SPSS command: SPSS GENLINMIXED
Generalized linear mixed models in your browser
Generalized linear mixed models handle binary, count or continuous outcomes measured repeatedly within subjects.
Free · No account · Files stay on your device · Updated 18 October 2026
What it tells you
Crosstabs fits the SPSS GENLINMIXED model on your device with random intercepts and slopes per subject, using the restricted pseudo-likelihood with penalized quasi-likelihood linearization (REPL-PQL) that SPSS uses.
- Which fixed effects matter, with F tests on residual degrees of freedom.
- The fixed coefficients with standard errors, confidence intervals and odds or rate ratios.
- How much subjects vary in their baseline and in their response to covariates (random-effect variances).
When to use it
Use it when
- Respondents answer the same yes/no question several times, or survey waves are nested in panelists.
- Counts per store or per respondent, with stores or respondents measured repeatedly.
Use something else when
- You need repeated-measures (R-side) covariance structures or several random-effect blocks: only one subject block of G-side effects is supported.
- You need Laplace or adaptive quadrature estimates: the panel uses REPL-PQL, as SPSS does.
How to run it in Crosstabs
- 01
Open your data
Open a .sav, CSV or Excel file in the workspace and switch to Statistics mode. The file is read in your browser and never uploaded.
- 02
Choose Generalized linear mixed models
In the General linear model group, pick Generalized linear mixed models (GENLINMIXED).
- 03
Pick the distribution and target
Choose Distribution and link, the Target and, for binary models, the Event category; then choose the Subject.
- 04
Add fixed and random effects
Tick Fixed factors and Fixed covariates; under Random effects keep Random intercept, tick Random slopes and set the Covariance type.
- 05
Run
Click Run and read the Fixed Effects, Fixed Coefficients and Random Effect Covariance tables.
Output tables explained
- Fixed Effects
- An F test for each fixed term with residual degrees of freedom.
- Fixed Coefficients
- Coefficients on the link scale with standard errors, t tests and intervals; Exp(Coefficient) gives odds or rate ratios.
- Random Effect Covariance
- Variance and covariance estimates of the random effects with Hessian standard errors and Wald intervals.
Assumptions and notes
- The scale is fixed at 1 for binomial and Poisson targets with only G-side effects, as in SPSS.
- Random effects are normal and independent across subjects.
How the results are validated
A dense-matrix R implementation of REPL-PQL run in Docker matches fixed effects and their standard errors within 1e-5 and covariance parameters within 1e-5 for binomial logit and probit and Poisson models; the normal model matches nlme and lme4 REML. lme4::glmer Laplace estimates are recorded for comparison only. Validation evidence
Questions
Why do the estimates differ from lme4::glmer?
glmer uses the Laplace approximation; SPSS GENLINMIXED uses REPL-PQL linearization, which shrinks binary-outcome variances somewhat. Crosstabs follows SPSS.
Which category is the event?
You choose it under Event category; by default the last category is modeled.
Are weights applied?
No. GENLINMIXED uses analysis (scale) weights rather than frequency weights, so the workspace weight is not applied.
What do variance components and unstructured mean?
Variance components estimate a separate variance for the intercept and each slope; unstructured also estimates their covariances.
Related procedures
- Linear Mixed ModelsMIXED
- Generalized Linear ModelsGENLIN
- Binary Logistic RegressionLOGISTIC REGRESSION
- Repeated Measures ANOVAGLM WSFACTOR
- Free SPSS alternative: full coverage list
Run Generalized Linear Mixed Models (GENLINMIXED) on your own data
Open a .sav, CSV or Excel file. Everything runs in your browser; the file never leaves your device.
SPSS is a trademark of IBM Corp. Crosstabs is not affiliated with or endorsed by IBM.