Power Analysis w Monte Carlo Sim w 3+... PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Christopher Wiese posted on Friday, July 21, 2017 - 10:03 am
Hello Drs. Muthén,

I'm running a power analysis for a 18 latent factor structure with 3 items per each factor (54 items total). I've built the Monte Carlo Simulation (correctly I believe) and gotten it to run without issue. My questions concerns how do I know if I have enough power. I've read Muthén & Muthén (2002) and the results that I've met the three criteria (parameter, SE, and coverage). However, I'm unsure how I would determine the power for this model.

The Muthén & Muthén paper suggests I should look to column 7 (Sig Coef) as an indication of power. With the example given in the paper, I believe this means looking at the WITH statement (.812 in Appendix B) - But I'm not entirely sure about that. If this is the case, how would I determine power when there are multiple latent factors? I've pasted my the WITH statements below (ignore the set values, they are placeholders)

f1 WITH f2-f18*.25;
f2 WITH f3-f18*.25;
f3 WITH f4-f18*.25;
f4 WITH f5-f18*.25;
f5 WITH f6-f18*.25;
f6 WITH f7-f18*.25;
f7 WITH f8-f18*.25;
f8 WITH f9-f18*.25;
f9 WITH f10-f18*.25;
f10 WITH f11-f18*.25;
f11 WITH f12-f18*.25;
f12 WITH f13-f18*.25;
f13 WITH f14-f18*.25;
f14 WITH f15-f18*.25;
f15 WITH f16-f18*.25;
f16 WITH f17-f18*.25;
f17 WITH f18*.25;
 Bengt O. Muthen posted on Friday, July 21, 2017 - 2:13 pm
The power given in the %Sig column refers to a single parameter, not a set of parameters; it will be different for your different factor covariances.
 Christopher Wiese posted on Friday, July 21, 2017 - 2:28 pm
Great! Thank you for your quick response :D
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message