The 2003 Institute featured Drs. Willett and Singer presenting material from their Oxford University Press book:
Applied Longitudinal Data Analysis
Modeling Change and Event Occurrence
Judith D. Singer and John B. Willett, Harvard University
Here is a much-needed professional book that will instruct readers in the many new methodologies now at their disposal to make the best use of longitudinal data. This book explains how to select an appropriate method given a research question, including how to use both individual growth modeling and survival analysis. Throughout the chapters, the authors employ many cases and examples from a variety of disciplines, covering multilevel models, curvilinear and discontinuous change, in addition to discrete-time hazard models, continuous-time event occurrence, and Cox regression models. Using longitudinal data is a unique contribution to the literature on research methods and will be useful to a wide range of behavioral and social science researchers.
Judith Singer is an applied statistician whose research interests center on developing better statistical methodology for educational, psychological, and social research. Singer has written extensively about research design and educational policy, with a current emphasis on methods for analyzing longitudinal data on systematic change and event occurrence. In addition to her methodological interests, Singer conducts research on a variety of educational policy issues including special education, child care, and early-childhood education.
John Willett's work focuses on the effective application of quantitative methodology to critical substantive problems in education and the social sciences. He is particularly interested in longitudinal research design, methods for analyzing the timing and occurrence of events, and methods for the measurement of change, learning, and development.