Visualizing scientific evidence
Choose a plot to answer a question: histogram for distribution, scatter for relationship, box or violin for group spread.
Loading video…
Choose a plot to answer a question: histogram for distribution, scatter for relationship, box or violin for group spread. Show individual points when feasible, use units on axes and explain error bars. A truncated axis can exaggerate changes, while hiding raw observations behind a bar can conceal outliers. A correlation plot cannot by itself establish causation.
Worked example
Plot expression versus dose as a scatter plot with dose units and individual replicates. A fitted trend may summarize association, but check whether batch or time could explain both variables.