Wikipedia:Don't draw misleading graphs
This is an essay. It contains the advice or opinions of one or more Wikipedia contributors. This page is not an encyclopedia article, nor is it one of Wikipedia's policies or guidelines, as it has not been thoroughly vetted by the community. Some essays represent widespread norms; others only represent minority viewpoints. |
This page in a nutshell: Graphs should accurately represent the context of the data. |
Graphs that show a trend of data should illustrate the trend accurately in its context, rather than illustrating the trend in an exaggerated or sensationalized way. In short, don't draw misleading graphs.
Incorrect origin of the Y-axis
editThe most commonly seen "sensationalization" of graphs in the popular media is probably when the graph is drawn with the vertical axis starting not at 0, but somewhere just below the low point in the data being graphed. Both upward and downward trends are exaggerated, for a more exciting look.
For example, the employment data graph to the right, taken from the New Deal article in June 2008, purports to illustrate the level of employment in the United States of America over the years. The amount of employment visually appears to triple from 1933 to 1941. However, the vertical axis begins at 19 million rather than at 0; this disguises the fact that the rise was actually about 56%.
For this reason, the graph to the right is misleading. It is certainly more exciting than the "flatter" equivalent, but to be accurate, the graph should have been created with the vertical axis starting at 0.
However, one should also avoid insisting on a misleading "0". For example, when plotting the temperature history of Boston, it makes no sense to start the plot at 0 K, since 0 K is far removed from physically obtainable values and will only obscure the actual range of variation. In general, if one needs to use an offset 0, it is advantageous to use labels that are large enough that the offset is legible in the thumbnail.
Abuse of the X-axis
editManipulation of the graph's X-axis can also mislead; see the graph to the right. Both graphs are technically accurate depictions of the data they depict, and do use 0 as the base value of the Y-axis; but the rightmost graph only shows the "trough"; so it would be misleading to claim it depicts typical data over that time period.
The graph might not be misleading if it were specifically labeled in the caption as showing data only from 1/10/2008 to 1/13/2008 — but it's a judgment call whether the wider view would be better anyway.
Exceptions
editSometimes for the purpose of clarity and space efficiency, we need to vary from the above standards; however, an editor presenting the graph should be prepared to demonstrate a strong and valid reason for the departure from standards. The underlying reason should be to increase understanding rather than to distort the data.