Content area
Full Text
Structural equation modeling: Present and future. A Festschrift in Honor of Karl Joreskog. Edited by R. Cudeck, S. H. C. du Toit & D. Sorbom. 2001. Paper $40. Lincolnwood, Illinois: Scientific Software International, Inc. (URL: http://www.ssicentral. com). ISBN 0 89498 049 1.
It would be difficult to understand and appreciate much research in psychology without having a good understanding of the statistical concepts that have been central to the ongoing contribution that Karl Joreskog has made to quantitative research. Our understanding of concepts such as reliability, validity, factor analysis, measurement error, nonlinear effects, multi-group factorial models, latent variable models with ordinal variables, to name but a few, have all been influenced by his work. He is probably best known for the LISREL model and the associated computer program. Among the contributors to this festschrift in honour of Karl Joreskog, on his sixty-fifth birthday, are many of the `usual suspects'.
The book is divided into four section: (a) History and Perspectives; (b) Robustness, Reliability, Fit Assessment; (c) Repeated Measures, Experimental Designs; (d) Ordinal Data, Interaction Models. However, this does not adequately convey the rich diversity contained in the book's 22 chapters. Within each of these broad sections there is a veritable feast of information for the substantive researcher.
Dag Sorbom, the co-author of the LISREL software, opens the first chapters with some background information on Karl Joreskog and the development of the LISREL program. I particularly like the story of Karl being asked why he had added the goodness-of-fit index into the LISREL program. This pedagogical approach is taken in a number of other chapters. Gosta Hagglund gives a fascinating overview of how Joreskog's work fits into the history of factor analysis and how indeed one might `speak...