Transformation Plots

The transformation plots display the original category number on the horizontal axes; the vertical axes give the optimal quantifications.

Figure 1. Transformation plot for menstruation
Transformation plot for menstruation

Some variables, like Menstruation, obtained nearly linear transformations, so in this analysis you may interpret them as numerical.

Figure 2. Transformation plot for School/employment record
Transformation plot for School/employment record

The quantifications for other variables like School/employment record did not obtain linear transformations and should be interpreted at the ordinal scaling level. The difference between the second and third categories is much more important than that between the first and second categories.

Figure 3. Transformation plot for Binge eating
Transformation plot for Binge eating

An interesting case arises in the quantifications for Binge eating. The transformation obtained is linear for categories 1 through 3, but the quantified values for categories 3 and 4 are equal. This result shows that scores of 3 and 4 do not differentiate between patients and suggests that you could use the numerical scaling level in a two-component solution by recoding 4’s as 3’s.

Next