# Statistical Advisor, Stratified Regression

Use one of the following to test predictions about differences in relationships between variables in different groups.

## General Linear Models (GLM), Generalized Linear Models (GLZ)

One option is to use General Linear Models (GLM or Generalized Linear Models (GLZ)

The GLM chapter discusses an implementation of the general linear model, the GLZ chapter discusses an implementation of the generalized linear model. Both chapters discuss how to analyze analysis of covariance (ANCOVA), separate slopes, and homogeneity of slopes designs.

## Nonlinear Estimation

Another option is to use Nonlinear Estimation.

This chapter discusses techniques to estimate any kind of regression equation (relationship) between variables. Such equations may contain logical variables, and thus, stratified (by groups) analyses can easily be performed. For example, one may hypothesize that compensation and performance are related in one occupational group, but not another. Refer to the manual for details.

## Survival Analysis

Another option is to use Survival Analysis.

This chapter discusses techniques for performing linear regression as well as various nonlinear regression analyses. In addition, one can perform stratified (by group) analyses, that is, test whether or not the respective regression equation (and thus relationship between variables) is the same in all groups. SURVIVAL ANALYSIS can also be used for censored survival/failure times. A data set is said to be 'censored' if some observations are incomplete, but not missing. For example, in industrial reliability testing, a part or product may not fail (break) within the time span covered by a study. However, we do not know how long it will function properly thereafter, and thus, that observation is 'censored.'

## Graphical Analytic Techniques

Graphical analytic techniques allow you to produce scatterplots, surface plots, etc., stratified (categorized) by groups.