How to Report ANOVA Results in APA 7 (With Examples)
By Dr. David C., PhD in Biostatistics. Reviewed by Dr. Amanda R.. 2026-05-20.
Get expert help with this chapter | University guides | Dissertation writing services
Key takeaways
- Report F-statistic with df, p-value, and partial eta squared every time.
- Use 'p < .001', never 'p = .000'.
- Always run and report assumption checks (normality, homogeneity, sphericity).
- Effect size matters more than significance in modern APA 7 reporting.
- Provide the syntax or script in an appendix for reproducibility.
The APA 7 ANOVA reporting formula
Every ANOVA result in APA 7 includes: the type of ANOVA, the F-statistic with degrees of freedom, the p-value, and an effect size (typically partial eta squared for factorial designs).
Template: 'A one-way ANOVA showed a significant effect of [IV] on [DV], F(df1, df2) = X.XX, p < .XXX, partial eta squared = .XX.'
One-way ANOVA example
A one-way ANOVA was conducted to examine the effect of teaching method (lecture, blended, flipped) on final exam scores (N = 120). The analysis revealed a significant effect of teaching method, F(2, 117) = 8.45, p < .001, partial eta squared = .13.
Post-hoc comparisons using Tukey HSD indicated that the flipped classroom group (M = 84.2, SD = 6.1) scored significantly higher than the lecture group (M = 76.8, SD = 7.3), p < .001. The blended group (M = 80.5, SD = 6.9) did not differ significantly from either other group.
Two-way (factorial) ANOVA example
A 2 (gender: male, female) x 3 (teaching method: lecture, blended, flipped) between-subjects ANOVA was conducted on final exam scores. There was a significant main effect of teaching method, F(2, 114) = 7.92, p = .001, partial eta squared = .12. The main effect of gender was not significant, F(1, 114) = 1.45, p = .231. The interaction between gender and teaching method was not significant, F(2, 114) = 0.78, p = .461.
Repeated-measures ANOVA example
A one-way repeated-measures ANOVA was conducted to examine changes in anxiety scores across three time points (pre, post, follow-up; N = 45). Mauchly's test indicated that the assumption of sphericity had been violated, chi-square(2) = 8.34, p = .015, therefore Greenhouse-Geisser correction was applied (epsilon = .82). There was a significant effect of time, F(1.64, 72.16) = 12.47, p < .001, partial eta squared = .22.
Effect size thresholds (Cohen, 1988)
Partial eta squared: .01 small, .06 medium, .14 large.
Always report effect size in APA 7; significance alone is insufficient.
For mixed designs, report partial eta squared for each effect and omega squared if your committee prefers it.
APA 7 table template for ANOVA
Use Table 1 (or appropriate number), title in italics, source-SS-df-MS-F-p-partial eta squared columns. Notes below: 'Note. N = 120. *p < .05. **p < .01. ***p < .001.'
Common reporting mistakes
Forgetting to report effect size.
Using p = .000 (APA 7 requires p < .001).
Reporting F without degrees of freedom.
Skipping assumption checks (normality, homogeneity of variance, sphericity).
Reporting post-hoc tests without specifying the correction method.
How DissertlyPro handles ANOVA reporting
Our PhD biostatisticians run the analysis, produce APA 7-compliant tables, write the narrative in your voice, and supply the SPSS or R syntax for your appendix. Every ANOVA report includes assumption checks and effect sizes.
Need a statistician on your committee timeline? Book a free 20-minute call with a PhD biostatistician on the DissertlyPro team.
FAQ
Do I report eta squared or partial eta squared?
Partial eta squared is APA 7 standard for ANOVA. Eta squared is acceptable in some fields; check your discipline norms.
What if assumptions are violated?
Apply the appropriate correction (Greenhouse-Geisser for sphericity, Welch's for unequal variances) or switch to a non-parametric alternative (Kruskal-Wallis). Report the violation and the action you took.
Should I include confidence intervals?
Yes when reporting means. APA 7 encourages 95 percent confidence intervals alongside means in tables.
How do I report a non-significant ANOVA?
Same format: F(df1, df2) = X.XX, p = .XXX, partial eta squared = .XX. Do not say 'no effect found'; say 'no significant effect detected'.