On this Picostat.com statistics page, you will find information about the DrinksWages data set which pertains to Elderton and Pearson's (1910) data on drinking and wages. The DrinksWages data set is found in the HistData R package. You can load the DrinksWages data set in R by issuing the following command at the console data("DrinksWages"). This will load the data into a variable called DrinksWages. If R says the DrinksWages data set is not found, you can try installing the package by issuing this command install.packages("HistData") and then attempt to reload the data. If you need to download R, you can go to the R project website. You can download a CSV (comma separated values) version of the DrinksWages R data set. The size of this file is about 2,098 bytes.
Elderton and Pearson's (1910) data on drinking and wages
In 1910, Karl Pearson weighed in on the debate, fostered by the temperance movement,
on the evils done by alcohol not only to drinkers, but to their families.
The report "A first study of the influence
of parental alcholism on the physique and ability of their offspring"
was an ambitious attempt to the new methods of statistics to bear on
an important question of social policy, to see if the hypothesis that
children were damaged by parental alcoholism would stand up to
Working with his assistant, Ethel M. Elderton, Pearson collected voluminous
data in Edinburgh and Manchester on many aspects of health, stature,
intelligence, etc. of children classified according to the drinking
habits of their parents. His conclusions where almost invariably
negative: the tendency of parents to drink appeared unrelated
to any thing he had measured.
The firestorm that this report set off is well described by Stigler (1999),
Chapter 1. The data set
DrinksWages is just one of Pearsons
many tables, that he published in a letter to The Times,
August 10, 1910.
A data frame with 70 observations on the following 6 variables, giving the number of non-drinkers (
and drinkers (
drinks) in various occupational categories (
wage class: a factor with levels
a factor with levels
the number of non-drinkers, a numeric vector
the number of drinkers, a numeric vector
weekly wage (in shillings), a numeric vector
total number, a numeric vector
The data give Karl Pearson's tabulation of the father's trades from
an Edinburgh sample, classified by whether they dring or are sober,
and giving average weekly wage.
The wages are averages of the individuals' nominal wages.
Class A is those with wages under 2.5s.; B: those with wages 2.5s. to 30s.; C:
wages over 30s.
Pearson, K. (1910). The Times, August 10, 1910.
Stigler, S. M. (1999).
Statistics on the Table: The History of Statistical Concepts and Methods.
Harvard University Press, Table 1.1
M. E. Elderton & K. Pearson (1910). A first study of the influence
of parental alcholism on the physique and ability of their offspring,
Eugenics Laboratory Memoirs, 10.
plot(DrinksWages) # plot proportion sober vs. wage | class
with(DrinksWages, plot(wage, sober/n, col=c("blue","red","green")[class]))# fit logistic regression model of sober on wage
mod.sober <- glm(cbind(sober, n) ~ wage, family=binomial, data=DrinksWages)
op <- par(mfrow=c(2,2))
par(op)# TODO: plot fitted model
Dataset imported from https://www.r-project.org.