University of AlbertaGroup treatment data are nested by design, that is, clients nested in groups. Dependenceassociated with the nesting of group intervention data can insate Type I error rates,whichposesuniquechallengestogrouptreatmentresearchers.Thisarticleevaluatestheextent and variability of dependence in data taken from 3 previously publishedrandomized clinical trials of group psychotherapy. Three methods of assessing depen-dence by calculating intraclass correlation coefÞcients ( ) were examined. Resultsshowed great variability in s across studies, across methods of calculating , andacross outcome variables. The distribution of s suggested that the amount of depen-dence in the data was moderate. Two methods of addressing dependence in groupedtreatment data through multilevel modeling were used. These methods resulted inminimally compromised statistical power compared with results from uncorrected data.These 2 methods may allow researchers to reliably assess their group treatments. Groupintervention researchers are encouraged to consider their assumptions and conceptual-izations of treatment change, and to choose corresponding methods of assessing for andaddressing .
No takes yet. Share an insight, caveat, or question.
Tasca et al. (2009) studied this question.