Pitfalls of Using Proxy Variables in Studies of Guns and Crime

8 Pages Posted: 29 Nov 2003

See all articles by Carlisle E. Moody

Carlisle E. Moody

College of William and Mary - Department of Economics; Crime Prevention Research Center

Thomas B. Marvell

Justec Research

Date Written: November 10, 2003

Abstract

The magnitude of the estimated coefficient on a proxy variable in a regression is only useful in determining sign and significance. Two recent papers on the relationship between guns and crime incorrectly state that such coefficients have important policy implications.

Keywords: Crime, guns, proxy variables

JEL Classification: K42, C59

Suggested Citation

Moody, Carlisle E. and Marvell, Thomas B., Pitfalls of Using Proxy Variables in Studies of Guns and Crime (November 10, 2003). Available at SSRN: https://ssrn.com/abstract=473661 or http://dx.doi.org/10.2139/ssrn.473661

Carlisle E. Moody (Contact Author)

College of William and Mary - Department of Economics ( email )

Williamsburg, VA 23187-8795
United States
757-221-2373 (Phone)
757-221-1175 (Fax)

Crime Prevention Research Center ( email )

Missoula, MT
United States

Thomas B. Marvell

Justec Research ( email )

155 Ridings Cove
Williamsburg, VA 23185
United States
757-229-3531 (Phone)

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
319
Abstract Views
5,145
Rank
172,528
PlumX Metrics