I am analysing treatment group differences at time 2 (two time points only) data on a psych measure outcome (with five covariates: age, gender, CVD, BMI and F2s), and I am interested in one of the predictor’s influence on this (f2s). The scripting I am planning to use (from another study of similar structure) includes using baseline F2s (F2_v1), however my restructured SPSS dataset (wide) has only the transposed F2.
How best to run a LMM analysis TMT*TIME*F2_baseline on a restructured dataset?
How do I keep the baseline F2s only - is there a step in the restructuring? (syntax below)
Thank you in advance.
MIXED bdi BY pptID Gender TMT CVD TIME WITH age BMI F2_V1
/CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE)
/FIXED=Gender TMT CVD time Age BMI F2_V1
TMT*time TMT*F2_V1 F2_V1*time TMT*F2_V1*time | SSTYPE(3)
/METHOD=REML
/PRINT=SOLUTION
/RANDOM=pptID | COVTYPE(VC)
/EMMEANS=TABLES(TMT*time) COMPARE(TMT).
dear: It is clear that the model used in the analysis is not aggregate, so the appropriate conversion must be made to make it aggregate, for example taking the logarithm to both sides of the model and then analyzing the new converted data in any program
Dear Karen Savage , could you please show an example of hoe your dataset looks like?
Based on your overall question, the issue seems to be easy to fix. However, I am not being able to fully understand how your data looks like and what you actually mean by f2_v1 - is there f2_v2? What's the difference between f2 and f2_v1?
Please explain your data and goal a bit more so I can help you. Thanks, Nick
In my opinion it would be much easier to run this as a regression in something like SAS or R.. See the analysis of covariance for a model.with Post = Pre + the rest of the model as described in the Analysis of Covariance sect. of D C Montogomery's Design And Analysis of Experiments 8th ed.
What do you mean by "... , and I am interested in one of the predictor’s influence on this (f2s)". I apologize but I don't speak SPSS. Best wishes, David Booth
- Badges
- Science topic