Multilevel Modelling - 9781847872739

Editors Skrondal (epidemiology, Norwegian Institute of Public Health, Oslo) and Rabe-Hesketh (educational statistics, U. of California, Berkeley) have selected 66 articles or book excerpts for inclusion in this four-volume reference. They wanted to cover the most important topics, yet were conscious that their selections needed to be accessible to social scientists; some seminal works were rejected as being too technical. They intentionally included some articles critical of multilevel modeling since, as they say in their introduction, "we think it is dangerous to view this approach as a panacea for all challenges in quantitative social science." Panacea or not, the subject has received considerable attention in recent years, because multilevel models are particularly useful when data has a "hierarchical structure with units nested in clusters" (again from the introduction). The selections are organized as follows: Volume 1 introduces linear multilevel models; volume 2 discusses inference, diagnostics and design for linear multilevel models; volume 3, multilevel generalized linear models; and volume 4, more complex models and modeling issues. Sources include such journals as American J. of Political Science, Annual Review of Public Health, J. of Epidemiology and Community Health, and International J. of Educational Research, among many others. There is no index. Annotation ©2011 Book News, Inc., Portland, OR (booknews.com)