In this study, we used a paper-pencil questionnaire to investigate whether teachers, administrators, and parents differed in their preferences and accuracy when interpreting visual data displays for decision making. For the data analysis, we used nonparametric tests due to violations of distributional assumptions for using parametric tests. We found no significant differences between the three groups on graph preference, but within two groups, we found statistically significant differences in preference. We also found statistically significant differences in accuracy between parents and administrators for two graphs--grouped columns and stacked columns. Implications for researchers and education stakeholders, and recommendations for further research are provided.