RT Journal Article SR Electronic T1 Graphics and statistics for cardiology: comparing categorical and continuous variables JF Heart JO Heart FD BMJ Publishing Group Ltd and British Cardiovascular Society SP 349 OP 355 DO 10.1136/heartjnl-2015-308104 VO 102 IS 5 A1 Kenneth Rice A1 Thomas Lumley YR 2016 UL http://heart.bmj.com/content/102/5/349.abstract AB Graphs are a standard tool for succinctly describing data, and play a crucial role supporting statistical analyses of that data. However, all too often, graphical display of data in submitted manuscripts is either inappropriate for the task at hand or poorly executed, requiring revision prior to publication. To assist authors, in this paper, we present several forms of graph, for data typically seen in Heart, including dot charts, violin plots, histograms and boxplots for quantitative data, and mosaic plots and bar charts for categorical data. Justification for using these specific plots is drawn from the literature on visual perception; we also provide software instruction and examples, using various popular packages.