|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Sample size in each category |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
Dear Dr. Muthén, In a path analysis, how many participants do we need per category in order to test a model with a grouping variable (with 3 categories) and with 2 dependent variables, 2 independent variables and 2 control variables? We want to make sure that our results are not only due to one category having less respondents (our 3 categories have 89, 38 and 77 respondents). Thank you very much! |
|
|
Those group sizes seem fine because you don't have many parameters per group. The 38 group will have less influence but that's ok. |
|
|
Thank you - much appreciated! |
|
Back to top |
|
|