📐 20 · Sample Size for a Master's Thesis
Research Question:
A master's student compares blended vs. traditional teaching (two groups). What is the minimum sample size to detect a medium effect (d = 0.5) with 80% power at α = .05?
| Parameter | Value | Why |
| Effect size d | 0.5 | Medium — Cohen (1988) convention |
| α | 0.05 | Standard significance level |
| Power (1−β) | 0.80 | 80% — minimum recommended |
-
1
Find z-values:
-
2
Calculate n per group:
-
3
Add 15% attrition buffer (recommended):
-
4
Total sample to recruit:
APA-7
An a priori power analysis (Cohen, 1988) indicated that 63 participants per group (total N = 126) were required to detect d = 0.50 with 80% power (α = .05, two-tailed). To account for attrition, 75 participants per group (total N = 150) were recruited.