Full Citation

Title: Does Sample-Selection Bias Explain the Industrialization Puzzle? Evidence from Military Enlistment in the Nineteenth-Century

Citation Type: Miscellaneous

Publication Year: 2015

ISBN:

ISSN:

DOI:

NSFID:

PMCID:

PMID:

Abstract: I test whether sample-selection bias explains the industrialization puzzle. It is widely believed that this decline in average stature and stature-implied standards of living in the presence of rising output and real wages was caused by the deleterious health effects of industrialization, such as increased urbanization and changes in the availability and relative price of food; but it has recently been suggested that the puzzle may be an artifact of sample-selection bias, stemming from a reliance on records of military volunteers and other self-selected samples for stature data. In this paper I test this argument by using a semiparametric sample-selection model to estimate selection-corrected trends in average stature in the United States. This estimation is based on a novel data set of my construction, consisting of military data including staturefor the birth cohorts of 18471860, linked to census data. These data are combined with similar data from the Union Army project for the birth cohorts of 18321846. Identification is aided through incorporation of voting data from the presidential elections of 1856 and 1860, which are argued to measure political motives for military enlistment. I find that the industrialization puzzleat least in the United Statesis robust to these corrections, and therefore is not an artifact of sampleselection bias. A decrease in average stature of approximately one inch in the 1830s and 1840s is present despite the correction. These results, however, do not imply that sample-selection bias is unimportant in understanding trends in average stature. On the contrary, the degree of sample-selection bias is shown to vary over time, and accounting for sample selection meaningfully alters the trend in average stature. This result shows generally that accounting for sample-selection bias can affect conclusions drawn from stature data and thus the importance of taking it into account in studying samples of stature that can be reasonably believed to not be representative of the population of interest. It also shows specifically that the decline in stature during industrialization may have been less severe than indicated by the raw data.

Url: http://aez.econ.northwestern.edu/zimran_height_selection.pdf

User Submitted?: No

Authors: Zimran, Ariell

Publisher: Northwestern University

Data Collections: IPUMS NHGIS

Topics: Methodology and Data Collection

Countries: