Outgoing Rotation Group/Earner Study User Notes
The outgoing rotation group questions, also known as the earner study, began in 1979. Households in the CPS are interviewed for four months, not interviewed for 8 months, and then interviewed again for 4 more months. Households that are interviewed for the fourth month or eighth month (that is, the households that are about to rotate out of interviews for eight months or indefinitely) are asked additional labor questions. The universe of these questions, in addition to being month-in-sample 4 or 8, includes only civilians age 15 and older who are currently employed as a wage or salaried worker (that is, not self-employed). The earner study questions concern the respondent's periodicity of pay, hourly wage, usual weeks worked per year at that rate, usual hours worked a week, and overtime pay. These work and income questions differ from ASEC variables because the ASEC questions use a reference period of the last week or last year, not current pay or usual hours.
The outgoing rotation group households, which comprise roughly one fourth of every basic monthly sample for a calendar year (months-in-sample 4 and 8), are extracted and merged together by the National Bureau of Economic Research (NBER) and are released as the Merged Outgoing Rotation Group file.
Researchers tend to use the outgoing rotation groups variables because the pooled sample contains almost three times the number of observations as the ASEC, which has similar variables. Also, matching current individual characteristics with the current income and work responses is sometimes more appropriate than matching current characteristics with last year's pay and work conditions. The earner study also includes a variable that provides information about union membership, which is not asked in the ASEC.
Weighting
Researchers should use the earnings weight EARNWT for the outgoing rotation group/earner study variables. According to the NBER CPS Merged Outgoing Rotation Groups documentation, the sum of EARNWT for each month is the sum of the total population, so weights will need to be divided by 12 when pooling over one year.
Users should note that the universe of EARNWT does not always match with the universe of the outgoing rotation group questions, also known as the earner study. The universe of the outgoing rotation group questions is civilians age 15 and older who are currently employed. In all CPS basic monthly and ASEC samples, employed 15-year-olds are eligible to answer questions in the earner study, including HOURWAGE. In the 1998-2010 ASEC samples, and in basic monthly samples from 1998 and later, the universe of EARNWT is employed persons in outgoing rotation groups age 16 and older. In those affected samples, 15-year-olds who are employed and in the outgoing rotation group survey will effectively be dropped from your analysis because of their zero weight in EARNWT. The documentation of the universe of EARNWT and the earner study variables accurately reflects the available data.
Users should note that households occur twice within a 2 year period, which will affect standard errors when pooling outgoing rotation groups for more than one year. The earnings topcodes for the outgoing rotation responses are also lower than the ASEC earnings responses.
Supported By
