The application of the Social Axiom Survey ll (SASll) in the South African context

Abstract

The SASII which is a measure of social axioms, a concept based on an understanding of the core construct of general beliefs, was investigated in this study. The objective was to determine whether the SASII model fits the data collected in a South African context and whether it measures the same social axiom constructs of the a priori five-factor model. Secondly, the study tested the SASII for configural, metric, scalar and full invariance. Thirdly, social axioms' nomological network was investigated by relying on personality factors to validate social axioms. This study also tested the linkages between social axioms and the Big-Five personality factors. Finally, social axioms' ability to predict personality across cultures was investigated. Beliefs are social in nature and are universally shared amongst individuals within cultures. Shared beliefs represent how people organise their world and make sense of and interpret social realities. Values are widely used in the conceptualisation of cultures as this allows comparisons of the value profiles of individuals socialised into different cultures. Personality represent self-views and is related to culture. Establishing the relationship between social axioms and the constructs of personality and values is an important area for research and intervention. The first article on generalised beliefs, often referred to as social axioms, was published in 2002. It described the axiomatic nature of beliefs based on truth assumptions, personal experience and socialisation. The article sparked a global research interest on social axioms designed to evaluate the universality and meaning of the structure of beliefs across cultures. This interest has resulted in an array of articles, chapters and even a book on the subject. South Africa is ideally positioned for cross-cultural research because of its 11 official languages, diversity in educational levels and unemployment rates, inequality and a divided society due to the historical legacy of apartheid. Language differences create specific environments, defined in terms of culture, race, ethnic grouping, values and attitudes. Globalisation has accelerated the rate of intercultural contact, guiding research attempts to comprehensively describe the emerging cross-cultural dynamics. Recent research findings indicate that South Africans exhibit extreme social axiom scores that are similar to trends in other developing countries. Social axioms research also seeks to assess overlap with other constructs such as values and personality. Efforts to refine social axioms and to expand their nomological network are therefore the main focus of this study. This study aims to contribute towards previous research efforts to improve the validity and expand the nomological network of social axioms through assessing the relationship between generalised beliefs, namely social axioms (labelled Social Cynicism, Reward for Application, Social Complexity, Fate Control and Religiosity) and personality factors (i.e., Extraversion, Agreeableness, Conscientiousness, Neuroticism and Intellect) in the South African context. A cross-sectional survey design was used to collect data regarding these constructs as experienced by students and their family members and friends. A convenience sample (N = 1567) of university students and their family members and friends participated in this study. The measuring instruments used were the Social Axioms Survey (SASII), Mini-International Personality Item Pool (Mini-IPIP) and a biographical questionnaire. The results of Study 1 indicated that the SASII five factor structure exhibited an unacceptable model fit within the South African context when conducting a confirmatory factor analysis (CFA). However, as an exploratory step, and by creating a parsimonious model, the improved CFA of the SASII presented a fit that improved on previous research findings, suggesting a reasonable fit, taking into account the complexity of the SASII model. Study 2 assessed the measurement invariance (configural, metric, and scalar invariance) of the five factor structure of the SASII parsimonious model across male and female groups in a South African sample. Evidence was obtained through multi-group confirmatory factor analyses, which supported a baseline configural model. This finding indicated that the number of factors and factor structure of the SASII parsimonious model are considered equivalent across male and female groups. Proof for partial metric invariance was found and scalar invariance could not be achieved. Results thus indicated that male and female respondents did not exhibit the same understanding of certain latent constructs and that some items did not have the same meaning for both groups. Males and females also differed in terms of their levels of trust on the SASII's scales of Social Cynicism, Social Complexity and Religiosity. Because scalar invariance could not be achieved, comparisons of the SASII five factors' means across groups could not be computed. This study provided support for previous research findings and indicated that some items and latent factors of the SASII need refinement.