TOPIC
This course will provide a survey of psychometric theory and applications to measurement in health and social sciences. It will cover classical test theory, generalizability theory, item factor analysis and item response theory. Special emphasis will be given to assessing reliability, construct validity and cross-population comparability of measurements. Psychometric theory is built on the proposition that responses people make to survey and health items are likely to be contaminated by random error and systematic misunderstanding. Replicate questions that are relevant to the target construct allow the estimation of the amount of error in the responses. Examples of replicate measurements include test-retest ratings as well as multiple items with overlapping content. Under certain circumstances the replicate questions provide a basis for estimating an objective measurement scale that allows comparisons between different populations using different items. The examination of the internal structure (i.e. correlations) of replicate items allows basic questions of construct validity to be examined.
During the course, Dr. Shrout will emphasize applications of psychometric principles to measurement challenges facing H&S Scholars in their own research. Scholars who have survey or psychometric data that contains replicate measurements are invited to bring these data to the lab sessions, which will comprise 1/3 to ¼ of the course. Participants will be shown how to carry out psychometric analyses using SPSS and a popular structural equation program called Mplus. To those without appropriate data, we will provide data-based exercises.
AUDIENCE
This course is designed for post-doctoral students and faculty who have basic doctoral training in social science statistics, such as that offered in programs in psychology, sociology, epidemiology and sociomedical sciences. Familiarity with regression analysis is assumed as is familiarity with a statistical package such as SAS or SPSS.
INSTRUCTOR
Patrick E. Shrout, Ph.D. is Professor of Psychology at New York University, where he teaches advanced quantitative courses including regression, psychometric theory, structural equation models and methods for the analysis of growth and change. Prior to going to NYU, Shrout was on the biostatistics faculty in the Columbia Mailman School of Public Health. He is Fellow of the American Psychological Association, Association for Psychological Science, and American Statistical Association. This year he is President of the American Psychopathological Association, for which he is organizing a meeting that focuses on causal analysis in psychopathology research. Shrout's own substantive research focuses on stress and coping, with studies that emphasize the use of daily diary methods.
REQUIRED READING
Cranford, J. A., Shrout, P. E., Iida, M., Rafaeli, E., Yip, T., & Bolger, N. (2006). A procedure for evaluating sensitivity to within-person change: Can mood measures in diary studies detect change reliably? Personality and Social Psychology Bulletin, 32(7), 917-929.
Reise, S. P., Widaman, K. F., & Pugh, R. H. (1993). Confirmatory factor analysis and item response theory: Two approaches for exploring measurement invariance. Psychological Bulletin, 114, 552-566.
Shrout, P. E. (2002). Reliability. In M. T. Tsuang & M. Tohen (Eds.), Textbook in Psychiatric Epidemiology (second ed., pp. 131-148). New York: Wiley.
Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86, 420-428.
Embretson, S. E. & Reise, S. (2000). Item response theory for psychologists. Mahwah, NJ: Erlbaum Publishers.**Suggested Only
To register
** Enrollment for this course is now closed**
This short course is open free of charge to faculty, postdoctoral fellows, and graduate students at Columbia University as well as faculty and postdoctoral fellows at other sites of the Robert Wood Johnson Health & Society Scholars (H&SS) Program. Enrollment is limited to 20; H&SS affiliates will have priority.
To register, please send an email to: chssp@columbia.edu.
