Premium
Using multilevel modeling to examine the effects of multitiered interventions
Author(s) -
Clements Melissa A.,
Bolt Daniel,
Hoyt William,
Kratochwill Thomas R.
Publication year - 2007
Publication title -
psychology in the schools
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.738
H-Index - 75
eISSN - 1520-6807
pISSN - 0033-3085
DOI - 10.1002/pits.20242
Subject(s) - psychological intervention , intervention (counseling) , context (archaeology) , psychology , hierarchy , multilevel model , response to intervention , computer science , machine learning , political science , psychiatry , paleontology , law , biology
Data collected in school settings are inherently hierarchical. At the same time, it is becoming increasingly common for interventions to be implemented within the context of a similar multitiered intervention framework where different interventions are provided at different levels of the hierarchy, often simultaneously. This prevalence of hierarchical structures in educational intervention research makes the use of multilevel models (MMs) an appealing methodology for the study of multitiered interventions. We provide an overview of (1) the hierarchical nature of educational intervention research, (2) relevant practical issues in the design of multitiered intervention studies, and (3) the use of MMs for providing a descriptive characterization of multitiered intervention effects. Findings from a hypothetical data example demonstrate that MMs provide noticeably more accurate estimates of effects than traditional single‐level analyses and clarify the advantages of MMs for examining data analysis issues relevant to multitiered intervention studies. Finally, the article provides recommendations for the use of MMs in future intervention studies, especially within the context of multitiered intervention models and response to intervention frameworks. © 2007 Wiley Periodicals, Inc. Psychol Schs 44: 503–513, 2007.