3P.7—Plotting and interpreting graphs

Plot a graph only after deciding which variable is changed, which is measured and what relationship the design can support. Put the independent variable on the horizontal axis and the dependent variable on the vertical axis.

Label axes and units, choose a scale that uses the plot area, plot points accurately and add a line or curve only when the pattern justifies it. Use a bar chart for categories and a scatter or line graph for continuous measurements.

A rate–temperature experiment should show a rise to an optimum and then a fall if enzyme structure is damaged; the graph should not imply a precise optimum beyond the measured points.

Correlation on a graph does not prove causation. Do not join disconnected categories or hide uncertainty with a decorative trendline.