Sensitivity Analysis for Hierarchical Models Employing T Level-1 Assumptions

Published in: Journal of Educational and Behavioral Statistics, v. 27, no. 2, Summer 2002, p. 181-222

Posted on RAND.org on January 01, 2002

by Michael Seltzer, John Novak, Kilchan Choi, Nelson Lim

Read More

Access further information on this document at jeb.sagepub.com

This article was published outside of RAND. The full text of the article can be found at the link above.

Much work on sensitivity analysis for hierarchical models (HMs) has focused on level-2 outliers (e.g., in multisite evaluations, a site at which an intervention was unusually successful). However, efforts to draw sound conclusions concerning parameters of interest in HMs also require that we attend to extreme level-1 units (e.g., a person in the treatment group at a particular site whose post-test score [y[subij]] is unusually small vis-a-vis the other members of that person's group). One goal of this article is to examine the ways in which level-1 outliers can impact the estimation of fixed effects and random effects in HMs. A second goal is to outline and illustrate the use of Markov Chain Monte Carlo algorithms for conducting sensitivity analyses under t level-1 assumptions, including algorithms for settings in which the degrees of freedom at level 1 (v[sub1]) is treated as an unknown parameter.

This report is part of the RAND Corporation external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.