|
Message/Author |
|
|
Dear Dr. Linda: Please help me with the below model: COND = 1 = family treatment; 0 = group COMDD = presence/absence of major depression; 1 = present; 0 = absent COXMDD = Product Term CONDXCOMDD RCMASPPO = anxiety at posttreatment). MODEL RESULTS Estimate P-Value RCMASPPO ON COND -0.861 0.215 RCMASP 0.577 0.000 COMDD -3.205 0.039 COXMDD 5.747 0.017 RCMASPFU ON COND -0.410 0.593 RCMASP 0.059 0.490 RCMASPPO 0.442 0.000 COMDD -2.813 0.025 COXMDD 3.553 0.146 Intercepts RCMASPPO 4.486 RCMASPFU 0.597 What I need help with is to interpret the meaning for COND; COMDD; COXMDD and the Intercept. I would like to get the mean values for the 2X2 table based on the results below, but I'm not sure how to get the corresponding values. I'm thinking that COND = family - group when MDD = 0. COMDD = MDD Present - MDD Absent for group Intercept = Value for residual change for MDD Present in group I trust this site the most and I hope you'll be able to help me. Thank you very much!! |
|
|
For simplicity, you could center the continuous covariates and evaluate the other covariates at 0 0, 0 1. 1 0 , and 1 1. |
|
|
Thanks a lot for your response. I started to pursue your suggestion with 1 of my 4 moderators I set out to test. Thus far, i had to create 4 different files to run the data for 0 0, 0 1, 1 0, and 1 1. Is there an easier way that you know of to do this without creating the different files? Also, how can I get the SD for each of the 4 groups in the 2X2 table. I can figure out the means, but I don't know how to get the SD. I want to create a table with the means and SD for each of the moderators. Thanks a lot! |
|
|
You don't need 4 runs - you have all the estimates you need in the run with the interaction. You just combine the estimates in 4 different ways in line with regular dummy variable regression: - the intercept gives the mean when both dummies (and therefore their interaction) are zero - the intercept plus one slope gives the mean for the cells where you have 0, 1 or 1,0 - the intercept plus both means gives you the mean for the cell where both dummies, and therefore their interaction, are 1. SD's are the same for each cell, again in line with regular regression. |
|
|
I am following up to say that the above was helpful and I was able to figure out the means. Thanks very much. What I do not know how/where to get from the output are the SDs. Also, if I wanted to look at the difference between the 2 treatment conditions at each level of the moderator, will I need to create a separate file and run the equivalent of a t-test in Mplus? I also would like to look at the difference between participants with and without major depression within each treatment condition (i.e., group; family). In other words, I want to explore simple main effects. I am asking because i don't think I can use the output on the moderation to answer this because the coefficients i get are 'conditional.' Thanks very much again. This helps me a lot toward my Dissertation. |
|
|
The SD is obtained as the square root of the residual variance (the variance given a certain cell). You can create an comparison you want using New parameters in Model Constraint, where model parameters are referred to by using labels in the Model command. To practice using Model Constraint, you can do the mean computations you just did. |
|
Back to top |
|
|