This section follows on directly after the results section using a centred bold heading (like for the Method and Results sections). After presenting the results in the previous section, you are required to evaluate and interpret the implications with respect to your original hypotheses. You will examine, interpret, and qualify the results and draw inferences and conclusions from them. A discussion should include a summary of the results, a statement of whether the hypothesis/es were supported or not, findings linked to past research, an interpretation of the research findings, implications, limitations, future directions, and concluding comments. Check your assessment outline for specific requirements relating to the discussion. It is okay to introduce new references in the discussion when new literature will enhance your explanation of a result, provide an explanation of unexpected findings, or can assist with providing implications or future recommendations. The components of a discussion in addition to appropriate writing style and formatting are explained below.
Writing style and formatting
All paragraphs should be indented with no spaces between paragraphs. You do not need to write a paragraph for each component of the discussion but all should be included in your write up. The way you write the discussion will depend on your research question, the number or research questions you have, the report length, and the results of the study. It is important that you do not simply repeat points already made; each new sentence should contribute to your interpretation and to the reader's understanding of the problem.
Summarise the Results
To refresh to reader, begin the discussion by restating the aim of the study. Follow this with a summary of the results making sure to point out the most relevant findings. Do not repeat statistics when summarising findings; instead use words. For example, instead of "There was a statistically significant moderate positive correlation between hours studies and exam marks (r = .45, p .001)”, write “There was a moderate positive correlation between hours studies and exam marks showing that both variables increased together”.
State support (or non-support) for the hypothesis/es
Next state whether your hypothesis/es were supported or not support. Remember to address each hypothesis and to avoid using the terms 'proved' or 'disproved'. Instead, you should state that the hypothesis/es were supported or not supported. To avoid being repetitive, providing a statement of support for your hypothesis/es can be done when summarising your results. For example, "Providing support for the first hypothesis, results of the experiment revealed...".
Link findings to past research
After discussing whether the hypothesis/es were supported, you need to state whether the experiment findings support the prior research. The prior research are the studies you used to justify your hypothesis/es in your literature review. Do not refer to a reference as simply 'the prior literature' - cite the reference. For example, "The finding that a meditation program showed reduced heart rate variability in participants supports prior literature by Anderson (2015)". Also, state how your research extended upon these past findings to highlight how the current study is building upon prior literature.
Interpreting findings
After your have given a summary of the findings, stated support for the hypothesis/es, and linked your findings to past research you are required to interpret your results. You need to interpret the findings in terms of 'what do they mean?'; do not simply restate the findings.

Expected Findings
If you have an expected finding (in other words, your result/s support the hypothesis/es) you could interpret the findings by explaining conclusions provided by past researchers. For instance, "Anderson (2015) concluded that reduced variability in heart rate could be the result of meditation decreasing stress and regulating breathing patterns". You can also base your interpretation off theory that supported your hypotheses or unique explanations that are specific to your outcomes.

Unexpected Findings
For unexpected findings (result/s that do not support the hypothesis/es), evaluate methodological, theoretical, or alternative explanations for why the hypothesis/es was not supported. A way to interpret an unexpected finding is by comparing the method of your study to methods of research that justified your hypothesis/es. You are looking for differences between the methodologies that could explain why you found contrary results. For example, your research may have used a different sample, measure, intervention, or conducted the study in a different geographical location.

Alternatively, you can use new sources to provide an explanation for unexpected findings. This is one of those cases where it is okay to cite a new reference! For example, lets say you hypothesised that there would be no difference on scores of the personality trait extraversion between male and female first year psychology students. However, your results showed there was a significant difference where females scored higher on extraversion than males in your sample. You could explain the finding with research you found (that you would also cite!) that describes females as being more inclined to reach out to others during new social situations more so than males. Therefore your unexpected finding can be explained using a new source that provided an alternative explanation.
Once you have interpreted your findings you need to discuss the implications of the findings for theory, clinical, or practical outcomes within the area of research. Implications are how your findings contribute to the wider community. To do this, evaluate what is important about the findings and communicate how the findings can be translated into an applied nature. Ask yourself how do the results contribute to the theoretical foundation of your topic? Do the findings have practical implications? How can these findings be of benefit to people? For example, your findings may have showed that an cognitive training program for children with AD/HD reduced their behavioural deficits. You could suggest that a clinical implication of these findings is that cognitive training could be considered an additional treatment in conjunction with medication for treating the symptoms of AD/HD. Avoid being absolute in describing implications as they are educated suggestions, not 'fact'.
Now that you have considered the implications, limitations of the experiment need to be considered when evaluating the findings. Limitations are possible methodological or statistical issues that systematically reduce the reliability or validity of the results. Limitations can also include procedural problems that occurred during the experiment. Example of methodological limitations could include narrow sample selection or using an measure that has poor reliability. Examples of statistical limitations include poor statistical power or small effect sizes. Procedural limitations are things that Once you have identified a limitation, explain how it could have impacted upon your results. For example, you may identify that "the current study was conducted using a university sample" and explain that due to this, "the findings may not be generalisable to the wider community".
Future research
Once you have outlined the limitations provide recommendations for future research. Suggestions for future research can be based on overcoming the research limitations. You can also propose new research directions that could extend the current knowledge.
Concluding comments
Lastly, you need to conclude with an overall summary of what the study found. The conclusion should reiterate the results, leave the reader understanding what the study contributed to the literature, and highlight the consequences of your research findings. If you are writing a research report you may only write on sentence providing a concluding statement, however, if you are writing a longer document your conclusion may be longer.
Common mistakes in writing the Discussion:

  • Not adequately linking your discussion to the hypotheses that you have tested. You need to make it clear whether the hypotheses are supported or not, and if not, why. If they are supported there still may be other explanations for why this occurred.

  • Your discussion needs to link in with the points raised in your introduction and it needs to “hang together” correctly. Make sure to refer to these points in light of your findings.

  • Incorrect use of the word “significant” when discussing findings. Only use “significant” if you have tested a difference between groups using statistics. Never use the word “insignificant” to describe your results.

  • Do not over claim about findings. It is better to understate your case by reporting exactly what the data indicate, no more and no less. Remain factual and objective.

  • Repeating the statistics from the results section instead of summarising the findings using words.

  • Not discussing unexpected findings. It is important to interpret both expected and unexpected results.

  • Stating that small sample size was a methodological limitation of the research without properly understanding what size sample is appropriate and how to determine the power of the analyses to justify this. Avoid using small sample size a limitation of the research unless you understand how to determine when a small sample size reduces statistical power.

  • Inappropriately introducing new literature.