The Bureau of the Census recently completed an expansion of the sample for its Current Population Survey, the source of the monthly statistics on total employment and unemployment and of periodic data on income, migration, education, family status, and many other social and economic characteristics of the population. The over-all sample size was increased by approximately two-thirds, from the previous monthly average of 21,000 interviewed households to an average of 35,000 interviewed households.¹ The former sample was located in 230 areas, comprising 453 countles and independent cities in the United States. The expanded sample is spread over 330 areas, comprising 638 counties and independent cities, with some coverage in every State and the District of Columbia. All of the 230 areas in the previous sample and all but about 1,500 of the current panel of households in that sample continued as part of the expended sample.

The first official results from the expanded 330-area sample—those for the month of May 1956—were announced on June 8 in the Bureau's "Monthly Report on the Labor Force" (Current Population Reports, Series P-57, No. 167). Comparable estimates for May from the 230-area sample were also presented for a few major items. Differences between the estimates from the two samples were small and within the range to be expected as a result of sampling variability alone. Thus, for all practical purposes, the data from the expanded sample starting in May 1956 can be used as a continuous series with the statistics from the 230-area sample for previous periods.

Detailed comparisons of the results from the two samples are presented later in this report, following a discussion of the objectives of the expansion and the methods used in achieving it.

__________
¹ Approximately 35,000 households are interviewed in a typical month in the expanded sample. Another 1,500 occupied households, on the average, are visited without obtaining interviews, because the occupants are not found at home after repeated calls or are unavailable for some other reason. There are also 5,500 sample units in an average month which are visited but found to be vacant or otherwise not to be enumerated.