In a-lenz/presize: Precision Based Sample Size Calculation

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

presize

Warning: R-package in development.

The goal of presize is to provide functions for precision based sample size calculation. For a given sample size, the functions will return the precision (half the width of the confidence interval), and vice versa.

Installation

You can install presize from github with:

# install.packages("devtools")
devtools::install_github("CTU-Bern/presize")

Overview

presize will provide functions for

• descriptive statistics

  • mean (prec_mean)
  • proportion (prec_prop)
  • rate (prec_rate)

• absolute and relative differences

  • mean difference (prec_meandiff)
  • risk difference (prec_riskdiff)
  • odds ration (prec_or)
  • risk ratio (prec_riskratio)
  • rate ratio
  • hazard ratio

• correlation measures

  • correlation coefficient (prec_cor)
  • Cohens kappa
  • ICC (prec_icc)
  • limit of agreement from Bland Altman plot

• diagnostic measures

  • sens
  • spec
  • positive LR
  • negative LR
  • AUC

Example

This is a basic example which shows you how to solve a common problem:

library(presize)

# calculate sample size for a proportion of 0.2, or 0.4 with a precision of 0.2
prec_prop(p = c(.2, .4), n = 10, method = "wilson")

a-lenz/presize documentation built on May 17, 2019, 7:44 a.m.