In lieu of an abstract, here is a brief excerpt of the content:

NOTES Introduction 1. Other commentaries on the theory of segmented assimilation are numerous; examples include Smith (2002), Alba and Nee (2003), Bean and Stevens (2003), and Jaynes (2004). 2. For more extensive praise of the IPUMS, see Perlmann (2003b). 3. Generally, given the wonders of computer power (and its falling costs), I used the largest available datasets through 1970. These typically include 1 percent of the American population (but 3 percent in 1970). In the last part of chapter 1, I used the 5 percent 1980 to 1990 datasets, and elsewhere the 1 percent. For all analyses of the 2000 census, I used all the cases from the 1 percent sample as well as all the cases of immigrants from the 5 percent sample. 4. The recent censuses have added a question on ancestry, which asks respondents to identify themselves in terms of countries or regions of origin. In one sense, this question is an advance, as it gives some information on origins that extend farther back than the parental generation. Yet the flip-side of this advance is the question’s great shortcoming: it does not specify when the individual’s family immigrated—two or twelve generations back—and it is also notoriously subjective in that people often mention some of their origins and not others. See Lieberson and Waters (1988) and Perlmann and Waters (2003). 5. For more detail, and a critique of CPS coding of the origin question, see Perlmann and Waters (2003, Introduction). 6. The samples included center on the year 2000, at least in terms of the number of subsamples drawn from before and after that year: one rotation each from 1998 and 1999 and two rotations from 2001. I relied on a version of NOTES the CPS datasets originally combined by Roger Waldinger and his colleagues at UCLA and kindly provided to me by them. 7. Also, the CPS is the result of a highly complex design; see, for example, Hicks (1997). 8. Fortunately, for the first time in many decades, the census asked immigrant respondents the specific year in which they arrived in the United States, rather than ranges of years. The appendix provides a full description and evaluation of the proxy. Chapter 1 1. Moreover, the story was sufficiently complex and the sample size sufficiently small, even in giant datasets, to suggest the aggregation into one group. 2. Four censuses provide useful mother-tongue data: 1910, 1920, 1940, and 1970. In identifying the SCEN, I used that data to supplement birthplace information by applying two rules to immigrants born in central or eastern Europe. First, those whose mother tongue was Yiddish or German were excluded from the SCEN (unless the German speakers had been born in Russia ); and second, those whose mother tongue qualified them as SCEN were included in that group regardless of where in central or eastern Europe they had been born. Many tables and figures relating to the SCEN in this chapter may be found in an early working paper (Perlmann 2001b) with a more detailed ethnic classification. 3. Figures by country of origin are summarized in Carter, Olmsted, Wright, and Haines ([1989] 1997); and these as well as race or people data for the period from 1899 to 1924 are summarized in Ferenczi (1929). See also Perlmann (2001a). 4. In 1910 and 1920, three in ten young children of immigrants from central and eastern Europe are listed with English (or nothing) as their mother tongue. In 1940 fully six in ten are, and in 1970 five in ten. In 1910 and 1920 the question asked about mother tongue, in 1940 about the “language spoken in the home in childhood” and in 1970 about “language spoken in the home in childhood other than English.” In 1950 and 1960, there was no comparable question (see variable descriptions in Ruggles et al. 2005). 5. From table 1.1. The Jews comprised 14 percent and the entire SCE group comprised 58 percent of total permanent immigration from 1899 to 1924. 6. The evidence of the 1897 census of the Russian Empire strongly suggests that the percentage of Russian Jewish immigrants who gave Yiddish as their mother tongue must have been over 95 percent; the proportion of Jews from central Europe—Austro-Hungary, Rumania, and the German Empire—who did so was lower, if still probably more than 75 percent. See, for example, Joseph (1914), Kuznets (1975), and Perlmann (2001b). 7. After World War I, when Poles were most likely to report that...

Share