How to write a results section
Short answer: The results section reports what you found, in a logical order and without interpretation. Organize it around your research questions or hypotheses, start with descriptive statistics and any preliminary checks, then report each main analysis with the test statistic, degrees of freedom, exact p value, effect size and confidence interval, and use tables or figures for detail rather than repeating every number in the text. Explanations of why results occurred and what they mean belong in the discussion.
On this page
The results section answers one question: what did you find? It is the most factual part of a paper. Readers should be able to see, quickly and without spin, whether your data support each hypothesis or answer each research question. Interpretation, comparison with other studies and implications come later, in the discussion.
What belongs in the results (and what does not)
| Include in the results | Save for the discussion | Put in the methods |
|---|---|---|
| Descriptive statistics for key variables | Why a result occurred | How data were collected |
| Preliminary checks: missing data, assumptions, reliability, manipulation checks | How results compare with earlier studies | Which analyses you planned and why |
| Each main analysis with its statistics | What results mean for theory or practice | Software and settings used |
| Results that do not support your hypotheses | Speculation about mechanisms | Exclusion criteria |
| Pointers to tables and figures | Limitations | Measures and their sources |
Organize around your questions
The clearest results sections follow the order of the research questions or hypotheses stated in the introduction. A common sequence is:
- Sample and descriptives: final sample size after exclusions, means, standard deviations and correlations for key variables.
- Preliminary analyses: missing data, assumption checks, measurement reliability, manipulation checks.
- Main analyses: one subsection per hypothesis or question, in the order you introduced them.
- Additional or exploratory analyses: clearly labeled as such, separate from the planned tests.
Use subheadings that tell readers which question each part answers, for example "Hypothesis 2: Boundary integration as a mediator".
Report each test completely
For each analysis, state which test you ran, the result in words, and the statistics that support it. APA 7 asks for exact p values (to two or three decimals, or p < .001 when smaller), effect sizes, and confidence intervals where possible. Statistical symbols such as M, SD, t, F, p and r are italicized.
| Statistic | APA 7 format example |
|---|---|
| Mean and standard deviation | M = 4.21, SD = 0.87 |
| Independent-samples t test | t(48) = 2.31, p = .025, d = 0.65 |
| One-way ANOVA | F(2, 117) = 5.43, p = .006, η² = .08 |
| Chi-square test | χ²(1, N = 200) = 4.12, p = .042 |
| Correlation | r(98) = .34, p < .001 |
| Regression coefficient | b = 0.21, SE = 0.05, 95% CI [0.11, 0.31], p < .001 |
Two details are easy to get wrong. First, values that cannot exceed 1, such as p and r, are written without a leading zero (.034, not 0.034). Second, a non-significant result is reported with its exact p value, not as "p = n.s.".
Text, tables or figures?
- Text works for a few numbers that tell the main story.
- Tables suit exact values that readers may want to look up: descriptives, correlation matrices, regression models.
- Figures show patterns, trends and interactions better than numbers do.
Do not repeat every number from a table in the text. Instead, point to the table and highlight what matters: "As Table 2 shows, remote-work intensity was positively associated with exhaustion, and this association remained after controlling for workload." Every table and figure needs a number, a clear title and a mention in the text.
Writing the sentences
Lead with the finding, then give the statistics in support. Write in the past tense for what you found.
Report results that do not support your hypotheses just as plainly: "Contrary to Hypothesis 3, team size was not associated with exhaustion, b = 0.02, SE = 0.04, p = .62." Readers trust a results section more when it reports what did not work.
Qualitative results
In qualitative studies, the results (often called findings) are organized around themes or categories rather than tests. For each theme:
- Name and define it in one or two sentences.
- Explain how widespread it was, without false precision (for example, "most participants" or "8 of 12 participants").
- Support it with short, well-chosen quotations, attributed with a participant code.
- Show variation or exceptions within the theme.
Many qualitative papers combine findings and interpretation in one section; follow the conventions of your field and target journal.
If you do not have your data yet
Proposals and early drafts often include a results template. That is fine as long as it is obviously a placeholder: table shells with empty cells and bracketed notes such as [insert coefficient, SE, p]. Never fill a results section with invented or "example" numbers that could be mistaken for real findings; presenting fabricated data as real is research misconduct.
Common mistakes
- Interpreting or explaining results instead of reporting them
- Reporting only significant results
- Giving p values without effect sizes or confidence intervals
- Repeating all table values in the text
- Introducing analyses that were not described in the methods without labeling them as exploratory
- Inconsistent numbers between text, tables and abstract
- Using "proved" or "confirmed" where "supported" or "was consistent with" is accurate
Quick checklist
- Is the section organized around your research questions or hypotheses?
- Did you report the final sample size and descriptive statistics?
- Does every test include the statistic, degrees of freedom, exact p value and an effect size?
- Are non-significant and unexpected results reported?
- Is every table and figure numbered, titled and referred to in the text?
- Do numbers match across the text, tables and abstract?
- Have you kept interpretation for the discussion?
Once the results are written, the next step is to interpret them; see how to write a discussion section and, for the weaknesses that shape interpretation, how to write the limitations section.