THE ESTIMATION OF CAUSAL EFFECTS FROM OBSERVATIONAL DATA

Annual Review of Sociology - Tập 25 Số 1 - Trang 659-706 - 1999
Christopher Winship1, Stephen L. Morgan1
1Department of Sociology, Harvard University, William James Hall, 33 Kirkland Street, Cambridge, Massachusetts 02138;

Tóm tắt

▪ Abstract  When experimental designs are infeasible, researchers must resort to the use of observational data from surveys, censuses, and administrative records. Because assignment to the independent variables of observational data is usually nonrandom, the challenge of estimating causal effects with observational data can be formidable. In this chapter, we review the large literature produced primarily by statisticians and econometricians in the past two decades on the estimation of causal effects from observational data. We first review the now widely accepted counterfactual framework for the modeling of causal effects. After examining estimators, both old and new, that can be used to estimate causal effects from cross-sectional data, we present estimators that exploit the additional information furnished by longitudinal data. Because of the size and technical nature of the literature, we cannot offer a fully detailed and comprehensive presentation. Instead, we present only the main features of methods that are accessible and potentially of use to quantitatively oriented sociologists.

Từ khóa


Tài liệu tham khảo

Allison PD. 1990. Change scores as dependent variables in regression analysis. InSociological Methodology 1990, ed. CC Clogg, 20:93–114. Washington, DC: Am. Sociol. Assoc.

Angrist JD, 1990, Am. Econ. Rev., 80, 313

10.2307/2291054

10.2307/2291629

10.2307/2937954

10.2307/2290263

10.2307/1924332

10.2307/1924810

10.2307/2291055

Cochran WG, 1950, Experimental Design.

Cook TD, 1979, Quasi-Experimental: Design and Analysis Issues for Field Settings.

Cox DR, 1958, Planning of Experiments.

10.2307/2333046

Fisher RA, 1935, The Design of Experiments.

Garfinkel I, Manski CF, Michalopoulos C. 1992. Micro experiments and macro effects. SeeManski & Garfinkel 1992, pp. 253–73

Goldberger AS, 1991, A Course in Econometrics.

Hamilton JD, 1994, Time Series Analysis., 10.1515/9780691218632

Harris RID, 1995, Using Cointegration Analysis in Econometric Modelling.

Harvey A, 1990, The Econometric Analysis of Time Series.

10.2307/1909757

10.2307/1912352

10.2307/1164605

Heckman JJ. 1992. Randomization and social policy evaluation. SeeManski & Garfinkel 1992, pp. 201–30

10.3386/t0185

10.2307/2109936

10.2307/146178

10.2307/2290059

10.2307/2999630

10.2307/2971733

10.1111/1467-937X.00044

Heckman JJ, 1998, Am. Econ. Rev., 88, 381

10.1017/CCOL0521304539.004

10.1007/978-1-4612-4976-4_7

Heckman JJ, Robb R. 1988. The value of longitudinal data for solving the problem of selection bias in evaluating the impact of treatment on outcomes. InPanel Surveys, ed. G Duncan, G Kalton, pp. 512–38. New York: Wiley

10.2307/2971729

10.1093/0198283164.001.0001

10.2307/2289064

Holland PW, Rubin DB. 1983. On Lord's Paradox. InPrinciples of Modern Psychological Measurement: A Festschrift for Frederic M. Lord, Ed. H Wainer, S Messick, pp. 3–25. Hillsdale, NJ: Erlbaum

10.1016/0304-4076(94)90065-5

Hood WC, 1953, Studies in Econometric Method.

10.2307/2951620

10.2307/2971731

Judd CM, 1981, Estimating the Effects of Social Interventions.

Judge G, 1985, The Theory and Practice of Econometrics.

Kempthorne O, 1952, Design and Analysis of Experiments.

LaLonde RJ, 1986, Am. Econ. Rev., 76, 604

Lieberson S, 1985, Making It Count: The Improvement of Social Research and Theory.

Maddala GS, 1993, The Econometrics of Panel Data, Vols. 1, 2.

Malinvaud EB, 1970, Statistical Methods of Econometrics.

10.1017/CCOL0521444594.004

Manski CF, 1995, Identification Problems in the Social Sciences.

10.2307/2171738

Manski CF, 1992, Evaluating Welfare and Training Programs.

10.1111/0081-1750.00043

Manski CF, Pepper JV. 1998. Monotone instrumental variables: with an application to the returns to schooling. Presented at Winter Meet. Am. Sociol. Assoc., Chicago

10.2307/2290448

Marcantonio RJ, Cook TD. 1994. Convincing quasi-experiments: the interrupted time series and regression-discontinuity designs. InHandbook of Practical Program Evaluation, ed. JS Wholey, HP Hatry, KE Newcomer, pp. 133–54. San Francisco: Jossey-Bass

McKim VR, 1997, Causality in Crisis: Statistical Methods and the Search for Causal Knowledge in the Social Sciences.

10.2307/2291632

Neyman JS. 1923. On the application of probability theory to agricultural experiments. Essay on principles. Transl. DM Dabrowska, TP Speed, 1990, inStat. Sci. 5:465–80(From Polish)

10.2307/2983637

Powell JL. 1987. Semiparametric estimation of bivariate latent variables models. Working pap. no. 8704. Madison, WI: Univ. WI, Soc. Syst. Res. Inst.

10.2307/2288326

10.1016/0304-4076(88)90039-5

10.2307/2284373

10.1016/0270-0255(86)90088-6

10.1016/0898-1221(87)90238-0

Robins JM. 1989. The analysis of randomized and nonrandomized AIDS treatment trials using a new approach to causal inference in longitudinal studies. InHealth Service Research Methodology: A Focus on AIDS, ed. L Sechrest, H Freeman, A Mulley, pp. 113–59. Washington, DC: US Public Health Serv.

10.1007/978-1-4612-1842-5_4

10.2307/2981697

10.2307/2288332

10.1007/978-1-4757-2443-1

10.2307/2335942

10.2307/2288398

10.2307/2683903

10.1093/oxfordjournals.oep.a041827

10.1037/h0037350

10.2307/1164933

10.1214/aos/1176344064

10.2307/2287653

10.2307/1164617

Rubin DB, 1986, J. Am. Stat. Assoc., 83, 396

10.1016/0378-3758(90)90077-8

10.2307/2532381

10.2307/2533160

Singer B, Marini MM. 1987. Advancing social research: an essay based on Stanley Lieberson'sMaking It Count. InSociological Methodology 1987, ed. CC Clogg, pp. 373–91. Washington, DC: Am. Sociol. Assoc.

10.1111/1467-9531.271030

10.1007/978-1-4899-1292-3_1

10.1177/0049124196024003004

10.2307/2287845

Winship C. 1998. Multicollinearity and model misspecification: a Bayesian analysis. Presented at Winter Meet. Am. Sociol. Assoc., Chicago

10.1146/annurev.so.18.080192.001551