摘要：The “erasing race” effect is the ability to reduce the salience of “race” as an alliance cue when recalling coalition membership, once presented with more accurate information about coalition structure. The effect is here meta-analysed with five US studies (and nine independent effect sizes), using a random effects model. The effect is found to be present (ρ=.137, K=9, 95% CI=.085 to .188). However, no moderation effects were present and no decline effect was noted either (i.e. the effect size did not decrease over time). Publication bias tests found little evidence for this. Synthetically correcting the effect size for bias stemming from the use of an older method for calculating error base-rates also reduced its magnitude, but the effect remained significant. Taken together, these findings indicate that the effect generalizes quite well across experimental contexts and would therefore appear to be quite robust. We reinterpret the theoretical basis for these effects in line with the Brunswikian evolutionary developmental theory and present a series of predictions to guide future research in this area.
关键词：alliance detection system; coalition; erasing race; social identity; meta-analysis
