Skip to main content

Table 3 Power analysis

From: Evaluating real-time internet therapy and online self-help for problematic alcohol consumers: a three-arm RCT protocol

F tests – ANOVA: Fixed effects, omnibus, one-way

Analysis:

A priori: Compute required sample size

 

Input:

Effect size f

0.25

 

α error probability

0.05

 

Power (1-β error probability)

0.80

 

Number of groups

3

Output:

Non-centrality parameter λ

9.938

 

Critical F

3.054

 

Numerator df

2

 

Denominator df

156

 

Total sample size

159

 

Actual power

0.805