Workspace Node: Goodness of Fit - Specifications - Quick Tab

The Goodness of Fit workspace node can be accessed from the Feature Finder, ribbon bar, or Node Browser. The Specifications - Quick tab is displayed by default when you double-click the node.

Note: Computations - In this module, casewise deletion of missing data is used to process the data. This may cause different results if you select different variables. For example, if you calculate goodness of fit statistics for Y, X1 and X2, then recalculate those statistics for Y, X1, X2 and X3, when X3 has missing data, the results will be different. If all cases are deleted by the casewise deletion method, an error message will be issued and the analysis will stop. Additionally, if a selected variable has no variance and correlation coefficient statistics have been selected, an error message will be issued and the analysis will stop.
Element Name Description
Variables Click this button to display a two-variable selection dialog box; specify the variable with the observed values or classifications, and a list of variables containing the predicted values or classifications.
Variable type Specify whether the observed (and predicted) values are continuous in nature (e.g., were computed for a regression-type problem) or categorical in nature (e.g., were computed for a classification problem). Different goodness of fit statistics can be computed for continuous and categorical variables (see the option descriptions in the Advanced tab topic).
Continuous Select this option button to treat the selected variables as continuous variables. Continuous variables are those measured on a continuous scale, such as Height or Weight.
Categorical (classification) Select this option button to treat the selected variables as categorical variables. Categorical variables are those measured on a nominal scale, such as Gender (Male, Female) or Educational Achievement (No High School, High School, College, Post-Graduate).
Statistics for Continuous DV Select the respective goodness of fit statistics to compute for the continuous (observed vs. predicted) variables; these options are not available if the Variable type is set to Categorical (classification). For more information about these statistics, see Computational Details.
Least square deviation (LSD) Select this check box to compute the ordinary least-squares goodness of fit statistics; the results summary spreadsheet will include the Mean square error goodness of fit statistic.
Average deviation Select this check box to compute the average absolute deviations of the predicted values from the observed values; the results summary spreadsheet will include the Mean absolute error goodness of fit statistic.
Statistics for Categorical DV Select the respective goodness of fit statistics to compute for the categorical variables (observed vs. predicted classifications); these options are not available if the Variable type is set to Continuous. For more information about these statistics, see Computational Details.
Chi-square Select this check box to compute the Pearson Chi-square statistic of goodness of fit.
G-square (Maximum-Likelihood) Select this check box to compute the G-square goodness of fit statistic.

Options / C / W. See Common Options.

OK Click the OK button to accept all the specifications made in the dialog box and to close it. The analysis results are placed in the Reporting Documents node after running (updating) the project.

See also, Specifications - Advanced tab, Results - Quick tab, Results - Advanced tab, Downstream tab, and Home tab.