propodds

Performs power and sample size calculation for a comparison of two groups on
an ordinal categorical response variable.
Assumes that response probabilities follow the proportional
odds assumption. Can solve for power, n1, n.ratio and alpha.

Power and sample size calculations for a variety of study designs and
outcomes. Methods include t tests, ANOVA (including tests for interactions,
simple effects and contrasts), proportions, categorical data (chi-square tests
and proportional odds), linear, logistic and Poisson regression, alternative
and coprimary endpoints, power for confidence intervals, correlation
coefficient tests, cluster randomized trials, individually randomized group
treatment trials, multisite trials, treatment-by-covariate interaction effects
and nonparametric tests of location. Utilities are provided for
computing various effect sizes. Companion package to the book
"Power and Sample Size in R", Crespi (2025, ISBN:9781138591622).
Further resources available at <https://powerandsamplesize.org/>.

Catherine M. Crespi

powertools

Power and Sample Size Tools

Zichen Liu

Kristen M. McGreevy

propodds function

<dl><dt>pC</dt>
<dd>Vector of response probabilities in control group (group 1). Must sum to 1. Categories are ordered from best to worst.</dd>
<dt>OR</dt>
<dd>Odds ratio when the alternative is true. Must be greater than 1.</dd>
<dt>n1</dt>
<dd>Sample size for group 1 (control group).</dd>
<dt>n.ratio</dt>
<dd>The ratio n2/n1 between the sample sizes of two groups; defaults to 1 (equal group sizes).</dd>
<dt>alpha</dt>
<dd>The significance level (type 1 error rate); defaults to 0.05.</dd>
<dt>power</dt>
<dd>The specified level of power; defaults to 0.8.</dd>
<dt>v</dt>
<dd>Either TRUE for verbose output or FALSE (default) to output computed argument only.</dd></dl>

Arguments

Power calculations for ordinal categorical variable under proportional odds assumption — propodds

<dl>

<dt>pC</dt>
<dd>Vector of response probabilities in control group (group 1). Must sum to 1. Categories are ordered from best to worst.</dd>


<dt>OR</dt>
<dd>Odds ratio when the alternative is true. Must be greater than 1.</dd>


<dt>n1</dt>
<dd>Sample size for group 1 (control group).</dd>


<dt>n.ratio</dt>
<dd>The ratio n2/n1 between the sample sizes of two groups; defaults to 1 (equal group sizes).</dd>


<dt>alpha</dt>
<dd>The significance level (type 1 error rate); defaults to 0.05.</dd>


<dt>power</dt>
<dd>The specified level of power; defaults to 0.8.</dd>


<dt>v</dt>
<dd>Either TRUE for verbose output or FALSE (default) to output computed argument only.</dd>

</dl>

propodds: Power calculations for ordinal categorical variable under proportional odds assumption

Description

Usage

Value

Arguments

Details

Examples