SSDbain

SSDbain computes required sample sizes for Bayesian hypothesis testing of independent means and ANOVA models using Bayes factors and the Approximate Adjusted Fractional Bayes Factor (AAFBF).

Key Features:

• Bayesian t-test and Welch's Test: SSDttest computes the required sample size per group for Bayesian t-tests and Bayesian Welch's tests and supports hypotheses H0: μ1 = μ2, H1: μ1 ≠ μ2, and H2: μ1 > μ2.
• Bayesian ANOVA: SSDANOVA and SSDANOVA_robust compute sample sizes for Bayesian ANOVA, Welch's ANOVA, and robust ANOVA and support constrained hypotheses such as H1: m1 = m2 = m3, H2: m1 > m2 > m3, and Hu (no constraints).
• Approximate Adjusted Fractional Bayes Factor (AAFBF): Implements AAFBF to ensure the probability of obtaining a Bayes factor exceeding a specified threshold is at least η under either the null or alternative hypothesis.
• Bayes factor evaluation: Employs Bayes factors for informative hypothesis evaluation across the implemented statistical models.

Scientific Applications:

• Experimental design for hypothesis testing: Determines sample sizes for studies comparing independent means and multiple-group ANOVA to support rigorous experiment design.
• Psychology and related fields: Applicable to psychological researchers and other scientists who require precise Bayesian sample size calculations for t-tests and ANOVA models.

Methodology:

Leverages Bayes factors for informative hypothesis evaluation and integrates the Approximate Adjusted Fractional Bayes Factor (AAFBF) to determine sample sizes that achieve a specified probability (≥ η) of exceeding a Bayes factor threshold.