Review by Choice Review
Written to address the paucity of graduate-level works on longitudinal and clustered data in the health and medical sciences, Fitzmaurice and colleagues (all, Harvard Univ.) provide solid coverage of both technique and the conceptual underpinnings of the material at hand. The 17 chapters rely on real data from the field to illustrate the applications and include well-founded, practical descriptions. (These data sets can be downloaded from the book's Web site.) Issues regarding missing data are addressed along with linear, covariance pattern, linear mixed effects, and marginal models. Naturally, residual analysis and related diagnostic concerns are also explored. Most chapters include problem sets, generally requiring statistical software for their completion (with SAS recommended by the authors). Of special value are case studies throughout the book, effectively illustrating the presented techniques and concepts. Appendixes include a brief review of essential matrix techniques, properties of expectations and variances, and critical points for a 50:50 mixture of chi-squared distributions. The authors' decision to include introductory comments at the start of each chapter and descriptive comments of sources for additional readings for each chapter lends additional solidity to the presentation. ^BSumming Up: Recommended. Graduate students through professionals. N. W. Schillow Lehigh Carbon Community College
Copyright American Library Association, used with permission.
Review by Choice Review
