The influence of graphical format on judgmental forecasting accuracy: Lines versus points

Abstract

People made forecasts from real data series. The points in the series were un‐trended and independent. Hence, forecasts should have been on the mean value. However, consistent with previous research on forecasting biases, forecasts were too close to the last data point. It appears that forecasters see positive sequential dependence where none exists. In three experiments, we examined this bias in different types of forecasting task: point forecasting, probability density forecasting, and interval forecasting. In all cases, we found that it was greater when the data series were displayed using continuous line graphs than when it was displayed using discrete point graphs. Consistent with arguments made by Zacks and Tversky (Memory and Cognition, 27:1073, 1999), we suggest that people are more likely to group data together and to see patterns in them when those data are presented in a continuous than in a discrete format. These findings have implications for forecasting practice.

Forecasting: theory and practice

“…Harvey and Bolger (1996) found that trends for instance were more easily discernible when the data was displayed graphically rather than tabular. Additionally, simple variations in presentation such as line graphs versus point graphs can alter accuracy (Theocharis et al, 2018).…”

Progress and opportunities in advancing near-term forecasting of freshwater quality

“…During model development, expert elicitation, a formal process of extracting expert knowledge while mitigating bias (Hemming et al, 2018), can be employed to inform model structure (e.g., Bertone et al, 2016). End users should also be consulted regarding forecast dissemination methods to ensure correct interpretation of forecast output and maximize forecast utility (Berthet et al, 2016;Gerst et al, 2020;Theocharis & Smith, 2019). For example, interviews and focus groups with end users of NOAA's Climate Prediction Center climate outlook visualizations guided updates of NOAA's air temperature and precipitation color maps for improved forecast interpretability (Gerst et al, 2020).…”

Analysis of The Influence of Graphic Financial Information Presentation; The Role of Human Information Modality, Experimental Approach

“…Therefore, graphs based on the characteristics inherent to them, the graph will be appropriate when the data presented has a trend and is used to forecast future conditions, for this type of data the graph can better show a general picture and the relationship between the data presented. Conversely, the use of tables is beneficial if used on data that requires an emphasis on data values at a particular point in time and will be more useful to facilitate the explanation of data that have specific benefits [12], [17] and [18].…”

“…Researchers in psychology find that humans tend to minimize cognitive efforts to achieve the accuracy of the decisions they make [17]. Several other studies have also found that managers take a lot of advantage from decision aids to reduce the effort needed to do a job ( [22] and [11].…”