Design and Analysis of Time-Series Experiments

By G. Glass, Victor L. Willson, J. Gottman. 2008

564 citations recorded by the source corpus.

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.

Subfields: Education, Sociology and Political Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/068bc25654e0cd8351c0d053254a6229e980f34c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Education

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Effects of school resource officers on school crime and responses to school crime (2020, methodology): “Generally, at least 50 time points (Npre + Npost ≥ 50) are required to fit ARIMA models with adequate power (McCleary & McDowall, 2012), but fewer time points are acceptable if the intervention ES is reasonably large (Glass et al., 1975).”
A Comparison of Missing-Data Procedures for Arima Time-Series Analysis (2005, methodology): “Among the many methods available for model identification, the most common method (Glass et al., 1975) consists of examining the patterns of autocorrelations and partial autocorrelations in data.”
A Comprehensive set of Postestimation Measures to Enrich Interrupted Time-series Analysis (2017, methodology): “74 A comprehensive set of postestimation measures Linden (2015) introduced the itsa command, which estimates the effect of an intervention on an outcome variable for a single treatment group or when compared with one or more control groups.”