Replication of “Instrumentally Inclusive: The Political Psychology of Homonationalism” (Turnbull-Dugarte and López Ortega 2024)

Turnbull-Dugarte and López Ortega (2024) argue that increasing exposure to sexually conservative ethnic out-groups causes instrumental support for LGBT+ rights among those predisposed to disfavor the ethnic out-group. The article presents results from two related experiments, one conducted in the UK and a follow up in Spain, where respondents were randomly assigned to read vignettes about anti-LGBT+ protests, and the identity of the protesters was varied. I outline a series of concerns with the article, primarily related to ad hoc empirical choices. The authors use weights (of undisclosed origin) that follow a peculiar bimodal distribution in the second study but use no weights in the first, and inconsistently use robust standard errors throughout. These choices create a pattern of statistically significant results consistent with their theory, a pattern that disappears when either choice is varied. Additional analyses show that, rather than supporting their theory, the second experiment contradicts it.