Hi
@Jeff Conlin,
Let's just get it out of the way: IBM SPSS Amos is a Windows only application and has never been tested nor is it supported under Parallels running on macOS.
OK, so let's pretend you are running Amos 28.0 on Windows 10 on dedicated physical intel/amd x86_64 architecture.
In the Amos Graphics path diagram file I would check:
- Are all groups in your path diagram pointing to the exact same file in the exact same location?
- Are all groups using the exact same grouping variable from this file.
- Are all groups using a unique value from the grouping variable in that data file?
- Do the combination of Grouping variable value and the other variables in your model actually result in data? I know you've said that you have no missing data. Is it possible that you actually do have missing observations for some variables across some level of you grouping variable - or even missing values in your grouping variable? How about some logical combination of categorical variables in your model that are implied, but do not actually occur in the data?
I hope this helps you to further troubleshoot your model and find the true cause of the error you are seeing.
------------------------------
David Dwyer
SPSS Technical Support
IBM Software
------------------------------
Original Message:
Sent: Fri February 03, 2023 08:52 PM
From: Jeff Conlin
Subject: AMOS 28.0 Error
I'm posting to get some help with resolving a technical error that I'm receiving that says, "An error occurred while attempting to read the data file for the group." My data is accessible in SPSS 28.0 but not in AMOS 28.0. I'm also using Windows 10 and the latest version of Parallels for Mac to run Windows and SPSS/AMOS. I'm running my Mac on Sierra 13.2.
My SEM models previously worked fine and there is no missing data in my dataset. I can open the data in SPSS. But I keep getting the same error message even when de/re-selecting different SPSS data files. I haven't found the issue elsewhere but this sure seems like a bug …
Is anyone able to help troubleshoot?
------------------------------
Jeff Conlin
------------------------------