Modeling Longitudinal Multiple-Group Data: Practical Issues, Applied Approaches, and Specific Examples

Modeling Longitudinal Multiple-Group Data: Practical Issues, Applied Approaches, and Specific Examples

Modeling Longitudinal Multiple-Group Data: Practical Issues, Applied Approaches, and Specific Examples

Modeling Longitudinal Multiple-Group Data: Practical Issues, Applied Approaches, and Specific Examples

Synopsis

This book focuses on the practical issues and approaches to handling longitudinal and multilevel data. All data sets and the corresponding command files are available via the Web. The working examples are available in the four major SEM packages--LISREL, EQS, MX, and AMOS--and two Multi-level packages--HLM and MLn. All equations and figural conventions are standardized across each contribution. The material is accessible to practicing researchers and students. Users can compare and contrast various analytic approaches to longitudinal and multiple-group data including SEM, Multi-level, LTA, and standard GLM techniques. Ideal for graduate students and practicing researchers in social and behavioral sciences.

Excerpt

Kai Uwe Schnabel Max Planck Institute for Human Development, Berlin

Todd D. Little Yale University

Jürgen Baumert Max Planck Institute for Human Development, Berlin

Both longitudinal and multilevel designs can provide invaluable empirical evidence for many, if not most, of the central assertions made by theories in the social and behavioral sciences (Baltes & Nesselroade, 1979; Menard, 1991). However, because such designs often involve considerable investments of time and money, their use is only justifiable if the resulting data can be analyzed adequately and thereby represented clearly. More often than not, however, disillusionment is part of the researchers' experience after the longitudinal or multilevel data are collected. For example, the puzzling number of different ways to analyze longitudinal data is likely to frustrate many researchers who neither consider themselves experts in statistics nor intend to become one. And multilevel designs also offer a plethora of complexities when it comes to decomposing the various sources of variability in the participants' responses.

Although researchers in the behavioral and social sciences are quite sophisticated when it comes to methodological and statistical issues, keeping up with the rapid advances and understanding the inherent complexities in the various analytic techniques for addressing longitudinal and multilevel data can be daunting. Such frustrations are made even more salient because more and more research questions lead researchers to use increasingly sophisticated longitudinal or multilevel designs. This volume is targeted to those researchers who wish to understand the practical issues and to learn from actual applications to address such data.

Many practical and theoretical issues are involved when addressing longitudinal and multilevel data. Some of these issues include (a) what information should be . . .

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.