Write research paper discussion better than 99% of researchers

A publishable discussion section hinges on four distinct moves—explaining results, interpreting what they mean, proposing future research, and spelling out practical implications for real stakeholders. The core workflow starts with a structured brainstorming session: spend one to two hours dumping ideas under those four headings, then turn the resulting notes into a discussion that reviewers can see as logically grounded, conceptually meaningful, and actionable.

The explanation component comes first. When results align with prior studies, the discussion should focus on why the findings match what the field already knows. When results diverge, the discussion needs plausible drivers—mechanisms, context differences, sampling issues, or other conditions that could produce the mismatch. Interpretation follows: the discussion should clearly state what the results suggest and how they should be understood, not just restate outcomes. From there, the discussion should look forward. Suggestions for future research identify what remains unknown and what new evidence would best resolve the study’s limitations or open questions. Finally, practical implications translate findings into decisions and behavior. The transcript emphasizes tailoring these implications to stakeholder groups—such as nurses, doctors, patients in medicine; teachers, students, and policy makers in education—so the discussion reads as useful beyond academia.

To make these moves concrete, the transcript uses a case study from English language teaching course book authorship. The study finds that roughly 90% of course book authors are white native speakers from the UK, despite the broader English-speaking population being largely non-native and not UK/white. A proposed explanation is “homophilious hiring,” a pattern where recruiters tend to select candidates who resemble themselves. The transcript notes similar evidence across disciplines: white male hiring groups are more likely to hire white male candidates, while groups dominated by black women tend to hire more black women. The course-book study did not directly collect data on commissioning editors or publishers, so the mechanism remains plausible rather than proven.

To strengthen the discussion when discipline-specific evidence is scarce, the transcript recommends using Consensus, an evidence-oriented search tool that returns references and summaries. The workflow is: query whether commissioning editors and publishers are predominantly white; verify relevance by checking abstracts; use international survey evidence from publishing employees when no English-language-teaching-specific data exists; and then incorporate relevant passages into the brainstorming notes. The process also relies on reference chaining: use “citing literature” to find additional papers that build out diversity, equity, and inclusion angles and practical implications.

In the example, the author uses external research on inequity in publishing to justify what publishers could do to promote diversity in authorship. The discussion then becomes more than a description of imbalance—it turns into a set of evidence-backed recommendations, with acknowledgments that the evidence comes from adjacent fields. The transcript frames this as a key reason Q1 journals are more likely to accept the work: reviewers see both breadth of knowledge and a discussion that connects results to established research and real-world action.