The importance of understanding the power of visualizations is increasing with their growing popularity in many fields. Deceptive visualizations are visualizations that distort the message in data with or without the intentions of the designer. This study aims to investigate the impact of truncated y-axis usage, a commonly used deception method, in visualizations that represent time series data sets. The study starts with an in-depth analysis of the existing literature in the visualization field. The theoretical part provides a summary of the existing literature for deceptive visualizations, with topics such as the goals and types of visualizations, visual perception process, the design principles for visualizations, and the types of deceptive visualizations for time series data sets. In this paper, an empirical analysis of truncated y-axis usage is presented. Also, not-connected axes usage is empirically tested as a solution to prevent the deceptiveness of truncated y-axis usage. An experiment is designed to measure the perceived message from visualizations. In total, six visualizations are created for the experiment. The experiment data is collected with two online questionnaires. In total, over 200 people participated in the study.The findings suggest that truncated y-axis usage is deceptive for the visualizations that represent time series data sets. Also, not-connected axis usage does not prevent the deceptiveness of truncated y-axis usage in these charts. Moreover, the results of the experiment shows that individual differences (gender, age, and education level) do not change the deceptiveness of truncated y-axis usage.
展开▼
