What should be included in the results section of a research paper?

The results section includes the preliminary analyses, such as descriptive statistics and assumption checks, then the main inferential results in the order of your research questions. Each finding is reported with its test statistic, exact p-value, and an effect size, with detail placed in tables and figures.

Should the results section include interpretation?

No. The results section states what you found without explaining why or comparing it to other studies, which belongs in the discussion. Reporting a difference and its statistics is a result; saying what that difference means about your hypothesis is interpretation.

What is the difference between the results and discussion sections?

The results section presents the evidence, the numbers and tests, in a neutral, factual way. The discussion interprets those results, relates them to prior research, acknowledges limitations, and draws conclusions. Keeping them separate is a basic expectation of academic writing.

Do you report non-significant results?

Yes. Report all results relevant to your research questions, including non-significant and unexpected ones, with their statistics. Omitting or burying null findings distorts the evidence and undermines the credibility of the paper.

Results Section of a Research Paper: Guide + Example

To write the results section of a research paper, you report what your analysis found in the past tense, in the same order as your research questions, presenting each finding with its test statistic, exact p-value, and an effect size, and using tables and figures for detail rather than repeating every number in prose. The section that passes review states the findings plainly and without interpretation, that belongs in the discussion. This guide covers the structure, the reporting format, and the line between results and discussion that trips most authors up.

What the results section is for

The results section presents the evidence, nothing more. It answers each research question with the relevant numbers and points the reader to the table or figure that holds the detail. It does not explain why a result came out as it did, compare it to past studies, or admit limitations; those are the discussion's job. A results section in a journal article is tighter than a dissertation results chapter, which carries more scaffolding, but the reporting rules are identical.

Structure it around your research questions

Order the results to mirror the questions or hypotheses set out in your introduction and named in your methods section. Open with the preliminary analyses, the descriptive statistics, assumption checks, and any data screening, then move to the main analyses in the same sequence as the hypotheses. This one-to-one mapping between question, method, and result is what makes a paper easy to review and hard to fault.

Report descriptives before inferential tests

Lead each finding with the descriptive statistics, the means, standard deviations, frequencies, or correlations, before the inferential test. A reader needs to see the pattern in the data before the test that evaluates it. The distinction between describing a sample and inferring to a population is set out in descriptive versus inferential statistics, and getting the order right is a small thing that signals statistical literacy.

How to report each statistic

Report every inferential result as a package: the test statistic with its degrees of freedom, the exact p-value, and an effect size with, where possible, a confidence interval. A t-test reads t(98) = 2.31, p = .023, d = 0.46; a regression coefficient reads b = 2.13, 95% CI [1.40, 2.86], p < .001. Never report significance without an effect size estimate, because a p-value alone says nothing about how large or important the effect is. The full formatting conventions, including italics and decimal places, are in the APA formatting rules for results.

Using tables and figures well

Put detail in tables and figures and keep the prose for the story. A table of means and correlations, or a figure of an interaction, carries more than a paragraph of numbers and is easier to scan. The rule is to never duplicate: if a value is in a table, summarize the pattern in the text and refer to the table rather than restating every cell. Number tables and figures in the order they are mentioned, and make each one self-explanatory through its title and notes.

Keep interpretation out

The most common error is drifting into the discussion. "Scores were higher in the treatment group, t(98) = 2.31, p = .023" is a result; "this shows the intervention works because it raised motivation" is interpretation and does not belong here. State what you found, including the non-significant and unexpected results, and save the meaning for later. Reporting null findings honestly, rather than burying them, is part of what makes a results section credible.

A short results section example

A compact results section example reads: "Preliminary checks showed the assumptions of normality and homogeneity of variance were met. Mean exam scores were higher for the blended group (M = 74.2, SD = 8.1) than the lecture group (M = 68.5, SD = 9.0). A one-way ANOVA found a significant effect of teaching method, F(2, 87) = 6.41, p = .003, omega squared = .11; Tukey comparisons showed the blended group outperformed the lecture group (p = .002)." Notice the order, descriptives then test then follow-up, and that not a word interprets why. Match that discipline and your results section will read as the work of someone who knows their data.

How to write the results section of a research paper