One thing we try to avoid when writing journal articles is to surprise readers (and before that, reviewers) about the limitations of the analysis. We can avoid such negative surprises by being open and forthcoming about limitations, explain them before the readers find out themselves. In a paper currently under review, we tried to be very transparent about the fact that we simply have cross-sectional survey data. While we could probably spin something about exogenous variation, we wanted to play it straight (this being science, not a game).
So we included a sentence saying, hey, we have no empirical basis to interpret the results causally. The only thing remotely causal here is what is implicit in the theoretical argument. It turns out for two of the reviewers, this was a bit too forthcoming, and they understood that we wanted to draw causal conclusions based on a theoretical argument and a bunch of cross-sectional data. No, we just wanted to highlight that we humans have this tendency to interpret associations causally, and we don't have much more to offer.
At first, I was a bit confused, because we simply spelled out what all these studies using cross-sectional survey data without an explicit causal identification strategy do -- possibly the staple of quantitative social science. Why would we get two reviewers commenting so negatively about this? We debated whether there's such as thing as being too transparent and forthcoming, but concluded that the exact wording was probably the reason.
Published 18 December 2025