Bar graphs are probably the most common type of graph in the biological sciences, and are excellent for representing data in an easily comparable way. However, there are some aspects to consider when making a bar graph that will allow your reader to get the most out of your data. Here are some of the key points to think about:
1) Horizontal or Vertical Bar Graph?
When trying to make this choice, you should consider the length of your data labels. If your labels are exceedingly long, squishing them into a vertical bar chart may not be advisable. Moreover, rotated titles can be awkward and difficult to read. In these cases, a horizontal bar graph is preferable.
2) Data Arrangement
How you arrange your data is critical for how easily your readers will grasp what you are trying to convey. For bar graphs, use your common sense. Start with the control value to which all other values are compared. Start with the highest value. Start with the lowest value. Pretty much any logical way is better than randomly arranged or alphabetically arranged for no apparent reason.
3) Start the Y-axis at Zero
This point is applicable to any plot. Basically, if you start at some non-zero point your data can be easily misrepresented. We naturally assume that the Y-axis starts at '0', and changing this can make the differences in your data seem huge, when in actual fact they are quite small.
4) Consider Different Types of Bar Graphs
Depending on what your data set looks like you may want to consider:
Grouped bar plots - good for comparing several categorical variables
Stacked bar plots - good for comparing values where the sum of the amounts represented in each bar is an important number.
You can also follow us on Twitter for helpful writing tips and the occasional discount! @ExcisionEditing