| Popis: |
This study presents the theoretical foundations and computational approaches to statistical power analysis. Ten hypothesis tests and their derivatives are reviewed, including the test of one proportion against a constant, difference between two proportions, one mean against a constant, difference between two means (independent and matched samples), one correlation against a constant, difference between two correlations, R-squared deviation from zero in linear regression, difference between two R-squared values in hierarchical linear regression, analyses of variance/covariance (one-way, two-way and three-way ANOVA or ANCOVA) for comparing means of two or more groups, and repeated measures analysis of variance (one-way repeated measures ANOVA). The concept of statistical power and sample size calculations for these tests are consolidated with practical examples. The hypothesis tests of non-inferiority, superiority, and equivalence, which are widely used in medical and pharmaceutical research are also introduced, and their applications are demonstrated using examples from behavioral and educational research. Calculations were performed with the pwrss R package (https://pwrss.shinyapps.io/lang-en/). |
