There are four key things to include in your discussion:
- Summary and explanation of key findings
- Engagement with the literature
- Synthesis and application of results
- Limitations and recommendations for future work
These can be broken into tasks (see Box 11.1.1).
Box 11.1 – Tasks of the Discussion
- Restate the research problem
- Recap the major findings
- Explain the meaning of the findings
- Highlight the relevance of your study and how it helps fill the gap
- Compare the findings with existing studies
- Acknowledge the limitations of the study and implication on the findings
- Recommend further research
Summary and Explanation of the Key Findings
There are two things that will help you to summarize your kind findings: first, revisit your research question(s) and ensure that the materials being discussed help to answer them; second, revisit your findings/results and highlight only the results that pertain to your research question(s). It is important that you are not discussing trivial data such as the demographic composition of your sample (unless it has a key bearing on the results) or reviewing central tendency data (unless you have tested an hypothesis which has a bearing on your research question). While there might be interesting insights in those data, typically, they do not answer the research question in and of themselves. The ease at which you can identify your key findings will depend on the kind of analysis that you have done. For qualitative research, you need to revisit the key themes. If you cannot itemize the key themes from your findings, now is the time to revisit your analysis, look at the codes and patterns and itemize the most important 5 to 7 themes. For quantitative data, it is important to visit your hypotheses and the statistical tests used to examine them. As Greetham (2019) argues, you need to remind yourself and your readers what the questions and sub-questions are.
After you have identified the key findings and themes that relate to your research question, it is important to not just repeat what was previously highted in your results section. You need to explain the results in a manner that tells a cohesive story. This means putting the findings into context or to “engage in productive speculation” (Hinshaw in APA, 2006). You need to highlight why these findings are important and the wider implications for knowledge, and their applications or other productive uses. For example, looking at the article from Hou, Shellenberg & Berry (2018) which discusses immigrants’ sense of belonging to Canada and their source country, we see that after discussing the findings they extrapolated a framework for categorization determination of immigrants’ belonging and they explained why this is important:
Looking at the determinants of membership in each of the four profiles, we can separate those factors that existed pre-migration from those that arose post-migration. The reason for this separation is that there are differing implications of the findings, because more can be done to improve outcomes when dealing with post-migration factors than for those that existed prior to migration (p. 1627)
Of course the example above is from professional and experienced writers, and it takes considerable skills to be able to extrapolate from findings. However, we want you to notice how they direct attention to the difference between pre and post-migration to justify the distinction. Through this, they are able to introduce the significance of their distinction, that “more can be done to improve outcomes” after migration than before. In two sentences, they are able to justify the central analysis of their paper. To do the same, it is important to ask yourselves the following questions:
- What do these findings mean? How/why are these findings useful?
- Why do these findings matter? Why should anyone care? Do they have the potential to impact structure, policy or change? Is there the potential for application?
- Are there new ways in which the findings can be categorized?
In essence, we do not want you to simply regurgitate the results. Try to answer the question: so what? And remember, limit your discussion to a few salient issues.
Engagement with Literature
This is perhaps the most important task in your discussion. The literature can be integrated in your paper in two ways: First you can relate your findings to previous works. Second, you might find that your literature review (as good as it might be in relation to your research question) does not anticipate your findings. This usually happens because your findings are either new or unexpected. While you may not need to re-do the literature review, you will definitely need to dig deeper into it. This digging will not produce a comprehensive literature review (certainly not as the original), but it will be more specific to the issues that you want to explain further. In that case, you will be introducing new sources to explain and/situate your findings. Please note that it is generally okay to introduce new literature in the discussion if the findings are surprising (just do not introduce new findings in the conclusion, which we will discuss later). Regardless of the approach taken, you must attempt to compare your findings with existing literature.
Engagement with the literature is important because it helps demonstrate why your results are significant. It establishes continuity with the scholarship and shows how your findings fit within it. It also justifies the relevance of the research by highlighting your contribution to the advancement of knowledge in the area. Going back to the discussion from Hou, Shellenberg & Berry (2018, p. 1629), we can see how this continuity/contribution is established:
…these findings in Canada provide further support to the assertion by Berry (2017) that there may be some general principles of intercultural relations in all societies that could provide the basis for developing policies intended to improve relations and outcomes for both immigrants and those already settled in a country.
In the above case, we see how Hou et al., (2018) findings built on their previous work and applied it to multicultural policy in Canada. Their findings are linked with Berry’s (2017) assertion that basic intercultural policies exist which can be appealed to in order to improve migrant-domestic relations. They, thus, are able to once again reaffirm, through their evidence, that Berry’s assertion is an attractive one. To similarly engage with the literature, we suggest that you ask these three questions about the relationship between your findings and the literature:
- Are my findings similar or different to previous studies? Why?
- Do other findings support the claims I am making?
- How might these findings be applied?
Synthesis and Application of Results
If you have not yet done so, review your discussion so far to ensure that you have connected all the previous chapters (e.g., theory, literature review, methodology, specific aspects of the research problems etc.). This will help to ensure that your story is cohesive, and will allow you to put it to practical application. Drisko (2005, p. 592) asserts this as follows:
Authors should make each major contribution of the study clear and explicit. Beyond linking the current work to the prior literature, the discussion may point out newly apparent definitional or conceptual limitations, illustrate the impact of context and population specific understandings, point out subjugated knowledge, or identify variation in processes unmentioned in the summative literature.
Drisko (2005) suggests that you not only engage with the literature, but go beyond it. This means that you not only anticipate conceptual limitations, but, where applicable, suggest new definitions and contexts for the field to consider. As we suggested in Chapter 1, the discussion should once again engage in that collective conversation that repeats: what next? It is your task as a researcher to not only engage with what has happened in the past, but to apply it to the future, to attempt to answer the question of what is next for your (sub)discipline.
In addition to stating the contributions, you should also consider lessons learnt and propose recommendations. Consider the statement by Hou et al (2018, p. 1628) who noted that the results have implications for addressing principles and policies for immigrants’ integration in Canada:
…general principles of intercultural relations in all societies that could provide the basis for developing policies intended to improve relations and outcomes for both immigrants and those already settled in a country…[include]: a culturally and economically secure place for both newcomers and members of the larger society; opportunities for mutual engagement and social interaction; and support for establishing and maintaining multiple identities and social interactions during and after the settlement process.
In building from the previous statement we showed how Hou et al (2018) applied their findings to recommending cultural and economic security, opportunities for interaction, and support for multiple identities. Hou et al (2018) concisely demonstrate how a straightforward connection to the literature can be immediately developed into recommendations for policy or future research. If you find a similar consensus in your research (agreement between your findings and other researchers), consider how this relative certainty should be acted upon. If all researchers agree on something, ask yourself: does policy already engage with this fact? And if not, attempt to answer: how could policy alter or expand its work to address this finding?
Limitations and recommendations for future studies
Regardless of how groundbreaking and innovative your research is, it will have limitations. Think about the design, methodology, and theoretical insights that might limit the generalizability of your results and highlight them. If you are unable to identify any limitations in your study, we suggest two things: first, ask your supervisor or research mentor for their honest feedback on what could have made the study even better. Second, think about what you would’ve liked to do but was not able to do in the current project. Your research is personal and is likely dear to you, so you might be unwilling to state anything that could potentially undermine it, yet you must demonstrate reflexivity and the willingness to adopt an outsider looking in posture (i.e. what might your readers criticize your study for?). The transparency that highlighting your limitations offers will enable your readers to have a better understanding of your work (Cömert & Al-Beyati, 2020). They might also enable you to answer why you got the results you presented.
- Do not have too many limitations. Pick out the main ones e.g., “the study used secondary data, so I had no control over the variables” or “the study is purely quantitative; however, qualitative data would have provided deeper insights from respondents”
- Try to limit your discussion of your limitation to a paragraph (maximum two, if you need to expand on a point).
- Do not point to limitations that could’ve been easily resolved e.g. do not say “More insights could have been gleaned about this relationship if a regression was done instead of a correlation.” Since it is possible for you to do it, then you should just run the regression.
- Do not discredit your findings while highlighting your limitations. E.g., instead of saying “due to the fact that the data was collected two decade ago, it has little ramifications on present situation”, you could say “despite the fact that the data in 2 decades old, it provides baseline data and insights into how presentation generations approach this problem”
- Highlight the strengths: as the above example demonstrates, show that despite the limitations, the study has merits. Do not state a limitation without reaffirming the merits of the study
- Common limitations include sample characteristics, how the participants were selected, measurement, general methodology and analytical approach (Cömert & Al-Beyati, 2020). Remember, emphasize the strengths of your work as well.
Stating the limitations offers an excellent bridge-in for your recommendations and agenda for future research. This is important because it allows you to take stock of your contribution to the scholarship and outline a vision of where you want it to go. It is also a crucial step in affirming in your own mind what your next project might be. Remember, no project answers all the questions there are about a topic. Be sure to point on where you see the field going next. Here are some pointers to help you make suggestions for future research:
- Limitations: Based on your limitations, what might future research do to improve on those and possibly expand the scholarship.
- Findings: Think about what aspects of your findings might be (a) surprising (maybe confirmatory studies are needed); embryonic (if you discovered a new idea, outline some of the potential applications or way it can develop); new questions raised by your findings (your findings might raise more questions e.g, why did you uncover the results you got); theory (what theoretical or analytical approach could elevate the field?).
- Trends in field: Your might indicate how your research could fit into an emerging trend (e.g., using big data or decolonizing methodology to understand a phenomena)
Box 11.3 – Assessing Your Discussion
Greetham (2019, p. 227) suggests that to assessessing the success of your discussion by asking the following questions: Have I addressed
- what difficulties I encountered and how they affected the work plan?
- the limitations of the research and how they might have affected the strength of the findings? Did any bias enter into the research process?
- the strengths and weaknesses of my research and the data in respect to answering the research questions and hypotheses?
- the extent to which the data supported my propositions? How strong is the evidence? Is it conclusive, nebulous or tentative?
- whether the data confirm or falsify my proposition and why?
Students invariably ask how long should the discussion be? There are no rules concerning the length. As a benchmark, in most journal articles of 35 pages double-space, discussions are usually 2-3 pages long. Şanlı et al (2013, p.22) offers that:
Generally the length of the ‘Discussion ‘ section should not exceed the sum of other sections (introduction, material and methods, and results), and it should be completed within 6-7 paragraphs [of no] more than 200 words each.
Instead of worrying about the length, we suggest that you examine the checklist in Box 10.2 and ensure that you have covered everything. We also suggest that you get feedback from your mentor or supervisor before submitting the final version. Like most other things in the social sciences, there are no definitive rules to follow. This means that the length of your discussion section will be contextual and will be determined by the nature of your findings, the amount of explanation required (which will depend on the extent to which your findings contradict the literature or are surprising).
Another question that students sometimes have is: how much literature do I need to include in the discussion? Again, there are no definitive answers here. The issue is not the quantity but the quality of your arguments. You need to identify the main findings and examine them in light of the applicable literature. You need to identify which support, which contradict and what might help you to understand why you got the results you present. It is only those sources that you need to cite. Remember, you do not need to go in depth into the sources that you quote (unless they explain your results). Instead, be concise and do not lose your voice when incorporating the literature.
Cömert, A., & Al‐Beyati, E. S. (2019). Writing the Discussion Section for Original Research Articles. A Guide to the Scientific Career. Virtues, Communication, Research and Academic Writing, 523-526
Drisko, J. W. (2005). Writing up qualitative research. Families in Society, 86(4), 589-593.
Greetham, B. (2019). How to write your undergraduate dissertation. Macmillan.
Shellenberg & Berry (2018)
Hou, F., Schellenberg, G., & Berry, J. (2018). Patterns and determinants of immigrants’ sense of belonging to Canada and their source country. Ethnic and Racial Studies, 41(9), 1612-163