ERIC Number: ED519763
Record Type: Non-Journal
Publication Date: 2011
Reference Count: 9
Under What Assumptions Do Site-by-Treatment Instruments Identify Average Causal Effects?
Reardon, Sean F.; Raudenbush, Stephen W.
Society for Research on Educational Effectiveness
The purpose of this paper is to clarify the assumptions that must be met if this--multiple site, multiple mediator--strategy, hereafter referred to as "MSMM," is to identify the average causal effects (ATE) in the populations of interest. The authors' investigation of the assumptions of the multiple-mediator, multiple-site IV model demonstrates that such models rely on a large number of non-trivial assumptions. Of most importance are the assumptions regarding the relationship between compliance and effect. Any correlation--whether within or between sites--between compliance (with respect to any mediator) and effect (again, of any mediator) will potentially bias the estimates of the effect of any mediator. Because individuals or groups (sites) with the most to gain from a specific practice (mediator) may be more likely to comply with treatment assignment, there is good reason to worry that compliance-effect covariances are non-zero in many cases.
Descriptors: Educational Research, Social Science Research, Research Design, Causal Models, Compliance (Psychology), Context Effect, Measures (Individuals)
Society for Research on Educational Effectiveness. 2040 Sheridan Road, Evanston, IL 60208. Tel: 202-495-0920; Fax: 202-640-4401; e-mail: email@example.com; Web site: http://www.sree.org
Publication Type: Reports - Evaluative
Education Level: N/A
Authoring Institution: Society for Research on Educational Effectiveness (SREE)