2부 · 원고 쌓아 올리기 — 13화

Writing the Discussion and Conclusion

Chapter purpose. By the end of this chapter, you should be able to apply the central concepts to a real assignment, research paper, thesis, or article in your own discipline.

From result to meaning

The discussion explains what the findings mean, why they may have occurred, how they relate to previous scholarship, and what they contribute. Begin with the principal answer rather than repeating the entire results section. Distinguish explanations supported by evidence from plausible alternatives. Return to the literature and framework. Identify where findings support, extend, complicate, or contradict prior work. Use the theoretical or conceptual framework as an analytical resource, not a vocabulary list. A surprising result is not automatically a contribution; explain its significance and consider methodological or contextual explanations.

Implications and limitations

Implications should identify who could reasonably act on the findings and under what conditions. Limitations define the boundaries of interpretation. Sample, setting, measurement, language, positionality, and design may all matter. A limitation does not require apology; it requires an account of how readers should calibrate the claim.

Concluding with proportion

A conclusion should synthesize the problem, approach, principal finding, and contribution. It may identify a focused direction for future work. Avoid introducing new evidence, making universal claims, or ending with a generic statement about the importance of research.

Discussion, transfer, and responsible limitation

A strong discussion begins with the most defensible answer to each research question, then explains how the results relate to prior scholarship and the conceptual framework. It distinguishes findings from possible explanations and identifies alternative interpretations. The conclusion should not introduce a new evidentiary claim. Calibrate transfer claims to the design. Statistical generalization, analytic generalization, theoretical transfer, and practical relevance are different. A small qualitative study may illuminate a mechanism or provide transferable insight without estimating prevalence. A controlled experiment may estimate an effect under specified conditions without establishing success in implementing it across every Korean university. Limitations should explain consequences, not merely list weaknesses. State which inference is restricted, whether the limitation likely changes magnitude, direction, credibility, or transferability, and what evidence would reduce the uncertainty. Avoid the ritual claim that "more research is needed"; specify the next design, comparison, population, or dataset.