The discussion section explains what your research findings mean, how they relate to previous studies, and why they matter. Reporting results is not enough. Authors need to interpret their findings, consider alternative explanations, and show how their study contributes to existing knowledge.
A strong discussion makes the significance of a study clear without claiming more than the evidence supports. It connects the findings to the research question, places them in the context of relevant literature, and acknowledges the limits of the conclusions that can be drawn.
This guide explains how to write a discussion section in a research paper, compare findings with previous studies, address unexpected results, explain limitations, and develop meaningful implications. Practical examples, a paragraph-writing framework, and a final checklist will help you prepare your manuscript for journal submission.
What Is the Discussion Section of a Research Paper?
The discussion section interprets the findings presented in the Results section. It explains how those findings answer the research question, how they compare with existing evidence, and what conclusions can reasonably be drawn.
Although the exact structure varies across disciplines and journals, a discussion section commonly addresses five questions:
- What are the main findings?
- How should these findings be interpreted?
- How do they compare with previous research?
- What are their implications and limitations?
- What questions should future research investigate?
The discussion should develop an evidence-based argument rather than simply summarize information.
For example, a study might find that farmers who have more frequent contact with agricultural extension services are more likely to adopt water-conservation practices. The Results section would report the observed relationship and relevant statistical evidence. The Discussion section would explore possible explanations, compare the finding with earlier studies, and consider what the relationship may mean for agricultural advisory services.
In short, the Results section describes what the study found; the Discussion section explains how those findings should be understood.
Results vs. Discussion: What Is the Difference?
| Results section | Discussion section |
|---|---|
| Reports findings and relevant analyses | Interprets the meaning of the findings |
| Presents statistics, tables, and observed patterns | Explains why the patterns may matter |
| Describes relationships or differences identified in the data | Compares findings with previous research |
| Reports the outcomes of the analyses | Considers implications, limitations, and alternative explanations |
The two sections serve different purposes, although some journals combine them. Always follow the structure required by your target journal.
1. Begin with the Main Findings
Start by briefly identifying the findings that most directly answer your research question or address your hypotheses.
You do not need to repeat every statistic, table, or secondary result. Focus on the central findings and explain their relevance to the research question.
Weak example:
“The results of this study are presented in the previous section. Several interesting findings were obtained.”
This opening is vague and gives readers little indication of the study’s contribution.
Stronger example:
“The findings indicate that access to agricultural extension services was associated with greater adoption of selected water-conservation practices among the farmers surveyed.”
The second example communicates a specific finding without exaggerating its significance.
Pay attention to the strength of your wording. If your study is observational, saying that two variables “were associated” is generally more appropriate than claiming that one caused the other.
2. Interpret What Your Findings Mean
After presenting the main findings, explain their meaning in relation to your research question, theoretical framework, or hypotheses.
A result may be statistically significant, but readers still need to understand its meaning and relevance. Interpretation involves considering plausible explanations, identifying what the evidence supports, and distinguishing established findings from possible explanations.
Ask yourself:
- Does the finding support or challenge the original hypothesis?
- Does it extend an existing theory or reveal a pattern that deserves further investigation?
- What explanations are consistent with the data?
- Could other factors account for the observed relationship?
- What can the study establish, and what remains uncertain?
Example:
Suppose a study finds that farmers who participate in agricultural training report greater use of water-saving practices.
A basic interpretation might state that training participation was associated with adoption. A stronger discussion would consider whether training may improve farmers’ awareness of available practices or their understanding of implementation costs. It would also recognize that farmers who are already interested in conservation may be more likely to attend training.
Unless the study directly tests these explanations, they should be presented as possibilities rather than established facts.
The key is to distinguish between what your data demonstrate and what you infer from them.
3. Compare Your Findings with Previous Research
A discussion should place the current study within the existing body of knowledge. Comparing your findings with previous research helps readers understand whether your study confirms, extends, or challenges what is already known.
Avoid simply listing studies that reached similar conclusions. Explain how their findings relate to yours and whether differences in methods or context affect the comparison.
When Your Findings Agree with Previous Studies
Identify the shared finding and explain how your study adds to the evidence.
Example:
“Our finding that access to agricultural advisory services was associated with greater adoption of water-conservation practices is consistent with previous research highlighting the role of agricultural extension in disseminating information about sustainable farming. The present study extends this line of inquiry by examining the relationship in smallholder farming systems facing water scarcity.”
In an actual manuscript, support this comparison with verified references and ensure that the cited studies genuinely support the statement.
When Your Findings Differ from Previous Studies
Differences are not necessarily weaknesses. They may reflect variations in study populations, research designs, measurements, or contextual conditions.
For example, a study in one region may identify a positive association between training and conservation practices, while another finds a weak or statistically nonsignificant relationship.
Consider whether differences in training quality, access to equipment, farm size, or water availability could help explain the contrasting results. Also examine whether the studies used comparable definitions and measurements.
Do not choose an explanation merely because it supports your preferred interpretation. If the available evidence cannot establish the reason for a difference, acknowledge that uncertainty.
When Previous Research Is Limited
If relatively few studies have examined your specific question, explain how your findings contribute to an underexplored area. Avoid claiming that your research is the first of its kind unless a sufficiently comprehensive literature search supports that claim.
A Practical Framework for a Discussion Paragraph
A useful discussion paragraph often moves through five steps. This is a flexible framework, not a mandatory formula for every discipline or journal.
- Finding: State the result that matters for your research question.
- Comparison: Explain whether it agrees with or differs from relevant previous studies.
- Interpretation: Offer a plausible explanation that is consistent with the evidence.
- Qualification: Acknowledge uncertainty, alternative explanations, or limits in the study design.
- Contribution: Explain what the result adds to current understanding.
For example, a researcher studying educational inequality might report that students’ access to academic support was associated with higher course completion. The discussion could compare this pattern with previous studies, consider whether access to support differs by socioeconomic background, and acknowledge that an observational design cannot establish that support alone caused the difference. The researcher should support comparisons with verified studies rather than insert citations that have not been checked.
Not every paragraph needs all five steps. Use the framework to develop a clear argument, not to make every paragraph sound identical.
4. Explain Unexpected or Contradictory Results
Not every finding will support your hypothesis or match your expectations. Unexpected results deserve careful consideration, particularly when they are relevant to the research question.
Describe the finding accurately, then consider plausible explanations and whether they can be evaluated using your data or the existing literature.
Example:
“Contrary to the initial expectation, participation in agricultural training was not significantly associated with the adoption of all conservation practices examined. One possible explanation is that some practices require equipment or financial resources that training alone cannot provide. However, the present study did not directly test this explanation.”
This approach is more credible than ignoring an unexpected finding or presenting an untested explanation as fact.
Also, do not automatically interpret a nonsignificant result as proof that no relationship exists. Its meaning depends on the estimated effect, confidence interval, statistical power, measurement quality, study design, and wider evidence.
A medical study, for example, might find no statistically significant difference between two treatment groups. The discussion should consider the estimated difference and its uncertainty rather than conclude automatically that the treatments are equivalent. A claim of equivalence requires an appropriate design and analysis.
A balanced discussion addresses findings that complicate the main argument instead of focusing only on results that support it.
5. Explain the Implications of Your Research
The implications section answers an important question: Why should readers care about these findings?
Depending on your discipline, implications may be theoretical, methodological, practical, clinical, educational, or relevant to public policy.
Theoretical Implications
Explain whether your findings support, refine, or challenge an existing theory or conceptual framework.
Practical Implications
Describe how the findings might inform professional practice, institutional decisions, or future interventions.
Policy Implications
Explain whether the evidence may be relevant to policy development while recognizing the limits of the study’s setting and design.
Example:
“If the observed relationship is confirmed in further studies, agricultural advisory programs may benefit from combining technical training with information about the resources required to implement water-conservation practices.”
This statement identifies a potential application without claiming that the study has demonstrated the effectiveness of a specific intervention.
Avoid broad claims such as “These findings will transform agricultural policy” unless the evidence genuinely supports such a conclusion. The strength of your implications should match the strength and scope of your findings.
6. Acknowledge Study Limitations
Limitations affect how readers should interpret your findings and how far the conclusions can be applied. They may arise from the sampling strategy, research design, measurement methods, missing data, analytical assumptions, or the context in which the research was conducted.
Rather than listing weaknesses without explanation, identify the limitations that materially affect interpretation and clarify their consequences.
Example:
“The study relied on self-reported information about farmers’ conservation practices, which may be affected by recall error or social desirability. Consequently, the reported adoption levels may not correspond precisely to observed farming practices. Future studies could combine survey data with direct field observations.”
Similarly, a cross-sectional study may identify associations but ordinarily cannot establish the temporal order needed to support causal conclusions. A sample drawn from a single institution or region may also limit the applicability of the findings to other settings.
For more practical guidance on preparing your manuscript, explore the related resources in the ManuscriptX Insight section.
7. Suggest Specific Directions for Future Research
Future research recommendations should follow logically from the findings, unanswered questions, or limitations of the current study.
Avoid ending the discussion with a generic statement such as “Further research is needed.” Instead, identify the question that remains unanswered, the population or context that deserves attention, or the research design that could provide stronger evidence.
Generic recommendation:
“Further research should investigate agricultural extension services.”
More specific recommendation:
“Longitudinal studies could examine whether changes in farmers’ access to extension services precede changes in water-conservation practices and whether the relationship differs by farm size or access to irrigation equipment.”
The second recommendation identifies a specific research need and suggests how future work could address it.
One or two well-justified recommendations are often more useful than a long list of generic suggestions.
8. Common Discussion Section Mistakes
Repeating the Results Section
Do not reproduce tables, statistics, and descriptive results without explaining their meaning. Refer to the most relevant findings and focus on interpretation.
Making Claims That Exceed the Evidence
Avoid causal language when the research design supports only an association. Do not generalize beyond the population or setting examined.
Ignoring Contradictory Findings
Discuss unexpected and nonsignificant results when they matter to the research question. Selectively emphasizing supportive findings can weaken the credibility of your argument.
Citing Studies Without Comparing Them
A list of references does not create a strong discussion. Explain how previous findings support, differ from, or help contextualize your results.
Introducing New Results Without Proper Presentation
The discussion should interpret findings reported in the Results section. If additional analyses are needed, present them in the appropriate section and explain their relevance clearly.
Listing Limitations Without Explaining Their Consequences
Statements such as “The sample was small” are incomplete unless you explain how the sample size affects precision, interpretation, or applicability.
Overstating Practical Implications
Distinguish between implications supported by the current evidence and recommendations that require further investigation.
9. A Practical Discussion Section Example
The following example is hypothetical. It illustrates how a discussion might be developed from an assumed research finding and is not a report of an actual study.
Research question: Is access to agricultural extension services associated with farmers’ adoption of water-conservation practices?
Assumed finding: Farmers with greater access to extension services reported higher adoption of selected conservation practices.
Example Discussion
“The findings indicate that access to agricultural extension services was associated with greater adoption of selected water-conservation practices among the farmers surveyed. This result suggests that agricultural advisory services may help farmers learn about and evaluate conservation options in water-limited farming systems.
One possible explanation is that extension services provide information about the availability, application, and potential benefits of conservation practices. However, access to advisory services may also be related to other characteristics, including farmers’ education, financial resources, or existing interest in sustainable agriculture. Because the study design does not establish causality, the observed association should not be interpreted as proof that extension access directly caused adoption.
The finding can be considered alongside previous studies examining the role of agricultural information and advisory services in technology adoption. Comparisons should account for differences in study populations, the types of practices examined, and the ways extension access and adoption were measured. These differences may help explain why the strength of the relationship varies across research settings.
The study also has limitations. Reliance on self-reported adoption may introduce measurement error, and findings from a limited geographical area may not be generalizable to farming systems with different resource constraints. Future research could combine longitudinal data with direct observation of farming practices to examine whether changes in extension access are followed by changes in adoption.
Taken together, the findings highlight the potential relevance of agricultural advisory services while indicating the need to consider the wider conditions that influence farmers’ decisions.”
Why Does This Example Work?
The discussion moves from the main finding to its interpretation, considers alternative explanations, relates the result to previous research, and identifies limitations and a specific direction for future investigation. It also avoids claiming causation from an assumed association.
In a real manuscript, each interpretive claim should be supported by the study’s evidence and relevant literature.
10. Final Checklist Before Submission
Before submitting your manuscript, check whether your discussion:
- Answers the research question introduced earlier in the paper.
- Highlights the main findings without repeating every statistic.
- Explains what the findings mean rather than merely restating them.
- Compares the findings with relevant previous studies.
- Addresses unexpected or contradictory findings where appropriate.
- Distinguishes evidence from speculation.
- Acknowledges limitations that affect interpretation.
- Explains implications in proportion to the evidence.
- Suggests specific and justified directions for future research.
- Follows the target journal’s structure and author instructions.
A useful final test is to ask: Could a reader understand what this study contributes and what its findings do not establish after reading the discussion? If not, revise the section to clarify both the contribution and the boundaries of the evidence.
How Academic Editing Can Improve Your Discussion Section
A strong discussion requires more than grammatical accuracy. It requires clear reasoning, appropriate interpretation, careful comparison with previous research, and language that reflects the strength of the evidence.
A professional editorial review can help identify repetitive passages, unclear explanations, unsupported or overstated claims, inconsistent terminology, and weak transitions between findings and implications. It can also help authors make the relationship between their findings and the relevant literature easier to follow, while preserving the intended scientific meaning.
At ManuscriptX, we provide human academic editing focused on clarity, consistency, and the effective presentation of scholarly research. Our editorial support can help authors express their findings more precisely while preserving the intended meaning and scientific content of their work.
Editing cannot replace sound research design, validate unsupported conclusions, or guarantee journal acceptance. Authors remain responsible for the accuracy and integrity of their research.
If you are preparing a manuscript for submission, request a Free Assessment. You will receive a tailored quotation before any payment is required.
You can also explore our Academic Editing services to identify the support that best suits your manuscript.
References & Further Reading
- Hess, D. R. (2023). How to write an effective discussion. Respiratory Care, 68(12), 1771–1774. https://doi.org/10.4187/respcare.11435
- McCombes, S. (n.d.). How to write a discussion section: Tips and examples. Scribbr. https://www.scribbr.com/dissertation/discussion/
- University of California, Irvine. (n.d.). Discussion: Writing a scientific paper. https://guides.lib.uci.edu/scientificwriting/discussion
Editorial note: Discussion section conventions differ across disciplines and journals. Always consult the current author instructions of your target publication before submitting your manuscript.