|
|
Standardized effects for Monte Carlo ... |
|
Message/Author |
|
|
Hello I'd like to conduct a Monte Carlo simulation to find about an appropriate sample size for my research (a conventional SEM). But how could I inform Mplus about the standardized effect sizes for my variables of interest? I'd like to use standardized regression coefficients b*(sd_x/sd_y) for the structural model of my SEM, but as far as I can see, the Monte Carlo inputs are based on unstandardized regression coefficients? Best Tino |
|
|
Your effects will be in standardized metric if you choose parameter values so that all variables have variances 1. For instance, with x*1; y on x*0.7; y*0.51; both x and y have variance 1 and 0.7 is the standardized effect. |
|
Back to top |
|
|