Effects of a brief sexual-orientation-focused instructional intervention incorporating critical-thinking elements on immediate self-reported prejudice against lesbians and gay men: Baseline-adjusted evidence from a randomized experiment

Prejudice against lesbians and gay men (PALG) is associated with poorer well-being among sexual-minority people and with broader intergroup tension. Efforts to reduce it may benefit from pairing accurate information about sexual orientation with guided evaluation of the evidence and reasoning behind prejudicial claims. This study tested whether participants assigned to a brief sexual-orientation-focused instructional intervention incorporating critical-thinking elements would display lower immediate self-reported PALG than participants assigned to a neutral-video control condition. In a two-condition, between-subjects randomized experiment, 112 Chinese college students were randomly assigned to the instructional intervention or to a neutral-video control condition matched in duration. PALG was assessed before and immediately after the assigned video; prior intergroup contact and critical-thinking disposition were measured at baseline. The primary analysis was a baseline-adjusted analysis of covariance (ANCOVA) with post-intervention PALG as the dependent variable, condition as the predictor, and centered baseline PALG and centered prior intergroup contact as covariates. A separate exploratory model added the condition × baseline PALG interaction. Relative to the neutral-video control condition, the instructional intervention condition showed lower post-intervention PALG, adjusted mean difference = − 20.36, 95% CI [− 25.32, − 15.40], p < 0.001. An exploratory interaction indicated that the adjusted condition difference varied across baseline PALG, F (1, 107) = 250.49, p < 0.001, partial η² = 0.70; this variation may partly reflect floor compression. Repeated-measures and expanded-covariate sensitivity analyses were consistent with the condition difference, and analyses by sexual orientation indicated that the difference was concentrated among heterosexual participants. Participants assigned to a brief sexual-orientation-focused instructional intervention that combined sexual-orientation information with two selected critical-thinking elements reported lower immediate self-reported PALG than participants assigned to a neutral-video control condition. The condition contrast did not isolate either selected element, and the outcome was measured immediately by self-report. The findings should therefore be treated as preliminary; active-control comparisons, delayed follow-up, more diverse samples, and less demand-sensitive outcomes are needed before stronger conclusions are drawn.