Statistically Lay Decision Makers Ignore Error Bars in Two-Point Comparisons

Abstract

An experiment tested statistically lay decision makers’ use of error bars in a graph reading task. Participants viewed two-point dot plots, with each point representing a sample mean. Across conditions, means were accompanied by error bars of different sizes. Graphs were described as plots of consumer product ratings, and participants were asked to judge whether the products represented in each graph were rated differently or about the same. Signal detection analysis showed no influence of error bars on participants’ sensitivity or bias, and on average performance was poorer and far more liberal than predicted by a simple decision rule that classified the two data points as different if their error bars did not overlap. Results suggest that non-expert graph readers make little use of error bars in drawing