 Honorable Mention

Average Estimates in Line Graphs are Biased Towards Areas of Higher Variability



Dominik Moritz, Lace M. Padilla, Francis Nguyen, Steven L Franconeri

 DOI: 10.1109/TVCG.2023.3326589

 Room: 109

2023-10-24T23:57:00ZGMT-0600Change your timezone on the schedule page
2023-10-24T23:57:00Z

Exemplar figure, described by caption below — Demonstration of the bias toward variability. The red line shows the mean estimated averages across all participants in one of two experiments. The line chart shows a bias of the estimated average toward higher variability in the higher y-values. The bias is smallest when the data is shown as points equally spaced along the x-axis and can be simulated using points sampled at equal intervals along the arc of the line.

Fast forward

Full Video

Keywords

bias, lines graph, ensemble perception, average

Abstract

We investigate variability overweighting, a previously undocumented bias in line graphs, where estimates of average value are biased toward areas of higher variability in that line. We found this effect across two preregistered experiments with 140 and 420 participants. These experiments also show that the bias is reduced when using a dot encoding of the same series. We can model the bias with the average of the data series and the average of the points drawn along the line. This bias might arise because higher variability leads to stronger weighting in the average calculation, either due to the longer line segments (even though those segments contain the same number of data values) or line segments with higher variability being otherwise more visually salient. Understanding and predicting this bias is important for visualization design guidelines, recommendation systems, and tool builders, as the bias can adversely affect estimates of averages and trends.