The thesis explores religious attitudes and Muslim identity in Turkey from a social psychological perspective with reference to university students. Religious attitudes are explored in relation to three components: cognitive, behavioural and affective religious attitudes, whereas Muslim identity is examined through macro and micro levels, and observations. In order to investigate these issues, qualitative and quantitative methods are employed. Research hypotheses are developed on the basis of a review of secondary materials related to Islam in the Turkish context, Muslim identity and the measurement of religious attitudes. Primary data for this study are gathered through standardised questionnaires, such as the Religious Attitude Scale, in-depth interviews and observations. The techniques of psychometrics are employed for the fieldwork of this study, carried out among 1149 students in two universities in Turkey. Using sophisticated statistical analyses, test variables are operationalised and research hypotheses are tested. In doing this, a number of demographic and contextual variations, namely gender, age, family incomes, social and educational backgrounds, supplementary religious education and orientations towards both the Diyanet and cemaats, are taken into account as independent variables. Conclusions are drawn on the basis of the results of statistical analyses, as well as using qualitative inferences from in-depth interviews. The thesis also investigates the predictors of religious attitudes. Using a stepwise multiple regression analysis, between about 55% and 75% of variance in religious attitudes of Turkish university students are explained. The greatest amount of variance in religious attitudes is explained by orientation towards the Diyanet, the formal religious institution in Turkey.