I am an SPSS user trying to migrate to JMP. I am having problems with a repeated measures experiment.
In my study, respondents select between various base model vehicles vs. more fuel efficient vehicles. In the study, I manipulated how the vehicle's fuel economy information was presented.
The design:
- Repeated measure: Metric: “Gallons” vs. “Dollars” (i.e., all choices between base models and more fuel efficient models were presented with fuel economy in terms of gallons and later again in terms of dollars, or reverse order).
- Between-subjects variable: Metric order: "Gallons --> Dollars" vs. "Dollars --> Gallons".
- Between-subjects variable: Scale: “per 100 miles” vs. “per 100,00 miles”.
- Between-subjects variable: Differences Calculated: “Yes” vs. “No”.
- The dependent variable was "Choice", which was the average of six different vehicle comparisons.
I have attached my SPSS file. I arrange the data like this: Participant / Scale / Differences_Calculated / Metric_Order / Choice_Gallons / Choice _Dollars ... and then ran a GLM Repeated Measures analysis with a full factorial design. My output is attached and it matches what I see in the figure perfectly.
I am now trying to run the same analysis with what I believe is called a mixed model in JMP. In order to do this, I have rearranged the data to look like this: Participant / Scale / Differences_Calculated / Metric_Order / Metric / Choice ... and then tried to fit a full factorial mode + Participant as a random effect. My output is saved as a script in the attached. Basically, I get a convergence error and I don't know why.
Can you please help me understand what I am doing wrong and how to conduct the equivalent analysis?
Thanks,
Adrian
