In jasonwgriffin/metapower: Power Analysis for Meta-Analysis

## update logo
library(badger)
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

metapoweR

r badge_cran_download("metapower", "grand-total", "blue")

The primary goal of metapower is to compute statistical power for meta-analyses. Currently, metapower has the following functionality:

Computation of statistical power for:

1. Summary main effects sizes
2. Test of homogeneity for between-group variance (for Random-effects models).
3. Test of homogeneity for within-study variance
4. Subgroup Analyses
5. Moderator Analysis

metapower can currently handle the following designs and effect sizes:

1. Standardized mean difference: Cohen's d
2. Correlation between two continuous variables: Correlation Coefficient (via Fisher's r-to-z transformation)
3. Probability of Success/Failure: Odds Ratio

Tutorial Paper

For detailed information about how to use metapower, see Calculating statistical power for meta-analysis using metapower

Installation

You can install the released version of metapower from CRAN with:

install.packages("metapower")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("jasonwgriffin/metapower")

Shiny Application

Check out the simple and easy to use shiny application

Example

library(metapower)
my_power <- mpower(effect_size = .3, study_size = 20, k = 10, i2 = .50, es_type = "d")
print(my_power)
plot_mpower(my_power)

See Vignette "Using metapower" for more information..

References

All mathematical calculations are derived from @hedges2004, @bornstein2009,@pigott2012, @jackson2017.

Citation

Griffin, J. W. (2021). Calculating statistical power for meta-analysis using metapower. The Quantitative Methods for Psychology, 17(1), 24–39. doi:10.20982/tqmp.17.1.p024

Issues

If you encounter a clear bug, please file a minimal reproducible example on github.

jasonwgriffin/metapower documentation built on April 30, 2021, 10:09 a.m.