Adjustment on the optimal layout: Visualizing standardized effect sizes using the framework of graph understanding and data visualization

Graphs are commonly used in scientific articles and public media. Different types of information have different graph layouts, some being favoured. However, there is no known optimal graph layout for appraising standardized effect sizes. This is the focus of this research proposal. We propose a framework of graph understanding and data visualization to connect numerical information and graph layout based on aesthetics and intention. We present different misleading visualizations and explain them. Based on the framework, we proposed ratios and formulas to set the y-axis range to improve appraisal of the standardized effect size. Study 1 aim to examine which combinations of ratios are critical for the judgement of standardized effect sizes. It will create a mapping of perceived levels of standardized effect sizes based on relevant information and quantify a free parameter to optimize y-axis range. Study 2 will compare optimal graphs with other graph layouts to confirm that the optimal graph works and allows observers to correctly assess the actual standardized effect sizes. Study 3 will generalize the optimal graph layout for other sample sizes. This research will result in a formula for creating optimal graphs for visualizing standardized effect sizes so that the scientists can better understand graphs and perceive correctly and efficiently standardized effect sizes, and also so that graph creators can generate better graphs. Study 4 generalized to laypeople, and Study 5 generalized to different sample size for laypeople. Study 6 will refine and verify the optimal layout based on limitations identified in previous Studies.