28. Design and Analysis of Observational Studies

  1. Bradley E. Huitema

Published Online: 14 OCT 2011

DOI: 10.1002/9781118067475.ch28

The Analysis of Covariance and Alternatives: Statistical Methods for Experiments, Quasi-Experiments, and Single-Case Studies, Second Edition

The Analysis of Covariance and Alternatives: Statistical Methods for Experiments, Quasi-Experiments, and Single-Case Studies, Second Edition

How to Cite

Huitema, B. E. (2011) Design and Analysis of Observational Studies, in The Analysis of Covariance and Alternatives: Statistical Methods for Experiments, Quasi-Experiments, and Single-Case Studies, Second Edition, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9781118067475.ch28

Author Information

  1. Department of Psychology, Western Michigan University, Kalamazoo, Michigan, USA

Publication History

  1. Published Online: 14 OCT 2011
  2. Published Print: 14 OCT 2011

Book Series:

  1. Wiley Series in Probability and Statistics

Book Series Editors:

  1. Walter A. Shewhart and
  2. Samuel S. Wilks

ISBN Information

Print ISBN: 9780471748960

Online ISBN: 9781118067475

SEARCH

Keywords:

  • analysis of covariance (ANCOVA);
  • observational studies;
  • propensity score methods;
  • propensity-based approaches;
  • random assignment;
  • randomized design

Summary

This chapter begins with a brief review of the randomized design, the conditions under which unbiased estimates can be obtained in the absence of random assignment, and a description of nonrandomized designs and observational studies. Rubin’s causal model and methods for designing observational studies that are consistent with this model are described next. A robust blocked version of analysis of covariance (ANCOVA) design and analysis approach is an alternative to applying some version of multiple ANCOVA. Certain practical advantages such as the transparency of the analysis and robustness to flaws in the model are associated with the propensity based analysis. Finally, new methods for the analysis of observational studies that have been designed using propensity scores are described and illustrated using data from a large observational study.

Controlled Vocabulary Terms

analysis of covariance; completely randomized design