delta_var {dmacs}R Documentation

Expected bias to total score variance

Description

delta_var computes the expected bias in total score variance due to measurement nonequivalence. delta_var will only work for unidimensional linear models (not categorical).

Usage

delta_var(LambdaR, LambdaF, VarF, categorical = FALSE)

Arguments

LambdaR

is the vector of factor loadings for the reference group.

LambdaF

is the vector of factor loadings for the focal group.

VarF

is the factor variance of the focal group.

categorical

is a Boolean variable declaring whether the variables in the model are ordered categorical. Categorical indicators are not supported for this function.

Details

delta_var is called by dmacs_summary_single, which in turn is called by lavaan_dmacs and mplus_dmacs, which are the only functions in this package intended for casual users

Value

The expected bias in total score variance due to measurement nonequivalence in equation 7, 8, and 9 of Nye & Drasgow (2011).

References

Nye, C. & Drasgow, F. (2011). Effect size indices for analyses of measurement equivalence: Understanding the practical importance of differences between groups. Journal of Applied Psychology, 96(5), 966-980.

Examples

LambdaF <- c(1.00, 0.74,  1.14, 0.92)
LambdaR <- c(1.00, 0.76,  1.31, 0.98)
VarF    <- 1.76
delta_var(LambdaR, LambdaF, VarF)


[Package dmacs version 0.1.0 Index]