Distinguishing between attitudes towards Muslims and tolerance of Islamic practices

Attitudes towards Muslims in Western society are often negative and pervasive. Traditionally, research has focused on identifying the causes and predictors of negative attitudes towards Muslims as a group. However, social psychologists are becoming increasingly interested in understanding the causes and predictors of tolerance of Islamic practices. From this research, a need has arisen to investigate both constructs simultaneously in order to disentangle their effects. Research into attitudes towards Muslims and tolerance of Islamic practices are largely independent in the literature. As such, these two constructs have traditionally been measured using separate scales. Existing scales which measure attitudes towards Muslims are problematic in that they often include items which conflate target categories, making it difficult to understand what specifically is being measured. Research exploring tolerance of Islamic practices is a relatively recent addition to the literature, and the construct has been measured inconsistently across the literature. To date, no single scale has been developed to measure both constructs simultaneously. To address this, my thesis had two major aims, which were addressed across two main parts. The first major aim of my thesis, tested across three studies in Part 1, was to develop and validate a psychometrically sound scale to measure both attitudes towards Muslims and tolerance of Islamic practices. Specifically, across Studies 1 to 3, I provided initial psychometric evidence for the 12 item Positive Attitudes and Tolerance towards Muslims Scale (PATMS). This included an exploratory factor analysis (Study 1, n = 315), a confirmatory factor analysis to confirm the dual dimensions of the scale, and validity analyses (Study 2, n = 338), and a test- retest analysis to confirm the temporal stability of the scale (Study 3, n = 40). The second major aim of my thesis, tested across three studies in Part 2, was to evaluate common predictors of negative attitudes and tolerance using the PATMS, and further disentangle the two constructs. Thus, across Studies 4 to 6, I provided additional evidence for the utility of the PATMS when measuring attitudes and tolerance, and provided evidence for the independence of the constructs. Specifically, in Studies 4 (n = 205) and 5 (n = 298), I evaluated the relationship between Christians’ belief in the representation of God and their attitudes towards Muslims and tolerance of Islamic practices. In these studies, I found that Christians’ belief in a Benevolent God (i.e., loving) predicts both positive attitudes towards Muslims and tolerance of Islamic practices, and belief in a Limitless God (i.e., omnipresent) positively predicts tolerance. On the other hand, Authoritarian God (i.e., vengeful) belief and Mystical God (i.e., part of nature) belief predicts less tolerance of Islamic practices. These findings provided evidence for the importance of investigating the impact of specific representations of the Christian God (i.e., rather than general Christian religious belief) when researching attitudes towards Muslims and tolerance of Islamic practices. In Study 6 (n = 327), I evaluated the mediating role of tolerance of Islamic practices in the established relationship between the factors of the Dual Process Model (i.e., Right-Wing Authoritarianism [RWA] & Social Dominance Orientation [SDO]) and attitudes towards Muslims. I found that the relationship between RWA and attitudes towards Muslims was fully mediated by tolerance of Islamic practices, and the relationship between SDO and attitudes towards Muslims was partially mediated by tolerance of Islamic practices. These findings suggest that tolerance of Islamic practices play a key role in the relationship between the ideological factors of the DPM and attitudes towards Muslims. Taken together, these findings provide preliminary evidence for the usefulness of the PATMS in measuring attitudes towards Muslims and tolerance of Islamic practices. These findings also provide evidence for the independence of the two constructs, and the need to measure both constructs separately.