Convergent validity of a reflective construct

Formula: AVE = sum lambda_i^2 / (sum lambda_i^2 + sum theta_i)

Usage

Convgs(loadings, residuals = NULL)

Arguments

loadings: Factor loadings of the indicators on their construct.
residuals: Indicator residual variances, or NULL for 1 - lambda^2.

Value

List with estimate (AVE), ave, cr, adequate, n_items, method.

Details

With standardised loadings the residual variance of an indicator is theta_i = 1 - lambda_i^2, which is what is used when residuals is not supplied. Composite reliability follows the same algebra, CR = (sum lambda_i)^2 / ((sum lambda_i)^2 + sum theta_i). Fornell and Larcker's rule is AVE >= 0.5 with CR >= 0.7.

References

Fornell & Larcker (1981), J. Marketing Research 18(1):39-50.

Examples

Convgs(loadings = c(0.7, 0.8, 0.75, 0.6))
#> $estimate
#> [1] 0.513125
#> 
#> $ave
#> [1] 0.513125
#> 
#> $cr
#> [1] 0.8066038
#> 
#> $adequate
#> [1] 1
#> 
#> $n_items
#> [1] 4
#> 
#> $method
#> [1] "convergent validity: AVE and composite reliability"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"