Presentation, communication and dissemination of vital statistics
Guidelines for good charts
Through the use of charts, data can often be understood better, because it becomes easier to see hidden patterns. Charts are particularly good when comparing magnitudes, changes over time, distributions, correlations and relative shares of a whole. The decision about the type of chart to decide to use should be based on the composition of the audience, what the data shows and what relationships are important to highlight. Depending on the above considerations, some charts will be more appropriate than others.
For presenters who are not very experienced in deciding which type of chart to use, it is recommended to experiment by drawing several types of charts to see which one is the most suitable. A trial-and-error approach can be very helpful in improving graphs. It is unlikely that you will find the right graph will be found the first time.
The following are some guidelines for charts inspired from the Secretariat of the Pacific Communities Data analysis and report writing course: 1
- Graphs should have a clear, self‐explanatory title
- The units of measurement should be stated
- Graphs should be simple and not too cluttered
- All axes should be carefully labelled
- Include the source of the data
- The scale on each axis should not distort or hide any information.
- Graphs should clearly show trends or differences between the data
- Use two‐dimensional designs
- Avoid data point markers on line graphs
- Graphs should be accurate in a visual sense (e.g. if one value on the chart is 15 and another 30, then the second value should appear to be twice the size of the first)
- Avoid abbreviations and acronyms
- Avoid legends except on maps
- Colour use should be consistent. For example, if males are shown in blue and females in red, this convention should be followed across all charts.
When selecting the scale to display, it is good practice to start at zero. When that is not possible, some symbol should be used to indicate a break in the line, this is usually done with a zigzag line calling attention to the fact that not all the scale is shown. This should also be used if there is a break in the axis at another point.
The temptation to use the option to produce three-dimensional charts that are included in the graphical software should be resisted, because they actually distort the data. A two-dimensional graph is always much easier to read correctly.
Keeping in mind the advice about designing charts, features in charts (including colour) that do not show something meaningful about the data should be avoided as they make the chart less readable.
Below are some examples of good charts that follow these guidelines:
1 Ryan C et al (2011). Workbook: Data analysis and report writing course, Secretariat of the Pacific Community.