I am trying to run a mixed effects model. in the estimates of fixed effects, I receive estimates for each value of the dependent variable separately (although defined as a scale), instead of an estimate for the total score. I assume the problem is in the definitions. I ran LLM, not attracting this problem. I would appreciate help identifying the problem in my definitions.
Could you provide more information about your model, data structure, and the statistic program (and packages) you are using?
I use the restricted maximum likelihood method.
the dependent variable is a scale, time is a dummy variable with two categories (time 1 and time 2, Time 3 as a reference)., and three variables as fixed variables.
individuals are defined as random. the statistic program is SPSS
Can you show us the output of the results from SPSS? From what you have said I would expect 6 fixed effects ( that is estimated means) - the intercept, the two differential intercepts associated with the time dummies, and the 3 slope coefficients associated with the three predictors. Correct?
I am presuming that it is a two level model with individuals at level 2 and occasions at level 1 and that there are two variance terms representing between individual variance and within individual between occasion variance.
Thank you. I am attaching, just for an example a model with only one predictor, the estimates look the same with the three predictors
I am not a SPSS user but it looks as if your predictor variables are being modelled as categorical (factor) and not continuous (scale) ; the problem is not your response variable but your predictors.
I agree with Kelvin. I think it would help if you change the scaling of the predictor variables ("measure" column in your dataset; change it to "scale"). If you have not already done so, you could also try to include the PANAS variable as a covariate (and not as a factor) in the model.
Thank you Marc and Kelvin for your help. indeed the variables are defined as scale.
Defining the predictors as covariance instead of as fixed factors solves the problem. Thank you again
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA