IBM SPSS Amos 25 was released on August 8th, 2017. The new features in version 25:

New Features:

  1. You can perform D-separation analyses.
  2. You can read R data files (*.rds and *.RData files) if R is installed on your computer.
  3. You can execute Amos from within an R program. (This is similar to the ability to execute Amos from the IBM SPSS Statistics main menu.)
  4. You can click File > File Explorer to show the current path diagram (*.amw) file in Windows File Explorer.
  5. Each user gets a separate copy of the example files. A copy of the example files from the User’s Guide is placed in each user’s Application Data folder. For a user named jim using Amos 25, this folder is located at C:\Users\jim\AppData\Local\AmosDevelopment\Amos\25\Examples.
  6. Environment variables help you locate important Amos folders.
  7. The list of variables displayed by View > Variables in Dataset now shows additional information about each variable, and allows sorting by variable name and variable label.
  8. IBM SPSS Statistics data files (.sav files) can now be viewed in Amos’s View Data window even if Statistics is not installed.

Other changes

  • Support has been discontinued for Lotus data files and for Excel 3 and Excel 4 data files.

SPSS Amos is found in the SPSS Statistics 25 Premium commercial bundle, and is also available to purchase separately. A trial version of Amos 25 will be available by mid-September. Check back here for a direct link to the trial version when it is ready to download.

4 comments on"What’s New in SPSS Amos 25"

  1. Christian Arnold September 10, 2017

    What absout robust ml estimation? Bootstrapped fit Indices?

  2. Robust ML estimation, other than what is currently available through bootstrapping, is on the to-do list, with no ETA at this time.

    Could you provide some references on how to perform a bootstrap for fit indices and how to interpret the results. I am pretty sure I know what you mean, and it would be easy to implement in Amos, but any references you can supply would be useful in documenting the procedure in the online help and the user’s guide.

  3. “Bootstrapping Goodness of Fit Measures in Structural Equation Models” (Bollen/Stine 1992). They do much more than calculating a p value. Assuming it is possible to calculate a bootstrapped chi square (which is possible) than it should be possible to scale fit coefficients like cfi, tli, rmsea… Regards

  4. Another reference September 22, 2017

    Kim, H. /Milliarden, R.: Using the Bollen-Stine Bootstrapping Method for Evaluating Approximate Fit Indices. In: Multivariate Behav. Res. 2014 Nov, 49(6): 581–596.

Leave a Reply