(Non-)Religious Bias in Secularization Research

The theory of secularization posits that religion declines as modernization unfolds. Once a dominant paradigm, it is now one of the most contested areas in the sociology of religion: even rigorous empirical studies reach contradictory conclusions. This PhD thesis asks whether part of this variability stems not from the social reality under study, but from researchers' own ideological orientations. Specifically, it tests whether scholars' religiosity (or lack of it) constitutes a source of bias in the secularization debate. Drawing on original data from researchers and their publications, the thesis combines survey methods, vignette experiments, and computational text analysis. Chapter 1 provides a systematic overview of the literature through Structural Topic Modeling; Chapters 2 and 3 examine religious (or secular) bias in research production and research evaluation. Results indicate that scholars’ religiosity significantly influences research production: religious researchers are less likely to report evidence for the secularization thesis, while secular researchers report more. When it comes to research evaluation, religiosity has limited impact on assessments of secularization studies, but more religious scholars rate studies portraying religion as socially beneficial more favourably. These findings suggest that controversies over religious change might not be resolved solely by accumulating more empirical evidence, as researchers’ own ideological orientations shape the research process as well. Beyond religious studies, the thesis contributes to broader debates on objectivity and bias, demonstrating that the influence of values in science can be empirically examined rather than merely debated philosophically. These insights are particularly relevant for ongoing efforts to foster greater transparency, openness, and replicability in the social sciences.