms of a foreign instrument do not take into account the sociocultural differences between the populations. Thus, unless the test items are appropriately adapted or developed to assess the new population and new University of Ghana http://ugspace.ug.edu.gh 17 . normative data are obtained, the results will be undoubtedly invalidated leading to errors in diagnoses. Van der Merwe (2008) emphasizes this in her study comparing the WISC-IV test performance for Afrikaans, English and Xhosa speaking South African Grade 7 learners. She notes that the available norms that come with imported westernized tests which, invariably, have not been standardized or normed for cross- cultural use, were more appropriate for use with the white population (Kanjee, 1999, as cited in Van der Merwe, 2008). The study thus concludes that in choosing appropriate norms for an instrument, the relevance of the norms ought to be carefully considered. For some purposes a broadly representative sample or nationally relevant norms may be justified whereas at other times, a specific subgroup sample defined by such demographic variables as ethnicity, culture, race, education, gender and socioeconomic status, may prove most appropriate. Even though the early definitions of intelligence were directly related to school performance, they have been modified, over time, to include culture, language, social class and associated issues. These modifications however make it difficult to determine the standard criteria upon which intelligence testing should be based (Benson, 2003). In the United States of America, for instance, traditional intelligence tests have been deemed culturally unfair as diverse and minority groups such as African Americans, Hispanics and Native Americans have persistently performed below the level of their White counterparts. This ubiquitous discrepancy in performance between Black and White Americans especially, pertinently demonstrated by many studies, has led to the proposition that IQ is genetically determined (Dickens & Flynn, 2006; Rushton & Jensen, 2006). This view has been opposed by other researchers who insist that these studies are not grounded in scientifically derived constructs but rather in folk beliefs and myths; as such, University of Ghana http://ugspace.ug.edu.gh 18 . they are unscientific, naïve and offensive (Sternberg, Grigorenko, & Kidd, 2005; Cooper, 2005; Cronshaw, Hamilton, Onyura & Wilson, 2006). 2.2 Cultural issues in cognitive testing In studying intelligence, Sternberg (2004) observes, it is often assumed that intelligence is perceived as, and indeed, is precisely the same across all cultures. To illustrate this point further, he refers to the empirical studies cited in contributions to the International Handbook of Intelligence (Sternberg, 2004b) most of which employed tests that were based on Western measures of intelligence. Admittedly, some, but not all of these tests were translated into other languages before administration, setting a somewhat unfair platform for the testing of cognitive abilities. Many intelligence theorists and researchers, who are of the mind that intelligence is best defined in terms of a universal general ability ‗g‘ that is fixed across cultures (Hermstein & Murray, 1994), have drawn rather sweeping and sometimes damaging conclusions based on these tests (Sternberg, 2004). This notion of a generalized cross-cultural intelligence however remains highly controversial and continues to be the subject of debate between the former group and several others who believe that many neuropsychological measures do not have acceptable diagnostic accuracy when used among culturally and socioeconomically different people (Ardila, Rodriguez-Menendez, & Rosselli, 2002; Cabo, & Manly, 2006; Boone, Victor, Wen, Razani, & Pontón, 2007; Dale, 2009). In support of the opinions of the latter group of intellectuals, Yan and Sternberg‘s (1997) review of Chinese philosophical conceptions of intelligence identified a Confucian perspective which emphasizes benevolence and doing right, and a Taoist tradition which places a high premium on humility, knowledge of oneself along with freedom from conventional standards of judgment. University of Ghana http://ugspace.ug.edu.gh 19 . They also found out that the Taiwanese Chinese conception of intelligence is based on five main factors; a general cognitive factor which is similar to the ‗g‘ factor in conventional Western tests, interpersonal and intrapersonal intelligence, as well as intellectual self-assertion and self- effacement (Sternberg, 2004). In further support of this, a study by Grigorenko et al., (cited by Sternberg, 2004) to determine the conceptions of intelligence among rural Kenya discovered four broad concepts, namely, ‗rieko‘ which is associated with knowledge-based skills, ‗luoro‘ meaning respect, responsibility and consideration for others, ‗paro‘ which represents initiative, and ‗winjo‘ which deals with the comprehension and handling of real-life problems. Several other studies conducted in Africa by Dasen (1984) throw more light on the cultural conceptions of intelligence. Among the Baoule` people in Ivory Coast for instance, intelligence in children is determined by one‘s willingness and readiness to serve their family and community. The word ‗ngware‘ which means intelligence in Zimbabwe is associated with prudence and caution especially in social relationships. The Chewas in Zambia highlight the importance of social responsibilities, respect, obedience and cooperativeness to intelligence. From these studies cited above, it is evident that even though most cultures acknowledge the importance of the cognitive ability aspect of intelligence, the African and Asian cultures place more emphasis on social competence skills than pertains in the western world (Serpell, 1994). These variations in the definition and conceptualization of intelligence across cultures certainly call for great care in the assessment of cognitive abilities or intelligence. To measure these constructs accurately and fairly, requires the use of culture-free or culture-fair tests which are designed in such a way that they are not biased in favour of any particular culture. However, Samuda (1998) cautions that ensuring accuracy of assessment findings goes further than simply University of Ghana http://ugspace.ug.edu.gh 20 . selecting an instrument that is marketed as culture-fair, because even the so-called culture-fair tests are really culture-reduced ones which suppose that examinees have been socialized and educated in the culture in which the test originated. Anastasi (1990) also asserts that, ―No intelligence test can be culture free, because human intelligence is not culture free‖. Thus, by their very own nature, no such constructed tests can be fair to more than one cultural group. Arguing along similar lines, Ardila (1996) opines that a person‘s cultural environment exerts an influence on the development of cognitive abilities by dictating what is relevant to be learned at each stage in one‘s development. Consequently, although cognitive processes are considered to be universal across cultures, the value placed upon the ability to perform certain tasks in a particular culture, provides the motivation to become skilled in those areas (Harris & Llorente, 2005; Daley & Ongwugbuezie, 2011). This leads to certain kinds of intelligences being more highly evolved in the people of one culture than another. The concept of intelligence is therefore culturally variable and cannot be measured by the same criteria in different societies. Based on arguments such as these, some researchers infer that instead of importing tests from the western world for our use, culture-specific tests which will be more appropriate for measuring cognitive abilities in Africans in Africa, should be employed (Opoku, in press; Ogbu, 2002; Sternberg, Grigorenko & Kidd, 2005). Others also maintain that with the globalization of the world and its attendant acculturation, tests developed in the western world are still applicable and appropriate for use in most parts of the world and should be used accordingly; any significant differences thus found, are due substantially to the different genetic constitutions (Jensen, 2005). University of Ghana http://ugspace.ug.edu.gh 21 . 2.3 Language and intelligence Language is without question, at the core of human existence, distinguishing us from the other creatures in the lower positions on the phylogenetic scale. Gollnick and Chinn (2006) aptly define language as a means of communication that shapes cultural and personal identity and socializes one into a cultural group. Language helps to transform, modify or shape our cognition. According to Mitrushina et al. (2005), the effect of language on cognitive test performance cannot be overemphasized. This is reiterated by Carstairs, Myors, Shores & Fogarty (2006) who proposed that language or proficiency in English has significant impact on verbal subtests, especially when the tests are written in English. Since most of the cognitive assessment instruments in use globally are in English, it stands to reason that people whose first language is English or who are fluent in the English language will have a better understanding of the test and hence perform better than those who have a different first language or are bilingual. This was indeed, proven by Fleisch (2007) in a study which revealed that despite receiving the same quality of education and in the same school, children assessed in a language other than their first language obtained lower scores than their counterparts who received the test in their first language. A more recent study of bilingual and monolingual children by Barac and Bialystok (2012) similarly showed that although the bilingual groups performed similarly as and exceeded monolinguals on executive control tasks, the best performance in verbal tasks was achieved by bilingual children whose language of instruction was the same as the language of testing and whose languages had more overlap. Citing South Africa as a typical example, Van der Merwe (2008) explains that despite having eleven official languages, nine of which are African in origin, few only a few tests are available in any of the African languages (Stead, 2002); thus English remains the main language of University of Ghana http://ugspace.ug.edu.gh 22 . assessment in South Africa. Though most of South Africans are taught and assessed in English, the variances in cultural background, language use, and socioeconomic status among the black African groups cause varying levels of competency in the English language, with the more impoverished having lower verbal skills and hence poorer performances on verbal tasks (Ardila, Rosselli, Matute & Guajardo, 2005; Skuy et al.,2001). Due to the lack of availability of cognitive assessment instruments in other languages aside English, many practitioners, globally, are often forced to informally translate tests into local languages or find already translated versions of standardized tests. Puente and Ardila (2000) however caution that translation and adaptation of tests is a rather complex and time-consuming task, requiring much expertise. Furthermore, such simple language translations and adaptations of English –standardized tests into other languages may compromise the validity of the tests as the impact of familiarity and import of the test items in different ethnic groups may be overlooked (Escandell 2000; Nell, 2000; Cattell & Ardila, 2000; Wong, 2000). Ghana, like South Africa, is faced with similar issues of a multilingual and multicultural society, socioeconomic inequalities and differences in access to quality education. English is the sole official language of instruction throughout the Ghanaian educational system, although for much of the first three years of school, students may study in any of eleven local languages. Most of the assessment instruments imported for use in the country, however, are all in English, making it difficult to accurately assess those who are not fluent in the English language, especially on the verbal components of intelligence tests. To solve this challenge, it is suggested that practitioners will have to both translate and adapt these tests into the local language or develop norms which take into cognizance the subject‘s level of education and English language proficiency (Stead, 2002). University of Ghana http://ugspace.ug.edu.gh 23 . 2.4 Sex differences in intelligence Sex differences in intelligence are among the most politically volatile topics in contemporary psychology and although they have been widely analyzed, in the psychological and neuropsychological literature (e.g., Hyde, 1981; Caplan, Crawford, Hyde, & Richardson, 1997; Deary, Thorpe, Wilson, Starr & Whalley, 2003), no single finding has sufficiently substantiated this claim (Halpern, 1997). However, conclusions from multiple studies suggest that ― females, on average, score higher on tasks that require rapid access to and use of phonological and semantic information in long-term memory, production and comprehension of complex prose, fine motor skills, and perceptual speed‖ than their male counterparts (Halpern, 1997). The latter also tend to perform better on tasks that involve ―transformations in visual–spatial working memory, motor skills involved in aiming, spatiotemporal responding, and fluid reasoning, especially in abstract mathematical and scientific domains‖ (Halpern, 1997). These differences are however not entirely stable throughout life. Some of them seem to get smaller during adolescence, whereas differences in other areas remain (Halpern, 2000). In support of this, Hyde and Linn, (1988), who conducted a meta-analysis of 165 language studies involving both children and adults, concluded that the magnitude of the sex difference in verbal ability is currently so small that it is negligible. Further to that, Wallentin (2009) performed an extensive review of gender differences in language among children and concluded that although a small difference does exist in early language development in favour of females, this invariably disappears during childhood, taking in its wake any sex differences in verbal abilities and in brain structure and function related to language processing. Accordingly, the review concludes that any differences, if existent at all, are not easily detected by current research methods. University of Ghana http://ugspace.ug.edu.gh 24 . In the field of mathematics, research reveals that although globally, no gender disparities are exhibited in the primary mathematical abilities inherent to every culture, they eventually show up in secondary mathematical capabilities that essentially arise in the formal education environment with males consistently outperforming females in geometry and word problems (Mullis et al., 2000 ; Geary, 1996). On interaction between sex and education, studies by Rosselli and Ardilla (2003) found that among the poorly- educated participants or those without any formal education at all, males tended to outperform their female counterparts in almost all cognitive ability tests. However, these sex differences in cognitive ability test scores decreased progressively with increasing educational levels, until they were no longer recognizable in subjects who had had 10 or more years of formal education. These sex differences in cognitive ability test performance have been ascribed to many causes including biological factors, environmental influences, educational background, and cultural idiosyncrasies; however, the actual causes remain unclear (Ardilla et al., 2011). 2.5 Socioeconomic status, parental education and intelligence Socioeconomic status is a multidimensional construct that includes not only measures material wealth (income), education and occupation but also physical health, early education and neighbourhood characteristics. The association between socioeconomic status and cognitive ability has been the object of numerous studies for many years (Jensen, 1968; Burnes, 1970; Boone & Adesso, 1974; Brooks- Gunn, Guo & Furstenberg, 1993; Liaw & Brooks-Gunn, 1994; Anum (1996); Smith, Brooks-Gunn & Klebanov, 1997; Turkheimer, Haley, Waldron, D'Onofrio & Gottesman, 2003; Hunt & Carlson, 2007; Hunt, 2012), with most asserting that parental University of Ghana http://ugspace.ug.edu.gh 25 . socioeconomic status can affect an individual beginning prior to birth and continuing into adulthood. Additionally, it has been reported that socioeconomic factors, like parental education, father‘s occupation and parental income are also related to intelligence and socioeconomic success (Asiedu, 2002; Strenze, 2007). Indeed, Duncan, Yeung , Brooks- Gunn and Smith ( 1998) found that socioeconomic status was more strongly associated with cognitive performance than with other variables such as health and behaviour ; accounting for as much as 20% of the variance in childhood IQ (Gottfried, Bathurst, Guerin and Parramore, 2003). Hackman and Farah‘s (2009) research posited that at least three cognitive systems (i.e. language, executive function and memory) are influenced by socioeconomic status partially explaining why children from low socioeconomic backgrounds perform more poorly than their peers from high or middle socioeconomic classes on tasks requiring selective attention, inhibition, cognitive control and working memory. Other researchers have identified relationships between different aspects of socioeconomic status and various facets of cognitive performance in children, verbal and numeric abilities (Noble, Farah, & McCandliss, 2006), language and working memory (Hackman & Farah, 2009), executive function (Ardila, Rosselli, Matute & Guarjardo, 2005), as well as attention and memory abilities (Villaseñor, Martín, Díaz, Rosselli & Ardila, 2009). In Edwin‘s research to determine the applicability of the Wechsler Intelligence Scale for Children- Third Edition (WISC III) in Ghana (Edwin, 2001), she posits that the disparity between the cognitive performance of children from low class families and those from the middle to high class ones is markedly obvious in rural-urban differences in school, and this increases with age. University of Ghana http://ugspace.ug.edu.gh 26 . A study on the culturally- fair Raven‘s Coloured Progressive Matrices (RCPM) in Ghana found highly significant differences between urban and rural children, with the economically - disadvantaged rural children performing poorer than their urban counterparts (Anum, 1996, as cited in Edwin, 2001). However, it was observed that the difference in performance between the two groups widened with increasing age, implying that the difference was more likely to be due to a slower cognitive development process in rural children, rather than impaired cognition. The slow rate of cognitive development, the study proposed, was as a result of the ―deprived environment‖ pertaining in many rural communities. Prifitera, Weiss, Saklofske & Rolfhus‘ (2005) observation that the persistent difference of 1SD between whites and African Americans on the Wechsler Scales is reduced to 8.8 points when the socioeconomic status and other pertinent variables are controlled for, certainly lends credence to the fact that socioeconomic status does indeed moderate cognitive ability. A recent study conducted in the UK by Hanscombe, Trzaskowski, Haworth, Davis, Dale and Plomin (2012) reiterates the view that parental socioeconomic status moderates the heritability of their children's intelligence, with a greater genetic contribution to IQ in high-socioeconomic families compared to low-socioeconomic ones. This lower inheritability in lower-socioeconomic families suggests that environmental interventions might be more effective than any other, in improving cognitive development of affected children. Substantial evidence exists in support of the assertion that education has a long term effect on cognitive ability (Rosselli and Ardila, 2003; Peters, Baker, Dieckmann, Leon & Collins, 2010; Hunt, 2012; Chin, Negash, Xie, Arnold and Hamilton, 2012). It is generally held that formal education instills a variety of skills which are required to perform well on psychological test of intellectual ability; and the influence of various educational variables such as level and quality of University of Ghana http://ugspace.ug.edu.gh 27 . education, access to formal education, effects of schooling and level of parental education on cognitive performance has been illustrated in many studies (Van der Merwe, 2008; Anger and Heineck, 2010; Matute, Montiel, Pinto, Rosselli, Ardila and Zarabozo, 2012). In their study to investigate the influence of parental educational level as well as type of school attended on executive function test performance, Ardila, Rosselli, Matute and Guarjardo (2005) posit that differences in executive function test performance between private and public school children are not wholly due to the school characteristics, but could be attributed to some other conditions potentially associated with attending a private or public school, such as the parents‘ educational level. The study further proposes an association between parents‘ educational level and home environmental circumstances which undoubtedly leads to the establishment of early skills when solving problems or performing executive function tasks. Ardila et al, (2005), also highlights Asiedu‘s (2002) suggestion that parents who have a higher level of education may have a value system or cosmovision which correlates with a more intellectually stimulating environment for their children; this ultimately results in the latter attaining higher performances in some executive function tests. Another study in this area by Villaseñor, Martín, Díaz, Rosselli and Ardila (2009) realizes that school type and parental educational level which is a good indicator of socioeconomic status, however seem to be related to attention and memory abilities and this correlation may be linked to a better development of verbal abilities and vocabulary acquisition. This is corroborated by Castillo, Ruiz, Chillón, Jiménez-Pavón, Esperanza-Díaz, Moreno and Ortega (2011) in the AVENA study which concluded that high parental educational and occupational levels which are also indicative of socioeconomic status, as well as the type of school attended (private vs. public), were positively associated with better cognitive performance in their sample of Spanish University of Ghana http://ugspace.ug.edu.gh 28 . adolescents. This was most pronounced in the verbal test scores. Comparison of the performance on the Griffiths Mental Development Scales of Black, South African infants with mothers who had twelve or more years of education and who were professionally employed, with infants of mothers with fewer than twelve years of education and who were employed in non-professional jobs yielded a significantly better performance in the former group than their counterparts, especially with respect to gross motor functioning (Cockcroft, Amod & Soellaart, 2008). This highlights the importance of taking into consideration, the maternal level of education and the socioeconomic status background of the child during neuropsychological assessments. Some tests however, are notoriously more sensitive to educational variables (e.g., language tests) than others (e.g., the Wisconsin Card Sorting Test). Illiterates and individuals with low levels of education have long been recognized to show low levels of performance on psychological tests in general. These low scores obtained by such people in current neuropsychological tests may be due in part to differences in learning opportunities of those abilities that the examiner considers relevant, to lack of exposure to these tests and the irrelevance of the test to them (Ardila, Ostrosky-Solis, Rosselli & Gomez, 2000). According to Ardila et al., (2005), although a big gap was noticed between the performances of private and public institutions in their research, this does not necessarily pertain in all societies. However, trends in academic performance over the years Ghana have similarly revealed a persistent achievement disparity between students in public and private basic schools (Quansah, 2000). Academic achievement as measured by Basic Education Certificate Examinations (BECE) and Criterion Reference Tests results of students in private schools consistently exceeds that of those in public schools (Quansah, 2000). Countless reasons have been ascribed to this, including lower socioeconomic backgrounds of parents, lesser investment and support in their University of Ghana http://ugspace.ug.edu.gh 29 . children‘s education, neglect and lack of maintenance of the physical infrastructure, and over- dependency on free government supplies in the public schools (Asiedu, 2002). As quality of education has a significant impact on cognitive performance, these differences in academic performances of private and public schools in the country should therefore not be unexpected. 2.6 The Kaufman Assessment Battery for Children- Second Edition (KABC-II) Designed for children aged three to eighteen years, the KABC-II falls within the top tier of cognitive ability tests for children available globally today. The test was which originally developed by Alan S. Kaufman and Nadeen L. Kaufman in 1983 and revised in 2004 incorporates several important developments in psychological theory and statistical methodology in its construction. According to its authors, it was designed to contribute to psychological, neuropsychological, clinical and psychoeducational evaluations. As a result of the minimized cultural content, as well as the reduced verbal instructions and responses, the KABC-II is capable of generating in-depth information about an examinee‘s cognitive abilities with less ―filtering‖ due to language and cultural differences (McKown, 2010). Unlike its original counterpart, the KABC-II is grounded in a dual theoretical foundation: Luria‘s Neuropsychological Model composed of 3 functional units and the Cattell-Horn-Carroll (CHC) hierarchical organization of broad and narrow cognitive abilities. Either approach gives you a global score that is highly valid and that shows small differences between ethnic groups in comparison with other comprehensive ability batteries. The KABC-II comprises 18 subtests which are of two types, core and supplementary. The core subtests provide all of the scale and global scale indexes, while the supplementary ones offer a wider coverage of the construct being measured. This allows a broader foundation for the University of Ghana http://ugspace.ug.edu.gh 30 . measurement of cognitive functioning and identification of any deficits that might be present. The number of supplementary subtests administered depends on the examinee‘s age and the examiner‘s discretion. They do not contribute to the examinee‘s scores except in the instance where they are used as replacements for spoiled core subtests or for the Nonverbal Scale. Some of subtests which are core at certain ages are supplementary at others (Kaufman & Kaufman, 2004). The Luria model of the KABC-II has between five and eight core subtests at each age. The CHC model on the other hand has two additional core subtests at each age, which measure knowledge/Gc. These two knowledge core subtests are supplementary for the Luria model. An exception to this rule occurs however, when during comparison between Initial Learning with Delayed recall, the supplementary subtests Atlantis Delayed and Rebus Delayed must necessarily be employed (Kaufman & Kaufman, 2004). For the KABC-II, scores for both the core and supplementary subtests are reported as scaled scores with a mean of 10 and a standard deviation of 3. All indexes (scores on the various scales and global scales) are expressed as standard scores with a mean of 100 and a standard deviation of 15. All the core subtests as well as all except one supplementary subtest are fully normed on the complete KABC-II standardization sample (Kaufman & Kaufman, 2004). The 18 subtests of the KABC-II are grouped into 5 scales that correspond to the processing areas and broad abilities of both models. The broad abilities are of primary importance in cognitive profile interpretation (Kaufman & Kaufman, 2004). Some subtests are core at some ages and supplementary for others. The core subtests provide all of the scale and global scale indexes. The supplementary ones enable a more extensive coverage University of Ghana http://ugspace.ug.edu.gh 31 . of the constructs measured by the KABC-II, thus allowing a broader base for cognitive function measurement and identification of process deficits. Administration of the instrument takes about 25 to 75 minutes depending on the interpretative model selected (Kaufman &Kaufman, 2004). The Luria model centers on mental processing and minimizes the importance of acquired knowledge, considering the latter to be more related to life experiences than to cognitive ability. This model comprising eight or nine subtests, depending on the age of the examinee, yields a Mental Processing index (MPI) global score composed of Sequential Processing, Simultaneous Processing, Learning Ability, and Planning Ability. It is recommended for situations where the validity of the global composite would be compromised by including acquired knowledge such as in the testing of bilinguals, non-mainstream culture individuals, and children with language disorders (McKown, 2010). The CHC model on the other hand, yields a Fluid-Crystallized Index (FCI) that emphasizes the importance of acquired knowledge and incorporates all the scales of the Luria model as well as an additional Knowledge/Gc scales. It comprises performance on indexes of Short-Term Memory (Gsm), Visual Processing (Gv), Long-Term Storage and Retrieval (Glr), Fluid Reasoning (Gf), and Crystallized/Verbal Ability (Gc). The inclusion of crystallized abilities in the FCI provides an alternative means of evaluating a child‘s cognitive abilities based on a theory which is consistent with several other Kaufman tests as well as other traditional views of cognitive testing. This model is employed in almost all other evaluations including children with disability in reading, written expression or mathematics, intellectual disability, attention deficit hyperactivity disorder, or emotional or behavioural disturbance. Either approach therefore, yields a global score that is highly valid and shows relatively smaller differences between ethnic groups when compared to other tests of cognitive ability (Kaufman & Kaufman, 2004). University of Ghana http://ugspace.ug.edu.gh 32 . The KABC-II also can also be used for nonverbal estimate of overall cognitive ability (Nonverbal Index, or NVI). This includes pantomime- like language-reduced instructions and nonverbal responses and is recommended for hearing loss, very limited English proficiency, moderate to severe impairments and language disorders (Kaufman & Kaufman, 2004). 2.6.1 Standardization Issues of the KABC-II Deriving from the 2001 US Census Bureau statistics, 3,025 examinees age 3 through 18 years from 39 states and the District of Columbia were randomly selected for the standardization of the KABC-II. These examinees were children who could speak English, were non-institutionalized and were not known to have any physical or perceptual impairment that would hinder their performance on the KABC-II. Samples were stratified based on gender, ethnicity and SES within ethnicity, parental education/SES, geographic region and special education (e.g. specific learning disability, speech/language impairment, intellectual disability, emotional/behavioural disturbance, ADHD, or gifted placement). The KABC-II was co-normed with the Kaufman Test of Educational Achievement, Second Edition (Comprehensive Form) at ages 4 years 6 months to 18 years 11months, in order to permit more direct ability-achievement comparisons. It is of great import to note that the performance of children from low SES-backgrounds is less affected with the KABC-II than pertains in most other traditional cognitive ability tests. Study findings reveal that the effects of SES correlated .39 on WISC-IV FSIQ, thereby attributing about 15% of variance in WISC-IV to effects of SES, as opposed to KABC-II‘s 6% (Fletcher- Janzen & Daniel, 2005). University of Ghana http://ugspace.ug.edu.gh 33 . 2.6.2 Reliability and Validity Issues of the KABC-II The reliabilities of the KABC-II global scale indexes of the KABC-II are particularly high, ranging from mid to upper .90s for the FCI and MPI and a little lower for the NVI. The average internal consistency coefficients are .95 for the MPI at both ages 3–6 and ages 7–18, and .96 at the same ages for the FCI. The average reliabilities of the scale indexes are .90 for both age ranges (Kaufman & Kaufman, 2004). Generally, the subtests display very good internal consistency across all ages, with a median reliability .85 for ages 3 to 6 years and .87 for ages 7 to 18 years for the core subtests. The supplementary subtests, however, show slightly lower reliabilities (Kaufman & Kaufman, 2004). The mean subtest reliabilities for ages 3 to 6 years range from .69 (Hand Movements) to .92 (Rebus). For this age- group, all the subtests, with the exception of Hand Movements, Face Recognition and Gestalt Closure, have mean coefficients of .80 or more. For ages 7 to 18 years, the mean subtest reliabilities range from .74 (Gestalt Closure) to .93 (Rebus). Aside Number Recall, Gestalt Closure, Story Completion and Hand Movements, all the other subtests have reliabilities of at least .80 (Kaufman & Kaufman, 2004). In terms of validity, the KABC-II presents substantial favourable evidence in terms of content, relationships within the test itself sand relationships with other tests. In the study by Bangirana et al., (2009) to assess the construct validity of the KABC-II in the Ugandan context, particular interest was paid to Luria‘s model of information processing comprising the four scales Sequential Processing, Simultaneous Processing, Learning and Planning. Results suggested that KABC-II tests truly measured those abilities in Ugandan children (Kaufman & Kaufman, 2004). Validation studies of the original KABC in children at risk proved that it was sensitive to University of Ghana http://ugspace.ug.edu.gh 34 . parental education and the type of home environment. In a similar vein, the subtests measuring similar abilities in the KABC-II loaded on the same factor, thus retaining construct validity in a group of children whose cognition had been compromised by cerebral malaria. In Fletcher-Janzen‘s study to assess the prepublication validity of the Lurian and CHC theoretical models with different cultural populations, the KABC-II was administered to a group of Taos Pueblo children. Due to the flexibility of reducing potentially confounding variables from the Mental Processing Index score, the results obtained actually reflected the child's individual information processing in contrast to how their verbal expression measured up to other children in the mainstream. The CHC index scores were generally lower than the Lurian index scores, thus supporting the validity of both models of the KABC-II (Fletcher-Janzen, 2003). In order to find out whether the KABC-II measures the same constructs across all ages, Reynold, Keith, Fine, Fisher, & Low (2007) conducted multiple-sample analyses to test for equality of the variance-covariance matrices across the 3- to 18-year-old sample. Higher-order confirmatory factor analyses comparing the KABC-II model with rival CHC models for children ages 6 to 18 confirms that the KABC-II measures the same constructs across all ages and further lends support for the construct validity of the test. . Like all other psychological tests, the KABC-II has its limitations and short-comings. These include but are not limited to the fact that although the test presents two distinct interpretive models for subscales each one is simply a ―renamed‖ version of the other. Another shortcoming is that although confirmatory analysis shows the KABC-II to measure the same constructs across all age-groups, it was noted that models which exclude time-bonuses tended to fit better than University of Ghana http://ugspace.ug.edu.gh 35 . when time-bonuses were included. According to Kaplan and Saccuzzo (2005), criticisms of the KABC such as lack of correspondence between the definition and measurement of intelligence (Sternberg, 1984) which are largely valid and generally apply to the KABC-II as well, remain unresolved. 2.7 Rationale for the study Even though the KABC-II is used extensively by practitioners in Ghana, searches of journals, texts and the internet shows that there is largely unavailable documentation about its validity among the children in this country. It is only assumed that since the test is purported to be culturally sensitive, it is wholly applicable to our peculiar sociocultural setting. This may not necessarily be so (Akhouri, Jahan, Singh, & Singh, 2006; Giordani, Boivin, Opel, Nseyila, & Lauer, 1996; Valencia, 1995; Fletcher-Janzen, 2003). A study conducted among Taos Pueblo Indian children in America supported the validity of the KABC-II in that, although the children had many distinctly different linguistic and cultural experiences from the majority of the normative sample for the test, their performance on the KABC-II compared favourably with same - aged children in the standardization sample. However, the Taos children's performance on the Verbal Knowledge ability scale of the KABC- II was somewhat lower, a probable reflection of the linguistic, cultural, and socioeconomic differences between them and the average American (Fletcher-Janzen, 2003). This further buttresses Dale‘s (2009) observation that in psychological testing, practitioners should take the cultural background of their examinees into consideration and determine how well that background matches the normative sample of the test being administered in order to prevent University of Ghana http://ugspace.ug.edu.gh 36 . misdiagnosis, misclassification or other such negative consequences. The purpose of this study therefore, is to ascertain the validity of the KABC-II in the Ghanaian context. 2.8 Statement of hypotheses The following hypotheses, based on the studies reviewed, were formulated and tested in the present study: Hypothesis One: There will be increasing performance on all the five scales of the KABC-II with increasing age such that: 1a. Children between the ages of 10 years and 12 years 11months are more likely to perform better than children between the ages of 6 years and 7 years 11months. 1b. Children between the ages of 10 years and 12 years 11 months are more likely to perform better than children between the ages of 8 months and 9 years 11months. 1c. Children between the ages of 8 years and 9 years 11months are more likely to perform better than children aged between the ages of 6 years and 7years 11months. Hypothesis Two: Children from private schools are more likely to perform significantly better than their counterparts from the public schools on the five scales of the KABC-II (a. Learning/Glr b. Sequential/Gsm c. Simultaneous/Gv d. Planning/Gf e. Knowledge/Gc). Hypothesis Three: The performance of children on the KABC-II scales will be such that: 3a. There will be no significant sex differences in the performance of male and female children on the Learning/Glr scale. University of Ghana http://ugspace.ug.edu.gh 37 . 3b. There will be no significant sex differences in the performance of male and female children on the Sequential /Gsm scale. 3c. There will be no significant sex differences in the performance of male and female children on the Planning/Gf scale. 3d. There will be no significant sex differences in the performance of male and female children on the Knowledge/Gc scale. 3e. Male children are more likely to perform better than females on the Simultaneous/Gv scale. Hypothesis 4: The performance of Ghanaian school children on the KABC-II will be such that: 4a. children from private schools will perform better on the FCI than the MPI 4b. children from public schools will perform better on the MPI than the FCI. 2.9 Operational Definitions: Private School: Schools where the cost of education is wholly borne by the parents or guardians of the school-children. Public School: Schools where the cost of education is solely funded by the national government. High socioeconomic status: Male parents or guardians whose professions require the tertiary (university, polytechnic, post-secondary) level of education. Medium socioeconomic status: Male parents or guardians whose professions require the secondary (senior high school) level of education. University of Ghana http://ugspace.ug.edu.gh 38 . Low socioeconomic status: Male parents or guardians whose occupations require education at the primary level or no formal schooling at all. The occupations of female parents or guardians were used in cases where the male counterpart was not available. Older children: Children aged 10 years up to 12 years 11 months. Mid- aged children: Children aged 8 years to 9 years 11 months. Younger children: Children aged 6 years up to 7 years 11 months. University of Ghana http://ugspace.ug.edu.gh 39 . CHAPTER THREE METHODOLOGY 3.1 Study Population A total of 90 school-children in the Greater Accra region, comprising 44 males and 46 females aged 6 to 12 years inclusive, with valid data on parental occupational levels were included in the present study. The participants were selected from both rural and urban areas as well as both private and public schools. No formal testing was done to diagnose intellectual or learning disabilities prior to the study; chronological age-grade disparity was ignored, preferring to impute it to a late start in school rather than to intellectual disability. 3.2 Study design The study is 2 (sex) x 2(school-type) x 3 (age groups) multifactorial design. The sexes are male and female, the school-types, private and public schools, and the age groups are children aged 6 years to 6 years 11months, 7 years to 7 years 11 months through to 12 years 11months. The age groups were further conveniently categorized into three sets; ages 6 years to 6 years 11 months, 7 years to 9 years 11 months and then 10 years to 12 years 11 months. The measures were the pupils‘ performances on the five scales that comprise the Fluid Crystallized Index (FCI) of the KABC-II, each represented by an aggregate of two of its component subtests. In analyzing the data obtained however, the KABC-II conversion scores using the standardized norms were employed. 3.3 School selection The Greater Accra region of Ghana is divided broadly into 6 districts namely the Accra Metropolis, Tema Metropolis, Ga West, Ga East, Dangme West and Dangme East. The original University of Ghana http://ugspace.ug.edu.gh 40 . intent of selecting two schools (one private and one public) from each district could not be followed through due to time constraints and the reluctance of some schools to participate in the study. Ultimately, six primary schools comprising four private and 2 public schools were selected by convenient sampling from all over the Greater Accra region for the study. 3.4 School type The schools were classified as public or private based on the source of funding for the institution. Public schools are funded by the national government, whereas the cost of education in private schools is completely borne by the parents or guardians of the pupils. 3.5 Sample size and sampling technique According to Strauss, Sherman and Spreen (2006), it is commonly assumed that the larger the sample size of a test, ―the more reliable and representative the scores derived‖. As a rule of thumb, they suggest at least 200 subjects for conducting item analysis conduction while others find a minimum of 150 subjects adequate. Strauss et al., (2006) however caution that even large normative samples yield small sizes when divided into subgroups based on demographic variables such as age, gender and education which ultimately may provide a better fit demographically than norms from a large, nationally representative test. Due to time constraints, the originally proposed sample size of 200 participants for this study could not be attained. Instead, the study comprised 90 school-children in the Greater Accra region, aged 6 years to 12 years inclusive. Fifty- four of the children, made up of 24 males and 30 females attended private schools while 36 of them, composed of 20 boys and 16 females were in public schools. The schools included in this study were selected by convenient sampling and University of Ghana http://ugspace.ug.edu.gh 41 . the participants, by simple random sampling from the school registers, using the Table of Random Numbers. For each class, pupils were given numbers with the first name in the class register being assigned the number 10, the next 11, until everyone had a double digit number. Then based on the sequence of the numbers in the Table of Random numbers, some of the students were chosen to participate in the study. For the purposes of this research, the school- children were divided into three age categories as shown in Table 1, based on the developmental, cognitive and psychosocial theories of the developmental stages of children with increasing age exhibiting increased concentration, more knowledge and greater ability to solve problems (Jackson & Warren, 2000). The summary of the sample characteristics are presented in the Table 1. Table 1: Frequency and Percentages of School- children in Private and Public Schools Variables Category Private School (N=54) Frequency (%) Public School (N=36) Frequency (%) Total (N=90) Sex Male 24 (44.4%) 20(55.6%) 44 (48.9%) Female 30 (55.6%) 16 (44.4%) 46 (51.1%) Age Group 6 years- 7 years 11 months 16 (29.6%) 11 (30.6%) 27 (30.0%) 8 years- 9 years 11 months 16 (29.6%) 13 (36.1%) 29 (32.2%) 10years–12years 11months 22 (40.8%) 12 (33.3%) 34 (37.8%) SES High 36 (66.7%) 1 (2.8%) 37 (41.1%) Medium 16 (29.6%) 9 (25.0%) 25 (27.8%) Low 2 (3.7%) 26 (72.2%) 28 (31.1%) University of Ghana http://ugspace.ug.edu.gh 42 . 3.6 Exclusion criteria Children with ill-health, physically or mentally, were excluded from the study. This was done in collaboration with the teachers and school-nurses who had written medical proof of any ailments the children with ill-health were suffering from. Any child who was selected for the study but reported ill had to produce a note from health personnel, a parent or the school-nurse where applicable to that effect. 3.7 Instruments The study employed the test tool kits for the Kaufman‘s Assessment Battery for Children- Second Edition (KABC-II). 3.7.1 The Kaufman’s Assessment Battery for Children, Second Edition (KABC-II) It is clearly stated in the KABC-II manual that ―the CHC model should generally be the model of choice, except in cases where the examiner believes that including measures of acquired knowledge/crystallized ability would compromise the validity of the Fluid-Crystallized Index,‖ in which case the Luria model (MPI) is favoured (Kaufman & Kaufman, 2004a). The CHC model is however preferred over the Luria model as according to the authors, Knowledge/Gc is a relevant part of cognitive functioning ((Kaufman & Kaufman, 2004a). For this study, the Cattell- Horn-Carroll model was employed for various reasons. First of all, the educational system in Ghana which is based on the British model is, according to the MOESS, considered to be one of the best in the West African region and the West African Examinations Council (WAEC) ensures that educational standards are maintained at their United Kingdom (UK) equivalents. For instance, senior high school graduates in Ghana hold the WAEC University of Ghana http://ugspace.ug.edu.gh 43 . senior secondary certificate, which is equivalent to the UK matriculation standard, requiring a combination of passes at GCE O-Level or its equivalent, and at least two subjects at GCE A- Level or its equivalent. This presupposes that Ghana has a high quality of education which is of international standards and thus, school-children can be fairly assessed using the CHC model of the KABC-II (Quansah, 2000). Secondly, even though English is a second language in Ghana, it is the medium of instruction at all levels of education (except the first few years of primary school where the main Ghanaian language of the region is used). As such, it is expected that the participants of this study are proficient enough in the English language for a fair assessment of their cognitive abilities (Quansah, 2000). Another reason for the employment of the CHC model for this study is that in Ghana, subjects such as mathematics, science, social studies, cultural studies, Ghanaian languages, English, agriculture, life skills and physical education are studied by all primary school pupils, inferring that the children under study have acquired some degree of formal knowledge which is expected to positively influence their cognitive performance (Quansah, 2000). The various scales of the KABC-II assessed in this study comprised the following: Sequential Processing/Short-term Memory scale (Gsm) In this, a child‘s ability to solve problems by remembering and using an ordered series of images or ideas is tested. Sequential processing and short-term memory are closely related to a variety of learning tasks, including mathematical calculation and spelling (Kaufman & Kaufman, 2004). The subtests assessed were: University of Ghana http://ugspace.ug.edu.gh 44 . Number recall (Core) - The child is required to repeat a series of numbers in the same sequence as the examiner recites them. The length of the sequence increases from two to nine digits. Hand movements (Supplementary) – this is a non-verbal test. The child performs a series of hand movements in the same sequence the examiner shows him or her. Simultaneous/Visual Processing Scale (Gv) The child‘s ability to consider an array of information to order to solve a problem is assessed. As this integrated form of thinking often requires visualization, it is closely related with visual processing ability (Gv). The subtests administered for this scale were: Triangles Children are asked to copy or build specified designs using coloured triangles or plastic shapes of different sizes and colours. The test can be administered nonverbally, and for the older children, was timed. Block Counting The child is shown a picture of a stack of blocks (some of which are hidden). He or she must then visualize the unseen blocks to determine how many are in the stack. Learning Ability/Long-Term Storage and Retrieval Scale (Glr) This scale measures a child‘s ability to successfully complete different types of learning tasks. Immediate-recall and delayed recall tasks are included. The subtests in this scale comprise: University of Ghana http://ugspace.ug.edu.gh 45 . Atlantis Examinees learn nonsense names of colourful pictures of fish, shells, and underwater plants. The child is then shown another picture with the same characters; he or she must point to the correct character out of many other options when the examiner says its name. Rebus After learning the word associated with each of a series of simple line drawings, the child must ―read‖ a series of drawings that form a meaningful phrase. Planning/Fluid Reasoning Scale (Gf) This scale presents nonverbal problems to the child, who must solve them using the set of mental operations that make up fluid reasoning namely, drawing inferences, understanding implications and applying inductive or deductive reasoning. The subsets administered comprised: Pattern Reasoning- The child is presented with a group of geometric designs or representational pictures, and asked to select an image to fill a gap in a series of pictures that forms a pattern. The child has to form and check hypotheses about the rule underlying the pattern in order to solve them accurately. Story Completion- the child is shown a series of pictures, with some gaps, that tell a story. The examinee must then fill in these gaps to complete the story. Images for filling in the gaps are selected from a group that includes both relevant and irrelevant pictures. Knowledge/Crystallized Ability Scale (Gc) This scale measures the breadth and depth of the child‘s general knowledge and the ability to apply it effectively. The questions assess knowledge of words and facts, using both verbal and University of Ghana http://ugspace.ug.edu.gh 46 . pictorial stimuli, and requiring either verbal or pointing response. The subtests in this scale s that were used in the study are: Expressive Vocabulary- This subtest measures the child‘s knowledge of the names of objects, which is one aspect of crystallized ability. For each item, the child is shown a colour drawing of an object and has to say its name. Verbal Knowledge – This measures the child‘s store of vocabulary and general information. The examiner says a vocabulary word or gives some general information prompt, and the child points to one out of six colourful art images or pictures that best illustrates the answer. This subtest does not require the child to give a verbal response. 3.8 Procedure Ethical clearance was first sought from Noguchi Memorial Institute for Medical Research Ethics Review Board. Granted that, consent for participation was then sought from the Ghana Education Service Director for Greater Accra Region, the head teachers of the selected schools, parents of the selected students and finally from the students themselves, in written and or verbal form. The study was then conducted by the principal researcher, aided by two assistants, both undergraduate Psychology students who had undergone rigorous training in the administration of the KABC-II test by the supervisor. Although the students, parents and heads of schools were assured of the strict privacy, confidentiality and anonymity of any information obtained or recorded in the study, consent to participate was generally difficult to obtain, especially from parents and guardians of children in the private schools. Thus, selected individuals who were not given approval to participate were University of Ghana http://ugspace.ug.edu.gh 47 . duly replaced with the next available pupil of the same age and sex in the class register. The research team was introduced to each participating student as they arrived for the study, and our mission and purpose were explained to them. They were reassured of the study being highly confidential and having no bearing on their academic work. On the average, it took about 45 to 90 minutes to administer the KABC-II. Altogether, this was sometimes too long for the subjects in which case breaks were scheduled in the course of the test for their convenience. Administration of the tests was slow and arduous, sometimes requiring up to 3 days for one student to complete all the tasks involved. This was mainly due to the fact that access to the students was restricted to only periods during which there was no academic activity going on, such as free periods or sports-time. 3.9 Data Management and Analysis The results were analysed using Independent t-test, Descriptive Analysis of Means and Standard deviations, and Three-way ANOVAs. University of Ghana http://ugspace.ug.edu.gh 48 . CHAPTER FOUR RESULTS The children were tested on the five subscales of the KABC-II‘s Fluid Crystallized Index (FCI), each represented by two of its component subtests. The Learning/Glr scale of the KABC-II was made up of the subtests Atlantis and Rebus. The Sequential/Gsm scale encompassed the subtests Number Recall and Hand Movements. Simultaneous/Gv included Block Counting and Triangles. The Planning/Gf scale incorporated the Story Completion and Pattern Reasoning subtests, whiles the Knowledge/Gc scale was an aggregate of the expressive Vocabulary and Verbal Knowledge subsets. The data gathered was analysed using the SPSS 20.00 to summarize the key findings by employing both descriptive and inferential statistics. Descriptive statistics such as means, standard deviations, standard error of the mean and internal consistencies of the subscales of KABC-II, as well as inferential statistics such as the Three-Way ANOVA were employed as the main statistical tools for testing the hypotheses. University of Ghana http://ugspace.ug.edu.gh 49 . Table 2: Descriptive Statistics and Internal Consistencies of KABC-II Scales and Subtests for 6-7 years 11 months Variables N Mean SD SE Min Max Skew Kurt’ Α LEARNING/Glr 27 113.33 38.68 7.44 28.00 170.00 -.22 -.63 .73 Atlantis 27 66.63 23.13 4.45 18.00 104.00 -.16 -.64 Rebus 27 46.70 20.34 3.91 10.00 78.00 -.26 -.87 SEQUENTIAL/Gsm 27 22.93 6.32 1.21 10.00 34 -.32 -.45 .68 Number Recall 27 12.52 3.39 .65 6.00 19.00 .07 -.88 Hand Movement 27 10.41 3.87 .74 1.00 16.00 -.65 -.10 SIMULTANEOUS/Gv 27 24.07 9.99 1.92 8.00 43.00 .29 -.71 .35 Triangles 27 16.33 7.74 1.49 4.00 34.00 .58 -.29 Block Counting 27 7.74 4.78 .92 1.00 18.00 .23 -.89 PLANNING/Gf 27 20.07 17.05 3.28 2.00 52.00 .88 -.67 .95 Story Comp. 27 9.26 9.57 1.84 0 31.00 1.18 -.06 Pattern Reasoning 27 10.81 8.21 1.58 1.00 26.00 .58 -1.06 KNOWLEDGE/Gc 27 45.85 18.07 3.48 17.00 85.00 .29 -.72 .71 Expressive Vocabulary 27 19.96 4.93 .95 12.00 30.00 .42 -.49 Verbal Knowledge 27 25.89 13.63 2.62 4.00 57.00 .25 -.49 On performance on the Learning/Glr scale of the KABC-II, which is an aggregate of the subtests Atlantis and Rebus, children aged between 6.00 years and 7 years 11 months obtained a mean score of 113.33 with a standard deviation of 38.6 and standard error of 7.44. The highest score obtained was 170.00 and the minimum, 28.00. The scores on the Learning/Glr scale followed the normal distribution with skewness of -.22 and kurtosis of -.63. The internal consistency of the Learning/Glr scale was .73. University of Ghana http://ugspace.ug.edu.gh 50 . On the Sequential/Gsm scale of the KABC-II, the 6.00 to 7 years 11 month olds obtained a maximum sore of 34.00 and a minimum of 10.00, with a mean score of 22.93, a standard deviation of 6.32 and standard error of 1.21. The scores on the Sequential/Gsm scale which is composed of the Number Recall and Hand Movement subtests were normally distributed, with skewness of -.32 and kurtosis of -.45. The internal consistency of the Sequential/Gsm scale was .68. For the performance of children on the Simultaneous/Gv scale, descriptive statistics showed that children between the ages of 6.00years to 7years 11months obtained a mean score of 24.07 with a standard deviation of 9.99, standard error of 1.92, minimum score of 8.00 and a maximum score of 43.00. The scores on the Simultaneous/Gv scale follow the normal distribution with skewness of -.29 and kurtosis of -.71. The internal consistency of the Simultaneous/Gv scale which is made up of Triangles and Block Counting was .35. For the performance of children on the Planning/Gf scale, descriptive statistics showed that children between the ages of 6.00 -7 years 11 months obtained a mean score of 20.07 with a standard deviation of 17.05, standard error of 3.28, minimum score of 2.00 and a maximum score of 52.00. The scores on the scale followed the normal distribution with skewness of -.67 and kurtosis of .95. The internal consistency of the Planning/Gf scale which consists of Story Completion and Pattern Reasoning was .95. Finally, on the Knowledge/Gc scale of the KABC-II, children between the ages of 6.00-7years 11months obtained a mean score of 45.85 with a standard deviation of 18.07 standard error of 3.48, minimum score of 17.00 and a maximum score of 85.00. The scores on the Knowledge/Gc scale followed the normal distribution with skewness of .29 and kurtosis of -.72. The internal University of Ghana http://ugspace.ug.edu.gh 51 . consistency of the Knowledge/Gc scale which is made up of the Expressive Vocabulary and Verbal Knowledge subtests was .71. Table 3: Descriptive Statistics and Internal Consistencies of KABC-II Scales and Subtests for 8-9 years 11 months Variables N Mean SD SE Min Max Skew Kurt (α) LEARNING /Glr 29 132.90 37.40 6.95 62.00 183.00 -.41 -1.21 .89 Atlantis 29 79.34 18.94 3.52 40.00 107.00 -.52 -.57 Rebus 29 53.55 20.49 3.81 11.00 78.00 -.48 -1.18 SEQUENTIAL/Gsm 29 27.34 8.75 1.62 16.00 59.00 1.62 4.90 .48 Number Recall 29 14.31 3.07 .57 9.00 20.00 .106 -.49 Hand Movements 29 13.03 6.98 1.30 5.00 42.00 2.65 10.19 SIMULTANEOUS/Gv 29 34.52 10.99 2.04 16.00 65.00 .67 .77 .48 Triangles 29 21.41 6.95 1.29 10.00 38.00 .55 -.48 Block Counting 29 13.10 6.62 1.23 2.00 30.00 .65 -.98 PLANNING/Gf 29 30.31 20.65 3.83 3.00 84.00 1.06 -.84 .87 Story Completion 29 12.31 10.89 2.02 0 42.00 1.51 1.88 Pattern Reasoning 29 18.00 11.06 2.05 3.00 42.00 .56 -.78 KNOWLEDGE/Gc 29 55.10 14.75 2.74 32.00 84.00 .09 -1.13 .72 Expressive Vocabulary 29 22.17 4.83 .89 13.00 31.00 -.01 -.92 Verbal Knowledge 29 32.93 10.77 2.00 15.00 54.00 -.05 -1.04 The descriptive statistics in Table 3 show that on the Learning/Glr scale of the KABC-II, children between the ages of 8.00 to 9 years11 months obtained a mean score of 132.90 with a University of Ghana http://ugspace.ug.edu.gh 52 . standard deviation of 37.40, standard error of 6.95, minimum score of 62.00 and a maximum score of 183.00. The scores on the Learning/Glr scale followed the normal distribution with skewness of -.41 and kurtosis of -1.21. The internal consistency of the of the Leaning/Glr scale which is made up of Atlantis and Rebus was .89. Additionally, on the Sequential/Gsm scale of the KABC-II, children between the ages of 8.00 to 9years 11months obtained a mean score of 27.34 with a standard deviation of 8.75, standard error of 1.62, minimum score of 16.00 and a maximum score of 59.00. The scores on the Sequential/Gsm scale did not follow the normal distribution with skewness of 1.62 and kurtosis of 4.90. The internal consistency of the Sequential/Gsm scale which is made up of Number Recall and Hand Movement is .48. For the performance of children on the Simultaneous/Gv scale, descriptive statistics showed that children between the ages of 8.00 years to 9 years 11 months obtained a mean score of 34.52 with a standard deviation of 10.99, standard error of 2.04, minimum score of 16.00 and a maximum score of 65.00. The scores on the Simultaneous/Gv scale followed the normal distribution with skewness of .67 and kurtosis of .77. The internal consistency of the Simultaneous/Gv scale which is made up of Triangles and Block Counting was .48. For the performance of children on the Planning/Gf scale, descriptive statistics showed that children between the ages of 8.00 years to 9 years 11months obtained a mean score of 30.31 with a standard deviation of 20.65, standard error of 3.83, minimum score of 3.00 and a maximum score of 84.00. The scores on the Planning/Gf scale did not follow the normal distribution with skewness of 1.06 and kurtosis of -84. The internal consistency of the scale which is made up of Story Completion and Pattern Reasoning was .87. University of Ghana http://ugspace.ug.edu.gh 53 . Finally, on the Knowledge/Gc scale of the KABC-II, children between the ages of 8.00 years to 9 years 11months obtained a mean score of 55.10 with a standard deviation of 14.75 standard error of 2.74, minimum score of 32.00 and a maximum score of 84.00. The scores on the scale followed the normal distribution with skewness of .09 and kurtosis of -1.13. The internal consistency of the Knowledge/Gc scale which is made up of Expressive Vocabulary and Verbal Knowledge was .72. University of Ghana http://ugspace.ug.edu.gh 54 . Table 4: Descriptive Statistics and Internal Consistencies of KABC-II Scales and Subtests for 10-12 years 11 months Variables N Mean SD SE Min Max Skew Kurt (α) LEARNING/Glr 34 142.76 29.03 4.98 80.00 182.00 -.49 -1.01 .79 Atlantis 34 82.71 17.70 3.03 40.00 106.00 -.88 .21 Rebus 34 60.06 14.03 2.41 24.00 77.00 -.71 -.36 SEQUENTIAL/ Gsm 34 28.53 5.22 .89 17.00 37.00 -.79 -.04 .53 Number Recall 34 15.21 3.05 .52 9.00 21.00 -.13 -.64 Hand Movements 34 13.32 3.28 .56 7.00 21.00 -.33 .06 SIMULTANEOUS/Gv 34 45.97 13.49 2.31 18.00 67.00 -.18 -1.03 .67 Triangles 34 26.44 8.29 1.42 7.00 41.00 -.27 -.55 Block Counting 34 19.53 7.22 1.24 9.00 31.00 .22 -1.59 PLANNING/Glr 34 49.00 23.31 3.99 13.00 87.00 -.04 -1.21 .91 Story Completion 34 22.12 11.69 2.00 4.00 43.00 .08 -1.12 Pattern Reasoning 34 28.88 12.17 2.09 8.00 48.00 -.05 -1.26 KNOWLEDGE/Gc 34 68.18 22.53 3.86 27.00 117.00 -.11 -.60 .69 Expressive Vocabulary 34 26.18 5.90 1.01 15.00 38.00 -.08 -.74 Verbal Know 34 42.00 17.22 2.95 10.00 79.00 -.05 -.56 On the Learning/Glr scale of the KABC-II, children between the ages of 10.00 years to 12 years 11months obtained a mean score of 142.76 with a standard deviation of 29.03, standard error of 4.98, minimum score of 80.00 and a maximum score of 182.00. The scores on the Learning/Glr scale followed the normal distribution with skewness of -.49 and kurtosis of -1.01. The internal University of Ghana http://ugspace.ug.edu.gh 55 . consistency of the Learning/Glr scale which is made up of the Atlantis and Rebus subtests was .79. For the Sequential/Gsm scale of the KABC-II, children between the ages of 10.00 years to 12 years 11months obtained a mean score of 28.53 with a standard deviation of 5.22, standard error of .89, minimum score of 17.00 and a maximum score of 37.00. The scores on the Sequential/Gsm scale followed the normal distribution with skewness of -.79 and kurtosis of - .04. The internal consistency of the Sequential/Gsm scale which is made up of Number Recall and Hand Movement is .53. On the performance of children on the Simultaneous/Gv scale, descriptive statistics showed that children between the ages of 10.00 years to 12 years 11months obtained a mean score of 45.97 with a standard deviation of 13.39, standard error of 2.31, minimum score of 18.00 and a maximum score of 67.00. The scores on the Simultaneous/Gv scale followed the normal distribution with skewness of -.18 and kurtosis of -1.03. The internal consistency of the Simultaneous/Gv scale which is made up of Triangles and Block Counting is .67. For their performance on the Planning/Gf scale, descriptive statistics showed that children between the ages of 10.00 years to 12 years 11months obtained a mean score of 49.00 with a standard deviation of 23.31, standard error of 3.99, minimum score of 13.00 and a maximum score of 87.00. The scores on the Planning/Gf scale followed the normal distribution with skewness of -.04 and kurtosis of -1.21. The internal consistency of the Planning/Gf scale which is made up of Story Completion and Pattern Reasoning was .91. Finally, on the Knowledge/GC scale of the KABC-II, children aged between 10.00 years and 12 years 11 months obtained a mean score of 68.18 with a standard deviation of 22.53 standard University of Ghana http://ugspace.ug.edu.gh 56 . error of 3.86, minimum score of 27.00 and a maximum score of 117.00. The scores on the Knowledge/Gc scale followed the normal distribution with skewness of -.60 and kurtosis of .60. The internal consistency of the Knowledge/Gc scale which is made up of Expressive Vocabulary and Verbal Knowledge is .69. Table 5 shows the breakdown of the children‘s performance on Learning/Glr as differentiated by age group, school type and sex. University of Ghana http://ugspace.ug.edu.gh 57 . Table 5: Descriptive Statistics of Age Groups, School Type and Sex on Learning/Glr Sex School Type Age Groups N Mean SD Female Private 6.00-7.11years 7 147.86 23.22 8.00-9.11years 12 164.08 10.19 10.00-12.11years 11 153.82 20.78 Total 30 156.53 18.56 Public 6.00-7.11years 6 99.50 24.78 8.00-9.11years 5 94.20 20.41 10.00-12.11years 5 125.20 35.46 Total 16 105.88 28.94 Total 6.00-7.11years 13 125.54 33.98 8.00-9.11years 17 143.53 35.40 10.00-12.11years 16 144.88 28.48 Total 46 138.91 33.11 Male Private 6.00-7.11years 9 123.56 31.37 8.00-9.11years 4 150.25 31.12 10.00-12.11years 11 150.82 27.15 Total 24 140.50 31.13 Public 6.00-7.11years 5 63.20 20.75 8.00-9.11years 8 101.63 27.44 10.00-12.11years 7 125.29 29.89 Total 20 100.30 35.26 Total 6.00-7.11years 14 102.00 40.48 8.00-9.11years 12 117.83 36.28 10.00-12.11years 18 140.89 30.21 Total 44 122.23 38.44 TOTAL PRIVATE 6.00-7.11years 16 134.19 29.92 8.00-9.11years 16 160.63 17.55 10.00-12.11years 22 152.32 23.64 Total 54 149.41 25.95 PUBLIC 6.00-7.11years 11 83.00 28.96 8.00-9.11years 13 98.77 24.34 10.00-12.11years 12 125.25 30.73 Total 36 102.78 32.27 TOTAL 6.00-7.11years 27 113.33 38.68 8.00-9.11years 29 132.90 37.40 10.00-12.11years 34 142.76 29.03 Total 90 130.76 36.58 Female children obtained a total mean Learning /Glr score of 138.91 with a standard deviation of 33.11, while male children obtained a total mean score of 122.23 with a standard deviation of University of Ghana http://ugspace.ug.edu.gh 58 . 38.44. It was also observed that children attending private schools had a mean Learning/Glr score of 149.41 with a standard deviation of 25.95 while those in public schools had a mean score of 102.78 with a standard deviation of 32.27. For the age categories, children between the ages of 6.00 years to 7 years 11 months had a mean Learning/Glr score of 113.33 with a standard deviation of 38.68. Those between the ages 8.00 years and 9 years 11 months had a mean Learning/Glr score of 132.90 with a standard deviation of 37.40 while children between 10.00 years and 12years 11 months had a mean score of 142.76 with a standard deviation of 29.03. These mean differences were subjected to further statistical analysis using the Three-Way ANOVA. Table 6: Three-Way ANOVA of Age Groups, School Type and Sex on Learning/Glr SOURCE Sum of Squares df Mean Square F ρ η2 Sex 2710.84 1 2710.84 4.32 .041 .052 School Type 43894.49 1 43894.49 69.93 .000 .473 Age Group 12901.74 2 6450.871 10.28 .000 .209 Sex * School Type 84.50 1 84.50 .14 .715 .002 Sex * Age Group 3463.53 2 1731.76 2.76 .070 .066 School Type * Age Group 4242.97 2 2121.49 3.38 .039 .080 Sex * School Type*Age Group 866.09 2 433.04 .69 .505 .017 Error 48961.22 78 627.71 Total 119116.62 89 Table 6 shows that age of children had a significant main effect on their performance in Learning/Glr, [F(2,78) = 10.28, ρ = .00], with a medium effect size of .209. It was also observed that the type of school children attend had a significant main effect on their performance in Learning abilities, [F(1,78) = 69.93, ρ =.00], with a medium effect size of .473. Comparison of University of Ghana http://ugspace.ug.edu.gh 59 . the means shows that children from private schools (Mean = 149.41) did significantly better than children from public schools (Mean = 102.78). Thus, the hypothesis 2a that children from private schools are more likely to perform significantly better than their counterparts from the public schools on the Learning/Glr scales is supported. Further, sex of participants had a statistically significant main effect on their performance on Learning/Glr, [F(1,78) = 4.32, ρ =.04] with a small effect size of .052. Comparison of the means showed that female children obtained a significantly higher mean score (138.91) than male children (122.23) on the Learning/Glr scale. This finding is contrary to the expectations of the study and thus, the hypothesis 3a that there will be no significant differences between the male and female children on the Learning/Glr subscale is not supported. The results showed that Sex and School Type did not have any significant interaction effect on Learning/Glr, [F(1,78) =.14, ρ = .72]. Similarly, Sex and Age did not have any significant interaction effect on children‘s performance on the Learning/Glr scale, F(2,78) =2.76, ρ .07]. However, School Type and Age had a significant effect on children‘s performance on Learning/Glr, [F(2,78) =3.38, ρ = .04] with a small effect size of .080 which indicates that age differences exist in performance of children from private and public schools. On the other hand, the combined effect of Sex * School Type*Age was not significant on children‘s performance in Learning, [F(2,78) = .69, ρ = .51]. Since there are more than two age groups, a multiple comparison was done using Bonferroni test and the results are summed up in Table 7. University of Ghana http://ugspace.ug.edu.gh 60 . Table 7: Multiple Comparisons of Age Groups on their Performance in Learning/Glr AGE GROUPS Mean SD 6.00-7.11years 8.00-9.11years 10.00-12.11years 6.00-7.11years 113.33 38.68 - 19.56* 29.43* 8.00-9.11years 132.90 37.40 - - 9.86 10.00-12.11years 142.76 29.03 - - - *= Significant at the .05 alpha level Children between the ages of 10.00-12.11years obtained a significantly higher mean (142.76) than children between the ages of 6.00 and 7.11years (Mean=113.33), t = 29.43, ρ < .05. Therefore, the hypothesis 1a which states that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 6.00-7.11years on the Learning/Glr scale is supported. A significant mean difference was also observed between children of 6.00-7.11years (Mean=113.33) and children between 8.00 and 9.11years (Mean=132.90), t = 19.56, ρ < .05, thus supporting the hypothesis 1c that children between the ages of 8.00-9.11years are more likely to perform better than children between the ages of 6.00- 7.11years on Learning/Gsm. Conversely, no significant mean difference was found between children of 8.00-9.11years (Mean=132.90) and children of 10.00-12.11years (142.76), t = 9.86, ρ > .05. Therefore, hypothesis 1b which states that children between the ages of 10.00-12.11 years are more likely to perform better than children between the ages of 8.00-9.11 years is not supported. University of Ghana http://ugspace.ug.edu.gh 61 . The effects of Age Groups, School Type and Sex on Sequential/Gsm, the means and standard deviations were computed and the results are presented in Table 8. Table 8: Descriptive Statistics of Age Groups, School Type and Sex on Sequential/Gsm SEX School Type New Age Group Mean SD N FEMALE Private 6.00-7.11years 26.29 6.92 7 8.00-9.11years 29.33 5.35 12 10.00-12.11years 31.55 1.69 11 Total 29.43 5.08 30 Public 6.00-7.11years 22.00 5.06 6 8.00-9.11years 19.60 5.32 5 10.00-12.11years 26.40 3.58 5 Total 22.63 5.24 16 Total 6.00-7.11years 24.31 6.29 13 8.00-9.11years 26.47 6.90 17 10.00-12.11years 29.94 3.38 16 Total 27.07 6.05 46 MALE Private 6.00-7.11years 22.89 6.49 9 8.00-9.11years 38.00 14.63 4 10.00-12.11years 30.18 5.23 11 Total 28.75 9.15 24 Public 6.00-7.11years 19.40 5.94 5 8.00-9.11years 23.88 5.00 8 10.00-12.11years 22.71 5.12 7 Total 22.35 5.31 20 Total 6.00-7.11years 21.64 6.31 14 8.00-9.11years 28.58 11.07 12 10.00-12.11years 27.28 6.28 18 Total 25.84 8.23 44 TOTAL PRIVATE 6.00-7.11years 24.38 6.68 16 8.00-9.11years 31.50 8.88 16 10.00-12.11years 30.86 3.86 22 Total 29.13 7.11 54 PUBLIC 6.00-7.11years 20.82 5.36 11 8.00-9.11years 22.23 5.36 13 10.00-12.11years 24.25 4.75 12 Total 22.47 5.21 36 TOTAL 6.00-7.11years 22.93 6.32 27 8.00-9.11years 27.34 8.75 29 10.00-12.11years 28.53 5.22 34 Total 26.47 7.18 90 University of Ghana http://ugspace.ug.edu.gh 62 . It was observed that female children obtained a total mean Sequential/Gsm score of 27.07 with a standard deviation of 6.05 while male children obtained a total mean score of 25.84 with a standard deviation of 8.23 on the same scale. It was also observed that children from private school had a mean Sequential/Gsm score of 29.13 with a standard deviation of 7.11, while children from public schools had a mean score of 22.47 with a standard deviation of 5.21. For the age categories, children between the ages of 6.00-7.11years had a mean Sequential/Gsm score of 22.93 with a standard deviation of 6.32; children between the ages of 8.00-9.11years had a mean Sequential/Gsm score of 27.34 with a standard deviation of 8.75 and children between 10.00- 12.11 years had a mean Sequential/Gsm score of 28.53 with a standard deviation of 5.22. Table 9: Three-Way ANOVA of Age Groups, School Type and Sex on Sequential /Gsm Source Sum of Squares Df Mean Square F ρ η2 Sex 1.99 1 1.99 .06 .809 .001 School Type 1085.42 1 1085.42 31.96 .000 .291 Age Group 449.43 2 224.71 6.62 .002 .145 Sex * School Type 19.41 1 19.41 .57 .452 .007 Sex * Age Group 360.92 2 180.46 5.31 .007 .120 School Type * Age 212.69 2 106.34 3.13 .049 .074 Sex * School * Age 21.48 2 10.74 .32 .730 .008 Error 2649.25 78 33.97 Total 4588.40 89 The age of children had a significant main effect on their performance in Sequential/Gsm, [F(2,78) = 10.28, ρ = .00], with medium effect size of .209. The type of school children attended had a significant main effect on their performance in Sequential/Gsm, [F(1,78) = 31.96, ρ = .00], with a medium effect size of .291. Comparison of the means shows that children from private University of Ghana http://ugspace.ug.edu.gh 63 . schools (Mean = 29.13) did significantly better than children from public schools (Mean = 22.47). Thus, the hypothesis 2b that children from private schools are more likely to perform significantly better than their counterparts from the public schools on the Sequential/Gsm scale is supported. In addition, sex of participants did not have any significant main effect on their performance on Sequential/Gsm, [F(1,78) = .06, ρ =.81] with a small effect size of .001. Comparison of the means showed that female and male children obtained similar mean scores on the Sequential/Gsm scale. Thus hypothesis 3b which states that there will be no significant difference in between the performance of male and female children on the Sequential/Gsm scale is supported. The results showed that Sex and School Type did not have any significant interaction effect on Sequential /Gsm, [F(1,78) =.57, ρ .45]. It was observed that Sex and Age had a significant interaction effect on children‘s performance on Sequential /Gsm, [F(2,78) =5.31, ρ = .01] with a small effect size of .120. This indicates that sex differences exist among the various age categories. Similarly, School Type and Age had a significant interaction effect on children‘s performance on Sequential/Gsm, [F(2,78) =3.13, ρ = .50] with a small effect size of .074 which indicates that age differences exist in performance of children from private and public schools. On the other hand, the combined effect of Sex * School Type*Age was not significant on children‘s performance in Sequential/Gsm, [F(2,78) = .32, ρ = .73]. Since the age groups are more than two, a multiple comparison was done using Bonferroni test and the results are summarized in Table 10. University of Ghana http://ugspace.ug.edu.gh 64 . Table 10: Multiple Comparisons of Age Groups on their Performance in Sequential/Gsm AGE GROUPS Mean SD 6.00-7.11years 8.00-9.11years 10.00-12.11years 6.00-7.11years 22.93 6.32 - 4.42* 5.60* 8.00-9.11years 27.34 8.75 - - 1.18 10.00-12.11years 28.53 5.22 - - - Children between the ages of 10.00-12.11years obtained a significantly higher mean (28.53) than children between the ages of 6.00 and 7.11years (Mean=22.93), t = 5.60, ρ < .05. Thus, the hypothesis 1a that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 6.00-7.11years is supported. A significant mean difference was also observed between children of 6.00-7.11years (Mean=22.93) and children between 8.00 and 9.11years (Mean=27.34), t = 4.42, ρ < .05. Consequently, the hypothesis 1c that children between the ages of 8.00-9.11years are more likely to perform better than children between the ages of 6.00-7.11years is supported. However, no significant mean difference was found between children of 8.00-9.11years (Mean=27.34) and children of 10.00-12.11years (28.53), t = 1.18, ρ > .05. Therefore, hypothesis 1b which expects that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 8.00- 9.11years is not supported. University of Ghana http://ugspace.ug.edu.gh 65 . Table 11 demonstrates the effects of Age Groups, School Type and Sex performance on Simultaneous/Gv. Table 11: Descriptive Statistics of Age Groups, School Type and Sex on Simultaneous/Gv SEX School Type AGE GROUP MEAN SD N FEMALE Private 6.00-7.11years 28.14 13.92 7 8.00-9.11years 38.00 12.19 12 10.00-12.11years 55.00 8.41 11 Total 41.93 15.42 30 Public 6.00-7.11years 22.33 8.98 6 8.00-9.11years 25.00 6.78 5 10.00-12.11years 41.60 17.01 5 Total 29.19 13.87 16 Total 6.00-7.11years 25.46 11.82 13 8.00-9.11years 34.18 12.28 17 10.00-12.11years 50.81 12.86 16 Total 37.50 15.97 46 MALE Private 6.00-7.11years 25.11 6.81 9 8.00-9.11years 40.00 7.62 4 10.00-12.11years 47.36 11.01 11 Total 37.79 13.56 24 Public 6.00-7.11years 18.60 9.61 5 8.00-9.11years 32.50 9.58 8 10.00-12.11years 32.71 10.70 7 Total 29.10 11.32 20 Total 6.00-7.11years 22.79 8.21 14 8.00-9.11years 35.00 9.37 12 10.00-12.11years 41.67 12.87 18 Total 33.84 13.20 44 TOTAL PRIVATE 6.00-7.11years 26.44 10.23 16 8.00-9.11years 38.50 11.02 16 10.00-12.11years 51.18 10.33 22 Total 40.09 14.64 54 PUBLIC 6.00-7.11years 20.64 9.00 11 8.00-9.11years 29.62 9.12 13 10.00-12.11years 36.42 13.73 12 Total 29.14 12.33 36 TOTAL 6.00-7.11years 24.07 10.00 27 8.00-9.11years 34.52 11.00 29 10.00-12.11years 45.97 13.49 34 TOTAL 35.71 14.71 90 University of Ghana http://ugspace.ug.edu.gh 66 . It was observed that female children obtained a total mean Simultaneous/Gv score of 37.50 with a standard deviation of 15.97 while male children obtained a total mean Simultaneous/Gv score of 33.84 with a standard deviation of 13.20. It was also observed that children from private schools had a mean Simultaneous/Gv score of 40.09 with a standard deviation of 14.64 while children from public schools had a mean Simultaneous/Gv score of 29.14 with a standard deviation of 12.33. For the age categories, children between the ages of 6.00-7.11years had a mean Simultaneous/Gv score of 24.07 with a standard deviation of 10.00; those between the ages of 8.00-9.11years had a mean Simultaneous/Gv score of 34.52 with a standard deviation of 11.00 while children between 10.00-12.11 years had a mean score of 45.97 with a standard deviation of 13.49. These mean differences were subjected to further statistical analysis using the Three-Way ANOVA. Table 12: Three-Way ANOVA of Age Groups, School Type and Sex on Simultaneous Processing Source Sum of Squares Df Mean Square F ρ η2 Sex 10565.04 1 105.39 .94 .334 .012 School Type 91555.06 1 2054.26 18.41 .000 .191 Age Group 105.39 2 2976.77 26.67 .000 .406 Sex * School Type 2054.26 1 6.98 .06 .803 .001 Sex * Age Group 5953.54 2 287.94 2.58 .082 .062 School Type * Age 6.98 2 108.07 .97 .384 .024 Sex * School * Age 575.88 2 22.53 .20 .818 .005 Error 216.13 78 111.61 Total 19270.49 89 University of Ghana http://ugspace.ug.edu.gh 67 . Age of children had a significant main effect on their performance in Simultaneous/Gv, [F(2,78) = 26.67, ρ = .00], with medium effect size of .409. The type of school children attended had a significant main effect on their performance in Simultaneous/Gv, [F(1,78) = 18.41, ρ = .00], with a small effect size of .191. Comparison of the means shows that children from private schools (Mean = 40.09) did significantly better than children from public schools (Mean = 29.14). Thus, the hypothesis 2c which premised that children from private schools are more likely to perform significantly better than their counterparts from the public schools on the Simultaneous/Gv scale is supported. Additionally, sex of participants did not have any significant main effect on their performance on Simultaneous/Gv, [F(1,78) = .94, ρ = .33] with a small effect size of .012. Comparison of the means showed that female and male children obtained similar mean scores on the scale. Therefore, hypothesis 3e which states that male children are more likely to perform better than females on the Simultaneous/Gc scale is not supported. The results showed that Sex and School Type did not have any significant interaction effect on Simultaneous/Gv, [F(1,78) =.06, ρ = .80]. It was also observed that Sex and Age did not have any significant interaction effect on children‘s performance on Simultaneous/Gv, [F(2,78) =2.58, ρ = .08]. This indicates that no sex differences exist among the various age categories. Similarly, School Type and Age did not have any significant interaction effect on children‘s performance on Simultaneous/Gv, [F(2,78) = .97, ρ = .38], which indicates that no age differences exist in performance of children from private and public schools. The combined effect of Sex * School Type*Age was not significant on children‘s performance on the Simultaneous/Gv scale, [F(2,78) = .20, ρ = .82]. Since the age groups are more than two, a multiple comparison was done using Bonferroni test and the results are summed up in Table 13. University of Ghana http://ugspace.ug.edu.gh 68 . Table 13: Multiple Comparisons of Age Groups on their Performance in Simultaneous/Gv AGE GROUPS Mean SD 6.00- 7.11years 8.00- 9.11years 10.00-12.11years 6.00-7.11years 24.07 10.00 - 10.44* 21.90* 8.00-9.11years 34.52 11.00 - - 11.45* 10.00-12.11years 45.97 13.49 - - - *= significant at the .05 alpha level Children between the ages of 10.00-12.11years obtained a significantly higher mean (45.97) than children between the ages of 6.00 and 7.11years (Mean=24.07), t = 21.90, ρ < .05. Thus, the hypothesis 1a that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 6.00-7.11years is supported. A significant mean difference was also observed between children of 6.00-7.11years (Mean=24.07) and children between 8.00 and 9.11years (Mean=34.52), t = 10.44, ρ < .05. Thus, the hypothesis 1c children between the ages of 8.00-9.11years are more likely to perform better than children between the ages of 6.00-7.11years is supported. A significant mean difference was found between children of 8.00-9.11years (Mean= 34.52) and children of 10.00-12.11years (Mean = 45.97), t = 11.45, ρ < .05. Thus, hypothesis 1b that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 8.00-9.11years is supported. University of Ghana http://ugspace.ug.edu.gh 69 . To examine the effects of Sex, School Type and Age Groups on Planning/Gf, the means and standard deviations were computed and the results are presented in Table 14. Table 14: Descriptive Statistics of Age Groups, School Type and Sex on Planning SEX School Type Age Group Mean SD N FEMALE Private 6.00-7.11years 29.86 17.88 7 8.00-9.11years 42.67 17.19 12 10.00-12.11years 59.09 13.25 11 Total 45.70 19.27 30 Public 6.00-7.11years 15.67 15.47 6 8.00-9.11years 12.80 8.76 5 10.00-12.11years 28.20 16.42 5 Total 18.69 14.75 16 Total 6.00-7.11years 23.31 17.72 13 8.00-9.11years 33.88 20.47 17 10.00-12.11years 49.44 20.19 16 Total 36.30 21.93 46 MALE Private 6.00-7.11years 20.67 19.34 9 8.00-9.11years 37.50 31.52 4 10.00-12.11years 64.73 19.80 11 Total 43.67 29.26 24 Public 6.00-7.11years 10.60 7.37 5 8.00-9.11years 19.13 10.87 8 10.00-12.11years 23.29 9.50 7 Total 18.45 10.41 20 Total 6.00-7.11years 17.07 16.49 14 8.00-9.11years 25.25 20.70 12 10.00-12.11years 48.61 26.36 18 Total 32.20 25.83 44 TOTAL PRIVATE 6.00-7.11years 24.90 18.70 16 8.00-9.11years 41.38 20.51 16 10.00-12.11years 61.91 16.70 22 Total 44.80 23.99 54 PUBLIC 6.00-7.11years 13.36 12.18 11 8.00-9.11years 16.69 10.23 13 10.00-12.11years 25.33 12.40 12 Total 18.56 12.33 36 TOTAL 6.00-7.11years 20.07 17.06 27 8.00-9.11years 30.31 20.65 29 10.00-12.11years 49.00 23.31 34 TOTAL 34.30 23.87 90 University of Ghana http://ugspace.ug.edu.gh 70 . Female children obtained a total mean Planning/Gf score of 36.30 with a standard deviation of 21.93 while male children obtained a total mean Planning/Gf score of 32.20 with a standard deviation of 25.83. It was also observed that children from private school had a mean Planning/Gf score of 44.80 with a standard deviation of 23.99 while children from public schools had a mean score of 18.56 with a standard deviation of 12.33. For the age categories, children between the ages of 6.00-7.11years had a mean Planning/Gf score of 20.07 with a standard deviation of 17.06. Those aged between 8.00-9.11years had a mean Planning/Gf score of 30.31 with a standard deviation of 20.65 while children between 10.00-12.11 years had a mean Planning/Gf score of 49.00 with a standard deviation of 23.21. These mean differences were subjected to further statistical analysis using the Three-Way ANOVA. Table 15: Three-Way ANOVA of Age Groups, School Type and Sex on Planning/Gf Source Sum of Squares Df Mean Square F ρ η2 Sex 84.93 1 84.93 .32 .576 .004 School Type 11629.85 1 11629.85 43.22 .000 .357 Age Group 8820.46 2 4410.23 16.39 .000 .296 Sex * School Type 14.22 1 14.22 .05 .819 .001 Sex * Age Group 252.42 2 126.21 .47 .627 .012 School Type * Age 2022.53 2 1011.27 3.76 .028 .088 Sex * School * Age 436.60 2 218.50 .81 .448 .020 Error 20988.05 78 269.08 .32 Total 50704.90 89 University of Ghana http://ugspace.ug.edu.gh 71 . Age of children had a significant main effect on their performance in Planning, [F(2,78) = 16.39, ρ = .00], with medium effect size of .296. The type of school attended had a significant main effect on the children‘s performance in Planning/Gf, [F(1,78) = 43.22, ρ = .00], with a medium effect size of .357. Comparison of the means shows that children from private schools (Mean = 44.80) did significantly better than children from public schools (Mean = 18.56), therefore, supporting hypothesis 2d which states that children from private schools are more likely to perform significantly better than their counterparts from the public schools on the Planning/Gf scale. Furthermore, sex of participants did not have any significant main effect on their performance on Planning/Gf, [F(1,78) = .32, ρ = .58] . Comparison of the means showed that female and male children obtained similar mean scores on Planning/Gf. Hypothesis 3c which postulates that there will be no significant difference in performance of male and female children on the Planning/Gf scale is supported. The results revealed that Sex and School Type did not have any significant interaction effect on Planning/Gf, [F(1,78) =.05, ρ = .82]. It was also observed that Sex and Age did not have any significant interaction effect on children‘s performance on the Planning/Gf scale, [F(2,78) = .47, ρ .63. This indicates that no sex differences exist among the various age categories in their performance in Planning/Gf. Conversely, School Type and Age had a significant interaction effect on children‘s performance on Planning/Gf, [F(2,78) = 3.76, ρ = .03], indicating that age differences exist in performance of children from private and public schools. The combined effect of Sex * School Type*Age was not significant on children‘s performance in Planning/Gf, [F(2,78) = .81, ρ = .45]. Since the age groups are more than two, a multiple comparison was done using Bonferroni test and the results are presented in Table 16. University of Ghana http://ugspace.ug.edu.gh 72 . Table 16: Multiple Comparisons of Age Groups on their Performance in Planning/Gf AGE GROUPS Mean SD 6.00- 7.11years 8.00- 9.11years 10.00-12.11years 6.00-7.11years 20.07 17.06 - 10.24 28.93* 8.00-9.11years 30.31 20.65 - - 18.69* 10.00-12.11years 49.00 23.31 - - - *= significant at the .05 alpha level Examination of the multiple comparisons Table 16 above shows that children between the ages of 10.00-12.11years obtained a significantly higher mean (49.00) than children between the ages of 6.00 and 7.11years (Mean=20.07), t = 28.93, ρ < .05. No significant mean difference was observed between children of 6.00-7.11years (Mean= 20.07) and children between 8.00 and 9.11years (Mean=30.31), t = 10.44, ρ > .05. Consequently, the hypothesis 1c that children between the ages of 8.00-9.11years are more likely to perform better than children between the ages of 6.00-7.11years is not supported. However, a significant mean difference was however was found between children of 8.00-9.11years (Mean= 30.31) and children of 10.00-12.11years (Mean = 49.00), t = 18.69, ρ < .05. Therefore, hypothesis 1b that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 8.00- 9.11years is supported. University of Ghana http://ugspace.ug.edu.gh 73 . To examine the effects of Sex, School Type and Age Groups on Knowledge/Gc, the means and standard deviations were computed and the results are presented in Table 17 below: Table 17: Descriptive Statistics of Age Groups, School Type and Sex on Knowledge/Gc SEX School Type New Age Group Mean SD N FEMALE Private 6.00-7.11years 58.86 17.51 7 8.00-9.11years 68.58 7.18 12 10.00-12.11year 80.73 11.77 11 Total 70.77 14.30 30 Public 6.00-7.11years 40.50 20.26 6 8.00-9.11years 40.40 6.43 5 10.00-12.11year 47.40 16.47 5 Total 42.63 15.20 16 Total 6.00-7.11years 50.38 20.37 13 8.00-9.11years 60.29 14.87 17 10.00-12.11year 70.31 20.48 16 Total 60.98 19.81 46 MALE Private 6.00-7.11years 50.11 11.43 9 8.00-9.11years 59.50 9.75 4 10.00-12.11year 80.64 20.25 11 Total 65.67 21.09 24 Public 6.00-7.11years 26.40 6.07 5 8.00-9.11years 41.88 6.96 8 10.00-12.11year 43.71 8.01 7 Total 38.65 9.96 20 Total 6.00-7.11years 41.64 15.19 14 8.00-9.11years 47.75 11.49 12 10.00-12.11year 66.28 24.64 18 Total 53.39 21.61 44 TOTAL PRIVATE 6.00-7.11years 53.94 14.57 16 8.00-9.11years 66.31 8.56 16 10.00-12.11year 80.68 16.16 22 Total 68.50 17.65 54 PUBLIC 6.00-7.11years 34.09 16.56 11 8.00-9.11years 41.31 6.52 13 10.00-12.11year 45.25 11.72 12 Total 40.42 12.53 36 TOTAL 6.00-7.11years 45.85 18.07 27 8.00-9.11years 55.10 14.75 29 10.00-12.11year 68.18 22.53 34 TOTAL 57.27 20.94 90 University of Ghana http://ugspace.ug.edu.gh 74 . From Table 17, it was observed that female children obtained a total mean Knowledge/Gc score of 60.98 with a standard deviation of 19.81 while male children obtained a total mean Knowledge/Gc score of 53.39 with a standard deviation of 21.61. It was also observed that children from private schools had a mean Knowledge/Gc score of 68.50 with a standard deviation of 17.65 whereas those attending public schools had a mean Knowledge/Gc score of 40.42 with a standard deviation of 12.53. For the age categories, children between the ages of 6.00-7.11years had a mean Knowledge/Gc score of 45.85 with a standard deviation of 18.07; those aged between 8.00-9.11years had a mean Knowledge/Gc score of 55.10 with a standard deviation of 14.75 while the children between 10.00-12.11 years of age had a mean score of 68.18 with a standard deviation of 22.53. These mean differences were subjected to further statistical analysis using the Three-Way ANOVA, with a summary of the results presented in Table 18. Table 18: Three-Way ANOVA of Age Groups, School Type and Sex on Knowledge/Gc Source Sum of Squares Df Mean Square F ρ η2 Sex 649.66 1 649.66 3.78 .055 .046 School Type 13862.95 1 13862.95 80.72 .000 .509 Age Group 5176.75 2 2588.38 15.07 .000 .279 Sex * School Type 1.45 1 1.44 .07 .927 .000 Sex * Age Group 343.02 2 171.51 1.00 .373 .025 School Type * Age 834.47 2 417.23 2.43 .095 .059 Sex * School * Age 240.09 2 120.04 .70 .500 .018 Error 13395.79 78 171.74 3.78 Total 39033.60 89 University of Ghana http://ugspace.ug.edu.gh 75 . An examination of Table 18 above shows that age of children had a significant main effect on their performance in Knowledge/Gc, [F(2,78) = 15.07, ρ = .00], with a medium effect size of .279. The type of school children attend had a significant main effect on their performance in Knowledge, [F(1,78) = 80.72, ρ .00], with a large effect size of .509. Comparison of the means showed that children from private schools (Mean = 68.50) did significantly better than children from public schools (Mean = 40.42). Hence, the hypothesis 2e that children from private schools are more likely to perform significantly better than their counterparts from the public schools on the Knowledge/Gc scale is supported. Additionally sex of participants did not have any significant main effect on their performance on the Knowledge/Gc scale, [F(1,78) = 3.78, ρ .06]. Comparison of the means showed that female and male children obtained similar mean scores on Knowledge/Gc, thus supporting the hypothesis 3d that there will be no significant difference between the performances of male and female children on the KABC-II scales except on the Simultaneous/Gv scale. The results further revealed that Sex and School Type did not have any significant interaction effect on Knowledge/Gc, [F(1,78) =.07, ρ = .93]. It was also observed that Sex and Age did not have any significant interaction effect on children‘s performance on the Knowledge/Gc scale, [F(2,78) = 1.00, ρ = .37]. This indicates that no sex differences exist among the various age categories in their performance in Knowledge/Gc. Additionally, School Type and Age did not have a significant interaction effect on children‘s performance on Knowledge, [F(2,78) = 2.43, ρ =.10], indicating that no age differences exist in performance of children from private and public schools on Knowledge/Gc. The combined effect of Sex * School Type*Age was also not significant on children‘s performance in Knowledge/Gc, [F(2,78) = .81, ρ = .50]. Since there are University of Ghana http://ugspace.ug.edu.gh 76 . more than two age groups, a multiple comparison was done using Bonferroni test and the results précised in Table 19. Table 19: Multiple Comparisons of Age Groups on their Performance in Knowledge/Gc AGE GROUPS Mean SD 6.00- 7.11years 8.00- 9.11years 10.00-12.11years 6.00-7.11years 45.85 18.07 - 9.25* 22.32* 8.00-9.11years 55.10 14.75 - - 13.07* 10.00-12.11years 68.18 22.53 - - - *= significant at the .05 alpha level Examination of the multiple comparisons in Table 19 shows that children between the ages of 10.00-12.11years obtained a significantly higher mean (68.18) than children between the ages of 6.00 and 7.11years (Mean=45.85), t = 22.32, ρ < .05. As a result, hypothesis 1a which proposes that children between the ages of 10.00-12.11years are more likely to perform better than those aged between 6.00-7.11years, is supported. As expected, children between sages 8.00 and 9.11years of age had a higher mean (Mean=55.10) than children between 6.00 and 7.11years of age (Mean= 45.85), t = 9.25, ρ < .05, thereby supporting, the hypothesis 1c that children between the ages of 8.00-9.11years are more likely to perform better than those aged between 6.00- 7.11years A significant mean difference was also found between children of 8.00-9.11years (Mean= 55.10) and children of 10.00-12.11years (Mean = 68.18), t = 13.07, ρ < .05. Thus, the hypothesis 1b that children between the ages of 10.00-12.11years are more likely to perform better than children between the ages of 8.00-9.11years is supported. University of Ghana http://ugspace.ug.edu.gh 77 . Table 20: Paired sampled t-test for differences in performance on MPI and FCI among private school children Variable N Mean SD Df t Ρ MPI FCI 54 54 23.40 22.58 4.00 4.02 53 2.24 .03 Analysis of the results from the correlated t-table above showed that a significant difference exists in performance of children from private schools on the MPI and FCI at the .05 alpha level, t(53) = 2.24, ρ < .05. Comparing the two means (MPI=23.40 and FCI=22.58), it indicated that performance of the children from private basic schools on the FCI was significantly better than their performance on the MPI thereby supporting hypothesis 4a. Table 21: Paired sampled t-test for differences in performance on MPI and FCI among public school children Variable N Mean SD Df t Ρ MPI FCI 36 36 16.92 16.61 3.27 3.11 35 3.33 .002 Analysis of the results from the correlated t-table above showed that a significant difference exists in performance of children from public schools on the MPI and FCI at the .05 alpha level, t(35) = 19.36, ρ < .05. Comparing the two means (MPI=16.92 and FCI=16.61), it was shown that performance of the children from public schools on the MPI was significantly better than their University of Ghana http://ugspace.ug.edu.gh 78 . performance on the FCI. This consequently supports hypothesis 4b which expected a better performance on the MPI than the FCI. 4.1 Summary of Key Findings Children between the ages of 10 and 12 years 11months did significantly better than children between the ages of 6 years and 7 years 11months on all the five scales (Learning/Glr, Sequential/Gsm, Simultaneous/Gv, Planning/Gf and Knowledge/Gc) of the KABC-II. However, children between the ages of 10 and 12years 11 months performed significantly better than children between the ages of 8years and 9 years 11 months only on the Simultaneous/Gv, Planning/Gf and Knowledge/Gc scales but not on the Learning/Glr and Sequential/Gsm scales of the KABC-II. Similarly, children between the ages of 8 and 9years 11years did significantly better than those aged between 6 and 7 years 11months on the Learning/Glr, Sequential/Gsm, Simultaneous/Gv and Knowledge/Gc but not on the Planning/Gf scales. No significant sex differences were found between male and female children in their performances on Sequential, Simultaneous, Planning and Knowledge subscales of KABC-II. However, female children performed significantly better than male children on their performance on the Learning/Glr scale of KABC-II. Children from private schools performed significantly better than children from public schools on all the five scales (Learning/Glr, Sequential/Gsm, Simultaneous/Gv, Planning/Gf and Knowledge/Gc) of the KABC-II. School type and age categories had significant interaction effects on children‘s performances on Learning/Glr, Sequential/Gsm and Planning/Gf. It was also observed that sex and age categories of children had a significant interaction effect on their performance on Sequential/Gsm. For the different global index scales, children from private schools performed significantly better on the FCI than the MPI whereas those in the public schools did significantly better on the MPI than the FCI. University of Ghana http://ugspace.ug.edu.gh 79 . CHAPTER FIVE DISCUSSION Misra, Sahoo, & Puhan (1997) observe that many tests that originate from the United States or Europe are used extensively in developing countries without any prior scientific or practical examination of their suitability for these countries or cultures. Similarly, the use and application of Western psychological tests in Ghana has become a common phenomenon with majority of them not having been validated to obtain local norms for the Ghanaian population (Edwin, 2001). One such psychological test of interest is the Kaufman Assessment Battery for Children, Second Edition (KABC-11), which is increasingly being employed by many practitioners both globally and locally. The study examined the performance of Ghanaian school-children on five scales of the KABC-II and investigated whether children‘s demographic characteristics such as age, sex and type of school have any significant effect on their performances on the five subscales. Three major hypotheses were formulated and tested. The first hypothesis compared three age categories on their performances on the five scales of the KABC-II; the second compared children from private and public schools‘ performances on the five scales of KABC-II while the third hypothesis compared male and female children‘s performance on the same scales. 5.1 Age Differences and Performance on KABC-II Scales To examine whether age significantly affects children‘s performances on the five scales of the KABC-II, it was hypothesized that older children are more likely to perform significantly better than younger children. As previously reported, there were significant age differences between the performances of the children aged 10 to 12 years 11 months and the 6 to 7 years 11 month olds. These significant age differences in cognitive abilities among the school children could be partly attributable to better control and direction of their thinking as the prefrontal cortex matures. University of Ghana http://ugspace.ug.edu.gh 80 . Naturally, children grow and as they do, maturation and experience combine to make them more articulate, reflective, effective and more competent (Bandura, 2001). As this takes place, problem solving abilities also improve. However, the improvement is not always linear (Salkind, 1981; Shaffer, 1989). In this study, children between the ages of 10.00 and 12.11years did significantly better than children between the ages of 8.00 and 9.11years on Simultaneous/Gv, Planning/Gf and Knowledge/Gc but not on the Learning/Glr and Sequenstial/Gsm scales of KABC-II. This means that while we expect differences across various ages of children in their cognitive abilities, the differences were not pronounced on every facet of their cognitive domain. As this study pointed out, no significant differences exist between the age categories in Learning and Sequential subscales which assess children‘s ability to successfully complete different types of learning tasks such Immediate-recall, delayed recall as well as ability to solve problems by remembering and using an ordered series of images or ideas respectively. This could be due in part to the different start points for examinees according to age, as stipulated by the KABC-II, or to Brunner‘s theory that anyone including very young children can learn any material so long as the instruction is appropriately organized (Wood, Bruner & Ross, 1976). Similarly, children between the ages of 8.00 and 9.11years did significantly better than children between the ages of 6.00-7.11years in Learning/Glr, Sequential/Gsm, Simultaneous/Gv and Knowledge/Gc but not on the Planning/Gf scales. This clearly indicates that at younger ages the differences in cognitive abilities are pronounced compared to when children are growing older with these differences not very apparent across the age groups. This may be because of the expectation that as children grow up to a certain age, their cognitive abilities remain relatively stable. An alternative explanation may be that the frontal lobes of the brain which are responsible University of Ghana http://ugspace.ug.edu.gh 81 . for the function of planning may not yet be fully developed for both of these age groups (Rowe, Bullock, Polkey & Morris, 2001). The age differences that were noted between younger and older children in their performance on the subscales of the KABC-II are consistent with some earlier studies that reported older children to perform significantly better than younger children on cognitive abilities. For instance, in their research which employed product knowledge as an explanation for age-related differences in children's cognitive responses to advertising, Costley and Brucks (1987) noted that older children were better able than younger ones to ―distinguish TV advertising from programming, exhibit a greater degree of skepticism towards advertising and recall a greater amount of commercial contents‖. These observed differences have frequently been attributed to cognitive development, most specifically, Piaget's theory of cognitive stages (Rubin 1974; Ward, Wackman & Wartella 1977). 5.2 Type of School and Performance on the KABC-II Scales Investigating the effect of school type on children‘s performance on the subtests of the KABC-II, the results showed that children from private schools performed significantly better than children from public schools on all the five scales of the KABC-II. Comparison of performance of type of school on the different global scales of the KABC-II also indicated that children from private schools did significantly better in the FCI than the MPI while the converse was true for those in public schools. These results can be attributed to the differences between the two types of schools in terms of the style of teaching, availability of essential facilities and parental SES (Asiedu, 2002; Edwin, 2001; Castilo et al., 2011). The teaching methods in private schools for instance, employ the incorporation of lessons that include pictures, video, and images via University of Ghana http://ugspace.ug.edu.gh 82 . technology which are more likely to engage children, in their curricula. Due to the perennial non- availability of such materials, the typical public school in Ghana relies largely on the traditional "tell-read-write" and ―run-o‘- the- mill‖ lessons which make learning difficult for children in such schools (Asiedu, 2002; Westera, 2011). Teachers in private schools are more qualified and are better supervised by both their principals and guardians of their students, than their colleagues in public schools. The parents of children in the private schools tend to be better educated, have better jobs and thus earn higher incomes than parents who patronize the public schools. As a result of their higher educational and SES, parents of private school children are able to afford the materials necessary to create a stimulating intellectual environment (e.g. TVs, computers, internet access) for their children in order to reduce possible ensure optimal and equal development, social support and opportunities to succeed related to cognitive performance in their offspring (Castilo et al., 2011). Higher parental SES implies more exposure for the private school children as they are able to travel around the world, and in the process, acquiring more exposure to current trends which positively impacts on their cognition. In contrast, children of low socioeconomic status are generally provided with few play materials at home, resulting in suboptimal stimulation at best (Malda, Van de Vijver, Srinivasan, Transler, & Sukumar, 2009). This coupled with an educational system which focuses on collective rote learning such as pertains in many public schools in Ghana, could account for the poor child‘s lack of experience with tools such as puzzles, building blocks, geometric figures and even individual test situations (Malda et al., 2009). The deprived environments in which most children from the public schools find themselves tend to slow down their cognitive developmental process (Anum, 1996). Generally, a person‘s level of income determines their quality of life; low income may lead to residence in very poor neighbourhoods, many of which are characterized by social University of Ghana http://ugspace.ug.edu.gh 83 . disorganization and limited resources for child development. In many countries, not excepting Ghana, public schools tend to attract parents who fall within the lower income bracket and invariably, who also have a lower level of education. Such parents are unable to provide adequate opportunities for optimum growth of their children. For instance, children in low income families are more likely to suffer nutritional and hence, growth problems than their counterparts in higher income families ((Brooks-Gunn & Duncan, 1997). Malnourished children are characteristically less responsive to adults, less inspired to learn, and less active in exploration than their more amply nourished counterparts in higher income families. Although the effects of poor nutrition on intelligence may be indirect, they are commonly associated with low achievement which is, in turn, linked to poor cognitive ability (Neisser, Boodoo, Bouchard, Boykin, Brody, Ceci, Halpern, Loehlin, Perloff, Sternberg & Urbina, 1996). The impact of this is felt in Africa much more than anywhere else in the world (Lynn, 2006). In their research, Brooks- Gunn and Duncan (1997) also ascertained that children living below the poverty threshold were 1.3 times more likely to experience learning disabilities and developmental delays than those above the poverty threshold. Furthermore, the effects of poverty on children‘s cognitive development occur early. The findings of this study thus concur with a recent one where school type and parental educational level were found to be related to attention and memory abilities, better development of verbal abilities and vocabulary acquisition in a sample of Spanish adolescents (Villaseñor, Martín, Díaz, Rosselli & Ardila, 2009). This indicates that the significant differences in cognitive abilities between children who attend private schools and those in public schools, is not limited to Ghana but pertains to many other cultures globally, as studies elsewhere (e.g. Brooks-Gunn & University of Ghana http://ugspace.ug.edu.gh 84 . Duncan, 1997; Anger & Heineck, 2010; Matute, Montiel, Pinto, Rosselli, Ardila & Zarabozo, 2012) have reported. 5.3 Sex Differences in performance on KABC-II Scales Sex differences in specific cognitive abilities are well documented in the literature (Halpern, 2000; Kaufman & Lichtenberger, 2002). According to Kaufman and Kaufman (2004), with the exception of the Knowledge/Gc scale where no sex difference is found, younger female children tend to perform slightly better than their male counterparts on all the global scales and on all the scales of the KABC-II. This observation is in line with the concept that young girls‘ cognitive development is advanced compared to that of boys of similar age (Kaufman & Kaufman, 2004). From the ages of seven to 18 years however, there is very little difference in their performance. Despite the overlap in cognitive performances of both sexes, several meta-analyses demonstrate that, on average, males outperform females in certain spatial tasks especially in mental rotation (Masters & Sanders, 1993; Voyer, Voyer, & Bryden, 1995). This was portrayed in the KABC-II, by males obtaining higher scores than females on the Rover subtest and Simultaneous/Gv scale (Kaufman & Kaufman, 2004). Females, on the other hand, perform better than males in specific aspects of verbal abilities, such as verbal fluency, verbal memory and perceptual speed (Feingold, 1992; Hedges & Nowell, 1995). On the KABC-II, females outperformed males on the Rebus and Story Completion subsets. The differences in performance between males and females on the KABC-II nonetheless offset each other resulting in a statistically insignificant sex difference in the 7 to 18 years age range (Kaufman & Kaufman, 2004). Based on this premise, it was hypothesized that there will be no significant sex differences in the performance of male and University of Ghana http://ugspace.ug.edu.gh 85 . female children on all the KABC-II scales except the Simultaneous/Gv scale where males will be more likely to perform better than females. The results from the analysis showed that male and female children did not differ significantly in their performances on four (Sequential/Gsm, Simultaneous/Gv, Planning/Gf and Knowledge/Gc) scales of the KABC-II. This nullifies hypotheses 3a, which expected males to perform better than females on the Simultaneous/Gv scale. From hypotheses 3a, 3b, 3c, 3d it was anticipated that there would be no significant difference between the performance of males and females on the Learning/Gf, Sequential/Gsm, Planning,/Gf and Knowledge/Gc scales respectively. However, contrary to expectations, female children performed significantly better than male children on the Learning/Glr scale. The unremarkable sex differences observed between male and female children in their cognitive abilities show that unlike previously thought, there are more similarities between males and females than differences in abilities across several domains. This could be attributed to the decreasing stereotyping of male and female roles children in the modern world. For instance, both sexes now engage in activities that were previously reserved for one particular sex such as playing football and other cognitive stimulating games without being frowned upon by society. The female superiority in their performance on the Learning/Glr scale comprising Atlantis and Rebus is worth highlighting. This scale measures a child‘s ability to successfully complete different types of learning tasks involving immediate recall and delayed recall tasks (i.e. the CHC narrowed ability of associative memory and learning abilities). The significantly better performance of the females can be attributed to the inherent verbal nature of the tasks involved in recalling previously memorized materials; a practice which usually favours females over males. It is however, worthy of note the effect size of sex on performance on the Learning/Glr subscale University of Ghana http://ugspace.ug.edu.gh 86 . is small (η2 = .052), suggesting that the difference between the male and female children is not very substantial. Relating the findings to previous literature on sex differences in the performance of males and females on cognitive tasks, these findings of no significant sex differences are consistent with some earlier studies. For instance, Ardilla et al., (2011) and Wallentin, (2009) found in their respective studies that no significant sex differences exist between male and female children in their cognitive abilities. This is however inconsistent with other studies such as Halpern (1997) who found that females performed better than males on verbal tasks while males performed better than females on visuo-spatial tasks. Although the origins of these cognitive sex differences are not yet fully understood, they seem to arise from a complex interaction of biological, social, and psychological factors (Halpern, 2000). 5. 4 The interaction effects of the independent variables (Age, Type of School and Sex) on the KABC-II scales To determine whether the independent variables in the study combined to significantly affect the performance of children on the subscales of the KABC-II, the results showed that only the interaction between School Type and Age categories had significant effects on children‘s performances on Learning/Glr, Sequential/Gsm and Planning/Gf. With regard to the Learning/Glr scale, the means showed that children between the ages of 8.00 and 9.11years from private schools had the highest mean (160.63), followed by children between the ages of 10.00-12.11years in private schools (Mean=152.32), children between the ages of 6.000 and 7.11years in private schools, children between the ages of 10.00-12.11years in public schools (Mean = 125.25), children between 8.00-9.11years in public schools (Mean = 98.77) and University of Ghana http://ugspace.ug.edu.gh 87 . the least performing group was children between the ages of 6.00-7.11years in public schools (Mean = 83.00). It is interesting to note that youngest age-group of children who attend private schools did significantly better than all the children who attend public schools including their young counterparts in the public schools as well as those in the older age-groups. These findings bring to bear the impact of socioeconomic factors on cognitive abilities the majority of children from the private schools are from homes that can be classified as socioeconomically very good, and most of those in public schools, from lower socioeconomic backgrounds. This concurs with earlier studies that reported on the impacts of school types and socioeconomic factors on children‘s intelligence (e.g. Edwin, 2001; Adote, 1996; Ardila, Rosselli, Matute & Guarjardo, 2005). Noteworthy, while the pattern of children from public schools followed the expected trend, that of private school children did not, as children between the ages of 8.00 and 9.11years from the latter performed (Mean = 160.63) better than their 10.00-12.11 years old peers in similar schools (Mean = 152.32). On the performances in Sequential/Gsm , the results showed that a similar pattern to that observed in Learning/ abilities, the youngest group (6.00-7.11years) from private schools performing significantly better than all the three age groups from public schools. Remarkably, the children between the ages of 8.00 and 9.11years from private schools had the highest mean followed by children between the ages of 10.00 and 12.11years from privates. This again, could be attributed to the effect of school types and socioeconomic status on cognitive performance. Performance on the Planning/Gf scale however, took a different trend with private-school children between the ages of 10.00-12.11years obtaining the highest mean of 61.91 followed by those aged between 8.00 and 9.11years (Mean = 41.38). The children between the ages of University of Ghana http://ugspace.ug.edu.gh 88 . 10.00-12.11years in public schools had the next highest mean, (25.33), closely followed by the 6.00-7.11year olds in private schools (Mean = 24.90). The groups with the lowest means were children in public schools aged 8.00 - 9.11years (Mean = 16.69) and 6.00-7.11years (Mean = 13.36). This finding indicates that though school type affected performance among children of different ages, the effect is not as much as that for children‘s performances on Learning/Glr and Sequential/Gsm. A second major interaction effect was observed for children‘s sex and age on their performances in Sequential/Gsm. The study found that differences exist between the age groups in their performance on Sequential/Gsm processing as a result of their sex. Females performed better than males in two out of the three age categories, thus highlighting their traditional superiority in verbal tasks. 5.5 Implications of the study Findings from the study imply that the KABC-II would be appropriate as a psychological measure in Ghana if some adaptations or modifications of the test could be done to make it as culturally sensitive as it proposes to be. It would also have to be accurately standardized and validated on Ghanaian children before being used. Taking the two school types investigated, the KABC-II appropriately discriminated between them, with the less economically advantaged groups performing worse as was expected. With the more culturally- fair attention tests, although significant differences existed between the two groups, this was not so pronounced. To be able to use it accurately and confidently in Ghana, the KABC-II will have to be standardized locally, taking into consideration the effect of school type, SES and the various cultural backgrounds of children in the country, on performance. University of Ghana http://ugspace.ug.edu.gh 89 . The study outcomes have implications for clinical practice, educational assessment, researchers interested in cognitive and intellectual assessment of children, as well as for the educational system in the country. The findings from this study imply that in assessing children with these psychological tests that have been developed and standardized for the Western countries, the interpretation should be done in respect of locally derived norms as the norms developed elsewhere may not be applicable to the Ghanaian child. In order to do this, such tests, invariably, need to be first standardized locally. In lieu of that, clinical decisions have to be carefully considered, taking into account the client‘s socio-demographic characteristics which can have a significant impact on their cognitive and intellectual abilities. For researchers interested in assessment of the cognitive abilities of children, the type of school (private/public), socioeconomic status and their ages should be taken into consideration as significant differences were found in the performances of children due to these characteristics. In other words, psychological assessment should not be merely focused on the individual child‘s learning problem but should also consider the structural and sociocultural inequalities that could hinder that child's ability to acquire knowledge. 5.6 Limitations and Recommendations The study was not without limitations. Foremost was the small sample size which was inadequate to establish local norms for the KABC-II in Ghana. This was as a result of the limited access to school children and the time required to comprehensively exhaust the tests administered. Another limitation was the lack of a screening test which would have enabled the appropriate subjects to be selected for inclusion in the study. University of Ghana http://ugspace.ug.edu.gh 90 . In order to achieve the aim of validating and establishing local norms for the KABC-II in Ghana, it is recommended that a brief screening of school children be conducted initially and then a large enough sample representative of the population of school children in Ghana be appropriately selected to participate in the study. Due to these shortfalls, it is encouraged that future studies build on this initial project by sampling more children from both healthy and clinical samples to provide practitioners and researchers with valid bases for their assessment of children using the KABC-II. 5.7 Conclusions This study sought to determine the performance of Ghanaian school-children on the KABC-II administering the test to a sample of pupils from public and private schools. Although the attempt at validating and obtaining local norms for the test in Ghana was not successful due to the challenges cited above, the objectives of investigating the effects of the socio-demographic characteristics of school-children on their performance on KABC-II were achieved and highlighted. The study provides the needed springboard for larger studies to be commenced in trying to make the KABC-II a valid test for clinical and educational purposes. It was found from the study that the age and type of school the children attend, as well as their socioeconomic backgrounds had significant impacts on their performance on the KABC-II scales. These findings suggest that in assessing children with the KABC-II, it is imperative to take into consideration the individual differences that could influence cognitive abilities to avoid drawing erroneous conclusions. University of Ghana http://ugspace.ug.edu.gh 91 . REFERENCES Anastasi, A. (1990). Psychological testing (6th ed.). New York: Macmillan Anastasi, A. & Urbina, S. (1997). Psychological Testing (7th ed.). Upper Saddle River (NJ): Prentice Hall. Akhouri, D., Jahan, M. R, & Singh, A. R (2006): Applicability on Kaufman Assessment Battery for Children (K-ABC) on Normal Children in India. Indian Journal of Clinical Psychology, 33 (1), 54-59. Anger, S., & Heineck, G. (2010a). Do Smart Parents Raise Smart Children? The Intergenerational Transmission of Cognitive Abilities, Journal of Population Economics, 23(3), 1105-1132. Anum, A. (1996). A normative Study of the Raven‘s Coloured Progressive Matrices among School Children in Ghana. (Master of Philosophy thesis, Department of Psychology, University of Ghana, Legon. Ardila, A. (1996). Towards a cross-cultural neuropsychology. Journal of Social and Evolutionary Systems, 19(3), 237-248. Ardila, A., Ostrosky-Solís, F., Rosselli, M.,, & Gómez, C. (2000). Age related cognitive decline during normal aging: The complex effect of education. Archives of Clinical Neuropsychology,15, 495–514. Ardila, A., Rodriguez-Menendez, G., & Rosselli, M. (2002). Current issues in neuropsychological assessment with Hispanics/Latinos. In: Ferraro F.R. (Ed.). Minority University of Ghana http://ugspace.ug.edu.gh 92 . and cross-cultural aspects of neuropsychological assessment. Lisse, Netherlands: Swets & Zeitlinger; pp. 161–179. Ardila, A., Rosselli, M., Matute, E., & Guajardo, S. (2005). The influence of the parents' educational level on the development of executive functions. Developmental Neuropsychology, 28(1), 539-560. Ardila, A., Rosselli, M., Matute, E., & Inozemtseva, O. (2011). Gender differences in cognitive development. Developmental Psychology, 47(4):984-90. Asiedu, S. (2002). A comparative study of public and private schools in the provision of quality education at the basic level in urban centres in Ghana. Unpublished MSc. Thesis, Department of Planning, Kwame Nkrumah University of Science and Technology. Bandura, Albert. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1-26. Bangirana P., Seggane-Musisi, Allebeck P., Giordani B., John C. C., Opoka O. R., Byarugaba J., Ehnvall A., & Boivin, M. J. (2009). A Preliminary Examination of the Construct Validity of the KABC-II in Ugandan Children with a History of Cerebral Malaria. African Health Sciences, 9(3),186–192. Barac, R., & Bialystok, E. (2012). Bilingual Effects on Cognitive and Linguistic Development: Role of Language, Cultural Background, and Education. Child Development, 83, 413– 422. Boone, J.A., & Adesso, V.J. (1974). Racial differences on a black intelligence test. Journal of Negro Education,43, 429-536. 362 University of Ghana http://ugspace.ug.edu.gh 93 . Bowman, M. (1989). Testing individual differences in ancient China. American Psychologist, 44, 576 – 578. Brickman, A. M., Cabo, R., & Manly, J. J. (2006). Ethical issues in cross-cultural neuropsychology. Applied Neuropsychology, 13, 91-100. Brooks-Gunn, J. & Duncan, G. J. (1997). The future of children. Future of Children & Poverty, 7(2),55-71. Brooks-Gunn, J., Guo, G., & Furstenberg, F. (1993). Who drops out of and who continues beyond high school? Journal of Research on Adolescence, 3, 271–294. Caplan, P. J., Crawford, M., Hyde, J. S., & Richardson, J. T. E. (1997). Gender differences in human cognition. New York: Oxford University Press. Carroll, J.B. (1976). Psychometric tests as cognitive tasks: A new ―structure of intellect‖. In L.B. Resnick (Ed.), The nature of intelligence (pp. 27-56). Hillsdale, NJ: Erlbaum. Carroll, J. B. (1997). The three-stratum theory of cognitive abilities. In Flanagan, Dawn P. (Ed); Genshaft, Judy L. (Ed); Harrison, Patti L. (Ed), (1997). Contemporary intellectual assessment: Theories, tests, and issues (pp. 122-130). New York, NY, US: Guilford. Carstairs, J. R., Myors, B., Shores, E. A., & Fogarty, G. (2006), Influence of language background on tests of cognitive abilities: Australian data. Australian Psychologist, 4, 48–54. Castilo, R., Ruiz, J. R., Chillon, P., Jimenez-Pavon, D., Esperanza-Diaz, L., Moreno, L. A., & Ortega, E.B. (2011). Associations between parental education /occupational levels and University of Ghana http://ugspace.ug.edu.gh 94 . cognitive performance in Spanish adolescents. The AVENA study. Piscotherma 23(3), 349-355. Chin, A. L., Negash, S., Xie,S., Arnold, S. E., & Hamilton, R. (2012). Quality, and not just quantity, of education accounts for differences in psychometric performance between African Americans and White Non-Hispanics with Alzheimer‘s disease. Journal of the International Neuropsychological Society, 18(2):277-285. Cooper, R. S. (2005). Race and IQ: Molecular Genetics as Deus ex Machina. American Psychologist, 60, 71–76. Costley, L.C., & Brucks, M. (1987). Product knowledge as an explanation for age-related differences in children's cognitive responses to advertising. In M. Wallendorf and P. Anderson (Eds.). NA - Advances in Consumer Research 14, (288-292) Provo, UT : Association for Consumer Research. Cronshaw, S. F., Hamilton, L. K., Onyura, B. R., & Winston, A. S. (2006). Case for non-biased intelligence testing against Black Africans has not been made. International Journal of Selection and Assessment, 14, 278–287. Das, J. P., Kar, B. C., & Parrila, R. K. (1996). Cognitive planning. New Delhi: Sage Publications. Das, J. P., Kirby, J. R., & Jarman, R. F. (1975). Simultaneous and successive syntheses: An alternative model for cognitive abilities. Psychological Bulletin, 82, 87-103. University of Ghana http://ugspace.ug.edu.gh 95 . Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment of Cognitive Processes. Allyn & Bacon, Publishers, Needham Heights: MA, USA. Dasen, P. (1984). The cross-cultural study of intelligence. Piaget and the Baoule. International Journal of Psychology. 19, 407-434. De Beer, M. (2000). The construction and evaluation of a dynamic computerised adaptive test for the measurement of learning potential. Unpublished doctoral dissertation, University of South Africa, Pretoria. Deary, I.J., Thorpe, G., Wilson, V., Starr, J.M., & Whalley, L.J. (2003). Population sex differences in IQ at age 11: the Scottish mental survey 1932. Intelligence, 31, 533–542. Debra, L. (2002). Standardization of the Multidimensional Aptitude Battery. (Unpublished M.Phil. thesis, Department of Psychology, University of Ghana, Legon. Dickens, W. T., & Flynn, J. R. (2006). Black Americans reduce the racial IQ gap: Evidence from the standardization samples. Psychological Science, 17, 913–920. Duncan, G. J., Yeung, W. J., Brooks-Gunn, J., & Smith, J. R. (1998). How much does childhood poverty affect the life chances of children? American Sociological Review, 63, 406 – 423. Edwin, A. K. (2001). Is the Wechsler Intelligence Scale for Children (WISC-IVUK) Applicable? Unpublished MPhil.thesis, Department of Psychology, University of Ghana. Escobedo, P. S., & Hollingworth, L. (2009). Annotations on the Use of the Mexican Norms for the WAISIII, Applied Neuropsychology,16 , 3, 223 — 227. University of Ghana http://ugspace.ug.edu.gh 96 . Feingold, A. (1992). Sex-differences in variability in intellectual abilities—A new look at an old controversy. Review of Educational Research, 62, 61–84. Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2007). Essentials of cross-battery assessment. (2nd Edition). New Jersey: John Wiley & Sons, Inc. Fleisch, A. (2007). Orientational clitics and the expression of path in TashelhitBerber (Shilha). Annual Publication in African Linguistics, 5, 55-72. Fletcher-Janzen, E. (2003). A Validity Study of the Kaufman Assessment Battery for Children, Second Edition (KABC-II) and the Taos Pueblo Indian Children of New Mexico. Circle Pines, MN: AGS Publishing. Geary, D.C. (1996). Sexual selection and sex differences in mathematical abilities. Behavioral and Brain Sciences, 19, 229-284. Giordani, B., Boivin, M., Opel, B., Nseyila, D., & Lauer, R. (1996). Use of the K-ABC with children in Zaire, Africa: an evaluation of the sequential-simultaneous processing distinction within intercultural context. International Journal of Disability, Development and Education, 43(1), 5-24. Gottfried, A.W., Gottfried, A.E., Bathurst, K., Guerin, D.W., & Parramore, M.M. (2003). Socioeconomic status in children‘s development and family environment: infancy through adolescence. In M.H. Bornstein & R.H. Bradley (Eds.), Socioeconomic status, parenting and child development (pp. 189–207). Mahwah, NJ: Lawrence Erlbaum Associates. University of Ghana http://ugspace.ug.edu.gh 97 . Gregory, R.J. (1992). Psychological Testing: Psychological Testing: History, Principles, and Applications. Boston: Pearson Education. Gregory, R. L. (1998). The Oxford Companion to the Mind. Oxford University Press, Oxford, UK. Grigorenko, E. L., Geissler, P. W., Prince, R., Okatcha, F., Nokes, C., & Kenny, D. A. (2001).The organization of Luo conceptions of intelligence: A study of implicit theories in a Kenyan village. International Journal of Behavior Development, 25, 367–378. Hackman D. A., & Farah, M.J. (2009). Socioeconomic status and the developing brain. Trends in Cognitive Science,13, 65–73. Halpern, D. F. (1997). Critical Thinking Across the Curriculum: A Brief Edition of Thought and Knowledge. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Halpern, D. F. (2000). Sex differences in cognitive abilities (3rd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Halpern, Diane F. (2000). Validity, fairness, and group differences: tough questions for selection testing. Psychology, Public Policy, and Law, 6, 56-62. Hanes, N., & Norlin, E. (2011). Psychological Traits and Earnings Differentials among Men: A Study of Second-generation Immigrants in Sweden. The Scandinavian Journal of Economics, 113(2), 318-341. Hanscombe, K. B., Trzaskowski, M., Haworth, C. M. A., Davis, O. S. P., Dale, P. S., & Plomin, R. (2012). Socioeconomic status (SES) and children's intelligence (IQ): In a UK- University of Ghana http://ugspace.ug.edu.gh 98 . representative sample SES moderates the environmental, not genetic, effect on IQ . PLoS ONE, 7, e30320. doi:10.1371/journal.pone.0030320. Harris, J. G., & Llorente, A. M. (2005). Cultural considerations in the use of the Wechsler Intelligence Scale for Children–Fourth Edition (WISC-IV). In A. Prifitera, D. H. Hedges, L. V., & Nowell, A. (1995). Sex-differences in mental test-scores, variability, and numbers of high-scoring individuals. Science, 269, 41–45. Helms, J.E. (1992). Why is there no study of cultural equivalence in standardizedcognitive ability testing? American Psychologist, 47(9), 1083-1101. Holding, P.A., Taylor, H.G., Kazungu, S.D., Mkala, T., Gona, J., Mwamuye, B., Mbonani, L., & Stevenson, J. (2004). Assessing cognitive outcomes in a rural African population: Development of a neuropsychological battery. Journal of the International Neuropsychological Society, 10, 246-260. Hunt, E. (2012). What Makes Nations Intelligent? Perspectives on Psychological Science, 7, 284-306. Hunt, E. & Carlson, J. (2007). Considerations Relating to the Study of Group Differences in intelligence. Perspectives on Psychological Science, 2, 194-213 Hyde, J. S. (1981). How large are cognitive gender differences? A meta-analysis using w² and d. American Psychologist, 36, 892-901. Hyde, J. S., & Linn, M. C. (1988). Gender differences in verbal ability: A meta-analysis. Psychological Bulletin, 104, 53–69. University of Ghana http://ugspace.ug.edu.gh 99 . Jackson, Y., & Warren, J. S. (2000). Appraisal, support and life events : Predicting outcome of behaviour in school-age children. Child Development, 71, 1441-1457. Jensen, A. R., & Reynolds, C. R. (1982). Race, social class and ability patterns on the WISCR. Personality and Individual Differences, 3, 423-438. Joppe, M. (2000). The Research Process. Retrieved February 25, 2013, from http://www.ryerson.ca/~mjoppe/rp.htm Kane, H., & Brand, C. (2003). The importance of Spearman‘s g as a psychometric, social and educational construct. The Occidental Quarterly, 3(1), 7-27. Kanjee, A. (1999). Assessment research. In M. Terre Blanche and K. Durrheim (Eds.), Research in practice: Applied methods for the social sciences (pp. 287-306). Cape Town: University of Cape Town Press. Kaplan, R.M., & Saccuzzo, D.P. (2005). Psychological Testing: Principles, applications, and issues. Belmont, CA: Thomson Wadsworth. Kaufman, A.S., & Kaufman, N. L. (2004). Kaufman Assessment Battery for Children- Second Edition Manual. Circle pines, MN: American Guidance Service. Klineberg, O. (1928). An experimental study of speed and other factors in ―racial‖ differences. Archives of Psychology, 15(93), 111. Legg, S., & Hutter, M., (2007). Artificial Intelligence (cs AI). Frontiers in Artificial Intelligence and Applications, 157,17-24. Lezak, M. D. (2005). Neuropsychological assessment (3rd ed.). Oxford: Oxford Press. University of Ghana http://ugspace.ug.edu.gh 100 . Liaw, F.-R., & Brooks-Gunn, J. (1994). Cumulative familial risks and low-birth weight children‘s cognitive and behavioral development. Journal of Clinical Child Psychology, 23 (4), 360–372. Luria, A. R. (1973). The working brain: An introduction to neuropsychology. New York, NY: Basic Books. Lynch, S.A., & Warner, L.(2012). A new theoretical perspective of cognitive abilities. Childhood Education. 88,6. Lynn, R. (2006). Race differences in intelligence: An evolutionary analysis. Augusta, GA: Washington Summit Books. Malda, M., Van de Vijver, F. J. R., Srinivasan, K., Transler, C., & Sukumar, P. (2009). Traveling with cognitive tests: testing the validity of a KABC-II adaptation in India. Assessment, 17(1), 107-115. Manly, J. J.,, & Echemendia, R. J. (2007). Race-specific norms: Using the model of hypertens- sion to understand issues of race, culture, and education in neuropsychology. Archives of Clinical Neuropsychology, 22, 319-325. Masters, M. S., & Sanders, B. (1993). Is the gender difference in mental rotation disappearing? Behavior Genetics, 23, 337–341. Matthews, J. S., Ponitz, C. C., & Morrison, F. J. (2009). Early gender differences in self- regulation and academic achievement. Journal of Educational Psychology, 101(3), 689- 704. University of Ghana http://ugspace.ug.edu.gh 101 . Matute, E., Montiel, T., Pinto, N., Rosselli, M., Ardila, A., & Zarabozo, D. (2012). Cognitive development on illiterate children. International Review of Education, 1-19. Mayer, J. D. & Salovey, P. (1993). The intelligence of emotional intelligence. Intelligence 17, 433-442. McGrew, K. S. (2005). The Cattell-Horn-Carroll theory of cognitive abilities: Past, present, and future. In D. P. Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), Contemporary intellectual assessment: Theories, tests, and issues (136–182). New York: Guilford. McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53(2),185-204. McKown, C., Gregory, A., & Weinstein, R.S. (2010). Expectations, stereotypes, and self- fulfilling prophecies in classroom and school life. In J. Meece & J. Eccles (Eds.) Handbook of research on schools, schooling, and human development. Misra, G., Sahoo, F. M., & Puhan, B. N. (1997). Cultural bias in testing: India. European Review of Applied Psychology, 47, 309-316. Mitrushina, M., Boone, K. B., Razani, J., & D'Elia, L. F. (2005). Handbook of normative data for neuropsychological assessment, (2nd ed.). New York: Oxford University Press. Mullis, I.V.S., Martin, M.O., Fierros, E.G., Goldberg, A.L., & Stemler, S.E. (2000). Gender Differences in Achievement IEA’s Third International Mathematics and Science Study (TIMSS). Boston: International Study Center, Lynch School of Education, Boston College. University of Ghana http://ugspace.ug.edu.gh 102 . Naglieri, J.A. (1997). Planning, attention, simultaneous, and successive theory and the Cognitive Assessment System: A new theory-based measure of intelligence. In D.P. Flanagan, J.L. Genshaft, & P.L. Harrison (Eds.), Contemporary intellectual assessment (pp.247-267). New York: The Guilford Press. Neisser, U., Boodoo, G., Bouchard T. J., Boykin, A.W., Brody, N., Ceci, S. J.,Halpern, D. F.,Loehlin, J. C., Perloff, R., Sternberg, R. J., & Urbina, S. (1996). Intelligence: Known and Unknowns. American Psychologist, 51(21), 77-101. Nell, V. (2000). Cross-cultural neuropsychological assessment. Theory and practice. Mahwah, NJ: Lawrence Erlbaum Associates. Noble, K.G., Farah, M.J., & McCandliss, B.D. (2006). Socioeconomic background modulates cognition-achievement relationships in reading. Cognitive Development, 21(3), 349-368. Ogbu J.U. (2002). Cultural Amplifiers of Intelligence: IQ and Minority Status in Crosscultural Perspective, J. M. Fish Race and Intelligence: Separating Science from Myth. Mahwah, NJ: Erlbaum. Opoku, J. Y. (in press). Some Theoretical And Practical Problems Associated With Using Western Instruments To Measure Cognitive Abilities On The African Continent. In K. Anyidoho & H. Lauer (Eds.), Re-appraising the Social Sciences and Humanities from an African Perspective (Vol.1). Accra: Sub-Saharan Publishers. Ostrosky, F., Gomez, E., Matute, E., Pineda, D, Rosselli, M., & Ardila, A. (2007). Neuropsi: Attention and Memory: A neuropsychological test battery in Spanish with norms by age and education. Applied Neuropsychology, 14, 3,156-70. University of Ghana http://ugspace.ug.edu.gh 103 . Ostrosky-Solís, E., Ramirez, M., & Ardila, A. (2004). Effects of culture and education on neuropsychological testing: A preliminary study with indigenous and nonindigenous - population. Applied Neuropsychology, 11(4), 186-193. Penny A., Holding, H., Taylor, G., Kazungu, S. D., Mkala, T., Gona, J., Mwamuye, B., Mbonani, L., & Stevenson, J. (2004). Assessing cognitive outcomes in a rural African population: Development of a neuropsychological battery in Kilifi District, Kenya. Journal of the International Neuropsychological Society, 10, 246-260. Perlman, M. D., & Kaufman, A. S. (2000). Assessment of Child Intelligence. In G. Goldstein & M. Hersen (Eds.), Handbook of Psychological Assessment (pp. 59-60). Oxford, UK: Pergamon. Peters, E., Baker, D. P., Dieckmann, N. F., Leon J., & Collins, J. (2010). Explaining the Effect of Education on Health: A Field Study in Ghana. Psychological Science, 21(10), 1369- 1376. Plucker, J. A. (Ed.). (2013). Human intelligence: Historical influences, current controversies, teaching resources. Retrieved February 14th, 2012, from http:/n/www.intelltheory.com Prifitera, A., Weiss, L. G., Saklofske, D. H., & Rolfhus, E. (2005). The WISC-IV in the clinical assessment context. In A. Prifitera, D. H. Saklofske, and L. G. Weiss (Eds.),(2005) WISC-IV clinical use and interpretation. Scientist-practitioner perspectives (pp. 3-32). San Diego, CA: Elsevier Academic Press. University of Ghana http://ugspace.ug.edu.gh 104 . Puente, A., & Ardila, A. (2000). Neuropsychological assessment of Hispanics. In E. Fletcher- Janzen, T. L. Strickland, & C. R. Reynolds (Eds.), The handbook of cross-cultural neuropsychology (pp. 87–104). New York: Plenum Press. Quansah, K. B. (2000). 1999 report on the administration of primary 6 criterion- referenced tests. Accra: Ministry of Education. Reynolds, M. R., Keith, T. Z., Fine, J. G., Fisher, M. E., & Low, J. A. (2007). Confirmatory Factor Structure of the Kaufman Assessment Battery for Children-Second Edition: Consistency with Cattell-Horn-Carroll Theory. School Psychology Quarterly, 22 (4), 511-539. Richardson, T. E. (2007). Howard Andrew Knox: Pioneer of Intelligence Testing at Ellis Island. Columbia University Press. Rosselli, M., & Ardila, A. (2003). The impact of culture and education on nonverbal neuropsychological measurements: A critical review. Brain and Cognition, 52, 326-333. Rowe, A. D., Bullock, P. R., Polkey, C.E., & Morris, R. G. (2001). Theory of mind impairment and their relationship to executive functioning following frontal lobe excision. Brain, 3, 600-616. Rubin, R. S. (1974). The effects of cognitive development on children‘s responses to television advertising. Journal of Business Research, 2, 409-419. Salkind, N. J. (1981) Arnold Gesell and the Maturational Approach.. Theories of Human Development. New York: D. Van Nostrand. University of Ghana http://ugspace.ug.edu.gh 105 . Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition and Personality, 9 (3),185-211. Schooler, C. (1996). Cultural and Social-Structural Explanations of Cross-National Psychological Differences. Annual Review of Sociology, 22, 323-349.. Seagull, L.R., & Emily, J. C. (2006). When a test turns into a trial: Things to keep in mind about psychological testing. Business Law Today, 15(3),12-17. Serpell, R. (1994). Negotiating a fusion of horizons: a process view of cultural validation in developmental psychology. Mind, Culture and Activity, 1, 43-68. Shaffer, David R. (1989). The Concept of Development. Developmental Psychology: Childhood and Adolescence, 4th edition. Pacific Grove, CA: Brooks/Cole, 1989.. Skuy, M., Schutte, E., Fridjhon, P., & O'Carroll, S. (2001). Suitability of published neuropsychological test norms for urban African secondary school students in South Africa. Personality and Individual Differences, 30, 1413-1425. Stead. G. B. (2002). The transformation of psychology in a post-apartheid South Africa: An overview. International Journal of Group Tensions, 31(1), 79-102. Sternberg, R. J. (1997a). The concept of intelligence and its role in lifelong learning and success. AmericanPsychologist, 52(10), 1030-1037. Sternberg, R. J. (1997b). Intelligence and lifelong learning: What's new and how can we use it? American Psychologist, 52(10),1134-1139. University of Ghana http://ugspace.ug.edu.gh 106 . Sternberg, R., Grigorenko, E. L., & Kidd, K. K. (2005.) Intelligence, Race, and Genetics. American Psychologist, 60, 46–59. Strauss, E., Sherman, E. M. S., & Spreen, O. (2006). A Compendium of Neuropsychological Tests: Administration, norms and commentary (3rd ed.). New York: Oxford University Press. Strenze, T. (2007). Intelligence and socioeconomic success: A metanalytic review of longitudinal research. Intelligence, 35, 401-426. Turkheimer, E., Haley, A., Waldron, M., D‘Onofrio, B., & Gottesman, I. I. (2003). Socioeconomic status modifies heritability of IQ in young children. Psychological Science, 14, 623–628. Valencia, R. R. (1995). K-ABC Content Bias: Comparisons between Mexican American and White Children. Psychology in Schools, 32(3), 153-169. Valencia, R. R., & Lopez, R. (1992). Assessment of racial and ethnic minority students: Problems and prospects. In M.Zeidner & R. Most (Eds.), Psychological testing: an inside view, pp. 309-439. Palo Alto, California : Consulting Psychologists Press. Van der Merwe, A. S. (2008). A Comparison of WISC-IV Test Performance for Afrikaans, English and Xhosa speaking South African Grade 7 Learners. Masters Thesis, Rhodes University. Villaseñor, E., Martín, A., Díaz, E., Rosselli, M., & Ardila, A. (2009). Effects of parents‘ educational level, school type and gender on the development of attention and memory. Revista Latinoamericana de Psicología, 41(2), 257-276. University of Ghana http://ugspace.ug.edu.gh 107 . Villaseñor, E., Martín, A., Díaz, E., Rosselli, M., & Ardila, A. (2009). Effects of parents‘ educational level, school type and gender on the development of attention and memory. Revista Latinoamericana de Psicología, 41(2), 257-276. Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex-differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117, 250– 270. Wartella, E., Wackman, D.B., & Ward, S. (1978). Children‘s Consumer Information Processing:Representation of Information from Television Advertisements. In K. Hunt & A. Abor (Eds.), NA-Advances in Consumer Research, 5, 535-539. MI: Association for Consumer Research. Wasserman, J.D., & Becker, K.A. (2000, August). Racial and ethnic group mean score differences on intelligence tests. In J.A. Naglieri (Chair) Making assessment more fair – taking verbal and achievement out of ability tests. Symposium conducted at the annual meeting of the American Psychological Association; Washington, D.C. Wechsler, D. (1944). The measurement of adult intelligence (3rd ed.). Baltimore: Williams & Wilkins. Wechsler, D. (2003). Manual for the Wechsler Intelligence Scale for Children (4th ed.). San Antonio: The Psychological Corporation. Westera, W. (2011). On the changing nature of learning context: Anticipating the virtual extensions of the world. Educational Technology and Society, 14(2), 201-212. University of Ghana http://ugspace.ug.edu.gh 108 . Wood, D., Bruner, J., & Ross, G. (1976). The role of tutoring in problem solving. Journal of Child Psychology and Psychiatry and Allied Disciplines, 17, 89-100. Yang, S., & Sternberg, R. J. (1997a). Conceptions of intelligence in ancient Chinese philosophy. Journbal of Theoretical and Philosophical Psychology, 17, 101–119. Yang, S., & Sternberg, R. J. (1997b). Taiwanese Chinese people‘s conceptions of intelligence. Intelligence, 25, 21–36. University of Ghana http://ugspace.ug.edu.gh 109 . APPENDICES Appendix I: Ethical Clearance from Noguchi Memorial Institute for Medical Research (NMIMR) University of Ghana http://ugspace.ug.edu.gh 110 . Appendix II: Consent Form University of Ghana http://ugspace.ug.edu.gh 111 . Appendix III: Departmental Introductory Letter University of Ghana http://ugspace.ug.edu.gh 112 . APPENDIX IV: KABC-II NB: The researcher could not scan the KABC-II test tool kit because permission to do so from the copyright owner was not attained. University of Ghana http://ugspace.ug.edu.gh