Writing a discussion that is not a second results chapter
What your findings mean, how they sit against other work, and what you cannot claim.
A discussion states the main finding in words, compares it with what other studies found and explains any disagreement, gives the mechanism or reason where one is plausible, states the limitations honestly, and says what should follow. It does not repeat numbers already in the results.
Why it matters
This is where most marks are won and lost, because it is the only chapter that shows you understand your own findings rather than having merely produced them.
An example
Weak: 'Our study found a significant difference in pain (p = 0.018). Sharma et al. also found a significant difference.'
Better: 'Supervised exercise reduced pain by about 1.3 points more than a home programme, a difference patients would notice. This is consistent with Sharma et al., who reported 1.1 points in a similar population, and larger than the 0.4 points reported by Menon et al. — whose home-exercise group received weekly telephone review, which may have narrowed the gap.'
Common mistakes
- Restating every result with its p-value.
- Limitations written as a formality — 'the sample size was small' with no statement of what that means for the conclusion.
- Claiming causation the design cannot support.
- Concluding more than the primary objective was powered to answer.
Read next
- Writing a statistical result the way a journal expects — Test statistic, degrees of freedom, p-value, effect size — in that order, in a sentence.
- Bias and confounding, in plain words — Bias is a fault in how you got your data. Confounding is a third variable explaining your result.
Chavery Research Companion applies this to your own study: it asks the questions in plain language, checks the assumptions against your data, recommends the test, and writes the sentence that reports it. Start free — planning and the master chart cost nothing.