Creating tables in scientific papers: row and column titles, units, error values and sample sizes

This is the second post in our series about creating and editing scientific tables. In the first post, we saw how basic table formatting and effective table titles could be used to improve an example of a poorly constructed table.

This post will deal with table row and column titles, units, error values and sample sizes. Let’s continue with the example table that we began to improve in the first post.

Fig. 1: Improved table after placing values in individual cells, formatting and double spacing, and adding an informative title.

Rule 4. Use short, descriptive row and column titles

The title of Table 1 (above) indicates the data in the table is about wheat plants exposed to salinity. Unfortunately, the row titles do not provide any useful information, except to show there were two groups in the experiment (control and test).

If this table was in a scientific paper, you could read the materials and methods section to find out how the control group and test groups were treated. However, every table should be understandable on its own, without having to look at other parts of the paper.

Therefore, the row titles in Table 1 should be the concentration of salt used in each group, perhaps Control (0 mM NaCl) and 50 mM NaCl (instead of control and test).

The column titles (light, 5 days and 10 days) in Table 1 are quite obvious: the researcher probably exposed the wheat plants to different periods of light each day, and then measured plant height after 5 and 10 days.

However, it is important to remember that simple titles such as “light” may be easily misunderstood by someone who is not familiar with your research.

Table 1 could be improved if the row titles provided a little more information, perhaps “Light exposure per day (hours)” or “Light/day (h)” instead of “light”. Similarly, “5 days exposure” and 10 days exposure” would be better than “5 days” and “10 days”.

If you need to use long or complicated titles that don’t easily fit in the column or row titles (for example, non-small cell lung carcinoma) then it is fine to use abbreviations (NSCLC), as long as you remember to define them in the table footnote.

Again, this makes the table easier to read and prevents your reader from having to look through the paper for the definitions for each abbreviation.

Rule 5. Always include the units, error values and number of samples

Although we have improved the content of Table 1 by changing the row and column titles, some very important information is still missing.

You could probably guess that the height of wheat plants is measured in centimeters, and the light exposure per day was measured in hours.

However, you may not be able to guess the correct units in every table (and your reader should never have to guess!!), so the units should be included in every table.

Additionally, it is not known what the numbers placed after the “±” represent in Table 1, as they could be the standard deviation or standard error of the mean. Therefore, every table should include the units (for example cm and hours) and define the error values (for example mean ± S.E.M.).

It is also important to show how many samples (or patients, cultivars, replicates) were in each group, especially if the sample sizes varied. You can choose where to include the units, error values and sample sizes, depending on the layout and information in your table.

For example, the units can be placed after every value, placed in a new row at the top of the table along with the type measurement as shown in Fig. 2 below or placed in a footnote, for example: “Values are mean centimeters ± SEM; (n = 5 per group).”

Fig. 2: Examples of different ways to include the units, error values and sample size information in a scientific table.

The table is improved by including more information in the row and column titles (rule 4), and defining units, error values and sample sizes (rule 5).

However, there is still some information missing and a few minor mistakes. Can you see any?

In the final post of this series, I will discuss the final pieces of information that should be included in every scientific table.

Welcome!

At Science Editing Experts, we help scientists like you to submit well-written journal papers with confidence and complete your thesis without headaches, so you can focus on your research and career.

Andrea Devlin PhD

Chief editor and owner of Science Editing Experts

The essential list of "Red Flags" in scientific writing:

348 words and phrases that scream "Written by ChatGPT or AI!"