This data set is from Costa et al. (1993) and it describes 21 welders and 26 potential controls.
All are men. The outcome is a measure of genetic damage;
specifically, dpc is a measure of DNA-protein cross-links. There are 3 covariates,
age, race and smoking. This tiny example is used to illustrate the concepts of
multivariate matching in Chapter 8 of Design of Observational Studies. The
example is useful because its tiny size permits close inspection of the details
of multivariate matching, but its small sample size and limited number of covariates
make it highly atypical of matching in observational studies.
Usage
data("costa")
Arguments
Format
A data frame with 47 observations on the following 6 variables.
subject
Within group ID number.
age
Age in years.
race
AA=African-American, C=Caucasian
smoker
Y=yes, N=no
welder
Y=yes/treated, N=no/control
dpc
DNA-Protein Cross-links (percent)
References
Costa, M., Zhitkovich, A. and Toniolo, P. (1993). DNA-protein cross-links in welders: molecular implications. Cancer research, 53(3), 460-463.
Rosenbaum, P. R. (2010). Design of Observational Studies.
New York: Springer. This example is discussed in Chapter 8.