escompute: escompute
In puniform: Meta-Analysis Methods Correcting for Publication Bias

View source: R/escompute.R

escompute

R Documentation

escompute

Description

Function that computes Hedges' g and its sampling variance for an one-sample mean and a two-independent means, Fisher's r-to-z transformed correlation coefficient and its sampling variance for a raw correlation coefficient and computes a p-value as in the primary studies was done.

Usage

escompute(
  mi,
  ri,
  ni,
  sdi,
  m1i,
  m2i,
  n1i,
  n2i,
  sd1i,
  sd2i,
  tobs,
  yi,
  vi,
  alpha,
  side,
  measure
)

Arguments

`mi`	A vector of group means for one-sample mean
`ri`	A vector of raw correlation coefficients
`ni`	A vector of sample sizes for one-sample mean
`sdi`	A vector of standard deviations for one-sample mean
`m1i`	A vector of means in group 1 for two-independent means
`m2i`	A vector of means in group 2 for two-independent means
`n1i`	A vector of sample sizes in group 1 for two-independent means
`n2i`	A vector of sample sizes in group 2 for two-independent means
`sd1i`	A vector of standard deviations in group 1 for two-independent means
`sd2i`	A vector of standard deviations in group 2 for two-independent means
`tobs`	A vector of t-values
`yi`	A vector of standardized effect sizes
`vi`	A vector of sampling variances belonging to the standardized effect sizes (`yi`)
`alpha`	A numerical value specifying the alpha level as used in primary studies
`side`	A character indicating the direction of the tested hypothesis in the primary studies (either "`right`" or "`left`")
`measure`	A character indicating what kind of effect size should be computed (Hedges' g or Fisher's r-to-z transformed correlation coefficients) and which arguments are used as input ("`M`", "`MT`", "`MD`", "`MDT`", or "`COR`"). See Details below.

Details

The measure argument has to be used to specify the desired effect size and what input parameters are used. There are six options:

"M" for one-sample mean with mi, ni, sdi, alpha, and side as input parameters
"MT" for one-sample mean with tobs, ni, alpha, and side as input parameters
"MD" for two-sample mean with m1i, m2i, n1i, n2i, sd1i, sd2i, alpha, and side as input parameters
"MDT" for two-sample mean with tobs, n1i, n2i, alpha, and side as input parameters
"COR" for raw correlation coefficients with ri, ni, alpha, and side as input parameters
"SPE" for user-specified standardized effect sizes and sampling variances with yi, vi, alpha, and side as input parameters

Value

Function returns a data frame with standardized effect sizes (yi), variances of these standardized effect sizes (vi), z-values (zval), p-values as computed in primary studies (pval), and critical z-values (zcv).