psychmeta (version 2.7.0)

create_ad_group: Generate an artifact distribution object for a dichotomous grouping variable.

Description

This function generates artifact-distribution objects containing either interactive or Taylor series artifact distributions for dichotomous group-membership variables. Use this to create objects that can be supplied to the ma_r_ad and ma_d_ad functions to apply psychometric corrections to barebones meta-analysis objects via artifact distribution methods.

Allows consolidation of observed and estimated artifact information by cross-correcting artifact distributions and forming weighted artifact summaries.

Usage

create_ad_group(
  ad_type = c("tsa", "int"),
  rGg = NULL,
  n_rGg = NULL,
  wt_rGg = n_rGg,
  pi = NULL,
  pa = NULL,
  n_pi = NULL,
  n_pa = NULL,
  wt_p = n_pi,
  mean_rGg = NULL,
  var_rGg = NULL,
  k_rGg = NULL,
  mean_n_rGg = NULL,
  var_unbiased = TRUE,
  ...
)

Value

Artifact distribution object (matrix of artifact-distribution means and variances) for use in artifact-distribution meta-analyses.

Arguments

ad_type

Type of artifact distribution to be computed: Either "tsa" for Taylor series approximation or "int" for interactive.

rGg

Vector of incumbent reliability estimates.

n_rGg

Vector of sample sizes associated with the elements of rGg.

wt_rGg

Vector of weights associated with the elements of rGg (by default, sample sizes will be used as weights if provided).

pi

Vector of incumbent/sample proportions of members in one of the two groups being compared (one or both of pi/pa can be vectors - if both are vectors, they must be of equal length).

pa

Vector of applicant/population proportions of members in one of the two groups being compared (one or both of pi/pa can be vectors - if both are vectors, they must be of equal length).

n_pi

Vector of sample sizes associated with the elements in pi.

n_pa

Vector of sample sizes associated with the elements in pa.

wt_p

Vector of weights associated with the collective element pairs in pi and pa.

mean_rGg

Vector that can be used to supply the means of externally computed distributions of correlations between observed and latent group membership.

var_rGg

Vector that can be used to supply the variances of externally computed distributions of correlations between observed and latent group membership.

k_rGg

Vector that can be used to supply the number of studies included in externally computed distributions of correlations between observed and latent group membership.

mean_n_rGg

Vector that can be used to supply the mean sample sizes of externally computed distributions of correlations between observed and latent group membership.

var_unbiased

Logical scalar determining whether variance should be unbiased (TRUE) or maximum-likelihood (FALSE).

...

Further arguments.

Examples

## Example artifact distribution for a dichotomous grouping variable:
create_ad_group(rGg = c(.8, .9, .95), n_rGg = c(100, 200, 250),
                mean_rGg = .9, var_rGg = .05,
                k_rGg = 5, mean_n_rGg = 100,
                pi = c(.6, .55, .3), pa = .5, n_pi = c(100, 200, 250), n_pa = c(300, 300, 300),
                var_unbiased = TRUE)
                
create_ad_group(ad_type = "int", rGg = c(.8, .9, .95), n_rGg = c(100, 200, 250),
                mean_rGg = .9, var_rGg = .05,
                k_rGg = 5, mean_n_rGg = 100,
                pi = c(.6, .55, .3), pa = .5, n_pi = c(100, 200, 250), n_pa = c(300, 300, 300),
                var_unbiased = TRUE)