The Psychometric Properties of the Christian Nationalism Scale

A growing body of research connects Christian nationalism—a preference for a religiously conservative political regime—to social and political beliefs. This paper raises questions about the validity of a popular scale used to measure those attitudes. I begin by exploring the factor structure of the six-item Christian nationalism index. I then show how semi-supervised machine learning can be used to illustrate classification problems within that scale. Finally, I demonstrate that this index performs poorly at the interval level, a combination of measurement error and the sorting out of religious and political preferences. These attitudes have become so bound up in conventional politics that they often exhibit a threshold rather than a linear relationship to political preferences. I conclude with an appeal for care in matching theory to empirics: Christian nationalism is a prominent political theology, but research must grapple with the limitations of prevailing measurement tools when operationalizing it.