How to Write a Discussion Section
A complete guide to writing a clear, insightful discussion section, from interpreting your results to discussing limitations and implications, with structure and common mistakes.
Guide
The discussion section is where your research findings become a scientific contribution. It is where you interpret your results in the context of existing literature, explain what they mean, acknowledge what they don't mean, and argue for their significance. Many researchers find the discussion hardest to write because it requires the most synthesis. This guide breaks down the discussion section into its essential components and explains how to execute each one effectively.
The standard discussion structure
An effective discussion section follows a recognizable pattern that mirrors the introduction in reverse, moving from specific findings back to broad implications:
Summary of main findings: restate (not repeat) your key results in plain language
Interpretation: explain what each finding means
Comparison to literature: how do your findings agree or disagree with prior work?
Mechanisms: why might these results be true?
Limitations: honest acknowledgment of study weaknesses
Implications: what do these findings mean for the field, practice, or future research?
Conclusion: a brief, definitive take-home message
How to interpret your results
The first paragraph of the discussion should state your most important finding clearly and directly: 'The main finding of this study is that X is associated with Y'. Then interpret it. What does this mean? Why might this be? Don't just restate what the results section already showed; add an interpretive layer that the results section lacks.
How to compare with existing literature
Every major finding should be situated against what prior studies have found. For consistent findings: 'This is consistent with previous studies showing X'. For inconsistent findings: explain possible reasons for the discrepancy (different population, different measures, different time period). Inconsistencies are not weaknesses; explaining them demonstrates scientific maturity.
How to write a limitations section
Limitations should be honest, specific, and proportionate. Avoid generic limitations ('our sample may not be generalizable') without explaining why and how. Good limitations acknowledge specific constraints (sample size, single-site study, cross-sectional design, self-report measures) and discuss how they might affect the conclusions. Do not omit significant limitations, as reviewers will identify them anyway.
Common discussion section mistakes
The most frequent discussion errors:
Repeating the results section without interpretation; narrate your numbers rather than just restating them
Overclaiming by asserting causation from correlational data or generalizing beyond your sample
Ignoring inconsistencies with prior literature
Vague limitations ('future research should address this') without specific guidance
Missing implications, such as failing to explain why your findings matter
Too long: discussions that repeat points or go in circles lose reviewer patience
Ready to check your manuscript?
Upload your paper and get a submission readiness report in under 2 minutes.
Check my manuscript, it's freeNo account · PDF, Word, LaTeX · Results in <2 min
Frequently asked questions
Typically 600–1,200 words for a journal article. The discussion should be thorough enough to interpret findings meaningfully but concise enough to maintain focus. Avoid repeating information from the results section at length.
No: the discussion section interprets results that were presented in the results section. Do not introduce new data or analyses that were not presented earlier. If you realize you need additional analysis, add it to the results section.
Present tense for interpreting findings and discussing their meaning ('These results suggest...', 'This finding indicates...'). Past tense when referring to specific study procedures or results from your study ('We found that...'). Present tense when discussing established literature ('Smith (2020) shows that...').
Yes: the discussion typically ends with a conclusion paragraph that states the key contribution of the paper and the main take-home message. Some journals require a separate Conclusions section; others include it at the end of the Discussion. Check your target journal's format.