Statoria Brand LogoStatoria
APA reporting

How to Write the Statistics Results Section of Your Thesis

5 min read

The results section is where most students lose marks without understanding why. The analysis is correct - but the reporting is incomplete, the format violates APA rules, or the narrative and numbers contradict each other. This guide shows you exactly what to include, in what order, and what not to write - with copy-paste templates for every common thesis test.

Free sample chapter

Data Analysis From Survey to Results

Step-by-step guidance for choosing the right test, running it, and writing up APA results - in plain language, not theory. Get the free sample chapter when you join the waitlist.

Key takeaways

  • No leading zero: write p = .032, not p = 0.032 - applies to all p-values, correlations, and effect sizes.
  • Exact p-values: report p = .032, not p < .05 - use p < .001 only when software shows .000.
  • Effect size is mandatory: d, η², r, or R² must accompany every test statistic - omitting it is an APA violation.
  • Report all tests: non-significant results must be reported - selectively omitting them is a form of research misconduct.
  • Results ≠ Discussion: the results section reports numbers only - all interpretation belongs in the Discussion chapter.

What the Results Section Should (and Should Not) Contain

The results section reports your findings in order. It should not interpret, discuss, or explain findings - that belongs in the discussion chapter.

  • Include in your results section:
  • Descriptive statistics (M, SD, n) for each group or variable
  • Assumption check results (Shapiro-Wilk, Levene's)
  • Test statistics and degrees of freedom
  • Exact p-values (p = .032, not p < .05 - use p < .001 only when below .001)
  • Effect size for every test you report
  • Post-hoc test results when applicable (ANOVA with 3+ groups)
  • Do not include:
  • Conclusions or interpretations ('This shows that...')
  • References to the literature
  • Recommendations or suggestions
  • Raw data tables (unless specifically required)

The Standard Structure for a Results Chapter

Most thesis results chapters follow this order.

SectionWhat to Write
Sample descriptionN, demographics (M age, gender split), any exclusions and why
Descriptive statisticsM and SD for all key variables - reference a table
Assumption checksShapiro-Wilk and Levene's results for each test
Main analysesTest statistic, df, p, effect size - one subsection per hypothesis
Post-hoc testsPairwise comparisons when ANOVA is significant

APA Format Rules for Statistical Reporting

These rules apply to every test you report - no exceptions.

  • Italicise all statistical symbols: t, F, r, χ², M, SD, p, η², d
  • No leading zero for values bounded between 0 and 1: p = .032 (not p = 0.032); r = .54 (not r = 0.54)
  • Report exact p-values, not p < .05 - use p < .001 only when software displays .000
  • Degrees of freedom in parentheses: t(48), F(2, 97), χ²(3)
  • Always report effect size alongside p-value: d, η², r, R²
  • Report means and SDs as: M = 74.2, SD = 8.3
⚠️

The three most common APA errors: (1) p = 0.032 instead of p = .032, (2) omitting effect size, (3) writing 'p < .05' instead of the exact value.

Copy-Paste Templates for Every Common Test

  • Independent t-test (significant):
  • "An independent samples t-test revealed a significant difference between [Group A] (M = 74.2, SD = 8.1) and [Group B] (M = 68.5, SD = 9.3), t(48) = 2.31, p = .025, d = 0.65."
  • One-way ANOVA:
  • "A one-way ANOVA revealed a significant effect of [factor] on [outcome], F(2, 87) = 8.42, p < .001, η² = .16. Post-hoc Tukey HSD indicated that [Group A] scored significantly higher than [Group C] (p = .003)."
  • Mann-Whitney U (non-parametric):
  • "A Mann-Whitney U test indicated a significant difference between groups, U = 234, z = −3.41, p = .001, r = .48."
  • Pearson correlation:
  • "Pearson correlation revealed a significant positive relationship between [X] and [Y], r(98) = .54, p < .001."
  • Linear regression:
  • "[Predictor] significantly predicted [outcome], B = 0.54, β = .45, t(98) = 5.12, p < .001. The model explained 20% of variance in [outcome], R² = .20, F(1, 98) = 26.2, p < .001."

Common Mistakes That Trigger Revision Requests

These are the patterns supervisors flag most frequently during thesis reviews.

MistakeWhat to Write Instead
p = 0.032 (leading zero)p = .032
p < .05 (non-exact value)p = .032 (exact); p < .001 only when below .001
'The result was significant' (no numbers)t(48) = 2.31, p = .025, d = 0.65
Missing effect sizeAlways add d, η², r, or R² for every test
'The test showed a big difference'Report M, SD, and effect size - let the numbers speak
Interpreting results in the results chapterMove all interpretation to the Discussion
Omitting non-significant resultsReport all tests run, regardless of p-value

Frequently asked questions

Should I report non-significant results in my thesis?

▾
Yes. Every test you run must be reported - whether significant or not. Omitting non-significant results is selective reporting and is considered a form of research misconduct. Non-significant results are scientifically meaningful and belong in your results chapter.

How do I report p = .000 as shown in SPSS output?

▾
Write p < .001. SPSS displays .000 when the p-value is below .001 because it rounds to three decimal places. Never write p = .000 in your thesis - the p-value is never actually zero.

Can I use a table for descriptive statistics instead of writing them out?

▾
Yes - and for four or more variables, a table is preferred over in-text reporting. APA requires that tables supplement, not duplicate, the text. Do not repeat every number from the table in your paragraph. Instead, refer to it: 'Descriptive statistics for all variables are presented in Table 1.'

What effect size measure should I report for each test?

▾
t-test → Cohen's d. ANOVA → eta-squared (η²) or partial eta-squared (ηp²). Mann-Whitney U → r (rank-biserial). Chi-square → Cramér's V. Pearson/Spearman → r. Regression → R² for the model and β for each predictor. Most statistics software calculates these automatically - check the effect size or additional statistics options in your software.

How long should the results section be?

▾
Typically 800–1,500 words for a bachelor thesis and 1,500–3,000 words for a master thesis. Each research question or hypothesis should have its own subsection. Length depends on the number of tests you ran - but prioritise clarity over length. Every sentence in the results chapter should contain numbers.

Further reading

Free tool

Not sure which statistical test to use?

Answer 5 quick questions about your research design and get the right test - with an explanation of why - in under two minutes.

Statoria Team

Statistics educators & software developers

We build Statoria to help bachelor and master students get through their thesis data analysis without stress. Our guides are written by researchers with experience in social science statistics and student supervision.

Related guides

APA reporting
APA reporting
APA Statistics Reporting: Copy-Paste Templates for Every Test in Your Thesis
Apr 2026 · 3 min read
Which Statistical Test to Use for Your Thesis: A Complete Decision Guide
Test selection
Which Statistical Test to Use for Your Thesis: A Complete Decision Guide
Feb 2026 · 3 min read
Thesis Data Analysis: The 5 Critical Steps Students Skip (With Checklist)
Data analysis
Thesis Data Analysis: The 5 Critical Steps Students Skip (With Checklist)
Mar 2026 · 3 min read
The 5 Thesis Statistics Mistakes That Cost Students Their Grade (And How to Catch Them Before Your Defense)
Common mistakes
The 5 Thesis Statistics Mistakes That Cost Students Their Grade (And How to Catch Them Before Your Defense)
Mar 2026 · 3 min read
Back to all tutorials