Testing the feasibility of small multiples of sparklines to display semimonthly income statement data

This paper tests the ability of 129 accounting students to look up data, perceive and compare data patterns, and detect anomalies when the amount of information in traditional single page three-year comparative income statements is expanded 25-fold by adding semimonthly data to the page in graphical form. Participants were able to use the graphs to perform most tasks correctly more often than not. Tabular presentation of the same 1100 to 2000 data points required six pages. Participants using the one-page income statements with embedded graphs were better able to identify and compare patterns than those using multipage tables, but had less success looking up precise data values. The results on pattern recognition and comparison are consistent with the proximity compatibility principle.

► Small multiples of sparklines are used to display semimonthly income statement data.
► The examples of displays contain up to 2000 data points per page.
► Tests using 129 students show good pattern recognition and anomaly detection.