Skip to Main Content

Ten tips for scientific writing

This is general guidance: module handbooks and assessment criteria may set very specific requirements for using diagrams and tables for certain assignments, so check what they require from you.

Many concepts and observations in science can be most clearly communicated using diagrams and tables. 

  • If you want to discuss changes in otter population in a particular location, then a graph or chart plotting the numbers of otters observed at that location at different times will be more effective than describing the same information in text.
  • If you want to discuss the population distribution of otters, a map shaded with different intensity to reflect this will be more impactful than a description on its own.

You must label and number diagrams, figures and tables so it is clear what information is being communicated. When creating or choosing a chart, graph or illustration to support your writing, you should aim to make it self-contained so that the meaning of the information it contains is clear to your reader without reference to the surrounding text.  Headings, labels and captions should be short but not so short that they leave the reader puzzled about what is being shown. 

This does not mean that there is no need for you to explain charts and other illustrations in the accompanying text. In the body of your essay explain why the information in the chart, table or diagram is important to the discussion in your essay. When using illustrations, it is good practice (unless told otherwise) to mention in the text the image before the image appears in the text e.g. text first introducing the image and then the image! If you are using statistical information discuss relevant details such as distribution and statistical significance. 

If you have reused someone else's image, chart or figure, make sure that you correctly acknowledge and reference where it has come from.  Likewise, if you use someone else's data to create your own charts and tables, make sure data sources are properly cited and referenced.

For graphs and charts:

  • Bar charts should start at zero as otherwise variations in quantity may look more significant than they are. 
  • Graphs: Choose suitable axes so that variations in data can be correctly interpreted.  Graphs can be used to show changes, in which case it may be reasonable to place your access axes at a suitable reference point.  This is especially the case when points are measured on a scale.  For example it is meaningful to show variation in human body temperature against a normal healthy temperature rather than against zero degrees centigrade (or Kelvin!). 

For tables:

  • Group and order data so that you can apply clear headings. 
  • If elements of data share common features you can layer headings so that the grouping is very clear.

Example tables

The two tables below compare actual and estimated calorie intake and use.  The arrangement of data in both is identical, but the layered headings in the second example make it easier to read. 

 

  Estimated calories eaten Estimated calories burned Estimated calorie gain/loss Actual calories eaten Actual calories burned Actual calorie gain/loss
Participant 1 1900 2100 -200 2000 2300 -300
Participant 2 2500 2350 +150 2600 2400 2400

Table 3.1: Table with unstructured headings

 

  Estimated calories Actual calories
  Eaten Burned Gain/loss Eaten Burned Gain/loss
Participant 1 1900 2100 -200 2000 2300 -300
Participant 2 2500 2350 +150 2600 2400 +200

Table 3.2: Table with layered headings (Both tables based on Boyle & Ramsey 2017, p. 90)

 

The values in the tables are centred in each cell.  This works well because the numbers are similar in each column.  If the numbers had been of different order or magnitude it would be better to justify to the right to align the values more meaningfully. 

Further reading

Graphs and charts:

Bergstrom, C., & West, J. (2017). Visualization: Misleading axes on graphs. Retrieved from http://callingbullshit.org/tools/tools_misleading_axes.html

Using data and all aspects of writing for science: