Coming Home: Attitudes toward U.S.
Veterans Returning from Iraq
Alair MacLean, Washington State University, Vancouver
Meredith Kleykamp, University of Maryland
In this article, we investigate public attitudes toward combat veterans returning from Iraq. Using data from a
nationally representative survey that incorporates an experimental design, we assess the extent to which attitudes
toward military veterans and private contractors differ, and whether public attitudes toward men vary based on combat and war zone experience. Drawing on social psychology and military sociology, we test hypotheses derived from a
conceptual model of stigma and from research on the cultural injunction to “support the troops.” Consistent with the
first portion of the stigma model, members of the public are not surprised to learn that men who went to a war zone
behave according to stereotypes that imply that such men have problems with mental health, substance abuse, and
violent behavior. Yet they do not discriminate against these men. Instead they favor men who went to Iraq compared
to those who stayed in the United States. They also favor veterans compared to contractors. While combat veterans may
be stereotyped, they are not stigmatized. They benefit from symbolic capital, which outweighs the effect of stereotypes on
discrimination. Keywords: military service; social inequality; stigma; symbolic capital; veterans.
Previous scholars have evaluated whether and how people who are mentally ill, particularly
those with violent tendencies, face stigma. According to this work, members of the public hold
stereotypes about the mentally ill and discriminate against them (Link et al. 1987; Pescosolido
et al. 2010). Yet, in certain cases, they hold stereotypes about how others will behave, but are less
likely than expected to discriminate (Corrigan et al. 2003; Link et al. 1999). They may hold unexpectedly positive views about those they believe are mentally ill because they believe the illness
is externally caused or about those who hold relatively greater power or capital (Link and
Phelan 2001).
Recent research has suggested that veterans returning from the war zones in Iraq and
Afghanistan are disproportionately likely to have poor psychological health, experiencing
symptoms, for example, of post-traumatic stress disorder (PTSD) and depression (Hoge et al.
2004; Tanielian and Jaycox 2008). U.S. troops have been at war for much of the last decade,
yet we still have little knowledge of how members of the public perceive these wartime veterans. In 2010, young veterans of this era were struggling to find jobs in the U.S. labor market;
the unemployment rate among 18- to 24-year-old veterans stood at 20.9 percent, while only
17.3 percent of comparable nonveterans were unemployed (Hefling 2011). Some observers
have explicitly and implicitly argued that these veterans have trouble finding jobs at least partly due to stigma. According to this argument proposed by some policymakers and veterans’
advocates, employers are reluctant to hire recent veterans, assuming that such veterans have
An earlier version of this article was presented at the 2010 meetings of the American Sociological Association in Atlanta,
GA. The authors are grateful to Jeremy Freese, Crosby Hipes, Robert Hughes, Monica Johnson, Jeff Lucas, Devah Pager, and
Jake Rosenfeld for input on earlier versions of this article. This research was partially funded by research grants from the
National Institute on Aging (R03 AG 029275) and the National Science Foundation (1048439). Data collected by the TimeSharing Experiments in the Social Sciences (TESS), NSF Grant 0818839, Jeremy Freese and Penny Visser, Principal Investigators. Direct correspondence to: Alair MacLean, Washington State University, Vancouver, 14204 NE Salmon Creek Ave.,
Vancouver, WA 98686. E-mail: alair.maclean@wsu.edu.
Social Problems, Vol. 61, Issue 1, pp. 131–154, ISSN 0037-7791, electronic ISSN 1533-8533. © 2014 by Society for the Study of
Social Problems, Inc. All rights reserved. Please direct all requests for permission to photocopy or reproduce article content
through the University of California Press’s Rights and Permissions website at www.ucpressjournals.com/reprintinfo/asp.
DOI: 10.1525/sp.2013.12074.
132
MACLEAN/KLEYKAMP
mental health problems, particularly PTSD (Gallucci 2010). Members of the public may also
fear and want to avoid combat veterans because of press reports that such veterans behave
violently (Mogelson 2011; Sullivan 2009). Yet little is known about whether and how these
combat veterans confront stereotypes and discrimination.
In the 1980s, researchers implicitly evaluated stigma directed at veterans by examining how
members of the public and employers responded to troops returning from Vietnam (Bordieri and
Drehmer 1984; D’Anton 1983). The findings of this research were mixed, suggesting that Vietnam
veterans experienced responses that were in some cases more negative than, while in others the
same as those experienced by their nonveteran peers (Bordieri and Drehmer 1984; D’Anton
1983). Apparently only one article assessed how employers respond to veterans of the contemporary era, suggesting that some of these veterans may not suffer if they served in combat occupations (Kleykamp 2009). Ten years after the wars in Afghanistan and Iraq began, however, we still
do not know whether people stereotype recent veterans, believing they have more symptoms of
mental illness due to combat exposure, and discriminate against them, treating them differently
than nonveterans. Few scholars have conducted research on how the public views veterans using
a quantitative approach. There is apparently only one survey that used these methods to examine
such attitudes toward veterans of the contemporary wars in Iraq and Afghanistan (Pew Research
Center 2011). During these wars, private military contractors worked in these war zones in similar
numbers to service members, yet very little research has examined attitudes toward them (Avant
and Sigelman 2010).
The following article assesses whether members of the general public view men differently
due to military service or combat exposure. It benefits from the fact that the contemporary wars
are the first in the modern era fought by the United States in which military troops have been
deployed alongside large numbers of civilians to war zones where they may be exposed to combat
(Congressional Budget Office 2008). We first elaborate previous theory and research that enable
us to predict how the public may view recent wartime veterans. We then test predictions drawn
from stigma and symbolic capital narratives using data from a survey experiment. According to
the findings, people stereotype but do not discriminate against men returning from a war zone,
and favor veterans compared to contractors with the same experiences.
A Conceptual Model of Stigma
We extend the social psychological concept of stigma to combat veterans, a group that is often
viewed as stigmatized, but that has apparently not been studied by stigma researchers. We draw on
elaborations of the theory of stigmatization in which scholars have delineated the necessary conditions for concluding that people stigmatize the members of a particular group (Link and Phelan
2001; Phelan, Link, and Dovidio 2008). We use this theory to develop our initial hypotheses.
Figure 1 represents a conceptual model of stigma consisting of three components: labels,
stereotypes, and discrimination (Link and Phelan 2001). The model draws on theories of labeling or societal reaction (Becker 1973; Goffman 1963; Scheff 1984), stigmatization, and prejudice (Phelan et al. 2008). As shown in the figure, people begin to stigmatize by first labeling
Labels
Figure 1 • Simple Conceptual Model of Stigma
Stereotypes
Discrimination
Attitudes toward U.S. Veterans Returning from Iraq
others. According to the first path in the figure, labels activate negative stereotypes about
groups, leading to assumptions about how the members of such groups behave. People expect
that someone labeled as mentally ill, for example, is dangerous and likely to behave violently
(Link et al. 1987; Link et al. 1999) and that someone described as a cocaine addict is more dangerous than someone described as a “troubled person” (Link et al. 1999). According to the second path in the figure, people are not stigmatized just because they are labeled and
stereotyped, but they must also experience discrimination. Two indicators of discrimination
are frequently used to evaluate stigma outside the lab context: support for help and social
closeness or distance. For example, mentally ill people are considered to be stigmatized because others are less likely to want to be neighbors and friends with them (Pescosolido et al.
2010). People are more likely to want to avoid children they believe are mentally ill than they
are children with physical ailments (Martin et al. 2007). They also do not want to provide the
mentally ill with help. They are less likely to say that they would provide men described as
schizophrenics with help if they believe the men are dangerous (Corrigan et al. 2003). This evidence has been used to support claims (Phelan et al. 2008) that people who are mentally ill
are stigmatized.
Wartime Veterans as a Stigmatized Group
Veterans have been characterized as a stigmatized group since at least the Vietnam War and
through the contemporary wars in Iraq and Afghanistan. When soldiers returned from the earlier
war in the 1960s and 1970s, they felt that the public condemned their behavior in combat (Shay
2002). Today, the public knows more about the psychological consequences of war than during the
Vietnam conflict. For example, PTSD was not formally considered a mental health diagnosis until
1980 (Yager, Laufer, and Gallops 1984). Such knowledge may increase the odds that people believe that combat veterans behave according to stereotypes. Combat soldiers and veterans of the
more recent wars have reported decreased mental health, and suffer from depression and PTSD at
relatively high rates, with as many as 1 in 5 experiencing either condition (Hoge et al. 2004; Kang
and Hyams 2005; Tanielian and Jaycox 2008). Researchers have begun to examine how these veterans may have been affected by traumatic brain injury (Institute of Medicine 2008). They have
also started to evaluate the social, psychological, and medical challenges faced by these veterans in
readjusting to their civilian lives (Institute of Medicine 2010). Newspaper stories report on the
violent behavior of combat soldiers abroad and veterans who have returned from war zones
(Mogelson 2011; Sullivan 2009). Several commentators have argued that increased attention to
the problems of PTSD among some veterans leads to stereotyping of all veterans as having mental
health problems (Balzar 2005). Based on these observations, we test the following prediction:
Stereotype hypothesis: Members of the public assume that combat veterans have more behavioral and
mental health problems than do men not exposed to combat.
As described by the conceptual model of stigma, stereotypes alone do not lead to stigma but
must be accompanied by discrimination. If people believe that combat veterans are more likely
than those who have not seen combat to be mentally ill, they may be less likely to want to be
socially close to and to provide help for such veterans. Some recent veterans suspect that employers are concerned about PTSD and mental instability and therefore discriminate against
them, making it harder for them to find jobs (Gallucci 2010). Because they fear discrimination,
some veterans say they do not include their military service on their resumes when they apply
for work in the civilian labor market (Hefling 2011). Veterans may also experience other forms
of discrimination, such as when they look for housing or simply interact with others (Balzar
2005). In addition, they may experience “self-stigma,” fearing the opinions of others. Recent
Iraq veterans report, for example, that they do not seek mental health assistance because they
fear stigma (Glantz 2009).
133
134
MACLEAN/KLEYKAMP
The anecdotal reports in the press and the suspicions of veterans would be consistent with
stigma research that finds that people discriminate against those believed to have mental health
problems. The simple model of stigma, therefore, suggests the following hypothesis:
Discrimination hypothesis: Members of the public are more likely to discriminate against others they
believe have behavioral problems, including combat veterans, than against
those without such problems.
How Stigma or Prejudice May Be Moderated or Modified
The conceptual model of stigma may be moderated or modified by attributions or different
forms of capital, which may apply to combat veterans. Several scholars have assessed whether the
stigma process is moderated by causal attributions. According to attribution theory, people may
view others as behaving according to stereotypes, but be less likely to discriminate against those
others if they believe that the stereotypical behavior was not internally but externally caused
(Corrigan et al. 2003; Martin et al. 2007). People want less social distance from, and support more
help for, a mentally ill person, for example, if they think that the mental illness was caused by an
accident or ordinary “troubles,” than if they think the illness was caused by drugs (Corrigan et al.
2003; Link et al. 1999). Scholars have argued that a number of potential causes of mental illness
are perceived as external, including combat (Scheff 1984).
People may also be less likely to discriminate against others they believe behave stereotypically if they believe those others have relatively more capital, either at the group or individual levels. This capital may be financial, but may also stem from social, cultural, or symbolic
resources (Bourdieu 1984). People have different levels of social capital based on the numbers
of friends, relatives, and acquaintances who can help them (Lin 2008; Mouw 2006). They have
different levels and types of cultural capital based on their knowledge and grasp of culture,
including culture’s material and nonmaterial aspects, as well as how to behave in different
social settings (Bourdieu 1984). They may also have different levels of symbolic capital based
on the symbolic boundaries between groups (Lamont and Molnar 2002; Sallaz and Zavisca
2007). Symbolic capital represents the “resources available to an individual on the basis of
honor, prestige or recognition, and functions as an authoritative embodiment of cultural value”
(Calhoun 2002:475).
We present these possible modifications or moderations to the conceptual model in graphic
form in Figure 2. The top path reflects the ideal type of the stigma process when people believe
that stereotypical behavior is internally caused, or when those who are stereotyped have limited
power or capital. The bottom path represents the same process when they believe the behavior is
externally caused, or when those who are stereotyped have more power or capital.
Internal cause,
limited power,
low capital
Labels
Discrimination
Stereotypes
External cause,
high power,
high capital
Figure 2 • Moderated/Modified Conceptual Model of Stigma
Attitudes toward U.S. Veterans Returning from Iraq
Wartime Veterans: Symbolic Capital
Due to symbolic capital, people may view veterans, particularly those who served in war
zones, in a relatively positive light. They may be less likely to discriminate against combat veterans
because they believe that such veterans have relatively more of one or another form of capital.
Individual combat veterans likely do not have more average economic, social, or cultural resources than people who did not see combat. Indeed, research has consistently demonstrated that such
veterans have lower economic capital, as they are disproportionately unlikely to come from the
top of the socioeconomic distribution (MacLean 2011; MacLean and Parsons 2010). As a group,
however, veterans may have access to symbolic capital stemming from the injunction to “support
the troops” (Coy, Woehrle, and Maney 2008).1 Since at least the Second World War, social psychologists have argued that citizens view wartime service members positively. According to this
argument, they see returning combat soldiers as having defended the nation, which leads to prestige (Schuetz 1945). During the Vietnam War, members of the Nixon administration claimed that
activists who opposed the war also undercut the soldiers who fought in it. Some peace activists
asserted, however, that they were protesting on behalf of the troops who, the activists argued, had
been betrayed by the government (Beamish, Molotch, and Flacks 1995). In 1991, both proponents and opponents of the first Persian Gulf War explicitly stated that they were supporting the
troops with their positions on the war. By the beginning of the more recent wars in Iraq and
Afghanistan, peace activists announced that they were supporting the troops before any service
members had even been deployed overseas. They have become increasingly likely to say that
while they do not agree that the United States should fight in wars, they admire the service members who fight (Coy et al. 2008; Leitz 2011).
While the word “troops” refers to people who are currently serving in the military, we argue
that the term draws on more long-standing cultural and political values that have benefited troops
who have returned from some wars, those who have become “veterans.” Scholars have argued
that veterans also benefit from the prestige accorded to returning service members, though this
prestige may not be as strong nor last long (Schuetz 1945). They have argued that veterans have
occupied a privileged position in U.S. culture and politics for more than a century (Skocpol 1993).
Veterans have received government assistance when such aid was not provided to civilians since
at least the Civil War. After that war, they received financial help in the form of pensions, while
their survivors received government benefits long before such programs were available to nonveterans (Skocpol 1993). During and after the First World War, they received a form of socialized
medicine from the government not available to other citizens (Linker 2011). After the Second
World War, they received funding for their education through the GI Bill, which some argue has
not been matched by programs available to nonveterans decades later (Skocpol 1996). Others theorize that military service rewards those who serve with both material and symbolic rewards, and
that the relative mix of the two forms of remuneration changes over time and depending on the
context of that service, with wartime service characterized as accruing high material and symbolic
rewards (or capital) (Levy 2007). According to this view, politicians reward wartime veterans generously because they view these veterans as having earned governmental support by fighting for
the nation.
The U.S. government has supported troops and veterans in different ways and to different extents across the last century and a half. After the First World War, it did not provide veterans with
outright benefits, but rather with medical rehabilitation. Indeed, in the 1930s, large numbers of
veterans of that war protested what they viewed as insufficient government support (Linker
2011:142). More recently, journalists have argued that the government has failed to adequately
1. We distinguish between how people respond to others as individuals (i.e., based on characteristics that they believe
individuals have) compared to how they respond to those others as members of groups (i.e., based on the average characteristics that they attribute to that group). We do not intend to assess attitudes that people have towards institutions, such as the
military.
135
136
MACLEAN/KLEYKAMP
support veterans returning from the wars in Iraq and Afghanistan (Glantz 2009). According to
this view, the Bush administration claimed to support the troops, but did not actually do so
(Glantz 2009).
Yet the “support the troops” narrative exemplifies a form of symbolic capital, suggesting
that people may view wartime military service as a source of honor or prestige based on sacrifice. Wartime veterans may benefit from an “economy of worth,” bearing the burden of potential danger and death in return for a later benefit (Boltanski and Thévenot 2006; see pp. 76, 93,
142). In this economy, people may be seen as more worthy if they offer to sacrifice their lives
by fighting in wars than if they do not. We draw on the long-standing political support for wartime veterans and the more recent cultural mandate to support the troops to derive the following prediction:
Symbolic capital hypothesis: Regardless of stereotypes, members of the public favor groups who have
more symbolic capital, such as combat veterans, relative to those who have
less, such as noncombat veterans and contractors.
The only data on public views of recent veterans comes from a Pew survey that finds
that more than 90 percent of the American public is proud of the troops serving in Iraq and
Afghanistan, while 33 percent report feeling ashamed of something the military has done in
those countries (Pew Research Center 2011). Although members of the public appear to support those who serve, they admit they know little about the problems facing those in the military (and we presume, by extension, those who have finished serving). Other than this study,
apparently no previous research has examined how the public views the troops returning
from the contemporary wars in Iraq and Afghanistan.
Previous Experimental Research on Attitudes toward Veterans
Previous researchers have explored the possibility that veterans suffer stigma especially in
employment-related outcomes (e.g., Bordieri and Drehmer 1984), demonstrating that employers
and the public may hold attitudes about veterans that are more negative or the same as those they
hold about nonveterans, while one article (Kleykamp 2010) finds that certain veterans may benefit from their status. Scholars have addressed this question since at least the Vietnam War by
comparing veterans who went to Vietnam to those who did not (Bordieri and Drehmer 1984).
Only one article has explicitly specified whether those veterans were exposed to combat (D’Anton
1983), which is an essential component of the stigma narrative.
One previous study suggests that Vietnam veterans faced stereotypes and discrimination
(Bordieri and Drehmer 1984). Experimental subjects were presented with identical resumes purportedly submitted by two types of veterans: the first had served in Vietnam and the second had
only served in the United States. The subjects viewed the two types of job candidates as equally
qualified, but they were more likely to assume that the Vietnam veteran had psychological problems and were less likely to recommend that that veteran be hired (Bordieri and Drehmer 1984).
Two studies suggest that employers hold neither negative nor positive views of veterans. One
study focused on Vietnam veterans and used a vignette approach (D’Anton 1983), while the other
study focused on veterans of the more recent era (without specifying whether the veterans saw
combat) using an audit approach (Kleykamp 2009). In the vignette study, personnel managers
were presented with 12 alternative descriptions of job candidates that varied whether the candidate: was a veteran or not, had served in Vietnam or not, was black or not, and had graduated
from high school or not. Managers viewed the Vietnam veterans no differently than the nonveterans (D’Anton 1983). In the audit study, employers were sent identical resumes in response to
advertisements for entry-level jobs. They responded to the resumes of veterans who were trained
in combat occupations and to resumes of nonveterans at statistically indistinguishable rates for
white and Hispanic veterans (Kleykamp 2009).
Attitudes toward U.S. Veterans Returning from Iraq
Only one article suggests that employers may have favored certain types of veterans of the
recent era. According to another audit study, employers preferred female veteran job seekers who
had been deployed to Iraq over their nonserving peers (Kleykamp 2010).
The Reference Group: Military Contractors
In the analyses described in the next section, we compare military veterans to the reference
group of former military contractors because the two groups resemble each other in three crucial
respects, with one important difference. Contractors served in similar numbers, performed similar
functions, and had the same potential experiences as service members, while differing in their institutional affiliation (Congressional Budget Office 2008; Singer 2003). As did service members,
contractors went to Iraq in large numbers. By 2008, they had come to represent half of U.S. personnel deployed to the wars in Iraq and Afghanistan (Avant and Sigelman 2010; Congressional
Budget Office 2008). These recent wars therefore became the first in which contractors served in
equal numbers as did military troops (Terry 2010). No other foreigners went to Iraq during that
war in anywhere near comparable numbers. Even United Nations (UN) peacekeepers, for example, only numbered 1,014 in 2009, of whom only 463 were not from Iraq (United Nations 2010).
As a consequence, contractors experienced relatively high numbers of casualties, with some estimating that such deaths constitute a third of the number of military deaths. According to previous
research, members of the public are as saddened and angered when contractors die as they are at
the deaths of service members (Avant and Sigelman 2010). Similar to service members, contractors filled various occupational roles in Iraq, ranging from those involving combat to those representing strategic and support functions (Congressional Budget Office 2008; Singer 2003). Because
of these large numbers and varied roles, they were subject to the same experiences as service
members, including potentially being exposed to combat.
Contractors differ in several important ways from service members, most importantly in that
they work not for the armed forces but for private firms, which implies less transparency and accountability. They do not report directly to congress, as does the Department of Defense (Avant
and de Nevers 2011). Compared to soldiers fighting in the Iraq war, contractors were the subject
of far fewer press stories. Scholars have estimated the press coverage of this group at less than
4 percent of the total wartime coverage (Avant and Sigelman 2010; Singer 2007). As a consequence, members of the public have been much less likely to find out when contractors were
killed. Indeed, up until recently, the numbers of contractors deployed alongside U.S. troops and
the deaths of those contractors were not tracked (Avant and Sigelman 2010; Terry 2010).
Contractors thus operated with far less public and congressional knowledge of their actions.
Data and Methods
Data
We evaluate whether the label “combat veteran” is associated with stereotypes and discrimination, and whether these associations are modified or moderated by causal attributions or symbolic capital, by analyzing data from a general population experiment. Such experiments combine
the broad coverage of a traditional survey administered to a random sample of the population
with the ability to isolate causal mechanisms provided by a traditional experiment. This experiment was conducted by the Time-Sharing Experiments in the Social Sciences (TESS) program,
which is funded by the National Science Foundation and managed by Knowledge Networks.2
2. For more information about the TESS program visit their website: www.tessexperiments.org (last retrieved
November 19, 2013).
137
138
MACLEAN/KLEYKAMP
Table 1 • Characteristics of Full and Analysis Samples Compared to those of the National Population
18 Years and Older
TESS data
Age
Male
Married
Race/ethnicity
White
Black
Other
Hispanic
2 + races
Highest degree
Less than HS
High school
Some college
BA+
N
ACS
Differences
Full Sample
Analysis Sample
3-Year Sample
Analysis - Full
46.36
(16.80)
.48
.51
46.37
(16.75)
.49
.51
46.26
(18.07)
.49
.52
−.01
.11
.00
.00
.00
−.01
.00
.00
.00
.00
.00
.02*
−.01
−.01
.00
−.01**
.01
.00
.00
.00
−.01
−.06***
.04***
.02**
.69
.11
.05
.13
.01
.69
.11
.05
.13
.01
.13
.32
.28
.28
3,021
.12
.32
.28
.28
2,892
.67
.12
.06
.14
.02
.13
.37
.24
.26
20,955,548
Analysis - ACS
Note: Standard errors in parentheses.
*p < .05 ** p < .01 ***p < .001 (two-tailed tests)
Knowledge Networks staff recruit panels of respondents from a sampling frame that encompasses
the population of people 18 and older in the United States. They are able to reach approximately
98 percent of the population by contacting potential subjects through a combination of random
digit dialing and address-based sampling (Brashears 2011). They ask potential respondents to respond to surveys that are collected through computer-assisted self-interview over the Internet. If
the respondents do not have computers or access to the Internet, they are provided with laptops
and free Internet access.
The data were collected in an online survey in September 2009. The survey questions were
asked of a TESS Knowledge Networks survey panel with 3,021 participants. Due to missing data
the final sample size is 2,892. We present demographic characteristics of the full sample and the
sample used for the analyses in Table 1. According to the table, the respondents in the analysis
sample had demographic characteristics that were not statistically different from those of the full
sample. In addition, they had similar characteristics to the U.S. population age 18 years and older
reflected in responses to the American Community Survey (ACS), with several exceptions. They
were the same average age as the respondents to the ACS, and they were as likely to be male,
married, black, Hispanic, and in the “other” racial ethnic category, as well as to have dropped out
of high school. They were slightly more likely to be white (69 percent compared to 67 percent),
and less likely to report being two or more races (1 percent compared to 2 percent). They were
also more likely to report having some college and above (28 percent reported having these levels
of education compared to 24 and 26 percent among the ACS), and less likely to report having just
a high school degree (32 percent compared to 37 percent).
In the survey, respondents first read a short vignette that described a man and then
answered questions about their attitudes toward that man (see Appendix A for an outline of
vignettes and questions). Approximately 500 respondents were randomly assigned to one of
six different vignettes.
We extend previous research by looking at three basic comparisons that test whether people
have different attitudes toward: (1) veterans compared to contractors, (2) men who went to a war
zone compared to those who did not, and (3) men who saw combat compared to those who did
not. These comparisons enable us to separate attitudes based on the type of worker, soldier, or
Attitudes toward U.S. Veterans Returning from Iraq
contractor, from the type of experience, combat or not. As elaborated above, contractors were
similar to service members in Iraq in their numbers, functions, and experiences. The following
analyses exploit these similarities to construct the experimental vignettes so as to differentiate between the attitudes that people hold toward men based on their experiences, such as being sent to
Iraq or seeing combat, from those they hold based on institutional affiliations, such as serving in
the military or working for a private firm.
According to the stigma narrative of combat, veterans who fought in wars are stigmatized.
People may assume that any person returning from a war zone has seen direct combat (although
not all have). If this is the case, they should view men who simply went to Iraq in the same light as
they view those who are specified to have seen combat. In order to test for this possibility, we
compare attitudes toward men who went to Iraq to attitudes toward men who went to Iraq and
came under enemy fire. Accordingly, the vignettes describe two thirds of the subjects as having
been sent to Iraq, one third with combat specified, and the other third without combat specified.
In these vignettes, we describe the vignette subjects as men because they have been the focus
of previous research and because male veterans constituted roughly 80 percent of the veterans
under 30 years old in 2010 (authors’ calculation from American Community Survey data
[Ruggles et al. 2010]). Men have also been more likely than women to be exposed to combat during wartime. We do not specify the race/ethnicity of the vignette subjects because this would have
added additional complexity and sample size demands to an already relatively complex research
design. The use of an experimental approach involves tradeoffs between variation in the demographic characteristics in the study and the size of the analytic sample; each inclusion of a new
variable cuts the sample by the number of categories in that variable of interest, or requires
multiplying the desired sample size by that number to maintain statistical power. Future researchers could elaborate the findings of this research by varying the vignette subjects’ gender and race/
ethnicity.
After reading the vignettes, respondents answered eight questions measuring stereotypes
and discrimination. To assess stereotypes, three questions asked whether respondents would be
surprised to learn that the vignette subjects have problems with mental health, substance abuse,
or violent behavior. The remaining questions reflect two dimensions of discrimination: social
closeness and support for help. Three questions assess social closeness, measured by how pleased
respondents are by the prospect of the man just described becoming a neighbor, friend, or
coworker. Two questions assess whether respondents think that the vignette subject should be
helped by the government or receive preferential hiring treatment from employers.
Methods
We test the preceding hypotheses in two steps: first, we assess whether the individual questions reflect underlying attitudes and, second, we evaluate whether these attitudes are affected by
the characteristics of the man described in the vignette. In the first step of the analysis, we assess
whether the answers to the questions are correlated enough to suggest that they reflect underlying attitudes about problematic behavior, social closeness, and support for help. The underlying
attitude that captures problematic behavior reflects whether respondents are not surprised to hear
that a vignette subject is being treated for a mental health problem, has abused drugs or alcohol,
and has acted violently.3 According to the social closeness attitude, respondents say they are
pleased to have the subject as a neighbor, coworker, and friend. According to the support for help
attitude, they think that the subject should receive help from the government and be treated
3. We reverse code the questions that load onto this factor. Thus, we refer, for the most part, to the extent to which the
respondents are not surprised to hear that the vignette subjects behave problematically. Because increases in the factor reflect
decreases in surprise at problematic behavior, or increases in expectations that the subjects behave problematically, we label
this factor “expected problematic behavior.”
139
140
MACLEAN/KLEYKAMP
preferentially by employers. In the second step of the analyses, we evaluate whether respondents
have different underlying attitudes toward the different vignette subjects.
We conduct these two steps using two different techniques: (1) structural equation models
and (2) factor analyses followed by OLS regressions. Both sets of analyses are based on the correlations among the items. Using the structural equation technique, we estimate models of the relationships between the attitudes across multiple groups defined by vignette subjects. Using the
second technique, we first conduct a factor analysis of the attitude questions. We then calculate
factor scores using the simple method of taking the average of the items for each factor to produce
a factor score (DiStefano, Zhu, and Mîndrila 2009). This method retains the original metric, from 1
to 9, of the questions. We also predict factor scores using regressions based on rotated, orthogonal
factors representing each attitude for each respondent, which produces similar estimates, though
in a standardized metric. We then regress these factor scores on the vignette subject characteristics, as well as other independent variables. These two techniques produce results that are substantively similar. Following previous research (Corrigan et al. 2003), we therefore report
results from the factor analyses and regressions using average factor scores because this technique
allows us to show how veteran status, war zone status, and combat status are related to the attitudes separately in a more easily interpreted metric (though we include results from the other
methods in appendices).
Findings
According to the findings, people view men returning from war zones more negatively than
they do men who stayed in the United States in terms of stereotypes, but more positively in terms
of desires for social closeness and support for help. They do not differentiate between veterans and
contractors in terms of stereotypes, but do view veterans more positively in terms of their desires
for social closeness and support for help. In the following section, we first describe differences in
the means of the eight observed variables across vignettes. We then describe estimates from the
models of the relationships between the factors, which allow us to assess whether people have different attitudes about how stereotypes relate to discrimination for the different types of vignette
subjects, enabling us to test our hypotheses directly.
Differences in Attitudes
Table 2 presents mean differences in the eight outcome measures across the six groups
defined by the vignettes, which imply that members of the public respond to war zone veterans
in ways that could be consistent with the stereotype, attribution, and symbolic capital hypotheses,
but not with the discrimination hypothesis. Our measures are coded so that higher numbers indicate that the respondent: is less surprised to hear that the person described in the vignette had
problems with the items; would be more pleased to be in the specified relationship with the
person; and is more supportive of the specified help for the person described in the vignette.
Respondents judge men differently based on institutional affiliation and war zone experience.
They are less surprised to hear that men who went to Iraq behave problematically than those who
did not go abroad. They also prefer to be socially closer to and support more help for men who
served in the military than contractors. They are also more positive toward men who went to Iraq
than to those who remained in the United States regardless of military service. These results
suggest that war zone exposure determines judgments about problematic behavior, and that such
exposure interacts with military service to produce views about closeness and help.
We calculated descriptive statistics with and without weights to account for differential response. The estimates of the characteristics of the sample with and without the weights differ by
trivially small amounts and the differences are not statistically significant. Therefore, we present
the unweighted statistics (see Appendix B for weighted statistics).
Attitudes toward U.S. Veterans Returning from Iraq
Table 2 • Descriptive Statistics of Variables by Vignette
Military
Panel 1: Expected problems
Mental health
Substance abuse
Assault
Panel 2: Social closeness
Neighbor
Coworker
Friend
Panel 3: Help
Government
Hiring preferences
Military Contractor
U.S. Only
Deployed
Combat
U.S. Only
Deployed
Combat
5.12
(2.37)
5.11
(2.26)
4.18
(2.29)
6.11***
(2.32)
5.87***
(2.21)
4.77***
(2.25)
6.36***
(2.17)
5.99***
(2.15)
4.82***
(2.31)
4.99
(2.27)
5.12
(2.30)
4.31
(2.27)
5.83***
(2.24)
5.83***
(2.26)
4.78***
(2.31)
6.27***
(2.17)
6.05***
(2.21)
4.99***
(2.37)
7.16
(1.65)
7.20
(1.59)
6.02
(1.88)
7.36
(1.63)
7.37
(1.69)
6.27*
(1.90)
7.33
(1.69)
7.33
(1.67)
6.32*
(1.91)
6.60***
(1.74)
6.61***
(1.68)
5.53***
(1.92)
6.91*
(1.77)
6.87**
(1.76)
5.86
(2.03)
6.92*
(1.76)
6.95*
(1.72)
6.02
(1.86)
6.94
(1.80)
5.01
(2.43)
7.33***
(1.70)
5.26
(2.61)
7.63***
(1.64)
5.53***
(2.49)
5.98***
(2.10)
4.16***
(2.54)
6.72
(1.96)
4.61*
(2.59)
6.64*
(2.08)
4.69*
(2.57)
Notes: Measures range from 1 through 9, with increases representing increases in: (1) respondent’s expectations that the
vignette subject exhibits the problematic behavior; (2) respondent’s pleasure at having the specified relationship of social
closeness: and (3) amount the respondent thinks that the vignette subject should receive the specified type of help (see
Appendix A for more detail). Standard errors in parentheses.
*p < .05 **p < .01 ***p < .001 (two-tailed tests of difference between specified vignette and person in the military who served in
the United States)
Panel 1 of Table 2 presents evidence that is consistent with the stereotype hypothesis. It
shows that respondents are less surprised that men who went to Iraq behave more problematically than those who stayed only in the United States, but that they do not distinguish
between veterans and former contractors on these measures. Respondents are less surprised
to learn that men are being treated for mental health problems, have issues with substance
abuse, or have been charged with assault if the men went to Iraq instead of staying in the
United States. Serving or working in Iraq is associated with greater assumptions of problematic
behaviors, even with no explicit mention of exposure to violence or combat. We interpret this
as evidence that people assume that men who spent time in a war zone were exposed to traumatic events that lead to problematic behavior.
According to the second panel in the table, respondents want to be socially closer to men who
served in the military than to those who worked for private military contractors. They also prefer
men who went to Iraq to those who stayed in the United States, though some of the differences
are not statistically significant. Respondents are more pleased to be neighbors or coworkers with
veterans than with former military contractors. They are also more pleased to be close to contractors if the contractors went to Iraq. They express slightly different patterns when describing their
thoughts about friendship; veterans who deployed or saw combat are seen as more desirable as
friends than are veterans who served only in the United States. Among men described as only
serving or working in the United States, respondents prefer friendships with veterans over military contractors. They are as pleased to be friends with contractors who worked or saw combat in
Iraq as they are with veterans who served only in the United States. Taken together, these results
suggest that members of the public have more negative views of contractors than of veterans, but
141
142
MACLEAN/KLEYKAMP
they have less negative views if the contractors spent time in a war zone. The findings contradict
the discrimination hypothesis, but could be consistent with either the attribution or symbolic capital hypotheses.
The third panel of the table presents results that also suggest people do not discriminate on
the basis of stereotypes about war zone survivors, but rather favor them, perhaps because of
causal attributions or symbolic capital. According to this panel, respondents have different opinions about whether men should receive help from the government and from employers if the
men were military veterans or former contractors and if the men served overseas or in the United
States. They are more supportive of help, within each category, for men who went to Iraq or saw
combat. Yet they express more support for giving government help and hiring preferences to veterans who served only in the United States than they do for giving such help to private military
contractors who went to Iraq.
We conducted factor analyses that suggest that the answers to these individual questions are
correlated in such a way that they are best represented by three factors. The problematic behavior
factor has an alpha of .82, the social closeness factor has an alpha of .87, and the support for help
factor has an alpha of .61.
Figure 3 therefore contains average scores for these factors and shows that people do not
simply discriminate against those they stereotype, which is inconsistent with the direct model of
stigma. It presents the factor scores by vignette subject on each of the three factors: problematic
behavior, support for help, and social closeness. The factor scores are calculated using the simple
method, which retains the metric of the original questions, ranging from 1 through 9. In the figure, the vignette subjects are ranked from highest to lowest values with respect to assumed problems. People are least surprised to find that men who experienced combat would behave
problematically, regardless of whether the men served in the military or not. They expect that
men who went to Iraq have nearly the same levels of problematic behavior even when men are
not specified to have seen combat. They expect that men are the least likely to behave problematically if the men stayed in the United States.
Problems
Help
Closeness
Military, combat
Contractor, combat
Military, Iraq
Contractor, Iraq
Military, U.S.
Contractor, U.S.
1
2
3
4
5
6
7
8
9
Figure 3 • Average Factor Scores by Vignette Subject Type
Note: Measures range from 1 through 9, with larger numbers representing increases in respondent’s: (1) expectations that the
vignetee subject exhibits problems, (2) pleasure at having the specified social closeness with the subject, and (3) support for
help (see Appendix A for more detail).
Attitudes toward U.S. Veterans Returning from Iraq
The patterns differ with respect to the order in which the vignette subjects are arrayed on the
other two factors. People support more help for and want to be closer to veterans than contractors,
though they support help for and want to be close to veterans in the following order: those who
were exposed to combat, those who went to Iraq, and those who served only in the United States.
They discriminate in favor of former contractors who experienced combat or went to Iraq relative
to contractors who only stayed in the United States. It appears that exposure to risk merits assistance for all, including contractors, but military service, even low-risk service, trumps exposure to
risk among private contractors, which is consistent with the view that veterans, especially those
who saw combat, benefit from an economy of worth based on their perceived sacrifices (Boltanski
and Thévenot 2006; Levy 2007).
According to the direct model of stigma, people should see those they believe behave more
problematically as less worthy of help and social closeness. Yet the figure shows that people want
to be closer to and support more help for some of the men they believe behave most problematically. These findings suggest that some men may suffer less stigma than would be expected according to the direct model of stigma because they benefit either from a moderated or modified
stigma process.
Relationships between Labels, Stereotypes, and Discrimination
Table 3 presents estimates from statistical models of the components of stigma and relationships between those components according to the conceptual model. The first three rows present
estimates of how people’s attitudes vary by the characteristics of the vignette subjects. The remaining rows describe how people’s attitudes vary by their own characteristics. The analyses suggest
that the stigma process for war zone veterans is not a moderated one. We tested for interactions
between veteran status and war zone status, as well as each of these variables and the problematic
behavior scale. None of these interactions were significant. Therefore we present models with just
these additive, main effects. (Estimates from the models with and without interactions are available in Appendix C.) These findings suggest that the effects of these variables are not moderated,
but are additive.
The first column presents estimates of the effects of the variables on expected problematic behavior, suggesting that people hold stereotypes about how men behave after returning from war
zones, but not about how veterans behave relative to contractors, which is consistent with the stereotype hypothesis. According to these estimates, people are more likely to believe that men behave problematically if the men were sent to Iraq whether or not the men experienced combat.
They do not, however, hold different opinions about whether veterans behave more or less problematically than do contractors. These findings are consistent with previous research, which
showed that experimental subjects assumed that Vietnam veterans had psychological problems
(Bordieri and Drehmer 1984). In addition, older or married people are less likely, on average, to
assume that the vignette subjects behave problematically than are younger or unmarried people.
More educated people are more likely to assume that the men behave problematically than are
less educated people.
The next four columns present estimates of the effects of the variables on social closeness and
support for help, which are consistent not with the discrimination, but with the symbolic capital
hypothesis. According to the estimates, people discriminate in favor both of men returning from
war zones and of men who served in the military. Columns two and three contain estimates of the
relationships between the independent variables and desires for social closeness. According to
these estimates, people are more pleased to be close to men who have returned from a war zone
than they are to those who remained in the United States. They also prefer to be closer to veterans
than to former contractors. According to the estimates in column three, people want to be less
socially close to men who they expect behave problematically. After accounting for the impact of
expected problematic behavior, people report being slightly more pleased to be close to men returning from war zones than they did in the model that did not include this behavior. Due to the
143
144
MACLEAN/KLEYKAMP
Table 3 • OLS Regression Coefficients from Models of Stereotypes and Discrimination
Assumed
Problems
Combat/deployment
status (ref: U.S. only)
Iraq
Iraq and combat
Veteran status (ref:
military contractor)
Assumed problems
Age
Male
White
Education (ref: high
school dropout)
High school graduate
Some college
College graduate
Married
Constant
Observations
R-squared
Social
Closeness
Support
for Help
Model 1
Model 2
Model 3
Model 4
Model 5
.728***
(.087)
.943***
(.087)
.009
(.071)
.244***
(.071)
.278***
(.071)
.479***
(.058)
.490***
(.084)
.633***
(.084)
.844***
(.069)
−.011***
(.002)
.009
(.071)
−.051
(.081)
.001
(.002)
.049
(.058)
.158*
(.066)
.387***
(.070)
.462***
(.070)
.481***
(.056)
−.196***
(.015)
−.001
(.002)
.050
(.056)
.148*
(.064)
.010***
(.002)
−.145*
(.069)
−.465***
(.078)
.600***
(.084)
.776***
(.084)
.845***
(.068)
−.152***
(.018)
.008***
(.002)
−.144*
(.068)
−.473***
(.077)
.085
(.121)
.334**
(.125)
.408**
(.126)
−.200**
(.073)
5.221***
(.177)
2892
.07
.168
(.099)
.053
(.102)
−.247*
(.103)
.248***
(.060)
5.964***
(.145)
2892
.05
.185
(.096)
.118
(.100)
−.168
(.100)
.209***
(.058)
6.984***
(.161)
2892
.10
−.066
(.117)
−.293*
(.121)
−.563***
(.122)
−.071
(.071)
5.320***
(.171)
2892
.10
−.053
(.116)
−.243*
(.120)
−.501***
(.121)
−.101
(.070)
6.113***
(.193)
2892
.12
Note: Standard errors in parentheses.
*p < .05 **p < .01 ***p < .001 (two-tailed tests)
counter-balancing preferences for veterans, on the one hand, and war zone survivors on the
other, people want to be just slightly closer to veterans who stayed in the United States than to
contractors who saw combat in Iraq.
Columns four and five contain similar models to those in columns two and three, and
suggest that people also support more help for men returning from Iraq than for men who
stayed in the United States and for veterans than for contractors. As with social closeness,
people think that men who went to Iraq deserve more help than those who stayed in the
United States, with those exposed to combat deserving the most help. They also think that
veterans deserve more help than do contractors. According to column 5, they support less
help for men they believe behave problematically. Whites tend to want to be socially closer
to the vignette subjects, but support less help for them. College graduates are also less likely
to support helping these men.
As mentioned above, we tested the robustness of these results in two ways, estimating the
same models using the regression-based factor scores, which present the effects in a standardized
metric. The results from this model may be found in Appendix D. We also estimated similar models in a structural equation framework. The results from this model may be found in Appendix E.
Attitudes toward U.S. Veterans Returning from Iraq
Conclusion
According to the preceding findings, people hold negative stereotypes about how men who
return from war zones behave, but have favorable attitudes about such men, as well as about veterans, likely due to these groups’ symbolic capital. If people believe that a man behaves more
problematically, they are more likely to stigmatize him. Yet they hold positive attitudes towards
the members of particular groups, such as war zone survivors, that can outweigh the stigma process at the individual level. Because of these preferences, people are as positive about veterans
who served only in the United States as they are about military contractors recently returned from
a war zone. We speculate that these positive attitudes stem from the increasingly common exhortation to “support the troops,” which may benefit people who go to war zones, even those who
are not in the armed forces.
The findings support the conclusion that people stigmatize others they believe behave problematically, wanting greater social distance from and supporting less help for those others, which
is consistent with a long line of research on stigma (Goffman 1963; Link and Phelan 2001). Yet the
effects of the labels “war zone” and “veteran” are powerful enough to counteract or overcome the
typical association between negative stereotyping and discrimination. Taken together, these findings present a paradox. Members of the public want to be more socially distant from and support
less help for men they believe behave problematically. They discriminate against those they believe have more problems. They also believe that war zone survivors behave more problematically
as a group than do men who did not go to war zones. Thus, they should want to be less close to
and support less help, on average, for veterans of the war in Iraq. And yet the opposite occurs:
people want to be socially closer to and support more help for these veterans compared to other
veterans and nonveterans.
We speculate that this paradox stems not from causal attributions, but from cultural values
about wartime veterans. According to attribution theory, people respond less negatively to others
who behave problematically if they believe that the behavior was not internally but externally
caused. Our analyses show, however, that people respond equally negatively to a man they believe behaves problematically regardless of whether that man has been to a war zone. War zone
survivors, however, appear to benefit from the imperative to support the troops, as do veterans in
general, not at the individual level but at the group level. People therefore want to be closer to and
support more assistance for such men. Both types of men may benefit from economies of
worth, because members of the public may perceive both groups as sacrificing (Boltanski and
Thévenot 2006). This elaboration is consistent with recent scholarship elaborating a Framework Integrating Normative Influences on Stigma (FINIS), which emphasizes macro-social
influences on stigma, particularly the normative cultural context in which the stigma process
occurs (Pescosolido et al. 2008). The preceding analyses of stigma among combat veterans are
noteworthy in that they demonstrate how cultural values can counteract the association
between stereotypes and discrimination.
Our findings are consistent with theory that suggests that the process by which people are
stigmatized is not a simple one, and that the link between stereotypes and discrimination can be
modified (Corrigan et al. 2003; Link and Phelan 2001). Scholars have demonstrated, for example,
that the process can be modified by causal attributions (Corrigan et al. 2003). Based on our findings, we suggest that the stigma process may also be moderated or modified by symbolic capital.
We have shown that it can be modified, sometimes in contradictory ways, by two different sets of
labels, those differentiating people who went to war zones from those who did not, and those differentiating veterans from contractors.
Our findings contribute to the literature that has assessed the impact of the recent shift from
fighting wars using service members toward prosecuting those wars with contractors (Avant and
Sigelman 2010; Schaub and Franke 2009–2010; Terry 2010), suggesting that this shift may also
change attitudes toward wartime survivors. Scholars have argued that this shift has reduced transparency and accountability (Avant and Sigelman 2010; Singer 2007). According to our findings,
145
146
MACLEAN/KLEYKAMP
people view wartime survivors as behaving in the same fashion regardless of whether the survivors are veterans or former contractors. Yet they hold unexpectedly positive views about combat
survivors and, in each case, they favor veterans over contractors. They are therefore more likely to
discriminate against wartime survivors who worked for contractors than those who served in the
armed forces. We have argued that they do so because they have been repeatedly exhorted to
support the troops. Yet they may prefer veterans to contractors for many other reasons. They may,
for example, view contractors in a particularly negative way. According to previous research, people believe that contractors are motivated more by money than by patriotism, while they view
veterans as holding the opposite motivations (Avant and Sigelman 2010). They may also view
former contractors less positively than former service members, because they are more familiar
with the struggles of veterans, about whom they have read in the press. Future research should
continue to evaluate attitudes towards military contractors more explicitly, especially in light of
their increased role in U.S military operations.
The preceding analyses do not separate the factors influencing people’s attitudes, such as
their images of the men, or their views based on news reports. As mentioned above, both contractors and service members performed multiple functions not strictly defined by combat while in
Iraq. The analyses do not test whether and how knowing about these functions might alter the
opinions that people have. We cannot tell, in either case, whether the respondents picture the
vignette subject as a medic or a warrior. This topic represents a promising avenue for future research. In addition, press coverage of both veterans and contractors may influence the attitudes
that people hold at particular points in time. People may hold more positive attitudes toward such
men when there have recently read or heard reports that elicit sympathy, such as stories about
casualties. They may, however, hold more negative attitudes toward such men after news reports
of atrocities committed by either group. The analyses cannot specify which images are most salient
in the minds of the respondents, though this question is also worthy of further research.
It is possible that our findings reflect, in part, social desirability bias in responding to questions about military personnel rather than true positive sentiment toward veterans. Preliminary research explicitly testing for social desirability suggests no major social desirability
effects with respect to social distance measures, and only small effects with respect to the provision of assistance to veterans (Kleykamp 2011).
Our analyses reach similar conclusions regarding attitudes toward men returning from Iraq
as some prior research about attitudes toward men returning from Vietnam. Vietnam veterans
were assumed to have more problems than veterans who did not go to Vietnam (Bordieri and
Drehmer 1984). Yet survey respondents thought that government programs should target such
veterans to help them reintegrate into civilian society (Veterans Administration 1980). Our findings suggest that these attitudes apply not just to veterans, but to all men returning from war
zones. People appear to believe that all war zone survivors behave according to negative stereotypes, but discriminate in favor of such survivors. Future researchers should evaluate if attitudes
toward men returning from war zones depend on the era in which the war occurred. They should
also evaluate whether attitudes toward contemporary veterans change, as research on Vietnam
veterans found increasingly negative attitudes toward veterans over time (Veterans Administration 1980).
Our analyses present a model for applying the concept of stigma to members of groups to
which this concept has not traditionally been applied. Useful theory should explain both positive
and negative cases, helping to understand what leads to the presence and absence of stigma when
conditions suggest that stigma is likely. Scholars could further elaborate the stigma model by examining the relationships between stereotypes and discrimination among members of other
groups who may benefit from positive cultural values. They could thereby specify the conditions
under which labels can be associated with both negative stereotypes and positive cultural values.
They could evaluate, for example, attitudes toward the firefighters who responded to the planes
flying into the World Trade Center, toward refugees from countries that are considered to be
enemies of the United States, or toward humanitarian workers experiencing mental illness from
Attitudes toward U.S. Veterans Returning from Iraq
vicarious trauma. They could also examine the complex interplay of stigma and prejudice for the
members of other groups. For example, researchers could test whether people hold more positive
attitudes about white mental patients than they hold about blacks without mental illness. Future
research could also evaluate how labels may interact in other contexts. Researchers could assess,
for example, whether people hold unexpectedly positive attitudes about others who work for the
government rather than for private entities, or just about those who serve in the military.
Future researchers could also assess how the opinions and attitudes that people hold
about combat veterans translate into how veterans are treated, by employers and by members
of the public. Despite the support expressed in the general population experiment for war
zone survivors, employers may still discriminate against wartime veterans when making actual hiring decisions. Audit studies of military veterans suggest, however, that employers do not
discriminate against contemporary veterans. Employers are more likely to call back veteran
applicants, including those who have been deployed, than their civilian peers (Kleykamp
2009, 2010). Researchers could also evaluate whether members of the public act to be socially
closer to war zone veterans than to other veterans or nonveterans, either in the context of lab
or field experiments. In addition, they could explore whether people are willing to see preferential hiring programs implemented not just in theory, but in practice, and to bear the costs of
such programs.
The current analyses focus on how members of the public view male veterans with race and
ethnicity unspecified, but future research should also examine attitudes toward female veterans as
well as toward veterans of particular racial and ethnic categories. Women constitute 15 percent of
the contemporary armed forces (Segal and Segal 2004) and have increasingly come into contact
with hostile forces in Iraq (Alvarez 2009). Yet little is known about whether and how members
of the public view female veterans differently from male veterans and from female civilians. In addition, the armed forces are very diverse with respect to race and ethnicity (Segal and Segal 2004).
Future researchers could evaluate whether members of the public respond differently to veterans
of different racial and ethnic categories.
While U.S. troops have officially ended combat operations in Iraq and will soon do so in
Afghanistan, more than two million service members have served in these war zones (Institute of
Medicine 2010). If history is a guide, many of these troops will return with mental health diagnoses (Tanielian and Jaycox 2008). Recent veterans have suffered from their service both mentally
and physically (Glantz 2009; Hoge et al. 2004). They may confront disability and unemployment
throughout their work lives (MacLean 2010). According to our findings, members of the public
will label such troops as suffering the mental health effects of combat. Yet they will also hold relatively positive attitudes toward these returning service members and veterans in terms of desire
for social closeness and of support for help.
Appendix A: Survey Items
Text in brackets represents the different presentations of test conditions.
“John is a single, 24-year-old man with an associate’s degree. For the past 4 years, he [served in the
military/worked as a private military contractor]. During that time, he [was (deployed/sent) to Iraq/came
under fire when his convoy was attacked in Iraq/(served/worked) only in the continental United States].
He is currently looking for a job and an apartment.”
After reading the vignettes, the subjects answered the following questions:
1. How would you feel about John moving in next door to you?
Very displeased 1 2 3 4 5 6 7 8 9 Very pleased
2. How would you feel about John being hired as someone you would work closely with on a daily basis?
Very displeased 1 2 3 4 5 6 7 8 9 Very pleased
147
148
MACLEAN/KLEYKAMP
3. If John did move into your neighborhood, how likely do you think it would be that you two would be
friends?
Very unlikely 1 2 3 4 5 6 7 8 9 Very likely
4. Government programs sometimes help people find jobs. How much should the government help
John?
Not at all 1 2 3 4 5 6 7 8 9 A great deal
5. If John and an equally qualified candidate were being considered for a job, how much preferential
treatment should John be given?
None at all 1 2 3 4 5 6 7 8 9 A great deal
6. How surprised would you be if you learned that John was receiving regular treatment for a mental
health problem?
Not at all surprised 1 2 3 4 5 6 7 8 9 Very surprised
7. How surprised would you be if you learned that John had a drug or alcohol problem?
Not at all surprised 1 2 3 4 5 6 7 8 9 Very surprised
8. How surprised would you be if you learned that John had recently been accused of assault?
Not at all surprised 1 2 3 4 5 6 7 8 9 Very surprised
Appendix B • Descriptive Statistics of Variables by Vignette (weighted)
Military
U.S. Only
Panel 1: Expected problems
Mental health
5.13
(2.36)
Substance abuse
5.11
(2.26)
Assault
4.18
(2.30)
Panel 2: Social closeness
Neighbor
7.18
(1.61)
Coworker
7.20
(1.58)
Friend
6.00
(1.89)
Panel 3: Help
Government
6.95
(1.81)
Hiring preferences
4.98
(2.43)
Military Contractor
Deployed
Combat
U.S. Only
Deployed
Combat
6.10***
(2.33)
5.88***
(2.21)
4.78***
(2.23)
6.40***
(2.15)
6.00***
(2.13)
4.84**
(2.31)
4.99
(2.26)
5.11
(2.28)
4.29
(2.25)
5.86***
(2.24)
5.85***
(2.27)
4.81***
(2.30)
6.29***
(2.15)
6.04***
(2.19)
4.98***
(2.36)
7.36
(1.63)
7.37
(1.69)
6.29*
(1.90)
7.33
(1.69)
7.33
(1.65)
6.31**
(1.91)
6.61***
(1.72)
6.61***
(1.66)
5.55***
(1.87)
6.90**
(1.78)
6.87**
(1.75)
5.82
(2.03)
6.89**
(1.75)
6.92*
(1.72)
6.00
(1.84)
7.34***
(1.69)
5.28
(2.59)
7.65***
(1.60)
5.51***
(2.48)
5.94***
(2.07)
4.13***
(2.52)
6.72
(1.95)
4.57**
(2.61)
6.61*
(2.08)
4.69
(2.55)
Note: Standard errors in parentheses.
*p < .05 **p < .01 ***p < .001 (two-tailed tests of difference between specified vignette and person in the military who served
in the United States)
Appendix C • OLS Regression Coefficients from Models of Stereotypes and Discrimination with Interactions
Expected Problems
Model 1a
Social Closeness
Support for Help
Model 1b
Model 2a
Model 2b
Model 3a
Model 3b
Model 4a
Model 4b
Model 5a
Model 5b
Combat/deployment status (ref: U.S. only)
Iraq
.728***
.675***
(.087)
(.122)
Iraq and combat
.943***
.952***
(.087)
(.123)
Veteran status (ref:
.009
−.020
military contractor)
(.071)
(.122)
Assumed problems
.244***
(.071)
.278***
(.071)
.479***
(.058)
.290**
(.100)
.363***
(.100)
.564***
(.099)
.387***
(.070)
.462***
(.070)
.481***
(.056)
−.196***
(.015)
.474*
(.196)
.480*
(.204)
.481***
(.056)
−.189***
(.024)
.490***
(.084)
.633***
(.084)
.844***
(.069)
.630***
(.118)
.619***
(.119)
.927***
(.118)
.600***
(.084)
.776***
(.084)
.845***
(.068)
−.152***
(.018)
.638**
(.236)
.849***
(.245)
.845***
(.068)
−.145***
(.029)
Interaction of veteran status with
Iraq
Iraq and combat
.107
(.173)
−.019
(.173)
−.090
(.142)
−.169
(.142)
−.285
(.168)
.031
(.167)
−.011***
(.002)
.009
(.071)
−.051
(.081)
.001
(.002)
.049
(.058)
.158*
(.066)
.001
(.002)
.049
(.058)
.159*
(.066)
−.001
(.002)
.050
(.056)
.148*
(.064)
−.017
(.035)
−.004
(.036)
−.001
(.002)
.050
(.057)
.148*
(.064)
.083
(.121)
.333**
(.125)
.408**
(.126)
.168
(.099)
.053
(.102)
−.247*
(.103)
.168
(.099)
.055
(.102)
−.247*
(.103)
.185
(.096)
.118
(.100)
−.168
(.100)
.187
(.096)
.120
(.100)
−.166
(.101)
Iraq and combat
Age
−.011***
(.002)
Male
.009
(.071)
White
−.051
(.081)
Education (ref: high school dropout)
High school graduate
.085
(.121)
Some college
.334**
(.125)
College graduate
.408**
(.126)
.010***
(.002)
−.145*
(.069)
−.465***
(.078)
.010***
(.002)
−.148*
(.069)
−.465***
(.078)
.008***
(.002)
−.144*
(.068)
−.473***
(.077)
−.008
(.042)
−.014
(.043)
.008***
(.002)
−.145*
(.068)
−.473***
(.077)
−.066
(.117)
−.293*
(.121)
−.563***
(.122)
−.061
(.117)
−.291*
(.121)
−.562***
(.122)
−.053
(.116)
−.243*
(.120)
−.501***
(.121)
−.052
(.116)
−.242*
(.120)
−.500***
(.121)
(continued )
Attitudes toward U.S. Veterans Returning from Iraq
Interaction of problems with
Iraq
149
150
MACLEAN/KLEYKAMP
Appendix C • OLS Regression Coefficients from Models of Stereotypes and Discrimination with Interactions (Continued)
Expected Problems
Married
Constant
Observations
R-squared
Social Closeness
Support for Help
Model 1a
Model 1b
Model 2a
Model 2b
Model 3a
Model 3b
Model 4a
Model 4b
Model 5a
Model 5b
−.200**
(.073)
5.221***
(.177)
2892
.07
−.199**
(.073)
5.235***
(.185)
2892
.07
.248***
(.060)
5.964***
(.145)
2892
.05
.249***
(.060)
5.918***
(.151)
2892
.05
.209***
(.058)
6.984***
(.161)
2892
.10
.209***
(.058)
6.951***
(.187)
2892
.10
−.071
(.071)
5.320***
(.171)
2892
.10
−.071
(.071)
5.278***
(.179)
2892
.10
−.101
(.070)
6.113***
(.193)
2892
.12
−.100
(.070)
6.080***
(.224)
2892
.12
Note: Standard errors in parentheses.
*p < .05 **p < .01 ***p < .001 (two-tailed tests)
Attitudes toward U.S. Veterans Returning from Iraq
Appendix D • OLS Regression Coefficients from Models of Stereotypes and Discrimination (standardized
factors)
Assumed
Problems
Social
Closeness
Model 1
Combat/deployment status (ref: U.S. only)
Iraq
.342***
(.039)
Iraq and combat
.433***
(.039)
Veteran status (ref:
.035
military contractor)
(.032)
Assumed problems
Age
−.004***
(.001)
Male
−.003
(.032)
White
−.026
(.036)
Education (ref: high school dropout)
High school graduate
.055
(.054)
Some college
.171**
(.056)
College graduate
.197***
(.056)
Married
−.084*
(.033)
Constant
−.120
(.079)
Observations
2892
Combat/deployment
.07
status (ref: U.S.
only)-squared
Support
for Help
Model 2
Model 3
Model 4
Model 5
.154***
(.042)
.163***
(.042)
.290***
(.035)
.166***
(.028)
.222***
(.028)
.240***
(.023)
.001
(.001)
−.017
(.035)
.095*
(.040)
.171***
(.043)
.185***
(.043)
.292***
(.035)
−.049*
(.020)
.001
(.001)
−.017
(.035)
.094*
(.040)
.002**
(.001)
−.021
(.023)
−.172***
(.026)
.180***
(.028)
.239***
(.028)
.241***
(.023)
−.039**
(.013)
.002*
(.001)
−.021
(.023)
−.173***
(.026)
.143*
(.059)
.087
(.061)
−.029
(.062)
.132***
(.036)
−.493***
(.087)
2892
.04
.145*
(.059)
.095
(.061)
−.019
(.062)
.128***
(.036)
−.499***
(.086)
2892
.04
−.057
(.039)
−.121**
(.040)
−.223***
(.040)
−.047*
(.023)
−.065
(.057)
2892
.09
−.054
(.039)
−.114**
(.040)
−.215***
(.040)
−.051*
(.023)
−.070
(.057)
2892
.09
Note: Standard errors in parentheses.
*p < .05 **p < .01 ***p < .001 (two-tailed tests)
Appendix E • Estimates from Best-Fitting Structural Equation Model of Labels, Stereotypes, and
Discrimination
Labels (ref: Veteran, U.S. only)
Veteran, Iraq
Contractor, U.S. only
Contractor, Iraq
Assumed Problems
Social Closeness
Support for Help
.73***
(.07)
.00
.33***
(.09)
−.59***
(.10)
−.13
(.09)
.37***
(.07)
−.56***
(.09)
−.11
(.07)
.00
(.02)
.73***
(.07)
Stereotypes
Assumed problems
—
−.22***
(.02)
Discrimination
Social closeness
—
—
Note: Standard errors in parentheses.
*p < .05 **p < .01 ***p < .001 (two-tailed tests)
.44***
(.03)
151
152
MACLEAN/KLEYKAMP
References
Alvarez, Lizette. 2009. “G. I. Jane Breaks the Combat Barrier.” New York Times, August 16. Retrieved November
19, 2009 (www.nytimes.com/2009/08/16/us/16women.html?scp=1&sq=gi%20jane&st=cse).
Avant, D. and L. Sigelman. 2010. “Private Security and Democracy: Lessons from the US in Iraq.” Security
Studies 19(2):230–65.
Avant, D. D. and R. de Nevers. 2011. “Military Contractors and the American Way of War.” Daedalus
140(3):88–99.
Balzar, John. 2005. “What Awaits at Home?” Los Angeles Times. Retrieved January 18, 2011 (http://articles.
latimes.com/print/2005/feb/23/entertainment/et-veterans23).
Beamish, Thomas D., Harvey Molotch, and Richard Flacks. 1995. “Who Supports the Troops? Vietnam, the
Gulf War, and the Making of Collective Memory.” Social Problems 42(3):344–60.
Becker, Howard Saul. 1973. Outsiders: Studies in the Sociology of Deviance. New York: Free Press.
Boltanski, Luc and Laurent Thévenot. 2006. On Justification: Economies of Worth. Princeton, NJ: Princeton
University Press.
Bordieri, James E. and David E. Drehmer. 1984. “Vietnam Veterans: Fighting the Employment War.” Journal
of Applied Social Psychology 14(4):341–47.
Bourdieu, Pierre. 1984. Distinction: A Social Critique of the Judgment of Taste. Cambridge, MA: Harvard University
Press.
Brashears, M. E. 2011. “Small Networks and High Isolation? A Reexamination of American Discussion
Networks.” Social Networks 33(4):331–41.
Calhoun, Craig J. 2002. “Symbolic Capital.” Pp. 475 in Dictionary of the Social Sciences, edited by C. J. Calhoun.
New York: Oxford University Press.
Congressional Budget Office. 2008. Contractors’ Support of U.S. Operations in Iraq. Washington, DC: Congressional Budget Office.
Corrigan, Patrick, Fred E. Markowitz, Amy Watson, David Rowan, and Mary Ann Kubiak. 2003. “An Attribution Model of Public Discrimination towards Persons with Mental Illness.” Journal of Health and Social
Behavior 44(2):162–79.
Coy, Patrick G., Lynne M. Woehrle, and Gregory M. Maney. 2008. “Discursive Legacies: The U.S. Peace Movement and ‘Support the Troops’.” Social Problems 55(2):161–89.
D’Anton, Michael A. 1983. “Personnel Officers’ Attitudes toward Vietnam Era Veterans.” Journal of Political and
Military Sociology 11(1):21–34.
DiStefano, Christine, Min Zhu, and Diana Mîndrila. 2009. “Understanding and Using Factor Scores: Considerations for the Applied Researcher.” Practical Assessment, Research & Evaluation 14(20). Retrieved March 5,
2012 (pareonline.net/getvn.asp?v=14&n=20).
Gallucci, Ryan. 2010. “Is Public Talk About PTSD Making It Harder for Vets?” Washington Post. Retrieved
January 18, 2011 (http://voices.washingtonpost.com/impact-of-war/2010/04/is_public_talk_about_ptsd_
maki.html).
Glantz, Aaron. 2009. The War Comes Home: Washington’s Battle against America’s Veterans. Berkeley: University of
California Press.
Goffman, Erving. 1963. Stigma: Notes on the Management of Spoiled Identity. Englewood Cliffs, NJ: Prentice-Hall.
Hefling, Kimberly. 2011. “2010 Jobless Rate for Young War Veterans at 20.9%.” Washington Post.
Retrieved March 18, 2011 (www.washingtonpost.com/rw/WashingtonPost/Content/Epaper/201103-18/Bx4.pdf).
Hoge, Charles W., Carl A. Castro, Stephen C. Messer, Dennis McGurk, Dave I. Cotting, and Robert L. Koffman.
2004. “Combat Duty in Iraq and Afghanistan, Mental Health Problems, and Barriers to Care.” New
England Journal of Medicine 351(1):13–22.
Institute of Medicine. 2008. Long-Term Consequences of Traumatic Brain Injury. Washington, DC: National Academy Press.
———. 2010. Returning Home from Iraq and Afghanistan: Preliminary Assessment of Readjustment Needs of Veterans,
Service Members, and Their Families. Washington, DC: National Academies Press.
Kang, Han K. and Kenneth C. Hyams. 2005. “Mental Health Care Needs among Recent War Veterans.” New
England Journal Of Medicine 352(13):1289.
Kleykamp, Meredith. 2009. “A Great Place to Start? The Effect of Prior Military Service on Hiring.” Armed
Forces & Society 35(2):266–85.
———. 2010. “Women’s Work after War.” SSRN eLibrary. Retrieved March 5, 2013 (http://ssrn.com/
paper=1676970).
Attitudes toward U.S. Veterans Returning from Iraq
———. 2011. “The Societal Context of Reception for Veterans.” Presented at the annual meeting of the
American Sociological Association, August 21, Las Vegas, NV.
Lamont, M. and V. Molnar. 2002. “The Study of Boundaries in the Social Sciences.” Annual Review of Sociology
28:167–95.
Leitz, Lisa. 2011. “Oppositional Identities: The Military Peace Movement’s Challenge to Pro-Iraq War Frames.”
Social Problems 58(2):235–56.
Levy, Yagil. 2007. “Soldiers as Laborers: A Theoretical Model.” Theory and Society 36(2):187–208.
Lin, Nan. 2008. “Social Networks and Status Attainment.” Pp. 580–82 in Social Stratification: Class, Race, and Gender in Sociological Perspective, edited by D. B. Grusky, M. C. Ku, and S. Szelényi. Boulder, CO: Westview
Press.
Link, Bruce G., Francis T. Cullen, James Frank, and John F. Wozniak. 1987. “The Social Rejection of Former
Mental-Patients: Understanding Why Labels Matter.” American Journal of Sociology 92(6):1461–1500.
Link, Bruce G. and Jo C. Phelan. 2001. “Conceptualizing Stigma.” Annual Review of Sociology 27:363–85.
Link, Bruce G., Jo C. Phelan, Michaeline Bresnahan, Ann Stueve, and Bernice A. Pescosolido. 1999. “Public
Conceptions of Mental Illness: Labels, Causes, Dangerousness, and Social Distance.” American Journal of
Public Health 89(9):1328–33.
Linker, Beth. 2011. War’s Waste: Rehabilitation in World War I America. Chicago: University of Chicago Press.
MacLean, Alair. 2010. “The Things They Carry: Combat, Disability, and Unemployment among U.S. Men.”
American Sociological Review 75(4):563–85.
———. 2011. “The Stratification of Military Service and Combat Exposure.” Social Science Research 40:336–48.
MacLean, Alair and Nicholas Parsons. 2010. “Unequal Risk: Combat Occupations in the Volunteer Military.”
Sociological Perspectives 53(3):347–72.
Martin, Jack K., Bernice A. Pescosolido, Sigrun Olafsdottir, and Jane D. McLeod. 2007. “The Construction of
Fear: Americans’ Preferences for Social Distance from Children and Adolescents with Mental Health
Problems.” Journal of Health and Social Behavior 48(1):50–67.
Mogelson, Luke. 2011. “A Beast in the Heart of Every Fighting Man.” The New York Times. Retrieved June 13,
2011 (www.nytimes.com/2011/05/01/magazine/mag-01KillTeam-t.html).
Mouw, Ted. 2006. “Estimating the Causal Effect of Social Capital: A Review of Recent Research.” Annual
Review of Sociology 32:79–102.
Pescosolido, Bernice A., Jack K. Martina, Annie Lang, and Sigrun Olafsdottir. 2008. “Rethinking Theoretical Approaches to Stigma: A Framework Integrating Normative Influences on Stigma (Finis).” Social Science &
Medicine 67(3):431–40.
Pescosolido, Bernice A., Jack K. Martin, J. Scott Long, Tait R. Medina, Jo C. Phelan, and Bruce G. Link. 2010.
“‘A Disease Like Any Other’? A Decade of Change in Public Reactions to Schizophrenia, Depression, and
Alcohol Dependence.” American Journal of Psychiatry 167(11):1321–30.
Pew Research Center. 2011. “The Military-Civilian Gap: War and Sacrifice in the Post-9/11 Era.” Retrieved
January 28, 2012 (www.pewsocialtrends.org/files/2011/10/veterans-report.pdf).
Phelan, Jo C., Bruce G. Link, and John F. Dovidio. 2008. “Stigma and Prejudice: One Animal or Two?” Social
Science & Medicine 67(3):358–67.
Ruggles, Steven, J. Trent Alexander, Katie Genadek, Ronald Goeken, Matthew B. Schroeder, and Matthew
Sobek. 2010. Integrated Public Use Microdata Series: Version 5.0 [MRDF]. Minneapolis: University of Minnesota [producer and distributor].
Sallaz, J. J. and J. Zavisca. 2007. “Bourdieu in American Sociology, 1980–2004.” Annual Review of Sociology
33:21–41.
Schaub, Gary, Jr. and Volker Franke. 2009–2010. “Contractors as Military Professionals.” Parameters
39:88–104.
Scheff, Thomas J. 1984. Being Mentally Ill: A Sociological Theory. New York: Aldine Pub. Co.
Schuetz, Alfred. 1945. “The Homecomer.” American Journal of Sociology 50(5):369–76.
Segal, David R. and Mady W. Segal. 2004. “America’s Military Population.” Population Bulletin 59(4):3–40.
Shay, Jonathan. 2002. Odysseus in America: Combat Trauma and the Trials of Homecoming. New York: Scribner.
Singer, P. W. 2003. Corporate Warriors: The Rise of the Privatized Military Industry. Ithaca, NY: Cornell University
Press.
———. 2007. “Can’t Win with ’Em, Can’t Go to War without ’Em: Private Military Contractors and Counterinsurgency.” Washington, DC: The Brookings Institution.
Skocpol, Theda. 1993. “America First Social-Security System: The Expansion of Benefits for Civil War Veterans.” Political Science Quarterly 108(1):85–116.
———. 1996. “Delivering for Young Families: The Resonance of the Gi Bill.” The American Prospect 28:66–72.
153
154
MACLEAN/KLEYKAMP
Sullivan, Julie. 2009. “An Effective Defense, an Uncertain Future: Iraq Vet Jessie Bratcher Will Go to the
Oregon State Hospital, Not Prison.” The Oregonian, December 8, pp. A1, A9.
Tanielian, Terri L. and Lisa Jaycox. 2008. Invisible Wounds of War: Psychological and Cognitive Injuries, Their
Consequences, and Services to Assist Recovery. Santa Monica, CA: RAND.
Terry, James P. 2010. “Privatizing Defense Support Operations: The Need to Improve Dod’s Oversight and
Management.” Armed Forces & Society 36(4):660–70.
United Nations. 2010. United Nations Peace Operations 2009: Year in Review. New York: United Nations.
Veterans Administration. 1980. Myths and Realities: A Study of Attitudes toward Vietnam Era Veterans. Washington,
DC: U.S. Government Printing Office.
Yager, Thomas, Robert Laufer, and Mark Gallops. 1984. “Some Problems Associated with War Experience in
Men of the Vietnam Generation.” Archives Of General Psychiatry 41(4):327–33.