alpha_cov: Calculate covariate-free and covariate-dependent Cronbach's...

In hplieninger/stylesim: Simulate (and Analyse) Data Distorted by Response Styles

Description

If a covariate (or possibly a set of covariates) is partialled out from a set of target variables, their corresponding coefficient alpha can be decomposed into a part that is dependent on that covariate and a part that is independent ("free") from that covariate (and the sum equals ordinary coefficient alpha).

Usage

alpha_cov(data, xx, zz = NULL, use = "pairwise")

Arguments

data	A data frame or matrix containing both the variables as well as the covariate(s). This can also be a covariance matrix.
xx	Numeric. A vector of indices to specify the columns in data that pertain to the target variables.
zz	Numeric. A vector of indices to specify the columns in data that pertain to the covariate(s).
use	an optional character string giving a method for computing covariances in the presence of missing values. This must be (an abbreviation of) one of the strings "everything", "all.obs", "complete.obs", "na.or.complete", or "pairwise.complete.obs".

Value

Returns a named vector with covariate-free, covariate-dependent, and total Cronbach's alpha (unstandardized).

References

Bentler, P. (2015). Covariate-free and covariate-dependent reliability. Manuscript submitted for publication. Retrieved from http://escholarship.org/uc/item/6v61v1pk

See Also

Cronbach's alpha in the psych package: alpha

hplieninger/stylesim documentation built on May 17, 2019, 4:54 p.m.