If your Results section has sentences like "this finding clearly shows that mindfulness works," you've crossed a line that reviewers will catch immediately. The Results section of a research paper reports what you found — numbers, tables, figures, and narrative that describes them — but it does not interpret. Interpretation belongs in the Discussion. Mixing the two is one of the fastest ways to draw "revise and resubmit" comments, because reviewers want to see the evidence cleanly before hearing what you think it means.
PaperDraft is a writing assistant, not a paper generator — the draft is your starting point, not your submission. You are responsible for editing, verifying sources, and following your school's academic integrity policy.
This guide walks through how to structure a results section, when to use tables vs figures, the narrative rules that make reporting readable, and the Results-vs-Discussion line. For paper-level context, see research paper structure.
What the Results Section Does (and Doesn't Do)
Results has one job: present what the data show, neutrally. A strong Results section does four things:
- Describes the sample. Demographics, response rates, attrition.
- Reports descriptive statistics. Means, standard deviations, frequencies, proportions.
- Reports inferential tests (for quantitative work). Test statistics, degrees of freedom, p-values, effect sizes, confidence intervals.
- Presents tables and figures with their narrative companions.
What it does not do:
- Interpret findings ("this suggests...").
- Compare to prior studies.
- Discuss implications.
- Report results that aren't connected to a research question or hypothesis.
The test for whether a sentence belongs in Results: does it describe what the data show, or does it say what the data mean? If the latter, move it to Discussion.
How to Structure the Results Section
A working order that fits most empirical papers:
1. Sample characteristics
Open with a brief description of who was in the sample. One short paragraph.
Example: "Of 154 recruited participants, 150 completed the post-test (97% retention). Participants had a mean age of 19.7 (SD = 1.2), and 62% identified as female. Demographic characteristics are reported in Table 1."
2. Descriptive statistics
Report means, standard deviations, and other descriptive measures for your key variables. A table usually handles this efficiently.
3. Analysis for each hypothesis or research question
Walk through each hypothesis in the same order as the Introduction. For each: state the test, report the numbers, cite any relevant table or figure.
Example: "To test H1, that the mindfulness condition would show lower post-test anxiety, we conducted an independent-samples t-test. Participants in the mindfulness condition (M = 14.2, SD = 3.1) reported significantly lower anxiety than those in the waitlist control (M = 18.7, SD = 4.0), t(148) = 7.63, p less than .001, Cohen's d = 1.25."
Write "p less than .001" or escape the sign. Never write a raw less-than next to a digit in MDX.
4. Secondary or exploratory analyses
Report last, and label them clearly as exploratory. Don't bury them among primary analyses.
Tables vs Figures vs Narrative: When to Use Each
The rule of thumb: if a reader could extract the number faster from a table than from a sentence, use a table. If the shape of the data matters (a distribution, a trend, a relationship), use a figure. Narrative connects the two.
Use a table when
- You have more than three or four related numbers (means and SDs across multiple conditions or variables).
- Readers may want to look up a specific value.
- Precision matters more than visual impression.
- You're reporting regression coefficients with their standard errors and p-values.
Use a figure when
- You're showing a trend over time.
- You're showing the shape of a distribution.
- You're visualizing an interaction (different slopes for different groups).
- A pattern would be hard to see in a table.
Use narrative when
- You have one or two headline numbers that readers need to notice.
- The numbers are connected by a short causal or comparative story.
- You're summarizing what a table or figure shows in words.
Every table and figure should have a caption that explains what it shows. Readers should understand the table or figure without reading the full text.
Narrative Rules That Keep Results Readable
Even with tables doing most of the lifting, the narrative has to land.
Report test statistic, degrees of freedom, p-value, and effect size. Every inferential test. APA format: t(148) = 7.63, p less than .001, d = 1.25.
Report the direction plainly. "Scores were lower in the mindfulness condition" is clearer than "a significant effect was observed."
Avoid causal language unless the design supports it. A correlation cannot say "X caused Y." An experiment with random assignment can.
Don't repeat table contents verbatim in the text. Summarize the pattern; refer to the table for details.
Use past tense. "Participants reported," not "participants report."
Example: Results Paragraph in Action
A condensed sample, from a psychology intervention study:
Sample description: "Of 154 enrolled participants, 150 (97%) completed the post-test. Demographic characteristics are reported in Table 1."
Descriptives: "Descriptive statistics for state anxiety scores are reported in Table 2. Baseline anxiety did not differ between conditions, t(148) = 0.42, p = .68."
Primary analysis: "To test the hypothesis that participants in the mindfulness condition would report lower post-test anxiety, we conducted an independent-samples t-test. Post-test anxiety was significantly lower in the mindfulness condition (M = 14.2, SD = 3.1) than in the waitlist control (M = 18.7, SD = 4.0), t(148) = 7.63, p less than .001, d = 1.25."
Exploratory: "Exploratory analyses examined whether the effect varied by baseline anxiety level. A regression with baseline anxiety as a moderator yielded a non-significant interaction (p = .42), suggesting a similar effect across baseline levels."
Four short paragraphs, each doing one job, each reporting numbers without interpreting them.
Numbers run but the blank Results section is still staring at you? PaperDraft gives you a structured first draft — IMRaD skeleton, Results scaffolding with table/figure placeholders in academic register — so you can spend your time on actual analysis instead of formatting. It's a drafting assistant, not a submission. Try PaperDraft — free
Common Mistakes in Results Sections
A few patterns keep showing up in first drafts.
Interpreting inside Results. "This suggests mindfulness reduces anxiety." Cut. Move to Discussion.
Reporting only the p-value. A p-value alone is not enough. Include the test statistic, degrees of freedom, and effect size.
Selective reporting. Dropping analyses because they weren't significant is a research-integrity problem. Report everything you planned to test.
No effect size. Statistical significance without effect size tells readers nothing about magnitude. Report d, r, eta-squared, or odds ratios depending on the test.
Dense paragraphs with too many numbers. If a paragraph has more than four or five numbers, consider moving them to a table.
Tables without captions. Every table and figure needs a standalone caption.
Using "significant" without "statistically." "Significant" in a statistical sense is different from "meaningful." Be precise.
How a Drafting Assistant Fits
A drafting tool can scaffold the Results section's subsection structure, give you placeholder narrative for common test types, and flag when your draft drifts from description into interpretation. What it cannot do is run your statistics, produce your tables and figures, or verify that your numbers match what your analysis actually yielded. PaperDraft handles the structure and register. You handle the data, the calculations, and the honest reporting of whatever came out.
FAQ
How long should the Results section be?
Varies widely. For an undergraduate paper, typically 500-1,200 words plus tables and figures. For a thesis or journal article, often longer.
Should I include descriptive statistics in the text or only in a table?
Put them in a table if there are more than two or three values. Mention the key takeaway in text and refer the reader to the table.
Can I include figures instead of tables?
Yes, for the right data — trends, distributions, interactions. Tables work better for precise values; figures work better for shapes and patterns.
How do I handle non-significant results?
Report them the same way as significant ones. A non-significant test is still a result, and hiding it is selective reporting. See null hypothesis examples for how to frame.
Where do raw data and full output go?
Not in Results. Summary statistics and key tests belong here. Full output (R scripts, full regression tables, data codebooks) belongs in supplementary materials — see research paper appendix.
Once Results is clean and neutral, the Discussion has room to do its job. For the next section — interpretation, limitations, and implications — see writing the discussion section.