Immigrants And Citizen Engagement essay paper
Immigrants And Citizen Engagement
Voting Behavior of Naturalized Citizens
Voting Behavior of Naturalized Citizens: 1996-2010
Sarah R. Crissey Thom File
U.S. Census Bureau Social, Economic and Housing Statistics DivisionImmigrants And Citizen Engagement
4600 Silver Hill Rd Washington, DC 20233
Please direct all correspondence to:
Thom File Thomas.a.file@census.gov
Keywords Voting, citizenship, naturalization, immigration
This report is released to inform interested parties of (ongoing) research and to encourage discussion (of work in progress). The views expressed on (statistical, methodological, technical, or operational) issues are those of the author(s) and not necessarily those of the U.S. Census Bureau. A previous version of this paper was presented at the Annual Meeting of the Population Association of America New Orleans, LA April 16-19, 2008.Immigrants And Citizen Engagement
Voting Behavior of Naturalized Citizens 1
Voting Behavior of Naturalized Citizens: 1996-2010
ABSTRACT
In 2010, 35.8 million foreign-born adults lived in the United States, including 16.9
million naturalized citizens. Research documents that naturalized citizens are less likely
to register and vote than native citizens. Since Bass and Casper’s (2001) baseline national
estimates from the 1996 Current Population Survey (CPS), the number of naturalized
citizens has doubled. Given population changes and the increasing political debate overImmigrants And Citizen Engagement
immigration, we explore how nativity influenced voting behavior from 1996 to 2010.
Using the Voting and Registration Supplement to the CPS, which surveys about 80,000
adults every other November, we address whether naturalized citizens continue to be less
likely to register and vote, and whether the nativity status effect is consistent across time
and in both presidential and congressional elections. Our regression results suggest that,
net of social and demographic factors, naturalized citizens are less likely to register and
vote than native citizens across all years. We find evidence that the magnitude of theImmigrants And Citizen Engagement
nativity association has increased over the course of the fourteen-year period –
particularly for presidential elections. Furthermore, we find tentative evidence that
nativity may have a stronger effect on voting behavior in congressional versus
presidential elections.
INTRODUCTION
According the 2010 Current Population Survey, approximately 38.7 million
foreign-born persons live in the United States, with 17.5 million naturalized citizens.
Naturalized citizens age 18 and over can participate in the political process, but research
Voting Behavior of Naturalized Citizens 2
suggests that they are less likely to register and vote than native citizens (Bass & Casper
2001; Casper & Bass 1998; Cassel 2002; Cho 1999; Day & Holder 2004; DiSipio,
Masouka, & Stout 2006; Garcia 2003; File 2008; File & Crissey 2010; Holder 2006;Immigrants And Citizen Engagement
Hero 1992; Jamieson, Shin, & Day 2002; Manza & Brooks 1999). Explanations for the
nativity gap vary, but a broad theoretical framework of political participation suggests
that citizens who are connected to and invested in society are more likely to participate in
political endeavors such as voting (Brady, Cogan, & Fiorina 2000; Campbell, Gurin, &
Miller 1960; Cho 2006; Lazarfeld, Berelson, & Gaudet 1948; Lipset 1980; Putman 2000),
while groups such as immigrants – who are potentially less connected and invested – are
less likely to participate (Alba & Nee 1997; Cain, Kiewet & Uhlaner 1991; DeSipio
1996). Political scholarship highlights the consistent importance of social predictors of
voting behavior, despite year-to-year variability in registration and voting rates (Jackson
& Carsey 1999; Manza & Brooks 1999). Researchers cite socioeconomic characteristics,
emerging cultural factors, and institutional barriers to participation – such as languageImmigrants And Citizen Engagement
differences – as reasons for the nativity gap in voting behaviors (Fraga & Segura 2006;
Jackson & Carsey 1999; Jones-Correa 2005; Manza & Brooks 1999; Miller & Shanks
1996; Segura, Barreto, & Woods 2004; Segura, Nicholson, & Pantoja 2007; Xu 2005).
This research builds on the work of Bass and Casper (2001), which assessed
naturalized citizens’ registration and voting using the 1996 Current Population Survey,1
and found that net of other factors, naturalized citizens were less likely to register and to
vote than native citizens. Since the production of these baseline estimates, the U.S.
population has added over 5 million new naturalized citizens (Rytina & Saeger 2005),
while little scholarly work has been produced to assess the continued impact or size of 1 In 1996 naturalization status was first available in the CPS during a presidential election year.Immigrants And Citizen Engagement
Voting Behavior of Naturalized Citizens 3
the nativity gap. Furthermore, the political debate about immigration has become
increasingly heated, as have academic debates regarding whether hot-button ideological
issues matter more than social factors in terms of predicting voting behavior. Given these
changes in the population, the centrality of the immigration issue, and the continued
debate concerning the causal impact of social factors as voting predictors, our research
seeks to extend Bass and Casper’s (2001) findings through the most recent election data
from the CPS.
RESEARCH QUESTIONS AND HYPOTHESES
Our analyses address three research questions. First, net of other predictors of
voting behavior, are naturalized citizens less likely to register and to vote in elections
held in the last decade? Second, has the magnitude of the nativity status effect changed
over this time span? Finally, does nativity status have the same effect in different types of
elections, namely between presidential and congressional election? We hypothesize that,
due to factors such as societal investment and barriers to voting, nativity will influenceImmigrants And Citizen Engagement
voting behavior across all years. For our second question, we speculate that as the
naturalized citizen population increases and diversifies, the magnitude of the nativity
effect will increase across election years. Finally, we hypothesize that nativity status will
exert a stronger effect on participation in congressional elections since voter turnout is
generally lower than in presidential elections, and is especially low for groups with large
naturalized citizen populations (File, 2008; Holder, 2006).
Voting Behavior of Naturalized Citizens 4
DATA AND METHODS
Data and Measures
Data: To explore these questions, we use data from the Current Population
Survey (CPS) – a nationally representative survey of the civilian, non-institutionalized
population administered monthly to a sample of approximately 72,000 housing units.
Since 1964, the November CPS has included a bi-annual supplement on voting and
registration to coincide with national congressional and presidential elections. This
includes questions about voting, registration, mode and timing of voting, and reasons for
not voting, and is asked of all household members who are U.S. citizens aged 18 and
over, although the CPS has only gathered citizenship data in a consistent way since 1994.
Our analyses use data from the eight November Voting Supplements collected from 1996
to 2010, with sample sizes ranging from about 77,000 to 89,000 unweighted cases (see
Tables 2 and 3 for sample sizes for each year).2
Dependent Variables: We focus on two dependent variables: voter registration
and voting. The first question in the supplement refers to voting in the most recent
election and asks; “In any election some people are not able to vote because they are sick
or busy or have some other reason, and others do not want to vote. Did (you/name) vote
in the election held on Tuesday, November (date), (year)?” Respondents who say “no”
are then asked the registration question: “(Were you/Was name) registered to vote in the
November (day), (year) election?” We constructed a dichotomous variable for each
2 The data in this report are from the 1996, 1998, 2000, 2002, 2004, 2006, 2008, and 2010 Voting supplements to the November Current Population Survey, and the estimates in it are based on responses from a sample of the population and may differ from actual values because of sampling variability or other factors. Further information about the source and accuracy of the estimates is available in Attachment 16 of www.census.gov/apsd/techdoc/cps/cpsnov06.pdf. Full documentation can be found in the complete document.
Voting Behavior of Naturalized Citizens 5
outcome (1=yes, 0=no). Respondents who report “yes” for the voting question are coded
as “yes” for registering to vote3. Table 1 shows the number and percentage of the adult
citizen population who registered to vote and voted across each year.
Independent Variables: The focal independent variable is nativity status, which
we operationalized as a dichotomous variable for whether the respondent was native or
was foreign born and became a citizen via naturalization (1=naturalized, 0=native). Table
1 displays the distribution of the native and naturalized citizens for the adult U.S. citizen
population across survey years, as well as the distribution across the dependent variables.
Apart from the nativity status variable, we also include a series of categorical
independent variables long established for being related to voting behavior (Bass &
Casper 2001). These include sex, race and Hispanic origin, marital status, educational
attainment, employment status, occupation type, home ownership, duration of residence,
and region of residence.4 We also include continuous measurements for both age and
income (as well as a dichotomous flag identifying imputed income values).
Analytical Plan
We estimate multivariate models predicting the effect of nativity status on the two
outcomes for each survey year. Due to the complex sampling design of the CPS, a
normalized person weight and design-effects are used to adjust estimates and standard
errors. Models for individual years address the first research question of whether, net of
the control variables, nativity status is statistically significantly associated with the voting
behavior outcomes. We address the second research question about variation in this
association across time by testing whether the nativity status coefficients vary from each
3 For all voting and registration questions in the CPS, respondents are able to offer proxy responses for other members of their household. 4 See Tables 2 and 3 for omitted categories.
Voting Behavior of Naturalized Citizens 6
other across years within the same type of election. Finally, we address the third research
question about the pattern of association between nativity status and voting across type of
election, by comparing the results from the presidential elections to congressional
election in proximate years.
RESULTS
The descriptive statistics presented in Table 1 show the size of the total U.S.
citizen population age 18 and over, by nativity status and year. Over this 14-year period,
the population eligible to vote has increased from 180 million to 211 million as the
number of naturalized citizens increased from around 8 million to 17 million5. While
naturalized citizens comprised a small percentage of the population overall, the
proportion of naturalized citizens in the voting population grew over this decade. In
1996, naturalized citizens represented about 5 percent of the adult citizen population. By
2010, this percentage had risen to about 8 percent. In each election year, a larger
percentage of native citizens reported registering to vote and voting compared to
naturalized citizens. For instance, in the most recent election of 2010, 66 percent of
native citizens registered to vote, compared to 54 percent of naturalized citizens. In the
same year, 46 percent of native citizens reported voting versus 37 percent of naturalized
citizens. Naturalized citizens have also been growing as part of the voting population.
They were four-percent of those who voted in 1998 and 7 percent in 2010.
The difference between turnout rates between natives and naturalized citizens
expanded from 6 percent in 1996 to 12 points in 2006, before receding to 9 points in
5 The estimates in this paper are based on responses from a sample of the population. As with all surveys, estimates may vary from the actual values because of sampling variation or other factors. All comparisons made in this paper have undergone statistical testing and are significant at the 95-percent confidence level unless otherwise noted.
Voting Behavior of Naturalized Citizens 7
2010. In sum, although naturalized citizens have grown as a part of the overall population
and eligible electorate, they have not kept pace with the rest of the nation in terms of
turnout.
Table 2 displays results from logistic regression analyses predicting voter
registration across survey year.6 In each year, naturalized citizens are less likely to vote
than native citizens. For instance, in 1996 the odds that a naturalized citizen registered to
vote were 36 percent less than the odds that a native citizen registered. In 2010,
naturalized citizens were about half as likely to vote as native citizens. Table 3 displays
these same analyses predicting reported voting. Similar to the findings for registration,
net of control variables, naturalized citizens were less likely to vote across all years
relative to native citizens. In 1996, naturalized citizens were about 25 percent less likely
to vote than native citizens. In 2010, the odds that a naturalized citizen voted were only
59 percent of the odds that a native citizen voted.
Tables 4 and 5 display results of the statistical tests used to compare logistic
regression coefficients across years. In these tables, cells marked with an asterisk indicate
a statistical difference between the coefficients at the 95-percent confidence level. Our
second research question asks about differences in the nativity effect over time and by
election year. We address this question by comparing the magnitude of the naturalized
citizen coefficient across years, but only within the same election type. The solid shaded
cells in Tables 4 and 5 represent comparisons of presidential years, while the striped cells
represent congressional years.
For registration models (Table 4), the magnitude (in terms of absolute values) of
6 Results are displayed as odds ratios, with values above 1 indicating higher odds of registering to vote compared to the reference group and values below 1 indicting lower odds. An asterisk indicates that the odds ratio is statistically different from 1 at the 95-percent confidence level.
Voting Behavior of Naturalized Citizens 8
the 1996 presidential year coefficient was statistically smaller than only one other
presidential year, 2008. This represented the only statistically significant difference in
registration across presidential years. While there was a nominal increase in the
magnitude of the coefficients across congressional elections, none of these differences
reached statistical significance.
For voting models (Table 5), there were only two statistically significant
differences between presidential years (1996 compared to 2008 and 2000 compared to
2008, respectively). For congressional years, meanwhile, there was only one statistically
significant difference (between 1998 and 2006). For presidential elections, these results
suggest that although there is not a strong voting or registration pattern emerging from
one single election to another, over time there has been a cumulative increase in the
magnitude of the citizenship status effect. This increased magnitude applies to both
registration and voting, as significant differences occur between the earliest and most
recent presidential elections on both outcomes. At the congressional level, evidence of
this effect increase is not present for registration but does exist for voting.
The final research question addresses variation in the nativity effect by election
type. In Tables 4 and 5, this research question is addressed in the top diagonal line of
boxes (outlined in bold), by comparing across election type within proximate years. In
the voter registration coefficients presented in Table 4, two comparisons in the diagonal
are statistically different: 1996-1998 and 2004-2006. In these pairings, the absolute value
of the congressional year coefficient is larger than the presidential year. This suggests
that in these two comparisons, the effect of nativity on voter registration was stronger in
congressional elections than presidential elections. The pattern of stronger effects in
Voting Behavior of Naturalized Citizens 9
congressional years was present in two comparisons in the models predicting voting
(Table 5): 2000 – 2002 and 2004 – 2006. This suggests that the effect of nativity is
generally stronger in congressional compared to presidential elections, although the effect
was not consistently statistically significant.
CONCLUSIONS
These results provide support for our first hypothesis that naturalized citizens are
less likely to register and to vote compared to native citizens. This suggests that Bass and
Casper’s (2001) findings for nativity status and voting behaviors were not isolated to the
1996 election and that the effect of nativity status remains a relevant predictor of voting
behavior across all elections since then. In the most recent election (2010), we found that
naturalized citizens are about half as likely to register to vote, and about 40 percent less
likely to vote compared to native citizens.
We also find support for our second hypothesis about the increasing importance
of nativity over time. For presidential elections, in particular, there has been an overall
increase in the magnitude of the nativity status effect for both voting and registration.
Although a similar increase did not occur for congressional election registration, the same
overall magnitude increase exists for congressional level voting. At a minimum, this
suggests that the effect of nativity has increased for voting in both types of elections since
1996.
For our third hypothesis, we found tentative evidence of differences across
election type. The magnitude of the nativity effect was at least as strong, if not more so,
in congressional versus presidential years. However, established election research
Voting Behavior of Naturalized Citizens 10
suggests that congressional and presidential elections are not directly comparable, so
these results should be interpreted with extreme caution.
Taken together, this research suggests that nativity status is an important
determinant of voting behavior. Despite growth in the foreign-born population and the
increased attention paid to immigration in American politics, naturalized citizens are not
capitalizing on their rights as citizens to participate in the electoral process. In fact, they
were less likely to do so in the most recent election than a decade ago. This finding
correlates strongly with the recent literature that identifies social factors as consistently
relevant predictors of electoral behavior.
Voting Behavior of Naturalized Citizens 11
REFERENCES
Alba, R. D. & Nee, V. (1997). Rethinking assimilation theory for a new era of immigration. International Migration Review, 31, 826-74.
Bass, L. E. & Casper, L. M. (2001). Difference in the registration and voting between
native-born and naturalized Americans. Population Research and Policy Review, 20, 483-511.
Brady, D. W., Cogan, J. F. & Fiorina, M. P. (2000). Continuity and change in house
elections. Stanford CA: Stanford University Press. Cain, B. E., Kiewet, R., & Uhlaner, C.J. (1991). The Acquisition of Partisanship Among
Latinos and Asian Americans. American Journal of Political Science, 35 (2), 390- 422.
Campbell, A., Gurin, G. & Miller, W. E. (1960). The American voter. New York: Wiley. Casper, L. M. and Bass, L. E. (1998). Voting and registration in the election of
November 1996 (P20-504). Washington, DC: U.S. Bureau of the Census. Cassel, C. A. (2002). Hispanic turnouts: Estimates from validated voting data. Political
Research Quarterly, 55, 391-408. Cho, W.K. (1999). Naturalization, Socialization, Participation: Immigrants and (Non-)
Voting. The Journal of Politics, 61 (4), 1140-1155. Cho, W.K. (2006). Residential Concentration, Political Socialization, and Voter Turnout.
The Journal of Politics, 68 (1), 156-167. Day, J. & Holder, K. (2004). Voting and registration in the election of November 2002.
(P20-552). Washington, DC: U.S. Bureau of the Census. DeSipio, L. (1996). Making citizens or good citizens? Naturalization as a predictor of
organizational and electoral behavior among Latino immigrants. Hispanic Journal of Behavioral Sciences, 18(2), 194-213.
File, T. A. (2008). Voting and registration in the election of November 2006. (P20-557).
Washington, DC: U.S. Bureau of the Census. File, T. A. and Crissey, S. R. (2010). Voting and registration in the election of November
2008. (P20-562). Washington, DC: U.S. Bureau of the Census. Fraga, L. R. & Segura, G. (2006). Culture clash? Contesting notions of American identity
and the effects of Latin American immigration. Perspectives on Politics, 4(2), 279-287.
Voting Behavior of Naturalized Citizens 12
Garcia, J. A. (2003). Latino politics in America: Community, culture, and interests.
Lanham, MD: Rowman and Littlefield, Inc. Hero, R. (1992). Latinos and the U.S. political system: Two-tiered pluralism.
Philadelphia: Temple University Press. Holder, K. (2006). Voting and registration in the election of November 2004. (P20-
556). Washington, DC: U.S. Bureau of the Census. Jackson, R. A. and Carsey, T. M. (1999). Group components of U.S. presidential voting
across the states. Political Behavior, 21 (2), 123-151. Jacobson, G. C. (1990). The electoral origins of divided government: Competition in U.S.
house elections, 1946-1988. Boulder, CO: Westview Press. Jamieson, A., Shin, H. B., & Day, J. (2002). Voting and registration in the election of
November 2000. (P20-542). Washington, DC: U.S. Bureau of the Census. Jones-Correa, M. (2005). Language provisions under the Voting Rights Act: How
effective are they? Social Science Quarterly, 86, 549-564. Lazarfeld, P. F., Berelson, B. R. & Gaudet, H. (1948). The people’s choice. New York:
Columbia University Press. Lipset, S. M. 1980 [1960]. Political man. Baltimore: Johns Hopkins University Press. Manza, J. & Brooks, C. (1999). Social cleavages and political change. New York:
Oxford University Press. Miller, W. E. & Shanks, J. M. (1996). The new American voter. Cambridge, MA:
Harvard University Press. Putman, R.D. (2000). Bowling Alone: The Collapse and Revival of American Community.
New York: Simon and Schuster. Rytina, N. F. & Saeger, C. (2005). Naturalizations in the United States: 2004.
Washington, DC: U.S. Department of Homeland Security, Office of Immigration Statistics.
Segura, G., Barreto, M. & Woods, N. D. (2004). The mobilizing effect of majority-
minority districts on Latino turnout. American Political Science Review, 98, 65- 76.
Voting Behavior of Naturalized Citizens 13
Segura, G., Nicholson, S. P. & Pantoja, A. D. (2006). Explaining the Latino vote: Issue voting among Latinos in the 2000 presidential election. Political Research Quarterly, 59, 259-271. Xu, J. (2005). Why do minorities participate less? The effects of immigration, education,
and electoral process on Asian American voter registration and turnout. Social Science Research, 34, 682-702.
Voting Behavior of Naturalized Citizens 14
Table 1. Voter Registration and Voting, by Nativity Status and Year
(Numbers in Thousands)
Year
U.S. Citizen Population 18+ Registered Voting Total Native Naturalized Total Native Naturalized Total Native Naturalized
1996 Number 179,936 171,713 8,223 127,611 122,431 5,180 105,017 100,623 4,334 Percent 95.4 4.6 70.9 71.3 63.0 58.4 58.6 52.7
1998 Number 183,451 173,862 9,588 123,104 117,847 5,257 83,098 79,453 3,645 Percent 94.8 5.2 67.1 67.8 54.8 45.3 45.7 38.0
2000 Number 186,366 175,679 10,687 129,549 123,337 6,212 110,826 105,420 5,406 Percent 94.3 5.7 69.5 70.2 58.1 59.5 60.0 50.6
2002 Number 192,656 180,473 12,183 128,154 121,526 6,628 88,903 84,490 4,413 Percent 93.7 6.3 66.5 67.3 54.4 46.1 46.8 36.2
2004 Number 197,005 183,880 13,125 142,070 134,039 8,030 125,736 118,693 7,042 Percent 93.3 6.7 72.1 72.9 61.2 63.8 64.5 53.7
2006 Number 201,073 187,132 13,941 135,847 128,282 7,565 96,119 91,010 5,109 Percent 93.1 6.9 67.6 68.6 54.3 47.8 48.6 36.6
2008 Number 206,072 190,683 15,390 146,311 137,001 9,310 131,144 122,839 8,305 Percent 92.5 7.5 71.0 71.8 60.5 63.6 64.4 54.0
2010 Number 210,800 193,897 16,903 137,263 128,098 9,165 95,987 89,740 6,247 Percent 92.0 8.0 65.1 66.1 54.2 45.5 46.3 37.0 Source: U.S. Census Bureau, Current Population Survey, November Voting Supplements: 1996-2010. Technical documentation for CPS is available at www.census.gov/apsd/techdoc/cps/cps-main.html
Voting Behavior of Naturalized Citizens 15
Table 2: Odds Ratios from Logistic Regressions Predicting Reported Voter Registration, by Year 1996 1998 2000 2002 2004 2006 2008 2010Nativity status
Native born REF
REF
REF
REF
REF
REF REF
REF Naturalized 0.64 ** 0.53 ** 0.56 ** 0.50 ** 0.56 ** 0.48 ** 0.50 ** 0.46 ** Sex
Female 1.22 ** 1.19 ** 1.23 ** 1.23 ** 1.26 ** 1.22 ** 1.38 ** 1.21 ** Age 1.03 ** 1.03 ** 1.03 ** 1.03 ** 1.03 ** 1.03 ** 1.03 ** 1.03 ** Race
White, non-Hispanic REF
REF
REF REF REF REF REF
REF Black, non-Hispanic 1.50 ** 1.49 ** 1.65 ** 1.46 ** 1.77 ** 1.42 ** 2.17 ** 1.79 ** Hispanic 0.98 1.00
0.87 ** 0.89 ** 0.81 ** 0.86 ** 0.90 ** 0.84 **
Other, non-Hispanic 0.64 ** 0.61 ** 0.59 ** 0.66 ** 0.55 ** 0.67 ** 0.66 ** 0.66 ** Martial status
Never married REF
REF
REF REF REF REF REF
REF Married 1.20 ** 1.43 ** 1.41 ** 1.42 ** 1.19 ** 1.28 ** 1.24 ** 1.47 ** Widowed, divorced, separated 0.87 ** 0.99
0.86 ** 0.88 ** 0.84 ** 0.83 ** 0.86 ** 0.97
Education
High school or less REF
REF
REF REF REF REF REF
REF
Some college or more 2.92 ** 2.61 ** 2.86 ** 2.69 ** 3.00 ** 2.65 ** 3.17 ** 2.83 ** Employment status
Employed 1.18 ** 1.25 ** 1.19 ** 1.30 ** 1.26 ** 1.32 ** 1.29 ** 1.44 ** Not employed 1.12 + 1.23 ** 0.86 * 1.20 ** 1.09
1.06 1.08
1.37 **
Not in labor force REF
REF
REF
REF
REF
REF REF
