However, they have somewhat jumped the shark with a new study in which they claim—or, rather, they don't quite claim; they maintain a fig-leaf of plausible deniability, but it's clear from the glee in this Twitter thread what they want readers of the paper to think—that male evangelical Christians are obsessed with having a bigger penis.
However, their study (full-text is paywalled, and so new that it's not on Sci-Hub yet, sorry) does not have big dick energy. Rather, it is a load of bollocks.
They used massively unreliable and biased primary data (Google Trends), ignored any sort of exploration of why their main dependent variable (searches for penis enlargement) has declined substantially over 15 years (clue: When were we all getting penis-enlargement spam?), threw a bunch of arbitrarily-specified and poorly-measure covariates into their regressions to get the suppression effects going, reported the results of only 8 search terms when they very likely tested hundreds, and attempted to draw individual-level conclusions from aggregate data (the "ecological fallacy") while making a lame handwaving claim that they were not actually doing so:
I expect this study to be used by progressive writers to point and laugh at evangelicals, right up to the point that someone demonstrates that it's total crap, at which point the "poor put-upon Christians" will be able to say, rightly, that they are the target of incompetent and politically-motivated scientists. Sigh.The “moral communities” approach provides a helpful framework for understanding the trends we observe because it avoids the ecological inference fallacy, in which inferences about individuals are inappropriately drawn from observations of aggregate groups. The moral communities view posits that religion itself is a group-level phenomenon, and thus, the theoretical link between group-level religious characteristics and individual behavior does not require researchers to measure the thought processes or personalities of individual actors. Our claim is not about which individual-level behaviors produce a particular group-level outcome, but rather, that the evangelical context can explain some of the variation in aggregated individual-level behaviors (Google searches; see Stephens-Davidowitz 2014).