Select menu: Stats | Regression Analysis | Linear Models
The Multiple linear regression with groups downdown list option fits a sequence of models to data values that are classified into groups.
Specifies the name of the response (or y-) variate.
Specifies the names of the explanatory (or x-) variate to be fitted. If there is more than one explanatory variate, their names must be separated by spaces or commas. Alternatively, you can select the explanatory variables within the Available data list and click the button to copy them across.
Specifies a factor defining the different groups.
For an analysis of parallelism the first model to be fitted is an ordinary multiple linear regression, ignoring the groups. Next the model is extended to include a different constant (or intercept) for each group, giving a set of parallel relationships one for each group. Then, the final model has both a different constant and a different regression coefficients (or slope) for the explanatory variates across the different groups. The list adjacent to the Groups field lets you select between the types of regression model that you want to fit.
For an analysis of parallelism, if the analysis shows that different intercepts are needed but not different slopes, you can use this option to select the final model and re-run the analysis to remove the interaction between the explanatory variates and the groups factor. Similarly, if different intercepts are not needed this option can be used to fit just the explanatory variates.
- Linear Regression for information on general options and other models
- Options for choosing which results to display
- Further Output for additional output subsequent to analysis
- Saving Results for further analysis
- Fitted Model for graphical display of the model
- Model Checking for diagnostic plots for model checking
- MODEL, FIT and ADD directives for fitting regression models with groups using the command language