Log-Linear Analysis of Frequency Tables

Select the More fit statistics tab of the Log-Linear Analysis Results dialog box to access the options described here.

Components of max-Likelihood Chi. Click the Components of max-Likelihood Chi button to display the Specify how to Review the Table dialog box, in which you specify how to review the table of components of the maximum likelihood ratio Chi-square, in the most flexible manner, slice by slice, as you specify. If there are more than two factors in the table, you can view any two-way table within each of the levels of the remaining factors.

These are the contributions of each cell to the overall maximum likelihood ratio Chi-square goodness-of-fit statistic:

cijk..=2*fijk..*ln(fijk../Fijk..)

In this equation, ln denotes the natural logarithm; Fijk.. denotes the fitted or expected cell frequency for cell i,j,k,...; fijk.. denotes the observed frequency.

Freeman-Tukey deviates. Click the Freeman-Tukey deviates button to display the Specify how to Review the Table dialog, in which you specify how to review the table of Freeman-Tukey deviates, in the same manner (slice by slice) as described above. Freeman-Tukey deviates (frijk..) represent a normalizing transformation that is appropriate when the frequencies in the table come from a Poisson

frijk..=fijk..½+(fijk..+1)½-(4*Fijk..+1)½

Plot of fitted vs. std. residuals. Click the Plot of fitted vs. std. residuals button to display a scatterplot of fitted frequencies versus standardized residuals. This graph is very useful for evaluating the overall fit of the model for different magnitudes of frequencies, and for detecting outliers.

Plot of fitted vs. m-L components. Click the Plot of fitted vs. m-L components button to display a scatterplot of fitted frequencies versus components of the maximum likelihood ratio Chi-square. This graph is very useful for evaluating the overall fit of the model for different magnitudes of frequencies, and for detecting outliers.

Plot of fitted vs. F-T deviates. Click the Plot of fitted vs. F-T deviates button to display a scatterplot of the fitted frequencies versus Freeman-Tukey deviates. This graph is very useful for evaluating the overall fit of the model for different magnitudes of frequencies, and for detecting outliers.