TY - JOUR
T1 - Multilevel models and unbiased tests for group based interventions
T2 - Examples from the safer choices study
AU - Carvajal, Scott C.
AU - Baumler, Elizabeth
AU - Harrist, Ronald B.
AU - Parcel, Guy S.
N1 - Funding Information:
Portions of this article were presented as part of the following presentation: Baumler, E. & Carvajal, S.C., The analysis of the Safer Choices Project: Multilevel models for program evaluation. Presented as part of a symposia Safer Choices: HIV/STD/ Pregnancy Prevention – Evaluation Issues and Results, at the 126th Annual Meeting of the American Public Health Association, November, 1998, in Washington, DC. The data were obtained from a project titled A Multicomponent School-Based HIV/STD and Pregnancy Prevention Program for Adolescents, funded by the Centers for Disease Control and Prevention.
PY - 2001
Y1 - 2001
N2 - For many large-scale behavioral interventions, random assignment to intervention condition occurs at the group level. Data analytic models that ignore potential non-independence of observations provide inefficient parameter estimates and often produce biased test statistics. For studies in which individuals are randomized by groups to treatment condition, multilevel models (MLMs) provide a flexible approach to statistically evaluating program effects. This article presents an explanation of the need for MLM's for such nested designs and uses data from the Safer Choices study to illustrate the application of MLMs for both continuous and dichotomous outcomes. When designing studies, researchers who are considering group-randomized interventions should also consider the features of the multilevel analytic models they might employ.
AB - For many large-scale behavioral interventions, random assignment to intervention condition occurs at the group level. Data analytic models that ignore potential non-independence of observations provide inefficient parameter estimates and often produce biased test statistics. For studies in which individuals are randomized by groups to treatment condition, multilevel models (MLMs) provide a flexible approach to statistically evaluating program effects. This article presents an explanation of the need for MLM's for such nested designs and uses data from the Safer Choices study to illustrate the application of MLMs for both continuous and dichotomous outcomes. When designing studies, researchers who are considering group-randomized interventions should also consider the features of the multilevel analytic models they might employ.
UR - https://www.scopus.com/pages/publications/0035535615
UR - https://www.scopus.com/pages/publications/0035535615#tab=citedBy
U2 - 10.1207/S15327906MBR3602_03
DO - 10.1207/S15327906MBR3602_03
M3 - Article
C2 - 26822108
AN - SCOPUS:0035535615
SN - 0027-3171
VL - 36
SP - 185
EP - 205
JO - Multivariate Behavioral Research
JF - Multivariate Behavioral Research
IS - 2
ER -