A one-way ANOVA tests whether three or more group means differ. The output looks busier than a t-test, so the write-up trips people up: which numbers go in the sentence, how to report the F statistic, what effect size to use, and how to handle post hoc tests. This guide gives you the exact APA 7 format, a worked example you can copy, and the mistakes reviewers catch.
What you need before you write a single word
An APA 7 write-up of a one-way ANOVA needs these values from your output:
- The mean (M) and standard deviation (SD) for each group.
- The F value.
- Two degrees of freedom: between-groups (df1 = number of groups minus 1) and within-groups (df2 = total N minus number of groups).
- The exact p value.
- An effect size, usually eta squared (eta-squared) or omega squared. APA 7 expects an effect size for the omnibus test.
- Post hoc comparisons (for example Tukey HSD) if the omnibus test is significant and you need to say which groups differ.
The APA 7 format template
Report the omnibus result in running text using this pattern:
A one-way ANOVA showed a significant effect of [independent variable] on [dependent variable], F(df1, df2) = X.XX, p = .XXX, eta-squared = .XX.
Formatting rules reviewers actually check:
- Italicize the statistical symbols: M, SD, F, p.
- Report both degrees of freedom in parentheses, between-groups first: F(2, 87).
- No leading zero on p or on eta-squared (neither can exceed 1).
- Round F to two decimals. Report p to two or three decimals, and if it is below .001, write p < .001.
- Give the group means and SDs somewhere, usually in a table, not only the F statistic.
A worked example
Suppose you compared exam scores across three teaching methods, with 30 students per group.
- Lecture (n = 30): M = 71.20, SD = 9.40
- Flipped classroom (n = 30): M = 78.90, SD = 8.10
- Self-paced (n = 30): M = 74.30, SD = 8.80
- Omnibus result: F(2, 87) = 5.83, p = .004, eta-squared = .12
Written up in APA 7, that becomes:
A one-way ANOVA showed a significant effect of teaching method on exam scores, F(2, 87) = 5.83, p = .004, eta-squared = .12. Tukey HSD post hoc comparisons indicated that the flipped classroom group (M = 78.90, SD = 8.10) scored significantly higher than the lecture group (M = 71.20, SD = 9.40), p = .003. No other pairwise comparison was significant.
The omnibus sentence tells the reader there is a difference somewhere; the post hoc sentence tells them where.
The APA 7 table (recommended for ANOVA)
Because there are three or more groups, a descriptives table earns its place:
| Group | n | M | SD |
|---|---|---|---|
| Lecture | 30 | 71.20 | 9.40 |
| Flipped classroom | 30 | 78.90 | 8.10 |
| Self-paced | 30 | 74.30 | 8.80 |
Note. Exam scores range from 0 to 100. The effect of teaching method was significant, F(2, 87) = 5.83, p = .004, eta-squared = .12.
Mistakes reviewers catch
- Reporting only one degree of freedom. ANOVA needs both, between and within: F(2, 87).
- Stopping at the omnibus test. A significant ANOVA says the means are not all equal; it does not say which groups differ. Run and report post hoc tests (or planned contrasts).
- No effect size. Report eta-squared or omega squared for the omnibus effect. Omega squared is less biased in small samples.
- Writing p = .000. Report p < .001.
- A leading zero on p or eta-squared. APA drops it for values that cannot exceed 1.
- Running many t-tests instead of an ANOVA. Multiple pairwise t-tests inflate the false-positive rate; that is what the ANOVA plus post hoc correction is for.
Before you report: did the test's assumptions hold?
A one-way ANOVA assumes:
- Independence of observations.
- Approximate normality of the dependent variable within each group.
- Homogeneity of variance across groups (checked with Levene's test). If Levene's is significant, report Welch's ANOVA and its adjusted degrees of freedom instead.
If normality is badly violated, the Kruskal-Wallis test is the usual non-parametric alternative.
Let KyroStat do the write-up for you
Formatting an ANOVA by hand, with its two degrees of freedom, effect size, and post hoc table, is where errors creep in. KyroStat runs the ANOVA on your data, checks the assumptions above, runs the post hoc comparisons, and hands you the finished APA 7 sentences, a publication-ready table, the plot, and the underlying Python or R code. Upload your spreadsheet, and the report is done in seconds.
Frequently asked questions
Which effect size should I report for ANOVA? Eta squared or, preferably, omega squared, which is less biased in small samples. Report it without a leading zero, for example eta-squared = .12.
Do I always need post hoc tests? Only if the omnibus F is significant and you want to say which specific groups differ. If it is not significant, you stop there.
What if Levene's test is significant? Use Welch's ANOVA and report its adjusted degrees of freedom, which are often decimals.
How do I report both degrees of freedom? Between-groups first, then within-groups, in parentheses after F: F(2, 87).