ckA {evident}R Documentation

Matching the Minimum Wage Data

Description

The data are Card and Krueger's (1994) minimum wage case study, prior to matching for CHAIN and HRSOPEN. The data illustrate matching to form three groups using Bikram Karmakar's approxmatch package. A similar matched data set after matching is in ck in this package: ck is a matched subset of all of the data in ckA.

Usage

data("ckA")

Format

A data frame with 351 observations on the following 9 variables.

SHEET

Identifier that may be used to link the limited data here to the full data set provided by Card and Krueger.

HRSOPEN

Number of hours the restaurant was open in February 1992

CHAIN

Restaurant chain, Burger King, Wendy's, KFC and Roy Rogers

chain1

Binary indicator, chain 1

chain2

Binary indicator, chain 2

chain3

Binary indicator, chain 3

FTE

Full time employees, Februrary 1992, before the increase in the minimum wage in New Jersey (NJ).

FTE2

Full time employees, November 1992, after the increase in the minimum wage in New Jersey (NJ).

grp

Group indicator NJhigh NJlow PA

Details

These data are used to illustrate matching to form matched triples from three groups, Pennsylvania restaurants, New Jersey restaurants with low starting wages before the wage increase, and New Jersey restaurants with high starting wages before the wage increase. The low/high wage cut is at 4.50 dollars per hour. The method illustrated is from Karmakar et al. (2019) as implemented in Karmakar's approxmatch package.

If you would like to create the match in the ck data using the ckA data, then you must install the optmach package, accept its academic license, and install the approxmatch package. The optmatch packages uses the relax code of Bertsekas and Tseng (1988). The following code will then recreate the matched data set. As the matched data is available in ck, there is no need to run this code, unless you are interested in how to use the approxmatch package.

dist<-approxmatch::multigrp_dist_struc(ckA,as.character(ckA$grp), list(mahal=c("chain1","chain2","chain3", "HRSOPEN")),wgts=1) mtch<-approxmatch::tripletmatching(dist,as.character(ckA$grp), indexgroup="PA",ckA,"CHAIN",design=c(1,1,1))

Used as an example in Rosenbaum (2021).

Using a different matched sample, an analysis of two evidence factors in Card and Krueger's study is given in Rosenbaum (2010).

Source

Card, D. and Krueger, A. B. (1994) <doi:10.1080/10618600.2019.1584900> "Minimum wages and employment: A case study of the fast food industry in New Jersey and Pennsylvania". American Economic Review, 84, 772-793.

References

Bertsekas, D. P. and Tseng, P. (1988) <doi:10.1007/BF02288322> "The Relax codes for linear minimum cost network flow problems". Annals of Operations Research, 13(1), 125-190.

Card, D. and Krueger, A. B. (1994) <doi:10.1257/aer.90.5.1397> "Minimum wages and employment: A case study of the fast food industry in New Jersey and Pennsylvania". American Economic Review, 84, 772-793.

Karmakar, B., Small, D. S. and Rosenbaum, P. R. (2019) <doi:10.1080/10618600.2019.1584900> "Using approximation algorithms to build evidence factors and related designs for observational studies". Journal of Computational and Graphical Statistics, 28, 698-709.

Rosenbaum, P. R. (2010). <doi:10.1093/biomet/asq019> "Evidence factors in observational studies". Biometrika, 97(2), 333-345.

Rosenbaum, Paul R. (2021) <doi:10.1201/9781003039648> Replication and Evidence Factors in Observational Studies. Chapman and Hall/CRC.

Examples

data(ckA)
table(ckA$CHAIN)
table(ckA$grp)

[Package evident version 1.0.4 Index]