This section if undoubtedly the most difficult to write. Take considerable care, and start by creating an outline of the points that you want to make, how the discussion relates to your Introduction, hypotheses and scope of the study, and the order in which you would like to present the ideas to the reader. Emphasis should be on interpretation and discussion of the results in light of your Introduction and other studies (references). Therefore, the discussion is a story linking all of the HOW, WHY, WHAT, WHERE and WHOs that you presented in the other sections.
Provide your interpretations of the data and relate them to the questions you posed in the Introduction, but limit your speculation to what can be supported by your results or the results of other studies. In other words, be reasonable! Organize the information in a logical order. Do not jump around from one topic to another. Finish one idea or concept and then go onto the next. Connect each idea or subject with the next to creating a flow to your writing.
- Do not indicate that your results provide ‘proof’.
- Do not introduce new material that was not provided in your results or that was not introduced in the Introduction.
- Do not repeat the exact material of the Introduction or repeat the specific findings of the results (however, trends are OK to repeat to remind the reader, but DO NOT cite the tables and figures as you did in the results section).
- If you have any data to explain away (such as outliers or values that are outside of a reasonable range), do it here or make a new hypothesis as to why the results came out in a way you did not expect.
- Do not refer to other studies as ‘The literature’ or use statements such as “The literature values …” or “Our values agree with the literature.”. The ‘literature’ is not a research group or author. Simply change the format of the sentence to refer to the author of the information. For example, “Our values agree with Gaston (2001)” or “Gaston (2001) found values of …”. See the references and citations section for more information.
The discussion should refer to the Introduction and aims of study making sure that all aims are discussed. Start by drawing some conclusions and supporting them with your data. Do the results agree with your hypotheses or answer your questions? If not, suggest possible reasons for the outcome. Are your results ambiguous because of limitations of the methods? Did anything unexpected happen and why might this have occurred, or were patterns observed as predicted?
You must also put your results into context with work done by others by making reference to other sources of information (i.e., do your results/interpretations agree or disagree with those in past studies? how do your results/interpretations compare with those expected?). It is quite acceptable to disagree with conventional wisdom, but you must state your reasons. When using information from other sources for purposes of comparison, remember to paraphrase and properly cite the information.
Suggest ways in which the study could be improved or further experiments that could be performed to clear up discrepancies or ambiguities in your results. How might your work best be continued or extended, or what other questions does this study suggest? What is the significance of your results in the general area you studied? What are the main principles demonstrated by your results? Finally, end with a summary of the significance of the work and the positive conclusions drawn from your study.
In summary, the discussion should always:
- Present principles, relationships, and generalizations
- Point out any exceptions or lack of correlation
- Examine possible sources of variability in your data
- Show how results and interpretation agree or conflict with published work
- State conclusions as clearly as possible