Magister Artium (Psychology) - MA(Psych) / There is an overwhelming concern about substance use amongst adolescents across the world, especially in low socio-economic status communities. The South African Substance Use Contextual Risk Instrument was developed to capture the contextual risk factors for adolescents in low socio-economic status South African communities. This study is part of a larger study which aimed to develop and conduct the initial validation of the instrument. In piloting the instrument, both English and isiXhosa mother tongue speakers were tested using the English version since the instrument is only available in English. The equivalence of the measure across the English and isiXhosa mother tongue speakers is however unknown. The purpose of this study is therefore to assess the scalar equivalence of the English version across the English and isiXhosa mother tongue speakers. The total sample was 674 consisting of 420 English and 247 isiXhosa language groups from low socio-economic status communities in Cape Town. The study employed the Hoteling’s T square test (to assess significant difference of means between the groups), the equality of reliabilities (to assess the significance of differences between the scale reliabilities) and the Tucker’s Phi coefficient of congruence (to assess the congruence of the construct across the two groups). In assessing the mean differences, the results revealed that there were significant mean differences, with the isiXhosa-speaking group performing significantly lower than the English-speaking group for most of the scales. Internal consistency was also generally lower for the isiXhosa group. The structural congruence revealed that there was incongruence between the two language groups for most of the scales with an exception of two of the twenty one scales. It can thus be concluded that the measure cannot be accepted as structurally equivalent across the two groups. It is clear that bias exists in the majority of the scales of the SASUCRI and that this version is thus not applicable for an isiXhosa speaking sample. The study recommends that the instrument is adapted for the isiXhosa speaking group. / National Research Foundation (NRF)

Magister Artium (Psychology) - MA(Psych) / The fall of apartheid and the rise of democracy have brought assessment issues in multicultural societies to the forefront in South Africa. The rise of multicultural assessment demands the development of tests that are culturally relevant to enhance fair testing practices, and issues of bias and equivalence of tests become increasingly important. This study forms part of a larger project titled the Additive Bilingual Education Project (ABLE). The Woodcock Munoz Language Survey (WMLS) was specifically selected to evaluate the language aims in the project, and was adapted from English to isiXhosa. Previous research has indicated that one of the scales in the adapted isiXhosa version of the WMLS, namely the Picture Vocabulary Scale (PV), displays some item bias, or differential item functioning (DIF), across rural and urban isiXhosa learners. Research has also indicated that differences in dialects can have an impact on test takers’ scores. It is therefore essential to explore the structural equivalence of the adapted isiXhosa version of the WMLS on the PV scale across rural and urban isiXhosa learners, and to ascertain whether DIF is affecting the extent to which the same construct is measured across both groups. The results contribute to establishing the scalar equivalence of the adapted isiXhosa version of the WMLS across rural and urban isiXhosa-speaking learners. Secondary Data Analysis (SDA) was employed because this allowed the researcher to re-analyse the existing data in order to further evaluate construct equivalence. The sample of the larger study consisted of 260 learners, both male and female, selected from a population of Grade 6 and 7 learners attending schools in the Eastern Cape. The data was analysed by using the statistical programme Comprehensive Exploratory Factor Analysis (CEFA) and the Statistical Package for Social Sciences (SPSS). Exploratory factor analysis and the Tucker’s phi coefficient were used. The results indicated distinct factor loadings for both groups, but slight differences were observed which raised concerns about construct equivalence. Scatter plots were employed to investigate further, which also gave cause for concern. It was therefore concluded that construct equivalence was only partially attained. In addition, the Cronbach’s Alpha per factor was calculated, showing that internal consistency was displayed only for Factor 1 and not for Factor 2 for the rural group, or both factors for the urban group. Scalar equivalence across the two groups must therefore be explored further.

Magister Psychologiae - MPsych / In many countries defined by multilingualism, language has been identified as a great influence during psychological and educational testing. In South Africa (SA), factors such as changes in policies and social inequalities also influence testing. Literature supports the translation and adaptation of tests used in such contexts in order to avoid bias caused by language. Different language versions of tests then need to be evaluated for equivalence, to ensure that scores across the different language versions have the same meaning. Differences in dialects may also impact on the results of such tests.Results of an isiXhosa version of the Woodcock Muñoz Language Survey (WMLS),which is a test used to measure isiXhosa learners’ language proficiency, show significant mean score differences on the test scores across rural and urban firstlanguage speakers of isiXhosa. These results have indicated a possible problem regarding rural and urban dialects during testing. This thesis evaluates the item bias of the subtests in this version of the WMLS across rural and urban isiXhosa learners. This was accomplished by evaluating the reliability and item characteristics for group differences, and by evaluating differential item functioning across these two groups on the subtests of the WMLS. The sample in this thesis comprised of 260 isiXhosa learners from the Eastern Cape Province in grade 6 and grade 7, both males and females. This sample was collected in two phases: (1) secondary data from 49 rural and 133 urban isiXhosa learners was included in the sample; (2) adding to the secondary data, a primary data collection from 78 rural isiXhosa learners was made to equalise the two sample groups. All ethical considerations were included in this thesis. The results were surprising and unexpected. Two of the subtests in the WMLS showed evidence of scalar equivalence as only a few items were identified as problematic. However, two of the subtests demonstrated more problematic items. These results mean that two subtests of the WMLS that demonstrated evidence of scalar equivalence can be used to measure the construct of language proficiency, while the other two sub-tests that showed problematic items need to be further investigated, as the responses given by learners on these items seem to be determined by their group membership and not by their ability.