|
|
 |
| Power analysis on Multiple group path... |
 
|
| Message/Author |
|
|
|
|
Dear Dr. Muthen
Hello, this is my model input.
Analysis:
Estimator is ML;
Model:
ERC on EPA EMA EOA Evo KMA;
Evo on EPA EMA EOA KMA;
EPA with EMA;
EMA with EOA;
EPA with EOA;
EPA with KMA;
EOA with KMA;
EMA with KMA;
EPA; EMA; EOA; KMA;
Evo; ERC;
Model indirect:
ERC ind EPA;
ERC ind EMA;
ERC ind EOA;
ERC ind KMA;
model ESL:
model EFL:
Output: standardized sampstat residual mod tech1;
The sample size for the ESL group is 50; and the EFL, 257. Since the sample size between the two groups is unbalanced, one of the reviewer of my manuscript asked me reporting the power of the analysis.
1. Could you help me how I can do the power analysis for the multiple group path analysis?
2. I did a bootstrap analysis for the model. If I report the 95% CI for the path coefficients, would it be one way to show the power of the model?
Thank you in advance for your help.
Sincerely,
Han Suk |
|
|
|
|
1. You can use 3 sources for guidance. First, each UG example has a Monte Carlo version on our website. Second, you can study the article on our website:
Muthén, L.K. & Muthén, B.O. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 4, 599-620. Mplus inputs and outputs used in this paper can be viewed and/or downloaded from the Examples page.
download paper contact first author show abstract
Third, you can study our new book:
http://www.statmodel.com/Mplus_Book.shtml
2. No, this does not show power. |
|
|
| ehrbc1 posted on Wednesday, May 31, 2017 - 1:07 am
|
|
|
Hello,
I have a similar question to Bae, Han Suk.
I've had a reviewer request some post-hoc power analyses. So based on my sample size and the effect size observed, what was the power (for certain regression coefficients in a multiple group path analysis)?
I'm assuming I don't need to a simulation/monte carlo for this? Could you please point me in the right direction?
Thank you. |
|
|
|
|
| You can do power in line with our Topic 4 handout and video, slides 110-120. Or you can do it via Monte Carlo simulation as we discuss in our new book. |
|
|
| Back to top |
|
|
