Message/Author |
|
JW posted on Monday, October 22, 2007 - 12:45 pm
|
|
|
I'm conducting a path model, with two mediating variables, but can't figure out how to obtain results for the test of significance for indirect effects i.e., the Sobel equation. I've imputed the IND command and the CINVTERVAL, but not sure how to obtain z score for significance. |
|
|
The IND option is part of MODEL INDIRECT. See MODEL INDIRECT in the user's guide. |
|
|
Hello, If I understand correctly, this means that by using the MODEL INDIRECT command in the input, the Z score (EST. / S.E.) that we obtain in the output under "Indirect effect" is equivalent to the one we would compute using Sobel's (1982) equation, z-value = a*b/SQRT(b2*sa2 + a2*sb2) Is that right? |
|
|
Yes. |
|
|
Dear Linda, I need to use the z-scores of the observed variables in my model. Could you please tell me how to get the z-scores and save them in Mplus? Regards, Prathibha |
|
|
I think you are asking how to standardize your observed variables. You can do this by using the STANDARDIZE option of the DEFINE command which is described in the Version 6.1 Language Addendum that is on the website with the user's guide. You can save these using the SAVEDATA command. See the user's guide for details. |
|
Back to top |