Quick answer: Your results chapter (often called Findings) has one job: report what you found, in the order your research questions ask about it, presented so a reader can verify every claim from your tables, figures or extracts — without interpreting it. Interpretation, comparison with the literature and explanation belong to the discussion chapter. For each research question or hypothesis: restate it briefly, present the relevant result with its table or figure, describe the result in words precisely, and move on. If you collected quantitative data, report descriptives before inferential tests; if qualitative, organise by theme with extracts as evidence. Write it, as a rule, in the past tense, and keep every “this suggests that” sentence for the discussion.
The results chapter is where dissertation students lose marks they have already earned. The data is collected, the analysis is done, the findings are genuinely there — and then the chapter either dumps output until the findings drown, or drifts into discussion until the marker cannot tell where reporting ends and arguing begins. Both failures are structural, not intellectual, and both are fixable with one principle: this chapter is evidence, presented in order, and nothing else. It sits in the middle of the chapter cluster — your methodology chapter promised an analysis, this chapter delivers its output, and your conclusion will eventually cash it in.
The boundary that organises everything: results versus discussion
Markers test one distinction relentlessly. Results answers what happened: the means, the test statistics, the themes, the pattern in the data. Discussion answers what it means: why the pattern appeared, how it compares with published findings, what it implies. Sentences like “this supports the theory that…” or “this may be because participants felt…” are discussion sentences wherever they appear, and in a results chapter they read as a writer who does not know which chapter they are in. The reverse error also costs marks: a discussion chapter that introduces results for the first time, because they were squeezed out of the results chapter, looks disorganised at exactly the point the argument should tighten.
A practical test while drafting: highlight every sentence in your results chapter that contains “because”, “suggests”, “indicates that” (used explanatorily) or a citation to another study. Most of those sentences belong one chapter later. Move them, and the results chapter usually clarifies itself.
Structure: follow your research questions, not your software
The chapter’s order should mirror the order of your research questions, hypotheses or objectives — not the order in which you ran analyses or conducted interviews. A reliable skeleton:
- A short opening that restates the chapter’s purpose, reminds the reader of the data and analysis (one short paragraph — the methodology chapter holds the detail), and states how the results are organised.
- One section per research question or hypothesis, each following the same rhythm: the question restated, the analysis or theme introduced, the evidence presented (table, figure or extracts), the finding described in words.
- A brief closing summary of the headline findings — two or three sentences per question, no new material — which hands over cleanly to the discussion.
Within a quantitative section, present descriptive statistics before inferential ones: the reader needs the shape of the data (means, standard deviations, frequencies, sample sizes) before a test result means anything. Report tests completely — the statistic, degrees of freedom, exact p-value and an effect size where your field expects one — because a bare “significant” is unverifiable and marked accordingly.
Tables and figures: the rules markers silently apply
Tables and figures are not decoration and not storage. Each one must earn its place by presenting data the text then uses, and each must be interpretable alone: a numbered caption that says what it shows, labelled columns and axes with units, and a note of sample size where it varies. Refer to every table and figure in the text (“Table 2 shows…”) and then say what matters in it — never “see Table 2” as the whole treatment, and never repeat the entire table in prose. Put the headline number in the sentence and let the table carry the detail. Raw software output — the unedited SPSS table with its default labels — should never appear in the chapter; rebuild the table in your document with the rows and columns your argument needs. If your analysis itself is the obstacle, our SPSS and data analysis help page covers results write-ups from your own data, and our dissertation chapter help service supports individual chapters like this one.
Qualitative results: themes with evidence
For interview, focus-group or observational projects, the findings chapter is organised by theme, not by participant and not by interview question. Each theme gets a section: what the theme is, the pattern across participants (how widespread, where it varied), and extracts chosen because they evidence the pattern — typical extracts for the main claim, and the divergent extract that complicates it, which markers reward because it shows the analysis is honest. Name participants by pseudonym or code consistently, keep extracts edited only for length and clarity (and say so in the methodology), and resist explaining the theme’s causes here; explanation is the discussion’s work.
Length and tense
There is no universal word count — findings chapters commonly run from around 1,500 words for a compact quantitative study to 4,000 or more for a rich qualitative one — and your supervisor’s guidance and your data’s complexity outrank any general figure. The test is completeness, not length: every research question answered with its evidence, nothing presented that is never used. Write in the past tense (“participants reported”, “the correlation was”), keep the reporting verbs neutral, and keep your own voice out of it beyond the organising sentences — your voice takes over, properly, in the discussion.
The mistakes that cost the most marks
- Software output pasted raw, with default variable names and irrelevant rows — it signals the writer did not curate their own evidence.
- Results without questions: analyses presented in run order, leaving the marker to work out which question each one answers.
- Significance theatre: p-values reported with no effect sizes, no descriptives and no sample sizes, so nothing can be judged.
- Interpretation smuggled in, chapter by chapter, until the discussion has nothing left to do but repeat it.
- Findings that vanish: a result presented here and never mentioned again — every presented result should be picked up in the discussion, or not presented at all.
If your results chapter is the one holding the project up — the data is fine but the chapter will not organise itself — that is precisely the shape of problem our dissertation chapter help exists for: one chapter, modelled properly, from your own analysis, which you then use as the pattern for the discussion. And if the analysis beneath it needs strengthening first, start with SPSS and data analysis help.
Results chapter questions
Should the results and discussion be one chapter or two?
Your handbook decides, and disciplines differ. Shorter undergraduate projects and some science degrees often combine them; Masters dissertations in the social sciences usually keep them separate. Either way, the order inside the chapter is fixed: findings first, clearly reported, and interpretation signposted as interpretation — never smuggled into the results as if it were data.
Do I report results that were not statistically significant?
Yes. A non-significant result is still a result, and leaving it out distorts the record your discussion then builds on. Report it honestly with the test statistic and exact p value, and let the discussion chapter explain what it might mean. Markers notice when a stated research question quietly produces no findings.
Can I paste tables straight from SPSS into the chapter?
No — raw software output is one of the fastest ways to lose presentation marks. Rebuild each table in a clean academic format: a numbered title, no vertical lines, abbreviations explained in a note underneath. If the full output matters for audit purposes, it belongs in an appendix, not the chapter.
How many tables and figures should a results chapter have?
There is no set number; the test is whether each one earns its place. Every table or figure must be introduced in the text, referred to by number, and say something the prose cannot say as economically. A table that repeats a paragraph word for word means one of the two should be cut.