This chapter discusses the usefulness of data on height for the analysis of the impact of long-term changes in nutritional status and health on economic, social, and demographic behavior. In this chapter, measures of height are used for two related purposes. First, mean height at specific ages is used as a measure of the standard of living. When used in this way, data on height supplement other evidence such as indexes of real wages, estimates of per capita income, and measures of food consumption. One advantage of height data is their abundance and wide coverage of socioeconomic groups. Consequently, it is possible to develop continuous series for a wide range of geographical areas as well as for quite refined occupational categories. It is also possible to develop far more refined measures of the extent to which particular classes and areas were affected by changes in economic fortunes than has so far been possible through the use of either real wage indexes or measures of per capita income. Although it is unlikely that large bodies of height-by-age data will be uncovered for periods before 1700, the data are abundant from 1700 on, adding more than a century to most of the current series on per capita income. The wide geographical and occupational coverage offers the possibility that aggregate indexes constructed from them will be more representative of national trends than are long-term wage indexes, which are composed of narrow, discontinuous series.

PRINCIPAL SAMPLES AND PROCEDURES

The chapter is based on a set of thirteen samples of data containing information on height-by-age and various socioeconomic variables that cover the period from 1750 through 1937 for the United States, Trinidad, Great Britain, and Sweden. Six of the samples are from U.S. military records for the period from 1750 to 1900. The other three U.S. samples contain information on both sexes: the sample of coastwise manifests provides information about slaves who boarded coastwise vessels between 1810 and 1863; the Fall River survey covers working children of school age during 1906–1907; and the cost of living survey covers all family members in a sample of households from 1934 to 1937. The Trinidad data set consists of complete censuses of the slaves on the island in 1813, with updates in 1814 and 1815, and then every three years until 1834.