Running a One-Way ANOVA (Analysis of Variance) in SPSS is one of the most common tasks for students analyzing survey data or experimental results. While t-tests compare two groups, ANOVA is used when you need to compare the means of three or more independent groups. This guide walks you through everything, from checking assumptions to writing up your APA results.
Our experienced tutors can clean your data, run the ANOVA, and write up the APA results for you quickly. Get a free quote.
1. When to Use ANOVA vs. T-Test
✅ Use ANOVA when...
- You have 3 or more groups to compare
- One categorical independent variable
- One continuous dependent variable
- Groups are independent (no repeated measures)
🔄 Use T-Test when...
- You have exactly 2 groups to compare
- Or comparing one group to a known value
- Or same group measured twice (paired)
You should run a One-Way ANOVA if you have:
- One continuous dependent variable (e.g., test scores, income, blood pressure)
- One categorical independent variable with three or more groups (e.g., diet type: Vegan, Vegetarian, Omnivore)
- Groups that are independent (no participant is in more than one group)
2. Assumptions to Check First
- Independence: Each participant appears in only one group.
- Normality: The dependent variable should be roughly normally distributed within each group. Test using Shapiro-Wilk in SPSS.
- Homogeneity of Variance: All groups should have similar variances. SPSS will run Levene's Test automatically when you check it. If Levene's p < .05, use Welch's ANOVA instead.
- No significant outliers: Check using boxplots in SPSS.
3. Running ANOVA in SPSS — Click-by-Click
Open the Analyze Menu
Go to the top menu bar and click Analyze → Compare Means → One-Way ANOVA...
Add Your Variables
Move your continuous variable (e.g., Test Scores) into the Dependent List box. Move your categorical grouping variable (e.g., Diet Type) into the Factor box.
Enable Descriptives & Levene's Test
Click Options, check Descriptive and Homogeneity of variance test. Click Continue.
Enable Post-Hoc Tests
Click Post Hoc..., check Tukey (the most common and robust). Click Continue.
Run the Analysis
Click OK. SPSS will generate the output in the Viewer window.
4. Interpreting the SPSS Output
The ANOVA Table
Find the main ANOVA table in your output. Look at the Sig. column (p-value):
| Sum of Squares | df | Mean Square | F | Sig. | |
|---|---|---|---|---|---|
| Between Groups | 245.6 | 2 | 122.8 | 4.56 | .019 |
| Within Groups | 726.4 | 27 | 26.9 | ||
| Total | 972.0 | 29 |
- If Sig. < .05 → statistically significant result. At least one group mean is different.
- If Sig. ≥ .05 → no significant difference between groups. You stop here.
Levene's Test of Homogeneity
Check this before reading the main ANOVA table. If Levene's Sig. > .05, your variances are equal and you proceed normally. If Sig. < .05, report Welch's ANOVA results instead.
5. Post-Hoc Tests (Tukey HSD)
If your ANOVA was significant (p < .05), the next question is: Which specific groups differ from each other? That's what the Post-Hoc table tells you.
Look at the Multiple Comparisons (Tukey HSD) table. Find rows where the Sig. column is < .05 — those specific pairs of groups are significantly different from each other.
Tip: Focus on the "Mean Difference" column to see which group scored higher or lower, and by how much.
6. Writing the APA Results
Use this template to write up your ANOVA results in APA format:
"A one-way ANOVA was conducted to compare [dependent variable] across [groups]. There was a statistically significant difference in [DV] between at least two groups, F(df_between, df_within) = __, p = .__. Tukey's HSD post-hoc test revealed that [Group A] scored significantly higher than [Group B] (p = .__). There was no significant difference between [Group B] and [Group C] (p = .__)"
7. Full Worked Example
Research Question: Does diet type (Vegan, Vegetarian, Omnivore) affect exam performance?
"A one-way ANOVA revealed that there was a statistically significant difference in test scores between at least two diet groups, F(2, 27) = 4.56, p = .019. Tukey's HSD Test for multiple comparisons found that the mean value of test scores was significantly different between the Vegan group (M = 82.4, SD = 4.9) and the Omnivore group (M = 74.1, SD = 5.2), p = .015. There was no statistically significant difference between the Vegetarian and Vegan groups (p = .124)."