A simple ratio-based approach for power and sample size determination for 2-group comparison using Rasch models

Table 3 Comparison of the required parameters and the results obtained using the linear regression model and the Raschpower method on the NHP data

σ²: variance of the latent trait; J: number of items; γ: difference between the mean values of the latent trait in the two groups (group effect); N_g: sample size per group providing a 1-β = 90% power with the classical formula and a 1-β_R = 80% power with Raschpower; δ: vector of the items parameters (δ₁, δ₂, δ₃, δ₄, δ₅, δ₆, δ₇, δ₈); Ra: ratio, N_R: sample size per group providing a 1-β ≈ 90% power for Rasch analysis with the linear regression model and Raschpower, /: not required.

ISSN: 1471-2288