The Demography of the Hispanic Population : Selected Essays [1 ed.] 9781617356827, 9781617356803

Citation preview

The Demography of the Hispanic Population Selected Essays

A volume in The Hispanic Population in the United States Richard R. Verdugo, Series Editor

This page intentionally left blank.

The Demography of the Hispanic Population Selected Essays

edited by

Richard R. Verdugo

INFORMATION AGE PUBLISHING, INC. Charlotte, NC • www.infoagepub.com

Library of Congress Cataloging-in-Publication Data The demography of the Hispanic population : selected essays / edited by Richard R. Verdugo. p. cm. -- (The Hispanic population in the United States) Includes bibliographical references. ISBN 978-1-61735-680-3 (pbk.) -- ISBN 978-1-61735-681-0 (hardcover) -ISBN 978-1-61735-682-7 (ebook) 1. Hispanic Americans--Population. 2. Hispanic Americans--Mortality. 3. Hispanic Americans--Social conditions. 4. Hispanic Americans--Ethnic identity. 5. Older Hispanic Americans--Social conditions. 6. United States--Population. 7. United States--Emigration and immigration. I. Verdugo, Richard R. E184.S75D464 2012 973’.0468--dc23                          2011043685

Copyright © 2012 Information Age Publishing Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the publisher. Printed in the United States of America

Contents Introduction.......................................................................................... vii Richard R. Verdugo 1 Hispanics, Immigrants, and the Growth of the U.S. Population: 1850–2007.......................................................................... 1 Richard R. Verdugo 2 Contemporary Migration and the Hispanic Population in the United States......................................................................................... 31 Karen A. Woodrow-Lafield 3 The Latino Paradox............................................................................. 47 Rogelio Sáenz and Trinidad Morales 4 Mortality Among the U.S. Hispanic Population................................ 75 Robert A. Hummer and Juanita J. Chinn 5 Disappearing Hispanics? The Case of Los Angeles County, California: 1990–2000......................................................................... 95 Matt Kaneshiro, Amanda Martinez, and David A. Swanson 6 Latino Baby Boomers: New Elders for a Diverse America.............. 125 Fernando Torres-Gil, Zachary Gasoummis, and Kathleen Wilber



v

This page intentionally left blank.

Introduction Richard R. Verdugo

In the early part of the 19th century, immigrants from Western and Eastern Europe contributed greatly to the growth of the U.S population. The United States was indeed a country of immigrants. But in the years leading up to WWII, immigrants from various countries in Central and South American, and Mexico began and continue to have an important impact on the size and growth of the U.S. population. In fact, by 2050, one in three Americans will be of Hispanic origin. As a population, Hispanics are a somewhat unknown group and have not received the same research scrutiny as the African American community. Topics for future research will focus on Hispanic education, their political behavior, their labor market experiences, and of course, the demography of the Hispanic population. Moreover, there is no reason why demographic research cannot be merged with any of these or other research disciplines. Demographically, Hispanics are an interesting group for study. In terms of population growth, we might ask which of two factors will contribute most to future Hispanic population growth: fertility or immigration. What about the significance of the Hispanic Paradox? That is, why are mortality rates lower or equal to those of non-Hispanic Whites? The somewhat recent trend of Hispanics moving into parts of the United States that have had little or no history with this population is also an interesting research topic because of their impact on local institutions, such as education and the labor market. Demographic analysis of the Hispanic population is wide open, The Demography of the Hispanic Population, pages vii–ix Copyright © 2012 by Information Age Publishing All rights of reproduction in any form reserved.

vii

viii   Introduction

especially research that would merge demography with other research disciplines. For example, the movement of Hispanics into certain localities has raised questions about Hispanics taking jobs away from native-born workers, about their impact on lowering wages, etc. In education, the movement of Hispanics into nontraditional destinations has challenged educational systems that lack resources to adequately educate this population. The present volume, in reality, is an eclectic collection of demographic research. On the one hand, the first four chapters are based on traditional demographic research and focus on immigration and mortality. On the other hand, the final two chapters address topics that are on the periphery of traditional demographic research: the relationship between vital records and census data, and the status of Latino baby boomers. Regardless of the topic being addressed, each chapter makes an important contribution to the growing body of research on the Hispanic population. The Chapters In the first chapter, Richard Verdugo opens the volume by asking a simple question, “How much have Hispanics and Hispanic immigrants contributed to the growth of the U.S. population?” His analysis begins in the middle of the 19th century, 2 years after the signing of the Treaty of Guadalupe Hidalgo, which ended the Mexican American War and turned over land that now comprises the United States west of the Mississippi. Using census data from 1850 to 2000 and the 2007 American Community Survey, Verdugo finds that Hispanics and Hispanic immigrants have made a substantial contribution to the growth of the U.S. population from 1850 to 2007. In fact, this contribution was particularly significant from about the middle of the 20th century onward. From 1850 to 2007, Hispanics (both native born and immigrants) contributed about 16% to the growth of the U.S. population. In Chapter 2, Karen A. Woodrow-Lafield examines the growth of the Hispanic population based on projection assumptions and on the uncertainty of such assumptions because of many complex factors, such as the economy and U.S. policies aimed at border crossings. Woodrow-Lafield concludes that forecasters need to be sensitive to immigration policies and to the efficacy of such policies in controlling migration. Rogelio Sáenz and Trinidad Morales examine the Latino Paradox in Chapter 3. The Paradox is that despite their lower socioeconomic status, Latinos have mortality rates equal to or more favorable than non-Latino Whites. They conclude that while the Paradox is real, additional research needs to be undertaken in order to find appropriate explanations. Robert Hummer and Juanita Chinn continue the mortality debate in Chapter 4. After reviewing the literature, and paying attention to the data,

Introduction    ix

methodologies, and a number of substantive issues in an attempt at understanding Hispanic mortality patterns, Hummer and Chinn draw several important conclusions. To begin with, they note that there are mortality differences by age groupings, sex, national origin, and nativity. Researchers should be careful in generalizing their results. Of particular significance are the mortality differences among young Hispanic males. Are Hispanics disappearing? So begins an interesting study by Matt Kaneshiro, Amanda Martinez, and David Swanson in Chapter 5. Essentially, Kaneshiro et al. compare the link between vital events data and census data. Using Hispanic birth data from the 1990 to 1999 period and the 2000 census data in Los Angeles County, the authors examine the correlation between both databases. The authors conclude that U.S.-born Mexican Americans are shifting their identities when it comes to census data. In such shifts, U.S.-born Mexican Americans may shift to one of two identities: pan-Hispanic or non-Hispanic. The implications for future counts of the Hispanic population are significant. Finally, in Chapter 6, Fernando Torres-Gil, Zachary Gasoummis, and Kathleen Wilber undertake a study of Latino baby boomers. To be sure, the baby boomer population has received a considerable amount of research scrutiny. And in 2011, the first stream of this cohort will be retiring. The authors, then, study one important subgroup within the larger baby boomer cohort, a subgroup that has not been much studied—Latino baby boomers. After providing a general description of this subgroup, the authors conclude that Latino baby boomers represent a microcosm of the challenges facing the United States as it addresses the many needs of a large retiring baby boomer population.

This page intentionally left blank.

Chapter 1

Hispanics, Immigrants, and the Growth of the U.S. Population 1850–2007 Richard R. Verdugo1,2

Introduction The United States is a country of immigrants. Immigrants have made important contributions to U.S. culture, politics, and many other aspects of American life. Immigrants also have contributed to the growth of the U.S. population during the 19th and 20th centuries. Since mid-20th century, one group that has contributed to U.S. population growth has been Hispanics. In fact, by the end of the 20th century, Hispanics had overtaken the Black population as the nation’s largest ethnic-racial minority group. The growth of the Hispanic population also has affected group relations, politics, and various social institutions, especially in many of America’s cities and smaller geographic localities. Included in this laundry list of entities that are being impacted by the Hispanic population are several important The Demography of the Hispanic Population, pages 1–29 Copyright © 2012 by Information Age Publishing All rights of reproduction in any form reserved.

1

2    R. R. VERDUGO

social institutions. For example, educational systems with large numbers of Hispanic students or educational systems that have experienced dramatic increases in their Hispanic student populations face cultural, economic, and language challenges that affect the teaching and learning environment (Verdugo, 2006). Hispanics have moved into these areas because they met a need for low-skilled, low-wage labor (Verdugo, 2006). Though the growth of the Hispanic population in the United States has been a major policy concern, one important question that seems to have been neglected is, “How has the growth of the Hispanic population affected the growth of the U.S. population?” The purpose of my chapter is to begin the process of framing an answer to this question. By this I mean I provide a general descriptive analysis of U.S. population growth, Hispanic population growth, and the contribution of the latter to the former. Data for my analyses are from the 1850 to the 2000 U.S. decennial census and from the 2007 American Community Survey. In describing the impact Hispanics have had on U.S. population growth, I address three broad topics: 1. Population change in the United States: My analysis focuses on Hispanic and non-Hispanic population growth in order to ascertain how much growth has occurred among both groups. In this section, I examine the total growth of the U.S. population from 1850 to 2007 as well as the growth of the Hispanic and non-Hispanic populations. My aim is to ascertain the impact each group has had on the total growth of the U.S. population, especially the impact of the Hispanic population. 2. Growth of the Hispanic population: In my second analysis section, I look at the growth rates of two Hispanic groups: Hispanic immigrants and Hispanic ethnic groups. The aim is to evaluate the growth of these two subpopulations, because they will be important in my final analysis on how much of an impact Hispanics, and these subgroups, have had on U.S. population growth from 1850 to 2007. 3. The impact of Hispanics on U.S. population growth: In this final section, I examine the impact Hispanics (immigrants, nonimmigrants, and Hispanic ethnic groups) and non-Hispanics (immigrants, and nonimmigrants) have had on U.S. population growth. Background Before examining and discussing Hispanic population growth and its impact on the growth of the U.S. population, I review what is currently known about the three main factors of population growth—fertility, immigration, and mortality—as they apply to the Hispanic population.

Hispanics, Immigrants, and the Growth of the U.S. Population    3

Fertility among the Hispanic Population Fertility research among the Hispanic population has produced at least two important findings.3 First, scholars have been interested in the role fertility plays in the growth of the Hispanic population. Prior to the 1960s, immigration appeared to have been the main driving force behind this growth. In the following decade, immigration and fertility equally shared in the growth of the Hispanic population, and from about 1980 to the present, immigration once again took the dominant role. Recently, some scholars have concluded that in the future, fertility will drive Hispanic population growth because a large second-generation cohort of Hispanics will reach childbearing age (Passel, 2004; Passel & Cohn, 2009; Suro & Passel, 2003; Tienda & Mitchell, 2006). Secondly, research points out that fertility within the Hispanic community is a complex combination of ethnicity, social-class background, and legal status (Durand, Telles, & Flashman, 2006; Tienda & Mitchell, 2006). For instance, Durand et al. (p. 78), among others, found that total fertility (TFR) varies by Hispanic ethnicity. Mexicans have the highest TFR (3.3), and they are followed by Puerto Ricans (2.6), Central/South Americans (2.1), and Cubans (1.9). Trend data are from Child Trends (see the Child Trends Web site: http://www.childtrendsdatabank.org/indicators/79BirthRates. cfm) tend to confirm these results. Immigration Research4 Push–Pull Factors Contributing to Hispanic Immigration Researchers have long suggested that push and pull factors account for Hispanic immigration. Pull factors are those forces associated with the United States that attract immigrants. Pull factors are such things as the availability of jobs, housing, and a political climate within the United States that ensures opportunities for immigrant economic success and an improved quality of life. Scholars have been able to identify at least two periods in U.S. history that have provided such opportunities for Hispanic immigrants. The first period occurred during WWII when a significant number of working-age men were off at war and labor was needed in the agricultural fields of California and in the industrial plants of the Midwest (Barrera, 1979; Cardoso, 1980; Garcia, 1981; McWilliams, 1948; Valdes, 1982, 1991, 2000; Vargas, 1993). For example, Valdes (1982) and Vargas (1993) both describe the role Hispanics (mainly Mexicans) took in the auto industry in Detroit and in other industries of the Midwest during the early and middle decades of the 20th century.

4    R. R. VERDUGO

A second important historical period occurred during the 1980s when the U.S. was undergoing significant economic restructuring (Durand et al., 2006; Harrison & Bluestone, 1990; Huber & Solt, 2004; Rumbaut, 2006; Walton, 2004). During the economic downturn and industrial restructuring, the United States had a need for cheap, unskilled labor, and for labor that delivered special services to skilled workers (Verdugo, 2006). Griffith (1995) discusses such pull factors as work in the poultry industry in southern rural areas and vegetable production in south Florida (Griffith, 2000; see also Barrera, 1979; Calavita, 1992; Garcia, 1981). Push factors also have led to the increase of Hispanics immigrants in the United States. By this I mean factors in the sending country that pushed many of its citizens to leave. This body of research suggests two broad push factors: economic depression and political instability. Thus, harsh economic times or economies in which workers, skilled and unskilled, are unable to earn a decent living for them and their families have pushed Hispanic immigrants to the United States. An example of political instability was the Cuban Revolution that sent many Cubans packing to the United States (see Portes & Stepick, 1993).5, 6 In the 1970s and 1980s, violent and turbulent political climates in Central America were events that pushed significant numbers of its population to the United States as well as to other places in the world (Hamilton & Stoltz-Chinchilla, 2001; Menjivar, 2000; Sullivan & Pedraza-Bailey, 1979). Settlement Patterns among Hispanic Immigrants Two issues have dominated recent discussions about the settlement patterns of the Hispanic population: discussions about traditional “ports of entry,” and the “new settlement patterns.” Traditionally, Hispanics have settled in areas that were ports of entry into the United States. For the most part, these areas were located in the Southwest where significant numbers of Mexicans and South/Central Americans reside (Bean, Telles, & Lindsay, 1987; Cardoso, 1980; Grebler, Moore, & Guzman, 1970; Gutierrez, 1995; Moore & Pachon, 1985). The Northeast, particularly New York and its neighboring states, are places with sizeable Puerto Rican populations. And finally, in the Southeast, primarily Florida and the Miami area are home to a large and politically powerful Cuban population (see Portes & Stepick, 1993; Sullivan & Predraza-Bailey, 1979; Verdugo, 2006). This body of research is important because it highlights not only the geographic diversity of the Hispanic population but also the economic, political, and cultural impact Hispanics are having in a state or local community. A second strand of research has discovered new ports of entry and new settlement patterns among Hispanics, both immigrants and nonimmigrants (Frey, 1994, 2002, 2003; Frey & Liaw, 1998). This body of research

Hispanics, Immigrants, and the Growth of the U.S. Population    5

has discovered that Hispanics are moving into areas that either lack or have limited histories with Hispanics. In many cases, the presence of Hispanics in these new areas has led to race relations problems as well as to important challenges for local and state government agencies. For example, Hispanics have been moving to many nontraditional areas in the South, such as North Carolina, Georgia, and Arkansas, as well as to northeastern states such as Maine (Durand & Massey, 2003; Economic Research Service, 2005; Effland & Kassel, 1996; Griffith, 1995; Hernandez-Leon & Zuniga, 2000; Kandel & Cromartie, 2004; Stull, Broadway, & Griffith, 1995). The two driving forces behind these new settlement patterns have been jobs and affordable housing. In Georgia, a revitalized carpet industry has attracted a large number of Hispanics, especially to Dalton (Hernandez-Leon & Zuniga, 2000). The agricultural industry in Florida continues to draw Hispanic workers (Griffith, 2000). In other parts of the South, the work-related attraction has been the poultry industry (Kandel & Parrado, 2004). By contrast, further up the East Coast, the attraction has been the fish canning industry (Smith-Nonini, 2002). In the Midwestern part of the United States, meat and poultry packing have drawn Hispanics (Durand & Arias, 2000; Stull et al, 1995; Valdes, 1982, 1991, 2000). Jobs and housing have been the main reasons Hispanic have been moving into new areas of the United States, and as they move into these new areas, cultural contact has created many social challenges. Mortality Research: The Hispanic Paradox and the Salmon Bias From a purely demographic standpoint, the most important research topics on Hispanic mortality have been the “Hispanic paradox” and its counterargument, the “salmon bias.” The paradox is based on the persistent finding that persons of lower economic status more than higher SES persons, tend to have higher mortality rates and are in poorer health. Hispanics, however, are the exception, and thus the paradox. As the originators of the concept say, increasingly controversial issue in the field of Hispanic health, the suggestion of an existence of a “Hispanic paradox” based on evidence that Hispanics have favorable health and mortality profiles relative to the non-Hispanic White population. That is the paradox because most Hispanics in the United States are socioeconomically disadvantaged vis-à-vis non-Hispanic Whites, and the literature from the United States and elsewhere has shown a consistent association between low socioeconomic status and poor health outcomes. (Markides & Eschbach, 2005, p. 68)

6    R. R. VERDUGO

Compared to non-Hispanic Whites, Hispanics have attained fewer years of education, earn less income, and work in less prestigious jobs. Yet the paradox is that Hispanics have lower all-cause mortality rates than nonHispanic Whites (see Markides & Coreil, 1986; Sorlie, Backlund, Johnson, & Rogot, 1993). There have been at least two explanations for this inconsistency. The first explanation argues that immigrants, as a group, are healthier than nonimmigrants and non-Hispanic Whites (Markides & Coreil, 1986; Marmot, Adelstein, & Bulusu, 1984; Scribner, 1996; Sorlie et al., 1933). For example, research has shown that immigrants have lower mortality rates than the populations in their own countries (Marmot et al., 1984), and that the foreign-born have better reported health than U.S.born respondents, and that Latino immigrants are healthier than Latinos residing in the United States for long periods of time (see Stephen, Foote, Hendershot, & Schoenborn, 1994). Healthier immigrant profiles thus account for the paradox; at least this is what some have argued. A second explanation is referred to as the “salmon bias” explanation. It argues that immigrants who are dying or are in poor health move back to their country of origin. Hence, their mortality is not counted in U.S. statistics, and available data are thus biased (Pablo-Mendez, 1994; Shai & Rosenwaike, 1987). Since the Hispanic paradox (HP) and the salmon bias (SB) hypotheses are so important in this particular area of demographic study, researchers have pursued various ways of examining which argument has the greater explanatory power. Abraido-Lanza, Dohrenwend, Ng-Mak, and Turner (1999) found that the SB explanation did not apply to Cubans and Puerto Ricans, and also had marginally mixed results in explaining mortality results for the remaining Hispanic groups in their study. In 2001, Franzini, Ribble, and Keddie conducted a comprehensive review of the research that spanned 20+ years and concluded that the HP was real and that there were no known hard, defensible explanations. More recently still, Markides and Eschbach (2005) reviewed the extant research and drew the following conclusions: • The Hispanic mortality advantage is persistent. • Though Hispanics have a mortality advantage, this does not translate into better health in old age. • There are mortality differences within Hispanic ethnic groups, and the salmon bias may be working in the older Mexican immigrant population. • Some of the Hispanic mortality advantage may be due to poor data. The research on the demography of the Hispanic population can be summarized in two statements: (a) the Hispanic population is growing at a

Hispanics, Immigrants, and the Growth of the U.S. Population    7

dramatic rate, and (b) while immigration has driven this growth in the past, it will be fertility that will determine the growth of this population in the future, while the influence of mortality appears to be minimal. What seems to be missing from this body of research on the demography of the Hispanic population is an examination of the impact it has on the growth of the U.S. population. Such an analysis should be tempered by two important issues: immigrant status and Hispanic ethnic groups. There are two reasons for my making such a statement. First, immigrant flows are notoriously volatile, they ebb and flow with the push and pull factors. For Hispanics, as we will see shortly, this has also been the case; in some years Hispanic immigration has been dramatic, and less so in other years. Second, Hispanic immigration to the United States has varied by Hispanic ethnicity. While Mexicans continue to be the largest group, “Other Hispanics” are now the second-largest Hispanic subgroup, and they represent different Hispanic countries and cultures. Both factors, then, are important to consider when studying population change in the United States that is affected by Hispanics. Methods Data7 Decennial Census: 1850–2000 The 18508 to the 2000 decennial census are the primary databases used in my study. Table 1.1 displays the sampling from each census, as well as the total Hispanic and non-Hispanic populations. I should also point out that the 1850 and 1860 census databases exclude data for the slave population in the United States. Thus, in 1850, the total population minus the slave population is 19,990,000. If I include the slave population, the total would be 23,054,000. In 1860, the numbers were 27,352,587 and 30,129,543. From 1870 forward, my estimates are well within reported estimates of the U.S. population. American Community Survey The 2007 American Community Survey (ACS) is the second database used in my study. The ACS is a 1-in-10 national random sample of the U.S. population. The data are weighted, and the smallest identifiable geographic unit is the PUMA, containing at least 100,000 persons. The total sample size of the 2007 ACS is 301.6 million, and there are 45.4 million persons of Hispanic origin in the sample: 27.4 million nonimmigrant Hispanics, and 18.0 million Hispanic immigrants.

8    R. R. VERDUGO Table 1.1  Databases Used in this Study Database and Year Population in 1,000s Description 1950, 1% sample T = 152,200 H = 3,300

1960, 1% sample T = 179,300 H = 5,800 1970, 1% sample, Form 1 state sample T = 202,500 H = 7,700 1980, 5% state sample T = 226,700 H = 14,700 1990, 5% state sample T = 248,100 H = 21,800 2000, 1% sample T = 281,400 H = 35,200 2007 American Community Survey T = 301,600 H = 45,400

1-in-100 national random sample of the population. Weighted sample. The 1930–1950 samples do NOT include data from Alaska and Hawaii. Samples from 1900 to 1920, and 1960 to the present do include data for Alaska and Hawaii. Every household has one “sample-line” person who answered additional census questions. Only places of at least 100,000 population can be identified with any geographic variable. 1-in-100 national random sample of the population. The smallest identifiable geographic unit is the state. 1-in-100 national random sample of the population. The smallest geographic unit is the state.

1-in-100 national random sample of the population. No place smaller than 100,000 can be identified with any geographic variable. 1-in-20 national random sample of the population. Data are weighted. No place smaller than 100,000 can be identified with any geographic variable. 1-in-100 national random sample. Data are weighted. The smallest identifiable geographic unit is the Super-PUMA containing at least 400,000 persons. Super-PUMAs do not cross state boundaries. See text for a description of the 2007 ACS.

Variables Hispanic Hispanic identity prior to the 1980 census was estimated from a constructed variable. The actual algorithm used in identifying Hispanics is presented in Appendix A. The algorithm is referred to as the “Hispanicrule.” In this chapter, Hispanics are persons who were identified as Mexican, Puerto Rican, Cuban, and those belonging to some Other Hispanic group. Immigrants Immigrant status is a second variable used in my analyses. The actual operationalization of immigrant status varies by census year in terms of the variables used to identity such status. Four variables were used: years in

Hispanics, Immigrants, and the Growth of the U.S. Population    9 Table 1.2  Operationalizing Immigrant Status: 1850–2007 Year

Variable

1850 to 1880

Place of birth: BPL

Operationalization

1 = nonimmigrant if born in the U.S. to American parents; 2 = immigrant if not born in the U.S., or born in the U.S. to immigrant parents. 1900 to 1930 Years living in the U.S.: 1 = nonimmigrant if response was “not YRSU.S.A2 applicable” because person was a U.S. citizen or born in the U.S.; 2 = immigrant if years in the U.S. was 0 or greater. (The 0 represents someone living in the U.S. less than 1 year.) 1940 and 1950 Citizenship status: 1 = nonimmigrant if the respondent was a citizen CITIZEN but not a naturalized citizen; 2 = immigrant if the respondent was not a citizen or a naturalized citizen. 1960 Status determined on 1 = nonimmigrant if the respondent was not foreign born and foreign born and if both parents were also not foreign-born status foreign born; 2 = immigrant if the respondent of parents: was foreign born and also if both parents were NATIVITY foreign born. 1970 and 1980 Years living in the U.S.: 1 = nonimmigrant if response was “not YRSU.S.A2 applicable” because person was a U.S. citizen or born in the U.S.; 2 = immigrant if years in the U.S. was 0 or greater. (The 0 represents someone living in the U.S. less than one year.) 1990 to 2007 Citizenship status: 1 = nonimmigrant if the respondent was a citizen CITIZEN but not a naturalized citizen; 2 = immigrant if the respondent was not a citizen or a naturalized citizen.

the United States, citizenship status, birthplace, and nativity. I also benchmarked my estimates with other published data, such as data from the Virginia Library Geostat Center (www.fisher.lib.virginia.edu/collections/stats/ histcensus/index.html). A complete description of how immigrants are identified in my sudy, both Hispanic and non-Hispanic, may be found in Table 1.2. It should be pointed out that my immigrant data are estimates. Findings Hispanics and Population Change in the United States: 1850–2007 In 1850, the census reported a total population of nearly 20 million persons living within the geographic boundaries of the United States. A majority of the population at that time was non-Hispanic, and the resident

10    R. R. VERDUGO

Hispanic population in the United States was less than 1% of the total, estimated to be 0.55%. Indeed, in 1850, there were an estimated 19, 990,000 persons living in the United States. Of this figure 19,880,000 were nonHispanic, and approximately 110,000 were Hispanic. But things began to change, first slowly and then dramatically. By 2007, the total U.S. population had grown to 301,600,000; the nonHispanic population was tabulated at 256,200,000, and the Hispanic population was 45,400,000. The share of the total U.S. population that was Hispanic grew from 0.55% in 1850 to 15.1% in 2007. Figure 1.1 exhibits the growth curves (1850 = 100) of the Hispanic, non-Hispanic, and total populations from 1850 to 2007, respectively. The growth of the Hispanic population has been greater than either the non-Hispanic or total U.S. populations. Of particular significance has been the rise of the Hispanic population since about 1950. At that point in history, the United States was completing its involvement in WWII, and during the war, employers had repeatedly turned to Mexican labor to fill severe labor shortages. Mexican immigrants had moved into such places as Michigan, Chicago, and of course, California to work in a variety of industries (see McWilliams,1990[1948]). Moreover, it was during this period in U.S. history that Puerto Ricans began taking advantage of lower airfares and began shuttling between the “Island” and the U.S. mainland, especially to the northeastern sections of the United States. These important histori-

Figure 1.1  Growth curves for total U.S. population, non-Hispanics, and Hispanics: 1850–2007.

Hispanics, Immigrants, and the Growth of the U.S. Population    11

cal events can be seen in the growth of the Hispanic population at certain points between 1850 and 2007. The Hispanic population continued to grow through the 1960s and 1980s as the U.S. economy started to “de-industrialize.” Hispanic population growth was spurred, in large measure, by Hispanic immigrants. Wellpaying manufacturing jobs began to disappear and the loss of these jobs in the North created serious economic problems (Harrison & Bluestone, 1990; Wilson, 1996). Moreover, technological and global competition increased the need for lowering production costs, and labor became a prime target.9 Cheap labor, whether involving moving plant operations, outsourcing to a Third World country, or hiring immigrant labor became a major strategy for many U.S. employers. In the United States, a large proportion of the Hispanic population (especially immigrants) filled these roles. Not only were immigrants a source of cheap labor, they also provided special services to skilled labor and their families, such as gardening, housecleaning, and childcare (Verdugo, 2006). It was during this period in U.S. history that large numbers of Hispanic immigrants began arriving from other Latin American countries, these specific sending countries faced harsh push factors, for example, hard economic times and/or significant political unrest. Between 1850 and 2007, the total U.S. population grew from 19,990,000 to 301,600,000. So the net growth over this time period was 281,610,000. Of this net growth, Hispanics accounted for 45,290,000 or 16.1% of the net growth in the U.S. population between 1850 and 2007. The remaining growth (83.9%) was due to non-Hispanics. Growth of the Hispanic Population: Immigrant Status, Hispanic Ethnic Groups, and Natural Increase Growth of Hispanic Ethnic Groups: 1850–2007 In 1850, of the roughly 110,000 Hispanics residing in the United States, about 79,773 were of Mexican origin, 2,143 of Cuban origin, and the remaining 28,084 were of Other Hispanic origin.10 By 2007, the distribution had changed: 29.2 million Mexicans, 1.6 million Cubans, 4.1 million Puerto Ricans, and 10.4 million Other Hispanics. Several important points in U.S. history precipitated the increase in the number of each Hispanic ethnic group. Some of these historical events are unique to a specific Hispanic group, while others may apply to at least two Hispanic groups. Immigrants from Mexico have been the driving force behind the growth in the Mexican population. Two world wars, a fundamental change in U.S. industrial structure, a tanking U.S. economy, as well as political turmoil in sending countries were all forces driving Mexican immigration to the Unit-

12    R. R. VERDUGO

ed States (see Barrera, 1979; Calavita, 1992; Cardoso, 1980; Garcia, 1981; Grebler et al., 1970; McWilliams, 1990[1948]). The Cuban Revolution uprooted its middle and upper classes to the United States, as Fidel Castro and Che Guevara took control of Cuban society and radically changed its class structure (see Portes & Stepick, 1993; Sullivan & Pedraza-Bailey, 1979). And throughout the 1980s and 1990s, Cuban refugees kept leaving the island, whether on their own or being pushed out by Cuban authorities.11 We can see the increase in the Cuban population residing in the United States during this time by noting the rise in the Cuban (immigrant) population during the 1960s and into the 1990s.12 Puerto Ricans have a different story. After the Spanish-American War of 1898, Spain turned over Puerto Rico to the United States as a territory under the Treaty of Paris on December 19, 1898.13 With the signing of that treaty, the foundation was set for Puerto Ricans to emigrate to the United States.14 It wasn’t until the 1950s, however, with the advent of cheap airfares, that movement to and from Puerto Rico truly became significant. This movement began during the late 1940s and has continued to the present (see Fitzpatrick, 1987 for an overview; also refer to Bonilla & Campos, 1981; Georges, 1990; Rivera-Batiz & Santiago, 1998; Rodriguez, 1991; Sanchez Korrol, 1983, 1984). Again, the data bear this out as the Puerto Rican population jumped from about 77,000 in 1940 to roughly 356,900 in 1950, and to about 1 million by 1960. By 2007, the Puerto Rican population on the U.S. mainland had quadrupled to 4.1 million. The pattern for Other Hispanics has been one of constant and then dramatic growth during the latter part of the 20th century (see Grasmuck & Pessar, 1991; Hamilton & Stoltz, 2001; Rumbaut, 1992). For example, between 1970 and 1980, the Other Hispanic population grew to over 3 million, and by 2007 there were nearly 10.5 million Other Hispanics in the United States. Other Hispanics are currently the second-largest Hispanic ethnic group in the United States.15 Figure 1.2 displays the growth curves of each Hispanic ethnic group from 1850 to 2007 (1850 = 100). I smoothed the curves by the log function (base 10). These results indicate that Other Hispanics have been the fastest growing Hispanic ethnic group, and that the growth curves of the remaining Hispanic ethnic groups have not only been steady but appear to have similar slopes. The Growth of the U.S. Hispanic Immigrant Population In this section, I examine the growth of the Hispanic immigrant population. Table 1.3 presents data for this section. In conducting my analysis, I will look at immigrants within each of the Hispanic ethnic groups.

Hispanics, Immigrants, and the Growth of the U.S. Population    13

Figure 1.2  Growth curves for the Hispanic population and subgroups: 1850 to 2007 (1850 = 100, log10).

Figure 1.3 exhibits the growth of the Hispanic immigrant population from 1850 to 2007. In 1850, there were an estimated 21,017 Hispanic immigrants in the United States; by 2007, this figure jumped to about 18 million. In 1850, Hispanic immigrants accounted for about 19.1% of the total Hispanic population in the United States, and by 2007, Hispanic immigrants accounted for 39.8% of the total Hispanic population residing in the United States. So, what has been the impact of Hispanic immigrants on the growth of the total Hispanic population? While the total Hispanic population grew by 45.3 million between 1850 and 2007, the Hispanic immigrant population grew by 18 million. Thus, about 39.8% of the total Hispanic population increase between 1850 and 2007 was due to Hispanic immigrants. The remainder (60.2%) was due to increases in the Hispanic nonimmigrant population. (Note that these figures are similar to those reported in the paragraph above. The reason for the close similarity is due to the exceedingly small numbers in 1850.) Are there differences in the impact immigrants from different Hispanic ethnic groups have had on the growth of the Hispanic population? There are, indeed. While the total Hispanic population grew by 45.3 million between 1850 and 2007, the growth in the immigrant populations

110,000 160,000 170,000 350,000 510,000 820,000 1,300,000 2,300,000 2,139,819 3,300,000 5,800,000 7,700,000 14,700,000 21,800,000 35,200,000 45,400,000

Year

1850 1860 1870 1880 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2007

Source: Ruggles et al., 2009.

Total Hispanic Population

21,017 37,785 52,484 97,746 140,000 300,000 620,000 840,000 480,000 630,000 1,535,121 1,818,200 4,230,700 7,810,000 14,130,000 18,050,000

88,983 122,215 117,516 252,254 370,000 520,000 680,000 1,460,000 1,659,819 2,670,000 4,264,879 5,881,800 10,469,300 13,990,000 21,070,000 27,350,000 50,000 10,000 180,000 160,000 310,000 480,000 1,000,000 –160,181 1,160,181 2,500,000 1,900,000 7,000,000 7,100,000 13,400,000 10,200,000

Hispanic Immigrant Nonimmigrant Change in Hispanic Population Hispanic Population Population

16,768 14,699 45,262 42,254 160,000 320,000 220,000 –360,000 150,000 905,121 283,079 2,412,500 3,579,300 6,320,000 3,920,000

33,232 –4,699 134,738 117,746 150,000 160,000 780,000 199,819 1,010,181 1,594,879 1,616,921 4,587,500 3,520,700 7,080,000 6,280,000

Change in Hispanic Change in Immigrant Nonimmigrant Population Hispanic Population

Table 1.3  Changes in the Native-Born and Foreign-Born Hispanic Population: 1850–2007

14    R. R. VERDUGO

Hispanics, Immigrants, and the Growth of the U.S. Population    15

Figure 1.3  Growth of Hispanic immigrant population: 1850 to 2007.

among the four Hispanic ethnic groups were: Mexican (11.6 million), Cuban (975,000), Puerto Rican (47,600), and Other Hispanics (5.4 million). The equation below displays the impact of Hispanic immigrants and Hispanic nonimmigrants on the growth of the total Hispanic population from 1850 to 2007: GHP1850–2007 = {MI ( 25.7 ) + CI (2.2) + PRI (.11 ) + OI (11.0 )} +  {MNI (38.6 ) + CNI (1.4 ) + PRNI (9.0 ) + ONI (11.2 )} where MI = Mexican immigrants, CI = Cuban immigrants, PRI = Puerto Rican immigrants, OI = Other Hispanic immigrants, MNI = Mexican nonimmigrants, CNI = Cuban nonimmigrants, PRNI = Puerto Rican nonimmigrants, ONI = Other nonimmigrants. Note that an important number of Puerto Rican respondents considered themselves immigrants; I did not recode these data. Thus, Mexican nonimmigrants have contributed the most to the growth of the total Hispanic population among nonimmigrant Hispanic ethnic groups. Other nonimmigrant Hispanics are next at approximately 11.2%. A different perspective can be garnered by looking at the percentage each Hispanic immigrant ethnic group has of the total Hispanic population from 1850 to 2007. Figure 1.4 displays these data. Mexican immigrants have the greater share of the total Hispanic population among the subgroups in all years, except in 1960 when it, essentially, shares the lead with Puerto Ri-

16    R. R. VERDUGO

Figure 1.4  Hispanic immigrants as a percentage total of the Hispanic population: 1850–2007.

can immigrants. The next-largest share of the total is held by Other Hispanics. Generally, Mexican and Other Hispanic immigrants have accounted for the greatest impact on the growth of the total Hispanic population from 1850 to 2007. The Impact of the Hispanic Population on Demographic Change in the United States The Impact of Hispanics on U.S. Population Growth: 1850–2007 Figure 1.5 displays the annual percentage of the total U.S. population composed of all Hispanics and selected Hispanic ethnic groups for the years 1850 to 2007. The impact of the Hispanic population has been particularly significant in the latter half of the 20th century. From 1850 to about 1910, the Hispanic population comprised less than 1% of the total U.S. population. The situation began to change between 1910 and 1920, though the percentage of the U.S. Hispanic population grew to over 1% and remained there for the next two decades. In 1950, Hispanics inched up to about 2% of the total U.S. population, over 3% in 1960, nearly 4% in 1970 (3.84%), about 6.5% in 1980, nearly 9% in 1990, and then to over 12.5% in 2000. By

Hispanics, Immigrants, and the Growth of the U.S. Population    17

Figure 1.5  Hispanics as a percentage of U.S. population: 1850–2007.

2007, the Hispanic population comprised slightly over 15% of the total U.S. population.16 Which Hispanic ethnic groups have contributed most to the growth of the U.S. population? From previous sections, I would hypothesize that it would be Mexicans. Data displayed in Figure 1.5 confirm my hypothesis. For the entire period represented in Figure 1.5, Mexicans have held a larger share of the total U.S. population than the other Hispanic subgroups. (In fact, the share of the U.S. population held by Mexicans is greater than the combined shares of the remaining three Hispanic ethnic groups.) For example, in 1850, Mexicans made up about 0.4% of the U.S. population; Cubans 0.01%; and Other Hispanics 0.1%. (Though data for Puerto Ricans were not available until 1860, the reported numbers are much too small for any meaningful discussion.) By 2007, Mexicans held 9.7% of the total U.S. population. The shares held by the remaining Hispanic ethnic groups were Cubans, 0.5%; Puerto Ricans climbed to 1.4%; and Other Hispanics held about 3.5% of the total U.S. population. Non-Hispanic and Hispanic Immigrants and the Growth of the U.S. Population: 1850–2007 Between 1850 and 2007, immigrants played a significant part in the growth of the U.S. population, especially in the early part of the 19th and 20th centuries. Of the nearly 281.6 million person increase in the U.S. population from 1850 to 2007, 35.8 million was due to immigrants. In other words, immi-

18    R. R. VERDUGO

grants accounted for 12.7% of the gain in the U.S. population between 1850 and 2007. The contribution of immigrants to population change in the United States was particularly important during the latter part of the 19th and 20th centuries, and in more recent decades. Figure 1.6 displays three curves: the proportion of the total U.S. population composed of all immigrants, of Hispanic immigrants, and non-Hispanic immigrants. In 1850, immigrants accounted for about 11.6% of the total U.S. population; and slightly over 15% in 1860. From 1860 to 1970, there were declines in the share of the U.S. population held by immigrants. In 1970, the percentage reached its all-time low of about 5%, and then it began to climb again, and by 2007, immigrants accounted for approximately 12.6% of the U.S. population. There are important reasons for these ebbs and flows of immigration to the United States. As I have discussed in earlier sections, such explanations are based on push and pull factors reflecting the economic and political circumstances, as well as foreign policy (see Footnote 4) in sending countries; the networks in the U.S. formed by immigrants in the sending country; and the availability of jobs in the United States in specific industries. All these factors have contributed to the volatile nature of immigration to the United States.

Figure 1.6  Total, non-Hispanic, and Hispanic immigrants as a percentage of total U.S. population: 1850–2007.

Hispanics, Immigrants, and the Growth of the U.S. Population    19

What has been the contribution of non-Hispanic immigrants to the growth of the U.S. population? Of the total U.S. population growth between 1850 and 2007, non-Hispanic immigrants accounted for about 9.7% of that growth, and their impact has varied by decade. In 1850, non-Hispanic immigrants composed slightly over 11% of the U.S. population. Their share of the U.S. population increased over the next decade to about 15%, and then it steadily declined until about 1910 when it picked up again to about 14.4%. Since 1910, though, there has been a steady decline in the share of the U.S. population held by non-Hispanic immigrants. In 1970, the percent reached an all-time low of about 4% and then began to increase again until 2007 when it climbed to 6.6%. The impact of the Hispanic immigrant population on U.S. population growth, in contrast, has been the direct opposite. Of the total U.S. population change between 1850 and 2007, the Hispanic immigrant population accounted for about 6.4% of that change. In 1850, Hispanic immigrants accounted for only 0.1% of the total U.S. population, and their share of the U.S. population remained below 1% until 1960. Between 1960 and 1970, the share of the U.S. population held by Hispanic immigrants climbed to 0.98 %. From about 1970, there has been a steady upward increase in the share of the U.S. population held by Hispanic immigrants. In 2007, the share of the U.S. population held by Hispanic immigrants was 5.98%. So which group has had the greater impact on the growth of the U.S. population from 1850 to 2007? If our gaze is on the overall change, we would conclude that non-Hispanic immigrants have had the greater impact. However, by stepping back and looking at change since the middle of the 20th century, we see that non-Hispanic immigrants had the greater impact in earlier decades, and Hispanic immigrants took the lead in later decades, especially after 1970. In any event, the equation below displays the contribution the 10 groups of interest have had to total U.S. population growth (U.S.PG) from 1850 to 2007. U.S.PG1850–2007 = {NHNI (77.61) + NHI (6.31)} + {MI (4.13) +  CI (.35) + PRI (.02) + OHI (1.91) + MNI (6.21) + CNI (.22) +  PRNI (1.44) + OHNI (1.80)} where NHNI = non-Hispanic nonimmigrants, NHI = non-Hispanic immigrants, MI = Mexican immigrants, CI = Cuban immigrants, PRI = Puerto Rican immigrants, OHI = Other Hispanic immigrants, MNI = Mexican nonimmigrants, CNI = Cuban nonimmigrants, PRNI = Puerto Rican nonimmigrants, OHNI = Other Hispanic nonimmigrants. The equation shows that non-Hispanic nonimmigrants (NHNI) have contributed the greatest to the growth of the U.S. population from 1850 to 2007. Indeed, NHNI have accounted for 77.61% of the growth of the

20    R. R. VERDUGO

U.S. population over that time period. But note that non-Hispanic immigrants (NHI) and Mexican nonimmigrants (MNI) each account equally for the growth of the U.S. population from 1850 to 2007: 6.31% and 6.21%, respectively. Moreover, by totaling percentages, we can get a sense of how much Hispanics (both immigrants and nonimmigrants) have contributed to U.S. population growth from 1850 to 2007. The equation below displays the contribution of the total non-Hispanic population (NH), the Hispanic nonimmigrant population, and the Hispanic immigrant population to the growth of the U.S. population from 1850 to 2007. U.S.PG1850–2007 = NH (83.92) + HNI (9.68) + HI (6.40) While the non-Hispanic population has contributed an overwhelming percentage to the total U.S. population growth from 1850 to 2007 (83.92%), the contribution of Hispanics, both immigrants and nonimmigrants, has been substantial. Indeed, Hispanic nonimmigrants have contributed 9.68% to the total growth of the U.S. population since 1850, and the immigrant Hispanic population another 6.40%. As can be seen from Figure 1.6 and the simple calculations I have just presented, both total immigrant and non-Hispanic immigrant shares of the U.S. population declined significantly between about 1910 and 1970. After 1970, both rates began to increase. Of the two curves, the total immigrant growth increased even more. By looking at the Hispanic immigrant share of the total U.S. population, an entirely different picture emerges. First, while the share of the U.S. population held by Hispanic immigrants was miniscule during the middle part of the 19th century and into the middle part of the 20th century, it started to pick up from about 1970 and has been increasing since. In fact, a comparison of their curve with those of NH and HNI suggests that since 1970, the increase in the total share of the U.S. population due to immigrants has been driven by Hispanic immigrants. Hispanic immigrants have had, since the latter part of the 20th century, a significant impact on U.S. population growth. Conclusion The United States is a country of immigrants. Immigrants have been and continue to be an important component of population growth in the United States. Yet, while immigrants are still an important part of the U.S. population, their presence in the United States has also become a point of contention and controversy.

Hispanics, Immigrants, and the Growth of the U.S. Population    21

In the early part of the 19th century in the United States, immigrants from Western and Eastern Europe made significant contributions to the growth of the U.S. population (not to mention their being an important cog in the rapid industrialization of the U.S. economy), but in the years preceding and following the Second World War, immigrants from Mexico and Central and South America began and have continued to make their presence felt in the United States. What, exactly, have been the contributions Hispanics and Hispanic immigrants have made to the growth of the U.S. population? In this chapter, I have attempted to frame an answer to such a question by examining historical trends and patterns among immigrants and Hispanics, both native born and immigrants, as well as Hispanic ethnic groups, from 1850 to 2007. In conducting my analyses, I estimated the number of all immigrants, non-Hispanic immigrants, Hispanic immigrants, and native-born Hispanics from the decennial census (1850 to 2000), and from the 2007 American Community Survey. There are three important findings from my study. First, the impact Hispanic immigrants have had on the growth of the U.S. population was minimal from 1850 to about the middle of the 20th century. Of the roughly 110,000 Hispanics residing in the U.S. in 1850, about 72% were Mexican. A major reason for the larger number of Mexicans, of course, was that the western region of the United States was part of Mexico until 1848 when the Treaty of Guadalupe Hidalgo was signed and ended the Mexican–American War. With the signing of that treaty, the United States annexed most of the lands west of the Mississippi. From 1850 to 2007, the Hispanic immigrant population grew by about 18 million persons and composed slightly less than 40% of the total immigrant population growth over that time. Moreover, Hispanic immigrants accounted for 6.4% of the growth of the U.S. population from 1850 to 2007, and they have had a significant impact on the growth of the U.S. population from middle of the 20st century to the present. From the middle of the 20th century to its conclusion were decades of substantial Hispanic immigration that contributed significantly to U.S. and Hispanic population growth. For example, the Hispanic population started to increase dramatically about 1920 when it reached a figure of slightly over 1.3 million; and by 1960, the total Hispanic population in the United States climbed to 5.8 million. Scholars are in agreement about the push and pull factors that accounted for this rapid increase in the size of the Hispanic population. Thus, the Second World War created severe labor shortages in many manufacturing and agricultural industries, and Mexican labor filled much of those needs. The Cuban Revolution sent many middle-class and upper-class Cubans to the United States, who had greatly benefited under the Batista regime, but were not likely to do so under the Castro regime.

22    R. R. VERDUGO

Inexpensive airfares from Puerto Rico to the United States made it easy for Puerto Ricans to move between the Island and the United States, particularly along the U.S. Eastern Seaboard. Economic and political instability in other parts of South and Latin America were causes for immigration from countries such as Chile, El Salvador, Columbia, Peru, Argentina, Guatemala, and Bolivia. Between 1950 and 2007, some 17.4 million Hispanic immigrants came to the United States, compared to about 24.7 million non-Hispanic immigrants over that same time span. Second, of the four immigrant Hispanic ethnic groups identified in my datasets, Mexican immigrants have had the most impact on U.S. population growth. Between 1950 and 2007, the years of greatest increase for the Hispanic immigrant population, the Mexican immigrant population grew by 11.2 million, the Cuban immigrant population by 947,000, the Puerto Rican immigrant population by 43,000, and the Other Hispanic immigrant population by 5.3 million. Mexican immigrants have and continue to have an important effect on Hispanic population growth as well an impact on U.S. population growth. Finally, after having parsed the contributions to U.S. population growth among eight groups, and thus controlling for Hispanic ethnicity and immigrant status, I find that Mexicans have contributed the most to U.S. population growth from 1850 to 2007. However, it should also be pointed out that immigrants, especially Mexican immigrants, have had a significant impact on U.S. population growth. What conclusions can we draw from this study? The clearest conclusion is that Hispanic immigrants have had a significant impact on the growth of the U.S. population. This impact was especially significant from about the middle of the 20th century to the present. Secondly, given what we know about push and pull factors, the patterns among Hispanic immigrant numbers agree with both these concepts: push factors include both economic and political instability that have pushed Hispanic immigrants to the United States, while pull factors have included the need for unskilled and semi-skilled labor in the United States. The trends in Hispanic labor confirm such a viewpoint. Thirdly, we know that new Hispanic immigration is not following previous settlement trends. While settlement patterns are not presented here, we do know that Hispanic immigrants are settling in areas without histories of Hispanics, and that their increasing numbers are challenging those localities. So, a third conclusion is that the increase in the Hispanic immigrant population and its impact on U.S. population and society needs more research. The dramatic growth in the Hispanic immigrant population, especially since the mid-20th century, needs research in the ever-changing contexts under which the Hispanic population grows, changes, and moves.

Hispanics, Immigrants, and the Growth of the U.S. Population    23

Appendix A: Hispanic Rule In the IPUMS dataset, a person was coded as Spanish/Hispanic/Latino with the variable HISPAN from 1850–1970. In contrast, Hispanic origin was asked directly on the census forms from 1980 onward. In the earlier samples, we inferred Hispanic origin from other variables The IPUMS established Hispanic origin in the pre-1980 samples according to eight basic rules, the HISPRULE algorithm. If Hispanic origin could be assigned according to more than one rule, the lowest-numbered rule was applied. See Table A.1. Table A.1  HISPRULE Algorithm Used in Identifying Hispanics, Pre-1980 Census 0 = Person is not Hispanic. 1 = Person was born in a Hispanic area. This covers two possibilities: (a) The person was born in a Hispanic country. Hispanic countries are Argentina, Bolivia, Canary Islands, Central America, Central America (n.s.), Chile, Columbia, Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Latin America (n.s.), Mexico, Nicaragua, Panama, Paraguay, Peru, Puerto Rico, South America, South America (n.s.), Spain, Uruguay, and Venezuela. (b) The person was born in Arizona, California, New Mexico, or New Mexico Territory while the area was still under Spanish/Mexican jurisdiction (i.e., before July 1848). Country of origin was coded as Mexican in these cases. 2 = The person’s father or mother was born in a Hispanic country. In cases where father or mother’s birthplaces were not available and the person was living with his/her father or mother, IPUMS used other geographic data to impute birthplace information. Although birthplace data are available for sample-line respondents in 1940 and 1950, this rule was not applied in 1940 and 1950 because of comparability issues. If both the father and the mother were Hispanic, the person received the country-of-origin code of his/her father. 3 = The person’s grandparent was born in a Hispanic country. Grandparent birthplace was imputed only in cases where father’s birthplace and mother’s birthplace were asked in the census and the person was living with his/her father and/or mother. Although birthplace data are available for sample-line respondents in 1940 and 1950, this rule was not applied in 1940 and 1950 because of comparability issues. If multiple grandparents were Hispanic, the person received the country-of-origin code of the Hispanic grandparent first on the following list: father’s father, father’s mother, mother’s father, mother’s mother. 4 = The person’s spouse is Hispanic because of rule 1, 2, or 3. The country of origin was coded to match the spouse. 5 = The person is a relative of a householder who is Hispanic because of rule 1, 2, 3, or 4. The country of origin was coded to match the householder. 6 = The person has a Spanish surname and the person was born in the United States and his/ her father was born in the United States (or has a missing value for birthplace) and his/ her father’s father was born in the United States (or has a missing value for birthplace). This rule only applies to males or females with no spouse in the household. For people who were allocated to Hispanic using rules 6–8, country of origin was assigned based on the predominating country of origin (if any) among Hispanics (as defined by rules 1–5) in that state in that year. 7 = A woman whose spouse qualifies as Hispanic through rule 6. 8 = The person is a relative of a householder who is Hispanic because of rule 6 or 7.

24    R. R. VERDUGO

Notes 1. The author is a Senior Research Scientist, National Education Association in Washington, DC. Address all correspondence to Richard R. Verdugo, National Education Association, 1201 16th Street, N.W., Washington, DC 20036. [email protected]. 2. Paper presented at the annual meeting of the European Population Conference, Vienna, Austria, September 1–4, 2010. 3. Another interesting finding is that the longer immigrants stay in the United States, the lower their fertility rate. Scholars have speculated about this pattern and the most common explanations are (a) that social norms about fertility take hold of immigrants’ behavior, and (b) greater economic opportunities in the United States for women translate to fewer children. 4. While I limit my discussion to the Push/Pull paradigm, there are at least nine theories of international migration. Their main contribution has been to add context to why people migrate, the involvement of networks (at home and abroad), the sort of jobs awaiting immigrants, and country-to-country relationships that affect the decision to immigrate and the sort of labor needed in the receiving country. For excellent discussions see Massey et al. (1993), and Portes and Borocz (1989). 5. There is another set of factors that appear to have increased the composition of Hispanic immigrants: U.S. laws regarding immigration. Three laws are most often cited for having such an effect: (a) The 1965 Immigration and Nationality Act restricted immigration from Eastern Europe and opened up the door for immigrants from Asia and ended the Bracero Program. (b) The 1968 Immigration Reform and Control Act limited immigration from Mexico and opened it up for immigrants from other Hispanic-dominated countries. (c) The 1996 Illegal Immigration Reform and Immigration Responsibilities Act restricted programs for immigrants, many of whom had depended on such programs, and opened up the possibility for undocumented Hispanic immigrants to be naturalized. In fact, the numbers applying for naturalization tripled over the year, and they did so with the full knowledge that their sending countries would not deny them their original citizenship and the rights attached to such citizenship. 6. Spain, for example, has a large Latin American immigrant population. See Verdugo and Swanson (2011). 7. Data were downloaded from the University of Minnesota’s Population Research Center. See Ruggles et al. (2009). 8. The 1850 census was the first to actually identify country of origin and so becomes a wonderful resource for immigration research. 9. The relevant topic here is globalization. Globalization is a concept that focuses on the elimination or reduction of national barriers in order that businesses might operate unfettered across international borders. As Saskia Sassen (2006) states, “a good part of globalization consists of an enormous variety of micro-processes that begin to denationalize what had been constructed as national—whether policies, capital, political subjectivities, urban spaces, temporal frame, or any other of a variety of dynamics and domains.”

Hispanics, Immigrants, and the Growth of the U.S. Population    25   Proponents of Globalization cite its benefits: increased economic prosperity, increased opportunities for prosperity, increased civil liberties, and the efficient use of resources (see Bhagwati, 2004; Croucher, 2004; Friedman, 2000, 2005; Sachs, 2005).   Opponents of globalization point out that it limits national sovereignty and decision making, increases the power of multinational corporations, damages the rights of citizens, labor, and the environment (see Stiglitz & Charlton, 2005; Hirst & Thompson, 2002).   For an alternate view, see Cohen (2006). Cohen appears to argue that globalization is not necessarily imposed or exploitative. He argues that poor countries simply do not have anything that rich countries want or need. Globalization can work if certain “levers” are pulled by poor countries, like improving their educational systems. The main problem I have with Cohen’s argument is that he is dead wrong about rich countries not wanting anything poor countries have—they want their cheap labor, for one thing, and their raw materials for another. 10. Data on the Puerto Rican population were not available from the 1850 census. 11. There is some sense that Cuban authorities encouraged an important number of its criminals to emigrate to the United States. Such action rid Cuba of a troublesome element of its society and placed it in the lap of the United States. 12. Once the revolutionaries had taken control of Cuba, and the Cuban middle and upper classes began emigrating to the United States, the U.S. government took great pains to ensure that Cuban refugees were successful in the United States. For example, as far as we know, the first English or Second Language program (ESL) was started to help Cubans learn English. It was not so much that U.S. officials were humanitarians as it was important to them that Cubans succeed in a democratic society. The Cold War and the Communist/Democratic ideological battles were in full swing (Sullivan & Pedraz-Bailey, 1979). 13. In 1900, the U.S. Congress established a civil government in Puerto Rico, and about 17 years later, under pressure from Puerto Rican activists, U.S. President Woodrow Wilson signed the Jones Act, which granted U.S. citizenship to all Puerto Ricans. Citizenship was an important factor because it allowed Puerto Ricans the ability to travel to and from the United States and Puerto Rica without any barriers. 14. Theoretically, Puerto Ricans are not immigrants, but many of those migrating from Puerto Rico consider themselves immigrants. In my analysis, I rely on a respondent’s self identification as an immigrant. 15. Keep in mind that many Central and South American countries were experiencing difficult economic and political circumstances. Both circumstances pushed many to the United States and to Europe as well. For example, a sizeable proportion of immigrants to Spain are from South America (Verdugo & Swanson, 2011). 16. Projections of the Hispanic population is expected to be about 29% of the U.S. population by 2050 (Passel & Cohn, 2009). What is also interesting about this report is that 82% of the population increase in the U.S. that will occur between 2005 and 2050 will be from immigrants and their descendants, and 19% of all Americans will be immigrants.

26    R. R. VERDUGO

References Abraido-Lanza, A. F., Dohrenwend, B. P., Ng-Mak, D. S., & Turner, J. B. (1999). The Latino mortality paradox: A test of the “salmon bias” and health migrant hypothesis. American Journal of Public Health, 89(10), 1543–1548. Barrera, M. (1979). Race and class in the Southwest. South Bend, IN: University of Notre Dame Press. Bean, F. D., Telles, E. E., & Lindsay L. B. (1987). Undocumented migration to the United States: Perceptions and evidence. Population and Development Review, 13(4), 671–689. Bhagwati, J. (2004). In defense of globalization. New York: Oxford University Press. Bonilla, F., & Campos, R. (1981). A wealth of poor: Puerto Ricans in the new economic order. Daedalus, 110, 133–176. Calavita, K. (1992). Inside the state: The Bracero Program, immigration, and the INS. New York: Routledge. Cardoso, L. (1980). Mexican immigration to the United States: 1897–1931. Tucson: University of Arizona Press. Cohen, D. (2006). Globalization and its enemies. Boston, MA: MIT Press. Croucher, S. L. (2004). Globalization and belonging: The politics of identity in a changing world. Lanham, MD: Rowman and Littlefield. Durand, J. (1994). Mas alla de la linea: Patrones migratorios en el occidente de Mexico. Guadalajra, Mexico: D. F. Conaculta. Durand, J., & Arias, P. (2000). La experiencia migrante: Historia e inconografia de la migracn Mexicao Estados Unidos. Guadalajara, Mexico: Universidad de Guadalajara. Durand, J., & Massey, D. S. (2003). Clandestinos: Migracion Mexico Estado Unidos en los albores del siglo xxi. Guadalajra, Mexico: D.F. Editorial Miguel Angel Porrua. Durand, J., Telles, E., & Flashman, J. (2006). The demographic foundations of the Latino population. In M. Tienda & F. Mitchell (Eds.), Hispanics and the future of America (pp. 66–99). Washington, DC: National Academies Press. Economic Research Service. (2005). Rural Hispanics at a glance. Washington, DC: U.S. Department of Agriculture. Effland, A. B. W., & Kassel, K. (1996). Hispanics in rural America: The influence of immigration and language on economic well-being. Washington, DC: Economic Resource Service, U.S. Department of Agriculture. Fitzpatrick, J. P. (1987). Puerto Rican Americans: The meaning of migration to the mainland (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall. Franzini, L., Ribble, J. C., & Keddie, A. M. (2001). Understanding the Hispanic paradox. Ethnicity and Disease, 11, 496–518. Frey, W. H. (1994). Immigration and internal migration from U.S. metro areas: 1990 census findings by race, poverty and education. Research Report No. 94-304. Ann Arbor: University of Michigan, Population Studies Center. Frey, W. H. (2002). Census 2000 reveals new native-born and foreign-born shifts across U.S. Research Report No. 02-520. Ann Arbor: University of Michigan, Population Studies Center.

Hispanics, Immigrants, and the Growth of the U.S. Population    27 Frey, W. H. (2003). Metropolitan magnets for international and domestic migrants. Living Cities Survey series. Washington, DC: Brookings Institution, Center on Urban and Metropolitan Policy. Frey, W. H., & Liaw, K. L. (1998). The impact of recent immigration on population redistribution within the United States. In J. P. Smith & B. Edmonston (Eds.), The immigration debate: Studies of economic, demographic and fiscal effects of immigration (pp. 388–448). Washington, DC: National Academies Press. Friedman, T. L. (2000). The Lexus and the olive tree. New York: Farrar, Straus and Giroux. Friedman, T. L. (2005). The world is flat. New York: Farrar, Straus and Giroux. Garcia, M. T. (1981). Desert immigrants: The Mexicans of El Paso, 1880–1920. New Haven, CT: Yale University Press. Georges, E. (1990). The making of a transnational community. New York: Columbia University Press. Grasmuck, S., & Pessar, P. R. (1991). Between two islands: Dominican international migration. Los Angeles: University of California Press. Grebler, L., Moore, J. W., & Guzman, R. (1970). The Mexican-American people: The nation’s second largest minority. New York: Free Press. Griffith, D. (1995). Hay trabajo: Poultry processing, rural industrialization and the latinization of low-wage labor. In D. D. Stull, M. J. Broadway, & D. Griffith (Eds.), Any way you cut it: Meat processing and small town America (pp. 129–151). Lawrence: University of Kansas Press. Griffith, D. (2000). Work and immigration: Winter vegetable production in South Florida. In R. Tardanico & M. B. Rosenberg (Eds.), Poverty or development (pp. 139–178). New York: Routledge. Gutierrez, D. G. (1995). Walls and mirrors: Mexican Americans, Mexican immigrants, and the politics of ethnicity. Berkeley: University of California Press. Hamilton, N., & Stoltz-Chinchilla, N. (2001). Seeking community in a global city: Guatemalans and Salvadorans in Los Angeles. Philadelphia: Temple University Press. Harrison, B., & Bluestone, B. (1990). The great u-turn: Corporate restructuring and the polarization of America. New York: Basic Books. Hernandez-Leon, R., & Zuniga, V. (2000). Making carpet by the mile: The emergence of a Mexican immigrant community in an industrial region of the U.S. historic South. Social Science Quarterly, 81, 49–65. Hirst, P., & Thompson, G. (2002). The future of globalization. Cooperation and Conflict, 37(3), 247–265. Huber, E., & Solt, F. (2004). Successes and failures of neoliberalism. Latin American Review, 39(3), 150–164. Kandel, W., & Cromartie, J. (2004). New patterns of Hispanic settlement in rural America. Washington, DC: U.S. Department of Agriculture. Kandel, W., & Parrado, E. (2004). U.S. industrial transformation and new Latino migration. Paper presented at the annual meeting of the Population Association of America, Minneapolis, MN. Markides, K. S., & Coreil, T. (1986). The health of Hispanics in the southwestern United States: An epidemiologic paradox. Public Health Report, 101, 253–265.

28    R. R. VERDUGO Markides, K. S., & Eschbach, K. (2005). Aging, migration, and mortality: Current status of research on the Hispanic paradox. Journal of Gerontology, 60B(11), 68–75. Marmot, M. G., Adelstein, A. M., & Bulusu, L. (1984). Lessons from the study of immigrant mortality. Lancet, 2, 1455–1457. Massey, D.D., Arrango, J., Hugo, G., Kaouaouci, A, Peligrino, A., & Taylor, J. E. (1993). Theories of international migration: A review and appraisal. Population Development Review, 19, 431–466. McWilliams, C. (1990 [1948]). North from Mexico: The Spanish-speaking people of the U.S. (Updated by M. S. Meier.) New York: Praeger. Menjivar, C. (2000). Fragmented ties: Salvadoran immigrant networks in America. Berkeley: University of California Press. Moore, J., & Pachon, H. (1985). Hispanics in the United States. Englewood Cliffs, NJ: Prentice-Hall. Pablos-Mendez, A. (1994). [Letter to the Editor]. JAMA, 271, 1237–1238. Passel, J. S. (2004). Methodology and assumptions for population estimates and projections by race and ethnicity, 1960 to 1970. Washington, DC: Pew Hispanic Center. Passel, J. S., & Cohn, D. (2009). A portrait of unauthorized immigrants in the United States. Report No. 4.14.2009. Washington, DC: Pew Hispanic Center. Portes, A., & Borocz, J. (1989). Contemporary immigration: Theoretical perspectives and its determinants and modes of incorporation. International Migration Review, 23, 606–630. Portes, A., & Stepick, A. (1993). City on the edge: The transformation of Miami. Los Angeles: University of California Press. Rivera-Batiz, F. L., & Santiago, C. E. (1998). Island paradox: Puerto Rico in the 1990s. New York: Russell Sage Foundation. Rodriguez, C. (1991). Puerto Ricans: Born in the U.S.A. Boulder, CO: Westview Press. Ruggles, S., Sobek, M., Alexander, T., Fitch, C. A., Goeken, R., Hall, P. K., King, M. et al. (2009). Integrated Public Use Microdata Series: Version 4.0 [Machine-readable database]. Minneapolis: Minnesota Population Center [producer and distributor], 2008. Retrieved from http://usa.ipums.org/usa/ Rumbaut, R. G. (1992). The Americans: Latin American and Caribbean peoples in the United States. In A. Stepan (Ed.), Americas: New interpretive essays (pp. 275–307). New York: Oxford University Press. Rumbaut, R. G. (2006). The making of a people. In M. Tienda & F. Mitchell (Eds.), Hispanics and the future of America (pp. 16–65). Washington, DC: National Academies Press. Sachs, J. (2005). The end of poverty. New York: Penguin Press. Sanchez Korrol, V. E. (1983). From colonia to community: The history of Puerto Ricans in New York City, 1917–1948. Westport, CT: Greenwood Press. Sanchez Korrol, V. E. (1994). In their own image: A history of Puerto Ricans in the U.S.A. In A. Jimeniz (Ed.), Handbook of Hispanic cultures in the United States (pp. 281–301). Houston, TX: Arte Publico Press. Sassen, S. (2006). Cities in a global economy. Thousand Oaks, CA: Pine Forge Press. Scribner, R. (1996). Paradox as a paradigm—The health outcomes of Mexican Americans. [Editorial.] American Journal of Public Health, 86, 303–305.

Hispanics, Immigrants, and the Growth of the U.S. Population    29 Shai, D., & Rosenwaike, I. (1987). Mortality among Hispanics in metropolitan Chicago: An examination based on vital statistics data. Journal of Chronic Disease, 40, 445–451. Smith-Nonini, S. (2002). Nadie sobe, nadie supo: El programa federal H2A y la explotacion de man de obra mediada por el estado. Relaciones, 27, 55–86. Sorlie, P. D., Backlund, E., Johnson, N. J., & Rogot, E. (1993). Mortality by Hispanic status in the United States. JAMA, 370, 2464–2468. Stephen, E. H., Foote, K., Hendershot, G. E., & Schoenborn, C. A. (1994, February 14). Health of the foreign-born population: United States, 1989–90. Advanced Data Vital Statistics, 241. Stiglitz, J. E., & Charlton, A. (2005). Fair trade for all: How trade can promote development. New York: Oxford University Press. Stull, D. D., Broadway, M. J., & Griffith, D. (Eds.). (1995). Any way you cut it: Meat processing and small town America (pp. 129–151). Lawrence: University of Kansas Press. Sullivan, T. A., & Pedraza-Bailey, S. (1979). Differential success among Cuban-American and Mexican-American immigrants: The role of policy and community. Chicago: National Opinion Research Center. Suro, R., & Passel, J. S. (2003). The rise of the second generation: Changing patterns of Hispanic population growth. Washington, DC: Pew Hispanic Center. Tienda, M., & Mitchell, F. (2006). Multiple origins, uncertain destinies: Hispanics and the American future. Washington, DC: National Academies Press Valdes, D. N. (1982). El pueblo Mexicano en Detroit y Michigan: A social history. Detroit, MI: Wayne State University Press. Valdes, D. N. (1991). Al norte: Agriculture in the Great Lakes region, 1917–1970. Austin: University of Texas Press. Valdes, D. N. (2000). Barrios nortenos: St. Paul and midwestern Mexican communities in the twentieth century. Austin: University of Texas Press. Vargas, Z. (1993). Proletarians of the north: A history of Mexican industrial workers in Detroit and the Midwest, 1917–1933. Berkeley: University of California Press. Verdugo, R. R. (2006). A report on the status of Hispanics in education: Overcoming a history of neglect. Washington, DC: National Education Association. Verdugo, R. R., & Swanson, D. (2011). Immigration and its impact on demographic change in Spain. The Open Demography Journal, 10, 22–33. Walton, M. (2004). Neoliberalism in Latin America: Good, bad, or incomplete? Latin American Review, 39(3), 165–184. Wilson, W. J. (1996) When work disappears: The world of the urban poor. New York: Knopf.

This page intentionally left blank.

Chapter 2

Contemporary Migration and the Hispanic Population in the United States Karen A. Woodrow-Lafield University of Maryland

Introduction The magnitude and diversity of the contemporary Hispanic population are linked with dramatic demographic change in the U.S. population profile and with deep introspection on the American identity. Having shown increasing numbers over decades, the Mexican-origin population represents the majority, about 60%, of the U.S. Hispanic population, with Puerto Rican-, Cuban-, and Other-origin populations accounting for much smaller shares. Migration has been a major component of Hispanic population growth, and explanations of migration are contrasting by origin. MexicoU.S. migration has become so culturally and socially accepted within Mexican communities that many young men aspire to migrate northward to achieve economic mobility rather than being oriented toward domestic strategies or completing school (Kandel & Massey, 2002). This yearning for The Demography of the Hispanic Population, pages 31–46 Copyright © 2012 by Information Age Publishing All rights of reproduction in any form reserved.

31

32    K. A. WOODROW-LAFIELD

economic opportunity is responsive to U.S. labor market demands despite the well-known risks of the border crossing. With collective redefinition of temporary migration as more permanent, demographic diversity and status heterogeneity on legality and generation became greater (Massey, 1995; Massey, Goldring, & Durand, 1994). After legal documents have been obtained, return migration is less likely (Massey & Espinosa, 1997). The case of the Cuban-origin population differs considerably. The initial influx of Cubans occurred in the late 1960s and 1970s after the Cuban Bay of Pigs crisis. High education and skill levels characterized these migrants, many of working and childbearing ages. More Cubans and Haitians arrived in early 1980, including the controversial Marielitos, and parole status was granted to these individuals. Later legislation granted lawful permanent residence. Subsequently, many more Cubans motivated by economic and political reasons sought to come into the United States, and their reception by U.S. authorities was dependent upon reaching and stepping on the mainland under because those intercepted at sea were not allowed into the United States under what became known as the “wet foot, dry foot” policy. Eventually, some Cubans may return or their descendants may migrate to Cuba following political changes. For the Puerto Rican-origin population, demographers refer to “net movement of persons” rather than migration because their choice of residence is less encumbered by immigration constraints. These patterns have varied over time. The “Other” Hispanic-origin category consists of Central American origins (Salvadoran, Nicaraguan, Guatemalan, and Honduran), South American origins (Colombian, Argentinean, Ecuadorian, and other), Spanish origin, and Portuguese origin (Guzmán, 2001;Guzmán & McConnell, 2002). The “Other” Latino- or Hispanic-origin groups showed increases, especially after the 1990s, although their precise share may have been misrepresented in 2000 census data (Chun, Kwon, Williams, & Yu, 2005; Logan, 2001, 2002; Suro, 2002). Motivations for migration are many, and as much economic as political or familial. The Central American diaspora groups of Salvadorans, Nicaraguans, and Guatemalans experienced migration and settlement that have not been easily discernible from official statistics. Many individuals were of unresolved or unauthorized status, and such smaller subpopulations were not well measured in national surveys, even if individuals vulnerable to deportation were participating in censuses or surveys. Nevertheless, researchers drew usefully from the 1990 and 2000 censuses (Lopez, Popkin, & Telles, 1996) and qualitative studies (Hagan, 1994; Mahler, 1996; Repak, 1995). Hispanic Migration, 1970–2000 Before changes to U.S. immigration laws in the 1960s and 1970s, migration from Latin American countries was already rising as a consequence of

Contemporary Migration and the Hispanic Population in the United States     33

regional economic policies, labor demand, and geographic proximity. The land area thought to be Aztlan, the mythical place of origin of the Aztec peoples, is the southwestern border areas that the United States took over after the Mexican-American War of 1846–1848 in the Treaty of Guadalupe Hidalgo in 1848 and the Gadsden Purchase in 1853. Many Mexican-origin persons were thus added to Spanish-origin descendants already living in the United States, although the census did not yet include such a question for counting Hispanics (Bean & Tienda, 1987; Bean, Corona, Turián, & Woodrow-Lafield, 1998). Mexican migrants were later recruited for working on the construction of the railroads, and they became entrenched in the urban industrial labor force in the North, although the greatest concentration of Mexican workers was in the Southwest. Mexican workers filled the gap after restrictive immigration legislation in 1921 until the Great Depression. Without demand for their labor, about one half million Mexicans then returned to Mexico, mostly forced to repatriate (Bean, de la Garza, Roberts, & Weintraub, 1997; Massey, Alarcon, Durand, & Gonzalez, 1987), often accompanied by their U.S.-born children. The notion of Mexicans as temporary workers evolved in 1942 as the Bracero Accord, and over more than 2 decades, about 4.5 million Mexicans worked in the United States, and some would settle, but the majority would return (Massey et al., 1987). With concerns about illegal immigration, Operation Wetback was a repatriation program in 1954 to deport about 1.1 million Mexicans. In the 1970s, apprehensions of Mexicans attempting to enter without inspection were escalating, and speculation arose about the numbers getting away from border patrol officers and numbers residing in southwestern states and elsewhere. Net population change due to unauthorized migration was emerging for the Hispanic population. The U.S. Hispanic population grew from slightly more than 2 million in 1950 to 9.1 million and 14.6 million in the 1970 and 1980 censuses, respectively, based on the new census question on Hispanic origin (Bean & Tienda, 1987; Choldin, 1986). Census coverage was better in 1980 than in 1970, and the 1990 census showed the Hispanic population at about 22.4 million, a rise that paralleled the rise in the foreign-born population. With passage of the Immigration Reform and Control Act of 1986 (IRCA), the Mexican population became predominantly authorized in status (Woodrow, 1992) with a transformed Mexican population more integrated within American communities, the labor force, schools, and churches (Durand, Massey, & Parrado, 1999). The IRCA legalization programs clearly diminished the number of unauthorized residents over 1989–1991. Among those receiving lawful temporary residence under general provisions and seasonal agricultural worker provisions, the majority were from Mexico and other Central American countries. From one study of INS apprehensions data and surveys of migrants, which enabled estimation of return migration, an estimate of net

34    K. A. WOODROW-LAFIELD

flow for the period 1965 to 1989 was 5.2 million, which was considerably less than the gross flows across the Mexico–U.S. border (Massey & Singer; 1995) and more than the number of Mexicans legalizing under IRCA provisions. The hidden presence had been partially disclosed by demographic studies (Passel & Woodrow, 1984, 1987; Warren & Passel, 1987), and the legalization programs brought further details on individuals through both administrative sources and the Legalized Population Surveys (U.S. Immigration and Naturalization Service, 1992). As of 1990, the Hispanic foreign-born population of 7.8 million may have included 1.5–2.5 million unauthorized residents after about 2.7 million Hispanic and non-Hispanic aliens gained temporary and permanent status under IRCA provisions (Passel & Woodrow, 1987). There was uncertainty about the precise number of Hispanic unauthorized residents because individuals who were eventually granted status in 1991 under agricultural worker provisions may or may not have been residing in the United States. Additionally, many other individuals became legalized long after the initial program through immigration policies and settlements of classaction lawsuits related to applying under IRCA. These administrative data demonstrated the impact of long-term residence of unauthorized migrants. The number of Hispanics reached a surprising 35.3 million in 2000, and about 14.2 million were foreign-born. About 15% to 20% of all Hispanics were likely of unauthorized status. About 75% to 80% of unauthorized residents were likely Hispanic or Latino, and unauthorized immigrants were primarily from Mexico and other Central America with non-Mexican origins perhaps accounting for 40% (Bean, Van Hook, & Woodrow-Lafield, 2002). The remarkable increase in the Hispanic population in the 2000 census is attributable to greater migration than anticipated by census demographers. For the early 1990s, migrant workers were staying longer in the United States (Cornelius, 1992), perhaps because IRCA had allowed many to obtain documents and bring their family members, and work situations in urban areas were more stable than seasonal work. Also, apparently, migrants were continuing to come, and unauthorized migration persisted in the 1990s partly because post-IRCA enforcement measures were gradually imposed at the southern border (Bean et al., 1998; Donato & Massey, 1992). The number of unauthorized Mexican residents increased from 1.4 million in 1990 to 2.5 million by 1996 (Bean et al., 2002). Not only unauthorized but also authorized Mexicans were increasing in number as family members joined those who had gained amnesty under IRCA provisions, and analyses of the Legalized Population Surveys had shown they expected their family members would intend to immigrate (Durand, Massey, & Zenteno, 2001; Woodrow-Lafield, 1995). For 1990–1996, net growth in the size of the Mexico-born population in the United States was nearly 2 million, including 510,000 legal immigrants, 630,000 unauthorized immigrants, 210,000

Contemporary Migration and the Hispanic Population in the United States     35

family members of generally legalized immigrants, and 550,000 migrants who had legalized under agricultural provisions (Binational Migration Study, 1997). This last statistic may have been conservative, and indeed, extrapolation of these findings would have been conservative in gauging change for the full decade. This was one of few empirical studies of undocumented migration for 1995–1999, as elaborated in Bean et al. (2002). A preliminary INS analysis was subpoenaed by a congressional office in 2000, but that study was for the period 1987–1997 and related to Warren (1997). Several immigration policies after IRCA made it possible for persons without legal documents to remain here in less precarious situations. One was the Section 245(i) provision added to the Immigration and Nationality Act under which individuals could file during October 1, 1994–January 15, 1998 (as extended from September 30, 1997). Eligible individuals, despite U.S. presence in an unauthorized or indefinite status, could apply for adjustment of status at an INS office and eventually obtain their immigrant visas in the United States and a pathway to citizenship. This was not a legal provision granting amnesty, because the individuals already had visa eligibility through either a labor certification or through a visa petition associated with a naturalized spouse, child, or parent. The law merely eased the visa process by allowing them to pay a fine and avoid having to travel abroad to the U.S. Department of State consular office through which their visa application had been filed. The largest impact may have occurred in immediate relatives categories that are exempt from numerical limitation. About one half million individuals may have utilized this processing mechanism over 1995–1998. As these provisions were permitting application and easier transition to lawful permanent residence upon receiving a visa notice, congressional debate arose over native-born persons’ and immigrants’ access to public benefit programs. Passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (also known as the 1996 Welfare Reform Act, PRWORA) and the 1996 Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) reflected definition of both legal and illegal immigration as problematic from a fiscal cost standpoint. U.S. welfare reform and restrictive legislation of the mid-1990s imposed restrictions on public benefits, specified income requirements for sponsoring family members, and made this financial responsibility effective for 10 years or until the family member became naturalized. Post-1996 immigrants would have more incentives to naturalize: for access to public benefits, to relieve their sponsoring family members of obligatory financial support, and for self-protection against deportation. IIRIRA also implemented several measures involving noncitizens and unauthorized residents. Shortly after passage of IIRIRA, certain Hispanic-origin groups benefited from the Nicaraguan Adjustment and Central American Relief Act (NACARA) of November 19, 1997, in gaining per-

36    K. A. WOODROW-LAFIELD

manent residence and a pathway to citizenship. This pertained to certain aliens who were long-term illegal residents and likely to be adversely affected by IIRIRA changes on hardship relief rules. Our law has for decades permitted the Attorney General to allow long-term illegal residents to stay on a case-by-case basis if their removal would cause undue hardship. The rules governing this hardship relief (known until recently as suspension of deportation) have changed over time, and in 1996, Congress significantly toughened hardship standards in IIRIRA amendments that could have precluded any hardship relief for many aliens who had resided here since well before IIRIRA was enacted. NACARA allowed approximately 150,000 Nicaraguans and 5,000 Cubans to adjust to permanent resident status without having to make any hardship showing. In addition, it allowed approximately 200,000 Salvadorans and 50,000 Guatemalans, as well as certain aliens from the former Soviet Union, to seek hardship relief under more lenient hardship rules than existed prior to IIRIRA amendments. Thus, NACARA enabled regularization of about 400,000 Hispanics. Census 2000-based studies revealed considerable change during the later 1990s that was not detected earlier, and a mid-decade study examined migration over more than a decade (Passel & Suro, 2005). Unauthorized immigration was higher in the 1990s than had been officially allowed (Bean, Corona, Turián, Woodrow-Lafield, & Van Hook, 2001; Passel, 2002, 2005, 2006), including more of Hispanic origin, due to economic reasons (low unemployment and high returns to low-skilled workers) and altered migrant settlement in response to enforcement and the greater risks of border crossing (Binational Migration Study, 1997; Durand et al., 2001; Marcelli & Cornelius, 2001). The increasing Mexican migration with longer stays occurred at the same time there were elevated rates of Mexican return migration for 1990–1994 (Durand et al., 2001), perhaps because IRCA-legalized individuals were then able to return more easily. Migration behavior was increasing in prevalence throughout Mexican communities with engagement of new sending communities (Massey et al., 1994). Immigration and Legalization, 2000–2010 Even after the 1998 lapse of the Section 245 (i) provision of the INA, new immigration policies allowed residents to change status from unauthorized to authorized. The Legal Immigration Family Equity Act (LIFE) of 2000 applied to several categories, including those individuals physically present on December 21, 2000, who qualified for permanent residency but were ineligible to adjust because of an immigration status violation. After Congress implemented the LIFE Act in December 21, 2000, about 400,000 individuals applied under Section 245(i) before expiration in April 2001

Contemporary Migration and the Hispanic Population in the United States     37

(Migration News, 2001) and perhaps as many more with extension of the deadline (Migration News, 2002), contributing to an estimated 1.5 million immigrant admissions in fiscal year 2001. The LIFE Act also was “a corrective policy response” to three “late amnesty” lawsuits (CSS v. Meese, LULAC v. Reno, or INS v. Zambrano) in regard to unauthorized aliens who had not been able to apply for IRCA legalization and had been living here in uncertainty, specifically applicable to persons who had filed for class membership before October 1, 2000. Settlements in those lawsuits stipulated that applications would be allowed from those individuals who had traveled abroad and thus did not meet the continuous residence requirement. For individuals eligible under the LIFE legalization provisions with their approval dependent on demonstrating knowledge of English and civics, their spouses and unmarried children would also be protected against certain categories of removal and given work authorization if they showed having begun unlawful U.S. residence before December 1, 1988. According to the 2008 Yearbook for Immigration Statistics (OIS), there were approximately 700,000 nonimmigrant admissions under LIFE over 1999–2008. These individuals would have had many years of residence upon eventually becoming legal immigrants. Considering the likely numbers adjusting status under Section 245(i) provisions over 1995–1998, the total number benefiting from these various policies is likely to be 1 million, essentially raising the legalization impact of IRCA to 4 million. Even 2 decades after IRCA, others were able to apply as IRCA beneficiaries following a December 2008 class-action settlement that opened up IRCA amnesty for tens of thousands who entered on valid visas and were without legal status between 1982 and 1988. That lawsuit challenged the rule that applicants had to show their shift to legal status was “known to the government.” Post-2000 Census: Counts, Estimates, and Projections The population estimate for the Hispanic population was 45.4 million for July 1, 2009, and the projection of 49.7 million Hispanics for 2010 (U.S. Census Bureau, 2010) was close to the Census 2010 count of 50.5 million Hispanics. There were some indications of lower migration of Hispanics and Asians due to the economic recession late in the decade (Yen, 2009). Understanding the demographic and social processes for the contemporary Hispanic population involves considering the demographic and ethnic composition of recent and earlier Hispanic immigrants and how fertility patterns of Hispanics and immigrants may converge with low fertility patterns of non-Hispanic White women (Hill & Johnson, 2002; Ventura, Mar-

38    K. A. WOODROW-LAFIELD

tin, Curtin, Mathews, & Park, 2000). Hispanic population growth greatly impacts school-aged populations and educational systems in new destination communities across the nation and traditional destination states and metropolitan areas in the West and South, with associated impacts for areas of declining school populations. Forecasting increases in the Hispanic population challenged U.S. demographers, while world migration increased from 156 million in 1990 to 214 million migrants in 2010 (United Nations, 2009). Population estimates and projections often misgauged the Hispanic population. For example, four separate Hispanic population projections for July 1, 2000, demonstrated the importance of the migration assumption, especially as to undocumented migration. Fortunately, each successive projected figure was closer to the 2000 census count of 35.3 million Hispanics, or 12.5%: 25.2 million or 9.4% (U.S. Census Bureau, 1986), 30.6 million or 11.1% (U.S. Census Bureau, 1993), 31.4 million or 11.4% (Day, 1996), and 32.5 million or 11.8% (Hollmann, Mulder, & Kallan, 2000). The earliest figure was based on methodologies and assumptions before estimates of undocumented migration in the 1970s were developed (Warren & Passel, 1987). The others allowed insufficiently for migration. Adding to the initial surprise of the high Census 2000 count overall, the census count of 35.3 million for the Hispanic population (Grieco & Cassidy, 2001) exceeded by more than 3 million the figure that census demographers had projected. Census counts evoked allegations that migration had been poorly measured in the 1990s (Cohn, 2001; Sum, Harrington, & Fogg, 2001). As had happened with the 1980 census (Passel, Siegel, & Robinson, 1982), the foreign-born count of 31.1 million would be higher than either the March 2000 CPS estimate (28.4 million; Lollock, 2001) or the national estimate (26.8 million; U.S. Census Bureau, 2000). The 2000 census counts of the Hispanic and foreign-born populations were not implausible given the extraordinary economic scenario in the 1990s with high returns to low-skilled workers. The growth of the Hispanic population over 1990–2000 heralded revisions to post-2000 national population projections by race and Hispanic origin. By late 2002, the initial mirroring of 2000 census results came in the International Data Bank (http://www.census.gov/ipc/www/idb/), with U.S. population projections calculated with the 2000 census base population and net immigration assumptions as consistent with official estimates in the 1990s. That projected population figure for 2020 (336,031,546) was about 3 million higher than the sum of the 2020 population figure from Hollmann et al. (2000; 324,926,000) and the Census 2000 error of closure (about 7,000,000). The 2008 population projections figure for 2020 was still higher, 341,387,000, reflecting assumptions of higher net international migration. The supplemental 2009 projections cites 334 million, assuming

Contemporary Migration and the Hispanic Population in the United States     39

constant international migration or 347 million, assuming higher international migration (Ortman & Guarneri, 2009). As the mid-century point approaches, the future population of the United States is likely to show minority groups (all except non-Hispanic Whites) in the majority (Hollmann et al., 2000). The timing of that majority-minority crossover has been predicted for varying dates according to immigration trends and may be 2042, as indicated by the population projections released in 2008, or later, as suggested by supplemental population projections in 2009, with assumptions of lower net international migration (Ortman & Guarneri, 2009). With a constant level of 1 million, the Hispanic population would more than double over 2000–2050, but the Hispanic population would increase more, even if immigration were declining, given the age distribution and expected higher fertility rates, than non-Hispanic Whites. Levels of net undocumented migration are not easily anticipated. Leading experts’ understanding and research guide their judgments on which to base assumptions. Key decisions are whether to rely upon trend data and evaluation of policies meant to control migration and reduce or avoid population change due to unauthorized migration. Federal agencies implement public policy in accordance with administration goals, which may constrain envisioning scenarios. For the middle series and the interval 1999 to 2020, the 2000 projections (Hollmann et al., 2001) used international migration amounts after consideration of current immigration trends by sending country with certain guiding assumptions. The first related to the long-term demographic consequences of the IRCA, deeming this flow, largely of Mexicans and Central Americans, as transitory. The report stated that legal migration from Mexico was assumed to return to the levels of the early 1990s by 2010, but this clearly did not occur. This was a flawed judgment because potential reunifications may not have been fulfilled quickly, and spouses who benefited from someone gaining legal status through IRCA provisions may have later sought to bring their parents and other relatives through numerically limited categories. A second assumption was technically correct: that immigration policy would not change as to the quantity of immigrant visas available in numerically limited legal categories. However, as noted earlier in discussing LIFE and NACARA, immigration policy would change in regard to the range of immigration statuses that allow long-term lawful residence. A third assumption was adequate: that the flow of refugees to the United States would tend to decline between 1998 and 2020; in part, this assumption may have been realized through the greater scrutiny of allowing refugees, given national security concerns after 2001. A fourth guiding assumption was problematic. The report stated, “Undocumented migration of people born in Mexico and Central America is viewed primarily as a function of the degree of success in controlling

40    K. A. WOODROW-LAFIELD

the Southwest border, and is not projected to change from levels assumed for the 1990’s base series.” This perspective is very limited because border enforcement is at the border, but, for those who succeeded in crossing the border, those border controls would result in reducing migrant trips and prolonging stays, as studies found increased net population change due to undocumented migration. Finally, those projections allowed for variability in legal migration from places other than Mexico, Central America, and refugee sources, depending on recent observed trends and the perceived demographic capacity of source countries to supply migrants. Basically, the projections for the middle series incorporated a modest rise in immigration from 1998 to 2002, a somewhat greater decline by 2010, and a gradual rise by 2020. These assumptions are conservative, presuming the U.S. government would achieve some success in controlling migration through Southwest border strategies. However, the model was not dynamic, and the methodology was not sensitive to the cumulative causation of migration that would lead to major shortcomings in projected population figures, as Massey and Zenteno (1999, p. 5335) stated, “demographers would be better served by developing dynamic projection algorithms that take into account the fact that each cohort of migrants is preceded by others who have arrived before, yielding a stock of human and social capital that raises the odds of future movement to dynamize the flows.” As Massey and Zenteno (1999) summarized, census population projections over recent decades were flawed in allowing for amounts of net immigration, sometimes based on administrative data and sometimes drawing on empirical research. There may have been better estimates of net immigration in the 1980s than in the 1970s, but this was certainly not the case for the 1990s relative to the 1980s. Soon after initial Census 2000 results, the INS, the lead agency on unauthorized statistics at the time of the 2000 census, quickly issued a memo raising its “estimate of the resident illegal population for 2000 from a range of about 5.5 to 6.5 million to a range of about 6.5 million to 7.5 million” (USINS, 2001). That statement continued, “INS has monitored and assessed the press reports of 9 to 11 million or more unauthorized residents in the country, and to date we have found no credible evidence to support these figures.” Eventually, after several studies affirmed such a likely range, similar findings were published (USINS, 2003). Conclusion Apart from the nuts and bolts of making estimates and projections for the Hispanic and overall population, broad findings are emerging as to the spatio-temporal change in net migration in the United States. There are large,

Contemporary Migration and the Hispanic Population in the United States     41

geographically contiguous regions of net in-migration, and there are similarly large geographically contiguous regions of net out-migration (Johnson, Voss, Hammer, Fuguitt, & McNiven, 2005). The latter, including the Great Plains, are likely unaffected by Hispanic migration and population growth in contrast with those regions in the Southwest and Florida that are showing net in-migration and becoming majority-minority through Hispanic migration. During the 2000–2010 decade, the Hispanic population seemed to be becoming more widely dispersed, and both new and traditional destination areas of Hispanic settlement depend on immigration for population growth due to domestic out-migration (Lichter & Johnson, 2009). An open question concerns the level of efficacy of any nation-state policies in controlling migration and settlement and how successful new ones might be for controlling migration and settlement in the future. Period effects may arise from changes to enforcement strategies and legislative reform of enforcement and lawful migration. Population projections may be flawed due to numerous complicating aspects of economic, social, and demographic behavior. Making assumptions about migration components brings uncertainty. U.S. policies were apparently not successful “in deterring undocumented migrants from seeking to enter the United States,” “in discouraging former undocumented migrants from undertaking return trips,” “in preventing illegal migrants from successfully crossing the border,” “in convincing settled undocumented migrants to return home,” or “in preventing employers from hiring unauthorized workers” (Massey, 1999), a recurrent theme (Massey, Durand, & Malone, 2002). More definitive analysis is always needed for this major segment of U.S. immigration policy. Official studies involved “several reports of the Government Accountability Office, formerly the U.S. General Accounting Office concerning evaluation of the deterrent effect of border strategies. Following issuance of the agency’s plans for a major study, an evaluation of the Southwest border strategy in 2002–2003 was underway in the former INS before creation of the Department of Homeland Security. Immigration from Latin American countries, especially Mexico and Central America, rivals the historical significance of the European migration a century ago. The Hispanic population of the United States has changed through authorized and unauthorized immigration, and uncertainty surrounds the precise parameters, although many unauthorized residents have progressed into lawful status through any of several mechanisms. The back-and-forth, circularity, or churning of migrants and potential settlers complicate the demography of the Hispanic population, and forecasters must attend to border-crossing successes and failures in the context of intensified border enforcement and apprehension machinery. Research on the demography of the Hispanic population must be sensitive to impacts

42    K. A. WOODROW-LAFIELD

from immigration policies as well as resilient to presuming efficacy in controlling migration. References Bean, F. D., Corona, R., Tuirán, R., & Woodrow-Lafield, K. A. (1998). Quantification of migration between Mexico and the United States. Migration between Mexico and the United States: Binational study, Volume I, thematic chapter (pp. 1–90): Mexican Ministry of Foreign Affairs and U.S. Commission on Immigration Reform. Retrieved from http://www.utexas.edu/lbj/uscir/binpapers/ v1-1bean.pdf Bean, F. D., Corona, R., Tuirán, R., Woodrow-Lafield, K. A., & Van Hook, J. (2001). Circular, invisible, and ambiguous migrants: Components of difference in estimates of the number of unauthorized Mexican migrants in the United States. Demography, 38(3), 411–422. Bean, F. D., de la Garza, R., Roberts, O. B. R., & Weintraub, S. (1997). At the crossroads: Mexico and U.S. immigration policy. London: Rowman & Littlefield. Bean, F. D. & Tienda, M. (1987). The Hispanic population of the United States. New York: Russell Sage Foundation. Bean, F. D., Van Hook, J., & Woodrow-Lafield, K. (2002, January 24). Estimates of numbers of unauthorized migrants residing in the United States: The total, Mexican, and non-Mexican Central American unauthorized populations in mid-2001. Washington, DC: Pew Hispanic Center http://pewhispanic.org/reports/report. php?ReportID=2 Binational Migration Study. (1997). Binational study on migration between Mexico and the United States. Mexico City and Washington, DC: Mexican Ministry on Foreign Affairs; U.S. Commission on Immigration Reform. Choldin, H. M. (1986). Statistics and politics: The “Hispanic” issue in the 1980 census. Demography, 23(3), 403–418. Chun, S., Kwon, J., Williams, C., & Yu, E. (2005, January 1). Latino-origin populations revisited: Estimating the Latino-origin populations at the national level and for selected states, counties, cities, and metro Chicago. Research Reports, Vol. 2005. South Bend IN: University of Notre Dame, Institute for Latino Studies. Cohn, D. (2001, March 18). Illegal residents exceed estimate. The Washington Post, p. A1. Cornelius, W. A. (1992, March 5). Potential impacts of North American free trade on Mexican migration to California. Presented at the International Conference on a North American Free Trade Agreement: Implications for California Agriculture. UC Agricultural Issues Center, Los Angeles. Day, J. C. (1996). Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050. Current Population Reports, P25-1130. Washington, DC: U.S. Census Bureau, U.S. Government Printing Office. Dillon, S. (1997, August 31). U.S.-Mexico study sees exaggeration in migration data. The New York Times, p. A1. Donato, K., & Massey, D. S. (1992). Stemming the tide? Assessing the deterrent effects of the Immigration Reform and Control Act. Demography, 29(2), 139–157.

Contemporary Migration and the Hispanic Population in the United States     43 Durand, J., Massey, D. S., & Parrado, E. A. (1999, September 2). The new era of Mexican migration to the United States. Journal of American History, 86, 518–536. Durand, J., Massey, D. S., & Zenteno, R. (2001). Mexican immigration to the United States: Continuities and changes. Latin American Research Review, 36(1), 107–127. Grieco, E. M., & Cassidy, R. C. (2001). Overview of race and Hispanic origin. Census 2000 Brief, CENBR/01-1. Washington, DC: U.S. Bureau of the Census Guzmán, B. (2001). The Hispanic population. Census 2000 Brief, CENBR/01-3. Washington, DC: U.S. Bureau of the Census. Guzmán, B., & McConnell E. D. (2002). The Hispanic population: 1990–2000 growth and change. [Special Issue] Population Research and Policy Review, 21(1–2), 109–128. Hagan, J. M. (1994). Deciding to be legal: A Maya community in Houston. Philadelphia: Temple University Press. Hill, L. E., & Johnson, H. P. (2002). Understanding the future of Californians’ fertility: The role of immigrants. San Francisco: Public Policy Institute of California. Hollmann, F. W., Mulder, T. J., & Kallan, J. E. (2000, January 13). Methodology and assumptions for the population projections of the United States: 1999 to 2100. Population Division Working Paper No. 38. Washington, DC: U.S. Bureau of the Census. Johnson, H. (1996). Undocumented immigration to California: 1980–1993. San Francisco: Public Policy Institute of California. Johnson, K. M., Voss, P. R., Hammer, R. B., Fuguitt, G. V., & McNiven, S. (2005, November). Temporal and spatial variation in age-specific net nigration in the United States. Demography, 42(4), 791–812. Kandel, W., & Massey D. S. (2002). The culture of Mexican migration: A theoretical and empirical analysis. Social Forces, 80(3), 981–1004. Lichter, D. T., & Johnson K. M. (2009, Fall). Immigrant gateways and Hispanic migration to new destinations. International Migration Review, 43(3), 496–518. Logan, J. (2001, September 10). The new Latinos: Who they are, where they are. Report prepared at the Lewis Mumford Center for Comparative Urban and Regional Research, University at Albany, SUNY. Logan, J. (2002, May 8). Hispanic populations and their residential patterns in the metropolis. Report prepared at the Lewis Mumford Center for Comparative Urban and Regional Research, University at Albany, SUNY. Lollock, L. (2001). The foreign-born population in the United States: March 2000. Current Population Reports, P20-534. Washington, DC: U.S. Census Bureau. Lopez, D. E., Popkin, E., & Telles, E. (1996). Central Americans: At the bottom, struggling to get ahead. In R. Waldinger & M. Bozorgmehr (Eds.), Ethnic Los Angeles (pp. 279–304). New York: Russell Sage Foundation. Mahler, S. (1996). Salvadorans in suburbia: Symbiosis and conflict. New York: Allyn and Bacon. Marcelli, E. A., & Cornelius, W. A. (2001, Fall). The changing profile of Mexican migrants to the United States: New evidence from California and Mexico. Latin American Research Review, 36(3), 105–131. Massey, D. S. (1995). The new immigration and ethnicity in the United States. Population and Development Review, 21(3), 631–652.

44    K. A. WOODROW-LAFIELD Massey, D. S. (1999). International migration at the dawn of the twenty-first century: The role of the state. Population and Development Review, 25(2), 303–322. Massey, D. S., Alarcon, R., Durand, J., & Gonzalez, H. (1987). Return to Aztlan: The social process of international migration from western Mexico. Berkeley: University of California Press. Massey, D. S., Durand, J., & Malone, N. J. (2002). Beyond smoke and mirrors: Mexican immigration in an era of economic integration. New York: Russell Sage Foundation. Massey, D. S., & Espinosa, K. E. (1997). What’s driving Mexico–US migration? A theoretical, empirical, and policy analysis. American Journal of Sociology, 102(4), 939–999. Massey, D. S., Goldring, L., & Durand, J. (1994). Continuities in transnational migration: An analysis of nineteen Mexican communities. American Journal of Sociology, 99(6), 1492–1533. Massey, D. S., & Singer, A. (1995, May). New estimates of undocumented Mexican migration and the probability of apprehension, Demography, 32(2), 203–214. Massey, D. S., & Zenteno, R. (1999). The dynamics of mass migration. Proceedings of the National Academy of Sciences, 96, 5328–5335. Migration News. (2001, May). INS: Border, 245, Asylum, Detention. Vol. 8. No 5. Retrieved from http://migration.ucdavis.edu/mn/more.php?id=2376_0_2_0 Migration News. (2002, February). Anti-Terrorism. Vol. 9. No 2. Retrieved from http://migration.ucdavis.edu/mn/more.php?id=2549_0_2_0 National Center for Education Statistics. (2000). Projections of education statistics to 2010. Washington DC: U.S. Department of Education. Ortman, J. M., & Guarneri, C. E. (2009). United States population projections: 2000 to 2050. Retrieved from http://www.census.gov/population/www/projections/ analytical-document09.pdf Passel, J. S. (1985, October 15) Report on expert working group on estimates of undocumented immigration. Unpublished internal memorandum. Washington, DC: U.S. Bureau of the Census. Passel, J. S. (1999). Undocumented immigration to the United States: Numbers, trends, and characteristics. In D. W. Haines & K. E. Rosenblum (Eds.), Illegal immigration in America: A reference handbook (ch. 3, pp. 27–111). Westport, CT: Greenwood Press. Passel, J. S. (2002, May 22). New estimates of the undocumented population in the United States. Migration Information Source. Retrieved from http://www.migration information.org/USfocus/display.cfm?id=19 Passel, J. S. (2005, June). Unauthorized migrants: Numbers and characteristics. Washington, DC: Pew Hispanic Center. Passel, J. S. (2006, March 7). Size and characteristics of the unauthorized migrant population in the United States. Washington, DC: Pew Hispanic Center. Passel, J. S., Siegel, J. S., & Robinson, J. G. (1982). Coverage of the national population by age, sex, and race in the 1980 census: Preliminary estimates by demographic analysis. P-23, No. 115. Washington, DC: U.S. Government Printing Office. Passel, J. S., & Suro, R. (2005, September 27). Rise, peak, and decline: Trends in U.S. immigration 1992–2004. Washington, DC: Pew Hispanic Center.

Contemporary Migration and the Hispanic Population in the United States     45 Passel, J. S., & Woodrow, K. A. (1984). Geographic distribution of undocumented immigrants: Estimates of undocumented aliens counted in the 1980 census by state. International Migration Review, 18, 642–671. Passel, J. S., & Woodrow, K. A. (1987). Change in the undocumented alien population in the United States, 1979–1983. International Migration Review, 21, 1304–1334. Repak, T. A. (1995). Waiting on Washington: Central American workers in the nation’s capital. Philadelphia: Temple University Press. Sum, A., Harrington, P., & Fogg, N. et al. (2001, February). An analysis of the preliminary 2000 census estimates of the resident population of the U.S. and their implications for demographic, immigration, and labor market analysis and policymaking. Boston: Northeastern University, Center for Labor Market Studies. Suro, R. (2002, May 9). Counting the “Other” Hispanics: How many Colombians, Dominicans, Ecuadorians, Guatemalans, and Salvadorans are there in the United States? Washington, DC: Pew Hispanic Center. United Nations. (2009). Trends in international migrant stock: The 2008 revision. United Nations Database, POP/DB/MIG/Stock/Rev.2008. United Nations, Department of Economic and Social Affairs, Population Division. U.S. Census Bureau. (1986). Projections of the Hispanic population: 1983 to 2080. Current Population Reports, Series P-25, No. 995. Washington, DC: Government Printing Office. U.S. Census Bureau. (1989). Projections of the population of the United States by age, sex, race: 1980 to 2080. Current Population Reports, Series P-25, No. 1018. Washington, DC: Government Printing Office. U.S. Census Bureau. (1993). Population projections of the United States, by age, sex, race, and Hispanic origin: 1992 to 2050. Current Population Reports, Series P-25, No. 1092. Washington, DC: Government Printing Office. U.S. Census Bureau. (2000). Foreign-born resident population estimates of the United States by sex, race, and Hispanic origin: April 1, 1990 to July 1, 1999 (Consistent with the 1990 population estimates base). Population Estimates Program, Population Division, Washington, D.C. April 11. U.S. Census Bureau. (2010). Projections of the population by net international migration series, race, and Hispanic origin for the United States: 2010 to 2050. Table 4 (NP2009-T4), Population Division. Release Date: December 16, 2009. Washington, DC: Government Printing Office. U.S. Immigration and Naturalization Service. (1992). Immigration Reform and Control Act: Report on the legalized alien population. Washington, DC: U.S. Government Printing Office. U.S. Immigration and Naturalization Service. (2001). New census data and INS estimates of the unauthorized immigrant population. Public Affairs, May 21. U.S. Immigration and Naturalization Service. (2003). Estimates of the unauthorized immigrant population residing in the United States: 1990 to 2000. January 31. Ventura, S. J., Martin, J. A., Curtin, S. C., Mathews, T. J., & Park, M. M. (2000). Births: Final data for 1998, national vital statistics Reports. Vol. 48 No. 3. Washington, DC: Department of Health and Human Services, Center for Disease Control, National Center for Health Statistics.

46    K. A. WOODROW-LAFIELD Warren, R. (1997). Estimates of the undocumented immigrant population residing in the United States: October 1996. Washington, DC: U.S. Immigration and Naturalization Service. Warren, R., & Passel, J. S. (1987). A count of the uncountable: Estimates of undocumented aliens counted in the 1980 United States census. Demography, 24, 375–393. Woodrow, K. A. (1992). A consideration of the effect of immigration reform on the number of undocumented residents in the United States. Population Research and Policy Review, 11, 117–144. Woodrow, K. A., & Passel, J. S. (1990). Post-IRCA undocumented immigration to the United States: Assessment based on the June 1988 CPS. In F. D. Bean, B. Edmonston, & J. S. Passel (Eds.), Undocumented migration to the United States: IRCA and the experience of the 1980s (ch. 2).Washington, DC: The Urban Institute Press. Woodrow-Lafield, K. A. (1995). Potential sponsorship by IRCA-legalized immigrants. Washington, DC: U.S. Commission on Immigration Reform. Retrieved from http://www.utexas.edu/lbj/uscir/respapers/ps-s94.pdf Yen, H. (2009, May 14). Growth of Hispanic, Asian population slows unexpectedly, Census Reports. The Washington Post, p. A1.

Chapter 3

The Latino Paradox Rogelio Sáenz University of Texas at San Antonio Trinidad Morales Texas A&M University

Abstract Over the last 3 decades, research findings have supported the unexpected finding that, despite their low socioeconomic standing, Latinos have mortality rates that are comparable or even more favorable than those of Whites. Three explanations have generally been proposed to account for the Latino paradox (also known as the epidemiological paradox). This chapter provides an overview of the literature on the Latino paradox and draws attention to key findings and theoretical, methodological, and data challenges that need to be addressed. Furthermore, the chapter also discusses some of the findings that have observed the immigrant mortality paradox outside of the United States. In addition, the chapter also points out paradoxes involving Latino immigrants that have been observed in other areas aside from mortality and health. Finally, the chapter concludes with a discussion of the direction that future research needs to take in order to advance our understanding of the Latino paradox.

The Demography of the Hispanic Population, pages 47–73 Copyright © 2012 by Information Age Publishing All rights of reproduction in any form reserved.

47

48   R. SÁENZ and T. MORALES

One of the most established research findings in the sociological and demographic literature is the negative relationship between socioeconomic status and mortality. For example, in their classic study on the relationship between socioeconomic status and mortality, Kitagawa and Hauser (1973) found that during the 1960s, persons with higher levels of socioeconomic status had lower levels of mortality. While the association has been consistently observed in the literature, it has gotten stronger since 1960 (Rogers, Hummer, & Nam, 2000). However, there is one major area of research that diverts from this established finding. Extending back to the mid-1970s research has observed that Latinos, despite their low socioeconomic status, have mortality rates comparable or even lower than those of Whites. Teller and Clyburn (1974) first observed this unexpected finding based on Latino infant mortality in Texas. About a decade later, Markides and Coreil (1986) discovered the unexpectedly low infant mortality among Hispanics in the southwestern part of the United States. They described this unforeseen and unexplained finding as an “epidemiological paradox.” Markides and Coreil (p. 253; see also Hummer, Powers, Pullum, Gossman, & Frisbie, 2007) described this paradox in the following way: Despite methodological limitations of much of the research, it can be concluded with some certainty that the health status of Hispanics in the Southwest is much more similar to the health status of other Whites than that of Blacks although socioeconomically, the status of Hispanics is closer to that of Blacks. This observation is supported by evidence on such key health indicators as infant mortality, life-expectancy, mortality from cardiovascular diseases, mortality from major types of cancer, and measures of functional health. On other health indicators, such as diabetes and infectious and parasitic diseases, Hispanics appear to be clearly disadvantaged relative to other Whites.

This pattern—extended beyond infant mortality, including adult mortality and low birthweight—has now been consistently observed in the last 25 years (Crimmins, Kim, Alley, Karlamangla, & Seeman, 2007; Gorman, Ecklund, & Heard, 2010; Hummer et al. 2007; Hummer, Rogers, Amir, Forbes, & Frisbie, 2007; Markides & Eschbach, 2005; Markides, Rudkin, Angel, & Espino, 1997; Turra & Elo, 2008; Turra & Goldman, 2007). Because this paradox has been associated with Latinos (Hispanics) and, more specifically, Mexicans, it has been used interchangeably with the terms “Latino paradox,” “Hispanic paradox,” and “Mexican immigrant paradox.” Indeed, it is Mexican immigrants in particular who are associated with the epidemiological paradox (Alegria et al., 2008; Markides & Eschbach, 2005). Sociologists, demographers, and epidemiologists have suggested three general explanations to account for the Latino paradox. These explanations can be placed in three broad categories—migration selectivity, a pro-

The Latino Paradox    49

tective immigrant culture, and a statistical artifact. Nonetheless, the verdict is still out in the attempt to unravel the Latino paradox. This chapter will provide an overview of the Latino paradox. In particular, the chapter highlights the explanations that have been proposed to explain the Latino paradox. In addition, we highlight key findings related to these explanations. Moreover, the chapter discusses the presence of the immigrant health paradox outside of the United States. Furthermore, the chapter provides a discussion of immigrant-related paradoxes in other areas aside from mortality and health. Finally, the chapter concludes with a discussion of the direction that future research needs to take in order to advance our understanding of the Latino paradox. However, before we discuss any of these objectives, we first document the existence of the Latino paradox. The Existence of the Latino Paradox Drawing on data from Arias (2010) to document the existence of the Latino paradox based on the latest morality data (2006), we first examine 1-year probabilities of dying across single ages from 0 to 100 for Latinos and Whites by sex. For each of the single ages, we compute the ratio of the Latino 1-year probability of dying to the White 1-year probability of dying for males and females from the q columns in the Arias tables. Figure 3.1 provides an illustration of the Latino-to-White 1-year probabilities of dying sorted by sex. Note that values under 1.0 indicate that Latinos have a lower probability of dying compared to Whites. The figure clearly shows that overwhelmingly across ages, Latinos have lower probabilities of dying compared to Whites. In fact, across the 100 single ages from 0 to 100, in 83% of the cases, Latino males have lower likelihoods of death compared to White males; in 91% of the cases, Latina females are less likely to die compared to their White counterparts. Nonetheless, Whites have lower probabilities of death rates compared to Latinos in the younger age groups. For example, White males have lower probabilities of death than Latinos at ages 1–3, 4–6, and 11–23, while White females have lower odds of dying than Latina females at ages 0–1, 2–5, and 9–14. Research has shown that the age band from 15 to 24 is a vulnerable period for Latino males. Indeed, Vaca, Anderson, & Hayes-Bautista (2011) have observed increasing deaths among Latino males 15 to 24 in California from homicide and motor vehicle accidents alongside higher mortality rates at these ages among Latino males compared to White males. In contrast, Latino males have lower death probabilities than White males from ages 23 to 100, while Latina females have lower death probabilities compared to White females from ages 14 to 100. The Latino male advantage in the probability of death is greatest in the ages of 77 to 79, with

50   R. SÁENZ and T. MORALES 1.8 Male 1.6 Female

Probability of Dying

1.4

1.2

1.0

0.8

0.6

0.4

0.2

96–97

100 and over

92–93

88–89

84–85

80–81

76–77

72–73

68–69

64–65

60–61

56–57

52–53

48–49

44–45

40–41

36–37

32–33

28–29

24–25

20–21

16–17

8–9

12–13

4–5

0–1

0.0

Figure 3.1  Ratio of Latino-to-White probabilites of dying by age and sex, 2006. Source : Arias (2010).

the probability of death of Latino men being 24.4 % lower than that of White men. The Latina female advantage is most pronounced at younger ages, with the probability of death of Latina women being 40% lower than that of White women in the ages of 31 to 33. The greater levels of survival among Latinos compared to Whites translate into longer life expectancies among Latinos. Figure 3.2 shows the sexspecific life expectancies at birth for Latinos, Whites, and Blacks. A Latina baby girl born in 2006 is expected to live, on average, 83.1 years, a 2.7 year advantage compared to White females and a 6.9 year edge relative to Black females. Similarly, a Latino baby boy born in 2006 is expected to live, on average, 77.9 years, an advantage of 2.3 years compared to White males and an edge of 8.7 years compared to Black males. Yet another manner of assessing the Latino paradox is through the examination of infant mortality rates across racial and ethnic groups. We draw on data from Mathews and MacDorman, (2007) based on the 2004 period linked birth/infant death dataset. The infant mortality rate refers to the number of deaths of babies before they reach their first birthday per 1,000 live births. Figure 3.3 shows that the infant mortality rates of Latinos (5.55)

The Latino Paradox    51 90 83.1

80.4

77.9

80

Male 76.2

75.6

Life Expectancy

Female

69.2

70

60

50

40

30

20

10

0 Latino

White

Black

Figure 3.2  Life expectancies at birth by race/ethnicity and sex, 2006. Source : Arias (2010).

16 13.60

Infant Mortality Rate

14 12 10 7.82

8

6.73 6

5.55

5.66

5.47 4.55

4.65

4 2 0 Total Latino

Mexican

Puerto Rican

Cuban

Central Other and and South Unknown American Latino

White

Black

Figure 3.3  Infant mortality rate by race/ethnicity, 2004. Source : Matthews et al. (2007).

52   R. SÁENZ and T. MORALES

is slightly lower than that of Whites (5.66) and significantly lower than that of Blacks (13.60). Furthermore, three Latino subgroups have lower infant mortality rates than Whites: Cubans (4.55); Central and South Americans (4.65); and Mexicans (5.47). Aside from general and infant mortality, the Latino paradox has also been observed in other health- and development-related areas. For example, a significant amount of research has found that Latina women, particularly the foreign-born, have a low prevalence of pre-term and lowbirthweight babies (Acevedo-Garcia, Soobader, & Berkman, 2007; FuentesAfflick, Hessol, & Pérez-Stable, 1999; Hessol & Fuentes-Afflick, 2000; Reichman, Hamilton, Hummer, & Padilla, 2008). In addition, research has also demonstrated that Latinos fare favorably compared to Whites and persons from other racial and ethnic groups in other health-related areas. For instance, Sanders (2010) discovered that Latino immigrants have more favorable oral health compared to Whites and Blacks, as well as U.S.-born Latinos. Latinos have also been observed to possess health advantages with respect to asthma (Moorman et al., 2007; Subramanian, Jun, Kawachi, & Wright, 2009) and psychiatric disorders (Alderete, Vega, Kolody, & AguilarGaxiola, 2000; Alegria et al., 2008; Escobar, Nervi, & Gara, 2000). Thus, it is clear that the Latino paradox first observed back in 1974 is still evident nearly 4 decades later. For example, Latinos tend to have slightly lower infant mortality rates compared to Whites. In addition, Latinos have lower adult mortality rates than Whites. Moreover, Latinos live longer than do Whites. The paradox, then, is that despite Latinos having low levels of education, high levels of poverty, high levels of non-insurance (see Sáenz, 2010), and working in strenuous and dangerous jobs (see Douglas & Sáenz, 2008), they have low mortality rates, rates that are even lower than those of Whites who are favorably endowed socioeconomically. While the Latino paradox is associated with immigrants (those born outside of the United States), unfortunately the vital statistics data that we have used above are not broken down by nativity. Thus, the Latino mortality advantages that we have observed above would be even greater when Latino immigrants are compared to Whites, but the Latino mortality advantages would be less prevalent when U.S.-born Latinos are compared to Whites. We now turn to a discussion of the common explanations that have been proposed to account for the Latino paradox. Explanations for the Latino Paradox Three broad categories of explanations have been put forth to explicate the Latino paradox with respect to adult mortality and infant mortality.

The Latino Paradox    53

These explanations are categorized as migration selectivity, a protective immigrant culture, and a statistical artifact. Migrant Selectivity The study of migration extending back to the work of E. G. Ravenstein (1885; see also Lee, 1966) has observed that migrants are not randomly distributed from the populations from where they originate. Rather, migrants are selective. For example, migrants tend to be young, with the probability of migration being relatively high from the late teens to the early to mid-30s. Similarly, people who migrate tend to be healthy. Indeed, much research has observed that immigrants are favorably selected in terms of health from the populations from which they originate (Franzini, Ribble, & Keddie, 2001; Jasso, Massey, Rosenzweig, & Smith, 2004; Landale, Oropesa, & Gorman, 1999; Markides & Eschbach, 2005; Palloni & Morenoff, 2001). This is especially the case when migration is arduous, as is the case with undocumented immigration. Thus, given that immigrants tend to be healthier compared to their counterparts from the countries from where they originate, it is likely that they will have more favorable health and lower levels of mortality in their country of destination as well. In addition, with respect to infant mortality, healthy women immigrants tend to give birth to healthy babies. Protective Immigrant Culture A second explanation for the Latino paradox focuses on the protective elements of the culture of immigrants. These cultural elements involve familism; social networks; social support; healthy lifestyles; healthy diets; healthy physical activities; and low levels of smoking, drinking, and drug use (Abraído-Lanza, Dohrenwend, Ng-Mak, 2005; Balcazar, Peterson, & Cobas, 1996; Guendelman, 1998; Guendelman & Abrams, 1995; Pérez et al., 2001; Reichman et al., 2008; Zambrana, Scrimshaw, Collins, & DunkelSchetter, 1997). For example, immigrants tend to be associated with active participation in strong social networks, including relationships with family members, which in turn helps cope with and adjust to the stresses of movement into a new country (Vega, Kolody, Valle, & Weir, 1991). During times of trouble, immigrants are likely to turn to members of their social network to seek assistance and social support when coping with their problems, including those related to health. Furthermore, the diets of Latino immigrants tend to be rich in vegetables, grains, and legumes and low in meat consumption. Moreover, due to lower availability of personal automo-

54   R. SÁENZ and T. MORALES

biles in developing countries, Latino immigrants are likely to have walked more in their daily activities in their countries of origin than in the United States. In addition, immigrants tend to work in physically demanding jobs, not only in their countries of origin, but also in their countries of destination. Finally, immigrants tend to be less likely to smoke, drink, or use drugs compared to people who were born in the United States. Such a bundle of traditional cultural patterns tends to be associated with more favorable health outcomes including lower mortality. Similarly, immigrant women with similar protective cultures tend to have healthy babies. For example, de la Rosa (2002) observed that Mexican-origin women with low levels of acculturation have the most favorable birthweight outcomes. Furthermore, foreign-born (Gibson-Davis & BrooksGunn, 2006) and less acculturated Latina women (de la Torre & Rush, 1987; Gibson, Diaz, Mainous, & Geesey, 2005; Sussner, Lindsay, & Peterson, 2008) are more likely to breastfeed their babies compared to U.S.-born and more acculturated women. Research has documented the variety of shortand long-term health benefits that breastfeeding provides (Hediger, Overpeck, Kuczmarski, & Ruan, 2001; Heinig, 2001; Sussner et al., 2007). Statistical Artifact Yet another explanation suggests that the Latino paradox does not exist, or rather that it is not as notable as it appears at first sight. The underlying reasoning for the statistical artifact explanation is that there are problems in the computation of the mortality rates due to age misreporting, inconsistencies in the reporting of race and ethnicity, and return migration. For example, in populations with lower levels of education as well as in situations where people do not directly answer questions themselves in a census or survey (filled out by one person in the household, such as the householder), or birth or death certificates (filled out by a physician, and in the cases of death, a funeral director), the probability of age misreporting increases as well as in the misrepresentation of race/ethnicity. Thus, errors in age and race/ethnic misreporting may lead to inaccuracies in mortality rates since there is a lack of consistency in the numerator (number of persons dying in race/ethnic group i and age group j ) and the denominator (persons in the population in race/ethnic group i and age group j ) used to generate mortality rates. Challenges to the validity of the Latino paradox occur when such inconsistencies result in artificially lower mortality rates among Latinos. A related issue associated with a statistical artifact resulting in artificially lower mortality rates among Latinos concerns the return migration of people when they are ill. Immigrants who become ill in the United States may

The Latino Paradox    55

opt to return to their country of origin for a variety of reasons, including the availability of family for support, access to more affordable health care, as well as the desire to die in their homeland. This type of return migration is called the “salmon bias” (a term coined by Pablos-Mendez, 1994), an analogy to the upstream return of salmon to spawn in their place of birth where they die shortly after. In cases involving ill immigrants who return to their country of origin and subsequently die, their death is not recorded or counted in the United States. As such, their deaths are not part of the numerator, although they are part of the denominator (population at risk of dying), with the result being an artificially lower mortality rate among Latinos. Given the proximity of Latin America—especially in the case of Mexico—to the United States, return migration is a potential reality for Latino immigrants. Palloni and Arias (2004) suggest that the out-migration of unhealthy Mexican older individuals from the United States to Mexico may result in the underreporting of mortality in the United States, resulting in an artificially lower Mexican-origin mortality rate in this country (one study that has not supported the salmon bias explanation is that of AbraidoLanza et al., 1999). Relatedly, there is a version of the salmon bias that affects the infant mortality rate. In this case, pregnant Mexican women may come across the border to give birth in the United States, particularly along the U.S.-Mexico border. However, after the birth occurs, some of these women return to Mexico with their babies. If some of these babies die before reaching their first birthday, their deaths are not recorded or counted in the United States, although they are part of the population at risk (births that occurred in the United States). Thus, such deaths are not part of the numerator, but the infants were part of the denominator in the computation of the infant mortality, with the result being that this rate is artificially lower among Latinos. Research in general has suggested that the statistical artifact explanation is not strong enough to invalidate the existence of the Latino paradox. For example, Turra and Elo (2008), using data from the Social Security Administration’s Master Beneficiary Record (MBR) and NUMIDENT data files, observe that Latino and White social security beneficiaries who migrated out of the United States have mortality rates that are higher than their counterparts who remain in the United States, consistent with the salmon bias explanation. However, even taking into account for such deaths, the Latino mortality rates among social security beneficiaries continue to be noticeably lower than those of Whites. The salmon bias effect is simply not large enough due to the relatively small volume of return migration (see also Hummer et al., 2007). In addition, through the calculation of detailed age-specific infant mortality rates, Hummer et al. observe that about half of all infant deaths occur within 1 week of birth, making it very unlikely that Mexican mothers of newborns will return to Mexico so quickly after giving

56   R. SÁENZ and T. MORALES

birth. Furthermore, Hummer et al. demonstrate that even at very young ages (first hour of birth, first hour, and first week), when return migration is not likely to occur, Latinos have infant mortality rates that are approximately 10% lower than those of Whites. Given these findings, Hummer et al. (p. 454) conclude that the epidemiological paradox for U.S. infant mortality among the Mexicanorigin population—both among immigrant women and among U.S.-born women—is real and not a data artifact. . . . Our findings should, at least for now, “close the case” . . . regarding the underreporting explanation for the epidemiological paradox of infant mortality.

Paradoxes to the Latino Paradox As if the Latino paradox is not enigmatic of its own accord, there are other unexpected features associated with the Latino paradoxes that have stumped researchers trying to unravel the paradox. There are three paradoxes to the Latino paradox we discuss here. Acculturation Paradox Research has consistently shown that the mortality and health advantage of Latino immigrants declines with increasing time in the United States as well as with an increasing level of acculturation (Antecol & Bedard, 2006; Collins & David, 2004; Collins, David, Mendivil, & Wu, 2006; de la Rosa, 2002; Elder, Broyles, Brennan, de Nuncio, & Nader, 2005; Ruiz, Dolbier, & Fleschler, 2006; Singh & Siahpush, 2002; Vega & Amaro, 1994). For example, Mexican immigrant farm workers have been observed experiencing rising levels of psychiatric disorders with greater levels of acculturation and time in the United States (Alderete et al., 2000). In addition, using data from the 1991 National Health Interview, Abraído-Lanza, Chao, and Flórez (2005) found that higher acculturation among Latino immigrants is associated with higher levels of alcohol use and smoking, and a higher body mass index (BMI). Moreover, higher levels of acculturation (Gibson et al., 2005) and longer residence in the United States (Gibson-Davis & Brooks-Gunn, 2006; Guendelman & Siega-Riz, 2002; Harley, Stamm, & Eskenazi, 2007) are associated with reduced probabilities of breastfeeding among Latina women. Furthermore, Kaplan, Huguet, Newsom, and McFarland (2004) discovered an association between length of U.S. residency and obesity among Latino immigrants. Latino immigrants living in the United States

The Latino Paradox    57

for more than 15 years are four times more likely to be obese than Latino immigrants residing in this country less than 5 years. In essence, the planting of deeper roots in the United States tends to be hazardous to the health of Latino immigrants. The reasoning is that with greater time in the United States and with greater levels of acculturation, Latino immigrants lose the protective elements of their culture. Thus, with the passage of time, immigrants are less likely to rely on their families or social support systems (Alba & Nee, 2003) and to take on the lifestyle of the host society with respect to diet (Jonnalagadda & Diwan, 2005), smoking, and substance use (Abraído-Lanza et al., 2005; Acevedo-Garcia, Pan, Jun, Osypuk, & Emmons, 2005; Scribner, 1996; Singh & Siahpush, 2002; Vega & Amaro, 1994; Vega, Pan, Jun, Osypuk, & Emmons, 1998). Accordingly, with greater time in the United States, Latino immigrants may forego their traditional healthy diets and opt for fast-food and processed food laden with saturated fats, and they tend to become more sedentary (Kaplan et al., 2004). These changes result in a deterioration of the health advantage Latinos possessed when they initially arrived in the United States, with the result being greater mortality risks. Still, increasing levels of acculturation do benefit Latinos in respect to engaging in physical activities in the form of exercise (Abraído-Lanza et al., 2005; Crespo, Smit, Carter-Pokras, & Andersen, 2001; Jonnalagadda & Diwan, 2005). However, some researchers are beginning to question the view that time in the United States and acculturation is hazardous to the health of Latino immigrants. It is suggested that perhaps the relationship is more complicated upon closer inspection. Ceballos and Palloni (2010) suggest that the relationship between time in the United States and birth outcomes may not be linear. They argue that there is likely to be a two-stage selectivity process associated with health and migration among Latino immigrants. At the early stage, Latino immigrants who experience health problems may return to their country of origin because they are likely to have better access to health care in their communities of origin than in their communities of destination (Abraido-Lanza et al., 1999; Ceballos & Palloni, 2010; Palloni & Arias, 2004). Thus, Latino immigrants who remain in the United States in the early stages of immigration represent a healthy population, consistent with the view that Latino immigrants who have spent little time in the United States have low levels of mortality. However, with the passage of time in the United States, these individuals become more entrenched in the lifestyles and activities of the host society, and they also come to have better access to health care in this country (Ceballos & Palloni, 2010; Durden, 2007; LeClere, Jensen, & Biddlecom, 1994). When they become ill at the later stage of the immigration process, they are more likely to remain in the United States in order to access health care in this country. As such, with longer stays in the United States, Latino immigrants, as would be ex-

58   R. SÁENZ and T. MORALES

pected, contract diseases and experience injuries that are likely to increase their mortality level. This pattern would inaccurately suggest that increasing time in the United States and rising acculturation levels are associated with worsening health and increasing risks of mortality. Morbidity and Disability Paradox Another enigma associated with the Latino paradox is that despite Latinos having low levels of mortality and high levels of longevity, they tend to have high rates of morbidity and disability (Hunt et al., 2003; LaVeist, 2005; Markides & Eschbach, 2005). As such, Latinos tend to live longer than persons from other racial and ethnic groups, but a greater portion of their lives is lived with illness and disabilities. This may reflect the lack of health insurance of Latinos—fully 30% of Latinos living in the United States do not have health insurance (Sáenz, 2010). In addition, while Latino immigrants have relatively low levels of disability as young adults, disability becomes more prevalent in later years, suggesting that the types of jobs that Latinos perform wear down the body over time. Furthermore, the health problems of Latinos are compounded by a high prevalence of diabetes, obesity, and low physical activities (Markides & Eschbach, 2005). The Ethnic Enclave Paradox Latino immigrants tend to live in areas with large concentrations of coethnics, characterized by high levels of poverty, limited resources, and environments often not conducive to favorable health (Osypuk, Galea, McArdle, & Acevedo-Garcia, 2009). Despite living in areas with limited access to the opportunity structure, Latinos who live in Latino enclaves tend to have more social support from family and their community compared to their counterparts who reside in other geographic settings, resulting in health advantages (Eschbach, Mahnken, & Goodwin, 2005; Eschbach, Ostir, Patel, Markides, & Goodwin, 2004; Patel, Eschbach, Rudkin, Peek, & Markides, 2003). While research suggests that Latino immigrants benefit from living in ethnic enclaves, other research shows that U.S.-born Latinos have better health outcomes when they live alongside Latino immigrants. For example, Osypuk, Bates, and Acevedo-Garcia (2010) observe that U.S.-born Mexican American women are less likely to have low-birthweight babies when they live in neighborhoods where they have high contact with Latino immigrants; conversely, they are more likely to have low-birthweight infants when they live in areas where they have high levels of interaction with U.S.born Mexican Americans.

The Latino Paradox    59

Promising New Lines of Research on the Latino Paradox Much of the existing research on the Latino paradox has tended to be based on traditional datasets. These include data based on secondary and primary quantitative sources that examine the links between factors traditionally associated with a wide variety of health outcomes. We see the emergence of new lines of research that are likely to help us gain a better understanding of the Latino paradox. In particular, we highlight below recent research that has used qualitative methodologies as well as studies that have incorporated the role of discrimination on health outcomes. Research that has attempted to bring structural elements into the analysis has critiqued the literature’s emphasis on acculturation. The ramification of this focus is that it makes central the individual as he or she adjusts to differing cultures rather than placing the individual within the larger structural elements within which he or she lives. Accordingly, Viruell-Fuentes (2007) points out that acculturation models assume that culture is located within an individual and that cultural traits are inherent to members of a particular group, instead of envisioning culture as a system that is socially constructed. . . . However, unintentional, acculturation models promote cultural facts as the focus of intervention at the expense of drawing attention to the structural constraints within which individuals operate. . . . By putting the onus on the individual, this treatment of culture runs the risk of lending support to victim-blaming explanations for health outcomes. The idea of culture as a “source of dysfunction” can easily follow this line of thinking. (p. 1525)

Furthermore, Viruell-Fuentes (2007) calls attention to the hostile and racialized environment in which Latino immigrants live, as evidenced by heightened levels of anti-immigrant sentiment and stress associated with the fear of apprehension and deportation (Sáenz, Menjivar, & Garcia, 2011). Viruell-Fuentes argues that the emphasis on acculturation and the lack of attention to structural forces results in questions about how Latinos and immigrants experience the racial and ethnic structure of the US; how they become racialized and negotiate US-based racial and ethnic identities; and how these processes translate into health status remain largely unexamined. For instance, while the role of discrimination on health has received increased attention . . . to date the immigrant and Latino health literatures, with few exceptions . . . have rarely highlighted discrimination as a determinant of health among these populations. Instead, acculturation is much more readily invoked in explaining the health outcomes of these groups. (p. 1525)

60   R. SÁENZ and T. MORALES

Research from the field of racial and ethnic relations has suggested that racial dynamics are much different in the country of origin of immigrants compared to the United States. Thus, we often find that first-generation Latinos are likely to report lower levels of discrimination compared to their counterparts who have been in this country for more generations (Telles & Ortiz, 2008). Accordingly, in her qualitative research of first- and second-generation Mexican-origin women in southeastern Michigan, Viruell-Fuentes (2007) observed generational differences in how these women view intergroup relations in the area. In particular, she found generational differences in how they perceived as being “othered” and the knowledge that they have an ascribed status of being a member of a minority group. As ViruellFuentes (p. 1525) points out, those are seen as “‘the other’ are located differentially in lower categories of a hierarchical system by both individual actors and institutions.” Moreover, Grove and Zwi (2006, p. 1933) add that the “person or group being ‘othered’ experiences a process of marginalization, disempowerment and social exclusion.” Viruell-Fuentes observed that only a few first-generation Mexican women in her study reported any “othering” experience. The following statement from Ernestina, who had been in the United States for 3 years, is more typical of the experiences of the first generation: “Everybody has treated me well here” (ViruellFuentes, p. 1526). Furthermore, Viruell-Fuentes noted that most of the first-generation women lived in the ethnic neighborhood surrounded by other Mexicans and had relatively little contact outside of the confines of this area. It is only when first-generation women venture out of the enclave that they encounter “othering” experiences (Viruell-Fuentes, 2007; see also Portes, 1984). For instance, Patricia, a woman who had resided in the United States for 12 years and who occasionally traveled to the Detroit suburbs, reported that You feel that people are looking at you ugly [because you are Mexican]. And sometimes people do show it openly, right? They look at you up and down . . . And [the other day at the store] there were two older White women behind me, and they were looking me up and down just like that; I tried smiling at them, but they wouldn’t smile back. They were staring at me; it upset me. (Viruell-Fuentes, p. 1527; bracketed remarks in original)

While “othering” was rare among first-generation Mexican-origin women, except when they ventured out of their neighborhood, it was a very common theme in the experiences of second-generation women in the Viruell-Fuentes (2007) study. For example, as children in school, they were subjected to hostility and taunting on the part of their White and Black classmates, which signaled to them that they were outsiders and did not belong in the community. Second-generation women reported being called “wetback,” told to

The Latino Paradox    61

“go back to Mexico,” and some people beaten up for speaking Spanish. The wounds as well as the remarks and gazes classifying them as outsiders continued to persist among these women well into adulthood. These patterns observed by Viruell-Fuentes (2007) obviously are important in light of the consistent research finding that the health of Latinos deteriorates with greater time in the United States as well as across generations. Viruell-Fuentes (p. 1531) notes that her findings point to “othering” and discrimination as potential pathways through which “becoming American” negatively impacts the health of immigrants and their descendants. The Paradox in Other Contexts The paradox has been observed in different contexts beyond the United States and beyond issues involving health-related outcomes. Below, we provide an overview of research that has observed the paradox outside of the United States as well as research that has observed the paradox in nonhealth related areas in the United States. The Paradox beyond the United States While the Latino paradox has been observed in the United States for nearly 4 decades, research has observed the immigrant paradox—whereby immigrants are healthier than native-born populations—in other parts of the world over the last couple of decades. This has been observed in Belgium (Guendelman et al., 1999), Canada (Ali, 2002; Chen, Wilkins, & Ng, 1996; Hyman, 2001; Pérez, 2002; Zheng and Schimmele 2005), France (Guendelman et al., 1999), Germany (Razum, Zeeb, Akgün, & Yilmaz, 1988; Razum, Zeeb, & Gerhardus, 1988; Razum, Zeeb, & Rohrmann, 2000), Ireland (Fong, 2008), Italy (Parkin, 1993), and the United Kingdom (Fong, 2008; Harrison, Owens, Holton, Neilson, & Boot, 1988; Hutchinson et al., 1996; Swerdlow, 1991; Williams, 1993). For example, immigrants have longer life expectancies than native-born individuals in Canada, with non-European immigrants having the highest life expectancy, a 6-year advantage in life expectancy compared to the native-born population (Chen et al., 1996). Some of the international research has assessed the extent to which the health advantage of immigrants declines over time. For example, Wu and Schimmele (2005) observed that while immigrants in Canada had more favorable mental health compared to native-born Canadians, their mental well-being declined for approximately 30 years.

62   R. SÁENZ and T. MORALES

The Paradox beyond Health A variety of research recently has observed the Latino paradox in areas beyond health, particularly in the areas of crime, violence, and education. For example, while Latino immigrants have low levels of socioeconomic status and live in neighborhoods that are considered dangerous, research has demonstrated that Latino immigrants have relatively low crime rates and that areas where they live exhibit relatively low levels of crime (Desmond & Kubrin, 2009; Hagan & Palloni, 1999; Lee & Martinez, 2002; Lee, Martinez, & Rosenfeld, 2001; Martinez, 2002, 2006; Martinez, Lee, & Nielsen, 2004; Morenoff, 2005; Morenoff & Astor, 2006; Nielsen, Lee, & Martinez, 2005; Peterson & Krivo, 2005; Sampson, 2008; Sampson & Bean, 2006; Sampson, Morenoff, & Raudenbush, 2005; Tapia, 2010; Velez, 2006; for a study that has not supported the Latino paradox in crime, see Feldmeyer, 2010). As is the case with the ethnic enclave paradox associated with health noted above, it is suggested that areas where Latino immigrants are concentrated contain significant levels of social networks and social support. In this environment, there is a constant social vigilance that keeps co-ethnics “in line” with respect to social norms and expectations (Warren, 2003). Accordingly, Aguilar-San Juan (2005, p. 37; see also Desmond & Kubrin, 2009) argues that such places represent “sites for the creation of social networks, aggregating devices, anchors for identity, and representations of culture.” Furthermore, Sampson (2008) suggests that immigrants may bring with them cultures that differ from more violent cultural perspectives found in low-income areas in the United States. In addition, Ousey and Kubrin (2009) argue that immigrants may be selective of individuals with a low propensity for crime. Hence, we find that areas where immigrants are clustered have relatively low levels of crime and violence. For example, Desmond and Kubrin (2009) find that youth who live in areas with a heavy representation of immigrants tend to report fewer acts of violence. Moreover, Wright and Benson (2010) observe that neighborhoods with greater concentrations of crime tend to have lower levels of intimate-partner violence. A significant amount of research has also found that immigrant children tend to have more favorable educational outcomes compared to their nativeborn counterparts, particularly the third-plus generations (Telles & Ortiz, 2008). For example, Kao and Tienda (1995) observe that second-generation students—the children of immigrant parents—tend to have more favorable academic achievement levels than first and third-plus generations. They suggest immigrant parents may impart their optimism to their children (see also Suárez-Orozco & Suárez-Orozco, 2001). In addition, research has also shown that immigrant students tend to have more positive outlooks regarding their schooling (Suárez-Orozco & Suárez-Orozco, 1995) and higher aspirations (Fuligini, 1997; Portes & Rumbaut, 2001). Furthermore, Valenzuela

The Latino Paradox    63

(1999) found that teachers tend to have more favorable views of Latino immigrant students compared to Latino native-born students. Nonetheless, as observed with respect to health (see above), with greater length of residence in the United States, immigrant children lose their favorable outlooks, aspirations, and optimism regarding school, and their academic performance deteriorates (Portes & Rumbaut, 2001; Suárez-Orozco & Suárez-Orozco, 1995; Suárez-Orozco, Rhodes, & Milburn, 2009). Yet another area in which the Latino paradox has been observed is homelessness. For long, research has observed that Latinos have significantly lower levels of homelessness compared to the African American population, despite both groups sharing similarly low socioeconomic levels (Conroy & Heer, 2003; Rossi, 1989). Baker (1996) used the term “Latino paradox” to describe this pattern. Two perspectives have been put forth to account for the Latino paradox (see Baker, 1996; Conroy & Heer, 2003). The first perspective suggests that Latinos are more likely than African Americans to tap social networks to find housing with friends and relatives in order to avoid living on the streets or in shelter homes. The second perspective argues that Latinos are disproportionately undercounted in homeless surveys because they tend to be located in hidden areas rather than in places where samples are traditionally taken (e.g., areas where the homeless congregate); this represents the statistical artifact explanation found in the Latino health and mortality paradox literature. Using a unique dataset of Mexican-born persons living in Los Angeles and a traditional survey based on the Los Angeles homeless population, Conroy and Heer found support for the latter explanation. Indeed, they observed that foreign-born Mexicans had homeless rates that were seven times greater than the population in Los Angeles County, and that the Mexican homeless were much more likely to be living in nontraditional homeless settings, where they are likely to be missed by traditional surveys of the homeless population that focus on areas where the homeless are concentrated. This literature focusing on the immigrant paradox beyond the United States and the Latino paradox beyond the confines of mortality and health offers important insights into our understanding of the lives of immigrants in a variety of settings around the world and in the different spheres of life in which they participate. Recommendations for Future Research on the Latino Paradox The major shortcoming associated with unraveling the Latino paradox concerns limitations in data. In particular, existing research on the Latino paradox has relied on cross-sectional data in a single country (mostly the

64   R. SÁENZ and T. MORALES

United States). Nonetheless, it is clear that we need binational and longitudinal data approaches that can capture the broad and dynamic settings in which health/mortality and immigration interact. To fully assess the Latino paradox, we need longitudinal data that tracks immigrants and nonimmigrants in their country of origin, immigrants and native-born persons in the country of destination, and the ongoing movement of persons between the country of origin and country of destination. Such an approach allows us to compare immigrants to persons who stayed behind in the country of origin as well as to persons who are natives of the country of destination in order to better capture the links between immigration; length of residence in the country of origin and country of destination; acculturation; changes in lifestyle, diets, drug and alcohol use; family and social networks; and health and mortality. In addition, as discussed earlier, there is a major need to incorporate qualitative and ethnographic approaches in the study of the Latino paradox. To date, with very rare exceptions (Fleuriet, 2009; Viruell-Fuentes, 2007), most research examining the Latino paradox has used quantitative data. It is likely that qualitative and ethnographic approaches can allow us to obtain a more in-depth understanding of behavior, including the use of nontraditional health care practices, which can help us gain an understanding of the health patterns of Latino immigrants that are not likely to be tapped by traditional quantitative approaches. For example, the research of Fleuriet provides an excellent example of the use of qualitative approaches to uncover the variety of ways in which Latinas deal with issues of pregnancy, including seeking social support from family members. In addition, the work of Viruell-Fuentes, as pointed out earlier, aptly demonstrates the usefulness of qualitative approaches in gaining an understanding of the racialization that Latina women experience across the life course. In particular, qualitative and ethnographic methodologies are likely to help us tap issues related to acculturation, racism, and discrimination more effectively than traditional quantitative methodologies. Finally, the study of the Latino paradox is likely to benefit from the incorporation of geographic space (e.g., the Geographic Information System [GIS] and Hierarchical Linear Analysis [HLM] as well as environmental factors; see Yabiku, Glick, Wendtz, Haas, & Zhu, 2009) into future research to detect and isolate factors that may be contributing to the paradox. Indeed, GIS and HLM techniques allow us to place Latino immigrants within the geographic spaces they inhibit to understand how place (in the country of origin and country of destination) may interact with commonly used factors to produce varying health and mortality outcomes.

The Latino Paradox    65

Conclusions This chapter has demonstrated that the Latino paradox is manifest in a plethora of contexts well beyond health and mortality outcomes. Researchers studying Latinos specifically or simply including Latinos in their studies have uncovered a paradox that begs further examination. Explanations are numerous with varying degrees of support, but the presence of a paradox should not be denied, as evidenced by the various studies cited in this chapter. Also, relegating the paradox to a statistical artifact entirely is not well supported. Thus, we suggest that researchers continue to make adjustments to account for statistical artifact factors in their analyses. Although this chapter may give the impression Latino paradox research is exhaustive, research is relatively recent and continues to encompass new avenues such as crime, homelessness, and discrimination. In addition, as the United States continues to grow in population size, much of the population growth is attributed to immigrants. Regardless of the political debate on immigration, immigrants continue to possess advantages usually reserved for people of high socioeconomic status. Perhaps an overdependence on technology in developed countries may be inadvertently truncating expected advantages associated with possession of wealth and access to resources. Public policy initiatives in the United States should nourish research on the various iterations of the paradox in an attempt to disseminate findings in order to benefit society at large. References Abraído-Lanza, A. F., Chao, M. T., & Flórez, K. R. (2005). Do healthy behaviors decline with greater acculturation?: Implications for the Latino mortality paradox. Social Science & Medicine, 61, 1243–1255. Abraído-Lanza, A. F., Dohrenwend B. P., Ng-Mak D. S., & Turner J. B. (1999). The Latino paradox: A test of the “salmon bias” and healthy migrant Hypotheses. American Journal of Public Health, 89,1543–1548. Acevedo-Garcia, D., Pan, J., Jun, H-J., Osypuk, T. L., & Emmons, K. M. (2005). The effect of immigrant generation on smoking. Social Science & Medicine, 61(6), 1223–1242. Acevedo-Garcia, D., Soobader, M-J., & Berkman, L. F. (2007). Low birthweight among U.S. Hispanic/Latino subgroups: The effect of maternal foreign-born status and education. Social Science & Medicine, 65, 2503–2516. Aguilar-San Juan, K., (2005). Staying Vietnamese: Community and place in Orange County and Boston. City & Community, 4(1), 37–65. Alba, R., & Nee, V. (2003). Remaking the American mainstream: Assimilation and contemporary immigration. Cambridge, MA: Harvard University Press.

66   R. SÁENZ and T. MORALES Alderete, E., Vega, W. A., Kolody, B., & Aguilar-Gaxiola, S. (2000). Lifetime prevalence of and risk factors for psychiatric disorders among Mexican migrant farm workers in California. American Journal of Public Health, 90(4), 608–614. Alegria, M., Canino, G., Shrout, P. E., Woo, M., Duan, N., Vila, D., Torres, M. et al. (2008). Prevalence of mental illness in immigrant and non-immigrant U.S. Latino groups. American Journal of Psychiatry, 165(3), 359–369. Ali, J. (2002). Mental health of Canada’s immigrants. Health Reports, 13(Suppl.), 101–113. Antecol, H., & Bedard, K. (2006). Unhealthy assimilation: Why do immigrants converge to American health status levels? Demography, 43(2), 337–360. Arias, E. (2010). United States life tables by Hispanic origin. Hyattsville, MD: National Center for Health Statistics. Balcazar, H., Peterson, G., & Cobas, J. A. (1996). Acculturation and health-related risk behaviors among Mexican-American pregnant youth. American Journal of Health Behavior, 20(6), 425–433. Baker, S. G. (1996). Homelessness and the Latino paradox. In J. Baumohl (Ed.), Homelessness in America (pp. 132–140). Phoenix, AZ: Oryx. Ceballos, M., & Palloni, A. (2010). Maternal and infant health of Mexican immigrants in the USA: The effects of acculturation, duration, and selective return migration. Ethnicity & Health, 15(4), 377–396. Chen, J., Wilkins, R., & Ng, E. (1996). Health expectancy by immigrant status, 1986 and 1991. Health Reports, 8, 29–37. Collins, J. W., Jr., & David R. J. (2004). Pregnancy outcome of Mexican-American women: The effect of generational residence in the United States. Ethnicity & Disease, 14(3), 317–321. Collins, J. W., Jr., David R. J., Mendivil, N. A., & Wu, S-Y. (2006). Intergenerational birth weights among the direct female descendants of US-born and Mexicanborn Mexican-American women in Illinois: An exploratory study. Ethnicity & Disease, 16(1), 166–171. Conroy, S. J., & Heer, D. M. (2003). Hidden Hispanic homelessness in Los Angeles: The “Latino paradox” revisited. Hispanic Journal of Behavioral Sciences, 25(4), 530–538. Crespo, C. J., Smit, E., Carter-Pokras, O., & Andersen, R. (2001). Acculturation and leisure-time physical inactivity in Mexican American adults: Results from NHANES III, 1988–1994. American Journal of Public Health, 91(8), 1254–1257. Crimmins, E. M., Kim, J. K., Alley, D. E., Karlamangla, A., & Seeman, T. (2007). Hispanic paradox in biological risk profiles. American Journal of Public Health, 97, 1305–1310. de la Rosa, I. A. (2002). Perinatal outcomes among Mexican Americans: A review of an epidemiological paradox. Ethnicity & Disease, 12, 480–487. de la Torre, A., & Rush, L. (1987). The determinants of breastfeeding for Mexican migrant women. International Migration Review, 21(3), 728–742. Desmond, S. A., & Kubrin, C. E. (2009). The power of place: Immigrant communities and adolescent violence. The Sociological Quarterly, 50, 581–607. Douglas, K. M., & Sáenz, R. (2008). No phone, no vehicle, no English, and no citizenship: The vulnerability of Mexican immigrants in the United States. In A.

The Latino Paradox    67 Hattery, D. G. Embrick, & E. Smith (Eds.), Race, human rights and inequality (pp. 161–180). Lanham, MD: Rowman and Littlefield. Durden, T. E. (2007). Nativity, duration of residence, citizenship, and access to health care for Hispanic children. International Migration Review, 41(2), 537– 545. Elder, J. P., Broyles, S. L., Brennan, J. J., de Nuncio, M. L. Z., & Nader, P. R. (2005). Acculturation, parent-child acculturation differential and chronic disease risk factors in a Mexican-American population. Journal of Immigrant Health, 7(1), 1–9. Eschbach, K., Mahnken, J. D., & Goodwin, J. S. (2005). Neighborhood composition and incidence of cancer among Hispanics in the United States. Cancer, 103(5), 1036–1044. Eschbach, K., Ostir, G. V., Patel, K. V., Markides, K. S., & Goodwin, J. S. (2004). Neighborhood context and mortality among older Mexican Americans: Is there a barrio advantage? American Journal of Public Health, 94(10), 1807–1812. Escobar, J. I., Nervi, C. H., & Gara, M. A. (2000). Immigration and mental health: Mexican Americans in the United States. Harvard Review of Psychiatry, 8, 64– 72. Feldmeyer, B. (2010). The effects of racial/ethnic segregation on Latino and Black homicide. The Sociological Quarterly, 51, 600–623. Fleuriet, K. J. (2009). Problems in the Latina paradox: Measuring social support for pregnant immigrant women from Mexico. Anthropology & Medicine, 16(1), 49–59. Fong, V. L. (2008). The other side of the healthy immigrant paradox: Chinese sojourners in Ireland and Britain who return to China due to personal and familial health crises. Culture, Medicine, and Psychiatry, 32, 627–641. Franzini, L., Ribble, J. C., & Keddie, A. M. (2001). Understanding the Hispanic paradox. Ethnicity and Disease, 11(3), 496–518. Fuentes-Afflick, E., Hessol, N. A., & Pérez-Stable, E. J. (1999). Testing the epidemiologic paradox of low birth weight in Latinos. Archives of Pediatrics & Adolescent Medicine, 153, 147–153. Fuligni, A. J., (1997). The academic achievement of adolescents from immigrant families: The roles of family background, attitudes, and behavior. Child Development, 2(68), 351–363. Gibson, M., Diaz, V. A., Mainous, A. G., & Geesey, M. E. (2005). Prevalence of breastfeeding and acculturation in Hispanics: Results for NHANES 1999–2000 Study. Birth, 32(2), 93–98. Gibson-Davis, C. M., & Brooks-Gunn, J. (2006). Couples’ immigration status and ethnicity as determinants of breastfeeding. American Journal of Public Health, 96(4), 641–646. Gorman, B. K., Ecklund E. H., & Heard, H. E. (2010). Nativity differences in physical health: The roles of emotional support, family, and social integration. Sociological Spectrum, 30, 671–694. Grove, N. J., & Zwi, A. B. (2006). Our health and theirs: Forced migration, othering, and public health. Social Science & Medicine, 62(8), 1931–1942. Guendelman, S. (1998). Health and disease among Hispanics. In S. Loue (Ed.), Handbook of immigrant health (pp. 277–301). New York: Plenum Press.

68   R. SÁENZ and T. MORALES Guendelman, S., & Abrams, B. (1995). Dietary intake among Mexican-American women: Generational differences and a comparison with White non-Hispanic women. American Journal of Public Health, 85, 20–25. Guendelman, S., Buekens, P., Blondel, B., Kaminski, M., Notzon, F.C., & MasuyStroobant, G. (1999). Birth outcomes of immigrant women in the United States, France, and Belgium. Maternal and Child Health Journal, 3(4), 177–187. Guendelman, S., & Siega-Riz, A. M. (2002). Infant feeding practices and maternal dietary intake among Latino immigrants in California. Journal of Immigrant Health, 4(3), 137–146. Guendelman, S., Buekens, P., Blondel, B., Kaminski, M., Notzon, F. C., & MasuyStroobant, G. (1999). Birth outcomes of immigrant women in the United States, France, and Belgium. Maternal and Child Health Journal, 3(4), 177–187. Hagan, J., & Palloni, A. (1999). Sociological criminology and the mythology of Hispanic immigration and crime. Social Problems, 46, 617–632. Harley, K., Stamm, N. L., & Eskenazi, B. (2007). The effect of time in the U.S. on the duration of breastfeeding in women of Mexican descent. Maternal Child Health Journal, 11(2), 119–125. Harrison, G., Owens, D., Holton, A., Neilson, D., & Boot, D. (1988). A prospective study of severe mental disorder in Afro-Caribbean patients. Psychological Medicine, 18(3), 643–657. Hediger, M. L., Overpeck, M. D., Kuczmarski, R. J., & Ruan, W. J. (2001). Association between infant breastfeeding and overweight in young children. Journal of the American Medical Association, 285(19), 2453–2460. Heinig, M.J., (2001). Host defense benefits of breastfeeding for the infant: Effect of breastfeeding duration and exclusivity. Pediatric Clinics of North America, 48(1), 105–123. Hessol, N. A., & Fuentes-Afflick, E. (2000). The perinatal advantage of Mexicanorigin Latina women. Annals of Epidemiology, 10(8), 516–523. Hummer, R. A., Powers, D. A., Pullum, S. G., Gossman, G. L., & Frisbie, W. P. (2007). Paradox found (again): Infant mortality among the Mexican-origin population in the United States. Demography, 44(3), 441–457. Hummer, R. A., Rogers, R. G., Amir, S. H., Forbes, D., & Frisbie, W. P. (2000). Adult mortality differentials among Hispanic subgroups and non-Hispanic Whites. Social Science Quarterly, 81, 459–476. Hunt, K. J., Resendez, R. G., Williams, K., Haffner, S. M., Stern, M. P., & Hazuda, H. P. (2003). All-cause and cardiovascular mortality among Mexican-American and non-Hispanic White older participants in the San Antonio Heart Studyevidence against the “Hispanic paradox.” American Journal of Epidemiology, 158, 1048–1057. Hutchinson, G., Takei, N., Fahy, T. A., Bhugra, D., Gilvarry, C., Moran, P., Mallett, R. et al. (1996). Morbid risk of schizophrenia in first-degree relatives of White and African-Caribbean patients with psychosis. British Journal of Psychiatry, 169(6), 776–780. Hyman, I. (2001). Immigration and health. Ottawa, Canada: Health Canada. Jasso, G., Massey, D. S., Rosenzweig, M. R., & Smith, J. P. (2004). Immigrant health: Selectivity and acculturation. In N. B. Anderson, R. A. Bulatao, & B. Cohen

The Latino Paradox    69 (Eds.), Critical perspectives on racial and ethnic differences in health in later life (pp. 227–266). Washington, DC: National Academies Press. Jonnalagadda, S. S., & Diwan, S. (2005). Health behaviors, chronic disease prevalence and self-rated health of older Asian Indian immigrants in the U.S. Journal of Immigrant Health, 7(2), 72–83. Kao, G., & Tienda, M. (1995). Optimism and achievement: The educational performance of immigrant youth. Social Science Quarterly, 76, 1–19. Kaplan, M. S., Huguet, N., Newsom, J. T., & McFarland, B. H. (2004). The association between length of residence and obesity among Hispanic immigrants. American Journal of Preventive Medicine, 27, 323–326. Kitagawa, E. M., & Hauser, P. M. (1973). Differential mortality in the United States: A study in socioeconomic epidemiology. Cambridge, MA: Harvard University Press. Landale, N., Oropesa, R. S., & Gorman, B. K. (1999). Immigration and infant health: Birth outcomes of immigrant and native women. In D. J. Hernandez (Ed.), Children of immigrants: Health, adjustment, and public assistance (pp. 244–286). Washington, DC: National Academy Press. LaVeist, T. A. (2005). Minority populations and health: An introduction to health disparities in the United States. San Francisco: Jossey-Bass. LeClere, F. B., Jensen, L., & Biddlecom, A. E. (1994). Health care utilization, family context, and adaptation among immigrants to the United States. Journal of Health and Social Behavior, 35(4), 370–384. Lee, E. S. (1966). A theory of migration. Demography, 3(1), 47–57. Lee, M. T., & Martinez, R. (2002). Social disorganization revisited: Mapping the recent immigration and Black homicide relationship in northern Miami. Sociological Focus, 35, 365–382. Lee, M. T., Martinez, R., & Rosenfeld, R. (2001). Does immigration increase homicide?: Negative evidence from three border cities. The Sociological Quarterly, 42, 559–580. Markides, K. S., & Coreil, J. (1986). The health of Hispanics in the southwestern United States: An epidemiological paradox. Public Health Reports, 101, 253– 265. Markides, K. S., & Eschbach, K. (2005). Aging, migration, and mortality: Current status of research on the Hispanic paradox. Journals of Gerontology, Series B: Psychological Sciences, 60B, 68–75. Markides, K. S., Rudkin, L., Angel, R. J., & Espino, D. (1997). Health status of the Hispanic elderly. In L. G. Martin & B. J. Soldo (Eds.), Racial and ethnic differences in the health of older Americans (pp. 285–300). Washington, DC: National Academy Press. Martinez, R., Jr. (2002). Latino homicide: Immigration, violence, and community. New York: Routledge. Martinez, R., Jr. (2006). Coming to America: The impact of the new immigration on crime. In R. Martinez, Jr. & A. Valenzuela, Jr. (Eds.), Immigration and crime: Race, ethnicity, and violence (pp. 1–19). New York: New York University Press. Martinez, R., Lee, M. T., & Nielsen, A. L. (2004). Segmented assimilation, local context and determinants of drug violence in Miami and San Diego: Does ethnicity and immigration matter? International Migration Review, 38, 131–157.

70   R. SÁENZ and T. MORALES Mathews, T. J., & MacDorman, M. F. (2007). Infant mortality statistics from the 2004 period linked birth/infant death data set. Hyattsville, MD: National Center for Health Statistics. Moorman, J. E., Rudd, R. A., Johnson, C. A., King, M., Minor, P., Bailey, C., Scalia, M. R. et al. (2007). National surveillance for asthma—United States, 1980–2004. MMWR Surveillance Summaries, 56(8), 1–54. Morenoff, J. D. (2005). Racial and ethnic disparities in crime and delinquency in the U.S. In M. Tienda & M. Rutter (Eds.), Ethnicity and causal mechanisms (pp. 139–173). New York: Cambridge University Press. Morenoff, J. D., & Astor, A. (2006). Immigrant assimilation and crime: Generational differences in youth violence in Chicago. In R. Martinez, Jr. & A. Valenzuela, Jr. (Eds.), Immigration and crime: Race, ethnicity, and violence (pp. 36–63). New York: New York University Press. Nielsen, A., Lee, M. T., & Martinez, R. (2005). Integrating race, place and motive in social disorganization theory: Lessons from a comparison of Black and Latino homicide types in two immigrant destination cities. Criminology, 43, 837–872. Osypuk, T. L., Bates, L. M., & Acevedo-Garcia, D. (2010). Another Mexican birthweight paradox? The role of residential enclaves and neighborhood poverty in the birthweight of Mexican-origin infants. Social Science & Medicine, 70, 550–560. Osypuk, T. L., Galea, S., McArdle, N., & Acevedo-Garcia, D. (2009). Quantifying separate and unequal: Racial-ethnic distributions of neighborhood poverty in metropolitan America. Urban Affairs Review, 45(1), 25–65. Ousey, G. C., & Kubrin, C. E. (2009). Exploring the connection between immigration and violent crime rates in U.S. cities, 1980–2000. Social Problems, 56, 447–473. Pablos-Mendez, A. (1994). Mortality among Hispanics. Journal of the American Medical Association, 271(16), 1237–1238. Palloni, A., & Arias, E. (2004). Paradox lost: Explaining the Hispanic adult mortality advantage. Demography, 41(3), 385–415. Palloni, A., & Morenoff, J. D. (2001). Interpreting the paradoxical in the Hispanic Paradox. Annals of the New York Academy of Science, 954, 140–174. Parkin, D. M.. (1993). Studies of cancer in migrant populations. In M. Geddes (Ed.), Cancer in Italian migrant populations (pp. 1–10). Lyon, France: International Agency for Research on Cancer. Patel, K. V., Eschbach, K., Rudkin, L. L., Peek, M. K., & Markides, K. S. (2003). Neighborhood context and self-rated health in older Mexican Americans. Annals of Epidemiology, 13(9), 620–628. Pérez, C. E. (2002). Health status and health behaviour among immigrants. Health Reports, 13(Suppl.), 98–109. Pérez, E. J., Ramirez, A., Villareal, R., Talavara, G. A., Trapido, E., Suarez, L., Marti, J. et al. (2001). Cigarette smoking behavior among US Latino men and women from different countries of origin. American Journal of Public Health, 91(9), 1424–1430. Peterson, R. D., & Krivo, L. J. (2005). Macrostructural analyses of race, ethnicity, and violent crime: Recent lessons and new destinations for research. Annual Review of Sociology, 31, 331–356.

The Latino Paradox    71 Portes, A. (1984). The rise of ethnicity: Determinants of ethnic perceptions among Cuban exiles in Miami. American Sociological Review, 49(3), 383–397. Portes, A., & Rumbaut, R. G. (2001). Legacies: The story of the second generation. Berkeley: University of California Press. Ravenstein, E. G. (1885). The laws of migration. Journal of the Statistical Society of London, 48(2), 167–235. Razum, O., Zeeb, H., Akgün, H. S., & Yilmaz, S. (1998). Low overall mortality of Turkish residents in Germany persists and extends into second generation: Merely a healthy migrant effect? Tropical Medicine and International Health, 3, 297–303. Razum, O., Zeeb, H., & Gerhardus, A. (1998). Cardiovascular mortality of Turkish nationals residing in West Germany. Annals of Epidemiology, 8(5), 334–341. Razum, O., Zeeb, H., & Rohrmann, S. (2000). The “Healthy Migrant Effect”—Not merely a fallacy of inaccurate denominator figures. International Journal of Epidemiology, 29, 191–192. Reichman, N. E., Hamilton, E. R., Hummer, R. A., & Padilla, Y. C. (2008). Racial and ethnic disparities in low birthweight among urban unmarried mothers. Maternal and Child Health Journal, 12, 204–215. Rogers, R. G., Hummer, R. A., & Nam, C. B. (2000). Living and dying in the USA: Behavioral, health, and social differentials of adult mortality. San Diego, CA: Academic Press. Rossi, P. (1989). Down and out in America: The origins of homelessness. Chicago: University of Chicago Press. Ruiz, R.J., Dolbier, C.L., & Fleschler, R. (2006). The relationships among acculturation, biobehavioral risk, stress, corticotrpin-releasing hormone, and poor birth outcomes in hispanic women. Ethnicity & Disease, 16(4), 926–932. Sáenz, R. (2010, December 2010). Latinos in the United States 2010. Population Reference Bureau Bulletin Update. Sáenz, R., Menjivar, C., & Garcia, S. J. E. (2011). Arizona’s SB 1070: Setting conditions for violations of human rights here and beyond. In J. Blau & M. Frezzo (Eds.), Sociology and human rights: A bill of rights in the twenty-first century. Newbury Park, CA: Pine Forge Press. Sampson, R. J. (2008). Rethinking crime and immigration. Contexts, 7, 28–33. Sampson, R. J., & Bean, L. (2006). Cultural mechanisms and killing fields: A revised theory of community-level racial inequality. In R. Peterson, L. Krivo, & J. Hagan (Eds.), The many colors of crime: Inequalities of race, ethnicity, and crime in America (pp. 8–36). New York: New York University Press. Sampson, R. J., Morenoff, J., & Raudenbush, S. (2005). Social anatomy of racial and ethnic disparities in violence. American Journal of Public Health, 95, 224–232. Sanders, A. E. (2010). A Latino advantage in oral health-related quality of life is modified by nativity status. Social Science & Medicine, 71, 205–211. Scribner, R. (1996). Paradox as paradigm: The health outcomes of Mexican Americans. American Journal of Public Health, 86(3), 303–305. Singh, G. K., & Siahpush, M. (2002). Ethnic-immigrant differentials in health behaviors, mortality, and cause-specific mortality in the United States: An analysis of two national data bases. Human Biology, 74(1), 83–109.

72   R. SÁENZ and T. MORALES Suárez-Orozco, C., Rhodes, J., & Milburn, M. (2009). Unraveling the immigrant paradox: Academic engagement and disengagement among recently arrived immigrant youth. Youth & Society, 41(2), 151–185. Suárez-Orozco, C., & Suárez-Orozco, M. (1995). Transformations: Immigration, family life, and achievement motivation among Latino adolescents. Stanford, CA: Stanford University Press. Suárez-Orozco, C., & Suárez-Orozco, M. (2001). Children of immigration. Cambridge, MA: Harvard University Press. Subramanian, S. V., Jun, H-J., Kawachi, I., & Wright, R. J. (2009). Contribution of race/ethnicity and country of origin to variations in lifetime reported asthma: Evidence of a nativity advantage. American Journal of Public Health, 99(4), 690–697. Sussner, K. M., Lindsay, A. C., & Peterson, K. E. (2008). The influence of acculturation on breast-feeding initiation and duration in low-income women in the US. Journal of Biosocial Science, 40, 673–696. Swerdlow, A. J. (1991). Mortality and cancer incidence in Vietnamese refugees in England and Wales: A follow-up study. International Journal of Epidemiology, 20, 13–19. Tapia, M. (2010). Untangling race and class effects on juvenile arrests. Journal of Criminal Justice, 38, 255–265. Teller, C. H., & Clyburn, S. (1974). Trends in infant mortality. Texas Business Review, 48, 240–246. Telles, E., & Ortiz, V. (2008). Generations of exclusion: Mexican Americans, assimilation, and race. New York: Russell Sage. Turra, C. M., & Elo, I. T. (2008). The impact of salmon bias on the Hispanic mortality advantage: New evidence from social security data. Population Research and Policy Review, 27(5), 515–530. Turra, C. M., & Goldman, N. (2007). Socioeconomic differences in mortality among U.S. adults: Insights into the Hispanic paradox. Journals of Gerontology, Series B: Psychological Sciences, 62(3), S184–S192. Vaca, F. E., Anderson, C. L., & Hayes-Bautista, D. E. (2011). The Latino adolescent male mortality peak revisited: Attribution of homicide and motor vehicle crash death. Injury Prevention, 17, 102–107. Valenzuela, A. (1999). Subtractive education: U.S. Mexican youth and the politics of caring. Albany: State University of New York Press. Vega, W., Alderete, E., Kolody, B., & Aguilar-Gaxiola, S. (1998). Illicit drug use among Mexicans and Mexican American in California: The effects of gender and acculturation. Addiction, 93(12), 1839–1850. Vega, W. A., & Amaro, H. (1994). Latino outlook: Good health, uncertain prognosis. Annual Review of Public Health, 15, 39–67. Vega, W. A., Kolody, B., Valle, R., & Weir, J. (1991). Social networks, social support, and their relationship to depression among immigrant Mexican women. Human Organization, 50(2), 154–162. Velez, M. (2006). Lower rates of homicide in Latino versus Black neighborhoods. In R. Peterson, L. Krivo, & J. Hagan (Eds.), The many colors of crime: Inequalities of race, ethnicity, and crime in America (pp. 91–107). New York: New York University Press.

The Latino Paradox    73 Viruell-Fuentes, E. A. (2007). Beyond acculturation: Immigration, discrimination, and health research among Mexicans in the United States. Social Science & Medicine, 65, 1524–1535. Warner, B. D. (2003). The role of attenuated culture in social disorganization theory. Criminology, 41, 73–97. Williams, R. L. (1993). Health and length of residence among South Asians in Glasgow: A study controlling for age. Journal of Public Health Medicine, 15, 52–60. Wu, Z., & Schimmele, C. M. (2005). The healthy migrant effect on depression: Variation over time? Canadian Studies in Population, 32(2), 271–295. Wright, E. M., & Benson, M. L. (2010). Immigration and intimate partner violence: Exploring the immigrant paradox. Social Problems, 57(3), 480–503. Yabiku, S. T., Glick, J. E., Wendtz, E. A., Haas, S. A., & Zhu, L. (2009). Migration, health, and environment in the desert southwest. Population Environment, 30, 131–158. Zambrana, R. E., Scrimshaw, C. M., Collins, N. L., & Dunkel-Schetter, C. (1997). Prenatal health behaviors and psychosocial risk factors in pregnant women of Mexican origin: The role of acculturation. American Journal of Public Health, 87(6), 1022–1026. Zheng, W., & Schimmele, C.M. (2005). Repartnering after first union disruption. Journal of Marriage and Family, 67(1), 27–36.

This page intentionally left blank.

Chapter 4

Mortality Among the U.S. Hispanic Population Robert A. Hummer University of Texas at Austin Juanita J. Chinn University of Texas at Austin

Introduction For over 2 decades, researchers have shown that Hispanics in the United States tend to exhibit age-specific mortality rates and an overall life expectancy that is similar to non-Hispanic Whites, despite the much overall low socioeconomic status of the Hispanic population (Markides & Coriel, 1986; Markides & Eschbach, 2005, 2011). This overall similarity in mortality between the two populations has been termed the “epidemiologic paradox,” and has garnered a great deal of research interest. In fact, the concept of the epidemiologic paradox has become the defining issue among researchers who study Hispanic mortality patterns and trends in the United States. Since the mid-1980s, demographic and public health researchers have exhibited sustained interested in Hispanic mortality patterns. This is both because of the intriguing idea of the epidemiologic paradox, as well

The Demography of the Hispanic Population, pages 75–93 Copyright © 2012 by Information Age Publishing All rights of reproduction in any form reserved.

75

76   R. A. HUMMER and J. J. CHINN

as because of the substantial growth and increasing diversity of the Hispanic population over the last 3 decades. In this chapter, we review recent literature and describe patterns of infant, child, young adult, and older adult mortality patterns among Hispanics. We provide as much detail regarding Hispanic-origin subgroups and nativity distinctions as the data or previous analyses allow. Throughout the chapter, we give significant attention to the data, as well as methodological and substantive issues that are important when estimating and understanding Hispanic mortality patterns. We also provide a new analysis of Hispanic adult mortality in the United States, focusing most specifically on the U.S.-born Mexican American and Mexican-born Mexican immigrant subpopulations in comparison to non-Hispanic Whites and non-Hispanic Blacks. Together, our review and new analysis provide continuing evidence that U.S. Hispanics have overall mortality levels that are similar to those of non-Hispanic Whites in the United States. At the same time, there is considerable diversity in the mortality experiences of Hispanics when national origin groups are compared with one another and when immigrant-native born distinctions are made. Moreover, there are important methodological and data issues that should be given serious consideration whenever Hispanic mortality estimates are presented and/or comparisons to other population groups are made. Hispanic Infant and Child Mortality Infant Mortality Infant mortality refers to deaths that occur to persons who are less than 1 year old. The most important measure of infant mortality among populations is the infant mortality rate, which refers to the number of infant deaths per 1,000 live births in a population during a given year. The infant mortality rate is often considered to be a key measure of the overall health of a population (Frisbie, 2005). Most research on Hispanic infant mortality has centered on the relatively low, or favorable, infant mortality rate exhibited by the Hispanic population in the United States. Evidence for relatively low infant mortality among the U.S. Hispanic population was first uncovered by Teller and Clyburn (1974). Based on Texas data from the mid-1960s, they found a slightly lower infant mortality rate for the Hispanic population than for non-Hispanic Whites. Later, historical demographic work found that the infant mortality rate for the Hispanic population was not always similar to or lower than non-Hispanic Whites. In fact, using historical data from Bexar County (San Antonio), Texas, the rate of infant mortality of Hispanics was found to be much

Mortality Among the U.S. Hispanic Population     77

higher than that of Whites throughout the first half of the 20th century, but was characterized by a rapidly closing disparity with Whites between 1950 and 1980 (Forbes & Frisbie, 1991). By 1980, data from both Texas and California exhibited near-parity between the infant mortality rates of the Hispanic and non-Hispanic White populations (Forbes & Frisbie, 1991; Williams, Binkin, & Clingman, 1986). This similarity in infant mortality rates between Hispanics and Whites was later echoed in national-level data on infant mortality from the 1980s and 1990s (Becerra, Hogue, Atrash, & Perez, 1991; Hummer et al., 1999a; Singh & Yu, 1996a). The most currently available infant mortality data for Hispanics, published by the National Center for Health Statistics (Mathews & MacDorman, 2010), are summarized in Table  4.1. Rates for non-Hispanic Blacks and non-Hispanic Whites are listed in Table 4.1 for comparison purposes. The overall Hispanic infant mortality rate for 2006 (not shown in the table) was 5.41. But this overall rate is characterized by some fairly extensive diversity in rates when both Hispanic national origin groups and maternal nativity are considered. For example, infants of Puerto Rican mothers have an overall infant mortality rate of 8.01, while the overall rate for infants of Central/South American mothers is 4.52. Thus, the overall infant mortality rate among infants born to Puerto Rican women is 77% higher than among infants born to Central/South American women. The largest Hispanic group in the United States, infants of Mexican-origin women, has an overall infant mortality rate of 5.34, which is 20% lower than the overall Table 4.1  Infant Mortality Rates per 1,000 Births for U.S. HispanicOrigin Subgroups and Comparison Populations by Nativity, 2006a

Mexican Origin Puerto Ricanb Cuban Central/South American Other Hispanicc Non-Hispanic Black Non-Hispanic White Total United States

U.S.-Born Mothers

Foreign-Born Mothers

Total

5.92 7.56 5.30 4.99 5.61 13.54 5.59 7.03

4.98 8.70 4.76 4.43 5.48 10.01 4.04 5.09

5.34 8.01 5.08 4.52 5.78 12.90 5.57 6.68

Data for this table are taken from Table 2 in the recent report by Mathews and MacDorman (2010). b The category of “Foreign-Born Mothers” for Puerto Ricans refers to women who were born in Puerto Rico and who gave birth in the mainland United States in 2006. c The specific rates for U.S.-born Mothers and Foreign-Born Mothers appear too low in this row, in comparison to the Total Infant Mortality Rate in this row. This is most likely because there is some missing data on maternal birthplace in the dataset used by Mathews and MacDorman (2010) in this report. See their note below Table 2 for more information. a

78   R. A. HUMMER and J. J. CHINN

2006 U.S. rate of 6.68. Infant mortality rates are also lower among most groups when foreign-born mothers are compared with U.S.-born mothers. For example, infants born to foreign-born Mexican-origin women have an infant mortality rate of 4.98, while infants born to U.S.-born Mexican-origin women have an infant mortality rate of 5.92. The overall picture, then, of infant mortality among Hispanics is quite favorable—or at least better than the U.S. overall rate—but is also characterized by sizable differences across both national origin and maternal nativity groups. Three explanations have been offered for the epidemiologic paradox of relatively favorable infant mortality among the Hispanic population given their overall low socioeconomic status. The first is immigration selectivity. This explanation argues that immigrants are positively selected on good health (Franzini, Ribble, & Keddie, 2001; Markides & Eschbach, 2005, 2010); given the relatively high levels of immigration among the Hispanic population in recent decades, such positive selection is thus reflected in their overall low infant mortality rate. Cultural factors have also been offered as an explanation for overall low Hispanic infant mortality. This perspective suggests that healthy behaviors (such as nonsmoking, refraining from drug and alcohol use, and good dietary practices) and strong family ties among Hispanics in the United States, particularly in the immigrant generation, are instrumental in helping to explain the relatively low infant mortality rates of Hispanics (Franzini et al., 2001; Scribner, 1996). However, some researchers also suggest that a negative acculturation process and the adoption of negative behaviors and practices of Americans by Hispanic immigrants may work to help deteriorate the favorable infant mortality outcomes among Hispanics over time and across generations (Hummer, Powers, Pullum, Gossman, & Frisbie, 2007), although data limitations have prevented a rigorous test of this hypothesis. Most recent demographic work has considered a third explanation for the favorable infant mortality outcomes among Hispanics: data-quality issues. That is, there has been and continues to be debate that the relatively low infant mortality rate among Hispanics is in part due to the underreporting of Hispanic infant deaths in the United States, particularly within the Mexican-origin population (Palloni & Morenoff, 2001; Williams et al., 1986). The most likely reason for underreporting is that women and their U.S.-born infants may out-migrate from the United States to Mexico or elsewhere; such out-migration of women and infants would result in an underregistration of infant deaths in the United States. Clearly, it is plausible that Mexican-origin women giving birth in the United States might later return to Mexico. Such a process would result in a birth being registered in the United States, which increases the denominator of the Mexican-origin infant mortality rate in the United States. However, if some out-migrating infants later died in the first year of life in Mexico, such deaths would not

Mortality Among the U.S. Hispanic Population     79

be recorded in the United States, and thus, the resulting count of deaths in the numerator of the U.S. infant mortality rate would be artificially low. And there is no data system in place for linking infant deaths that may occur in Mexico back to the corresponding birth certificate records in the United States. A recent study by Hummer et al. (2007) tested this data-quality explanation using U.S.-linked birth and infant death files from 1995 through 2000. The results clearly showed that mortality among infants born to Mexican immigrant women, even within the first few days of life, are about 10% lower than that experienced by infants of non-Hispanic White native-born women. It is implausible that such favorable rates are artificially caused by the out-migration of Mexican-origin women and their newborn infants. Further, infants born to Mexican American women, while exhibiting rates of mortality slightly higher than those of non-Hispanic White women during the first weeks of life, also fare considerably better than infants born to nonHispanic Black women, with whom the former share similar socioeconomic profiles. These patterns are completely consistent with the concept of the epidemiologic paradox. Additional simulation calculations based on different mortality rates for hypothetical out-migrant women demonstrated that, even though out-migration of Mexican immigrant women and their infants surely can and does occur, with some subsequent deaths of U.S.-born infants going unrecorded in U.S. data, the extent to which this would need to happen to even moderately influence recorded U.S. infant mortality rates is very substantial. Thus, Hummer et al. concluded that the epidemiologic paradox for U.S. infant mortality among the Mexican-origin population— both among immigrant women and among U.S.-born women—is real and not a data artifact. Future research should focus on understanding why the Mexican immigrant population in particular exhibits the low infant mortality rates that it does, with one area of focus being the exceptionally healthy behaviors (e.g., very low rates of smoking, alcohol use, drug use) exhibited by immigrants to the United States, particularly women (Lopez-Gonzalez, Aravena, & Hummer, 2005). Such healthy behaviors by immigrant women may be especially important in helping to account for the low infant mortality rate of Mexican-origin infants. Child Mortality The exact definition of child mortality varies across research studies, with the age range of what is considered child mortality varying across contexts (Frisbie, Hummer, & McKinnon, 2009). Here we define childhood mortality as deaths occurring among U.S. individuals between the ages of 1 and 14. We separate childhood deaths into those occurring between ages 1 and 4, and

80   R. A. HUMMER and J. J. CHINN

those occurring at ages 5–14. The latter of these age groups, 5–14, generally has the lowest overall death rate of any age group throughout the life course. Child mortality rates for U.S. Hispanic subgroups and comparison populations, shown in Table 4.2, are expressed as the number of deaths in specific childhood age groups (by race/ethnicity and sex) per 100,000 persons in that same age-race/ethnic-sex group. These data are also taken from a recently published report by the National Center for Health Statistics (Heron et al., 2009). Table 4.2 shows that child mortality rates for Hispanic subgroups are roughly on par with those of the total U.S. population and non-Hispanic Whites. For example, the age 1–4 death rate for Mexican American females is 24.1, compared to 23.2 among non-Hispanic Whites and 26.3 among the overall U.S. population of aged 1–4 females. The relative similarity in child mortality rates for Hispanic subgroups is generally evident at both age groups 1–4 and 5–14. There is some indication that aged 5–14 mortality rates for both female and male Other Hispanics are elevated in comparison to non-Hispanic Whites and in comparison to the overall U.S. population of aged 5–14 children. At the same time, mortality rates during this stage of the life course are quite low and can exhibit irregularities on a year-to-year basis. Notably, child mortality rates for Cuban Americans and for aged 1–4 Other Hispanics are not shown because the National Center for Health Statistics deemed these rates to be too unreliable to be shown in their report because of sparse data (Heron et al., 2009, Table 5). Given that death rates among children are the lowest in the life course, there is relatively little published literature that focuses on U.S. child mortality trends and differentials by race/ethnicity. In perhaps the most com-

b

a b

b

17.0 9.6

33.7

Total U.S.

23.8 14.1

22.8

NH White

Panel B: Males 1–4 30.2 5–14 17.1

20.6 5.5

NH Black

b b

Other Hispanic

b

10.0

C/S American

Puerto Rican

Panel A: Females 1–4 24.1 5–14 12.7

Age Group

Cuban

Mexican

Table 4.2  Child Mortality Rates per 100,000 for U.S. Hispanic-Origin Subgroups and Comparison Populations, by Sex, 2006a

b

40.3 17.9

23.2 11.8

26.3 12.8

48.2 25.8

26.7 16.2

30.5 17.6

b

Data for this table are taken from the report by Heron et al. (2009). Data for this specific rate were deemed to be too sparse to be reliable (see Heron et al., 2009, Table 5).

Mortality Among the U.S. Hispanic Population     81

prehensive assessment to date, Singh and Yu (1996b) provided documentation of trends and differentials in U.S. child mortality between 1950 and 1993, focusing on the 1–4 and 5–14 age ranges. Given immigration trends and data availability, their documentation for Hispanic subgroups began in 1979–1981 and extended through 1989–1991. They show that child mortality rates declined for Hispanic subgroups across that decade and, similar to the patterns demonstrated for infant mortality above, rates for Hispanics were generally similar to non-Hispanic Whites. For example, the aged 1–4 mortality rate for Hispanic children was about 10% higher than non-Hispanic White children in 1989–1991, while the aged 5–14 mortality rate for Hispanic children was about 5% lower than non-Hispanic White children in 1989–1991. Singh and Yu also showed that mortality rates for both females and males were generally higher among Puerto Rican and Other Hispanic children than among Mexican American and Cuban American children, with Cuban American children demonstrating the lowest child mortality rates in 1989–1991 among the Hispanic subgroups. Innovative demographic work has used historical U.S. census data sources to estimate child mortality levels in 1890–1910 for Hispanics in comparison to other U.S. population subgroups (Gutmann, Haines, Frisbie, & Blanchard, 2000). This study found that Hispanics exhibited much higher child mortality than non-Hispanic Whites during this period, with rates among Hispanic children similar to those estimated for non-Hispanic Black children. Estimated mortality rates for Hispanic children were uniformly high across most states that the researchers examined (Arizona, California, Texas), although they found exceptionally high rates at the time among Hispanic children in New Mexico and in Florida (outside of the city of Tampa). Thus, like the historical work on Hispanic infant mortality discussed above (Forbes & Frisbie, 1991), the historical work on Hispanic child mortality strongly suggests that the similar contemporary death rates between Hispanics and non-Hispanic Whites was not the case in an earlier period of U.S. history. Hispanic Adult Mortality Officially Reported Adult Mortality Rates for Hispanics Table 4.3 uses the most recent version of official U.S. mortality data (Heron et al., 2009) to display 2006 age-specific adult mortality rates per 100,000 persons for Hispanic-origin subgroups and comparison populations. Numerator data for official mortality rates are taken from death certificates, while denominator data are taken from population estimates provided by the U.S. Census Bureau. The use of these two distinct sources results in somewhat lower mortality rates for Hispanic subgroups than should most

82   R. A. HUMMER and J. J. CHINN

Panel A: Females 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85+ Age-Adjusted Ratec Panel B: Males 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85+ Age-Adjusted Ratec a b

c

36.0 41.3 84.6 210.2 507.1 1319.1 3124.2 9423.5 480.6

36.3 49.2 153.0 300.0 634.7 1423.9

576.0

478.3

318.6

605.5

125.8 112.7 169.9 382.7 809.3 1836.3 4725.1

109.0 144.8 309.8 703.4 1243.1 2743.8

124.4 83.1 139.9 462.5 1056.5 2323.7

86.1 80.8 110.6 228.3 438.4 1116.7

197.2 274.1 393.0 766.4 1442.1

b

b

b

b

b

b

682.8

b b

910.8

68.0 229.5 398.4 1225.4 3231.1

20.5 34.8 50.3 134.6 296.9 719.2 2611.0

b

b

b b

676.9

b

452.4

85.1 113.2 193.7 377.2 727.6 1555.4 b b

Total U.S.

NH White

NH Black

Other Hispanic

C/S American

Cuban

Puerto Rican

Age Group

Mexican

Table 4.3  Age-Specific Adult Mortality Rates per 100,000 for U.S. Hispanic-Origin Subgroups and Comparison Populations, 2006a

52.8 42.9 42.8 111.4 62.5 64.3 254.1 136.3 141.6 562.2 299.8 317.7 1100.2 668.0 687.0 2281.4 1677.4 1677.9 5107.2 4460.7 4388.3 12350.5 13150.7 12759.0 828.4 660.0 657.8

176.9 107.6 119.3 266.0 141.1 146.8 405.6 233.1 238.7 944.2 515.1 541.0 1935.2 1064.0 1110.0 b 3736.8 2490.3 2516.2 b 7520.4 6278.3 6177.7 b 13403.1 14841.1 14309.1 900.8 1241.0 922.8 924.8

Data for this table are taken from Table 5 in the report by Heron et al. (2009). Data for this specific rate were deemed to be too sparse to be reliable (see Heron et al., 2009, Table 5). The age-adjusted death rate standardizes for age structure differences across populations. The age-adjusted mortality rates reported here pertain to the complete life course and not just adult age groups. However, because U.S. mortality is so heavily concentrated in the adult ages, these age-adjusted rates largely reflect the mortality levels of U.S. adults.

likely be the case (Arias, Eschbach, Schauman, Backlund, & Sorlie, 2010; Rosenberg et al., 1999), a methodological issue to which we return below. It is clear from Table 4.3 that for most age groups, adult mortality rates are modestly-to-moderately lower among Hispanic subgroups than non-Hispanic Whites and substantially lower than non-Hispanic Blacks. Lower age-specific adult mortality for Hispanic subgroups in comparison to non-Hispanic Whites and Blacks is especially pronounced among Central/South Americans, Cuban Americans, and Mexican Americans, with Central/South Ameri-

Mortality Among the U.S. Hispanic Population     83

cans displaying the lowest adult mortality of any subgroup examined. Relatively low age-specific mortality for most Hispanic subgroups is also especially pronounced among women, with only two minor exceptions (Puerto Rican women aged 35–44 and those aged 45–54, compared to non-Hispanic White women); Mexican American, Puerto Rican, Cuban American, and Central/ South American women consistently exhibit lower mortality rates than both non-Hispanic Whites and non-Hispanic Blacks in every adult age group. Among men, Mexican Americans and Cuban Americans exhibit modestly higher mortality than non-Hispanic Whites at ages 15–24, most likely due to the higher Hispanic rate of homicide mortality among young men (Miniño, 2010). But throughout the remainder of the life course, age-specific mortality rates are higher among non-Hispanic White and non-Hispanic Black men than among Mexican American and Cuban American men. In contrast, both Puerto Rican and Other Hispanic men exhibit modest-tomoderately higher mortality than non-Hispanic White men in the 15–24, 25–34, 35–44, 45–54, and 55–64 age groups. In most of these age groups, however, mortality rates among Puerto Rican and Other Hispanic men are lower than those of non-Hispanic Black men. Relatively low age-specific mortality rates for Hispanic subgroups are also reflected in each subgroup’s age-adjusted mortality rate, which is listed in the bottom row for both women and men in Table 4.3. The age-adjusted mortality rate standardizes for age structure differences across populations and thus provides a useful summary indicator of the overall mortality level of populations in comparison to one another (McGehee, 2004). The ageadjusted mortality rates reported in Table 4.3 pertain to the complete life course and not just adult age groups. However, because U.S. mortality is so heavily concentrated in the adult ages, the reported age-adjusted rates largely reflect differences in the mortality levels of U.S. adults. Specifically, females’ and males’ age-adjusted death rates for all of the Hispanic subgroups are lower than among non-Hispanic Blacks and non-Hispanic Whites, with Central/South Americans, Cuban Americans, and Mexican Americans exhibiting the lowest age-adjusted rates for both women and men. The lower rate of age-adjusted mortality among Mexican Americans in comparison to both non-Hispanic Whites and non-Hispanic Blacks is again illustrative of the epidemiologic paradox, particularly given the overall low socioeconomic status of the Mexican American population (Markides & Eschbach, 2010; Tienda & Mitchell, 2006, Chapter 5). Data, Methodological, and Substantive Issues Regarding Hispanic Adult Mortality An important body of research has been critical of official adult mortality estimates among Hispanics and has offered alternative explanations

84   R. A. HUMMER and J. J. CHINN

(e.g., racial/ethnic misclassification and return-migration selectivity) for the relatively low adult mortality that has been found for the Hispanic population (Palloni & Arias, 2004; Palloni & Morenoff, 2001). Indeed, recent demographic work in this area has attempted to account for the effect of out-migration on Hispanic mortality estimates in the United States and has dealt with the issue of disparate race/ethnic reporting across different data sources and across time (Abraido-Lanza, Dohrenwend, Ng-Mak, & Turner, 1999; Arias et al., 2010; Elo, Turra, Kestenbaum, & Ferguson, 2004; Eschbach, Kuo, & Goodwin, 2006; Eschbach, Stimpson, Kuo, & Goodwin, 2007; Hummer, Benjamins, & Rogers, 2004; Liao et al., 1998; Palloni & Arias, 2004; Rosenberg et al., 1999; Smith & Bradshaw, 2006; Swallen & Guend, 2003; Turra & Elo, 2008). These studies arrive at a consensus that official U.S. government reported Hispanic adult mortality rates are too low because of data-quality issues. Most recently, Arias and colleagues (2010) assessed the extent of underreporting of Hispanic deaths in official mortality statistics by comparing official mortality estimates to data from the National Longitudinal Mortality Study (NLMS). The NLMS is extremely useful for the ascertainment of Hispanic mortality because it is very large, includes self-reports of Hispanic identity from multiple Current Population Surveys, and is characterized by very high quality mortality follow-up. Their comparison of official mortality data to the NLMS found that Hispanic-origin reporting on recent U.S. certificates is reasonably good, resulting in 5% net underascertainment of Hispanic mortality in the official data. Once Arias and colleagues adjusted the official data based on this level of 5% underascertainment of mortality, overall Hispanic mortality at every age group throughout the life course remained lower than that for non-Hispanic Whites in the official data. Arias et al. further note that Hispanic subgroup mortality estimates in the official U.S. data remain problematic, particularly for subgroups other than Mexican Americans, Puerto Ricans, and Cuban Americans. That being said, the Arias et al. study is currently the best available that has tackled the underascertainment of Hispanic deaths in official U.S. data and, most importantly, continues to find an overall Hispanic mortality advantage compared to non-Hispanic Whites. The Arias et al. findings were also supported by recent work on the same topic specific to the states of California and Texas (Eschbach et al., 2006, 2007). The extent to which out-migration selectivity influences estimates of Hispanic adult mortality was most effectively tackled in recent work by Turra and Elo (2008). They examined the impact of out-migration from the United States (to Mexico and elsewhere) on reported mortality rates among primary social security beneficiaries (ages 65 and older) in the United States. They found evidence that both Hispanic and non-Hispanic White out-migrants from the United States do, indeed, exhibit higher mortality than persons in their same racial/ethnic group who do not out-migrate. Thus, deaths to these relatively high-risk out-migrants are not included in typical U.S. datas-

Mortality Among the U.S. Hispanic Population     85

ets and, as a consequence, U.S. datasets may tend to underestimate Hispanic adult mortality. However, even when Turra and Elo took such out-migration into account, Hispanic mortality in the United States for social security beneficiaries aged 65 and older remains between 11% and 18% lower than that of non-Hispanic Whites—depending on the specific age/sex group in question—in large part because the volume of out-migration is simply not large enough to significantly influence adult mortality rates in the United States. Thus, while they show evidence for what is consistent with what some have referred to as a “salmon bias” in terms of Hispanic mortality in the United States, the overall magnitude of its effect on Hispanic versus non-Hispanic White mortality patterns was found to be very slight. Beyond the data and methodological issues, a substantial literature has developed over the last 20 years providing evidence from a number of different datasets that Hispanics appear to have somewhat favorable adult mortality risks when compared to non-Hispanic Whites and much more favorable risks compared to non-Hispanic Blacks. At the same time, such patterns have been found to vary by Hispanic national-origin group and nativity and to some extent by gender and age (Abraido-Lanza et al., 1999; Borrell & Crawford, 2009; Elo et al., 2004; Hummer, Rogers, Amir, Forbes, & Frisbie, 2000; Rosenwaike, 1987; Sorlie, Backlund, Johnson, & Rogot, 1993). Most recently, Borrell and Crawford investigated Hispanic adult mortality patterns by national origin, nativity status, and racial identification using data from the National Health Interview Survey Linked Mortality File (NHIS-LMF). Like the National Longitudinal Mortality Study, the NHIS-LMF is very useful for estimating Hispanic adult mortality patterns because of its very large size, self-reports of racial and ethnic identification, and high-quality matches to subsequent mortality records. Their findings are consistent with other previous studies in finding lower overall adult mortality risk for Hispanics relative to non-Hispanic Whites and Blacks, with particularly low mortality among older (aged 65 and above) foreign-born Hispanics. At the same time, Borrell and Crawford find higher mortality among Hispanics aged 25–44 years of age compared to nonHispanic Whites and Blacks, in large part due to the heightened mortality risk in that age range among Puerto Rican-born and Mexican-born women and men. This finding of higher mortality among young adult Hispanic immigrants contrasts with that of Eschbach et al. (2007), who used official U.S. mortality data from California and Texas and demonstrated lower mortality among young adult Hispanic immigrants compared to non-Hispanic White native-born men and women. The Borrell and Crawford study is also the first to examine racial identification and adult mortality among Hispanics, and they show that Hispanic Black adults exhibited similar death rates to non-Hispanic White adults for both genders and all age groups examined. Further, adult mortality patterns among Black Hispanics were found to be much more similar to White Hispanics than to non-Hispanic Blacks.

86   R. A. HUMMER and J. J. CHINN

A New Analysis of Hispanic Adult Mortality In this section, we provide a new analysis of Hispanic adult mortality patterns in the United States using the Third National Health and Nutrition Survey Linked Mortality Files (NHANES-LMF). This dataset includes individuals from NHANES surveys conducted between 1988 and 1994, with a mortality follow-up period through the end of 2006 to analyze mortality patterns for adults who were aged 35–89 at the time of the survey. This dataset is particularly useful for mortality analysis because it includes a richer set of social integration and health behavior measures than is available in other datasets that are typically used to examine Hispanic mortality patterns. At the same time, we limit our focus in this new analysis to the U.S.-born Mexican American (Mexican American) and foreign-born Mexican immigrant (Mexican immigrant) Hispanic subpopulations due to the small number of individuals and deaths in the dataset for other Hispanic subgroups. NonHispanic Blacks and non-Hispanic Whites are included here for comparison. We estimate hazard ratios for all-cause mortality; the results are presented in Table 4.4. A hazard ratio greater than 1.000 demonstrates a higher risk of death when compared to the reference category (for race/ethnicity, the reference category is non-Hispanic White) and a hazard ratio less than 1.000 shows a lower, more favorable risk of death than the reference category. Table 4.4  Hazard Ratios for All-Cause Mortality among U.S. Adults Aged 35–89 Variable Age (Continuous) Sex   Female (ref.)  Male Race/Ethnicity   MexAm   MexIm   nhBlack   nhWhite (ref.) Education   High Sch or More (ref.)   Less Than High Sch Poverty Status   Not in Poverty (ref.)   Near Poverty   In Poverty   No Poverty Info

Model 1

Model 2

Model 3

1.092***

1.088***

1.096***

— 1.490***

— 1.551***

— 1.571***

0.978 0.913 1.337*** —

0.842** 0.731*** 1.160*** —

0.808*** 0.732*** 1.081+ —

— 1.102**

— 1.032

— 1.355*** 1.468*** 1.243***

— 1.234*** 1.170** 1.056

Mortality Among the U.S. Hispanic Population     87 Table 4.4 (continued)  Hazard Ratios for All-Cause Mortality among U.S. Adults Aged 35–89 Variable Marital Status   Married (ref.)   Unmarried Household Size   1   2–4 (ref.)   5 or more Visit Family   More than Weekly (ref.)   Weekly or Less Visit Neighbors   More than Weekly (ref.)   Weekly or Less Club Membership   Yes (ref.)  No Time Lived in City   5 or More Yrs (ref.)   1–4 Years   < 1 Year Religious Attendance   More than Weekly (ref.)   Weekly or Less   Never Activity Level   More than Peers (ref.)   Same as Peers   Less than Peers Smoking   Never (ref.)   Former   Current BMI   30 or greater   20–29 (ref.)   Less than 20 –2 Log Likelihood Degrees of Freedom Unweighted N

Model 1

Model 2

Model 3

— 1.371*** 0.830*** — 1.018 — 1.035 — 0.946+ — 1.148*** — 1.007 1.207 — 1.097* 1.200*** — 1.262*** 1.850*** — 1.209*** 1.701***

78,244.66 5 11,745

77,096.48 9 11,644

Source: NHANES III 1988–1994 Linked Mortality Files through 2006 + p