Message/Author |
|
Alex posted on Thursday, January 26, 2006 - 2:14 pm
|
|
|
I am working on a multilevel analysis and need to control for the effects of an extraneous between level variable on the between level variance of a dependent variable (the dependent variable exists on the between and witnin levels). For various reasons, I cannot include the independent variable directly in the full analysis. Hence, I am attempting to control for the effects of the independent variable by creating a new variable. This variable would be created by regressing (at the between level) the dependent variable on the independent variable and then using the regression residuals as the new variable. Is there any way to save the regression residuals for each case in the dataset using Mplus 3.12? Any thoughts/help would be appreciated! |
|
|
No, it is not possible to save the regression residuals for each case. |
|
|
I am analyzing school data with a two level model (students in classes) with a random intercept and a random slope. I need the estimated intercepts and slopes of the single classes for specific analyses. Is it meanwhile with Mplus 7 possible to output the class-specific level2 residuals (residuals of the random intercept and the random slope)? If not: is there any other way to get the estimations of the intercepts and the slopes of single classes? Thank you in advance! |
|
|
Estimates of the intercepts and slopes are obtained by requesting factor scores. Level-2 residuals are obtained as described in the FAQ on our website: Between-level residual for each cluster |
|
|
Is the method described in the FAQ 'Between-level residual for each cluster' also applicable for multilevel logistic regression analyses? |
|
|
Yes |
|
Back to top |