Adding up to good Bayesian reasoning: Problem format manipulations and individual skill differences

Abstract

Bayesian reasoning is ubiquitous in everyday life. However, how people perform in formal Bayesian reasoning tasks depends on both how the task is presented and individual differences in general ability. Different theoretical views predict that these individual difference factors should either interact with, or be independent of, presentation formats. This research first established changes in reasoning performance across different presentation formats (information presented either in natural frequencies or in percentages and either with pictorial aids or with no pictures). Concurrently, participants were assessed for their general level of numerical literacy and level of spatial ability. Multiple analyses indicate that the contributions of numerical literacy and spatial ability to Bayesian reasoning success were largely independent of the presentation of the tasks (numerical format and picture presence). These findings are shown in a subsequent experiment to hold, broadly, across 4 different assessments of numerical literacy and 4 different assessments of spatial ability. This result is most consistent with an ecological rationality view of statistical reasoning. (PsycINFO Database Record