Your goal in communicating science should always be to make the topic you are discussing as clear and accessible to your audience as possible. Although your writing is a critical element of this, you should not underestimate the importance of producing effective tables and figures.
Some people find it easier to understand and interpret results and/or the relative magnitude of any trends by looking at figures as opposed to reading explanations in text. Similarly, tables should be used to summarize the most important, specific results in a way that makes it easy for anyone to read and interpret these. Even if you provide similar information in words, it can help to include tables and figures as well. This is particularly true when you are communicating uncertainty (see our resource on this here), and, because virtually all data analyses are based on probability, there will almost always be an element of uncertainty in the results you present in lab reports or research-based journal articles.
On the other hand, some people struggle to interpret figures (and information contained in tables), which means there are some guidelines you should follow when producing these, to minimize the likelihood of confusing your readers. Remember that you should be producing a table or figure to help show a pattern and/or specific information in a way that is easier to interpret than if it was simply written. With this in mind, you should always try to:
1) Be clear and concise.
2) Be unique.
3) Give clear titles and legends that are concise and easy to read.
4) Be consistent.
Deciding Whether to Use Figures or Tables
Whether you use a figure or a table in any given situation should depend on what data you have to present and on what you want your reader(s) to focus on. This often comes down to whether you want your reader(s) to focus on a pattern, or a trend, or whether you want them to focus on very specific details (such as the absolute numbers generated by experiments).
To help further, Table 1 (below, note the clear title!) should allow you to match your data to the most appropriate option for displaying that data.
Table 1: When to use tables and figures based on the different types of data you may wish to present.
|Use A Table When You Want To||Use A Figure When You Want To|
|1: Compare data values among related items or groups||1: Summarize trends and patterns|
|2: Display specific numerical values (or many data values)||2: Show these trends and patterns when these are more important than the specific numerical values|
|3: Highlight a Yes/No effect (or a Present/Absent effect)||3: Show explanations of procedures (methods) or responses (e.g. images)|
Some Examples (Using Table 1 To Help Decide Whether To Use Figures or Tables)
You should display these results in a figure (a bar chart would work well) because the most important thing to display is the pattern (did people taking the drug sleep longer than those that took the placebo?) rather than the specific data values.
You should display these results in a table (to provide the specific numerical data (ages of all 27 participants) and whether they got to sleep faster after taking the drug (Yes/No).
Specific Tips for Producing Effective Figures
1) Simplicity is important. Try to show the main pattern or trend as simply as possible.
Figure 1: Bar Charts are Generally Easier to Interpret than Pie Charts.
The same data are displayed in the two charts but while it is easy to rank the three groups in terms of performance in the bar chart, it is impossible in the pie chart.
2) Label all axes and legends. In Figure 1, note how it is clear what the bars and pie slices represent (Groups A, B and C), and how the data are measured (%).
3) Limit white space. There is little point in taking up valuable space in any publication with blank space.
Specific Tips for Producing Effective Tables
1) Divide categories logically. Think about the order in which data should be presented so that it tells the tale you would like your audience to read.
Table 2: Presenting Data in Rank Order in Tables Can Help to Show the Important Patterns As Well As The Specific Values.
|Example A||Example B|
|Group||Performance (%)||Group||Performance (%)|
2) Format your table so that it is easy to read. Make sure there is sufficient space between rows and columns because it is hard to distinguish between data or see patterns in an overly cluttered table.
3) Be as concise as possible. You would often show data in a table and not a figure because it is important to show the specific data values rather than the pattern in these data. But this does not mean it is OK to include a huge amount of raw data that is not imperative for the reader to see. Try to minimize the amount of data you actually present (e.g. if you measured the height of 197 university students, you don’t need to present a table with 197 rows; instead show the mean height, the range, and the standard deviation for males and females in your sample).