Updated citation information following publication in JOSS.
Added Japanese documentation to the package website.
pwranova 1.1.4 (2026-06-19)
Fixed an issue in pwranova() where epsilon adjustments
were not fully incorporated into noncentrality parameter
calculations.
Expanded validation tests, including additional comparisons with
G*Power.
pwranova 1.1.3 (2026-06-18)
Added conventional effect-size benchmarks based on Cohen (1988) in
pwranova(), pwrttest(), and
pwrcortest() documentation.
Clarified the relationship between eta-squared and partial eta
squared in pwranova() documentation.
pwranova 1.1.2 (2026-06-18)
Clarified effect-size definitions and noncentrality parameter
calculations in pwrttest() documentation.
pwranova 1.1.1 (2026-03-05)
Strengthened input validation across functions. Added checks for
finite values, parameter bounds, and boundary cases in effect-size
conversion utilities and power analysis functions.
Clarified documentation for pwrcortest() and
pwrttest() to state that signed effect-size inputs
(rho, delta) are converted to their absolute
values.
pwranova 1.1.0 (2026-03-05)
Added ncp_scale argument to pwrcortest()
for alternative scaling of the noncentrality parameter in correlation
power analysis.
pwranova 1.0.3 (2026-01-19)
Fixed a bug in the denominator df for paired/repeated-measures
contrast tests.
pwranova 1.0.2 (2025-10-01)
Minor adjustments in DESCRIPTION and documentation links.
Small fixes in preparation for CRAN submission.
pwranova 1.0.1 (2025-09-30)
Removed obsolete comments in code.
Updated citation information.
Minor updates to DESCRIPTION.
pwranova 1.0.0 (2025-09-28)
First stable release of the package.
Provides power analysis functions for:
Between-, within-, and mixed-factor ANOVA designs (main effects and
interactions).
Planned contrasts with user-defined weights.
t-tests (one-sample, paired, and two-sample).
Pearson correlations (using either the t-distribution or Fisher’s
z-transformation).
Supports solving for:
Statistical power
Required total sample size
Significance level (alpha)
Minimal detectable effect size (Cohen’s f, partial eta squared, or
Cohen’s d-type indices).
Validated against G*Power for supported ANOVA designs, t-tests,
and correlation tests.