Hostname: page-component-7b7567c8c-w7t8q Total loading time: 0 Render date: 2026-11-19T06:28:20.434Z Has data issue: false hasContentIssue false

Merging Graphics and Text to Better Convey Experimental Results: Designing an “Enhanced Bar Graph”

Published online by Cambridge University Press:  12 June 2017

William D. Berry
Affiliation:
Florida State University
Matthew Hauenstein
Affiliation:
Florida State University
Rights & Permissions [Opens in a new window]

Abstract

We propose a format for presenting experimental results that combines a graph’s strength in facilitating general-pattern recognition with a table’s strength in displaying numerical results. The format supplements a conventional bar graph with additional text labels and graphics but also can be based on a dot plot. The resulting enhanced bar graph conveys general patterns about treatment effects; displays point estimates and confidence intervals for all key quantities of interest relevant to testing hypotheses (e.g., first differences in the mean of the dependent variable); and clarifies the interpretation of these quantities as treatment effects. Presenting information in a single figure avoids the need to devote scarce journal space to both a graph and a table. Moreover, an enhanced bar graph prevents readers from having to move back and forth between a graph and a table of numerical results—thereby reducing their cognitive load and facilitating their understanding of the findings.

Information

Type
Articles
Copyright
Copyright © American Political Science Association 2017 
Figure 0

Table 1Results from a Fictitious 2x2 Factorial Experiment to Test a Hypothesis that the Two Factors Interact in Influencing a Dependent Variable

Figure 1

Figure 1A Conventional Bar Graph Showing Results from Table 1Note: Each bar shows the estimated mean of the dependent variable among subjects in an experimental condition. The vertical line overlaid on a bar shows the boundaries for a 95% confidence interval.

Figure 2

Figure 2An Enhanced Bar Graph Showing Results from Table 1Note: Each bar shows the estimated mean of the dependent variable among subjects in an experimental condition. Each Δ value next to a U-shaped arrow is a first difference in means reflecting a treatment effect; a solid arrow indicates an effect deemed substantively significant. The ΔΔ value next to the double arrow is a second difference (i.e., a difference between two differences in means: 7 ≈ 9 – 3) reflecting the strength of interaction. (ΔΔ could be computed equivalently as a difference between the other two first differences portrayed: 7 = 22 – 15.) Each estimated quantity of interest is reported along with the boundaries for a 95% confidence interval in parentheses. An asterisk (*) indicates statistical significance at the 0.05 level (two-tailed test).

Figure 3

Figure 3An Enhanced Dot Plot Showing the Same Results as Figure 2 Using Straight Arrows To Depict First and Second Differences.Note: Each dot shows the estimated mean of the dependent variable among subjects in an experimental condition. Each Δ value to the right of an arrow is a first difference in means reflecting a treatment effect; a solid arrow indicates an effect deemed substantively significant. The ΔΔ value to the right of the double arrow is a second difference (i.e., a difference between two differences in means: 7 = 22 – 15) reflecting the strength of interaction. (ΔΔ could be computed equivalently as a difference between the other two first differences portrayed: 7 ≈ 9 – 3.) Each estimated quantity of interest is reported along with the boundaries for a 95% confidence interval in parentheses. An asterisk (*) indicates statistical significance at the 0.05 level (two-tailed test).

Supplementary material: PDF

Berry and Hauenstein supplementary material

Online Appendix

Download Berry and Hauenstein supplementary material(PDF)
PDF 1 MB