The Pull of the Parties for Immigrants in the United States

gaybayberryΤεχνίτη Νοημοσύνη και Ρομποτική

17 Νοε 2013 (πριν από 3 χρόνια και 10 μήνες)

133 εμφανίσεις


The Pull of the Parties for Immigrants in the United States




Katsuo Nishikawa

Department of Polit
ical Science

Trinity University

San Antonio, TX 78212

knishika@trinity.edu


James A. McCann

Department of Political Science

Purdue University

West
Lafayette, IN 47907

mccannj@purdue.edu



Stacey Connaughton

The Brian
Lamb School of

Communication

Purdue University

West Lafayette, IN 47907

sconnaug@purdue.edu






Abst
ract
:

Immigrants in the United States frequently express a desire to return to their country of birth, an attitude
that could undermin
e political

acculturation and
democratic
inclusion in the American context.
This mindset has
been labeled an “ideology of

return.”

Drawing from original surveys of the Mexican
-
born population

conducted

during the 2006 and 2008 elections, we show that partisan competition during major national campaigns
has the
potential to pull immigrants towards U.S. civic life and make th
e prospect of remaining in the country over the long
-
term a more attractive
option.


Prepared for presentation at the annual meeting of the Western Political Science Association, San
Antonio, TX, April 21
-
23, 2011.


1


Introduction
1

Approximately one
-
eighth of the current U.S. resident population
is foreign
-
born, a

nearly
unprecedented

proportion that stands in marked contrast to the 1960s and 1970s, when
immigrants numbered fewer than one in twenty (Batalova and Terrazas 2010). As t
he foreign
-
born population has grown, social scientists have devoted increasing atte
ntion to questions of
inclusion in American society and public affairs. One recurring theme that emerges in field
interviews and opinion surveys is the ambivalence that imm
igrants often feel about living in the
United States. Many of the foreign
-
born carry with them a deep sense of nostalgia and
admiration for the
ir

country of birth, work hard to keep the country’s

distinctive

traditions alive
in their households and commun
ities, and hope one day to return to it (
Arthur 2000;
Jones
-
Correa 1998
a, 1998b
;
Massey and Sánchez 2010;
Portes and Rumbaut 2006;
Wampler, Chávez
and Pedraza 2009; Wong 2006). As one author put it, it is common for immigrants to refer to the
United State
s as “
this

country,” and the country from which they came as “
my

land” or “
my

country” (Duany 1994, referenced in Jones
-
Correa 1998
a
, 91). Such a mindset,
which has been
labeled an “
ideology of return
,

could depress naturalization rates and limit
political engagement
in the U.S. (Jones
-
Correa 1998
a
; DeSipio 2006; McCann, Cornelius and Leal 2009; Pantoja
2005).

In this paper, we consider the effects of po
litical parties on this

ideology of return.


Can the
signals sent by the Democrats and Repub
licans during contemporary national election campaigns

cause immigrants to update their views of

the U.S. and
make the prospect of remaining in the
country

over the long
-
term more

desirable
? C
an partisan mobilization
influence

attitudes that
are
closely tied to residential preferences, such as general
feelings towards
the American people
and U.S. governing institutions,
beliefs regarding

personal
competence and influence

in the
2


American context, and
a sense of immigrant group consciousness

and sol
idarity
? Are
immigrants who are less

acquainted with the U
nited States



relative
newcomers, the non
-
naturalized, and those who live outside of traditional settlement areas


more responsive

to
major
party
campaign stimuli?

When examining the institutions

that pull immigrants toward
s American civic life

and
encourage them to put down roots
, scholars

give

great weight to a variety

of nonpartisan
organiza
tions


l
abor unions and worker centers, social service agencies, ethnic associations, and
churches
, amon
g others

(
e.g.,
Andersen 2008a and 2008b; Andersen and Cohen 2005; Bada,
Fox, and Selee 2006; de Graauw 2008; Hamlin 2008; Jones
-
Correa 2005;
Verba et al. 1995;
Wong 2006; Wong, Rim and Perez 2008). The two major parties figure far les
s prominently in
these accounts
. T
he

powerful

urban machines tha
t
selectively

incorporated immigrant groups
through

the distribution of goods and services
in exchange for support
during

the
first half of the
twentieth century

have
all but
disappeared
.

In this post
-
reform era, parties seek first and
foremost to serve the short
-
term interests
of office
-
seekers
. This means targeting established
constituencies during electoral campaigns,
typically through short candidate
-
centered
advertisements

and photo
-
o
pportunities
,
rather than cultivating

personal
ties
with

immigrants
,
many of whom are ineligible

to vote

or do not participate regularly for other reasons.


The
replacement of the patronage system with the merit system
,


write Scott and Hrebenar

(1984, 15
-
16; quoted in Wong 20
06, 58)
, has “
reduced the parties’ opportunities to function as socializers.

.
. .
Are any of today’s immigrants introduced to American politics and political traditions through
the medium of the Democratic or Republican par
ties?”

While the

historic

movement

away from patronage
-
based p
ar
t
y politics is beyond dispute,
a
nd for most observers would be a
sign of health in American democracy
,

w
e do not see the
3


question that Scott and Hrebenar raise as purely rhetorical. E
ven in a

political environment
where the foreign
-
born are not the primary subjects of p
artisan

mobilization,

and a good many
lack formal standing to take part,

it is still possible for immigrants to be

exposed to
much of the

fanfare of
campaign events and appeals
during
major
national elections in the United States
.

Drawing from

an original panel survey of the Mexican
-
born population over the course of the
2006 midterm elections and a randomized survey experiment conducted within th
is same
population in September

2008,
we demonstrate that
election
-
year political

messages
have

the
potential to

counteract

the “id
eology of return
.


In this regard, the
party system
of the United
States may continue

to function
, at least

after a fashion
,

as a
conduit

for immigrant inclusion.


Theoretical Background


Much is known about

the factors
that
are responsible for

the recent surge in

migration
from
the
lesser developed nations of Latin America, Asia, and Africa to the United States and
other industrialized dem
ocracies.

For many

migrants
, economic calculations
, specifically the
expected difference in wages and opportunities in the sending versus the host country,

carry the
most we
ight in the decision to relocate
. Q
uality of life
matters a great deal

as well,
with crime,
social unrest, or political violence driving many to seek refuge and stability abroad.

Transnational family conn
ections and social networks further

facilitate movement from one
country to another; it is far easier to contemplate
leaving one’s
native land if family members
and friends have previously done so

(
Castles and Miller 2003; Massey and Sánche
z 2010; Portes
and Rumbaut 2006
)
.


Substantially less research has been
conducted on what causes migrants to return to the
country of birth
once
they have settled

abroad
. E
ven less is known about the
desire
to repatriate

4


in the first place
, a
preference that
has been

liken
ed

to an ideology because
it
surfaces

often and
forcefully

in
discussions with
immigrant
s
.
The
limited
work
in this
area

stress
es

economic
considerations. Immigrants who are
dissatisfied

with their employment status

or
anxious about
long
-
term job prospects
, for example, are
more
likely

than the financially secure to
wish to
return
, even if their current living arrangements
would not permit
this

(
Waldorf 1995
).
Socio
-
p
sychological factors,
however, might also
play an important

part in shaping views on
repatriation

among the foreign
-
born
.
E
mpirical
investigations

based on
interviews with

im
migrants in the United States

point
to three

interrelated
types of attitudinal

variables that
sha
pe
residential preferences
.

The first of these is the degree of affect

felt

towards Americans. Quite
understandably, immigrants who hold the U.S.
-
born in high esteem and are more trustin
g of
Americans tend to be less inclined to want to repatria
te at some point in the future (Massey and
Sánchez 2010; Wampler, Chávez, and Pedraza 2009).


Perceptions of status and efficacy in the United States might also have an effect on long
-
term resident
ial preferences. Aguilera (2004) finds, for example, in an examination of Mexican
migrants who originally entered the U.S. without papers that self
-
described proficiency in
English is a major predictor of planning to stay in the country through retirement
.
More
generally, DeSipio (2006) reports that Latino immigrants who believe that they have more
influence in the United States vis
-
à
-
vis the country of origi
n are much more likely to wish

to
remain rather than repatriate (see also Jones
-
Correa 1998b).


In

additi
on to these beliefs,
grou
p consciousness and identities have

been found to bear
on dispositions towards staying or re
turning
. Immigrants typically identify with their native
country
; such
self
-
designation
s

may

remain salient
long after settlement i
n the

Un
ited States
.
This does not prevent the foreign
-
born, however, from developing an understanding of their
5


distinctive
social
position and
group
interests in the American context, a recognition that is an
important step in civic acculturation
.

Masse
y and Sánchez
(
2010
, Chapter 1
)

note that becoming
an “immigrant” in this sense


a
s opposed to a “sojourner” or “temporary guest”


c
ould result
,
somewhat counterintuitive
ly,

in identifying

less as an

American” even as familiarity with the
U.S.

and the desire to remain in
the country

grow.


Since most of the fo
reign
-
born living in the United States
today spent their formative
childhood years outside of the country, they may not possess the repertoire of
fundamental core

beliefs about
American po
litics and

civic
involvement

that native citizens
have

(
Abrajano and
Alvarez 2010;
Easton and Dennis 1969;
Hajnal and Lee 2011;
Krosnick 1991; Sears 1983
).

This
could make the political attitudes of immigrants less stable compared to the American
-
born, and

more resp
onsive to current events and to

signals sent
from mainstream society

and governing
institutions

(Mass
e
y and Sánchez 2010).
In the United States and other established democracies,
such signals are most pronounced during major national election campaigns. In
these periods,
parties conduct

an internal dialogue to define the
ir

positions and strategies, and then turn their
attenti
on to reaching out to the
country
. Whether a particular party is successful or not in
persuading voters to back its candidate
s, a great deal of
research on the effects of
the
c
ontemporary
electora
l process

itself

shows that
spirited conflict
and mass mobi
lization can
pay

democratic dividends within the mass public
.

As parties seek to sell their programs

and attract followers
, they

also implicitly
reinforce
a sense of political
legitimacy
,
civic
competence

and
duty
, interpersonal trust, and group
attachments



all essential ingredients in the maintenance of a democratic system

(
see, e.g.,
Anderson and Paskeviciute 2009
; Banducci and Karp 2003;

Cl
arke and Acock 1989;
Conover
1988;
Finkel 1985
;
Kam 2007;
Mansbridge 1999;
Rahn et al. 1999;
Valentino 2001; Verba et al.
6


1995). These effects are not necessarily limited to those who turn out to vote or work in a
political campaign. Rahn et al. (1999) show, for
instance
, that
Americans who are merely
“psychologically involved” in electoral poli
tics


that is, they are attentive to campaigns


become more trusting of their fellow citizens and feel more efficacious.

Expanding on this work
, electoral politics in the United States
might

likewise make long
-
term settlement a more attractive propositio
n for that segment of the public

for whom living

in
another country

is a realistic option
.
M
essages of democratic inclusion, solidarity, and
representation during a campaign could cause immigrants to see the U.S. in a more positive light,
feel more confide
nt in their ability to make a difference

in the country
, and become more aware
of
particular
group identities



a
ll of which
should lessen the desire to repatriate and lead to more
substantive democratic engagement within the U.S.

Of course, this could only happen if the
foreign
-
born are inclined
to follow public affairs and be
exposed to campaign stimuli.
Recent
large
-
scale surveys of Latin American and Asian immigrants suggest that this is generally the
case (Fraga et al. 2007;

Junn et al. 2008
; McCann, Cornelius and Leal 2009
). To illustrate, in the
2006 Latino National Survey
, over 80 percent of the foreign
-
born respondents
stated that they
watch television news “daily” or “most days,” as opposed to “once or twice a week” or
“never”;
over two
-
thirds of these immigrants were “somewhat” or “very” interested in politics
.

It appears
safe to assume that immigrants in the United States
develop some awareness of American
campaign po
litics each time a major national election takes pl
ace.
2

In the following section,

we probe the impact of electoral

outreach and
advertising more
deeply

with a focus on the Mexican
-
born population.
Mexicans
are

by far the largest group of
immigrants in the United States
; a
pproximately one
-
third of the
current foreign
-
born population
emigrated from Mexico. Given the country’s close proximity to the U.S.

and the
considerable

7


differences between the two nations
, a number of authors, notably
Samuel P.
Huntington (2004),

see the
ultimate

incorporation of Me
xican immigrants

into American civic life

as

particularly
challenging.

If the ebb and flow of partisan competi
tion in the United States has a salutary

effect
on long
-
term residential preferences

for this population
, such concerns might well become less
pressing.

As explained in greater detail below, we pay particularly close attention to potential
variations in campaign effects, since the
Mexican immigrant population is strikingly diverse with
respect to length of

time in the United States, place of residence, and naturalization status

(Massey 2008)
.

Much research on public opinion in the U.S. and elsewhere demonstrates that
attitudes across a wide array of domains (e.g., partisanship, self
-
positioning along a lef
t
-
right
ideological continuum, and political values) become more stable as individuals gain experience
in public affairs. In keeping with this literature, it could be the case that
Mexicans who have
lived in the U.S. for many years, have settled in locati
ons that are rich in Mexican
-
American
culture
and civic organizations
(e.g., Texas or southern California), or have become naturalized
American citizens, a long process that requires
a great deal of commitment,
ha
ve a relatively
well
-
developed

attitude tow
ards

staying or repatriating
. C
ampaign communication could have
far less
of an effect
for these immigrants compared to relative newcomers, noncitizens, or
Mexicans who
live outside of traditional settlement destinations and would not be exposed as
regular
ly to political stimuli.


Research Design and Findings


The potential effects of American political campaigns on immigrants’ long
-
term
residential preferences are examined through two surveys. The first
bracketed

the historic
8


midterm elections

of 2006
, which brought about a transition in the U.S. Congress from
Republican to Democratic control. The second was fielded in September of 2008,
when the
political climate was

quite

different but

equally historic
.

Study 1

In June of 2006,
telephone interviews were conducted with 753 Mexican immigrants
across three sampling sites, Dallas, San Diego, and north
-
central Indiana, including Indianapolis
but excluding the Chicago region.
R
espondents
were
recruited randomly through records
obtained

from a well established marketing research firm specializing in the Latino community.
3

Immediately after the November elections
, a
s many immigrants as possible were
contacted for
another interview

(
N

= 264).
4


Panel data such as these allow
researchers

to track changes in
attitudes

with a great deal of precision

(Bartels 2006; Finkel 1995).


The three sampling areas were selected to maximize variation in
settlement areas

(King,
Keohane and Verba 1995)
. Dallas and San Diego are major traditional destina
tions for migrants,
with a current combined Mexican
-
born population of over one million (Batalova 2008). North
-
central Indiana is typical of “new” settlement destinations for Mexicans and other immigrant
groups. Between 2000 and 2004, the number of India
na
-
based Mexicans rose by approximately
60,000. Out of all metropolitan areas in the United States, Indianapolis had the fifth
-
highest rate
of Latino population growth during this period (
Menéndez Alarcón and Novak 2010;
Sagamore
Institute for Policy Rese
arch 2006). While the number of Mexicans now living in Indiana is
much smaller than in California or Texas, this rapid expansion of immigrant communities was
unprecedented in the Hoosier State
.


The item used to gauge long
-
term residential preferences
was

comparable to what other
scholars have employed (
e.g.,
DeSipio 2006;
Wampler, Chávez, and Pedraza 2009
; Waldorf
9


1995):

Do you want to return to Mexico permanently to live one day, or are
you planning to
remain in the United States
?”

Table 1 reports the

frequency of responses in each of the survey
waves, with study participants in the three regions pooled.
5

In June, two out of three immigrants
stated a preference for returning
. This percentage varied somewhat across sampling regions,
with Mexicans in I
ndiana being more inclined to repatriate. In each location, however, a solid
majority
wished to return.

[Table 1 about here]

When interviewed five months later, the thought of
remaining in the U.S. had become

more attractive. In this instance, the pooled sample was nearly divided in half, with only 56
percent now wishing to repatriate.
As noted at the bottom of this table, this increase is not likely
to have come about by chance; a paired
-
sample t
-
test is hi
ghly
statistically
significant.
At the
individual
-
level
, responses to this item changed a fair amount
. The test
-
retest correlation is
moderately
-
sized at .35,

underscoring

that for Mexican immigrants, there is

a measure of
uncertainty or

flexibility in l
ong
-
term
residential

preferences.

The overall gain in desire to
remain
suggests that
the signals and outreach of contemporary
electoral
campaigns

ha
ve

the
potential to pull immigrants


many of whom may see themselves as living “in between” two
countries (Jones
-
Correa 1998a)


mo
re

closely towards the United States.



Immigrants with less experience living in the U.S., those who have not become America
n
citizens, or those who have settled in regions nontraditional destinations such as Indiana may
have been particularly sensitive to campaign effects. The regression analysis in the lower part of
Table 1 examines
these differences
. In this model, residen
tial preferences as measured in the
post
-
election wave (dummy
-
coded, with 1 =

remain permanently

)

are regressed on preferences
from the June wave plus sampling region, naturalization status (1 = citizen), and years spent in
10


the United States.
6


By inclu
ding the lagged dependent variable as a predictor, we are modeling
changes

in settlement preferences (Finkel 1995).
The findings from this analysis indi
cate that the
rise in desire

to remain in the country did not vary with respect to sampling location, length of
time in the U.S., and formal civic status. If the dynamics of the midterm elections were indeed
responsible for making the United States a more attractive option for long
-
term settlement, even
immigrants with a great deal of prior exposure to American politics and culture were affected

at
the same rate
.

Other explanations for this change in attitude are certainly possible, however. One

plausible

rival interpretation is tha
t the

turn towards

wishing

to live permanently
in

the United
States between June and November reflects a

particular type of

transnational
de
mobilization
effect in the Mexican context. The data collection for the first survey wave took place on the eve
of
the July 2 presidential election in Mexico.
Even though Mexican political parties are legally
barred from canvassing voters outside of the country and the costs of casting an international
absentee ballot in this election were inordinately high, m
any of t
he immigrants in the study
closely followed this election and were aware of the main themes of th
e campaigns
(McCann
,
Cornelius, and Leal 2009). L
arge number
s

engaged i
n informal participation

in this context
,
such as

encouraging friends and famil
y in
Mexico to turn out to vote
.
By the time of the
November survey wave, mass electoral mobilization

within Mexico
had long since
ceased.
7

This
dynamic, more than the fanfare
and pull
of American electoral politics, could have tilted
immigrants away from Mex
ico.

To delve more deeply into
the effects of U.S. campaigns within this population, and to
assess whether changes in residential preferences coincide with changes in
attitudes in
the three
closely related domains

discussed above


general feelings towards Americans, beliefs about
11


one’s ability to succeed in the United States, and a sense of group consciousness
--

w
e conducted
a
follow
-
up
study
during

the
American presidential
election

two years later.


Study 2.

8


As in 2006, randomly selected
Mexican immigrants were interviewed

by telephone
.

N
early all surveys
again
were in Spanish.
In total, 1,023 respondents took part in
the study
,
which commenced after the Labor Day weekend, the traditional start of fall pres
idential
campaigns.

Sampling this time took place in two regions rather than three: north
-
central Indiana,
again excluding the Chicago
area, and San Antonio, TX. The

reduction in survey sites raised the
number of respondents in each area (
N

= 501 in Indiana and 522 in San Antonio), which serves
to increase the statistical power for regional comparisons.

San Antonio was chosen in place of
Dallas because its Mexican population is larger and somewhat better established
.
9


T
he city has
long bee
n
recognized

as a leading center of Mexican
-
American culture
. For this reason,
candidates wishing to show their commitment to the Mexican community frequently campaign in
San Antonio.
10

Th
ese features offer

a more advantageous contrast w
ith the emerging

Mexican
immigrant population of Indiana.


Researchers investigating the impact of campaign stimuli on political attitudes and
perceptions
via conventional

survey methods frequently wrestle with thorny issues concerning
measurement error and causal inferenc
e.
Respondents may be unable to state how many
commercials they have seen or heard (Ansolabehere, Iyengar and Simon 1999). Furthermore,
even if they can accurately report exposure, evaluations of the candidates or identification with a
party may prompt i
ndividuals to watch particular television programs or listen to radio stations
where exposure to political information is likely to occur.
In the

2006

panel stud
y
,
long
-
term
residential preferences for immigrants
were

seemingly

prone to updating, and

part
isan
12


mobilization signals

c
ould

have
exert
ed

strong effects on these attitudes.

Here we put a finer
point on this claim.


A few minutes in
to the interview
s
, survey participants were randomly divided into three
groups. Approximately half were chosen to
listen either to an ad for Barack Obama (
N

= 262,
Treatment Group 1), or an ad for John McCain (
N

= 250, Treatment Group 2). These were
actual ads produced for Spanish
-
language radio stations. The third group (
N

= 511) served as the
control. For the treat
ment groups, exposure was prefaced by the remark, “Recently political
parties in the U.S. began their presidential campaigns. I am going to play a brief ad from Senator
[Obama / McCain]. You may have already heard it on the radio. After listening to it,

I will have
a couple questions for you about the ad.” Following exposure, respondents were queried on
whether they remembered hearing such a commercial before, whether they believe
advertisements influence voters in general, and whether the ad gave a posi
tive or negative
impression of the candidate. The lead
-
in to the sound clip and questions afterwards were
intended to supply a plausible rationale for the treatments.
11

Only a very small number of subjects in each treatment group stated that the advertisem
ent
left a bad impression. Overwhelmingly, the ads were seen as showing the candidate in a positive
light. This was our intention. In selecting political spots for the treatments, the goal was to
expose respondents to that most ubiquitous form of campai
gn communication in the Latino
media market: the upbeat personal endorsement from someone who is a part of that ethnic
community.
12

Wordings for the Obama and McCain ads are given in Table 2. Each was played
to the randomly selected respondents in its ent
irety.
13


[Table 2 about here]

13


If short candidate
-
centered partisan messages implicitly draw immigrants close
r

to American
society over the course of a campaign cycle, we should see evidence of this effect in
attitudes
that were gauged several minutes
after

subjects
in the treatment groups
listened

to an ad from
Obama or McCain.

Using the same
instrumentation

as in the 2006 study,
all
immigrants stated
towards the end of the interview
whether they wish to return to Mexico one day or remain in the
U.S. for the rest of their lives.
S
ubjects also
offer
ed

summary
affect
ive
ratings of

the
American
people, U.S.
governing institutions,

and Mexicans
;

judgments concerning their

personal
compete
nce and e
fficacy in the American context;

and
responses to several items
pertaining

to

immigrant group consciousness.


All of these attitudes are elements in an “ideology of return.”
If the residential preferences that study participants express are more

than top
-
of
-
the
-
head

responses


that is, if they are indicative of a coherent mindset


responses across these multiple
domains should move in sync following exposure to campaign stimuli. T
he following wordings
were used

to capture the different attitude

elements
:



Affect
ive
Ratings
.
On a scale of 0 to 10, where 0 means that your opinion is very bad and
10 means that it is very good, where would you place
Americans
?
T
he U.S. government
in Washington, DC
?
Democrats
?
Republicans
?
Mexicans
?




Personal
Competence
.

o

Agree or disagree:
American politics is so complicated that I
sometimes
do not
understand it well.

o

Do you feel that your English is good, so
-
so, or not so good?




Immigrant
Group Consciousness
.

o

Have you been badly treated in the U.S. because
of being an immigrant? If so,
where was this? Finding employment? Arranging housing? Seeking medical
attention? Shopping? Just going about personal business in the street?


o

Which of these words would you use to describe yourself?
American
?
Mexican
-
Americ
an
?
Hispanic
?
Latino
?

o

How well do the two parties represent the interests of Latinos? Average
American voters? Large corporations? Immigrants?


14


The
zero to ten
-
point scale for affective ratings

is comparable to those that have been
used in public opinion surveys in Mexico (e.g.,

Lawson et al. 2007
).
14

The first item to tap into
personal competence is adapted from the American National Election Study series; it is a
decades
-
old measure of persona
l political efficacy.

The second would be

relevant only within
immigrant populations. Language skills are a key marker of social incorporation in the United
States and an important determinant of status among the foreign
-
born.
Appraisal of language
abil
ity, however, is inherently inexact. Skills can fluctuate over time and depend on one’s
general level of confidence in a given context. If political campaigns can instill a sense of
accomplishment and empowerment (
Cl
arke and Acock 1989; Finkel 1985
; Mans
bridge 1999
),
these effects for this population could extend beyond political efficacy to judgments about
nonpolitical skills.

The concept of group consciousness appears frequently in the scholarly literature on
African
-
American participation and has been
fruitfully
applied to a wide range of ethnic groups

(Sanchez 2006)
. The three

sets of items

in this category are derived from this
expansive
literature: interpreting the treatment one receives in the course of daily activities as a
consequence of membership in a specific social category,
namely
“immigrant”;
reluctance to
identify oneself as “American,” but
instead
to prefer

labels
that are
tied

to ethnicity; and
evaluating the major parties based on how well they represent immigrants, as opposed to Latinos,
average American voters, and business corporations.

Table 3

shows the impact of exposure to one of the campaign ads on long
-
term
residential preferences.

Two effects are apparent; listening to either advertisement reduced
uncertainty about living in the United States versus returning to Mexico, and increased the
attractiveness of remaining in the U.S. In the control group, 47 percent stated a preference for
15


remaining. This figure rises eight points for immigrants in the “Obama ad” group and 13 for
those exposed to the McCain commercial. This change is highly si
gnificant (
p

< .01) and
comports w
ith the trend from 2006. The kinds of mobilization messages that are aired during
American campaigns can orient Mexican immigrants towards life in the United States.

There is
little difference in effect between the two a
ds. Outreach for the Republicans counted as much as
messages from the Democrats.
15

[Table 3 about here]

The campaign ads further affected summary evaluations of the two major parties,
Americans, and the U.S. government.
In the case of the parties, a
distinction emerges. Listening
to the Obama advertisement significantly improved assessments

of the Democrats, even though
the party was not specifically mentioned in the commercial. The McCain ad did not have a
noteworthy impact on ratings of the Democr
ats, but it did cause immigrants to give Republicans
somewhat more

positive ratings relative to both the “Obama ad” and the control groups (
p
=
.088).

Respondents in the two

treatment groups, however, evaluated both “Americans” and the
“U.S. government” m
ore highly compared to those who had not heard an ad (
p

< .05 in each
case).

On the other hand, exposure to an ad had no appreciable impact on affect towards
“Mexicans” as a group.

These shifts in opinions further underscore the potential for campaign communication to
contribute to the social incorporation of immigrants, even
if the
goal of the candidates is
first and
foremost
to move
an

audience towards one party in particular.

Cha
nges in the two efficacy items
are presented in the

lower part of Table 3. In the

case of political efficacy, the ad treatments did
not significantly alter appraisals. For respondents in all experimental groups, the conse
nsus was
that American politics

c
an sometimes be too complicated to understand. Such a breakdown in
16


responses

typically surfaces

in ANES surveys of the mass public.
Exposure to the McCain
advertisement boosted political efficacy to some extent, but compared to the control group, this
is

not a significant change.

On the other hand, there i
s some evidence that
after hearing one of
the campaign commercials, immigrants felt mo
re confident in their English language skills, a key
indicator of perceived status in the United States (
p

= .088).
As with the zero to ten
-
point
evaluation scales and the question of whether to remain in the country, the McCain and
Obama
ads had comparable effects
.

The impacts

of the commercials on the multiple indicators of immigrant group
consciousness are

presented in Table 4. Perceptions of group
-
based mistreatment were tallied
and coded into three categories (no discrimination, fewer than three types of mistreatment, or
more than five types).


The pattern that emerges

after breaking down this tally by e
xperimental
treatments

demonstrates that political outreach to the Latino community can cause the foreign
-
born to interpret social interactions in a way that reinforces identification as an
immigrant
. At
the same time, the campaign messages significantly
decreased the odds of identifying as an
“American.” In the control group, 36 percent designated themselves as such, as opposed to 28
percent in both the Obama and McCain treatment groups (
p

< .01).

No similar

reductions
emerged for other labels (Mexican
-
American, Latino, and Hispanic). The results for these latter
terms imply that
ethnic group attachments are more strongly felt among the foreign
-
born and
are
less bound to particular contexts, a point that has

been raised in previous work on immigrant
identities (e.g., Jones
-
Correa 1998a).

[Table 4 about here]

The regression analyses in the lower half of the table speak to another facet

of campaign
effects
, the priming of group
-
based considerations when
express
ing

summary evaluations of the
17


two major parties

(
Iyengar and Kinder 1987)
. As shown

in Table 3, study participants who
listened to an advertisement updated their
general
impressions of the parties to a significant
degree, with those in the “Obama” treatm
ent group liking the Democrats more and those in the
“McCain” group rating Republicans more highly.

These evaluations were differenced (i.e., the
rating for Republicans was subtracted from that for the Democrats) and re
gressed on four items
that touched o
n

group representation within the party system: whether the Democratic Party,
relative to the Republican Party, represents the interests of “Latinos,” “average American
voters,” “business corporations,” and “immigrants in the United States.”

Each judgment

was
measured on a three
-
point scale (
-
1 = only the Republican Party represents the particular group,
+1 = only the Democratic Party represents the group, 0 = all other responses).


In the “no ad” control group, the most relevant constituency groups for r
espondents when
evaluating the parties were

Latinos


and

average American voters.


Both regression slopes are
highly significant (
p

< .01) and run in the expected direction
: perceptions that the Democratic
Party is more representative than the Republica
n Party are linked to higher general evaluations of
the Democrats relative to the Republicans. The impact of perceptions
of “immigrant”
representation

is also positive, an
d the coefficient for


business corporations


has a negative
sign. These signs

woul
d have been e
xpected, but neither slope

is large or statistically significant.

Turning to the two treatment groups, a much different pattern appears. Whether subjects listened
to an ad for Obama or McCain,
perceptions of how well immigrants were
represented took on
much greater weight and became the largest coefficient out of the set.
16


Taken together, the findings in Tables 3 and 4 provide a framework for interpreting the
change in long
-
term settlement pre
ferences
over the course of the 2006 camp
aign. Conventional
political outreach

during elections

can shape basic orientations towards U.S. society and sharpen
18


group attachments that are r
elevant in the American context, with the ultimate effect of making
immigrants feel more at home in the countr
y.
For each of the items in these two tables,
subsequent models were estimated to exam
ine variations in causal outcomes
: advertising
treatments by sampling region, by length of stay in the United States, or by naturalization status.

In no case did signif
icant interactions appear.
17

As in the panel su
rvey two years earlier
, it
appears that Mexican immigrants are sensitive to the cues and signals from partisan elites even if
they have resided in the country for many years, live in an

environment that is rich in bi
cultural
civic organizations and Latino leaders, or have become American citizens.


Conclusion


For the Mexican immigrants we studied, feelings towards the United States, beliefs about
one’s ability to navigate within mainstr
eam American society,
the

recognition
that one is

part of
a particular group within the larger public, and
a desire

to remain in the country far into the
future are all
interlinked. These are the principal elements in an ideology

of settlement or
repatriation.
As the five
-
month panel study of 2006 suggests and the survey experiment of 2
008
more thoroughly documents
, this ideology is flexible to a degree. To apply a phrase from
Converse’s classic (1964)

piece on mass belief syste
ms, immigrants respond to campaign stimuli
in a manner that typifies

dynamic
ideological c
onstraint

;

changes

in one attitude domain are
matched by changes in other related areas. Mexicans in the U.S. possess rather coherent
orientations towards settleme
nt and incorporation, but these views are hardly
static
.

As they
attempt to draw as much of the public as possible into the partisan fold, Democratic and
Republican campaigners can pull

immigrants

closer towards American society.


19


The fact that this move
ment occurred without simultaneously causing the Mexican
-
born
to lower their
generally positive
evaluations of “Mexicans” to a significant degree could imply
that esteem for
one

nation need not come at the expense of feelings towards the others.

Many of
the major migrant
-
sending countries in the developing world, including Mexico, go to great
lengths to maintain ties to emigrant populations in the United States and the other industrialized
democracies. Whether these efforts undermine or facilitate civic
incorporation in the receiving
country


or whether they simply have no effect at all


is an important question worthy of
further research.


Also worthy of future work is whether the effects of campaigns vary depending on the
communication choices made by

candidates. The lesson from the 2006 study is that campaigns
in the aggregate
could

to an extent orient the foreign
-
born towards life in the United States. The
2008 experiment
probes this effect more deeply
, showing the Democratic and Republican
message
s
can be

equally consequential.
The
two
ad
vertisements

used as experimental treatments
were selected because they exemplify the parties’ traditional approach to mobilizing ethnic
constituencies: sell the candidate in highly personalized terms, avoiding exp
licit discussion of
partisanship or polarizing ideological disagreements;
keep the tone upbeat; and rely heavily on
endorsements from members of the ethnic community.
Such styles of communication will surely
reappear in future election cycles. Of course,

other political outreach strategies are employed as
well, in particular various forms of negative campaigning.
Exposure to advertisements that are
meant to inspire fear or anger
,
presumably directed against the opposing party
,

could have the
unintended effect of tilting immigrants away from U.S. civic life altogether.


Finally, the consistency of campaign effects across the full range of the Mexican
immigrant population


U.S. citizens and noncitizens, newcomers and immigrants who have
20


lived in the country for many years, and Mexicans based in San Antonio, Dallas, San Di
ego, or
Indiana


implies that the Mexican
-
born may never feel truly settled in the United States nor
fully
committed to repatriating.
This ambivalence may be an inescapable feature of migrant life

across
all
nationalities
.
As o
ur findings
make clear
,

th
e two major political parties

play an
important part

in immigrants’ assessments amid the
se

uncertainties.



21


References


Abrajano, Marisa A. 2010.
Campaigning to the New American Electorate: Advertising to
Latino Voters
. Palo Alto, CA: Stanford
University Press.


Abrajano, Marisa A. and R. Michael Alvarez. 2010.
New Faces, New Voices: The Hispanic
Electorate in America
. Princeton, NJ: Princeton University Press.


Aguilera, Michael B. 2004. “Deciding Where to Retire: Retirement Location Choices of

Formerly Undocumented Mexican Migrants
,” Social Science Quarterly

85 (2): 340
-
360.


Andersen, Kristi. 2008a. “In Whose Interest? Political Parties, Context, and the Incorporation of
Immigrants,” in Jane Junn and Kerry L. Haynie, eds.
New Race Politics in
America:
Understanding Minority and Immigrant Politics
. New York: Cambridge University
Press.


Andersen, Kristi. 2008b. “Parties, Organizations, and Political Incorporation: Immigrants in Six
U.S. Cities,” in S. Karthick Ramakrishnan and Irene Bloemraad, e
ds.
Civic Hopes and
Political Realities: Immigrants, Community Organizations, and Political Engagement
.
New York: Russell Sage Foundation.


Andersen, Kristi and Elizabeth F. Cohen. 2005. “Political Institutions and Incorporation of
Immigrants,” in Christin
a Wolbrecht and Rodney E. Hero, eds.
The Politics of
Democratic Inclusion
. Philadelphia: Temple University Press


Anderson, Christopher, and Aida P
askeviciute. 2009. “Partisan Legitimacy: Parties,
Partisanship, and Support for the Political System,” paper

prepared at the conference on
Citizen Preferences, Political Institutions, and Democratic Performance, Center for the
Study of Democratic Performance, Florida State University, Tallahassee, FL, Feb. 27
-
28,
2009.


Ansolabehere, Stephen, Shanto Iyengar, and

Adam Simon. 1999. “Replicating Experiments
Using Aggregate and Survey Data: The Case of Negative Advertising and Turnout.”
American Political Science Review

93 (December): 901
-
9.


Arthur
, John.

2000
.
Invisible Sojourners: African Immigrant Diaspora in
the United States
.
Westport, CT: Praeger.


Bada, Xóchitl, Jonathan Fox, and Andrew Selee, eds. 2006.
Invisible No More: Mexican Migrant
Civic Participation in the United States
. Washington, DC: Woodrow Wilson
International Center for Scholars Mexico
Institute.


Banducci, Susan A. and Jeffrey A. Karp. 2003. “How Elections Change the Way Citizens View
the Political System: Campaigns, Media Effects and Electoral Outcomes in Comparative
Perspective.”
British Journal of Political Science

33 (1): 1
75
-
99.


22


Bartels, Larry

M. 2006. “Three Virtues of Panel Data for the Analysis of Campaign Effects,”
in
Henry Brady and Richard Johnston, ed.
Capturing Campaign Effects
. Ann Arbor:
University of Michigan Press.


Batalova
, Jeanne. 2008. “Mexican Immigrants in the United States,” Migration Policy Institute,
available at
http://www.migrationinformation.org
.


Batalova Jeanne, and Aaron Terrazas. 2010. “
Frequently Requested S
tatistics on Immigrants and
Immigration in the United States
” Migration Policy Institute, available at
http://www.migrationinformation.org
.


Camp, Roderic Al, ed. 2003. “Forum, Three Views of Mexican Demo
cracy,”
Mexican Studies
,
19:1, 1
-
106, special issue.


Castles
, Stephen

and
Mark
Miller
.

2003
.
The Age of Migration
. New York: Guilford Press.


Clarke
, Harold D.

and
Alan C.
Acock
.

1989
. “National Elections and Political Attitudes: The
Case of Political Efficacy,”
British Journal of Political Science.

19: 551
-
62.


Conover
, Pamela J. 1988. “The Role of Social Groups in Political Thinking,” British Journal of
Political Science.

18:1, 51
-
76.


Converse, Philip E. 1964. “The Nature of Belief Systems in Mass Publics.” In
Ideology and
Discontent
, ed. David E. Apter. New York: Free Press, 206
-
61.


de Graauw, Els. 2008. “Nonprofit Organizations: Agents of Immigrant Political Incor
poration in
Urban America,” in S. Karthick Ramakrishnan and Irene Bloemraad, eds.
Civic Hopes
and Political Realities: Immigrants, Community Organizations, and Political
Engagement.
New York: Russell Sage Foundation.


DeSipio, Louis. 2006. “Do Home
-
Country

Political Ties Limit Latino Immigrant Pursuit of U.S.
Civic Engagement and Citizenship?” In Taeku Lee, S. Karthick Ramakrishnan, and
Ricardo Ramírez, eds.
Transforming Politics, Transforming America: The Political and
Civic Incorporation of Immigrants in
the United States
, pp. 106
-
126. Charlottesville, VA:
University of Virginia Press.


Dominguez, Jorge I., Chappell Lawson, and Alejandro Moreno, eds. 2009.
Consolidating
Mexico’s Democracy
. Baltimore: Johns Hopkins University Press.


Duany, J. 1994.

Quisqu
eya on the Hudson: The Transna
tional Identity of Dominicans in
Washington Heights
,


The CUNY Dominican Studies Institute, Dominican

Research
Monographs.


Easton, David, and Jack Dennis.


1969.


Children in the Political System: Origins of Political
Legitimacy.


Chicago: University of Chicago Press.


23


Finkel, Steven E. 1985. “Reciprocal Effects of Participation and Political Efficacy
.” American
Journal of Political Science.

29:4, 891
-
913.


Finkel, Steven E.


1995.


Causal Analysis with Panel Data.


Sag
e University Paper Series on
Quantitative Applications in the Social Sciences.


Beverley Hills: Sage Publications.



Fraga, Luis R., John Garcia, Rodney Hero, Michael Jones
-
Correa, Valerie Martinez
-
Ebers, and
Gary Segura. 2007.
Latino National Survey
(LNS), 2006
.

ICPSR Study No. 20862.

Hajnal
Zoltan
and
Taeku
Lee
.

2011
.
Why Americans Don’t Join the Party
. Princeton, NJ:
Princeton University Press.


Hamlin, Rebecca. 2008. “Immigrants at Work: Labor Unions and Noncitizen Members,” in S.
Karthick
Ramakrishnan and Irene Bloemraad, eds.
Civic Hopes and Political Realities:
Immigrants, Community Organizations, and Political Engagement.

New York: Russell
Sage Foundation.


Huntington, Samuel P. 2004.
Who Are We? The Challenges to America’s National Iden
tity
. New
York: Simon and Schuster.


Iyengar, Shanto and Donald R. Kinder. 1987.
News that Matters
. Chicago: University of Chicago
Press.


Jones
-
Correa, Michael, 1998a.
Between T
wo Nations: the Political Predicament of Latinos in
New York City.

Ithaca:
Cornel University Press.


Jones
-
Correa, Michael, 1998b. “Different Paths: Immigration, Gender, and Political
Participation
.

International Migration Review
.

32:2, 326
-
49.


Jones
-
Correa, Michael. 2005. “Bringing Outsiders In: Questions of Immigrant Incorpo
ration,” in
Christina Wolbrecht and Rodney E. Hero, eds.
The Politics of Democratic Inclusion
.
Philadelphia: Temple University Press.


Junn, Jane, Karthick Ramakrishnan, Taeku Lee, and Janelle Wong.

2008
.
National Asian
American Survey
,
http://www.naasurvey.com/assets/NAAS
-
nyc
-
presentation.pdf

(last
accessed 4/10/11).


Kam, Cndy D. 2007. “When Duty Calls, Do Citizens Answer?”
Journal of Politics
. 69:1 (Feb),
17
-
29.


King, Gary, Robert Keohane and Sidney Verba. 1994.
Designing Social Inquiry.

Princeton, NJ:
Princeton University Press.


King, Gary, James Honaker, Anne Joseph, and Kenneth Scheve.
"
Analyzing
Incomplete Political
Science Data: An Alternative Algorithm for Multiple Imputation
.
"

American Political
Science Review
. 95:1 (March),

49
-
69

24



Krosnick, Jon. 1991. “The Stability of Political Preferences: Comparisons of Symbolic and
Nonsymbolic Attitudes,”
American Journal of Political Science
. 35:3 (August) 547
-
76.


Lawson
, Chappell, et al. 2007.
Mexico 2006 Panel Study
.
http://web.mit.edu/polisci
.


Long
, J. Scott.

1997
.
Regression Models for Categorical and
Limited Dependent Variables
.
Beverly Hills: Sage Publications.


Mansbridge, Jane.


1999. “On the Idea that Participation Makes Better Citizens,” in S. Elkin and
K. Soltan, ed.
Citizen Competence and Democratic Institutions. University Park, PA:
Pennsylvani
a State University Press.


Massey, Douglas S., ed. 2008.
New Faces in New Places: The Changing Geography of American
Immigration
. New York: Russell Sage Foundation.


Massey, Douglas S., and Magaly Sánchez R.

2010.

Brokered Boundaries: Creating Immigrant
Identity in Anti
-
Immigrant Times
.

New York: Russell Sage Foundation.


McCann, James A., Wayne Cornelius, and David Leal. 2009. “Absentee Voting and
Transnational Civic Engagement among Mexican Expatriates,” in Jorge I. Dominguez,
Chappell Lawson, and Alej
andro Moreno, eds.,
Consolidating Mexico’s Democracy: The
2006 Presidential Campaign in Comparative Perspective

(Baltimore: Johns Hopkins
University Press)


Mendez

Alarcon
, Antonio

and
Katherine
Novak
.

2010
. “Latin American Immigrants in
Indianapolis.”
La
tino Studies
. 8:1, 93
-
120.


Moreno, Alejandro. 2005.
Nuestros Valores
. Mexico, DF: Banamex.


Pantoja, Adrian. 2005.

Transnational Ties and Immigrant

Political Incorporation:

The Case of
Dominicans

in Washington Heights, New York
.”
International Migration

43:4, 123
-
146.


Pew Hispanic Center. 2006. “The 2006 National Survey of Latinos,” survey data and study
documentation available at:
http://pewhispanic.org/datasets/
.


Popkin, Samuel L. 1991.
The Reasoning Vo
ter
. Chicago: University of Chicago Press.


Portes, Alejandro and Rubén G. Rumbaut. 2006.
Immigrant America: A Portrait, Third Edition
.
Berkeley: University of California Press.


Rahn, Wendy M., John Brehm, and Neil Carlson.

1999
. “National Elections as In
stitutions for
Generating Social Capital,” in T. Skocpol and M. Fiorina, ed.
Civic Engagement in
American Democracy
. Washington, DC: Brookings Institution Press.


Sagamore Institute for Policy Research. 2006. “Connecting Mexico and the Hoosier Heartland: P
olicy
Briefing,” available at
www.sipr.org
.

25



Sanchez, Gabriel R. 2006. “The Role of Group Consciousness in Political Participation Among
Latinos in the United States.”
American Politics Research

34 (4): 427
-
51.


Scott, R. K. and R. Hrebenar. 1984.
Parties in Crisis
. New York: Wiley.


Sears
, David

O.

1983
. “Persistence of Early Political Predispositions.” In Ladd Wheeler, ed.,
Review of Personality and Social Psychology
. Beverly Hills: Sage Publications.


Subervi
-
V
élez, Federico A. 2008. “What is Known?” in Federico A. Subervi
-
Vélez, ed.
The
Mass Media and Latino Politics
. New York and London: Routledge.


Valentino, Nicholas A. 2001. “Group Priming in American Elections.” In
R. Hart and D. Shaw,
ed.,
Communication
and U.S.

Elections: New Agendas
.
Lanham, MD: Rowman and

Littlefield Publishers.


Verba Sidney, Kay Lehman Schlozman, Henry E. Brady. 1995.
Voice and Equality: Civic
Voluntarism in American Politics
. Cambridge, MA: Harvard Univ. Press.


Waldorf, Brigitte.
1995
.

“Determinants of International Return Migration Intentions.”
Professional Geographer

47(2), 125
-
136.


Wampler, Brian, Maria Chavez, and Francisco I. Pedraza. 2009. “Should I Stay or Should I Go?
Explaining Why Most Mexican Immigrants are Choosing
to Remain Permanently in the
United States.”
Latino Studies

7: 83
-
104.


Wong, Janelle. S. 2006.
Democracy’s Promise: Immigrants and American Civic Institutions.

Ann Arbor: University of Michigan Press.


Wong, Janelle, Kathy Rim, and Haven Perez. 2008. “Pr
otestant Churches and Conservative
Politics: Latinos and Asians in the United States,” in S. Karthick Ramakrishnan and Irene
Bloemraad, eds.
Civic Hopes and Political Realities: Immigrants, Community
Organizations, and Political Engagement.

New York: Russe
ll Sage Foundation.







26



Table 1.
Changes in Long
-
Term Residential

Prefe
rences: June to

November, 2006









June




November




Stay in the U.S.




34 %




44 %

Return to Mexico



66 %




56 %



**********************************************


Logistic r
eg
ression model of preferences,
November









b (se)


p
-
value



Preferences, June




.86 (.23)


< .01


U.S. citizen





.03 (.41)


.949

Years spent in the U.S.



-
.01 (.01)


.318

Dallas
resident





.22 (.29)


.454

Indiana resident




-
.32 (.29)


.280

Constant




1.48 (.29)


<.01



Note: Survey respondents were Mexican immigrants living in Dallas, San Diego, or north
-
central Indiana,
interviewed by telephone.
N

= 753 in the June wave and
264 in Novembe
r
.
To retain all cases, missing values were
imputed via
Amelia

(King et al. 2001). The increase in desire to remain in the U.S. from June to November is highly
significant based on a paired
-
sample t
-
test (
p

< .01
,
test
-
retest
r

=.35
). In t
he logistic regression model, the
dependent variable is coded 1=Stay in the U.S., 0 = Return to Mexico.






27


Table 2
. Experimental Treatments
: Exposure to a Sixty
-
Second Presidential Campaign Radio Ad


Treatment Group 1: Obama Ad Exposure (N = 262)

[Barack Obama speaks in English, with dramatic orchestral music playing in the background.]


There is not a liberal America and a conservative America. There is the
United States of
America
!


[Crowds cheer.]

[A youthful
-
sounding man then speaks Spanish wi
th a middle class Mexican accent, as dramatic music
continues to play in the background.]

“Barack Obama is talking to me. He has confronted a lot of the same difficulties my family has
faced. His parents weren’t rich. But working hard, Obama won a scholar
ship and made his own
path. After graduating from Harvard Law School, instead of taking a job which offered more
money, Obama chose to work with churches, helping those who are the least protected in his
community.”

[Barack Obama speaks in English, with dr
amatic music still playing in the background]

“In this election, in this moment let us reach for what we know is possible”

[The voice of the young man speaking Spanish resumes.]

“Obama is talking to me, to give us the opportunity of a university educatio
n, and to assure me
that my parents and grandparents have healthcare. That’s why I’m spreading the word, to my
parents, to my uncles, and friends. Because, politics isn’t just for those who like to argue. It’s for
those who want to construct a better futur
e. Obama is talking to me, and he is talking to you too.”

[Obama speaks in Spanish with a strong American accent.]

“I’m Barack Obama, and I approve this message.”

[The voice of a young man speaking Spanish resumes.]

“Paid for by Obama for America”



28


Tab
le 2, continued.


Treatment Group 2: McCain Ad Exposure (N=250)

[An older male speaks Spanish with a slight American accent as soothing acoustic guitar music plays in
the background. After thirty seconds, light keyboard music is added.]


My name is Frank
Gamboa, a proud Latino who wants the best for our country and our
community. My roommate when I was in the Naval Academy wants to be president, and he also
wants what is best for Hispanics. His name is John McCain, and he has always been with us even
in th
e hardest of times. This is because he shares our conservative values and faith in God. He
knows that family is the most important thing we have, and that we value working hard. As
someone from Arizona, John understands this and has won the respect of Lati
nos and has a long
history of supporting us. In this election, it seems the other candidate has just discovered the
importance of the Hispanic vote. That’s why when it comes to our values and to understanding
us, I know that for John this is not politics.
It comes from his heart, and that’s why I’m going to
vote for John McCain.



[John McCain speaks in English]

“I’m John McCain and I approve this message”

[Announcer speaks in English]

“Paid for by John McCain 2008”


Control, Group 3
: No Ad Exposure
(N=511)



Note: Survey participants were randomly assigned to one of the groups. Those chosen for one of the treatment
categories were exposed to the ad a few minutes into the interview.



29


Table 3
.

Settlement Preferences, Group Evaluations, and Personal Efficacy after
Exposure to a
Campaign Advertisement:
Randomized Telephone Survey Experiment in
September, 2008





Obama Ad

McCain Ad

No Ad


Significance

Residential

Preferences (%)


Stay in the U.S.




55



60



47



Return to Mexico someday


34



31



35

Undecided




11



8



17


χ
2
=17.9,
p

< .01



Evaluations on a 0
-
10 Scale,

Means and Standard Errors



Democrats



7.66 (.15)

7.06 (.16)

6.96 (.13)

F
=5.94,
p
=.003


Republicans



5.51 (.19)

5.95 (.19)

5.55 (.19)

F
=2.44,
p
=.088

Americans



7.84 (.13)

7.59 (.14)

7.39 (.11)

F
=3.23,
p
=.040


U.S. Government


6.69 (.17)

6.82 (.18)

6.26 (.09)

F
=3.72,
p
=.025


Mexicans



8.31 (.13)

7.93 (.13)

8.14 (.11)

F
=1.83,
p
=.161


Politic
al Efficacy: Can U.S. Politics

Sometimes Be Too Complicated?

(%)



Yes





61



56



61




No





31



38



32




Unsure





8



6



7


χ
2
=4.4,
p

= .361



Linguistic Efficacy
: Believes English Skills

Are
Good

(3),
So
-
So

(2), or
Not Good

(1)



Means and Standard Errors

2.05 (.04)

2.02 (.04)

1.94 (.03)


F
=2.44,
p
=.088



Note: R
espondents were Mexican immigrants living in
San Antonio, Texas (
N
=522), or

north
-
central Indiana

(
N
=501), all
interviewed by telephone.

For those

selected to listen to an advertisement, exposure occurred
approximately five minutes into the survey (
N
=262 in the “Obama Ad” group, 250 in the “McCain Ad” group, and
511 in the control group).

Responses to all of these items were gauged several minutes
after exposure.



30



Table 4
.

Immigrant Group Consciousness by Experimental Treatment Group



Obama Ad

McCain Ad

No Ad


Significance


Perceived Mistreatment

Due to Being an Immigrant

(%)



Three or more settings



12



8



8

Fewer than three
settings


12



18



11


No mistreatment



77



74



81


χ
2
=10.8,
p
=.029



Self
-
Designations (%)


American




28



28



36


χ
2
=13.1,
p

<.01


Mexican
-
American



48



47



53



χ
2
=3.0,
p
=.226


Latino





87



84



88



χ
2
=2.2,
p
=.329


Hispanic




93



91



93



χ
2
=.9,
p
=.623




**********************************************


Impact of Beliefs Concerning Group

Representation on Partisan Evaluations:

Multivariate OLS Effects and SEs


Latinos





1.05
(.42) *


.61 (.41)

1.06 (.18) **

Average American Voters



.85 (.37) *


.55 (.41)

1.13 (.30) **

Business Corporations



-
.57 (.35)


.28 (.38)

-
.22 (.27)

Immigrants




1.37 (.41) **

1.24 (.45) **


.54 (.32)


Constant




1.31 (.24) **


.86 (.24) **


.98
(.18) **






Note:
Perceived mistreatment

was measured in the context of finding employment, arranging housing, seeking
medical attention, shopping, and going about personal business “in the street.” Self
-
designations were measured by
reading a list of t
erms to respondents and asking them to state “which they would use to describe themselves.” In
the regression models,
the dependent variable is a
-
10 to + 10
relative

evaluation
scale for

the
two major
parties

(i.e.,
ratings

of
Democrats
on a 0
-
10 scale



ratings

of
Republicans

on a comparable scale
). Perceptions of partisan
representativeness were measured by asking responde
nts whether

each party “represented the interests of [
Latinos,
average American voters, business corporations,
and
immigrants
].” R
esponses
to each item
were dummy
-
coded
(1=yes) and differenced (perception of

Democratic Party representation


perception of Republican Party
representation
).


* =
p

< .05; ** =
p

< .01.



31


Notes




1

Much of the research for this paper was conducted while McCann was a visiting scholar at the Russell Sage
Foundation in New Yo
rk City (Fall
2010) and

the Institute for Policy Research at the Catholic University of
America in Washington, DC (Spring 2011).
We thank Aixa Cintrón
-
Vélez,
Geri Mannion,
Ann Marie Clark, and
Ronald B. Rapoport for helpful comments and assistance. Replication files are

available upon request:
mccannj@purdue.edu
.


2

Indeed, the number of immigrants in the
2006 Latino National Survey

who were “somewhat” or “very” interested
in politics closely matches the proportion of U.S. citizens who reported being “somewhat” or “very” interested in the
campaigns of 2004 and 2008, according to the American National Election Study.



3

Funding for t
hese surveys was provided by the Carnegie Corporation of New York, the Public Policy Institute at
the University of Texas, the College of Liberal Arts at Purdue, and the Center for Comparative Immigration Studies
at the University of California
-
San Diego.

N

= 350 in Dallas, 125 in San Diego, and 277 in Indiana. David Leal and
Wayne Cornelius collaborated in the design of the study, but w
e alone are responsible for the findings and
interpretations presented here. Nearly all interviews were conducted in Spa
nish, and on average lasted just over
twenty minutes. Interviewing Services of America (Van Nuys, CA), a firm with a long track
-
record of academic
survey research on Latinos, administered the interviews. Since no ready
-
to
-
use listings of immigrants are a
vailable
in th
e three regional sites,
random samples of “Mexican heritage” households
were obtained
from Geoscape
International (Miami, FL). Up to fifteen attempts were made to reach a respondent. Because the telephone records
contained both U.S.
-
born Mex
ican
-
Americans and immigrants in unknown proportions, and many lines were out of
service, there is no straightforward way to calculate a rate of response. If disconnected telephone lines, calls that
were never answered, busy signals, and individuals who as
ked to be contacted again before interviewers could
determine whether they fit the study profile are counted as “non
-
responses” (per the RR1 calculation in American
Association for Public Opinion Research 2006, 32), the estimated response rate is a rather
low 11 percent.
However, if the response rate is defined as the ratio of completed interviews / attempted interviews of subjects
known to fit the study protocol (i.e., RR5 in the AAPOR guide), this figure is dramatically higher at 89 percent.
Whatever the

method for calculating response rates, it is worth noting that with respect to key background variables
such as gender, age, level of education, church attendance, and language use at home, study participants are similar
to the Mexican
-
born respondents in

other large
-
scale surveys (e.g., Camp 2003; Moreno 2005; Pew Hispanic Center
2006).


4

Near the end of the

753 interviews

in June,
respondents were asked if they would be willing to take part in another
survey during the fall; 655 (or approximately 87 per
cent) agreed to this, left their first name or nickname, and gave
up to two telephone numbers where they could be called. In November, we were able to reach and interview 264
Mexicans for the second round. This represents a successful contact rate of 40
percent, using a baseline of 668
potential interviewees, or a rate of 35 percent if the baseline is the original 753 who were queried about participating
in a follow
-
up survey wave. The most common reason for a respondent to be dropped from the panel was
a
telephone line that was no longer in service. In cases where the line was still active and the respondent could be
reached (with up to fifteen attempts) nearly all (97 percent) participated in the survey. To our knowledge, in
political science this was

the first survey of Latino immigrants that has included a panel component. Panel attrition
is not significantly related to most of the socioeconomic and demographic variables (level of affluence, gender, level
of education, language use, religious practi
ce, or time spent in the United States). There is a small but statistically
significant correlation between age and being included in the second survey wave. As in the first wave, nearly all
interviews in November were conducted in Spanish, with each las
ting approximately sixteen minutes on average.


5

To retain all cases, missing values were imputed via
Amelia

(King et al. 2001). This has only a marginal effect on
the percentages of respondents wishing to remain or repatriate. Multiple imputation does,

however, slightly sharpen
the regression estimates in the lower part of the table.


6

Approximately one out of ten respondents had become a naturalized U.S. citizen, and the average length of stay in
the United States was 9.3 years.


32








7

The outcome of the

election, however, remained contested long after the July 2 contest, given the razor
-
thin margin
separating the first
-
place and second
-
place finishers (Felipe Calderón of the National Action Party and Andrés
Manuel López Obrador of the Party of the Democr
atic Revolution). But by September


well before the second
survey wave


popular mobilization against Mexico’s electoral authorities had largely ceased, though López
Obrador continued to protest the decision to declare Calderón the next president (Doming
uez, Lawson, and Moreno
2009).


8

Funding for Study 2

was provided by the Carnegie Corporation of New York, the Russell Sage Foundation, and
the
College of Li
beral Arts, Purdue University.


9

In the San Antonio sample, respondents had lived in the United

States an average of 23 years, which is
significantly longer than in either the Dallas or San Diego samples in 2006. Moreover, nearly half of the immigrants
in San Antonio reported being naturalized U.S. citizens, a proportion that is substantially highe
r than in the Indiana
sample.


10

This was the site of Gerald R. Ford’s storied tamale
-
eating incident. In an attempt to woo Mexican voters during
the 1976 campaign, the president learned the hard way, on camera in living color, that one should shuck tamal
es
before biting (Popkin 1991, 1).


11

As Iyengar and Kinder (1987
,

11) note in their “living room”

experiments involving
network news broadcasts, the
application of a treatment

must be accompanied with a reasonable explanation for why subjects are being as
ked to
behave in a certain way. For present purposes, the few questions that came after exposure to an ad are not of
substantive interest.


12

At least since the 1970s, presidential campaigns have attempted to reach out to Latinos through the strategic
deployment of “surrogates,” i.e., members of the Latino community who can sing the praises of the party or vouch
for its nominee. See Subervi
-
Vél
ez (2008, 34).


13

Before fielding the study, we listened to the advertisements through a conventional telephone line to confirm that
the sound quality was equivalent to what might be heard on a standard portable radio.


14

Prior research suggests that such
scales are more approachable for Mexicans than the 0
-
100 feeling thermometers
that commonly appear in American National Election Studies.



15

Differences in the causal effects of the two ads can be examined through equality constraints. Fixing the Obama
a
nd McCain commercials to have the same impact on residential preferences does not result in a significantly poorer
fit, as judged by the χ
2

goodness of fit (Long 1997).


16

Post
-
estimation tests of the coefficients show that the effect of perceptions regard
ing “immigrant representation”
in the two treatment groups could safely be assumed to be equivalent, with the estimated impact being significantly
higher than that for the control subjects.


17

These supplemental models are available upon request (
mccannj@purdue.edu
).