Separating Results and Discussion
Just as the introduction and methods sections can sometimes be
blurred when writing the thesis, especially when aspects of the
field site are of prime importance to the study, it can also be
difficult to work out what should go in your results and discussion
sections. This is especially so when you have a complex dataset, the
experiment(s) encountered problems, or the results of one part of
the study inform the next part of the study. The basic rules to
Have a look at the following phrases and work out whether they
belong in a results section or a discussion section.
- The results section should only contain results! Obvious?
Perhaps! This means NO interpretation of results; don’t be
tempted to run straight from a statement of the significance of
a test into something like ‘suggesting that individuals in
population X are more vulnerable to Y predators’. You don’t want
to give away the ending!
- In a neat reversal, the discussion section should not contain
results! What does this mean? NO numerical or graphical results
(e.g. means with standard errors, P values), and NO new outcomes
that haven’t been introduced in previous sections.
Of course, this is easier to do in a hypothetical exercise than in
the reality of thesis-writing, when you might have difficulty
teasing out the outcomes and the implications of research that
you’ve been so intimately involved with. Communication and revision
with your advisor and mentor on what’s appropriate to include in
each section will be important.
- ‘Important differences were evident in the patterns of
mortality among the studied species. The Acropora had high rates
of whole-colony mortality and little partial mortality. In
contrast, rates of whole-colony mortality were low in Platygyra
daedalea and Porites lobata.’
- ‘The power to detect an 80% change in density between
treatments was high for Diplosoma literianum, didemnids, Pyura
stolonifera, Tricellaria porter and serpulid polychaetes
(>0.8; Table 1).’
- ‘Damaged colonies were consistently smaller than the controls
across the entire experiment, and at the final census, this
effect was statistically significant (Table 1; Fig. 3a).’
- ‘It appears that colonies that originate from larger larvae
have a selective advantage when mortality is low and occurs
early in post-metamorphic life’.
If you are contemplating publication of your study, bear in mind
that some journals, particularly those with extreme space
limitations – generally high-profile journals such as Nature,
Science, Ecology Letters and others – will require authors to submit
a combined results and discussion section. After working so hard to
keep your results and interpretation in defined sections, it can be
extremely difficult to integrate them into one section! Doing so
usually requires some creativity and flair, which are really
important in scientific communication, even if the public may not
think of us as creative!