turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- Problem with repeated measures / mixed model

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

May 23, 2012 7:01 PM
(2456 views)

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

Solved! Go to Solution.

1 ACCEPTED SOLUTION

Accepted Solutions

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

May 24, 2012 12:23 PM
(4312 views)

The modeling type for Participant needs to be "nominal", not "continuous". See if that helps. I couldn't compare the results with yours because it appears SPSS is necessary to read the .spv and .sav files.

2 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

May 24, 2012 12:23 PM
(4313 views)

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

May 24, 2012 1:47 PM
(2271 views)

Thank you! That was indeed the problem.