Hyde, Experiments in IR

PL18CH22-Hyde ARI 6 April 2015 10:8

ANNUAL
REVIEWS Further Experiments in International
Click here for quick links to
Annual Reviews content online,
including:
Relations: Lab, Survey,
Other articles in this volume
Top cited articles
Top downloaded articles
and Field
Annu. Rev. Polit. Sci. 2015.18:403-424. Downloaded from www.annualreviews.org

Our comprehensive search

Susan D. Hyde
Department of Political Science, Yale University, New Haven, Connecticut 06520;
Access provided by CAPES on 11/06/17. For personal use only.
email: [email protected]
Annu. Rev. Polit. Sci. 2015. 18:40324 Keywords

First published online as a Review in Advance on international relations theory, audience costs, bargaining, elites
February 2, 2015
The Annual Review of Political Science is online at Abstract

polisci.annualreviews.org
At conferences, at seminars, and on political science blogs, the potential util-
This articles doi: ity of experimental methods for international relations (IR) research contin-
10.1146/annurev-polisci-020614-094854
ues to be a hotly contested topic. Given the recent rise in creative applications
Copyright c 2015 by Annual Reviews. of experimental methods, now is a useful moment to reect more generally
All rights reserved
on the potential value of experiments to study international affairs, how these
inherently micro-level methods can shed light on bigger-picture questions,
what has been learned already, what goals are probably out of reach, and how
various research agendas in IR might productively incorporate experiments.
403
INTRODUCTION
At conferences, at seminars, and on political science blogs, the potential utility of experimental
methods for international relations (IR) research continues to be a hotly contested topic. Among
some IR scholars, the debate focuses on the difculty in establishing causal relationships. Because
experimental methods can be used most convincingly to test causality, they are held up by propo-
nents as a solution to persistent methodological problems in IR. For example, McDermott (2011)
describes the advantages of experiments to IR researchers: They can provide precise method-
ological control, unparalleled causal insight, and innovative theoretical clarication and direction
(p. 504). Yet among other scholars, recent enthusiasm for experimental research methods (partic-
ularly from the related political science subelds of comparative politics and American politics)
has been met with signicantly more skepticism, and the debate tends to focus on whether or
not experiments can be used (or used ethically) to study any meaningful questions central to IR.
Ulfelder (2014) summarizes this perspective in a blog post, regarding eld experiments specically:
Im skeptical that eld experiments will shed much light on many topics of interest to students of
international politics, mostly because I dont think those eld experiments will ever happen. . .I dont
see how researchers are going to create and reliably observe experimental and control groups for things
like war between states, participation in insurgencies, or protests against authoritarian regimes, given
the political sensitivity and ethical dilemmas involved.
Other concerns raised by IR scholars are that a focus on experiments will lead the subeld away
from questions of greater substantive importance (Mearsheimer & Walt 2013), that experiments
likely fail to lead to the accumulation of knowledge (Deaton 2010), and that experiments are a fad
that will pass along with other high-tech methods (Saideman 2013).
I will say upfront that although I am a proponent of expanding the use of experiments in IR
research, I do not think experiments can or should be used to answer all important questions in
the eld. However, given the recent creative applications of experimental methods, now is a useful
moment to reect more generally on the potential value of experiments to study international
affairs, how these inherently micro-level methods can shed light on bigger-picture questions,
what has been learned already, what goals are probably out of reach, and how various research
agendas in IR might productively incorporate experiments.
Too often, seasoned IR scholars reject experimental methods out of hand due to a lack of
understanding, a lack of thoughtful consideration about whether such methods could be useful
in their corner of IR, or both. There is also a great deal of inconsistency and/or confusion about
what can and cannot be learned from experimental research in IR, how micro-level ndings
might be integrated into broader research agendas, and the relative advantages of various types of
experimental methods, including lab, survey, and eld experiments.
To some extent, the eld of IR has already changed, and ndings from experimental research
have contributed signicantly to research agendas in several areas of IR, discussed below. Knowl-
edge about experimental methods has also changed. Before 2000 it would have been relatively
rare for a PhD program in political science to offer IR students methodological training in a
diverse array of experimental methods; now coursework and mentorship in lab, survey, and eld
experimental approaches are available to students in many leading programs.1 A growing number
of scholars include experiments as one potential methodological tool at their disposal and now
incorporate experimental research into their broader research agendas.
1
New textbooks are aimed at this audience (Gerber & Green 2012, Morton & Williams 2010). In some programs, training
for IR students in laboratory experiments has existed for much longer than training in survey and eld experiments.
404 Hyde
In a more general sense, the intellectual sands have also shifted such that experimental research
design, even when impractical for a particular research question, now serves as a hypothetical
benchmark when evaluating research design. Even when a researcher cannot realistically control
the randomization of variables with potentially important causal effectssuch as natural resource
endowments, the occurrence of foreign military intervention, participation in war, regime type,
military alliances, currency crises, or membership in international institutionsthe language of
experimentation can be useful for students and researchers to think through the empirical impli-
cations of an argument as if randomization of the explanatory variable of interest were possible
in the real world. For example, Gerring & McDermott (2007) recommend the adoption of an
experimental language for presenting case study research design. This eld experimental baseline
provides an important common language for discussing the strengths and limitations of methods
for causal inference.
Yet better integration of experiments into IR is still needed, and it will require a shift in
perspective by both researchers and consumers of IR research. This review focuses exclusively
on lab, survey, and eld experimental research in IR. All three methods are united by the fact
that the researcher controls the random assignment of relevant units in the study to treatment
and control groups. Natural experiments, quasi-experiments, and other related methods are not
discussed here (cf. Druckman et al. 2011, Dunning 2012, Morton & Williams 2010), though there
are clear connections between the discussion below and these methodological approaches. This
is also not the rst review on experiments in IR (McDermott 2002, 2011; Mintz et al. 2011), and
it builds on existing insights. In addition, because other scholars have made clear the potential
limitations of experiments, I do not repeat them here (for example, see Druckman et al. 2006,
2011; Humphreys & Weinstein 2009; King et al. 2011; McDermott 2002; Mutz 2011).
POTENTIAL VALUE OF EXPERIMENTS IN

INTERNATIONAL RELATIONS
Lab, survey, and especially eld experiments are still relatively rare in IR, and their scarcity may
lead to misconceptions about their potential value. As McDermott (2011) points out, [t]heir use
remains far from commonplace in international relations; a recent survey of international relations
scholars. . .found that only 4% of over 1,000 respondents used experimental methodology at all
(McDermott 2011, p. 503; see also Peterson et al. 2005).2 For IR experimentalists, one of the
biggest challenges stems from the type of questions the eld tends to explore. It is often difcult
to provide a convincing link between macro-level IR theories, some of which focus on the nature
of the international system itself, and experimental studies, which are most commonly carried
out on individuals or small groups. The most recognizable subjects in IR research are sovereign
states, of which there are now only about 200, and at the highest level of analysis there is only
one international system. Because experiments require sufciently large treatment and control
groups, the most common subjects of experimental research are units below the country level of
analysis. Thus, unless an experiment randomizes a treatment across many or all countries in the
international system, IR scholars who incorporate experiments into their research must somehow
connect micro-level ndings to macro-level questions. Squaring this circle is possible, at least in
many substantive areas of IR research, but the best ways of doing so are not widely agreed upon,
and there are few examples of research agendas to use as models.
2
This number has grown slightly since the rst Teaching, Research & International Policy (TRIP) survey, and it has ranged
from 4% to 8% depending on the year of the survey ( Jordan et al. 2009; Maliniak et al. 2007, 2012).
www.annualreviews.org Experiments in International Relations 405

The level-of-analysis problem has long been recognized in IR, although it has not yet been tied
explicitly to the challenges of integrating experimental research into IR. As Singer (1961) wrote,
[i]n any area of scholarly inquiry. . .the observer may choose to focus upon the parts or upon the
whole, upon the components or upon the system. He may, for example, choose. . .the owers or
the garden, the rocks or the quarry, the trees or the forest (p. 77).
It may go without saying that experimental analysis of the international system is much less
likely to be fruitful than the application of experimental methods to an empirical analysis of
its components. It is true that many of the so-called components of IR are difcult to imagine
in an experiment, particularly if the most important components are sovereign states dened
primarily by their relative power (Waltz 1979). Yet, if researchers move away from the billiard
ball assumption in IR, and instead focus on agents within states or on transnational or nonstate
actors (Findley et al. 2013c), as many contemporary areas of IR already do, there are at least four
distinct ways of incorporating experiments into broader research agendas in IR, each of which
has different strengths and weaknesses. The four approaches are (a) assuming that experimental
subjects (such as university students) will behave like relevant actors in IR, (b) using elites as
subjects in lab or survey experiments, (c) focusing experiments on existing IR theories that explicitly
involve public opinion or mass behavior, and (d ) building experiments into a broader multimethod
approach so that they test clear empirical implications of specic midrange theories.
All four of these approaches are united by the idea that many research agendas in IR can
progress by better linking macro-level theories to their micro-level implications and then testing
those implications with methods that maximize researchers ability to make causal claims. Pro-
viding convincing causal tests is not the only goal of IR research, but such tests are currently
underprovided in the eld. In some research agendas, such testing of micro-level implications can
also make research more clearly relevant to policy, particularly with eld experiments, as I discuss
below. This shift will be easiest for scholars who work in substantive areas that lie at the inter-
section of IR and comparative politics or who consider domestic politics, subnational variation,
or transnational actors to be relevant to their specic research questions. Making experiments
more relevant in IR will require that many bigger-picture theories in IR be more fully specied at
the micro level and that IR experimentalists work harder to link their ndings to bigger-picture
theories and to adjudicate more explicitly between competing hypotheses, many of which will be
generated at a higher level of analysis.
Assuming Subjects in the Lab Are Like Relevant IR Populations

The longest-standing tradition of IR experimentalism has grown primarily out of laboratory ex-
periments and involves the assumption that experimental subjects, who are more easily accessible
to researchers than are subjects in a eld setting, will behave in some meaningful way like non-
experimental subjects who are direct participants in international affairs, foreign policy decision
making, or other activities of interest (Schelling 1961). For example, if scholars can learn why
individuals are more or less likely to cooperate in a controlled laboratory setting, such ndings
may be informative to scholars thinking about the conditions under which states, armed groups,
or rms are more likely to cooperate (McDermott 2011). Just as importantly, if scholars cannot
replicate important theorized dynamics of decision making or rational behavior on any subjects or
under any conditions in highly controlled lab settings, then such dynamics probably do not exist
in the real world.
Many lab experiments examine concepts that are highly relevant to IR, such as decision-making
behavior (Boettcher 2004, Geva et al. 2000, McDermott et al. 2002, Mintz et al. 1997, Moxnes &
Van der Heijden 2003, Redd 2002), threat perception (Kemmelmeier & Winter 2000, Rousseau &
406 Hyde
Garcia-Retamero 2007), reputation (Renshon 2015), gender differences in foreign policy opinions
(McDermott & Cowden 2001), support for war or intervention (Beer et al. 1995; Berinsky &
Kinder 2006; Gartner 2008, 2011; Van der Heijden & Moxnes 2013), bargaining (Dickson 2009),
and cognitive biases (Boettcher 1995, 2004). Some of this research assumes that the empirical
implications of formal models of state behavior can be tested on individuals in a lab who pretend to
be states or other aggregate actors. Examples of such research relating to terrorism are reviewed by
Arce et al. (2011). They include global security games, in which individual lab subjects imagine they
are countries and choose whether to invest in productive activities which increase their GDP,
preemptively invest in attacking the terrorist directly, or invest in protective measures for their
own population (p. 377). These lab games are intended to approximate decision making within a
national bureaucracy, but for skeptical readers of experimental research, the articiality of the lab
settingespecially when subjects pretend to be aggregate actors such as states, bureaucracies, or
militariesmakes it hard to rule out the possibility that such games are not useful approximations
of real-world situations.
Although such laboratory experiments offer numerous advantages, including the ability to
control many, if not all, aspects of the study, the central criticism of this type of research is well
known: The external validity of such experiments can be seriously questioned, and the assumption
that undergraduates or other laboratory subjects will behave like decision makers in the real
world, or that the nuances or the stakes of the real world can be effectively recreated in the lab,
may lead to faulty conclusions. To be clear, undergraduates and other easily accessible lab subjects
are not necessarily problematic (Druckman & Kam 2011), but this is ultimately an empirical
question, and in the absence of comparisons to other subject populations of interest, laboratory
experiments may be less persuasive. Scholars including Mintz et al. (2006), Hafner-Burton et al.
(2014), Gray & Hicks (2014), and Renshon (2015) have compared undergraduates and other more
realistic subject populations, such as military ofcers, in parallel laboratory experiments on foreign
policy opinions, and they have found signicant differences in their behaviors. More generally,
there is often a trade-off between the ease with which an experimental researcher can access an
experimental population and the degree to which that subject population is directly relevant to
IR. Challenges in linking the behavior of individuals to that of aggregate actors such as states are
not unique to experiments: A similar debate arose around the growth of rational-actor models in
IR (e.g., Kahler 1998) and the assumption that both states and individuals will behave rationally.
There are conditions under which it is less problematic to assume that laboratory subjects
will behave like relevant actors in international affairs. Whereas it might be controversial to
assume that undergraduates reasonably approximate state bureaucracies in a laboratory setting, it
appears far less controversial to assume that some characteristics of human cognition and decision-
making ability are more or less common across individuals, perhaps varying by degree rather than
kind. If these characteristics are (or can be assumed to be) present across all individuals and
contexts, researchers need not experiment directly on the relatively small population that is active
in the practice of IR, such as national leaders or military commanders, in order to learn about,
for example, their likely reaction to increased perceptions of risk or their ability to incorporate
new information. Studying elites directly is difcult. As Hafner-Burton et al. (2013) describe the
problem, experienced elites are difcult to obtain as subjects because they are generally busy, wary
of clinical poking, and skittish about revealing information about their decision-making processes
and particular choices (p. 368).
Thus, laboratory experiments have the greatest comparative advantage over other experimental
methods when they can show that some characteristics of human psychology or cognition are con-
stant across individuals, or when they can demonstrate that particular IR theories assume a model
of human cognition that is clearly not constant across individuals and thereby challenge existing

theories (Hafner-Burton et al. 2013). For example, in a review essay Rathbun (2009) discusses the
potential for social psychology and experimental studies of trust to inform IR, theorizing about
the prospects for international cooperation and the existence of enduring rivalries. Laboratory
experiments to study trust reveal that most people are inherently more cooperative or competi-
tive in their behavior across a variety of contexts and respond in systematically different ways to
various strategic settings. Rathbun (2009) argues that these differences line up well with IRs three
major research traditions: structural realism, neoliberal institutionalism, and constructivism. All
individuals (citizens, scholars, and foreign policy elites) may therefore carry a crude paradigm of
IR in their minds (see also Kertzer & McGraw 2012), which is likely to condition their views of
the potential for conict and cooperation as well as the likelihood that they will be able to avoid
conict in a given setting.
As a number of scholars have argued, there is still more potential for a productive exchange be-
tween social psychology and IR, partly based on laboratory experiments about individual variation
in the human propensity to trust and cooperate (Goldgeier & Tetlock 2001, Hafner-Burton et al.
2013, McDermott 2011, Rathbun 2009). To the extent that IR theories require some understand-
ing of human cognition, these experiments represent an important and ongoing contribution.
One of the clearest areas in which IR theory has already been strongly inuenced by exper-
imental ndings from the lab is prospect theory (Boettcher 1995, 2004; Kahneman & Tversky
1979; Levy 1997; McDermott 2004; Tversky & Kahneman 1992). Developed in economics and
referenced widely in IR, prospect theory suggests that individuals place a disproportionate value
on losses from the status quo relative to equal gains from the status quo. By clearly outlining
a theory of individual decision making and conditions under which the assumptions underly-
ing rational-actor models may be problematic, ndings from laboratory experiments in this area
highlight the potential for IR theory to be informed by a better understanding of human cogni-
tion. And although the loss aversion that drives prospect theory may be common across nearly
all individuals, some experimental results from economics show that, compared to nonelite deci-
sion makers, experienced decision makers are less prone to loss aversion and may actually behave
more like rational decision makers on this dimension (e.g., List & Mason 2011; see discussion
in Hafner-Burton et al. 2013). Thus, making the assumption that readily accessible laboratory
subjects behave similarly to relevant international actors can be useful under specic conditions,
but it remains a somewhat risky methodological bet (Lake & Powell 1999, p. 32).
An important issue associated with IR lab experiments is viewed by proponents as one of their
chief advantages: In a laboratory, most elements are controllable, including the environment.
However, such ne-grained control may make studies unrealistic. If the context in which decisions
are made is important in IR, then even if humans share some fundamental characteristics relevant
to decision making, it would be necessary to replicate in the lab the complicated world of foreign
policy decision making because getting the context right may be just as important as getting
the sample population right. Perfectly imitating the context of international affairs in the lab may
never be possible; however, the realism problem may be at least partially addressed by experimental
research that directly includes elites or relevant subject populations in the lab, in surveys, or in
the eld, as discussed in the next sections.
Bringing Experiments to Elites

A second way in which experiments can better connect to bigger-picture IR theories is by engag-
ing in lab or survey experimentation on elites or other IR-relevant subject populations directly.
Berinsky (2007), Mintz (2004), and Mintz et al. (1997, 2006) provide examples of lab experiments
that focus on military elites. Hafner-Burton et al. (2014) have conducted survey experiments on
408 Hyde
elites and on undergraduates, examining how personality traits such as patience and strategic rea-
soning inuence their preferences for international legal agreements. Herrmann et al. (2001; see
also Herrmann & Shannon 2001) have run a survey experimental study on samples of US elites
(broadly dened) and the general population to evaluate attitudes toward trade, including hypothe-
ses about relative gains, distributive justice, and international trade institutions.3 Other scholars
approaches have begun to move in this direction, such as Dietrichs (2014) use of interviews with
development aid ofcials about the way donors allocate aid within recipient countries. Renshon
(2015) has conducted laboratory experiments on political, military, and economic elites enrolled
in the Harvard Kennedy Schools Senior Executive Fellows Program. Incorporating elites into
experiments in IR is an important step. Scholars should continue to clearly justify why their chosen
population of elites is appropriate for their study, ensuring that the sample population is not just
dened as elite but that it incorporates the particular elites most relevant to theory testing.
In an innovative set of eld experiments, Nielson, Findley, and coauthors have collected con-
tact information, primarily email addresses, for subject populations relevant to different topics
in transnational relations and have used this universe as a sampling frame for conducting experi-
ments in which randomly varied (but real) treatments are administered via email to the relevant
subject population (Brigham et al. 2014, Findley et al. 2013ac). These studies have yielded im-
portant ndings about money laundering, factors that inuence compliance with international
law, desirability of various types of foreign investment in the United States, and the willingness
of nongovernmental organizations (NGOs) to learn from evidence. Additionally, in a set of eld
experiments in Uganda, Findley et al. (2014a), Milner et al. (2014), and Blaschke et al. (2014) used
survey experiments with some behavioral outcomes to study both mass and elite attitudes toward
foreign aid.
Research agendas in IR that incorporate experiments with elite subjects remain in the early
stages, and it is too soon to draw overarching conclusions. But these studies have already made it
clear that focusing directly on elite decision makers is possible in lab, survey, and eld experiments;
that such studies have signicant potential; and that they can yield conclusions that are distinct from
the ones afforded by lab experiments alone. There are, by denition, relatively few elites available
for participation in studies, and their participation may carry greater opportunity costs than those
borne by regular citizens. Even so, when a particular research agenda is concerned with behaviors
or decisions that may be unique to elites or experts in IR, experimenting on undergraduates or the
general population and then generalizing the ndings to elites may carry signicant risks of being
wrong, although in some cases the ndings can be quite similar (Hafner-Burton et al. 2013, 2014).
Thus, for a subset of research agendas that focus in part on elite decision making, conducting
experimental research directly with elites is one way to make experiments in IR more realistic
and potentially more relevant; however, this is far from guaranteed, because the decision-making
context may be just as important as the subject population.
Focusing on Public Opinion or Mass Behavior in IR

A third way to increase the bigger-picture relevance of IR experiments is to focus experimental
methods on IR theories that explicitly incorporate public opinion or have clearly observable im-
plications for mass behavior. In this sense, survey and eld experiments can target the populations
relevant to specic theories directly rather than assuming that the relevant population behaves like
a convenience sample in the lab. For example, many international political economy (IPE) theories
3
US elites were recruited from the list generated by Holsti & Rosenau (1984, 1993).

have implications for individual and group behavior within states, such as how individuals may re-
act to globalization or who is most likely to support trade liberalization and immigration (Ardanaz
et al. 2013, Ehrlich & Hearn 2014, Hainmueller & Hiscox 2010, Hiscox 2006, Margalit 2012).
Another area of IR research suggests that citizens can constrain their leaders international
behaviors and decision making (Fearon 1994), particularly in democracies. This assumption is
prominent in theories focusing on military crises, the democratic peace, alliances, economic
sanctions, foreign trade, foreign direct investment, monetary commitments, interstate bargaining,
and international cooperation more generally (Tomz 2007, p. 821). Testing the existence and
manipulability of so-called audience costs is one area in which experimental methods have already
provided important causal leverage in evaluating the empirical implications of existing theories
and in adjudicating among competing hypotheses where observational research has fallen short.
Even more importantly, experimental results concerning audience costs have begun to feed back
into renements of the original theories, particularly by suggesting how political elites can shape
(and sometimes escape) audience costs during potential crises rather than be constrained by them
(Davies & Johns 2013, Guisinger & Saunders 2013, Levendusky & Horowitz 2012, Trager &
Vavreck 2011).
Observational research can show, for example, that democracies and autocracies vary system-
atically in their international behaviors in a manner consistent with audience costs, but it is far
less convincing in showing that audience costs actually cause the difference in behavior. Many
other differences correlate with variation in regime type. Some inuential experimental research
in this area uses laboratory experiments with primarily undergraduate subjects to explore how
subjects respond to variations in the framing of foreign policy problems (Berinsky & Kinder 2006;
Boettcher & Cobb 2006, 2009; McDermott et al. 2002; Pronin et al. 2006); this approach is
based on assumptions about experimental subjects similar to the ones discussed above (see section
entitled Assuming Subjects in the Lab Are Like Relevant IR Populations). Yet, as already men-
tioned, convenience samples may differ in important ways from representative populations within
countries, and such differences could undermine the external validity of these ndings.
In part to address these issues, Tomz (2007) uses survey experiments to study audience costs,
changing the subject population from undergraduates to a representative sample of citizens in the
United States. This shift from laboratory to survey experiments increases the external validity of
the study and makes the connection to IR theory clearer because the study deals directly with the
relevant theoretical populationdomestic audiences. By randomizing the presentation of infor-
mation to survey respondents, Tomz (2007) convincingly shows that audience costs exist across a
wide range of conditions (p. 836) within the United States and also that these audience costs vary
depending on whether the conict escalates, are greater among certain types of respondents, and
likely arise because respondents are concerned about the international reputation of the country
and its leaders (p. 836). The focus on survey experiments to evaluate audience costs has sparked
a productive collection of articles that together are beginning to provide a more nuanced picture
of the role that public opinion can play in constraining a countrys international behavior as well
as the role that elites can play in shaping public opinion.
Levendusky & Horowitz (2012) follow up on Tomzs investigation of audience costs by focusing
on variation in domestic political conditions. They show that presidents who back down in a foreign
policy crisis are less likely to be punished in the court of public opinion when they explain why
they backed down. Davies and Johns use a similar setup but extend the analysis from the United
States to Great Britain and add an investigation of whether domestic audience costs depend on
the type of crisis (Davies & Johns 2013) or on the dominant religion in the target state ( Johns
& Davies 2012). Trager & Vavreck (2011) use survey experiments to test how US presidential
approval ratings change during a hypothetical crisis-bargaining scenario, and they nd support for
410 Hyde
several common assumptions in the crisis-bargaining literature, most importantly that audience
costs in democracies can be extremely high, potentially driving leaders to participate in conicts
they are unlikely to win. They also nd, like Levendusky & Horowitz (2012), that the magnitude of
audience costs can be mitigated by presidential rhetoric and other aspects of the domestic political
context.
With reference to the democratic peace, Tomz & Weeks (2013) also use representative public
opinion polls in the United States and the United Kingdom, and they nd experimental support
for the theory that the democratic peace is driven in part by public reticence to go to war with other
democracies. Desposato et al. (2013) compare whether citizens in a newer democracy (Brazil) and
an autocracy (China) differ in their willingness to support a war against another democracy. They
nd that the two countries are more similar than anticipated in the degree to which individuals
are willing to support going to war against a democratic state, and that public aversion to going to
war against democracies may be more universal than previously thought. Several related survey
experimental studies have focused on other factors that affect public support for war and other uses
of force (Berinsky 2007; Boettcher & Cobb 2006, 2009; Gartner & Gelpi 2012; Gelpi 2010a,b;
Grieco et al. 2011; Herrmann et al. 1999; Horowitz & Levendusky 2011; Tingley & Tomz 2012;
Wallace 2013, 2014).

Most survey experiments on audience costs and other related phenomena to date utilize repre-
sentative samples of the US population or the US voting population. Depending on the research
question and on whether the results will be generalized to all citizens in a particular country,
representative samples of the population may not be strictly necessary for survey experiments,
and new techniques for recruiting convenience samples for surveys, such as Amazon.coms Me-
chanical Turk, provide very low-cost access to experimental populations for a signicantly wider
range of researchers. Such samples may or may not be meaningfully different from representative
samples for many types of survey experiments (Berinsky et al. 2012). Using a convenience sample
for a survey experimental study is not necessarily a problem for IR researchers (e.g., Chaudoin
2014), but it may push the research closer to the laboratory studies with experimental subjects
discussed above (see section entitled Assuming Subjects in the Lab Are Like Relevant IR Popula-
tions), because the most relevant subject population is not necessarily studied directly. Authors of
studies that rely on such Internet-based convenience samples rather than representative samples
of the theoretically derived subject population should use caution in generalizing their ndings
to broader audiences and should pay careful attention to how convenience sample properties may
interact with treatment conditions (Tingley 2014).
Public opinion is important to theories outside of audience costs and the democratic peace. A
relatively new research agenda uses survey experimental methods in much more difcult contexts to
study counterinsurgencyfor example, the willingness of Afghan citizens to participate in wartime
informing on insurgent groups (Lyall et al. 2013b), support for insurgents or international forces
in Afghanistan (Hirose et al. 2014), and support for militant group policies in Pakistan (Blair et al.
2013, Fair et al. 2014, Lyall et al. 2013a). Many of these studies use survey experimental techniques
to measure sensitive opinions rather than to adjudicate between competing hypotheses. Because
the bigger-picture theories motivating these studies include the general public as a relevant actor,
survey experiments conducted on representative samples directly target the appropriate population
in a subset of countries.
Of course, public opinion plays no role in many IR theories, and scholars considering survey
experiments should pay careful attention to whether public opinion is actually relevant to their
theories. To be clear, the relevance of public opinion should be predicated on specic hypotheses
or empirical implications rather than on generalizations by issue area. Simply conducting survey
experiments that mention foreign policy is insufcient to make experimental results relevant to

bigger questions in international affairs. For example, public opinion does not play a signicant
role in most theories about nuclear weapons proliferation, and in most cases a survey experiment
that referenced nuclear weapons proliferation would not be testing micro-level implications on
theoretically relevant populations.4
Two studies by Wallace (2013, 2014) provide an excellent example of how identifying the
relevant subjects for experimental work makes the theoretical contribution signicantly greater.
Wallace focuses on international law and the use of torture. In one article, the author is primarily
interested in how public opinion about torture can be inuenced by reference to international
law (Wallace 2013). In this study, a random sample of American adults is directly relevant to the
research question. In a second study, Wallace (2014) focuses on the role played by military expe-
rience in conditioning attitudes toward torture and international legal conventions about torture.
For this second research question, the relevant population includes individuals with military expe-
rience, and Wallace successfully recruited two subject populations to make comparisons between
civilians and veterans.

Thus, scholars should pay careful attention to whether public opinion is relevant to their re-
search questions, and therefore whether survey experiments are likely to be useful tools. Because
only a subset of theories in IR explicitly incorporate public opinion, an extreme focus on survey
experiments in IR could lead researchers away from a number of potential explanations that are
not amenable to experimental testing. Awareness of this potential bias is important, and it is part
of a broader recommendation, discussed below, to consider experimental ndings in their appro-
priate theoretical context. For hypotheses that incorporate public opinion, survey experiments are
probably low-hanging fruit and an important empirical focus in future research.
As literature on the role of public opinion in IR continues to develop, attention to whether
differences in self-reported attitudes translate into differences in behavior (such as voting, political
participation, and economic decision making) will be important. It will require that researchers
focus on behavioral indicators, whether by developing eld experiments (e.g., Hainmueller et al.
2014), by using existing ndings to motivate eld experimental studies, or by using behavioral
indicators as outcome measures within survey experiments in addition to self-reported behavior
(e.g., Bursztyn et al. 2014, Findley et al. 2014a). Across all of these possibilities, IR experimentalists
should maintain careful links between bigger-picture theories, theoretically relevant populations,
and observable implications of competing explanations.
Theory First, Experiments Second

The fourth and nal recommendation to strengthen links between bigger-picture IR theories
and experimental research methods does not really constitute a separate approach but rather a
distinct way of viewing the contribution of experiments. The value of an experiment relative to
a particular research agenda in IR can become clear when lab, survey, and/or eld experiments
are integrated into a broader project in which the observable implications of a theory are clearly
spelled out; when experiments are clearly connected to these observable implications; and when
experimental ndings are replicated across theoretically relevant contexts. This way of examining
the contribution of experiments to bigger questions in IR has the most untapped potential, but it
also requires a signicantly longer time horizon, and is perhaps the most difcult to t into a short
journal article. It requires signicant additional effort on the part of the researcher to link bigger-
picture theory with appropriate empirical tests (see Berinsky 2009, Findley et al. 2014b). It may also
4
But see Press et al. (2013) for relevant research on nuclear weapons that considers the role of public opinion.
412 Hyde
require a change in perspective among consumers of IR research such that experiments are viewed
more systematically within their broader empirical and theoretical context. The successful use of
experiments to test empirical implications of bigger-picture theories may also require replication
of experimental studies across theoretically relevant contexts, as the audience costs literature has
begun to do.
For example, in my own research on international election observation, I was interested in
explaining why inviting foreign election observers became an international norm. I developed a
theory of unintended norm formation in which signaling behavior on the part of states (i.e., inviting
foreign observers) can generate a new international norm (Hyde 2011). One of the important
empirical implications of this theory is that the presence of international election observers should
be costly for governments that invite observers and cheat in front of them. Yet providing convincing
cross-national evidence for this implication was problematic for a number of reasons, including
potential bias in the type of countries that invite international observers. A focus on within-country
variation and random assignment of international observers to polling stations offered one way
around this problem and allowed the study to show that international election observers can cause
a measurable reduction in the (fraudulent) vote share of the incumbent candidate (Hyde 2007,
2011). A separate study showed that no such observer effect was apparent in an election that
was widely viewed as clean (Hyde 2010a, 2011). As stand-alone articles, the experimental studies
of international election observation appear theoretically much narrower, showing for example
that international observers can reduce election day fraud. Within the context of a larger project
(Hyde 2011), however, the experiments (one of which was admittedly a natural experiment) provide
much-needed causal leverage on a hypothesis that would be difcult to test convincingly using
other methods: that inviting foreign observers and cheating in front of them is costly to leaders.
To date, eld experiments are by far the least common type of experiment in IR. Their strength
lies in their realism and their ability to introduce causal leveragethrough randomizationto
the study of complex real-world phenomena. Problems of external validity still exist, but they
are substantively different from those that characterize lab and survey experiments. Whereas
a lab or survey experiment might not usefully approximate any real-world phenomenon, eld
experiments, if conducted properly, represent at least one real-world situation. Questions about
external validity in eld experiments center on which (if any) real-world situations are sufciently
similar to the context in which the eld experiment was conducted. Replication across theoretically
important contexts provides an empirical response to questions about external validity, though
such replications are still nearly unheard of for IR eld experimentation.
Field experimental studies in IR typically involve randomized interventions below the state
level or randomization of treatments across nonstate actors such as NGOs or transnational rms
(Findley et al. 2013c). However, eld experiments need not take the analysis all the way down to the
individual level, as some scholars assume (Pepinsky 2014). Alongside lab and survey experiments,
as well as other methods, eld experiments have the potential to enhance current understandings
of the micro-level implications of a number of IR theories.
In outlining the observable implications of specic theories, scholars should focus on whether
any important aspects of their theories might be tested experimentally. To the extent that a complex
reality may be an important part of the theoretical context, eld experiments can be usefully
considered part of a research agenda. Readers of eld experimental research should exercise caution
in judging whether a single study is important, for rarely will a single eld experiment provide
the denitive word on a subject. It can show, often beyond the shadow of statistical doubt, that
a particular causal relationship can exist, at least in one place at one moment in time, but it has
greater difculty showing the conditions under which a particular causal relationship is likely to
hold. Thus, IR eld experiments should be viewed in the context of a broader research agenda,

and are more substantively meaningful if they are replicated across time and theoretically relevant
contexts (which is admittedly not possible for historical phenomena). In some areas it will also be
informative to compare eld experimental results with relevant survey and observational studies
(Findley et al. 2013b).
To illustrate, one area of IR research focuses on the efcacy of international organizations
(IOs) and their ability to carry out their stated goals, such as the ability of the World Bank or the
United Nations to facilitate postwar reconstruction and economic recovery. There are dozens of
eld experiments, carried out primarily by economists, on the efcacy of postconict intervention
and of programs funded by the United Nations or the World Bank (e.g., Avdeenko & Gilligan
2014; Beath et al. 2012, 2013; Blattman & Annan 2014; Blattman et al. 2014; Fearon et al. 2009;
Gilligan et al. 2014; Humphreys & Weinstein 2009; Mvukiyehe & Samii 2013). However, many
of these studies are authored outside of IR and have not necessarily been read by IR scholars,
creating an opportunity for researchers to articulate how these and related studies shed light on
theories about the efcacy of IOs (Hyde 2010b).

There are other areas of research in which existing experimental ndings could be connected
to bigger-picture IR theories, such as ongoing research on the diffusion of innovation, the spread
of disease, migration, consumer behavior, and ethnicity. However, the task of connecting these
theories, their observable implications, and existing experimental ndings requires additional work
and a change in perspective that remains relatively rare in the subeld.
WHAT HAVE WE LEARNED ALREADY?

IR research using experimentsespecially eld experimentsis just getting started, and it is
too early to judge the utility of experiments as an IR method in any general sense. Yet some
research agendas have already yielded important insights that may prove to be instructive models,
particularly when compared to what could be learned from observational studies alone.
To the extent that human cognition and psychology are important for understanding IR,
laboratory experiments have much to offeralthough as Hafner-Burton et al. (2013) argue, lab and
survey experiments on elites, in addition to the relatively large number of studies already conducted
on university students and other easily accessible populations, are an important and underprovided
part of this research agenda. Areas in which individual lab experiments have been particularly
important for IR research include threat perception, evaluation of risk, framing, deterrence, and
bargaining behavior more generally, as discussed above (Goldgeier & Tetlock 2001, Hafner-
Burton et al. 2013, McDermott 2004, Mintz et al. 2011).
As also summarized above, the literature on audience costs has undergone something of a revival
due in part to recent survey experimental work. Not only has recent work provided important
causal leverage on empirical tests of several components of audience cost theories, but some studies
have also contributed to further renement of the theories. Questions about whether and how
political elites are able to manipulate public opinion in their favor are crucial for understanding any
potential constraining effect of audience costs on the behavior of leaders (Guisinger & Saunders
2013)including work on media framing (Gelpi 2010a) and presidential justications for crisis
decisions (Levendusky & Horowitz 2012, Trager & Vavreck 2011), or research analyzing whether
shifts in public opinion based on security or trade policy decisions are sufcient to bring about
real electoral consequences (Saunders 2013).
A quickly developing research agenda examines the consequences of efforts by international
actors to change politics or policies within sovereign states. Observational research can show
that international efforts to bring about change in particular countries are correlated with a vari-
ety of positive and negative consequences, but it faces signicant barriers in demonstrating that
414 Hyde
international intervention actually causes any intended or unintended consequences. In most ob-
servational studies of international intervention, any observed correlation may always be spurious
because some omitted variable may cause both the intervention and the changes in the outcome
of interest. Corstange and Marinov use survey experimental methods to evaluate how foreign
publics respond to efforts by outsiders to inuence their internal elections, either by favoring one
political party or candidate or by supporting democratic elections (Corstange & Marinov 2012,
Marinov 2013). Hyde & Lamb (2013) use eld experimental methods, in collaboration with an
international NGO, to show that in the authoritarian electoral context of Cambodia, international
democracy promotion programs aimed at rural villages are successful in changing citizen willing-
ness to engage in democratic civic action that requires individual agency. However, such programs
do not convince citizens that Cambodia is democratic or that Cambodian political institutions will
be responsive to their needs (Hyde & Lamb 2013).
In a separate but related area of research in security studies, scholars have begun to examine the
micro-level implications of various components of counterinsurgency and international efforts to

reduce civil conict. The most common tools are survey experiments or lab-in-the-eld exper-
iments on populations of interest, including citizens in countries such as Afghanistan, Pakistan,
and Uganda (Beath et al. 2012; Berman et al. 2014; Blair et al. 2014; Fearon et al. 2009; Lyall et al.
2013a,b).
Other experimental studies have begun to explore the political effects of foreign aid (Beath
et al. 2012, 2013; Dietrich & Winters 2014; Findley et al. 2014a; Milner et al. 2014). These have
shown that foreign aid can help improve social cohesion and repair mistrust following conict
(Fearon et al. 2009), that branding foreign aid may not always improve public attitudes toward
donors (Dietrich & Winters 2014), and that knowledge of who pays for public goods projects may
inuence elites and elite support for those projects (Findley et al. 2014a, Milner et al. 2014). All
these ndings carry important implications for the political effects of foreign aid and the efcacy
of aid programs.
Several recent studies use experimental methods, including some eld experiments, to examine
international law. They have begun to evaluate which factors may increase or explain compliance
with international law across a variety of issue areas (Findley et al. 2013b,c; Jensen & Malesky
2013; Putnam & Shapiro 2013; Tingley & Tomz 2012, 2014; Tomz 2008; Tomz & Weeks 2012;
Wallace 2013, 2014). Chilton & Tingley (2014) argue that international law is set to benet from
more experimental research because selection problems are so acute in the use of observational
data and because understanding the causal effect of laws on behavior is so important for the eld.
Within IPE and comparative political economy, a number of lab, survey, and eld experimental
studies have made important contributions by testing micro-level implications ( Jensen et al. 2014,
Pepinsky 2014, Tingley 2014). For example, scholars have used survey experiments to investigate
public preferences toward trade, particularly how framing affects public attitudes (Hiscox 2006),
how socioeconomic traits inuence sensitivity to such framing effects (Ardanaz et al. 2013), and
why publics in advanced industrial economies are willing to tolerate high levels of agricultural
protectionism (Naoi & Kume 2011). Other political economy studies have used experiments to
answer questions related to the individual consequences of globalization (Ehrlich & Hearn 2014,
Hainmueller & Hiscox 2010, Margalit 2012), attitudes toward migrants during economic crises
(Goldstein & Peters 2014), foreign direct investment (Findley et al. 2013a, Gray & Hicks 2014,
Jensen et al. 2013), and consumer response to fair labor and environmentally conscious labeling
(Hainmueller & Hiscox 2012a,b; Hainmueller et al. 2014). Because many prominent questions
in IPE involve clear theoretical expectations for individual and group behavior, there is excellent
potential to further connect the future use of experiments in these and related research questions
to bigger-picture theories (e.g., Hainmueller & Hiscox 2010). Recent experimental work has

also inspired methodological innovation. Dafoe et al. (2014) have shown that the experimentally
manipulated survey primes about democracy versus nondemocracy used to study audience costs
and the democratic peace may be fundamentally inuenced by other parts of the vignettes and
therefore that the magnitude of audience costs may be underestimated. Facing the difculty of
measuring public opinion in experimental work, other scholars have developed new methods of
measuring attitudes about sensitive topics, such as support for insurgents in the midst of violent
conict (Blair et al. 2014, Bullock et al. 2011).
The above examples are not intended to constitute a comprehensive review of all that has been
learned to date from experiments in IR but rather to illustrate several clusters of experimental
research in which themes have begun to develop and in which the connection to bigger-picture
theory is reasonably clear.
WHAT IS THE INTERNATIONAL RELATIONS FIELD

UNLIKELY TO LEARN?
Fundamental questions about what qualies as IR theory and how empirical research can best
evaluate important questions underlie critiques about the broader theoretical relevance of exper-
iments in IR. This debate is well outside the scope of this article, yet it is important to make a
few points that are relevant to experiments potential. One common misperception is that exper-
iments connect to bigger-picture IR theories only if they somehow test the leading paradigms
in IR against one another. For example, in what is otherwise an excellent study, Findley et al.
(2013c) characterized their experimental treatments as realist, liberal, and constructivist in
an effort to connect their study of international law and anonymous incorporation to IR theory.
Such micro-level evidence is not relevant to the debates between the research traditions, and it
distracts from the more important connections made in their study regarding international law
compliance and the problems of money laundering and shell corporations.
More generally, as many prominent scholars continue to advocate, the portrayal of IR theory
as a debate between isms (or paradigms or research traditions) is not constructive (e.g., Bennett
2013, Lake 2011, Sil & Katzenstein 2010). Continued portrayal of IR as an interparadigm debate
that ends up producing self-afrming research and then wag[ing] theological debates between
academic religions (Lake 2011, p. 465) is unlikely to teach much about how the world works or to
contribute to the scientic study of IR. Thus, for scholars interested in connecting experimental
research to IR theory, simply referencing a few of the isms is insufcient, and shoehorning research
questions or hypotheses into the various research traditions is counterproductive and undervalues
the advantages of experiments.
Thankfully, although the interparadigm debate has continued to limp along (Bennett 2013,
p. 460), most contemporary IR research has moved beyond arguing about assumptions that char-
acterized this third Great Debate (see Lake 2013, pp. 57071) in IR and has been productively
replaced with analytic eclecticism (Sil & Katzenstein 2010, p. 411), problem-driven research,
and contingent, midlevel theories of specic phenomena (Lake 2011, p. 466). Although none of
these approaches to IR requires experimental research methods, experiments can t well into these
alternative conceptions of what IR theory can look like and how it can be most useful. Experimen-
tal methods, and eld experiments in particular, have the greatest potential to contribute to IR
when they are connected to the midlevel theories discussed by Lake (2011) and Sil & Katzenstein
(2010) and to Bennetts (2013) related recommendation that IR focus on causal mechanisms. For
scholars interested in pluralism and bridge building across the theoretical traditions, Checkels
(2012) recommended focus on causal mechanisms and attention to research design in the early
stages of a project are relevant ways in which experimental research can play a larger role in IR.
416 Hyde
The testing of midrange theory is also the area in which experiments are most likely to be
policy relevant (see Walt 2005) and in which high-quality academic work can be connected to
real-world policy making. For example, rather than investigating whether Americans think like
intuitive neorealists, intuitive Rawlsians, or intuitive neoclassical economists in relation to factors
that interact with trade policy preferences (Herrmann et al. 2001), a more midlevel question that
is also more policy relevant would instead focus on what sorts of arguments from policy makers
have the greatest effects on American support for or opposition to trade liberalization.
Field experiments in particular have signicant potential to be policy relevant, in part because
they can involve randomizing real programs or policies. And, although this may sound even more
difcult, eld experimental studies have the potential to test the empirical implications of bigger-
picture IR theories precisely because they are designed around real-world interventions. But
even for policy-relevant eld experiments in IR, showing clear connections to specic theories
and empirically testing the observable implications of broader theories remains a difcult but
important step. As Deaton (2010) argues in a critique of experiments in development economics,

which could just as easily be applied to IR, [t]he demand that experiments be theoretically driven
is, of course, no guarantee of success, though the lack of it is close to a guarantee of failure (p. 450).
Taking a more pragmatic approach to IR theory is likely to be a much more fruitful starting
point for squaring the micro-macro circle outlined at the beginning of this essay, and for clarifying
the connections between experimental research and bigger-picture IR theory, than forcing exper-
iments into so-called tests of the isms. This is not to say that realism, neoliberal institutionalism,
and constructivism are no longer relevant to IR. To the contrary, these research traditions are
quite useful in generating hypotheses, discussing potential explanations for specic outcomes of
interest, and thinking about which assumptions may be more or less appropriate in a particular
context (see treatments by Press et al. 2013, Rathbun 2009). However, the research traditions
should not be mistaken for well-specied theories (Lake 2011) with empirical implications that
can be tested with experimental methods.
FUTURE RESEARCH AGENDAS

The use of experiments in IR, particularly lab and eld experimentation, is in its infancy, and an
evaluation of their utility is premature. However, it is clear that there is much untapped potential
for learning about important topics through experimental research in IR as well as in areas at the
intersections of IR and comparative and American politics.
Moving forward, researchers should keep in mind that best practices continue to evolve. The
community of researchers will innovate in the use of experimental methods and develop standards
for high-quality research, led in part by the new Experiments Section of the American Political
Science Association (APSA) and groups such as Experiments in Governance and Politics (EGAP).
As these organizations have begun to advocate, learning in experimental research is enhanced by
other factors that will be important for IR researchers to keep in mind. Preregistration of exper-
imental protocols, ideally before experiments are implemented but crucially before experimental
results are analyzed, as outlined by Humphreys et al. (2013), will likely become an expected step
in any experimental study.
Replication should also be a valuable part of any research agenda that incorporates experiments.
Consumers of experimental research (including peer reviewers) should recognize the crucial
role that replication plays in building knowledge based on experimental ndings and in linking
experimental ndings to bigger-picture theories. Because a single experiment in a particular
substantive area is unlikely to provide the nal word on any subject, consumers of experimental
research should keep in mind that learning from experiments might best be done in the context

of iterated studies over time (Goldstein & Peters 2014, Jensen & Malesky 2013) or in multiple
studies replicated across contexts. Experimentalists and agencies that fund research can contribute
to the accumulation of knowledge by helping to coordinate research across studies to maximize
opportunities for replication (Dunning & Hyde 2014). Building a coherent research agenda, with
clear links to theory, will allow the ndings of experimental research to feed more clearly into our
understanding of the big questions in IR. Experimental methods are unlikely to touch all corners
of IR and may not even be fruitful in the majority of areas, but it is clear that they are dramatically
underutilized and often poorly understood, and there is great potential to expand their use.
DISCLOSURE STATEMENT
The author is a member of the Experiments Section of the American Political Science Association
and of the Experiments in Governance and Politics (EGAP) network and currently serves as an
elected member of EGAPs Board of Directors.

ACKNOWLEDGMENTS
I thank Jonathon Baron and Lauren Pinson for excellent research assistance and comments. For
helpful comments and conversations I thank Nuno Monteiro, Alex Debs, Bruce Russett, and
Jessica Weiss.
LITERATURE CITED
Arce DG, Croson RT, Eckel CC. 2011. Terrorism experiments. J. Peace Res. 48(3):37382
Ardanaz M, Murillo MV, Pinto PM. 2013. Sensitivity to issue framing on trade policy preferences: evidence
from a survey experiment. Int. Organ. 67(2):41137
Avdeenko A, Gilligan MJ. 2014. International interventions to build social capital: evidence from a eld experiment
in Sudan. Work. Pap., World Bank, Washington, DC
Beath A, Christia F, Enikolopov R. 2012. Direct democracy and resource allocation: experimental evidence from
Afghanistan. Work. Pap., World Bank Off. Chief Econ. East Asia Pac. Reg., Washington, DC
Beath A, Christia F, Enikolopov R. 2013. Empowering women through development aid: evidence from a
eld experiment in Afghanistan. Am. Polit. Sci. Rev. 107(3):54057
Beer FA, Sinclair GP, Healy AF, Bourne LE Jr. 1995. Peace agreement, intractable conict, escalation trajec-
tory: a psychological laboratory experiment. Int. Stud. Q. 39(3):297312
Bennett A. 2013. The mother of all isms: causal mechanisms and structured pluralism in International Relations
theory. Eur. J. Int. Relat. 19(3):45981
Berinsky AJ. 2007. Assuming the costs of war: events, elites, and American public support for military conict.
J. Polit. 69(4):97597
Berinsky AJ. 2009. In Time of War: Understanding American Public Opinion from World War II to Iraq. Chicago:
Univ. Chicago Press. 1st ed.
Berinsky AJ, Huber GA, Lenz GS. 2012. Evaluating online labor markets for experimental research: Ama-
zon.coms Mechanical Turk. Polit. Anal. 20(3):35168
Berinsky AJ, Kinder DR. 2006. Making sense of issues through media frames: understanding the Kosovo crisis.
J. Polit. 68(3):64056
Berman E, Callen MJ, Gibson C, Long JD. 2014. Election fairness and government legitimacy in Afghanistan.
Work. Pap., Natl. Bur. Econ. Res., Cambridge, MA
Blair G, Christine Fair C, Malhotra N, Shapiro JN. 2013. Poverty and support for militant politics: evidence
from Pakistan. Am. J. Polit. Sci. 57(1):3048
Blair G, Imai K, Lyall J. 2014. Comparing and combining list and endorsement experiments: evidence from
Afghanistan. Am. J. Polit. Sci. 58(4):104363
418 Hyde
Blaschke SM, Carroll PP, Chaves DR, Findley MG, Gleave MC, et al. 2014. Extrinsic, intrinsic, and
social incentives for crowdsourcing development information in Uganda: a eld experiment. Work. Pap.,
UNICEF/Columbia Univ., New York
Blattman C, Annan J. 2014. Can employment reduce lawlessness and rebellion? A eld experiment with high-risk
youth in a fragile state. Work. Pap., Sch. Int. Public Aff., Columbia Univ., New York
Blattman C, Hartman AC, Blair RA. 2014. How to promote order and property rights under weak rule of
law? An experiment in changing dispute resolution behavior through community education. Am. Polit.
Sci. Rev. 108(1):10020
Boettcher WA. 1995. Context, methods, numbers, and words prospect theory in International Relations.
J. Con. Resolut. 39(3):56183
Boettcher WA. 2004. The prospects for prospect theory: an empirical evaluation of international relations
applications of framing and loss aversion. Polit. Psychol. 25(3):33162
Boettcher WA, Cobb MD. 2006. Echoes of Vietnam? Casualty framing and public perceptions of success and
failure in Iraq. J. Con. Resolut. 50(6):83154
Boettcher WA, Cobb MD. 2009. Dont let them die in vain: casualty frames and public tolerance for
escalating commitment in Iraq. J. Con. Resolut. 53(5):67797

Brigham M, Egbert A, Findley M, Petrey C, Nielson D. 2014. International actors willingness to update: a global
eld experiment on micronance institutions. Work. Pap., Dep. Polit. Sci., Brigham Young Univ., Provo,
UT
Bullock W, Imai K, Shapiro JN. 2011. Statistical analysis of endorsement experiments: measuring support for
militant groups in Pakistan. Polit. Anal. 19(4):36384
Bursztyn L, Callen MJ, Ferman B, Hasanain A, Yuchtman N. 2014. A revealed preference approach to the elicitation
of political attitudes: experimental evidence on anti-Americanism in Pakistan. Work. Pap., Natl. Bur. Econ.
Res., Cambridge, MA
Chaudoin S. 2014. Promises or policies? An experimental analysis of international agreements and audience
reactions. Int. Organ. 68(1):23556
Checkel JT. 2012. Theoretical pluralism in IR: possibilities and limits. In Handbook of International Relations,
ed. W Carlsnaes, T Risse, BA Simmons, pp. 22041. Thousand Oaks, CA: Sage
Chilton AS, Tingley DH. 2014. Why the study of international law needs experiments. Columbia J. Transnatl.
Law 52(1):173239
Corstange D, Marinov N. 2012. Taking sides in other peoples elections: the polarizing effect of foreign
intervention. Am. J. Polit. Sci. 56(3):65570
Dafoe A, Zhang B, Caughey D. 2014. Confounding in survey experiments. Presented at Annu. Meet. Soc. Polit.
Methodol., 31st, Atlanta, GA
Davies GAM, Johns R. 2013. Audience costs among the British public: the impact of escalation, crisis type,
and prime ministerial rhetoric. Int. Stud. Q. 57(4):72537
Deaton A. 2010. Instruments, randomization, and learning about development. J. Econ. Lit. 48(2):42455
Desposato S, Gartzke E, Suong C. 2013. How democratic is the democratic peace? A survey experiment of foreign
policy preferences in Brazil and China. Work. Pap., Dep. Polit. Sci., Univ. Calif. San Diego
Dickson ES. 2009. Do participants and observers assess intentions differently during bargaining and conict?
Am. J. Polit. Sci. 53(4):91030
Dietrich S. 2014. Donor political economies and the pursuit of aid effectiveness. Work. Pap., Dep. Polit. Sci., Univ.
Mo., Columbia, MO
Dietrich S, Winters MS. 2014. Foreign aid and government legitimacy. Work. Pap., Dep. Polit. Sci., Univ. Mo.,
Columbia, MO
Druckman JN, Green DP, Kuklinski JH, Lupia A. 2006. The growth and development of experimental
research in political science. Am. Polit. Sci. Rev. 100(4):62735
Druckman JN, Green DP, Kuklinski JH, Lupia A. 2011. Cambridge Handbook of Experimental Political Science.
New York: Cambridge Univ. Press
Druckman JN, Kam CD. 2011. Students as experimental participants: a defense of the narrow data base.
See Druckman et al. 2011, pp. 4157
Dunning T. 2012. Natural Experiments in the Social Sciences: A Design-Based Approach. New York: Cambridge
Univ. Press

Dunning T, Hyde S. 2014. Replicate it! A proposal to improve the study of political accountability. The
Monkey Cage blog, Washington Post, May 16. http://www.washingtonpost.com/blogs/monkey-cage/
wp/2014/05/16/replicate-it-a-proposal-to-improve-the-study-of-political-accountability/
Ehrlich SD, Hearn E. 2014. Does compensating the losers increase support for trade? An experimental test
of the embedded liberalism thesis. Foreign Policy Anal. 10(2):14964
Fair CC, Malhotra N, Shapiro JN. 2014. Democratic values and support for militant politics: evidence from
a national survey of Pakistan. J. Con. Resolut. 58(5):74370
Fearon JD. 1994. Domestic political audiences and the escalation of international disputes. Am. Polit. Sci. Rev.
88(3):57792
Fearon JD, Humphreys M, Weinstein JM. 2009. Can development aid contribute to social cohesion after civil
war? Evidence from a eld experiment in post-conict Liberia. Am. Econ. Rev. 99(2):28791
Findley MG, Harris AS, Milner HV, Nielson DL. 2014a. Resource curse or cure? Experimental evidence about
elite and mass behavior toward foreign aid versus government programs in Uganda. Work. Pap., Dep. Gov.,
Univ. Tex. Austin, Austin, TX
Findley MG, Jensen NM, Nielson DL. 2013a. A eld experiment on the liability of foreignness: country of origin
and U.S. inward investment promotion. Work. Pap., Dep. Gov., Univ. Tex. Austin, Austin, TX
Findley MG, Laney B, Nielson DL, Sharman JC. 2013b. Deceptive studies or deceptive answers? Alternative global
eld and survey experiments on anonymous incorporation. Work. Pap., Dep. Gov., Univ. Tex. Austin, Austin,
TX
Findley MG, Nielson DL, Sharman JC. 2013c. Using eld experiments in international relations: a randomized
study of anonymous incorporation. Int. Organ. 67(4):65793
Findley MG, Nielson DL, Sharman JC. 2014b. Global Shell Games: Experiments in Transnational Relations,
Crime, and Terrorism. New York: Cambridge Univ. Press
Gartner SS. 2008. The multiple effects of casualties on public support for war: an experimental approach. Am.
Polit. Sci. Rev. 102(1):95106
Gartner SS. 2011. On behalf of a grateful nation: conventionalized images of loss and individual opinion
change in war. Int. Stud. Q. 55(2):54561
Gartner SS, Gelpi CF. 2012. The affect and effect of images of success and failure in war on public opinion. Work.
Pap., Penn. State Univ., State College, PA
Gelpi C. 2010a. Performing on cue? The formation of public opinion toward war. J. Con. Resolut. 54(1):88
116
Gelpi C. 2010b. The two-front homefront: American attitudes toward Afghanistan in the shadow of Iraq. Work.
Pap., Dep. Polit. Sci., Duke Univ., Durham, NC
Gerber AS, Green DP. 2012. Field Experiments: Design, Analysis, and Interpretation. New York: W.W. Norton
& Co.
Gerring J, McDermott R. 2007. An experimental template for case study research. Am. J. Polit. Sci. 51(3):688
701
Geva N, Mayhar J, Skorick JM. 2000. The cognitive calculus of foreign policy decision making: an experimental
assessment. J. Con. Resolut. 44(4):44771
Gilligan MJ, Pasquale BJ, Samii C. 2014. Civil war and social cohesion: lab-in-the-eld evidence from Nepal.
Am. J. Polit. Sci. 58(3):60419
Goldgeier JM, Tetlock PE. 2001. Psychology and international relations theory. Annu. Rev. Polit. Sci. 4:6792
Goldstein JL, Peters ME. 2014. Nativism or economic threat: attitudes toward immigrants during the Great
Recession. Int. Interact. 40(3):376401
Gray J, Hicks RP. 2014. Reputations, perceptions, and international economic agreements. Int. Interact.
40(3):32549
Grieco JM, Gelpi C, Reier J, Feaver PD. 2011. Lets get a second opinion: international institutions and
American public support for war. Int. Stud. Q. 55(2):56383
Guisinger A, Saunders EN. 2013. Mapping the boundaries of elite cues: how elites shape mass opinion across interna-
tional issues. Presented at Annu. Meet. Am. Polit. Sci. Assoc., 31st, Chicago
Hafner-Burton EM, Hughes DA, Victor DG. 2013. The cognitive revolution and the political psychology of
elite decision making. Perspect. Polit. 11(2):36886
420 Hyde
Hafner-Burton EM, LeVeck BL, Victor DG, Fowler JH. 2014. Decision maker preferences for international
legal cooperation. Int. Organ. 68(4):84576
Hainmueller J, Hiscox MJ. 2010. Attitudes toward highly skilled and low-skilled immigration: evidence from
a survey experiment. Am. Polit. Sci. Rev. 104(1):6184
Hainmueller J, Hiscox MJ. 2012a. Buying green? Field experimental tests of consumer support for environmentalism.
Work. Pap., Polit. Sci. Dep., Mass. Inst. Technol., Cambridge, MA
Hainmueller J, Hiscox MJ. 2012b. The socially conscious consumer? Field experimental tests of consumer support for
fair labor standards. Work. Pap., Polit. Sci. Dep., Mass. Inst. Technol., Cambridge, MA
Hainmueller J, Hiscox MJ, Sequeira S. 2014. Consumer demand for the fair trade label: evidence from a
multi-store eld experiment. Rev. Econ. Stat. In press
Herrmann RK, Shannon VP. 2001. Defending international norms: the role of obligation, material interest,
and perception in decision making. Int. Organ. 55(3):62154
Herrmann RK, Tetlock PE, Diascro MN. 2001. How Americans think about trade: reconciling conicts
among money, power, and principles. Int. Stud. Q. 45(2):191218
Herrmann RK, Tetlock PE, Visser PS. 1999. Mass public decisions to go to war: a cognitive-interactionist
framework. Am. Polit. Sci. Rev. 93(3):55373
Hirose K, Imai K, Lyall J. 2014. Can civilian attitudes predict civil war violence? Work. Pap., Dep. Polit., Princeton
Univ., Princeton, NJ
Hiscox MJ. 2006. Through a glass and darkly: attitudes toward international trade and the curious effects of
issue framing. Int. Organ. 60(3):75580
Holsti OR, Rosenau JN. 1984. American Leadership in World Affairs: Vietnam and the Breakdown of Consensus.
Boston: Unwin Hyman
Holsti OR, Rosenau JN. 1993. The structure of foreign policy beliefs among American opinion leadersafter
the Cold War. Millenn. J. Int. Stud. 22(2):23578
Horowitz MC, Levendusky MS. 2011. Drafting support for war: conscription and mass support for warfare.
J. Polit. 73(2):52434
Humphreys M, de la Sierra RS, van der Windt P. 2013. Fishing, commitment, and communication: a proposal
for comprehensive nonbinding research registration. Polit. Anal. 21(1):120
Humphreys M, Weinstein JM. 2009. Field experiments and the political economy of development. Annu. Rev.
Polit. Sci. 12(1):36778
Hyde SD. 2007. The observer effect in international politics: evidence from a natural experiment. World Polit.
60(1):3763
Hyde SD. 2010a. Experimenting in democracy promotion: international observers and the 2004 presidential
elections in Indonesia. Perspect. Polit. 8(2):51127
Hyde SD. 2010b. The future of eld experiments in international relations. Ann. Am. Acad. Polit. Soc. Sci.
628(1):7284
Hyde SD. 2011. The Pseudo-Democrats Dilemma: Why Election Monitoring Became an International Norm. Ithaca,
NY: Cornell Univ. Press
Hyde SD, Lamb E. 2013. The micro-level consequences of democracy promotion: a eld experiment in rural Cambodia.
Work. Pap., Dep. Polit. Sci., Yale Univ., New Haven, CT
Jensen NM, Malesky E, Medina M, Ozdemir U. 2013. Pass the bucks: credit, blame, and the global competition
for investment. Int. Stud. Q. 58(3):43347
Jensen NM, Malesky EJ. 2013. Does the OECD convention affect bribery? Investment liberalization and corruption
in Vietnam. Work. Pap., Dep. Int. Bus., Wash. Univ., St. Louis, MO
Jensen NM, Mukherjee B, Bernhard WT. 2014. Introduction: survey and experimental research in interna-
tional political economy. Int. Interact. 40(3):287304
Johns R, Davies GAM. 2012. Democratic peace or clash of civilizations? Target states and support for war in
Britain and the United States. J. Polit. 74(4):103852
Jordan R, Maliniak D, Oakes A, Peterson S, Tierney MJ. 2009. One discipline or many? TRIP survey of in-
ternational relations faculty in ten countries. Rep. Inst. Theory Pract. Int. Relat., Coll. William Mary,
Williamsburg, VA
Kahler M. 1998. Rationality in international relations. Int. Organ. 52(4):91941

Kahneman D, Tversky A. 1979. Prospect theory: an analysis of decision under risk. Econometrica 47(2):26391
Kemmelmeier M, Winter DG. 2000. Putting threat into perspective: experimental studies on perceptual
distortion in international conict. Pers. Soc. Psychol. Bull. 26(7):795809
Kertzer JD, McGraw KM. 2012. Folk realism: testing the microfoundations of realism in ordinary citizens.
Int. Stud. Q. 56(2):24558
King G, Nielsen R, Coberley C, Pope JE, Wells A. 2011. Avoiding randomization failure in program evaluation,
with application to the Medicare Health Support program. Popul. Health Manag. 14(Suppl. 1):S1122
Lake DA. 2011. Why isms are evil: theory, epistemology, and academic sects as impediments to understand-
ing and progress. Int. Stud. Q. 55(2):46580
Lake DA. 2013. Theory is dead, long live theory: the end of the Great Debates and the rise of eclecticism in
international relations. Eur. J. Int. Relat. 19(3):56787
Lake DA, Powell R. 1999. Strategic Choice and International Relations. Princeton, NJ: Princeton Univ. Press
Levendusky MS, Horowitz MC. 2012. When backing down is the right decision: partisanship, new informa-
tion, and audience costs. J. Polit. 74(2):32338
Levy JS. 1997. Prospect theory, rational choice, and international relations. Int. Stud. Q. 41(1):87112
List JA, Mason CF. 2011. Are CEOs expected utility maximizers? J. Econom. 162(1):11423
Lyall J, Blair G, Imai K. 2013a. Explaining support for combatants during wartime: a survey experiment in
Afghanistan. Am. Polit. Sci. Rev. 107(4):679705
Lyall J, Imai K, Shiraito Y. 2013b. Dropping a dime: coethnic bias and wartime informing. Work. Pap., Dep. Polit.
Sci., Yale Univ., New Haven, CT
Maliniak D, Oakes A, Peterson S, Tierney MJ. 2007. The view from the ivory tower: TRIP survey of international
relations faculty in the United States and Canada. Rep. Inst. Theory Pract. Int. Relat., Coll. William Mary,
Williamsburg, VA
Maliniak D, Peterson S, Tierney MJ. 2012. TRIP around the world: teaching, research, and policy views of interna-
tional relations faculty in 20 countries. Rep. Inst. Theory Pract. Int. Relat., Coll. William Mary, Williams-
burg, VA
Margalit Y. 2012. Lost in globalization: international economic integration and the sources of popular dis-
content. Int. Stud. Q. 56(3):484500
Marinov N. 2013. Voter attitudes when democracy promotion turns partisan: evidence from a survey-
experiment in Lebanon. Democratization 20(7):1297321
McDermott R. 2002. Experimental methods in political science. Annu. Rev. Polit. Sci. 5(1):3161
McDermott R. 2004. Prospect theory in political science: gains and losses from the rst decade. Polit. Psychol.
25(2):289312
McDermott R. 2011. New directions for experimental work in international relations. Int. Stud. Q. 55(2):503
20
McDermott R, Cowden J, Koopman C. 2002. Framing, uncertainty, and hostile communications in a crisis
experiment. Polit. Psychol. 23(1):13349
McDermott R, Cowden JA. 2001. The effects of uncertainty and sex in a crisis simulation game. Int. Interact.
27(4):35380
Mearsheimer JJ, Walt SM. 2013. Leaving theory behind: why simplistic hypothesis testing is bad for interna-
tional relations. Eur. J. Int. Relat. 19(3):42757
Milner HV, Nielson DL, Findley MG. 2014. Which devil in development? An experiment on citizen actions toward
foreign aid and government programs in Uganda. Work. Pap., Dep. Polit., Princeton Univ., Princeton, NJ
Mintz A. 2004. Foreign policy decision making in familiar and unfamiliar settings: an experimental study of
high-ranking military ofcers. J. Con. Resolut. 48(1):91104
Mintz A, Geva N, Redd SB, Carnes A. 1997. The effect of dynamic and static choice sets on political decision
making: an analysis using the decision board platform. Am. Polit. Sci. Rev. 91(3):55366
Mintz A, Redd SB, Vedlitz A. 2006. Can we generalize from student experiments to the real world in political
science, military affairs, and international relations? J. Con. Resolut. 50(5):75776
Mintz A, Yang Y, McDermott R. 2011. Experimental approaches to international relations. Int. Stud. Q.
55(2):493501
Morton RB, Williams KC. 2010. Experimental Political Science and the Study of Causality: From Nature to the
Lab. New York: Cambridge Univ. Press
422 Hyde
Moxnes E, Van der Heijden E. 2003. The effect of leadership in a public bad experiment. J. Con. Resolut.
47(6):77395
Mutz DC. 2011. Population-Based Survey Experiments. Princeton, NJ: Princeton Univ. Press
Mvukiyehe E, Samii C. 2013. Promoting democracy in fragile states: insights from a eld experiment in Liberia.
Work. Pap., Dep. Polit. Sci., Columbia Univ., New York
Naoi M, Kume I. 2011. Explaining mass support for agricultural protectionism: evidence from a survey
experiment during the global recession. Int. Organ. 65(4):77195
Pepinsky TB. 2014. Surveys, experiments, and the landscape of international political economy. Int. Interact.
40(3):43142
Peterson S, Tierney MJ, Maliniak D. 2005. Teaching and research practices, views on the discipline, and policy
attitudes of international relations faculty at US colleges and universities. Program Theory Pract. Int. Relat.,
Coll. William Mary, Williamsburg, VA
Press DG, Sagan SD, Valentino BA. 2013. Atomic aversion: experimental evidence on taboos, traditions, and
the non-use of nuclear weapons. Am. Polit. Sci. Rev. 107(1):188206
Pronin E, Kennedy K, Butsch S. 2006. Bombing versus negotiating: how preferences for combating terrorism
are affected by perceived terrorist rationality. Basic Appl. Soc. Psychol. 28(4):38592
Putnam TL, Shapiro JN. 2013. International law and voter preferences: the case of foreign human rights violations.
Work. Pap., Dep. Polit. Sci., Columbia Univ., New York

Rathbun BC. 2009. It takes all types: social psychology, trust, and the international relations paradigm in our
minds. Int. Theory 1(3):34580
Redd SB. 2002. The inuence of advisers on foreign policy decision making: an experimental study. J. Con.
Resolut. 46(3):33564
Renshon J. 2015. Losing face and sinking costs: experimental evidence on the judgment of political and military
leaders. Int. Organ. 69(3):In press
Rousseau DL, Garcia-Retamero R. 2007. Identity, power, and threat perception: a cross-national experimental
study. J. Con. Resolut. 51(5):74471
Saideman S. 2013. Lamenting the loss of the light, the ebbing of grand theory and the decline of old
boy networks. The Duck of Minerva blog, Jan. 5. http://www.whiteoliphaunt.com/duckofminerva/
2013/01/lamenting-the-loss-of-the-light-the-ebbing-of-grand-theory-and-the-decline-of-old-
boy-networks.html
Saunders EN. 2013. The electoral disconnection in U.S. foreign policy. Work. Pap., Dep. Polit. Sci., George Wash.
Univ., Washington, DC
Schelling TC. 1961. Experimental games and bargaining theory. World Polit. 14(1):4768
Sil R, Katzenstein PJ. 2010. Analytic eclecticism in the study of world politics: reconguring problems and
mechanisms across research traditions. Perspect. Polit. 8(2):41131
Singer JD. 1961. The level-of-analysis problem in international relations. World Polit. 14(Spec. Issue 01):7792
Tingley D. 2014. Survey research in international political economy: motivations, designs, methods. Int.
Interact. 40(3):44351
Tingley D, Tomz M. 2012. How does the UN Security Council inuence public opinion? Work. Pap., Dep. Gov.,
Harvard Univ., Cambridge, MA
Tingley D, Tomz M. 2014. Conditional cooperation and climate change. Comp. Polit. Stud. 47(3):34468
Tomz M. 2007. Domestic audience costs in international relations: an experimental approach. Int. Organ.
61(4):82140
Tomz M. 2008. Reputation and the effect of international law on preferences and beliefs. Work. Pap., Dep. Polit.
Sci., Stanford Univ., Stanford, CA
Tomz M, Weeks JLP. 2012. Human rights, democracy, and international conict. Work. Pap., Dep. Polit. Sci.,
Stanford Univ., Stanford, CA
Tomz M, Weeks JLP. 2013. Public opinion and the democratic peace. Am. Polit. Sci. Rev. 107(4):84965
Trager RF, Vavreck L. 2011. The political costs of crisis bargaining: presidential rhetoric and the role of party.
Am. J. Polit. Sci. 55(3):52645
Tversky A, Kahneman D. 1992. Advances in prospect theory: cumulative representation of uncertainty. J. Risk
Uncertain. 5(4):297323

Ulfelder J. 2014. Digging down to the micro-foundations. Dart-Throwing Chimp blog, Apr. 3. http://
dartthrowingchimp.wordpress.com/2013/04/03/digging-down-to-the-micro-foundations/
Van der Heijden E, Moxnes E. 2013. Leading by example to protect the environment: Do the costs of leading
matter? J. Con. Resolut. 57(2):30726
Wallace GPR. 2013. International law and public attitudes toward torture: an experimental study. Int. Organ.
67(1):10540
Wallace GPR. 2014. Martial law? Military experience, international law, and support for torture. Int. Stud. Q.
58(3):50114
Walt SM. 2005. The relationship between theory and policy in international relations. Annu. Rev. Polit. Sci.
8(1):2348
Waltz KN. 1979. Theory of International Politics. Boston: Addison-Wesley
424 Hyde
PL18-FrontMatter ARI 6 April 2015 13:4
Annual Review of
Political Science
Contents Volume 18, 2015
A Conversation with Hanna Pitkin

Hanna Pitkin and Nancy Rosenblum p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Income Inequality and Policy Responsiveness
Robert S. Erikson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p11

How Do Campaigns Matter?
Gary C. Jacobson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p31

Electoral Rules, Mobilization, and Turnout
Gary W. Cox p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p49
The Rise and Spread of Suicide Bombing
Michael C. Horowitz p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p69
The Dysfunctional Congress
Sarah Binder p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p85
Political Islam: Theory
Andrew F. March p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 103
Borders, Conict, and Trade
Kenneth A. Schultz p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 125
From Mass Preferences to Policy
Brandice Canes-Wrone p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 147
Constitutional Courts in Comparative Perspective:
A Theoretical Assessment
Georg Vanberg p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 167
Epistemic Democracy and Its Challenges
Melissa Schwartzberg p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 187
The New Look in Political Ideology Research
Edward G. Carmines and Nicholas J. DAmico p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 205
The Politics of Central Bank Independence
Jose Fernandez-Albertos p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 217
What Have We Learned about the Resource Curse?
Michael L. Ross p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 239
v
PL18-FrontMatter ARI 6 April 2015 13:4
How Party Polarization Affects Governance

Frances E. Lee p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 261
Migration, Labor, and the International Political Economy
Layna Mosley and David A. Singer p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 283
Law and Politics in Transitional Justice
Leslie Vinjamuri and Jack Snyder p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 303
Campaign Finance and American Democracy
Yasmin Dawood p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 329
Female Candidates and Legislators
Jennifer L. Lawless p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 349
Power Tool or Dull Blade? Selectorate Theory for Autocracies

Mary E. Gallagher and Jonathan K. Hanson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 367
Realism About Political Corruption

Mark Philp and Elizabeth David-Barrett p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 387
Experiments in International Relations: Lab, Survey, and Field
Susan D. Hyde p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 403
Political Theory as Both Philosophy and History: A Defense Against
Methodological Militancy
Jeffrey Edward Green p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 425
The Empiricists Insurgency
Eli Berman and Aila M. Matanock p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 443
The Scope of Comparative Political Theory
Diego von Vacano p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 465
Should We Leave Behind the Subeld of International Relations?
Dan Reiter p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 481
Indexes
Cumulative Index of Contributing Authors, Volumes 1418 p p p p p p p p p p p p p p p p p p p p p p p p p p p 501

Cumulative Index of Article Titles, Volumes 1418 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 503
Errata
An online log of corrections to Annual Review of Political Science articles may be found
at http://www.annualreviews.org/errata/polisci
vi Contents

Hyde, Experiments in IR

Uploaded by

Copyright:

Available Formats

Hyde, Experiments in IR

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hyde, Experiments in IR

Uploaded by

Copyright:

Available Formats

PL18CH22-Hyde ARI 6 April 2015 10:8

Our comprehensive search

Annu. Rev. Polit. Sci. 2015. 18:40324 Keywords

The Annual Review of Political Science is online at Abstract

POTENTIAL VALUE OF EXPERIMENTS IN

www.annualreviews.org Experiments in International Relations 405

Assuming Subjects in the Lab Are Like Relevant IR Populations

www.annualreviews.org Experiments in International Relations 407

Bringing Experiments to Elites

Focusing on Public Opinion or Mass Behavior in IR

www.annualreviews.org Experiments in International Relations 409

Wallace 2013, 2014).

www.annualreviews.org Experiments in International Relations 411

civilians and veterans.

Theory First, Experiments Second

www.annualreviews.org Experiments in International Relations 413

theories about the efcacy of IOs (Hyde 2010b).

WHAT HAVE WE LEARNED ALREADY?

micro-level implications of various components of counterinsurgency and international efforts to

www.annualreviews.org Experiments in International Relations 415

WHAT IS THE INTERNATIONAL RELATIONS FIELD

important step. As Deaton (2010) argues in a critique of experiments in development economics,

FUTURE RESEARCH AGENDAS

www.annualreviews.org Experiments in International Relations 417

elected member of EGAPs Board of Directors.

escalating commitment in Iraq. J. Con. Resolut. 53(5):67797

www.annualreviews.org Experiments in International Relations 419

www.annualreviews.org Experiments in International Relations 421

Work. Pap., Dep. Polit. Sci., Columbia Univ., New York

www.annualreviews.org Experiments in International Relations 423

A Conversation with Hanna Pitkin

Robert S. Erikson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p11

Gary C. Jacobson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p31

How Party Polarization Affects Governance

Power Tool or Dull Blade? Selectorate Theory for Autocracies

Realism About Political Corruption

Cumulative Index of Contributing Authors, Volumes 1418 p p p p p p p p p p p p p p p p p p p p p p p p p p p 501

You might also like