The quantitative aspects of this work have rested most heavily on U.S. Census Bureau publications, particularly the reports of the fifteenth and sixteenth censuses. The lack of comparability between the classification systems used for the two censuses presented a number of difficulties. The absence of separate reporting of Hispanics in 1940 made it impossible to document fully the Mexican-American experience through the Depression. Other differences between the two censuses could be minimized by a readjustment of categories, but it should be recognized that no amount of juggling will make the 1930 and 1940 counts entirely comparable.
The differences in the enumeration of workers and of occupations between the two censuses of workers and of occupations between the two censuses are explored fully in the U.S. Bureau of the Census, Sixteenth Census of the United States, 1943: Population, vol. 2, Comparative Occupation Statistics for the United States. In this volume, prepared by Alba M. Edwards, changes both in classification and in the instructions to enumerators are discussed.
The most important single difference between the two censuses is the replacemerit of the "gainful worker" concept that had been used in 1930 with the "labor force," a term that applies beginning in 1940. In 1930 census enumerators were instructed to count as a gainful worker each person ten years old or older who was usually employed in a specific occupation, even if the individual was not employed or seeking work at the time of enumeration. The instructions to enumerators in 1940 singled out the last week of March, 1940, as the period that determined labor- force status. All persons aged fourteen years and older who were either employed or seeking work were counted as labor-force members.
Specific groups included in the 1930 count were deleted from the 1940 count. Inmates in institutions who worked at a specific occupational task were counted as workers in 1930 but not in 1940. Seasonal workers were more likely to be counted in 1930 than in 1940 since they were included in 1940 only if they were at work, were seeking work, or were on temporary layoff during the last week of March. Persons with no previous work experience who were seeking work were enumerated in 1940 but not in 1930. With respect to the effect of the 1940 procedural changes on the enumeration of women, the Census Bureau estimated that some women may have been recorded as paid housekeepers in 1930 who were full-time housewives not earning wages.
In an attempt to minimize the differences between the two censuses, I calculated work rates controlling for age, but calculations controlling simultaneously for age and for occupation could not be made. in 1940 the categories into which specific occupations were grouped were also changed from 1930. In comparing 1930 and 1940, I also rearranged the classifications of specific occupations in larger occupational or industrial categories used in 1940 to conform to the 1930 categories.