Research Seminar, Special Topic: Causal Inference in the Social Sciences

Statistics 711, Section 01 (Susan Murphy, samurphy@umich.edu)Sociology 897, Section 02 (Yu Xie, yuxie@psc.isr.umich.edu)Education 737, Section 02 (Steve Raudenbush rauden@umich.edu)

Time and Location:

Main Seminar: 1-3 Thursday, 4050 LSA

Group meetings: 1-3 Tuesday (it may vary depending on arrangement with individual instructors).

In this course we explore and critique methods for conducting causal inference in the social sciences. These methods will be drawn from a wide variety of disciplines, including economics, sociology, statistics, education, psychology, and epidemiology. Particular attention will be paid to causal inference from quasi-experimental and observational research designs.

This course is part of the Michigan Methodology Seminar. It provides an interdisciplinary forum for researchers and graduate students in several related disciplines at Michigan to be engaged in discussing cutting-edge issues in social science methodology.

Office Hours:

Susan Murphy: Tuesday 3-5, ISR 2038. (Tel: 763-5046)

Raudenbush: Thursday 3-5, ISR 2037. (Tel: 936-0462)

Yu Xie: Tuesday 3-5, ISR 2040. (Tel: 936-0039)

Home Page: http://www.psc.isr.umich.edu/~yuxie/897/

 

 

 

 

Course Outline:

Section 1: Introduction

Week 1 (January 6)

Topic: Introduction

Faculty Leader: Xie

Reading: Manski, C.F., and Garfinkel, I. 1992. "Introduction." Pp.1-21 in Evaluating Welfare and Training Programs, edited by Manski, Charles F. and Irwin Garfinkel. Cambridge, MA: Harvard University Press.

Week 2 (January 13)

Topic: Potential Outcomes

Faculty Leader: Murphy

Reading: Holland, P. 1986. "Statistics and Causal Inference" (with Discussion) JASA 81:945--970.

 

Section 2: Adjustment for Observables in Observational Studies

Week 3 (January 20)

Topic: Control with Observed Covariates

Faculty Leader: Raudenbush

Reading: Cochran, W. 1965. "The planning of observational studies of human populations (with discussion)." Journal of the Royal Statistical Society, A 128:234-255.

Cochran, W.G. 1957. "Analysis of covariance: Its nature and uses." Biometrics, 13:261-281.

Week 4 (January 27)

Topic: Structural Equation Models

Faculty Leader: Xie

Reading: Duncan, O. D. 1966. "Path Analysis: Sociological Examples." American Journal of Sociology 72:1-16.

Freedman, D.A. 1987. "As Others See Us: A Case Study in Path Analysis." Journal of Educational Statistics 12:101-128.

Week 5 (February 3)

Topic: Choice in Observational Studies

Faculty Leader: Murphy

Reading: Rosenbaum, P. 1999. "Choice as an Alternative to Control in Observational Studies (with discussion)." Statistical Science 14:259--304.

Week 6 (February 10)

Topic: Matching, Propensity Score, and Other Methods Compared

Faculty Leader: Xie

Reading: Winship, C.. and Morgan, S.L. "The Estimation of Causal Effects from Observational Data." Annual Review of Sociology 25:659-707.

Week 7 (February 17)

Topic: Causal Generalization in Social Science and Epidemiology

Faculty Leader: Raudenbush

Reading: Cook, T.D. 1990. "The generalization of causal connections: Multiple theories in search of clear practice. Keynote address." Research Methodology: Strengthening Causal Interpretations of Non-Experimental Data. Conference Proceedings, US Department of Health and Human Services, Washington, DC.

Kraemer, H. C., Kazdin, A. E., Offord, D. R., Kessler, R. C., Jensen, P. S., & Kupfer, D. J. 1996. "Coming to terms with the terms of risk." Archives of General Psychiatry.

Section 3: Accounting for Unobservables in Observational Studies

Week 8 (February 24)

Topic: Parametric Approach in Econometrics

Faculty Leader: Murphy

Reading: Willis, R. and Rosen, S. 1979. "Education and Self-Selection." Journal of Political Economy 87:S7-36.

[Spring Break]

 

Week 9 (March 9)

Topic: Non-parametric Methods and Bounds

Faculty Leader: Xie

Reading: Manski, C. and Nagin, C. 1998. "Bounding Disagreements about Treatment Effects: A Case Study of Sentencing and Recidivism." Pp.99-137 in Sociological Methodology, edited by Adrian Raftery. Washington D.C.: The American Sociological Association.

Guest Speaker: Dan Nagin on March 10, 12:00-2:00.

Week 10 (March 16)

Topic: Instrumental Variables in Experiments with Non-Compliance

Faculty Leader: Raudenbush

Reading: Little, Roderick J. & Yau, Linda H.Y. (1998). "Statistical techniques for analyzing data from prevention trials: Treatment of no-shows using Rubin's causal model." Psychological Methods 3(2):147-159.

Guest Speaker: Rod Little

 

Week 11 (March 23)

Topic: More on Instrumental Variables

Faculty Leader: Raudenbush

Reading: Angrist, J.D., Imbens, G.W., and Rubin, D.B. (1996). "Identification of causal effects using instrumental variables." Journal of the American Statistical Association 91:434,444-471. (With discussion).

Section 4: Longitudinal Observational Studies

Week 12 (March 30)

Topic: Causal Inference when Participants are Changing.

Faculty Leader: Raudenbush

Reading: Bryk, A., & Weisberg, H. 1977. "Use of the nonequivalent control groups design when subjects are growing." Psychological Bulletin, 84, 950-962.

Cherlin, A. J., Furstenberg, F. F., Jr., Chase-Lansdale, P. L., Kiernan, K. E., Robins, P. K., Morrison, D. R., & Teitler, J. O. (1991, 7/June). "Longitudinal studies of effects of divorce on children in Great Britain and the United States." Science 252:1386-1389.

Week 13 (April 6)

Topic: Confounding when Covariates are Time Varying

Faculty Leader: Murphy

Reading: Robins, J.M. 1987. "A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods." Journal of Chronic Disease (40, Supplement), 2:139s-161s.

Week 14 (April 13)

Topic: Marginal Structural Models

Faculty Leader: Murphy

Reading: Hernan, B. and Robins, J. Forthcoming. "Marginal Structural Models to Estimate the Causal Effect of Prophylaxis Therapy for Pneumocystis carinii Pneumonia." Epidemiology.

Section 5: Conclusion

Last Week (May 2)

Topic: Symposium on Causal Inference in the Social Sciences

Faculty Leader: Xie

Reading: None.