Admissions assessment batteries at most South African higher education institutions augment matriculation results with assessment data from cognitive and non-cognitive measures, in order to identify potentially successful students. The Motivated Strategies for Learning Questionnaire (MSLQ) is a non-cognitive measure, developed in the United States of America to assess the motivational orientation and use of learning strategies among college students. Numerous international studies have confirmed the reliability and validity of the MSLQ. Presently only three quantitative studies have been conducted in the South African higher education context, and in these, researchers have focused on the predictive validity, construct validity and reliability of the MSLQ when used with multicultural and multilingual groups of test-takers. An overview of these studies revealed a confirmation of adequate reliability of the MSLQ, and whereas it was established that nine of the fifteen subscales of the MSLQ are significantly related to the academic success of undergraduate students, limited support was found for the construct validity of the measure. As a result of these findings, a lack of qualitative methods for confirming or disconfirming the quantitative results was noted. The general aim of this study was to explore the applicability of the MSLQ in a higher education South African context, with respect to aspects related to bias, through assessing the item content with specific reference to language and cultural appropriateness. This was to ensure that items on the MSLQ did not favour or disadvantage any cultural and language groups. Investigations in this regard are a necessity in a country like South Africa in an effort to promote fair assessment practices. In order to achieve the above-mentioned aim, a quantitative and qualitative approach was utilised following an exploratory-descriptive research method. The researcher employed a non-probability, purposive sampling procedure to select post-graduate participants employed at the Nelson Mandela Metropolitan University (NMMU) to act as expert reviewers. A Bias Review Form was the primary mode of data collection and this enabled the researcher to collect the biographical details of the expert reviewers, their ratings of the language and cultural appropriateness of each item, and their comments, suggestions and recommendations for improving identified problematic items. Descriptive statistics were used to analyse the reviewers’ responses to the short biographical questionnaire included on the Bias Review Form, while frequency counts were calculated based on the reviewers’ responses to the language and cultural appropriateness of each item. The reviewers’ comments were analyzed into themes using Tesch’s model of content analysis. Three main themes relating to potential bias were identified, namely, language, culture and item characteristics. Guba’s model was employed for assessing the trustworthiness of the qualitative data and for data verification.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:nmmu/vital:9925 |
| Date | January 2008 |
| Creators | Somtsewu, Nolwazi |
| Publisher | Nelson Mandela Metropolitan University, Faculty of Health Sciences |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis, Masters, MA |
| Format | xii, 156 leaves, pdf |
| Rights | Nelson Mandela Metropolitan University |
Page generated in 0.0024 seconds