In this web content, the author presents a list of common mistakes in data visualization and explains why these practices are ineffective. The author emphasizes the importance of using appropriate visualization techniques to accurately convey data. Some key points include avoiding bar plots for mean separation, not using violin plots for small sample sizes, being cautious with color scales, avoiding bar plot meadows for multi-factorial experiments, and considering the ordering of rows and columns in heatmaps. The author also highlights the need to check for outliers, consider data range at each factor level, try different layouts for network graphs, not confusing position-based visualizations with length-based visualizations, avoiding pie charts, and using color scales that are colorblind-friendly and grey scale-safe. Additionally, the author mentions the importance of reordering stacked bar plots for effective visualization. The content is written in an opinionated and informative tone, providing examples and references to support the author’s points.
https://github.com/cxli233/FriendsDontLetFriends