Repeated Measures ANOVA in SPSS

Sometimes you want to know whether something changes – e.g., stress levels before, during, and after an intervention. Since you measure the same person multiple times, you need a repeated-measures ANOVA, also called a Repeated Measures ANOVA.

---

Example: Mindfulness Training and Stress

Research question:
Does the stress level of the participants change across three time points (before, during, and after the training)?

---

1. Entering data in SPSS

You need a dataset with one row per person and one column per measurement time point, e.g.:

ID	before	during	after
1	75	70	65
2	80	75	68
…	…	…	…

Tip: You need at least two measurement time points, but with SPSS you can also analyze 3 or more.

---

2. Conducting a Repeated Measures ANOVA

Analyze → General Linear Model → Repeated Measures ANOVA

a) Defining the factor

• Within-subject factor name: e.g., “Time”
• Number of levels: 3 (for “before,” “during,” and “after”)
  → Continue

b) Assigning the measurement time points

• Assign the time points to the variables in your dataset:
  • Time1 = before
  • Time2 = during
  • Time3 = after
    → Continue

c) Setting the options

• Models: Leave as “Full” (for interaction with time)
• Options:
  • “Descriptive statistics”
  • “Estimated means”
  • “Display effects” (optional)
  • You do not need “Test homogeneity of variances” – this is not required for repeated measures

→ Then click OK.

---

3. Interpreting the results

a) Mauchly’s test of sphericity

Definition (box):
Sphericity means that the variance of the differences between the time points is equal.
Violations lead to biased F-values.

• $p > 0{,}05$ → The assumption of sphericity is met
• $p < 0{,}05$ → Assumption violated → Consider the corrected tests:
  • Greenhouse-Geisser (conservative)
  • Huynh-Feldt (more liberal)

b) ANOVA table (within-subjects effects)

• If the time factor is significant ($p < 0{,}05$), the outcome variable has changed over time.

Example:

• F = 18.3, df = 2, p = 0.000 → significant

---

4. Post hoc tests (comparing time points)

SPSS does not offer a conventional post hoc selection here.
However, the output automatically includes the pairwise comparisons of the time points.

• You will find these in the table “Tests of Simple Main Effects” or “Pairwise Comparisons”
• Pay attention to the Bonferroni-corrected p-values!

---

5. Visualization (optional, recommended!)

Graphs → Means Plot → Simple

• Category axis: Time
• dependent variable: e.g., “before”, “during”, “after”
  → SPSS creates a line chart with means and error bars – ideal for presentations!

---

Summary: This is how to run a repeated-measures ANOVA in SPSS

Step	What you do
1.	Enter data in wide format (one row per person)
2.	Choose “Repeated Measures ANOVA” under “General Linear Model”
3.	Define the within-subjects factor (e.g., Time)
4.	Enable the options and run the analysis
5.	Check sphericity (Mauchly’s test) and interpret the results
6.	Create a chart and, if necessary, use pairwise comparisons

---

Q&A – Test Your Knowledge

Question 1: When do you need a Repeated Measures ANOVA?
Answer: When you measure the same people multiple times – e.g., before, during, and after an experiment.

Question 2: What does Mauchly’s test assess?
Answer: Whether sphericity – that is, equality of the variances of the differences – is met.

Question 3: What happens when sphericity is violated?
Answer: SPSS automatically calculates corrected F-tests (Greenhouse-Geisser, Huynh-Feldt).

Question 4: How do you need to structure the data?
Answer: Wide format – one row per person, columns = measurement time points.

Question 5: Can you perform post-hoc comparisons for repeated measures in SPSS?
Answer: Yes – SPSS automatically displays pairwise comparisons with adjusted p-values.