I am currently using JMP Student Edition to perform Structural Equation Modeling (SEM) analyses, and I am writing to inquire about the multiple group analysis (Measurement Invariance testing) feature.
[My Environment]
[Analysis Objective] I am trying to perform a measurement invariance test (evaluating configural, metric, scalar, and strict invariance, and calculating differences like ΔCFI, ΔRMSEA) across sex using data from a psychological scale (14 items, 3-factor structure).
[Current Steps and Issue] I have followed the steps below:
-
Selected Analyze > Multivariate Methods > Structural Equation Models.
-
Assigned the 14 item variables to Model Variables.
-
Assigned the categorical group variable (Sex, with the red nominal bar chart icon) to the Grouping role. (Note: The By role was left completely empty).
-
Built the path diagram and clicked Run.
After running the model, tabs for each group (1 and 2) appear at the top left of the report window, indicating that the multiple groups are correctly recognized. However, when I click the red triangle menu (▼) next to "Structural Equation Model: Model 1", the "Measurement Invariance" option does not appear in the menu.
[My Questions] Based on the above, could you please clarify the following points?
-
Does the current SEM platform in "JMP Student Edition" include the automated Measurement Invariance testing feature (the model comparison macro)?
-
If the feature is included, could you please let me know what might be causing the menu not to appear, or if there is a specific configuration I might be missing?
I would greatly appreciate your clarification on this matter so I can determine how to proceed with my analysis.
Thank you for your time and support.
To my knowledge JMP Student is same as JMP Pro with a different license (JMP® Student Edition - FAQ).
I'm not familiar with that platform but here is an example Multivariate Methods > Structural Equation Models > Additional Examples of Structural Equation Models > Example of Multigroup Measurement Invariance
Thank you very much for your advice!
You were absolutely right. I followed the example you pointed out and used the "Model Shortcuts" (Multigroup Measurement Invariance) instead of drawing the path diagram manually, and it worked perfectly!
Thanks to your help, I was able to successfully run the invariance tests (Configural, Metric, Scalar, and Strict) and obtain all the necessary model comparison tables for my research.
I truly appreciate your prompt and accurate support. It saved me a lot of time!
Best regards,