psychmeta (version 2.3.4)

var_error_rel: Estimate the error variance of reliability estimates

Description

Estimate the error variance of reliability estimates

Usage

var_error_rel(rel, n, rel_type = "alpha", k_items = NULL)

Arguments

rel

Vector of reliability estimates.

n

Vector of sample sizes.

rel_type

Character vector indicating the type(s) of reliabilities being analyzed. See documentation for ma_r for a full list of acceptable reliability types. NOTE: Currently, only alpha has its own dedicated error-variance estimate; the error variance of other reliability types is estimated using the generic definition of reliability as the squared correlation between observed scores and true scores.

k_items

Optional numeric vector indicating the number of items in each scale for which reliabilities are being analyzed.

Value

A vector of sampling-error variances.

Details

The sampling variance of a reliability coefficient is:

$$var_{e}=\frac{4r_{XX}(1-r_{XX})^{2}}{n-1}$$

For the equation to estimate the variance of coefficient alpha, see Duhachek and Iacobucci (2004).

References

Dahlke, J. A., & Wiernik, B. M. (2017). One of these artifacts is not like the others: New methods to account for the unique implications of indirect range-restriction corrections in organizational research. Unpublished manuscript.

Duhachek, A., & Iacobucci, D. (2004). Alpha<U+2019>s standard error (ASE): An accurate and precise confidence interval estimate. Journal of Applied Psychology, 89(5), 792<U+2013>808. https://doi.org/10.1037/0021-9010.89.5.792

Examples

# NOT RUN {
var_error_rel(rel = .8, n = 100)
var_error_rel(rel = .8, n = 100, rel_type = "alpha", k_items = 10)
# }