Download this Paper Open PDF in Browser

Instrumental Variables Methods in Experimental Criminological Research: What, Why, and How?

27 Pages Posted: 14 Aug 2007  

Joshua D. Angrist

Massachusetts Institute of Technology (MIT) - Department of Economics; National Bureau of Economic Research (NBER); IZA Institute of Labor Economics

Date Written: September 2005

Abstract

Quantitative criminology focuses on straightforward causal questions that are ideally addressed with randomized experiments. In practice, however, traditional randomized trials are difficult to implement in the untidy world of criminal justice. Even when randomized trials are implemented, not everyone is treated as intended and some control subjects may obtain experimental services. Treatments may also be more complicated than a simple yes/no coding can capture. This paper argues that the instrumental variables methods (IV) used by economists to solve omitted variables bias problems in observational studies also solve the major statistical problems that arise in imperfect criminological experiments. In general, IV methods estimate the causal effect of treatment on subjects that are induced to comply with a treatment by virtue of the random assignment of intended treatment. The use of IV in criminology is illustrated through a re-analysis of the Minneapolis Domestic Violence Experiment.

Suggested Citation

Angrist, Joshua D., Instrumental Variables Methods in Experimental Criminological Research: What, Why, and How? (September 2005). NBER Working Paper No. t0314. Available at SSRN: https://ssrn.com/abstract=1006502

Joshua Angrist (Contact Author)

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

IZA Institute of Labor Economics

P.O. Box 7240
Bonn, D-53072
Germany

Massachusetts Institute of Technology (MIT) - Department of Economics ( email )

50 Memorial Drive
E52-353
Cambridge, MA 02142
United States
617-253-8909 (Phone)
617-253-1330 (Fax)

Paper statistics

Downloads
52
Rank
325,212
Abstract Views
554