Do the Maasai perceive weak walkers to be stronger and more attractive than strong walkers? A re-analysis of Fink et al. (2019)

摘要

Fink et al. [1] report that Maasai men and women rated strong walkers as both weaker and less attractive than weak walkers: a surprising reversal of the effects found in previous research [2,3]. These data from a non-WEIRD population are highly valuable for assessing the universality gait perception [4]. However, the conducted analyses relied on participants' ratings of attractiveness and strength that were averaged across groups of stimuli (i.e. by-participant aggregation), which is problematic for two reasons (see [5–7] for a full discussion of these issues). First, aggregating compounds error variance and severely inflates false-positive rates. Second, inferences from by-participant aggregation can only be applied to those exact stimuli used in the study because averaging treats the stimuli as a fixed effect rather than a random sample from the population of potential stimuli. These problems are pervasive in the literature and not unique to Fink et al. [1], but given the current state of psychological research, it is critical to begin to examine and address these issues where they occur.