Do Americans Know More Than We Think about Public Budgets?
Reconsidering Popular Overestimates of Foreign Aid and Government Waste
Vanessa Williamson
Brookings Institution
Using public opinion data from the Kaiser Family Foundation and a new survey of attitudes
about government, I examine how Americans think about government waste and foreign aid, and
how their definitions of these terms affect their estimates of and attitudes towards those
expenditures. I find that “foreign aid” is widely understood to encompass overseas military
spending for humanitarian purposes, and that the term “government waste” is popularly used to
discuss systemic failures of the democratic process. Respondents who use broader definitions for
these terms have substantially higher estimates of their respective costs. Taking account of what
members of the public mean by “waste” and “foreign aid,” existing studies overestimate public
ignorance and obscure the substance of public critiques of U.S. policy.
Introduction
How much do Americans know about government spending? Two commonly cited
survey results suggest that the answer is “very little indeed.” For decades, it has been reported
that Americans vastly overestimate foreign aid spending. As one recent study notes, “On
average, Americans think 28 percent of the federal budget is spent on foreign aid, when it is
about 1 percent.” (Kaiser Family Foundation, 2013) Similarly, U.S. survey respondents tend to
dramatically overestimate government waste. While expert assessments put federal government
waste at a few pennies on the dollar, survey respondents think about half the federal budget is
wasted, on average (Riffkin, 2014). These misperceptions are commonly understood as evidence
of profound public ignorance of fiscal matters.
Building on intuitions developed in open-ended interviews, a re-examination of the
survey data suggests a different interpretation of these overestimates. Many Americans,
especially those of lower education levels, conceive of “foreign aid” as overseas military
spending, and those who think of foreign aid as including military investments pick significantly
higher estimates of the foreign aid budget. Similarly, what Americans mean by “waste” includes
more than just inefficiency. Taking account of the popular terminology substantially reduces the
extent to which these estimates can be defined as a misperception and suggests that scholars both
reconsider how public policy knowledge is measured and re-examine popular critiques of
government expenditures.
A Review of Public Misinformation
Though members of the public may be able to use information shortcuts to compensate
for a lack of factual information (e.g. Popkin, 1991, but c.f. Lau and Redlawsk, 2001), it is
1
widely agreed that an informed citizenry is a good thing for democracy. “Democracy functions
best when its citizens are politically informed,” as Delli Carpini and Keeter put it (1996).
Given this normative appeal, there is an immense literature examining what Americans
know and do not know about politics. Much of this literature is pessimistic (for a thorough
review, see Lupia and McCubbins, 1998: 3). Most famously, Philip Converse (1964) argued that
many Americans hold “non-attitudes” on major political issues of the day. Many scholars have
critiqued the characterization of Americans as “know nothings,” (Bennett, 1988), most
commonly by re-examining survey methodology (e.g. Achen, 1975). A more theoretical critique
comes from Lupia (2006), who notes that assessments of voter competence can be undermined
by “elitism” because “questions test information that academics, journalists, and politicos value.”
Whether theoretical or methodological, the debate about public misinformation has
typically centered on whether, in assessing the public’s level of knowledge, scholars are asking
the right questions. This paper addresses a different but related concern; I argue that in certain
contexts we may not be correctly interpreting survey respondents’ answers. I re-examine popular
overestimations of government waste and foreign aid. These particular overestimations are
worthy of attention for three reasons. First, overestimates of foreign aid and government waste
are extreme. Second, these errors can plausibly be imagined to influence voters’ political
judgments in substantive ways. Finally, scholars and reporters treat these overestimates as
evidence of innumeracy or ignorance, and therefore these statistics may have a meaningful
impact on elite perceptions of democratic responsiveness.
Survey respondents overestimate many, many things. In the United States and abroad,
poll respondents overestimate the unemployment rate, life expectancy, the average age of the
population, and the size of minority and immigrant groups, to name just a few well documented
2
examples (Duffy and Stannard, 2014). Still, the extent to which Americans overestimate foreign
aid and waste are extreme; in some polls, the difference between perception and reality is a
factor of twenty to one or more. Similarly, government waste is regularly estimated at
approximately half of the federal budget – an overestimation by an order of magnitude. While a
smaller mistake might reasonably be overlooked, errors of this scale imply major
misunderstandings of government policy and are therefore worthy of scholarly study.
Moreover, the foreign aid and waste errors are on a subject about which many people
have strong opinions. Over half of Americans support “major reductions” in foreign aid spending
(Kaiser Family Foundation, 2013). But given accurate information about the foreign aid budget,
survey respondents revise their views. Told that the U.S. spends about 1% of its budget on
foreign aid, about half of Americans who had previously said the U.S. spends too much on
foreign aid changed their views and said the U.S. spends about the right amount or too little. To
adopt the terminology of Hochschild and Einstein (2015), foreign aid and waste are subjects
about which Americans appear to be actively misinformed; their incorrect impressions are
associated with specific policy preferences. A similar dynamic may be at play with government
waste; voters’ support for major tax cuts has been associated with perceptions of government
inefficiency (Ladd and Wilson, 1983). It seems especially worthwhile to examine misperceptions
that appear to have an effect on attitudes and policy outcomes.
Finally, public estimates of foreign aid and waste have become touchstones for
academics, journalists and policymakers. In his analysis of government trust, Marc Hetherington
repeatedly refers to survey data showing high estimates of government waste, which he deems a
“misperception” (2005, 10). Derek Bok uses the same survey data to assert a “widespread
impression” of government spending for which there is “little evidence” (1997: 62). Similarly,
3
Manza, Cook and Page (2002) explain public opposition to foreign aid as a result of “extreme
overestimation” of the foreign aid budget. David Hudson and Jennifer vanHeerde-Hudson
(2012), despite noting the problem of measurement error in surveys about foreign aid, look at
U.S. public estimates of foreign aid and conclude that “Americans, apparently, are particularly
ignorant.” As McDonnell, Lecomte and Wegimont (2003: 228) argue, topline survey data about
foreign aid spending can reinforce “long-held views – prejudices, perhaps – about American
introspection and isolationism.”
In the popular media foreign aid and government waste estimates are also widely cited.
“Yet another depressing survey was released last week that attests to the failure of most
Americans to understand the basics of the federal budget,” wrote Washington Post correspondent
Glen Kessler in 2011. “Four Pinocchios to the American public,” he concludes. Thus these
survey results play a role in elite debates about public misinformation about government, and
underwrite doubts about the prospects for democratic accountability.
For all of these reasons, misperceptions of public waste and foreign aid are worthy of
attention. To do so, I use a process of qualitative hypothesis generation followed by quantitative
testing in nationally representative survey samples. I begin by noting attitudes about foreign aid
and government waste, as I heard them expressed in the course of 49 open-ended phone
interviews about taxation and government that I conducted with Americans in 21 states in the fall
of 2013 and the spring of 2014.1 The words of these interviewees, drawn from respondents to an
online survey, provide useful insight into the potential mechanisms and motivations behind their
individual survey responses. But no small-n qualitative sample can be deemed representative. I
therefore turn to large-n national surveys to test how Americans’ definitions of “government
waste” and “foreign aid” affect their estimates of these expenditures. First, I examine perceptions
4
of foreign aid, qualitatively and then quantitatively. Then, I follow the same structure in
assessing the public understanding of government waste.
What Americans Mean By Foreign Aid: Interviews Provoke New Hypotheses
As my interviewees elaborated on their feelings about foreign aid, it became clear that
their concerns were not primarily about humanitarian spending. Erick2 is a 43-year-old man from
Michigan. Asked what he disliked about government spending, he says:
Erick: Our money is sent overseas. That bothers me.
Interviewer: Anything else?
Erick: Like money to Israel I don’t like. We send money to Egypt. And the wars that are
going on over there still.
Erick mentions “money to Israel” and “money to Egypt” – both part of the foreign aid budget –
and also “the wars that are going on over there still.” For him, our interventions in the Middle
East seem to be one category of spending, money that is “sent overseas.” Interestingly, this
interviewee was already familiar with the budget data about foreign aid. Without prompting, he
went on to mention that “if you look at a pie chart, foreign aid is, like, a sliver” of the federal
budget – a fact that left him confused.
Another interviewee also volunteered her thoughts about the federal budget pie chart, and
expressed the same bemusement. Marjorie, 53-year-old woman from Ohio, named foreign aid
among the programs she opposed, and then continued, “They say that it’s only one percent of the
budget or something like that, but like I said, it seems like we’re giving billions of dollars to
people in Afghanistan.” Marjorie is correct that Afghanistan receives billions in foreign aid; all
told, the country received about $13 billion in U.S. aid in 2012. But she may also thinking more
5
generally of the decade-long American military presence in the region; she does not make a clear
distinction.
The open-ended discussions I had with my interviewees suggest a new interpretation of
the opposition to foreign aid. Is it possible that a substantial portion of the public are eliding
between foreign aid and defense spending? The quantitative survey data suggests that many
Americans’ conception of “foreign aid” is tied to U.S. military interventions.
Do Survey Respondents See “Foreign Aid” as Overseas Military Spending?
Though many surveys ask Americans to estimate foreign aid, hardly any have asked
Americans what they mean by the term. One exception is a February 2012 study by the Kaiser
Family Foundation. The survey was conducted among a nationally-representative random digit
dial landline and cell phone sample of 1205 adults and weighted to match 2011 Census data on
sex, age, education, race, Hispanic origin, and region. The AAPOR RR3 response rate was 24
percent for the landline sample and 21 percent for the cell phone sample.
In this survey, respondents were asked an open-ended question: Just your best guess,
what percentage of the federal budget is spent on foreign aid? Half the survey respondents were
later asked, Thinking about U.S. spending on foreign aid, what types of things do you think this
money is spent on? This question was also open ended, and up to three responses were recorded
for each survey participant. Their answers were then sorted into fourteen categories, including
“food,” “health care,” “education/schools,” “military/weapons/defense,” and “clean water.”
Strikingly, more respondents thought of foreign aid as military spending than disaster
relief, education, and economic development combined. Only food aid and health care were as
common a response as military spending when it came to respondents’ first thoughts about
6
foreign aid. In all, more than a quarter of respondents mentioned military spending as one of the
things they consider foreign aid.3
If overestimates of foreign aid spending were driven in part by a confusion of aid and
military budgets, one would expect those who think of foreign aid in military terms to have
substantially higher estimates of the foreign aid budget. I test this hypothesis using a beta
regression of the relevant demographic and attitudinal variables. Beta regression, rather than a
transformation, is recommended for modelling proportion outcomes between 0 and 1 because the
results are directly interpretable (Cribari-Neto and Zeileis, 2010).
Table 1 reviews the correlates associated with respondents’ estimates of foreign aid. All
models include demographic variables commonly associated with differing estimates of foreign
aid; as a rule, people of higher socio-economic status pick lower estimates. Income is measured
on an eight-point scale from “less than $20,000” to “$150,000 or more.” College graduate is an
indicator for receipt of a four-year college degree, white is an indicator variable for being white
and non-Hispanic, and male is also an indicator variable. Party ID is measured on a 5-point scale
from 1, strong Democrat, to 5, strong Republican. Ideology is measured on a 3-point scale from
1, liberal to 3, conservative.4 I also include a measure of respondents’ preferences for
isolationism; those who believe the U.S. should be a world leader tend to pick higher estimates
of foreign aid spending.
[Table 1]
Model 2 adds an indicator variable for whether the respondent thought of foreign aid in
terms of military spending. Those who think of foreign aid as military spending do indeed pick
substantially higher estimates of the foreign aid budget. All else being equal, a person who
7
thought of foreign aid as military spending picked an estimate of foreign aid 55% higher than
someone who thought of other uses for foreign aid. The difference between thinking of foreign
aid as military spending versus humanitarian spending is as big as the impact of having a fouryear college degree. Moreover, as Model 3 reveals, the effect is concentrated among those with
lower levels of education – those least likely to be familiar with the definition of foreign aid
employed by policy experts.
Thus the tendency to overestimate foreign aid can be explained in part by the tendency of
Americans to think of foreign aid in military terms. Those who pick higher estimates may be
defining the term more broadly than policymakers do. As we will see in the following section, a
similar dynamic exists when it comes to “government waste.”
What Americans Mean By Government Waste
As we saw with foreign aid, my interviews suggest that popular understandings of waste
differ from the experts’ definition. Gabriel, a 28-year-old man from Utah, was one of several
interviewees who thought of government waste differently from policy experts:
8
Interviewer: And, then, this is just a personal opinion, but how much of every dollar do
you think the government wastes?
Gabriel: 30 cents.
Interviewer: And—go on—
Gabriel: How do you define waste?
Interviewer: That’s what I was going to ask you.
Gabriel: Yeah, I guess, to me, waste would be any frivolous spending, even if it is, you
know, going back to the military, even if it is going to something tangible, you know, I
think that’s wasteful. But, even if they were to keep up everything that’s going on, I
think a certain percentage of that is just falling through the cracks to inefficiency. If I had
to define the things I disagree with as waste, then probably 30 percent, I would say.
Interviewer: And, if you were just thinking about inefficiency, would you have a
different number?
Gabriel: Yeah, inefficiency, probably around 15 percent.
Gabriel knows that military spending produces “tangible” results, but still sees this spending as
wasteful, because he disagrees fundamentally with American military policy. Similarly, Lucy, a
59-year-old living in Chicago, sees waste almost exclusively in terms of military spending. But
she does not mean military spending is wasteful in the sense a policymaker might. American
military spending, she believes, is not “a good idea” even in principle. “Inefficiency,” she says
when I inquire, “doesn’t even play into it.”
Military spending is not the only government investment that my interviewees saw as
wasteful in principle, rather than just in its execution; other interviewees describe social
9
spending, arts spending and spending on family planning as “waste” because they object on
principle to these endeavors.
For these interviewees, the idea of “government waste” provokes a much more
substantial critique of government than annoyance over duplicative services or bureaucratic
delays. In the following section, I use survey data to examine whether the broader U.S.
population thinks about government waste in a comparable way to these interviewees.
A Broader Definition of Government Waste
Unlike in the case of foreign aid, I could find no existing survey that offered Americans
the opportunity to explain what they meant by government waste. So, working with an opt-in
panel from the survey firm Qualtrics, I conducted an online survey of 1,000 U.S. adults
November 5-19th, 2014. The population was balanced to match the U.S. adult population in
terms of gender, partisanship and education, and results are weighted to match the broader public
in terms of age, employment status, and ethnicity.5 Respondents were first asked to estimate
government waste: How many cents out of every tax dollar do you think the government wastes?
The mean answer was 53 cents, a result comparable to the Gallup survey question on
government waste.6 Respondents were then asked When you were thinking of government waste,
what specifically came to mind? They recorded their answers in an open-response text box.7
By far, the respondents’ most common definition of government waste category is simply
programs of which they entirely disapprove. For instance, a 68-year-old independent from
Illinois lists “Afghanistan, Iraq, the UN, Medicaid,” as examples of government waste. This
respondent’s list was relatively typical; military and humanitarian interventions overseas, along
with social welfare spending, were the most common types of specific spending deemed
wasteful. But other specific programs also came in for criticism. Another respondent, from New
10
York, sees “gay rights” and “abortion support” as wasteful. For a plurality of people, thinking
about waste appears to have provoked thoughts about entire programs of which one disapproves,
rather than efficiency concerns.
Far less common are complaints about waste as it is considered by policymakers –
inefficiencies like overpayment or redundancy. The survey respondents did make mention of
certain widely reported examples of government overpayment, like “$300 toilet seats,” as well as
occasional references to government having “too many employees,” “slow processes,” and
“duplicate services.” But inefficiency is a consideration for only 10% of respondents. When most
people talk about government waste, they are thinking of something very different from the
official definition. Public estimates of government waste are not, as Bok (1997) argues, primarily
a misperception of “inefficient administration.”
Moreover, there is a strong correlation between how one defines waste and the
percentage of government spending one sees as wasteful. As with the analysis of the foreign aid
survey data, I use a beta regression to model this outcome. Table 2 reports the factors correlated
with picking a higher estimate of waste. Model 1 provides a baseline for the demographic
factors. Republicans and conservatives think government waste is higher. Men and college
graduates tend to choose lower estimates. Model 2 includes those same control variables, but
adds variables for how the respondent described waste, including “disliked programs,”
“redundancy” and “overpayment.” I also include two additional variables, “partisan comment,”
which identifies the 5% of responses in which the respondent’s description of waste included an
explicitly partisan remark; and “perks” which refers to respondents’ considerations about the
benefits given to officeholders, such as “big fancy dinners for politicians.”
11
[Table 2]
All else being equal, a person who thinks of waste in terms of programs they dislike will
pick a 40% higher estimate of waste. Those thinking of waste in terms of overpayment pick
estimates about 30% lower. And, as Model 3 would suggest, this effect is concentrated among
those who have lower levels of education, the people least likely to have encountered elite
definitions of “government waste.” Thus public overestimation of government waste is due in
part to differing interpretations of the phrase.
Discussion and Conclusions
In two categories commonly used to describe federal spending, experts and amateurs are
talking past one another. A substantial portion of public overestimates of waste and foreign aid
can be explained by the fact that members of the public often define these terms more broadly
than policymakers do. Respondents thinking of foreign aid in terms of military spending pick
substantially higher estimates of foreign aid. Those thinking of government waste in terms of
programs they dislike think waste is higher than those thinking in terms of government efficiency
or administration. The impact of these broader definitions is concentrated among those with
lower levels of education, the people least likely to be familiar with elite definitions of waste and
foreign aid.
These results should not be taken to mean that Americans are fully informed, or even
well informed, on the subject of government efficiency or foreign policy. There is also much
more to be done to assess the dimensions driving public support for foreign aid (e.g. Boas, 2002)
and government trust (Hetherington, 2005). But these examples provide a cautionary tale for
those attempting to assess public policy knowledge. Researchers should be alert to the possibility
12
that what appears to be misinformation is in fact a miscommunication. More generally, scholars
should recognize and take seriously the substantive critiques of U.S. policy implicit in these poll
results. For instance, anger at “foreign aid” includes a substantial amount of opposition to U.S.
military interventions, and estimates of “government waste” includes meaningful opposition, not
to government processes but to major policy initiatives. Taking account of how the public
conceives of “foreign aid” and “government waste” allows us to extract a meaningful critique of
government policies from these estimates.
13
References
Achen, Christopher H. 1975. "Mass political attitudes and the survey response." American
Political Science Review 69(04): 1218-1231.
Bennett, Stephen Earl. 1988. “’Know-Nothings’ Revisited: The Meaning of Political Ignorance
Today.” Social Science Quarterly 69(2): 478.
Bøås, Morten. 2002. “Public attitudes to aid in Norway and Japan.” Center for Development and
the Environment, University of Oslo.
Bok, Derek. 1997. "Measuring the performance of government.” Pp. 55-76 in Joseph S. Nye,
Phillip D. Zelikow, and David C. King (eds.), Why people don't trust government.
Cambridge, MA: Harvard University Press.
Converse, Philip E. 1964. "The nature of belief systems in mass publics.” Pp. 75-169 in David E.
Apter (ed.), Ideology and Discontent.
Cribari-Neto, Francisco, and Achim Zeileis. 2010. "Beta Regression in R." Journal of Statistical
Software 34(2): 1.
Delli Carpini, Michael X., and Scott Keeter. 1996. What Americans don’t know about politics
and why it matters. New Haven: Yale University Press.
Duffy, Bobby and James Stannard. 2014. “Perceptions are not reality: things the world gets
wrong.” IPSOS Mori. 29 October.
Henry J. Kaiser Family Foundation. 2012. Kaiser US Role in Global Health Survey, February
2012. USPSRA.12GLOBEH. Princeton Survey Research Associates International
[producer]. Storrs, CT: Roper Center for Public Opinion Research, iPOLL [distributor],
accessed Oct-21-2015.
14
Henry J. Kaiser Family Foundation. 2013 Survey of Americans on the U.S. Role in Global
Health, August 2013. USPSRA.2013GHP005. Princeton Survey Research Associates
[producer]. Storrs, CT: Roper Center for Public Opinion Research, iPOLL [distributor],
accessed Oct-21-2015.
Hetherington, Mark J. 2004. Why Trust Matters: Declining Political Trust and the Demise of
American Liberalism. Princeton, NJ: Princeton University Press.
Hochschild, Jennifer, and Katherine Levine Einstein. 2015. Do Facts Matter?: Information and
Misinformation in American Politics. Vol. 13. Norman, OK: University of Oklahoma
Press.
Hudson, David and Jennifer vanHeerde-Hudson. 2012. “‘A Mile Wide and an Inch Deep’:
Surveys of Public Attitudes towards Development Aid.” International Journal of
Development Education and Global Learning 4(1): 5-23.
Kessler, Glenn. 2011. “Four pinocchios for the American public on the budget.” Washington
Post. March 3.
Ladd, Helen and Julie Wilson. 1983. ”Who Supports Tax Limitations: Evidence from
Massachusetts’ Proposition 2½.” Journal of Policy Analysis and Management 2(2): 256279.
Lau, Richard R., and David P. Redlawsk. 2001. "Advantages and disadvantages of cognitive
heuristics in political decision making." American Journal of Political Science 45(4):
951-971.
Lupia, Arthur. 2006. "How elitism undermines the study of voter competence." Critical
Review 18(1-3): 217-232.
15
Lupia, Arthur, and Mathew D. McCubbins. 1998. The democratic dilemma: Can citizens learn
what they need to know? Cambridge, UK: Cambridge University Press.
Manza, Jeff, Fay Lomax Cook, and Benjamin I. Page. 2002. Navigating public opinion: Polls,
policy, and the future of American democracy. New York: Oxford University Press.
Mc Donnell, Ida, Henri-Bernard Solignac Lecomte, and Liam Wegimont. 2002. “Public Opinion
and the Fight against Poverty.” Organisation for Economic Co-operation and
Development.
Popkin, Samuel. L. 1991. The Reasoning Voter: Communication and Persuasion in Presidential
Campaigns. Chicago: University of Chicago Press.
Riffkin, Rebecca. 2014. “Americans Say Federal Gov’t Wastes 51 Cents on the Dollar.” Gallup.
September 17.
Williamson, Vanessa. 2016. "On the ethics of crowdsourced research." PS: Political Science &
Politics 49(1): 77-81.
16
Intercept
College graduate
White
Party (more Republican)
Ideology (more conservative)
Income
Male
Agrees U.S. should take “the leading role in world affairs.”
1
-0.78
(0.10)
-0.44
(0.05)
-0.28
(0.05)
0.01
(0.02)
0.11
(0.03)
-0.02
(0.01)
-0.41
(0.04)
0.13
(0.02)
Describes foreign aid as military spending
*
*
*
*
*
*
2
-0.69
(0.15)
-0.45
(0.08)
-0.52
(0.07)
0.02
(0.02)
0.06
(0.05)
-0.02
(0.02)
-0.40
(0.07)
0.23
(0.04)
0.44
(0.10)
College graduate * foreign aid as military spending
*
*
*
*
*
*
3
-0.83
(0.15)
-0.37
(0.08)
-0.55
(0.07)
0.02
(0.02)
0.07
(0.05)
-0.01
(0.02)
-0.39
(0.07)
0.24
(0.04)
0.80
(0.13)
-0.83
(0.20)
*
*
*
*
*
*
*
Table 1: Factors Correlated with Foreign Aid Estimates
Beta regression of KFF 2012 survey, * = p<.05
17
(Intercept)
College graduate
White
Party (more Republican)
Ideology (more conservative)
Income
Male
“disliked programs”
“redundancy”
“overpayment”
“perks”
partisan comment
1
-0.25
(0.14)
-0.36
(0.09)
0.17
(0.08)
0.07
(0.03)
0.07
(0.03)
0.00
(0.01)
-0.26
(0.08)
*
*
*
*
2
-0.34
(0.14)
-0.31
(0.09)
0.17
(0.08)
0.07
(0.03)
0.07
(0.03)
0.00
(0.01)
-0.27
(0.08)
0.32
(0.08)
-0.24
(0.12)
-0.37
(0.15)
0.07
(0.10)
-0.06
(0.17)
*
*
*
*
*
*
*
*
“disliked programs” * College graduate
-0.39
(0.14)
-0.14
(0.11)
0.18
(0.08)
0.07
(0.03)
0.06
(0.03)
-0.00
(0.01)
-0.27
(0.08)
0.45
(0.10)
-0.25
(0.12)
-0.36
(0.15)
0.09
(0.10)
-0.06
(0.17)
-0.37
(0.16)
*
*
*
*
*
*
*
*
*
Table 2: Correlates of Respondents’ Estimates of Waste
Beta regression of Qualtrics 2014 survey, * p<.05
18
Notes
1
Details of the interview recruitment and interviewee demographic makeup are available in
Williamson 2016.
2
To protect their privacy, all interviewees are given pseudonyms.
3
Moreover, the survey had already primed respondents to think about the problems in
“developing countries such as those in Africa, Asia, and Latin America,” but did not mention the
Middle East, which may have discouraged respondents’ thoughts about war.
4
Including a variable for the respondent’s age has no substantive effect on these results. In
models run without a measure of ideology, the estimate on partisanship gets somewhat larger but
still does not reach statistical significance.
5
Pre- and post-weighting demographics compared to the U.S. population as a whole can be
provided as a technical appendix on request.6 Asked about the federal government, “How many
cents of each dollar would you say are wasted?” the mean response is 51 cents.
http://www.gallup.com/poll/176102/americans-say-federal-gov-wastes-cents-dollar.aspx
7
These categories are not mutually exclusive and respondents could offer as many examples of
waste as they wanted. Each complaint was counted, and about a quarter of responses were coded
in more than one category.
19