QUANTITATIVE ANALYSES: DATA AND METHODS
In trying to understand the Crown Mill workers, I sought to learn as much as possible about the individuals within their community. Only then was it possible to fathom the complex ways in which family relationships, gender, age, and income shaped their collective response to textile work and mill-village life. One reason local analyses are so valuable is that they make it possible to examine the interaction between individual and group behavior. They offer the statistical data necessary to intersect private circumstances with community action. Quantification is the art of using that data to re-create an accurate and detailed portrait of a community. Quantification need not turn complex historical actors into lifeless statistics. It can instead breathe life into a richly textured past that, lacking data and quantitative methods, would remain forever buried. There are a wide variety of ways to use statistics to study community. This appendix describes the data on individuals that was available to me as I sought to understand the Crown Mill community, and it discusses some of the ways I used the data to re-create the evolution of that community.
My understanding of Crown's millhands in the early twentieth century, especially as presented in chapter 5, is based largely on data collected from two different sources: the federal manuscript census schedules for the years 1900 and 1910 and the biweekly payroll records of the Crown Cotton Mill. Manuscript census data is a gold mine of information on the lives of ordinary people, but its usefulness is naturally circumscribed by the years under study. Crown was not operating in 1880, so the census from that year was useful only in tracing mill workers to their pre-Crown locations. Virtually all of the 1890 census records, including those for Georgia, were long ago destroyed by fire. The 1920 census was not open to the public at the time I conducted this study, although by the time this book is in print, it will be. As for the Crown Mill payroll records, there are substantial gaps in coverage, but the surviving volumes provided information for a variety of different analyses. For linking payroll data to the census data, I used the payrolls from 1900 and 1912 (the 1910 payroll was not available). For an investigation into the origins of the first Crown Mill workers, I used data from the 1888 and 1890 payrolls. To examine the evolution of wage rates and labor turnover, I used data from the following payroll years: 1890, 1896, 1902, 1908, 1915, 1922, and 1927. (For an explana-