How Are Immigrants Faring After Welfare Reform?. Endnotes


2.  Los Angeles County encompasses a number of cities, including Los Angeles City, Glendale, Inglewood, Long Beach, Pico Rivera, Pomona, the San Fernando Valley and Santa Monica. The five boroughs of New York City are Manhattan, Bronx, Brooklyn, Queens, and Staten Island.

3.  Both Los Angeles and New York City are important entry points for immigrants, who often migrate to other areas of the country. Thus, the immigrants in these areas may have entered more recently than, say, immigrants in the midwest or deep south areas of the United States and might have a different set of experiences. Moreover, the nationalities of immigrants vary across the nation, e.g., the large proportion of Cubans in Florida, which also affects many components of their social and economic status.

4.  In 1997, SSI eligibility was restored to immigrants who were already on SSI before August 1996 or who were in the United States on that date and subsequently became disabled. In addition, Amerasians and Cuban-Haitian entrants were to be considered similar to refugees, so that they would also be given a grace period for benefits. In 1998, food stamp eligibility was restored to children, some elderly people, and disabled immigrants who were living in the United States before August 1996. American Indians born in Mexico or Canada and Hmong immigrants were also accorded a grace period.

5.  Some immigrants, such as refugees, diversity immigrants and most immigrants admitted under employment preferences, do not have sponsors, so these rules do not apply to those groups.

6.  In 2001, the poverty line for a family of three is $14,630. In 1998, it was $13,650.

7.  Legal challenges and court decisions kept most provisions of Proposition 187 from being implemented. In 1999, Governor Davis ceased efforts to implement the final provisions.

8.  A 1997 review of 46 RDD surveys concluded that "response rates in RDD surveys over 70 percent are the exception and not the rule." The average response rate for the 46 surveys was 62 percent, and only one-sixth of the surveys had response rates above 70 percent (Massey et. al. 1997).

9.  If a person who spoke another language was contacted, the interviewers attempted to conduct the survey in one of the languages available (generally English). In a few cases, it was not possible to conduct an interview in any of the languages.

10.  We exclude from this definition the small number of native citizens born abroad, who are considered native-born citizens because their parents were citizens when they were born and because they have elected to retain American citizenship.

11.  The respondent decides whether to report someone as being his or her nonmarital partner.

12.  In this report, "children" means less than 18 years old; offspring who were 18 or older were categorized as adults.

13.  When there were two parents with different status, we assigned status using a hierarchy in which being undocumented was the highest priority and being a native citizen was the lowest. That is, if one parent was an LPR and the other was undocumented, we classified the family as "undocumented." We designed the hierarchy in this order to best capture mixed-status families in which at least one parent is ineligible for benefits due to immigration status.

14.  Because of the small sample sizes, in the remaining tables we do not include data based on groups or categories for which there are fewer than 50 unweighted observations, as shown in Table 1.1. (In some tables, the unit of analysis is the family, with a different count of unweighted observations.) This is to reduce the risk of drawing inferences from samples that are too small to be statistically reliable.

15.  In subsequent reports, we may be able to refine these estimates somewhat by seeing whether recipients were eligible under federal rules.

16.  The Urban Institute has defined "low-income" as income below 200 percent of the federal poverty level.

17.  Since NSAF II did not collect complete data on elderly households, we exclude households with no adult under the age of 65 from the LANYCIS data in Table 1.2.

18.  The survey data for TANF participation may be underestimated in New York in both NSAF II and LANYCIS. Comparisons of administrative and survey data for NSAF show particular shortfalls in TANF participation. The reason is not clear.

19.  The primary factor driving the increase in naturalization applications was a record number of legal permanent resident admissions in the early 1990s, due in part to the legalization of 2.7 million undocumented immigrants under the amnesty provisions of the Immigration Reform and Control Act of 1986 (IRCA). Legal immigrants are eligible to naturalize five years after their admission to permanent residence. In fact by 1999, naturalization applications fell to 760,000 because most of the IRCA amnesty population had already naturalized. Thus, restrictions on noncitizen eligibility for public benefits explain only a fraction of the rise in naturalization applications. See generally, Fix and Passel 2002.

20.  A related possibility is that we may have incorrectly imputed either immigration status or Medicaid participation. As described in Appendix 2, we imputed some data in these areas and may have erred in a few cases.

21.  To ascertain the periods of participation, respondents were asked if anyone in their family received food stamps any time in 1996-97, any time in 1998 or after, and at the time of the interview. Thus, someone who answered 'yes' to the first question, but 'no' to the second two, would be counted as having lost food stamps by 1998. See Appendix 4 for details.

22.  A few respondents reported that their food stamp allotments were reduced, but could not say when this happened. To be conservative, we present this as a benefit reduction between 1998 and the interview date, but it seems likely that in some cases it happened earlier.

23.  The proportion of the sample who are immigrants arriving after 1996, for all ages, is 8 percent in Los Angeles and 11 percent in New York City.

24.  These are measures of total gross income, before taxes and including welfare, SSI, or other cash transfer payments. See Appendix 2 for more discussion about editing income data.

25.  Comparable or higher poverty rates among post-enactment immigrants exist despite the new statutory provisions that sought to limit the entry of post-enactment low-income immigrants into the country.

26.  A small portion of the difference in income levels of pre- and post-enactment immigrants might be their eligibility for cash assistance programs like TANF or SSI. Since participation rates in these programs are relatively low, however, this is a minor component of the income differential.

27.  The respondent was usually a parent but sometimes the spouse or partner of a parent.

28.  See Shetterly et al. (1996). An alternative way to assess health status might be to measure the prevalence of diagnosed chronic illnesses or disabilities (Hogan, et al. 1997). But this might be biased in the opposite direction: if immigrants have less access to health services, they are probably less likely to be diagnosed with illnesses, regardless of their actual clinical status.

29.  LANYCIS asked about health insurance at the time of the survey, while the CPS asked whether people had insurance at any time during the prior year. When a LANYCIS respondent reported that a family member had more than one type of insurance, the type of insurance was assigned by a hierarchy with Medicaid at the top. For example, people with job-based insurance and Medicaid as well as people with both Medicare and Medicaid were classified as having Medicaid. This hierarchy was designed to best capture the share of the sample receiving Medicaid, one of the benefits affected by the immigrant eligibility restrictions in welfare reform.

View full report


"report.pdf" (pdf, 2.23Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®