In the current era of more extreme political and religious views coming into the mainstream, this paper investigates whether party affiliation is associated with the concept of religious morality more than political ideology. To answer this research question, we construct an empirical scale of religious morality based on data from opinion polls, which remain among the most efficient and widespread survey methods to capture psycho-social data at large scales. Classification models, using ideology and religious morality as predictors, respectively, were compared on their performance in predicting party affiliation. Response data on just four questions from a national opinion poll were used to demonstrate the viability of constructing a parametric scale using item response modeling approaches. Results show that even with 'small data', a measurement-based and data-driven approach can produce classification models that are both statistically precise and accurate. Finally, we discuss the findings and the implications for both politics and psycho-social measurement.