The study of construct validity is particularly relevant in the twenty-first century, as more and more entities in South Africa are using psychometric instruments – instruments which have to be valid and reliable in accordance with the requirements of the Health Professions Council of South Africa (HPCSA). Even though validity and reliability, together with their accompanying aspects, are the two main considerations of a psychometric instrument, the construct validity of an instrument is one of the most important aspects to consider. This study’s focus is therefore, for the most part, based on the construct validity of the Life-Style Questionnaire. Even though the Life-Style Questionnaire is not a very distinguished questionnaire in the South African context, it is, however, a classified and useable questionnaire to determine into which of the five trait-descriptive lifestyle categories an individual belongs. In recent years, the Professional Board has become increasingly concerned about the misuse of assessment measures in South Africa, while recognising the important role of psychological assessment in the professional practice of psychology, as well as for research purposes (Foxcroft&Roodt, 2005:20). The need therefore existed to determine whether or not the Life-Style Questionnaire actually measures what it is supposed to measure, in other words to determine its construct validity. “The Life-Style Questionnaire was developed as an objective means to measure into which of five trait-descriptive lifestyle categories (aggressive, conforming, defensive, individualistic, or resistive) participants belong” (Driscoll&Eckstein, 2007:1). The purpose of this study was to determine the construct validity of the Life-Style Questionnaire, developed by Driscoll and Eckstein, in the South African -Questionnaire, developed by Driscoll and Eckstein, in the South African context. A quantitative descriptive survey design was used to conduct the research. The Life- Style Questionnaire was administered to a non-probability convenience sample consisting of 301 individuals living in South Africa and the results were subjected to factor analysis (FA), item analysis and confirmatory factor analysis (CFA). Various iterations of the (FA) indicated the primary factors for each of the components of the Life-Style Questionnaire. The final FA yielded a questionnaire consisting of five factors. This was confirmed through Kaiser’s eigenvalues and Cattell’s scree plot. The item analysis indicated that it should be considered to remove items 2, 3, 5, 7, 12, 27, 28, 30 and 32 from the Life-Style Questionnaire. After the factor and item analysis, a CFA was conducted. The purpose of the CFA was to determine whether the postulated theoretical model actually fits the observed data. The most common test used to measure the goodness-of-fit of an instrument is the chi-square test. The chi-square test was conducted by using the EQS programme. The results indicated a poor model fit. However, the reliability of the Life-Style Questionnaire was determined and a Cronbach alpha of 0.853 established the instrument’s high level of reliability. Thus, the Life-Style Questionnaire is a reliable, but invalid instrument. This research study contributed to the understanding and importance of construct validity in psychological instruments. It is important to realise that instruments do not necessarily measure what they are intended to measure and therefore they have to be investigated. Lastly, this study not only emphasises the importance of psychometric properties of psychological instruments, but also the important role that psychometrists and industrial psychologists play with regard to the development and use such instruments.
Dissertation (MCom)--University of Pretoria, 2013.