Does Evidence Presentation Format Affect Judgment? An Experimental Evaluation of Displays of Data for Judgments

Information relevant to a prediction was presented in one of eight formats: a table of numbers, a brief text, a longer biographical story, and five different types of bar graphs. Experimental participants made judgments of marathon finishing times based on information about the runners' ages, prior performance, training, and motivation. A regression analysis was used to assess the individual judges' relative weighting of the various types of information relevant to their predictions. The different formats for displaying information yielded different levels of accuracy and patterns of information utilization. In accordance with an explanation-based decision model, the text and story displays induced the heaviest reliance on information about runners' motivation and prior performance and produced the most accurate judgments of marathon finishing times.