General CHAID Models Quick Specs - Quick Tab
Select the Quick tab of the General CHAID Models Quick specs dialog box to access the options described here.
- Variables
- Click this button to display a variable selection dialog box from which you select the dependent variable, categorical predictor variables, continuous predictors, and count variable (optional).
From the first list, select the dependent variable. If you are analyzing a classification problem with a categorical response variable, the dependent variable must contain text or numeric (integer) codes that identify the class or group to which each case or observation belongs.
From the second list, select the categorical variables (factors). Note that these variables must contain text or numeric codes that identify the classes or groups to which each case or observation belongs.
From the third list, select the continuous predictors.
From the fourth list, select the count variable. This variable is optional. If selected, it will be used as a case-multiplier in the analysis. Thus, by using a count variable you can analyze input data that have previously been aggregated (tabulated).
- Categorical Response
- Select this check box to analyze a dependent (criterion) variable that is categorical in nature, i.e., to perform a classification analysis. When you select this check box, the Response codes button will also become active, provided you have specified a dependent variable. Note that if this check box is not selected, even the categorical response variable selected for the analysis will be treated as a continuous response variable. Likewise, if the selected dependent variable is continuous and the Categorical response check box is set, the dependent variable will be treated as categorical.
- Response codes
- This button is available only after a dependent variable has been specified, and if the Categorical response check box is selected. In that case, the dependent variable is assumed to be categorical in nature, and the object of the analysis is to correctly classify the cases (observations) into groups specified by the dependent variable. Click the Response codes button to select codes for the Categorical response from the list of codes for the Dependent variable selected in the current analysis. The list of codes can be displayed by clicking on the Zoom button of the Select codes for the categorical response dialog. Note that you must select at least two codes for the selected Dependent variable.
- Factor codes
- This button is available after variables have been selected. Click this button to select codes for the Categorical factors. The list of codes can be displayed by clicking on the Zoom button of the Select codes for indep. vars (factors) dialog. Note that you must select at least two codes for each of the selected Categorical factors.
- Between effects
- In Standard CHAID analysis, the current "design" will be shown, which in this case amounts to a simple combination of the selected predictor variables and the factors (categorical predictors).
In CHAID with coded design analysis, click the Between effects button to display the General CHAID Between Effects dialog box. Use this dialog box to specify an ANOVA/ANCOVA-like design for the selected predictor variables. The design for the factor effects can be specified in the same manner as described in the context of the General Linear Models. In that case, the categorical factor effects and the interactions are transformed into the coded vectors of a design matrix, and the analysis is performed on the vectors of the resulting design matrix; see also, Analyzing ANCOVA-like designs. To learn more about ANOVA/ANCOVA designs, and general linear models, see the Introductory Overview for General Linear Models.