SPSS Statistics Subscription, launching on March 28, includes several key new features. Last time we reviewed the new chart options launching in SPSS Statistics Subscription. Today’s blog will focus on enhancements to existing statistics. All of these enhancements were driven directly by customer feedback.

The Mixed Linear Models (MIXED) & Generalized Linear Mixed Models (GENLINMIXED) procedures now provide:

  • Random effects solution results (EBLUPs).
  • Continuous time spatial covariance structures.

The General Linear Model (GLM) and UNIANOVA procedures have been enhanced with new features including:

  • Profile Plots with error bars, bar/line charts, an option to include grand mean, and an option to force the chart to include 0 on Y axis.
  • Added new tests for heteroskedasticity and model specification.
  • Robust standard errors.
  • Modified versions of Levene’s test.

All these features will be available at launch time for SPSS Statistics Subscription. UNIANOVA is found in the Base Edition of the Subscription while MIXED, GENLINMIXED, GLM and REPEATED MEASURES are found in the Custom Tables & Advanced Statistics Add-on. See this related blog if you’re interested in the packaging options.

3 comments on"Enhanced Statistics for SPSS Statistics Subscription"

  1. Richard Anderson June 14, 2017

    Hi. Regarding “Profile Plots with error bars:” I understand that that works for GLM: UNIVARIATE. Does it also work for GLM: REPEATED MEASURES?

  2. It’s great that the EBLUPs can finally be produced in SPSS. However, they seem to be available in the output only. This generally creates lots of output: the number of rows in the produced table is equal to the number of random effect parameters times the number of observations on the defined level. This is of course not very useful, it would be awesome if one could use OMS to save them into a file. However, this option doesn’t seem to be available in the OMS Control Panel. Is this something that could be added in the future? Or is there another way of achieving this (besides copying and pasting from the output table…)?

    Many thanks 🙂

Join The Discussion

Your email address will not be published. Required fields are marked *