The Current Population Survey (CPS) identifies the foreign born population through questions on country of birth and citizenship. However, the CPS does not differentiate among legal immigrants, undocumented migrants, and legal temporary residents. CPS data also tend to overstate the number of naturalized citizens, especially among recent arrivals, in comparison with official administrative data on naturalizations. For this analysis, the reporting of naturalization has been corrected and two groups not eligible to naturalize— unauthorized migrants and legal temporary migrants (or “non-immigrants”)— have been identified in the data so that they can be excluded. The resulting comparisons of naturalized citizens and legal aliens (or “legal permanent residents”) are thus more accurate and precise.

Legal permanent residents (LPR) aged 18 and over who have been in the U.S. for more than 5 years (based on the question, “When did you come to live in the U.S.?”) are classified as “eligible” to naturalize. In addition, LPR adults who have been in the U.S. 3 to 5 years and are married to a U.S. citizen are also “eligible.” Adult legal aliens who have been in the U.S. less than 6 years are part of the “soon to be eligible” population. LPRs ages 13 to 17 will become eligible to naturalize in 5 years and are also part of the “soon to be eligible” population.

The techniques employed to augment the data were developed initially at the Urban Institute by Passel and Clark.4 First, the number of unauthorized migrants included in the CPS is estimated using residual methods. In this process, the CPS data are first corrected for over-reporting of naturalized citizenship on the part of recently-arrived aliens.

Then, persons entering the U.S. as refugees and with certain kinds of temporary visas (including students, diplomats, and “high-tech guest workers”) are identified in the survey and assigned an immigration status using information on country of birth, date of entry, occupation, education, and various family characteristics.

Individuals who are definitely legal and those who are potentially unauthorized are then identified in the CPS (based on state of residence, age, sex, occupation, country of birth, and date of entry). Finally, using probabilistic methods, enough are selected and assigned to be unauthorized so as to agree with the residual estimates of legal and unauthorized immigrants.

The last step involves a consistency edit to ensure that the family structure of both legal and unauthorized populations “make sense.” The whole process requires several iterations to produce survey-based estimates that agree with the demographically-derived population totals. For the 2000–2005 CPS datasets, the populations of legal immigrants and unauthorized migrants are adjusted to account for CPS omissions, as measured in the residual estimation process.