Writing a discussion section
One of the most difficult parts of writing an article based on your research is the discussion section. My PhD students struggle with it, but established researchers have difficulty with it as well because the discussion section is where you have to put your money where your mouth is, so to speak. You did the background research, you designed and conducted the study, and you wrote up your findings; now you have to explain why all of that mattered, how it all fits into the existing landscape, and what should be done because of it. In this post, I go over some tips for writing a strong discussion section.
General Guidance for Discussion Sections
Your discussion section is where you get to discuss your contribution to the existing body of knowledge. You get to explain why your research is important and what it means in the context of what we already know. It’s also a place for you to make theoretical contributions and to provide specific recommendations for future research, and for policy and practice. It’s actually exciting when you think about it. This is an opportunity to use your research to make a difference. Below are some general guidelines for discussion sections.
Your discussion section should situate your findings in the existing research (use your literature review).
Guiding questions for situating your findings in the literature:
What does my study contribute to the existing body of knowledge?
How do my findings support what is found in the existing literature? Where do my findings converge with what we already know?
Where do my findings diverge from the existing literature? Why might that be?
Don’t neglect your theoretical framework. Your discussion section should also situate your findings in the existing theory and apply your framework.
Guiding questions for applying your theoretical framework:
How do my findings fit within my theoretical framework?
Are there any places where my findings do not fit? What does that mean? Why might that be?
What, if any, are my theoretical contributions?
Whether included in the same section as the discussion or in a separate section, you should include implications (Why is your research important? Why should we care? What does it all mean?)
Recommendations are important! Think in terms of what your study means for research, policy, and practice.
Be specific – what EXACTLY should be done?
Recommendations for future research:
What should future researchers on this topic study?
How should they study it?
Implications/Recommendations for Policy:
How, if at all, should existing policy change because of what you found in your study?
Should new policy be created? What should it look like?
Based on your findings, what should policymakers do?
Implications/Recommendations for Practice:
What recommendations do you have for practice/practitioners in your area of study?
What changes should be made? How should these changes be made? By whom?
Examples from Discussion and Implications Sections
Below are some examples from a few of my previous studies.
Situating Findings in the Literature (from Bingham & Ogunbowo Dimandja, 2017)
At CTA, teachers adjusted their lessons or curricula based on continuous analysis of student assessment data. This consistent data use fostered teachers’ understanding of students’ progress and instructional needs and aligned with the organization- al emphasis on holding students accountable. This supports prior research that teachers in PL models utilize students’ mastery and achievement data to provide educational recommendations and tailor instructional content for individual students (Bill & Melinda Gates Foundation, 2014; Pane et al., 2015). However, our research expands on these previous studies by comparing how veteran teachers engaged in and responded to this emphasis on student data and accountability with how newer, more inexperienced teachers responded to these key elements of PL. Indeed, as sensemaking theory indicates, teachers drew on their prior experiences and constructions of “good teaching” to enact the elements of PL.
Applying the Theoretical Framework (from Bingham & Ogunbowo Dimandja, 2017)
Through the lens of sensemaking theory, newer teachers’ positive responses to the analysis and use of student data may be based on superficial knowledge related to their belief of what they viewed as the most important aspect of the personalized learning model – increasing student test scores based on the school leadership’s enforcement of student data analysis and accountability. This may be due to a more limited capacity to access abstract principles as a result of their relatively novice level of understanding (Spillane et al., 2002). Conversely, veteran teachers’ advocacy for improving students’ overall development may be dependent on their expert ability to focus on deeper principles and pay attention to conceptually significant aspects of the personalized learning model that would result in permanent changes, rather than superficial changes. Indeed, as many scholars have indicated, teachers’ professional identities can influence how they respond to reforms (Schmidt & Datnow, 2005; Vähäsantanen, 2015).
Making a Theoretical Contribution (from a working paper)
Integrating CHAT and sensemaking theory to analyze change at multiple levels. Using CHAT and sensemaking together to make sense of the case of Blended Academy allowed me to analyze changes at multiple levels. In using these two theories together, I was able to create a fuller picture of the evolution of Blended Academy, moving between an organizational-level analysis and an individual-level analysis. In so doing, I was able to not only see how the organization changed as an activity system – experienced a process of expansive learning – but also keep an eye on how the individuals within the system shaped those changes and how they themselves experienced change. The use of sensemaking theory showed me that teachers themselves did not experience substantive change (schema reconstruction) even though the system did experience substantive change. The use of either one theory or the other would have painted a different picture, perhaps obscuring key organizational changes or individual sensemaking processes. Further, the use of sensemaking theory allowed me to zoom in on how individuals constructed and reconstructed the Object of activity – a key process in CHAT. Thus, using sensemaking in conjunction with CHAT may be key to future studies that require analysis of multiple levels of change.
Recommendations for Practice and Policy (from Bingham et al., 2018)
In the context beyond the school, it is important to identify teacher needs in a personalized learning model and develop digital content and learning platforms that satisfy those needs. For example, it would be useful for schools to have access to learning platforms that can integrate data from multiple sources. Second, it is critical that we begin to identify best practices and create teacher training and professional development around those practices. Finally, it will be important to rethink how school and student success is measured to make room for innovative school models. For example, rather than multiple-choice assessments tied to particular grade levels, state assessment could include more flexible options for demonstrating progress, and ultimately competency, through mastery of standards or objectives.
Recommendations for Future Research (from Bingham et al., 2018)
Future research on personalized learning school models should aim to identify exemplar schools and teachers by exploring student outcomes in these models and linking these outcomes to specific practices and strategies. Once these effective schools and teachers are identified, researchers should also seek to provide a descriptive analysis of these exemplar schools and teachers to contribute some understanding of best practices in innovative school models. To understand larger contextual factors influencing innovation in schools, future research should examine what districts and states are doing to facilitate the implementation and success of innovative school models. Finally, if there is to be innovation in traditional school contexts, new methods of measuring school and student success in innovative school contexts should be explored. Generative questions for future research might include: (1) What supports are necessary for teachers to make high-tech changes to their practice? (2) How do students adapt to new forms of high-tech instruction? (3) How do personalized learning models influence various student outcomes? (4) What are some best practices for personalizing learning? (5) How can districts or states align their measurement processes with those of innovative school models that use alternative grading practices? (6) How are districts supporting teachers in developing and finding digital resources and implementing innovative teaching strategies?
Ultimately, your discussion section pulls it all together for your readers. Look at it as an opportunity to amplify your study’s contribution to the knowledge base.
*photo by Nick Morrison on Unsplash