Design and Analysis of Time-Series Experiments

Edited by:
Gene V Glass, Arizona State University
Victor L. Willson, Texas A&M University
John M. Gottman, The Gottman Institute, Seattle, Washington

Published 2008

Hailed as a landmark in the development of experimental methods when it appeared in 1975, Design and Analysis of Time-Series Experiments is available again after several years of being out of print. Gene V Glass, Victor L. Willson and John M. Gottman have carried forward the design and analysis of perhaps the most powerful and useful quasi-experimental design identified by their mentors in the classic Campbell & Stanley text Experimental and Quasi-experimental Design for Research (1966). In an era when governments seek to resolve questions of experimental validity by fiat and the label "Scientifically Based Research" is appropriated for only certain privileged experimental designs, nothing could be more appropriate than to bring back the classic text that challenges doctrinaire opinions of proper causal analysis. Glass, Willson & Gottman introduce and illustrate an armamentarium of interrupted time-series experimental designs that offer some of the most powerful tools for discovering and validating causal relationships in social and education policy analysis. Drawing on the ground-breaking statistical analytic tools of Box & Jenkins, the authors extend the comprehensive autoregressive-integrated-moving-averages (ARIMA) model to accommodate significance testing and estimation of the effects of interventions into real world time-series. Designs and full statistical analyses are richly illustrated with actual examples from education, behavioral psychology, and sociology.

"...this book will come to be viewed as a true landmark. ... [It] should stand the test of time exceedingly well." James A. Walsh in Educational & Psychological Measurement

"Ordinary least squares estimation is usually inapplicable because of
autoregressive error... Glass, Willson, and Gottman have assembled the best approach." Donald T. Campbell