Class Evolution Tree

Abstract

The aim of our study was to develop a graphical tool that can be used in addition to standard statistical criteria to support decisions on the number of classes in explorative categorical latent variable modeling for rehabilitation research. Data from two rehabilitation research projects were used. In the first study, a latent profile analysis was carried out in patients with cancer receiving an inpatient rehabilitation program to identify prototypical combinations of treatment elements. In the second study, growth mixture modeling was used to identify latent trajectory classes based on weekly symptom severity measurements during inpatient treatment of patients with mental disorders. A graphical tool, the Class Evolution Tree, was developed, and its central components were described. The Class Evolution Tree can be used in addition to statistical criteria to systematically address the issue of number of classes in explorative categorical latent variable modeling.

Drinking patterns of alcohol intoxicated adolescents in the emergency department: A latent class analysis

“…Entropy values indicate how well classes are separated with values N0.8 indicating good separation (Tein, Coxe, & Cham, 2013). Model selection was based on fit indices as well as practical interpretability and research objective as recommended (Jung & Wickrama, 2008;Kriston et al, 2011;Muthén & Muthén, 2000). Models with 1 through 9 classes were tested using MPlus Version 5 (Muthén & Muthén, 2011).…”

“…A 7-class model was suggested by BIC and a 9-class model by AIC. Because different statistical fit indices suggested different class solutions research objective and practical interpretability received more weight in deciding on the number of classes as recommended (Jung & Wickrama, 2008;Kriston et al, 2011;Muthén & Muthén, 2000). In the 2-class solution, 283 subjects (89.6%) were assigned to one class.…”