Sample-size salience and statistical inference

Abstract

In this study, students were asked to judge which of two samples was most likely to produce data exceeding the population mean. In two experiments, we examined the importance of the discriminability of the sample sizes for students’ judgments. The findings conflict with claims that students make their judgments based on a representativeness heuristic and support the view that errors reflect inattention to important but nonsalient dimensions of the task.