ArticlePDF Available

God and the Government: Testing a Compensatory Control Mechanism for the Support of External Systems

Authors:

Abstract and Figures

The authors propose that the high levels of support often observed for governmental and religious systems can be explained, in part, as a means of coping with the threat posed by chronically or situationally fluctuating levels of perceived personal control. Three experiments demonstrated a causal relation between lowered perceptions of personal control and the defense of external systems, including increased beliefs in the existence of a controlling God (Studies 1 and 2) and defense of the overarching socio-political system (Study 4). A 4th experiment (Study 5) showed the converse to be true: A challenge to the usefulness of external systems of control led to increased illusory perceptions of personal control. In addition, a cross-national data set demonstrated that lower levels of personal control are associated with higher support for governmental control (across 67 nations; Study 3). Each study identified theoretically consistent moderators and mediators of these effects. The implications of these results for understanding why a high percentage of the population believes in the existence of God, and why people so often endorse and justify their socio-political systems, are discussed.
Content may be subject to copyright.
God and the Government: Testing a Compensatory Control Mechanism for
the Support of External Systems
Aaron C. Kay and Danielle Gaucher
University of Waterloo
Jamie L. Napier
New York University
Mitchell J. Callan
University of Western Ontario
Kristin Laurin
University of Waterloo
The authors propose that the high levels of support often observed for governmental and religious
systems can be explained, in part, as a means of coping with the threat posed by chronically or
situationally fluctuating levels of perceived personal control. Three experiments demonstrated a
causal relation between lowered perceptions of personal control and the defense of external systems,
including increased beliefs in the existence of a controlling God (Studies 1 and 2) and defense of the
overarching socio-political system (Study 4). A 4th experiment (Study 5) showed the converse to be
true: A challenge to the usefulness of external systems of control led to increased illusory
perceptions of personal control. In addition, a cross-national data set demonstrated that lower levels
of personal control are associated with higher support for governmental control (across 67 nations;
Study 3). Each study identified theoretically consistent moderators and mediators of these effects.
The implications of these results for understanding why a high percentage of the population believes
in the existence of God, and why people so often endorse and justify their socio-political systems,
are discussed.
Keywords: system justification, control, God, religion, governmental support
Approximately 95% of the American population believes in
the existence of God (Gallup & Castelli, 1989). Given the
prevalence of this belief, understanding its psychological foun-
dations has obvious implications for our understanding of basic
psychological functioning. Surprisingly little is known, how-
ever, about why the majority of people, in the majority of the
cultures around the world, believe in a higher order controlling
influence such as God. Likewise, people have a remarkable
ability to justify and defend the sociopolitical systems that
control, at least in part, their fate (see Jost, Banaji, & Nosek,
2004; Kay et al., 2007). In this context too there is little
empirical evidence examining why this is the case. In noting the
dearth of research on the psychological bases of religion,
Baumeister (2002) drew particular attention to the utility of
research in this domain for uncovering broad psychological
principles and for developing social psychological theory. Fol-
lowing that perspective, and inspired by previous psychological
theory (Rothbaum, Weisz, & Snyder, 1982), in this article we
describe a general social psychological model developed to help
explain the relationship people hold with external systems of
control—systems such as governments, societal institutions,
religious ideologies, and the like—and then test this model first
in the context of religious beliefs and then in the context of
beliefs about governmental systems.
Compensatory Control and the Endorsement of External
Systems
A long line of research has made it clear that people are often
motivated to perceive that they possess personal control over their
social environments and outcomes (Kelly, 1955; Perkins, 1968;
Presson & Benassi, 1996; Seligman, 1975, 1976; Skinner, 1995;
White, 1959; but see Burger, 1989). One reason for this motiva-
tion, it has been proposed, is to help prevent feelings of random-
ness and chaos in the social world. Such perceptions can be
psychologically stressful, traumatic, and anxiety provoking (e.g.,
Janoff-Bulman, 1992; Pennebaker & Stone, 2004) and are there-
fore often avoided in favor of perceptions of structure and order
(Kruglanski, 1989; Kruglanski & Webster, 1996; Landau et al.,
2004; Landau, Greenberg, Solomon, Pyszczynski, & Martens,
2006). From this perspective, the motivation to perceive personal
control is considered a subgoal of the larger and more inclusive
motivation to defend against perceptions of randomness and chaos
Aaron C. Kay, Danielle Gaucher, and Kristin Laurin, Department of
Psychology, University of Waterloo; Jamie L. Napier, Department of
Psychology, New York University; Mitchell J. Callan, Department of
Psychology, University of Western Ontario.
This research was supported by a Social Sciences and Humanities
Research Council of Canada research grant to Aaron C. Kay. We would
like to thank Gra´inne Fitzsimons, John Holmes, Ian McGregor, and Steven
Spencer for their constructive comments on this article and the ideas
within.
Correspondence concerning this article should be addressed to Aaron C.
Kay, Department of Psychology, University of Waterloo, 200 University
Avenue West, Waterloo, Ontario, Canada N2L 3G1. E-mail:
ackay@uwaterloo.ca
Journal of Personality and Social Psychology Copyright 2008 by the American Psychological Association
2008, Vol. 95, No. 1, 18 –35 0022-3514/08/$12.00 DOI: 10.1037/0022-3514.95.1.18
18
within the social environment.
1
Such a perspective is particularly
common among social justice theorists (e.g., Jost et al., 2004;
Lerner, 1980).
However, at the same time that the field has demonstrated the
tendency to prefer feelings of personal control, it has also made it
clear that perceptions of personal control, the motivation to
achieve such perceptions, and the positive effects of such percep-
tions all fluctuate greatly, both situationally and chronically, be-
tween and within cultures (e.g., Burger, 1989; Burger & Cooper,
1979; Ji, Peng, & Nisbett, 2000; Dweck & Leggett, 1988; Pepitone
& Saffiotti, 1997; Rodin, Rennert, & Solomon, 1980; Sethi &
Lepper, 1998; Weisz, Rothbaum, & Blackburn, 1984; Wohl &
Enzle, 2003). This latter set of findings— coupled with the extent
to which perceptions of personal control have been asserted as a
primary means of fulfilling the motivation to guard against ran-
domness in the social world (Hafer & Be`gue, 2005; Janoff-
Bulman, 1992; Lerner, 1980)—raises an intriguing question: How
might people maintain beliefs in order and structure, and defend
against perceptions of randomness, in the face of fluctuating per-
ceptions of personal control?
We propose that perceptions of personal control, although a very
effective means of preventing feelings of randomness in the social
world, are only one (substitutable) means of doing so. The en-
dorsement of external systems that impose order on one’s social
world, we believe, can also help people to meet this goal. In other
words, because of a psychological need to insulate the self from
feelings of randomness and chaos (and, conversely, to promote
feelings of order and structure), we posit a substitutability of the
belief in personal control with the belief that things are under
control (Antonovsky, 1979). That is, to the extent that external
systems of control, such as religions, governments, or broad soci-
etal institutions, suggest a reduction of randomness in the social
order, people should show an increased tendency to rely on and
endorse these systems when levels of personal control are low or
under threat.
2
As an analogy to this model, imagine a glass that only when full
represents a given individual’s preferred level of perceived order
within his or her environment. As one way to fill this glass, this
individual might rely on his or her perceptions of personal control
over his or her outcomes: To guard against perceptions of random-
ness in the world, that is, he or she can affirm the belief that
whatever happens, good or bad, will be due to his or her own
actions and therefore not random. Often, however, such percep-
tions will only partially fill the glass because levels of perceived
personal control, for a variety of reasons, tend to fluctuate (see
Burger, 1989, 1992). Much of the time, therefore, to fill the glass
completely, he or she will need to complement these beliefs in
personal control with one or more external systems of control.
Thus, to the extent that an individual is motivated to maintain a
psychologically comfortable level of perceived order in his or her
environment, when levels of personal control dip, the endorsement
and defense of external systems of control should increase.
This prediction is based on a rich history of theory and empirical
research suggesting that in times of threatened personal control,
people will increasingly align themselves with external sources of
control. Most notably, Rothbaum, Weisz, and Snyder’s (1982)
influential two-process model of perceived control suggests that
when people’s means of primary control—that is, people’s ability
to directly influence their environment in line with their personal
desires—are thwarted, they may engage in any number of second-
ary control strategies (for a review, see Morling & Evered, 2006;
Rothbaum et al., 1982). Although Rothbaum et al. outlined a
number of such strategies, most relevant to this line of research is
their notion of vicarious control, a secondary control strategy in
which people are presumed to gain a sense of personal control by
identifying with powerful others. Although we are unaware of any
experimental tests of this model in the context of beliefs about the
existence of God or the defense of sociopolitical systems, in
discussing the likely manifestations of their two-process model of
perceived control Rothbaum et al. noted that “the religious sphere
may ultimately prove to be one of the most fruitful for the assess-
ment of the relations among secondary control phenomena”
(p. 20).
The empirical predictions offered here, therefore, have not
emerged out of the blue, so to speak. They are instead borne out of
previous theoretical approaches, the most important of which is
that of Rothbaum et al. (1982). Our aim here is not to compete with
or supplant these previous approaches, but to build on and apply
them in such a way that we can provide empirical and theoretical
explanations for important social psychological phenomena that
have yet to be adequately understood. (For a discussion of the
ways in which our model also differs from this approach and other
related conceptions of compensatory control, see the General Dis-
cussion.)
Despite this solid theoretical footing, empirical (and, especially,
experimental) evidence for this phenomenon is scarce. There are,
however, data suggesting that under times of low personal control
people do rely on higher level sources of control. For example,
increases in beliefs and appeals to higher, supernatural sources of
control have been shown to occur among Melanesian fisherman in
foreign and dangerous waters (Malinowski, 1954); among Israelis
when under military threat (Keinan, 1994); among those with a
high chronic need for personal control (Keinan, 2002); and among
baseball players who more strongly believe that personal, primary
control only goes so far in controlling their performance on the
field (Burger & Lynn, 2005).
An important corollary to the specific model we are proposing
is that this process should be most likely engaged for certain types
of external systems. In this set of experiments, we focus primarily
on the role of perceived system benevolence—the extent to which
people believe the system holds an actual intention of serving the
interests of its constituents. A specific external system should be
less likely to be relied on if it is not perceived as having the goal
of serving the interests of the system it is governing. For example,
the control and order offered by an invading regime— one that is
seen as present only to help serve its own interests elsewhere—is
1
There are, of course, other proposed functions for the motivation to
perceive personal control (see Skinner, 1995).
2
To be clear, we are not suggesting that only negative instances of
uncontrollability lead to increased dependence on external systems. Rather,
any outcome that signifies a lack of personal control should lead to this
phenomenon. Although people may very well prefer a positive random
outcome (e.g., a $20 bill blowing in through the window) to a negative
controllable one (e.g., forgetting to remove a $20 bill before throwing a
pair of pants in the wash), any event that serves to lower feelings of
personal control, regardless of valence, should instigate processes of com-
pensatory control.
19
GOD AND THE GOVERNMENT
unlikely to be endorsed as a means of psychological coping. We
therefore predict that the extent to which an external system is
perceived as benevolent will moderate the extent to which it will
be increasingly relied on following reductions in personal control.
Studies 3 and 4 directly test this notion. We are not proposing,
however, that perceived benevolence is the only potential moder-
ator of this relationship. Other moderators also likely exist. For
example, an external system may also need to be perceived of as
capable of providing order, or as being efficacious, to serve as a
likely source of compensatory control. Moreover, in extreme cir-
cumstances, such as following dramatic reductions in perceived
personal control, people may rely on any external system avail-
able. We discuss these possibilities further in the General Discus-
sion.
In the remainder of this article, we review empirical evidence
that supports the notion that when personal control is low or
threatened, people increasingly support external sources—most
typically, God and government. Then, in five studies, we empiri-
cally test the compensatory control model in the context of both
religious and governmental systems of control.
Religious Beliefs as a Compensatory Mechanism of
Control
We believe that this good God, after He had created all things, did not
abandon them or give them up to fortune or chance [italics added], but
that according to His holy will He so rules and governs them that in
this world nothing happens without His direction. . . . This doctrine
gives us unspeakable consolation, for we learn thereby that nothing
can happen to us by chance [italics added], but only by the direction
of our gracious heavenly Father. He watches over us with fatherly
care, keeping all creatures so under His power that not one hair of our
head—for they are all numbered—nor one sparrow can fall to the
ground without the will of our Father. (de Bres, 1561/1984, pp.
449 450)
To the extent that people are motivated to preserve beliefs in
order and structure, and thereby prevent beliefs in randomness and
chaos (e.g., Dechesne, Janssen, & van Knippenberg, 2000; Green-
berg, Porteus, Simon, & Pyszczynski, 1995; Kay, Jimenez, & Jost,
2002; Kruglanski, 1989; Kruglanski & Webster, 1996; Landau et
al., 2004, 2006), religious beliefs may represent a particularly
effective means for doing so—a point that is illustrated vividly in
the quotation above. Believing in a God who exerts some sort of
control over the universe—or even a religious doctrine that posits
a particular spiritual order, such as Karma—should therefore be
particularly likely to surface as a compensatory resource during
times of lowered personal control.
Although there have been a number of correlational and, to a
much lesser extent, experimental investigations into the underpin-
nings of religious beliefs, none have tested the specific model we
offer here. A number of studies have investigated the links be-
tween religious beliefs and generalized threat or stressful and
negative life experiences (for a review, see Pargament, 2002).
Some of these studies are complementary to our approach, insofar
as stressful life experiences are often accompanied by reductions
in personal control. In fact, despite the generality of much of this
research, it is within this tradition that perhaps the most supportive
evidence for the theory we are proposing can be found.
Sales (1972, 1973), using archival data, conducted a fascinating
examination of the relationship between broad societal threats and
temporary increases in authoritarianism. Although Sales’s interests
lay in demonstrating the relationship between threatening experi-
ences and the display of behaviors typically associated with au-
thoritarianism, and not on the relationship between levels of per-
sonal control and the endorsement of external systems of control,
some of his data are very supportive of the latter position. Namely,
Sales demonstrated that during times of threat, such as the Great
Depression and other economic downturns, (a) conversion rates
into religious sects offering high levels of imposed order increased,
(b) conversion rates into religious sects offering low levels of
imposed order decreased, and (c) interest in supernatural sources
of order in general (as measured by the publication of books
dealing with such topics) increased. Although these data are of
course correlational, to the extent that such threatening and restric-
tive historical incidents may have reduced feeling of personal
control, it is suggestive of our general model.
Other findings within this tradition are also worth noting. Again,
although none of them test our exact model, and most are corre-
lational, many are supportive of the notion that believing in reli-
gious sources of control, such as God, can compensate for de-
creased perceptions of personal control. Work on elderly people,
for example, has demonstrated that as people age, their beliefs in
God-mediated control increase, an increase that is associated with
increased well-being (Krause, 2005). Indeed, this increase in be-
liefs in God-mediated control may explain why well-being does
not generally decrease with age, despite the fact that levels of
personal control do (Shapiro, Sandman, Grossman, & Grossman,
1995). Other research has demonstrated that marginalized social
groups tend to more strongly believe in religious doctrine (Argyle
& Beit-Hallami, 1975; Gurin, Veroff, & Feld, 1960; Pargament,
1997). To the extent that the life experiences of such groups can be
assumed to produce lower perceptions of personal control, these
findings are also in line with our theoretical model.
Finally, research on the relationship between well-being and the
different types of religiosity has also yielded some supportive
findings. For example, it has been demonstrated that the associa-
tion between intrinsic religiosity and adjustment to uncontrollable
life events is much stronger for those who ascribe to religious sects
that are high in faith (such as Protestantism). Along the same lines,
it has also been shown that irrespective of religious sect, those who
hold less superficial and more intrinsically driven beliefs in the
existence of God tend to experience more health and well-being
benefits. This pattern of data has been found in examinations of the
differing outcomes associated with intrinsic versus extrinsic reli-
giosity (Allport & Ross, 1967), internalization versus introjection
of religious ideals (Ryan, Rigby, & King, 1993), and positive
versus negative religious coping methods (Pargament, Smith, Koe-
nig, & Perez, 1998). Each of the latter religious orientations, which
involve more doubt as to the true existence of God, has been
shown to be less psychologically beneficial (for a review of these
findings, see Pargament, 2000).
Within the domain of terror management theory, there have
been several notable tests of the psychological function of reli-
gious beliefs (Greenberg et al., 1995; Jonas & Fischer, 2006;
Norenzayan & Hansen, 2006; also see Osarchuk & Tatz, 1973).
The emphasis in that work has been on the role that religion and
belief in an afterlife can play in reducing existential angst associ-
ated with the fear of death, and not, as we are suggesting, on the
role such beliefs can play in compensating for lowered perceptions
20
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
of personal control. We do not dispute the important role that
mortality concerns play in the adoption of religion, especially as it
pertains to beliefs in an afterlife and the cultural norms supported
by religious ideology. Our emphasis here is merely on a different,
and likely complementary, existential basis for the belief in God.
Although it is conceivable that mortality concerns (or, for that
matter, many other sets of concerns, including uncertainty needs,
beliefs in the ability of the individual to seek positive outcomes
and avoid negative ones, etc.) are themselves a significant con-
tributor to the desire to protect one’s self from perceptions of
randomness, the goal of the present set of studies is to gain a
deeper understanding of the role external systems may play in
helping us meet this desire to perceive order as opposed to ran-
domness. It is not on why we hold this desire—which is a topic
that has received considerable attention elsewhere (e.g., Dechesne
et al., 2000; Greenberg et al., 1995; Kay et al., 2002; Kruglanski,
1989; Kruglanski & Webster, 1996; Landau et al., 2004, 2006)—
nor on documenting the desire itself, which also has been widely
addressed (for reviews, see Janoff-Bulman, 1992; Skinner, 1995).
Compensatory Control, Governmental Control, and the
System Justification Motive
Dozens of experiments under the umbrella of system justifica-
tion theory (Jost & Banaji, 1994) have demonstrated that people
often defend, justify, excuse, and trust the social systems within
which they function (see Jost et al., 2004; Kay et al., 2007, for
reviews). Whether operationalized as society in general, the gov-
ernment, or specific institutions such as one’s university, people
have been shown to engage in psychological processes aimed at
defending the legitimacy of their social systems (Jost & Kay, 2005;
Kay et al., 2002; Kay & Jost, 2003; Kay, Jost, & Young, 2005;
Lau, Kay, & Spencer, 2008). Although several theoretical expla-
nations have been put forth to explain this phenomenon (see Jost
& Hunyady, 2002, 2005; Kay & Zanna, in press), there is a dearth
of empirical evidence offered to account for this motivated defense
of external systems.
The model of compensatory control we have proposed above
and the test of it we provide below can, we believe, help to explain
why people are motivated to place their faith in these types of
external systems—that is, because of their potential for compen-
sating for lowered levels of personal control. Although not tradi-
tionally viewed as such, governmental and organizational institu-
tions, we believe, can serve very similar existential needs to those
served by religious ideologies. Given the clear rules, guidelines,
norms, and structure formal systems provide, governments and
organizational systems, much like religions, hold the potential to
help people imbue their worlds with order and control. Indeed, it
has been explicitly suggested that existential needs are a key
underlying driver of the system justification motive (Jost & Hu-
nyady, 2005). To this end, although the first set of studies pre-
sented investigates the relation between personal control and be-
liefs in religious control, the later set tests this model specifically
in the context of defending societal and governmental systems,
making explicit the contribution of our model of compensatory
control to system justification theory.
Although believing in God and defending the legitimacy of
one’s social system may seem like conceptually unrelated depen-
dent measures, they both fit our main construct of interest: the
defense of external systems. When motivated to defend God, the
most pressing argument one generally needs to make is to con-
vince others (or one’s self) that he or she exists. When motivated
to defend governmental and sociopolitical systems, however, peo-
ple do not need to argue for their existence (as this is self-evident);
instead, in such cases, defending them is generally manifested as
defending their legitimacy and promoting their continued exis-
tence.
Overview of Studies
In five studies, we systematically tested the proposition that
individuals are more likely to endorse and defend external systems
of control, such as God and government, when personal control is
low or threatened. In Study 1, we experimentally manipulated
feelings of personal control (via a memory task, in which partic-
ipants were asked to remember events from their recent past over
which they did or did not have control) and then assessed the
effects of this manipulation on beliefs in the existence of God as an
external source of control. To the extent that beliefs in God serve
a compensatory control function, we predicted that beliefs in the
existence of God would increase after the manipulation designed
to temporarily decrease feelings of personal control. To more
precisely examine the compensatory control mechanism we sug-
gest underlies this effect, we also introduced a moderator variable
into this design. The dependent measure was varied such that half
the participants were asked about beliefs in the existence of God
framed in a way that emphasized a controlling nature (i.e., God as
controller), whereas the other half were asked about beliefs in the
existence of God framed in a way that deemphasized a controlling
nature (i.e., God as creator). We expected this to moderate our
findings, such that beliefs in the existence of God would show
significantly more movement as a function of the control manip-
ulation when the controlling, rather than creationist, nature of God
was emphasized in the dependent measure.
In Study 2, we sought to demonstrate that the effects of the
control manipulation on beliefs in the existence of a controlling
God occurred, at least in part, for the reason we presumed: the
threat that this manipulation posed to participants’ general beliefs
in order, structure, and nonrandomness in their world. Thus, we
introduced a mediator variable into our design that assessed the
extent to which the control manipulations threatened beliefs in
order and nonrandomness. If the effect of the control manipulation
on beliefs in the existence of God observed in Study 1 was due to
the threat it posed to participants’ overarching beliefs in the order,
structure, and nonrandomness that exists in their world, we ex-
pected the threat variable to mediate the relevant effect.
In Studies 3 and 4, we sought to demonstrate that processes of
compensatory control can also explain the defense of external
systems of control other than religion, such as governmental and
societal institutions. In Study 3, using a representative, cross-
national, and cross-cultural data set, we examined whether beliefs
in personal control predict support for governmental systems that
exert control over people’s lives. If it is the case that people rely
on governmental systems to compensate for lowered levels of
perceived personal control, then we would expect these two vari-
ables to be significantly and negatively correlated: Lower percep-
tions of personal control should predict higher beliefs in the extent
to which governmental systems should have control. Also, given
21
GOD AND THE GOVERNMENT
the moderator we have proposed in regards to perceived system
benevolence, we would expect this correlation to be moderated by
perceptions of government corruption, such that the relevant rela-
tion should be weakest in those countries with governments seen
as most corrupt.
In Study 4, we assessed the effects of our control manipulation
on explicit defense of the sociopolitical system (or system justifi-
cation), operationalized as resistance to societal change. Our pre-
diction was that participants in the low perceived control condition
would defend the sociopolitical system more strongly than those in
the high perceived control condition. Additionally, to test our
hypothesis that perceived system benevolence will moderate the
likelihood that a specific external system will be used to compen-
sate for lowered levels of perceived personal control, we included
a moderator variable in this design, namely, the extent to which
participants believe the relevant political system to be benevolent.
We predicted the perceived personal control manipulation would
be most likely to lead to increased system justification for those
who believe this particular system to be a benevolent force.
3
In Studies 1 to 4, we examined whether low or threatened
personal control will lead or relate to the endorsement and justi-
fication of external systems of control. In Study 5, we examined
the converse prediction that a threat to the justness and orderliness
of an external system of control would cause enhanced perceptions
of personal control, using a paradigm known to be effective for
measuring illusions of personal control (Alloy & Abramson,
1979). Our compensatory control model suggests that because
people are motivated to defend and justify external systems that
can add order and control to their lives, threatening the viability
and effectiveness of such social systems should lead to enhanced
perceptions of personal control.
Across these five studies, therefore, we aim to provide novel
empirical evidence for two very important, but still not satisfac-
torily understood, psychological phenomena: (a) the prevalence of
religious beliefs and (b) the operation of the system justification
motive.
Study 1
In Study 1, we experimentally assessed our hypothesis that
lowered levels of perceived personal control would predict higher
levels of belief in religious sources of control, that is, the existence
of God. In particular, we experimentally assessed whether a per-
sonal control manipulation could actually shift people’s reports of
the extent to which they believe God as a controlling entity exists.
Using a 2 2 factorial design, we exposed participants to one of
two memory tasks that asked them to recall recent positive events
over which they did or did not have control. All participants were
then asked to rate their beliefs in the existence of God. For half the
participants, the dependent measure referred to God in such a way
that deemphasized a controlling nature (God as creator), and for
the other half of the participants, the dependent measure referred to
God in such a way that emphasized a controlling nature (God as
controller). To the extent that beliefs in the existence of God can
serve as an external source of compensatory control, we expected
a two-way interaction to emerge, such that participants asked about
a controlling God would endorse the existence of God more
strongly following the no-control memory task, whereas partici-
pants asked about the existence of a creationist God would not be
affected by the manipulation. To adjust for preexisting levels of
religious endorsement, we also included a measure of religiosity.
Method
Participants. Thirty-six participants (19 women and 17 men;
17– 45 years of age; 31 Canadian and 5 other) were recruited from
a public venue on a southern Ontario university campus and
participated in exchange for a chocolate bar.
Procedure. Participants were asked to fill out a questionnaire
about self-thoughts and attitudes towards religion on site. The first
page in the questionnaire booklet was our manipulation of personal
control. Participants were exposed to one of two memory tasks
asking them to recall a recent positive event over which they did
or did not have control. Specifically, participants read and com-
pleted the following: “Please try and think of something positive
that happened to you in the past few months that was [not] your
fault (i.e., that you had [absolutely no] control over). Please
describe that event in no more than 100 words.”
On the following page, participants were asked to rate their
beliefs in the existence of God. Half of the participants completed
the dependent measure that referred to God in a way that deem-
phasized a controlling nature (God as creator). These God-as-
creator items were “To what extent do you think it is feasible that
God, or some type of nonhuman entity, created the universe?” and
“To what extent do you think that it is feasible that God, or some
type of nonhuman entity, created all life on the planet?” Items used
a 7-point response format (1 tremendously doubtful,7 very
likely) and were averaged to form a highly reliable composite
(␣⫽.96).
The other half of the participants completed the dependent
measure that referred to God in a way that emphasized God’s
controlling nature (God as controller). These God-as-controller
items were “To what extent do you think it is feasible that God, or
some type of nonhuman entity, is in control, at least in part, of the
events within our universe?” and “To what extent do you think that
the events that occur in this world unfold according to God’s, or
some type of nonhuman entity’s, plan?” Again, items used a
7-point response format (1 tremendously doubtful,7 very
likely) and were averaged to form a highly reliable composite
(␣⫽.94).
Last, participants completed a demographics page that assessed
their age, gender, and religiosity. The one-item measure of religi-
osity was “How religious do you consider yourself?” The item
used a 5-point response format (1 not at all religious,5
extremely religious).
Once finished, participants were probed for awareness or sus-
picion of our hypotheses and any presumed relation between our
independent and dependent variables, which none reported. Fi-
nally, they were thanked, debriefed, and given a chocolate bar as
payment.
Check on the manipulation of personal control. Using a sep-
arate sample of participants, we examined whether reporting pos-
3
In Studies 1, 2, and 4, our manipulation of perceived personal control
involved memories only for positive events. This was done to distinguish
our approach from those that emphasize the role that beliefs in God,
religion, and the government play in helping people to cope solely with
negative and traumatic life events and experiences.
22
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
itive events over which participants either did or did not have
control (a) affected their perceptions of personal control, as we
assumed it would; (b) did so in the direction we presumed; (c) did
not produce contrast effects in the belief that everybody and
everything has more control (recall that our prediction is that
threats to personal control are only compensated via the defense of
external systems that the self may be under the control of, not just
any system); and (d) did not affect other variables that might have
been related to our effects, such as mood or self-esteem.
Using a neutral cover story (i.e., study of daily events), we asked
participants to complete one of our two manipulations, embedded
within other questions about daily events (e.g., “Please list the last
three movies you saw”). They were then asked to complete one of
two sets of dependent measures. In one sample (n 29), partic-
ipants were asked to rate their agreement (1 strongly disagree,
7 strongly agree) with four items related to beliefs in the control
of irrelevant agents (e.g., “The Chinese government has control
over its citizen’s lives,” “Seniors (i.e., people over 70 years of age)
have control over their lives,” and “Hockey coaches have control
over their team” [this was, after all, conducted in Canada]), and
items related to their control over their own lives (i.e., “The events
in my life are mainly determined by my own actions” and “I am
not in control of most things that occur in my life” [reverse
scored]). In another sample (n 39), participants were asked to
complete standard measures of self-esteem (Rosenberg, 1979) and
mood (Positive and Negative Affect Schedule; Watson, Clark, &
Tellegen, 1988).
As expected, participants who were asked about a time when
they did not have control reported lower levels of belief in personal
control (M 4.1) compared with participants who were asked
about a time when they did have control (M 4.9), t(27) 2.45,
p .02, d 1.10. However, there was no effect of our manipu-
lation in terms of generalized control of irrelevant agents, t(29)
1, ns, nor in terms of self-esteem, t(37) 1, ns; positive mood,
t(37) 1; or negative mood, t(37) 1. These findings confirm
that our manipulation is indeed a specific threat to participants’
sense of personal control.
Results
We conducted a two-way univariate analysis of variance in
which the memory for instances of personal control manipulation
(two levels: control vs. no control) and the moderator (God as
controller vs. God as creator) were entered as fixed factors and
religiosity was entered as a covariate.
4
The dependent measure was
belief in the existence of God (or similar supernatural beings).
The analyses yielded a main effect of religiosity, indicating that
participants who were more religious more strongly endorsed the
existence of God, F(1, 31) 66.65, p .001. Although no other
main effects were obtained, the two-way interaction between the
control–no-control memory manipulation and the dependent mea-
sure manipulation also attained statistical significance, F(1, 31)
6.32, p .02.
As can be seen in Figure 1, this interaction unfolded in the
predicted direction. Whereas the control versus no-control memory
manipulation exerted no effect on participants’ ratings of the
existence of God when God’s controlling nature was deempha-
sized (adjusted Ms 4.27 and 4.06, respectively), F(1, 14) 0.03,
ns, when God’s controlling nature was emphasized, the personal
control–no personal control memory manipulation exerted the
hypothesized effect: Participants who first remembered events
over which they had no control were significantly more likely to
subsequently believe in the existence of a God who has control
than those participants who first remembered events over which
they had control (adjusted Ms 4.89 and 3.08, respectively), F(1,
16) 5.12, p .04.
Discussion
Study 1’s findings support our compensatory control model.
Participants asked to recall events from their recent past over
which they had no control subsequently reported stronger beliefs
in the existence of (a controlling) God (or a similar supernatural
controlling influence) than those participants asked to recall in-
stances of controllability from their recent past. Given the extent to
which beliefs in the existence of God are something most people
have spent a good deal of time thinking about in their lives, the fact
that our manipulation significantly shifted these beliefs is rather
striking. In addition, and in line with our theoretical explanation,
this effect was also moderated by the manner in which God was
framed within the dependent measure: Only when the controlling
nature of God was emphasized did participants show the relevant
effect.
Why in the God-as-controller condition did we observe increas-
ingly strong beliefs in the existence of God following the no
personal control manipulation? We believe this was due to the
threat this manipulation posed to participants’ overarching sense of
4
The manipulation of personal control did not significantly affect this
covariate (Ms 2.93 and 2.59 in the no-control and control conditions,
respectively). Theoretically, this is not surprising. Our predicted effect of
the control manipulation on beliefs in the existence of God is specific to
God as a controlling agent (indeed, we are not predicting it will affect
ratings on beliefs in a strictly creationary God). A general measure of
religiosity such as the one we used as the covariate likely gauges several
aspects of religiosity unrelated to beliefs in a controlling God (see Allport
& Ross, 1967).
Figure 1. Mean endorsement of God as controller or creator as a function
of personal control (adjusted for religiosity; Study 1).
23
GOD AND THE GOVERNMENT
order and structure in their environment and the usefulness of God
as an external source of control in compensating for this threat.
That the findings from Study 1 were moderated by the extent to
which God was framed as a controlling entity is supportive of this
reasoning. However, this notion can be addressed more directly. In
particular, a mediational design that measures the extent to which
the control manipulation actually threatened overarching beliefs in
order and structure and the extent to which such threat, in turn,
influenced beliefs in the existence of God, would accomplish this.
Study 2 provides such a test.
Study 2
When important motivations or core beliefs are threatened,
people often respond defensively, strengthening their original con-
victions and beliefs through both direct and indirect means. For
example, when meaningful identities, such as one’s self- or group
identity, are attacked, the result is often an increased belief in the
worth of that particular identity (e.g., Crocker & Luhtanen, 1990;
Dunning, 2003; Mullen, Brown, & Smith, 1992; Tesser, Crepaz,
Beach, Cornell, & Collins, 2000). Similarly, when epistemic belief
systems are threatened, such as beliefs in personal uncertainty
(McGregor & Marigold, 2003; McGregor, Zanna, Holmes, &
Spencer, 2001), system legitimacy (Jost et al., 2004), cognitive
consistency (Tesser et al., 2000), and belief in a just world (Lerner,
1980), similar processes of defensive zeal and conviction often
ensue.
One way to test whether a particular manipulation has threat-
ened a given motivated belief, therefore, is to assess the extent to
which the relevant threat manipulation produces defensive convic-
tion in the domain it is presumed to threaten (in this case, general
beliefs in nonrandomness and order). Thus, if the personal control
manipulation used in Study 1 threatened the overarching motiva-
tion to guard against randomness and chaos in one’s environment
(via lowered levels of personal control), then following that ma-
nipulation, we would expect participants to react by increasingly
defending their beliefs in generalized order and structure in their
lives (or, likewise, by increasingly denying the role of chance).
Moreover, if this threat was what caused participants to increas-
ingly believe in the existence of a controlling God, then we would
also expect the magnitude of this defensive response to mediate the
effect of the personal control manipulation on the relevant depen-
dent measure—belief in the existence of a controlling God. Thus,
in Study 2, we introduced a measure into our design designed to
tap explicit denial of randomness and chance in the participants’
lives.
The pretest data in Study 1 demonstrated that the personal
control manipulation lowered feelings of personal control, and so
one might assume it should have the same effects on a scale
measuring general beliefs in nonrandomness. However, beliefs in
personal control and beliefs that the world is nonrandom do not sit
at the same end of one continuum. Why is this? Although beliefs
in personal control are one way to fulfill the overarching motiva-
tion to believe the world is not random, many other beliefs can also
serve the same function. In other words, although increased levels
of perceived personal control naturally imply increased levels of
perceived nonrandomness, decreased levels of perceived personal
control do not necessarily imply decreased levels of perceived
nonrandomness.
As an example, imagine an individual, such as a devout Cal-
vinist, who very strongly believes God is in complete control of his
fate. This person would be very low on a scale of perceived
personal control, but also very high on the overarching belief that
the world is nonrandom—that is, he could very easily see himself
as having little personal control without also believing that the
world operates randomly. These two constructs, therefore, al-
though related, should not be thought of as one and the same.
Indeed, as will be seen, and as our model predicts, the same
manipulation can have different, even opposite, effects on mea-
sures designed to gauge beliefs in personal control versus beliefs
that outcomes are due to randomness or accidental happenings.
Method
Participants and procedure. Thirty-eight participants (12
women, 10 men, and 16 gender unknown; 19 –23 years of age; 16
Canadian, 6 other, and 16 unknown; 10 Christian, 4 not religious,
5 other, and 19 religion unknown) completed our materials in
exchange for course credit. All participants were recruited from a
university campus in southern Ontario. Participants were told that
the study concerned how individuals think about themselves and
various social institutions. All of the measures were completed
online.
First, participants were exposed to one of the same two memory
task manipulations of personal control that were used in Study 1.
On the following Web page, participants completed two items that
were designed to assess feelings of perceived randomness in their
lives. The two items, each answered on 7-point scales (ranging
from 1 tremendously doubtful to 7 very likely), were “To a
great extent my life is controlled by accidental happenings” and
“The things that occur in my life are mostly a matter of chance.”
The responses to these items were reverse coded so that higher
responses indicated increasing defense against randomness and
averaged to form a reliable “defensive reactions” composite
(␣⫽.88).
Last, to measure beliefs in the existence of God, participants
were asked “To what extent do you think it is feasible that God, or
some type of nonhuman entity, is in control, at least in part, of the
events within our universe?” The item used a 7-point response
format (1 tremendously doubtful,7 very likely). Once fin-
ished, participants were thanked for their participation and de-
briefed.
Results
To investigate whether the personal control manipulation did in
fact produce defensive reactions regarding overarching views of
the role played by randomness and chance in the participants’
lives, and whether these reactions mediated the effect of the
personal control manipulation on belief in the existence of God as
an external source of control, we conducted a series of separate
regressions, as depicted in Figure 2.
Two separate regressions showed an effect of the personal
control manipulation on the denial of randomness, ␤⫽0.362,
t(36) 3.13, p .003, and the belief in the existence of a
24
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
controlling God, ␤⫽0.278, t(36) 1.78, p .086,
5
indicating
that following the low personal control manipulation, compared
with the high personal control manipulation, participants more
strongly denied randomness and chance (Ms 4.12 and 3.05,
respectively) and more strongly endorsed the existence of God
(Ms 4.78 and 3.70, respectively).
When personal control and defensive reactions were entered
simultaneously into the regression to predict the belief in the
existence of a controlling God, the direct association between
personal control and the belief in the existence of God was greatly
diminished, whereas the association between defensive reactions
and the belief in the existence of God remained statistically sig-
nificant, ␤⫽0.132, t(36) 1, ns, and ␤⫽0.415, t(36) 2.61,
p .013, respectively. To test the significance of the indirect path,
we used the bootstrapping procedure described by Preacher and
Hayes (2004), who suggested that the bootstrapping method is the
most appropriate option when testing indirect paths with small
samples. The bootstrapping procedure was used to test the null
hypothesis that the indirect path from personal control to the belief
in the existence of a controlling God through defensive reactions
was not significantly different than zero. As per the bootstrapping
procedure, the 95% confidence interval values did not cross zero
(1.64 and 0.01), which confirms that the defensive reactions
did mediate the effect of the personal control manipulation on
beliefs in the existence of God ( p .05). To rule out a possible
reverse mediation interpretation (i.e., control manipulation 3 be-
liefs in God 3 denial of randomness), we also conducted analyses
of the reverse mediation path. This path did not reveal significant
mediation, allowing for confidence regarding the direction of this
effect.
Discussion
Study 2, therefore, provides further support for our theoretical
reasoning. The personal control manipulation was shown to be
threatening to participants’ overarching sense of order and struc-
ture (insofar as it engaged defensive reactions surrounding the role
that chance and randomness play in the participants’ lives), and the
extent of these defensive reactions mediated the effect of the
personal control manipulation on beliefs in the relevant external
source of control. In Studies 3, 4, and 5, we test this model in the
context of other potential sources of external control and also make
explicit the potential for our model to contribute to our understand-
ing of other important, but not fully understood, social psycholog-
ical phenomena, such as system justification.
Studies 3 and 4
Beyond religion, processes of compensatory control should also
generalize to other external systems that can offer order and
control to people’s lives. System justification theory research has
documented numerous cases in which people defend external
systems of control, ranging from their governments to societal
norms and institutions (for reviews, see Jost et al., 2004; Kay et al.,
2007). Why might people do so?
Despite the abundance of research on system justification phe-
nomena, there are virtually no experimental tests of why this
occurs. We believe that the motivated defense of societal arrange-
ments and the systems and institutions that uphold these arrange-
ments (i.e., system justification) may represent a means of com-
pensatory control. That is, we believe the compensatory control
model is generalizable beyond religion; just as we have observed
in Studies 1 and 2 in the context of religion, people should also
endorse and defend other types of external sources of control—
such as the government and prevailing sociopolitical systems—to
compensate for fluctuating levels of personal control.
In Study 3, we first test this hypothesis correlationally. Using
nationally representative samples from 67 countries, we assessed
the extent to which beliefs in personal control predict support for
increased governmental intervention in people’s lives. We pre-
dicted that these variables would correlate negatively and signifi-
cantly: The less people perceive personal control over their own
lives, the more they will endorse governmental control. In addi-
tion, to assess our prediction that these effects should be strongest
for those external systems perceived as benevolent, we investi-
gated the moderating role of reports of government corruption.
In Study 4, we test this hypothesis again. In this study, however,
we used an experimental rather than a correlational methodology,
introduced a moderator to the design (i.e., perceived system be-
nevolence), and used a different measure of support for the socio-
political system (i.e., resistance to system change).
Study 3
In this study, we used data from the European and World Values
Survey to examine whether low feelings of personal control are
associated with increased dependence on governmental institu-
tions. The purpose of this study is twofold. First, it is important to
establish that the findings from Studies 1–3 are not limited to
beliefs about God, per se, but rather that low personal control
should be associated with greater dependence on external systems
of control in general (see also Rothbaum et al., 1982). Second, if
this process is part of basic psychological functioning, it should not
be limited to a specific culture or type of governmental system. Of
course, the concept of personal control may vary in different
cultures around the world, but we expected to find that the rela-
tionship is at least not limited to one particular governmental
system. Furthermore, in this study, we tested the notion that
external sources of control will be most useful in compensating for
5
Although this effect is only marginal, when meta-analyzed across
Studies 1 and 2, the effect of our personal control manipulation on the
belief in God produces a moderate to large effect size and is highly
significant (d 0.634, Z 2.32, p .01).
Figure 2. Mediation model. The no personal control and personal control
conditions were dummy coded as 1 and 0, respectively (Study 2).
p .10.
*
p .05.
25
GOD AND THE GOVERNMENT
low personal control to the extent the external source is deemed
benevolent.
Method
Data for this study came from the third and fourth waves of the
World Values Survey (World Values Survey Association, n.d.).
These surveys were conducted by face-to-face interviews in coun-
tries around the world from 1994 to 2003, and the sample is
representative of about 85% of the world’s population (see Ingle-
hart, Basan˜ez, Diez-Medrano, Halman, & Luijkx, 2004, for a more
detailed description of this dataset).
Participants. The variables that we were interested in (de-
scribed below) were available for 93,122 individuals from 67
countries.
6
Person-level variables. The dependent measure, increased
support for governmental control, was operationalized with one
item that asked respondents to place themselves on a scale ranging
from1to10(1 People should take more responsibility to
provide for themselves, 10 The government should take more
responsibility to ensure that everyone is provided for).
Personal control was operationalized with one item that read,
“Some people feel they have completely free choice and control
over their lives, while other people feel that what they do has no
real effect on what happens to them”; respondents were asked to
indicate how much control they feel on a scale ranging from 1
(none at all)to10(a great deal).
In addition to our two main variables of interest, we included
many adjustment variables in our models, including age, income,
education, and sex. Age was coded in six categories (15–24,
25–34, 35– 44, 45–54, 55– 64, and 65 and older). Income was
coded with three steps representing the lower, middle, and upper
income levels on the basis of the standard of living in the survey
country. Education was coded in eight categories representing the
level of schooling obtained. Sex was dummy coded so that males
were 0 and females were 1. Because views about government
responsibility could be, in part, ideological, we also adjusted for
left–right political orientation in all of our models. This was
measured on a scale ranging from 1 (left wing)to10(right wing).
National-level variables. We used the 1999 Corruption Per-
ceptions Index from Transparency International (n.d.) as a measure
of perceived benevolence on a country level. The Corruption
Perceptions Index is calculated on a yearly basis for a large set of
countries and is considered one of the most reliable cross-national
comparison indexes (see You & Khagram, 2005). The index
ranges from 1 (most corrupt)to10(least corrupt).
The data were weighted with a weight provided by the European
and World Values Survey to reflect national distributions of de-
mographic variables. All continuous variables were centered on
the scale midpoint.
Results
We tested a two-level model in HLM version 5.05 (Raudenbush,
Bryk, & Congdon, 2001). On the individual level, we included
personal control, left–right orientation, and demographic adjust-
ments as predictors of preference for governmental responsibility.
On the country level, we adjusted the intercept with the national
mean of the government responsibility item to account for country-
level differences in how much (or how little) the government is
perceived to be taking responsibility for. We also included random
effects for the intercept and slope between personal control and
preference for governmental responsibility.
The results from this model (Model 1) are shown in Table 1.
Above and beyond the effects of random variation between coun-
tries (
2
.01, p .001) for both the intercept and slope, personal
control is significantly and negatively associated with preference
for governmental responsibility (b ⫽⫺0.07, SE .01, p .001),
supporting the notion that those with a low sense of personal
control are more likely to prefer governmental involvement.
According to the compensatory control model, people with low
feelings of personal control should be most likely to look to a
given external source of control to the extent it is seen as benev-
olent. To examine this, we ran a second multilevel model with
Transparency International’s (n.d.) Corruption Perceptions Index
included as a moderator of the relationship between personal
control and governmental responsibility.
As shown in Table 1, the results from this model (Model 2)
support our reasoning. Above and beyond the effects of random
variation between countries (
2
.01, p .001, for the intercept
and
2
.005, p .001, for the slope between personal control
and governmental responsibility), we found that personal control
was again significantly and negatively associated with preference
for governmental responsibility (b ⫽⫺0.07, SE .01, p .001).
This effect is qualified by a significant interaction between per-
sonal control and governmental benevolence (b ⫽⫺0.01, SE
.00, p .05). As shown in Figure 3, individuals living in countries
with benevolent (as opposed to corrupt) governments are much
more likely to turn to the government when personal control is
lacking. Simple slope analyses (Preacher, Curran, & Bauer, 2006)
show that the relationship between personal control and preference
for government responsibility is negative and significant for all
countries, but is stronger in countries with benevolent governments
(b ⫽⫺0.14, SE .03, p .001) as compared with corrupt
governments (b ⫽⫺0.09, SE .01, p .001).
6
Specifically, the analyses include the following countries: Albania
(n 1,787); Argentina (n 1,530); Armenia (n 1,448); Australia (n
1,589); Austria (n 992); Belgium (n 1,256); Azerbaijan (n 1,154);
Belarus (n 1,619); Brazil (n 960); Bulgaria (n 1,342); Canada (n
1,488); Chile (n 1,721); Colombia (n 2,732); Croatia (n 764); Czech
Republic (n 2,391); Denmark (n 768); El Salvador (n 821); Estonia
(n 1,334); Finland (n 1,575); France (n 1,058); Georgia (n
1,561); Germany (n 2,818); Great Britain (n 507); Greece (n 829);
Hungary (n 753); Iceland (n 811); India (n 1,908); Indonesia (n
727); Ireland (n 695); Italy (n 1,205); Japan (n 897); Jordan (n
393); Korea (n 1,131); Kyrgyzstan (n 946); Latvia (n 1,517);
Lithuania (n 1,262); Luxembourg (n 447); Macedonia (n 1,220);
Mexico (n 2,580); Moldova (n 1,358); Morocco (n 479); Nether-
lands (n 901); New Zealand (n 700); Nigeria (n 3,375); Norway
(n 1,015); Pakistan (n 177); Peru (n 2,060); Philippines (n
1,133); Poland (n 806); Romania (n 1,479); Russia (n 2,516);
Serbia (n 1,645); Slovakia (n 1,814); Slovenia (n 490); South
Africa (n 4,187); Spain (n 1,909); Sweden (n 1,804); Switzerland
(n 758); Tanzania (n 744); Taiwan (n 701); Turkey (n 5,521);
Uganda (n 69); Ukraine (n 1,854); United States (n 2,268);
Uruguay (n 833); Venezuela (n 1,645); Vietnam (n 891); and
Zimbabwe (n 699).
26
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
Discussion
In this study, we found additional evidence that is consistent
with the compensatory control model. By examining people’s
feelings about government responsibility, we have shown that the
need for personal control is associated with increased support for
external systems of control, not only religion. The relationship
between feelings of personal control and support for governmental
control was negative and significant after adjusting for random
variation between countries and was above and beyond the effects
of ideological orientation.
We also predict that people who have low levels of personal
control will be most likely to turn to a given external system of
control to the extent it is deemed benevolent. The data from this
study seem to support this notion. For those living in countries in
which the government is perceived as benevolent (as measured by
lack of corruption), the relationship between personal control and
preference for governmental responsibility was stronger than for
those living in countries in which the government is perceived as
corrupt. Still, it should be noted that even for those countries
deemed highly corrupt, the relationship remained significant (al-
beit at a considerably weaker level). This is a testament to the
power of the motivation to guard against randomness.
Study 4
Not all external sources of control are equally likely to be relied
on following a threat to personal control. Some should lend them-
selves more to this role than others. Although there are likely many
factors that can determine this, one likely candidate is the extent to
which an external system is perceived as benevolent— or as rep-
resenting its constituents’ best interests. Because our two experi-
mental studies (Studies 1 and 2) involved an external source of
control that—if it exists—is mostly thought to be benevolent, those
studies did not provide the ideal context in which to test this
particular aspect of our model. That is, although beliefs in the
extent to which God exists certainly run the gamut, beliefs about
the extent to which God, in real or fictional form, cares about the
welfare of humanity obviously vary much less.
7
Beliefs in the benevolence of governmental and federal systems
and institutions of control, however, are likely to show much more
of a range—indeed, system justification theory does not posit that
all people tend to uniformly believe that their systems hold their
best interests at heart, just that structural aspects of systems are
often deemed more justifiable than one would expect and that
people tend to resist large systemic changes (Jost et al., 2004; Kay
& Zanna, in press). Thus, in our fourth study, we experimentally
tested the extent to which the tendency to defend one’s sociopo-
litical federal system following a threat to personal control would
be moderated by views of the benevolence of said system.
To this end, in Study 4, we first threatened participants’ sense of
personal control and then gauged the extent to which participants
engaged in system justification (operationalized as resistance to
system change). We also measured participants’ beliefs in the
benevolence of their national system. We expected responses to
this measure to moderate the effects of the personal control ma-
nipulation on system justification, such that those participants who
perceive the system to be most benevolent would be most likely to
defend it following a threat to personal control.
Method
Participants. Twenty-eight participants (12 women and 16
men; 16 –27 years of age; 15 Canadian, 8 East Asian, 4 other, and
1 unknown) were recruited from a public venue on a southern
Ontario university campus and participated in exchange for a
chocolate bar.
7
Many conceptions of the Judeo-Christian God involve a God who is
vengeful, strict, and punishing; this does not, however, imply any lack of
benevolence or interest in the welfare of humanity. One can be strict,
punishing, and vengeful with the intentions of doing so for the greater
good. A strict parent is an excellent example of this. Of course, there may
be individuals or groups who view God as nonbenevolent. For these people,
we would not expect a replication of the effects from Studies 1 and 2.
Figure 3. Preference for governmental responsibility as a function of
personal control and government benevolence (Study 3).
Table 1
Feelings of Personal Control and Governmental Benevolence
(Vs. Corruption) as Predictors of Preference for Governmental
Responsibility in 59 Countries (Study 4)
Predictors
Fixed effects b (SE)
Model 1 Model 2
Country level
Intercept 0.41 (.02)
**
0.42 (.02)
**
Mean government
responsibility 0.96 (.02)
**
0.96 (.02)
**
Government benevolence
(vs. corruption) index 0.00 (.01),ns
Individual level
Sex 0.14 (.02)
**
0.14 (.02)
**
Age 0.01 (.01),ns 0.01 (.01),ns
Income 0.17 (.02)
**
0.17 (.02)
**
Education 0.08 (.01)
**
0.08 (.01)
**
Right-wing orientation 0.07 (.01)
**
0.07 (.01)
**
Personal control 0.07 (.01)
**
0.07 (.01)
**
Cross-level interaction
Personal Control
Government
Benevolence 0.01 (.00)
*
*
p .05.
**
p .01.
27
GOD AND THE GOVERNMENT
Procedure. Participants were asked to fill out a questionnaire
about self-thoughts and social institutions on site. First, partici-
pants were exposed to one of the same two memory task manip-
ulations of personal control that were used in Studies 1 and 2.
Participants then completed four items that were designed to
assess their beliefs in the benevolence of the national system. The
four items, each answered on 9-point scales ranging from 1
(strongly disagree)to9(strongly agree), were “The current federal
government has my best interests at heart,” “I feel the current
federal government, all in all, is doing a very good job,” “The
current federal government ensures that its citizens are, for the
most part, well taken care of,” and “All in all, I feel that the current
federal government is run for the benefit of its citizens.” The
responses to these items were averaged to form a reliable “benev-
olence of national system” composite (␣⫽.88).
8
Next, participants completed three items that were designed to
assess defense of the current sociopolitical system, operationalized
as resistance to system change. The three items, each answered on
9-point scales ranging from 1 (strongly disagree)to9(strongly
agree), were “Societal change is disruptive,” “When societal
change occurs, the consequences of this change on society are
almost always negative,” and “Promoting national stability should
be the Canadian government’s main goal.” The responses to these
items were averaged to form a “system justification” composite
(␣⫽.62).
Once finished, participants were probed for awareness or sus-
picion of our hypotheses or any presumed relation between our
independent and dependent variables.
9
Finally, participants were
thanked, debriefed, and given a chocolate bar as payment.
Results
We hypothesized that (a) people would increasingly defend the
legitimacy of their sociopolitical system—that is, they would resist
system change—following a threat to personal control and (b) this
effect would be moderated by views of the benevolence of this
system. To test this, we used regression analyses to predict par-
ticipants’ resistance to social change. Predictors were the control–
no-control memory manipulation (contrast coded), mean-centered
ratings of the benevolence of the sociopolitical system, and the
interaction between these two variables. There were no main
effects of the control–no-control memory manipulation or ratings
of the benevolence of the national system on participants’ resis-
tance to system change. However, the predicted two-way interac-
tion between the personal control–no-personal control memory
manipulation and ratings of the benevolence of the national system
attained statistical significance, ␤⫽⫺0.467, t(25) ⫽⫺2.64,
p .015.
As can be seen in Figure 4, this interaction unfolded in the
predicted direction. For those participants who did not perceive the
national system to be benevolent, the control manipulation did not
affect defense of this external system of control, ␤⫽0.297,
t(25) 1.19, ns. For those who believed this system to be
benevolent, however, the control manipulation exerted the hypoth-
esized effect: Participants defended their sociopolitical system to a
significantly stronger extent following the no personal control
manipulation compared with the personal control manipulation,
␤⫽⫺0.691, t(25) ⫽⫺2.63, p .015.
Discussion
This fourth study further corroborates our compensatory control
model of support for external systems in two important ways. First,
and most generally, it provides experimental evidence for our
proposition that people will respond to threats to personal control
via increased support for external systems in a context outside of
religious beliefs. In this study, following a threat to personal
control, participants increasingly defended their sociopolitical sys-
tem. This finding both complements the correlational data regard-
ing the relation between chronic levels of perceived personal
control and support for governmental control attained in Study 3
and also provides a noteworthy link to system justification theory
(Jost & Banaji, 1994).
Second, we obtained important experimental evidence support-
ing our prediction that the more an external system of control is
deemed benevolent—that is, perceived to hold the goal of serving
a system’s constituents to the best of its ability—the more likely it
will be relied on during times of low perceived personal control.
Indeed, those participants who did not a priori deem the Canadian
sociopolitical system as representative of their interests did not
increasingly defend it following the personal control manipulation.
This suggests a key, theoretically sensible moderator. We would to
remind the reader, however, that we are not suggesting that per-
ceived system benevolence is the only moderator of this phenom-
enon. Other factors, such as perceptions regarding the system’s
ability to competently and effectively meet its goals, are also likely
to play a moderating role.
Study 5
Our first four studies provided consistent and converging evi-
dence that people are motivated to enhance their faith in external
8
It is important to note that there were no effects of our manipulation on
the benevolence of the national system measure, F(1, 24) 0.025, ns.
9
Two participants suspected a connection between the independent and
dependent measures and were therefore excluded from the analyses. Nei-
ther participant, however, suspected our specific prediction.
Figure 4. Mean resistance to system change scores as a function of the
personal control manipulation and belief in the benevolence of govern-
ment. Belief in the benevolence of the government was plotted at 1
standard deviation below and 1 standard deviation above the mean (Study 4).
28
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
systems of control—such as in God and the government—as
means of compensating for lowered levels of personal control.
Given that the primary purpose of this article is to test and develop
a theoretical model that can help explain why people so often
endorse and defend external systems of control, we have chosen to
test this particular direction of causality in the majority of our
experimental studies. However, if our general theoretical model is
correct, the converse should also be true: That is, if endorsing
external systems of control and believing in personal control do in
fact serve substitutable means (i.e., maintaining overarching be-
liefs in nonrandomness and order), then when external systems of
control are challenged, people should increase their perceptions of
personal control. Thus, to offer even more converging evidence for
the general model tested in Studies 1–4, we included a fifth study
that experimentally tests this reverse path.
To test this hypothesis, in Study 5 we presented participants
with a video that portrayed the governmental system as either
capable or incapable of restoring order following an injustice and
then measured illusory perceptions of personal control. If our
model of compensatory control is in fact correct, and beliefs in
external systems of control are, to some extent, substitutable with
perceptions of personal control, then exposing participants to an
instance in which the system could not help to alleviate a random
injustice should cause these participants to enhance their percep-
tion of personal control. That is, we should observe the converse
effect of that obtained in Studies 1– 4.
Following the video presentation, we examined participants’
perceptions of personal control by using a modified version of
Alloy and Abramson’s (1979) classic contingency task, which has
been successfully used in a number of studies examining illusions
of control (e.g., Alloy, Abramson, & Viscusi, 1981; Presson &
Benassi, 2003; Tennen & Sharp, 1983; Thompson et al., 2004). In
this task, we asked participants to “control” the onset of a green
circle appearing on a computer screen by either pressing the space
bar or not. After the task was completed, participants judged the
extent to which they believed they had controlled the onset of the
green circle.
If enhanced personal control serves a system justifying function,
then participants who learn that the system could not rectify the
injustice should engage in the compensatory process of believing
that they had more control over the onset of the green circle than
participants who learn that the system was effective (despite the
fact that the amount of control they actually had over the appear-
ance of the green circle was held constant by the experimenter).
Method
Participants. Fifty-one undergraduate students participated
for course credit. Two participants’ data were removed because of
their suspicions and another participant’s data were removed be-
cause the participant failed to follow the proper procedures. The
resulting sample consisted of 41 women and 7 men with a mean
age of 20.75 years (SD 2.85).
Procedure. Participants entered the laboratory under the guise
that the study was about attentional distractions and the processing
of emotional cues. They were told that they would watch an
emotionally involving video presentation and then complete an
attention-distracting problem-solving task before answering ques-
tions about the video. The problem-solving task was a modified
version of Alloy and Abramson’s (1979) illusions of control par-
adigm. Participants were told that that the problem-solving task
was designed to temporarily shift their attention away from the
video presentation. To facilitate the credibility of the cover story,
participants then completed a bogus Emotions and Feelings of
Others Scale that we developed.
Participants were then given instructions on how to complete the
problem-solving task. Specifically, they were told that they were
required to make a green circle appear on a computer screen on the
basis of whether they pressed the space bar or not. Participants
were informed that they could press or not press the space bar and
that the green circle would appear or not appear. Because it was
their task to make the green circle appear, we told the participants
that it would be to their advantage to press the space bar on some
trials and not on other trials to see what happened when they did
and did not press the space bar. Participants then completed four
practice trials of the task.
Participants were then given a description of the video presen-
tation. They were informed that the video presentation involved a
two-part interview with a young woman named Kerry who is
infected with HIV (Fisher & Fisher, 1992). They were told that
because of time constraints, they would only be able to view a few
clips from the initial interview session, but that they would receive
a summary after the video presentation of what the young woman
discussed in the follow-up interview. Before the video, participants
were told that Kerry contracted HIV after protected sex in which
the condom broke. The experimenter then started the video pre-
sentation and left the room. The video ran for approximately 4
min. During the video, Kerry discusses the many ways in which
her life has now been negatively affected as a result of this extreme
misfortune. Kerry also discusses how the system has tried to come
to her aid by prescribing various antiviral medications. (In Canada,
the medical system is wholly governmental, and all our partici-
pants were Canadian.) Once the video was finished, the experi-
menter reentered the room and orally administered the manipula-
tion. In the ineffective system condition, participants learned that
the treatment recommended by the medical (i.e., governmental)
system had done little to minimize the consequences of Kerry’s
misfortune. In the effective system condition, participants learned
that the recommended course of action had effectively and signif-
icantly minimized her suffering and bad luck—the virus was now
in a total state of remission.
Using this manipulation, Callan, Ellard, and Nicol (2006) found
that knowledge of the ineffectiveness (vs. effectiveness) of the
treatment led to an implicit activation of the justice motive as
revealed by a modified Stroop task (cf. Hafer, 2000; Kay & Jost,
2003). Moreover, despite the potential association between HIV
and death, during the modified Stroop task participants did not
evidence greater concerns with mortality as a function of the
system’s effectiveness in alleviating the young woman’s unjust
suffering.
Participants then completed the problem-solving task, which
consisted of 40 trials. Participants were reminded that their task
was to make the green circle appear on the screen on the basis of
their responses or nonresponses. The green circle appeared on the
screen 75% of the time for each participant. Because we were
interested in perceptions of control between system threat condi-
tions, each participant received the same ordering of green circle
appearance or nonappearance, which was determined by a random
29
GOD AND THE GOVERNMENT
list generated before the commencement of the study. Each trial
during the task began with a beeping sound that signaled to
participants that they had to choose to press the space bar or not.
When the circle appeared, it remained on the screen for 500 ms and
was followed by an intertrial interval that ranged between 500 ms
and 2,500 ms. The green circle appeared at the center of the screen
and was 4 cm (1.5 in.) in diameter. On the trials in which the circle
did not appear, the participants saw a blank screen until the
beginning of the next trial.
Immediately following the problem-solving task, participants
were asked to complete a survey about the task itself, ostensibly
because we were still testing its attention-shifting capabilities and
were interested in their opinions of it. In the survey, participants
were asked questions about how often they thought about the video
presentation while completing the task and whether they thought
the task effectively shifted their attention. Within this survey,
participants completed an item assessing our critical dependent
measure of compensatory control. Participants were asked to rate
on a scale ranging from 1 (no control)to10(complete control)
how much control they had over the onset of the green circle
during the task. Participants then completed an Emotional Cues
Questionnaire and a number of items related to the video to further
facilitate the credibility of the cover story. Finally, participants
were debriefed and thanked for their time.
Check on the Manipulation of System Effectiveness
Using a separate sample of participants (n 26), we examined
whether people’s concerns associated with the medical system’s
ineffectiveness in alleviating the young woman’s suffering were
specific to fairness or were more generally related to self-esteem or
self-enhancement concerns. Research has demonstrated that illu-
sions of control in nondepressed individuals stem, in part, from the
desire to protect self-esteem (e.g., Alloy & Abramson, 1982;
Koenig, Clements, & Alloy, 1992; Mikulincer, Gerber, & Weisen-
berg, 1990; but see Msetfi, Murphy, & Simpson, 2007). Thus, if
our manipulation of system threat produces changes in self-esteem
or self-enhancement when the system is ineffective, then our
predicted results (i.e., increased illusions of personal control when
the system is ineffective) might be more straightforwardly ex-
plained in terms of self-esteem maintenance. Thus, in this manip-
ulation check study we included measures of state self-esteem and
self-enhancement to examine whether our manipulation alterna-
tively produced reduced self-esteem or increased self-
enhancement.
Using a nearly identical cover story and procedure as described
above (i.e., processing of emotional cues), after the video presen-
tation participants completed an established measure of state self-
esteem (Heatherton & Polivy, 1991), a “better-than-average” self-
enhancement measure, and an item assessing the perceived
unfairness of the young’s woman suffering status (in counterbal-
anced order). The self-enhancement measure asked participants to
rate themselves relative to the “average university student” on 20
personality characteristics (10 positive and 10 negative; e.g., po-
lite, lazy, friendly, and snobbish) using scales ranging from 0
(much less than average)to10(much more than average; ␣⫽
.79). The measure of perceived fairness asked participants to rate
the unfairness of how Kerry’s condition has progressed on a scale
ranging from 1 (not at all unfair)to7(a great deal unfair).
As expected, participants rated the progress of the young wom-
an’s suffering as more unfair when the system was ineffective
(M 4.54, SD 1.45) than when it was effective (M 3.17,
SD 1.34), t(23) 2.45, p .02, d 0.98 (one participant did
not complete this item). However, there were no significant dif-
ferences between the ineffective system and the effective system
conditions in terms of state self-esteem (Ms 3.89 and 3.90,
respectively, p .98), or self-enhancement (Ms 6.93 and 7.36,
respectively, p .29). These findings corroborate and extend
Callan et al.’s (2006) Stroop task findings that the ineffectiveness
of the system to alleviate the young woman’s unjust suffering
produces concerns specific to justice.
Results and Discussion
We hypothesized that participants would enhance their percep-
tions of personal control following the condition in which the
system was shown to be ineffective at minimizing the unfairness of
Kerry’s situation. Our hypothesis was supported. Participants in
the ineffective system condition reported having more control over
the onset of the green circle (M 6.67, SD 1.43) than partic-
ipants in the effective system condition (M 5.38, SD 2.10),
t(46) 2.49, p .02, d 0.72.
This fifth and final study, although using a very different meth-
odology and approach, nicely complements the results of our first
four studies. We found that participants exposed to a situation in
which the system was wholly ineffective at alleviating a woman’s
unfair suffering subsequently enhanced their perceptions of per-
sonal control relative to participants who learned the system could
help to restore fairness. Furthermore, our manipulation check
study, combined with the results of prior research (Callan et al.,
2006; Kay & Jost, 2003; Kay et al., 2005), bolsters our interpre-
tation that enhanced perceptions of personal control were guided
by the unfairness of the system’s ineffectiveness in alleviating the
young woman’s suffering (and not concerns associated with self-
esteem or self-enhancement).
Interestingly, increased compensatory control was instigated by
an outcome (i.e., the onset of the green circle) that was noncon-
tingently related to the participants’ responses. That is, their per-
ceptions of control were illusory, but nonetheless varied as a
function of the success of the woman’s treatment. Moreover,
enhanced perceptions of control occurred even though the outcome
participants were pursuing was rather trivial. Indeed, requiring
participants to make a green circle appear on a computer screen
was sufficient to instigate compensatory control processes when
the effectiveness of the medical– governmental system at restoring
order and fairness was threatened. This finding may speak to the
substitutable nature of control beliefs, such that people’s compen-
satory strivings to restore a belief in an orderly and nonrandom
world may potentially occur in any domain relevant to control,
including those related to God and the government.
General Discussion
Across five studies, we tested a compensatory control model of
support for external systems, which suggests that in coping with
the existential threat posed by lowered (or chronically low) levels
of personal control, people increase their support for broad exter-
nal systems that impose order and control on their personal lives,
30
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
such as religious and sociopolitical systems. In three experimental
studies, we demonstrated that a manipulation designed to remind
people of their lack of personal control led to increased beliefs in
the existence of a controlling God and increased defense of the
overarching sociopolitical system. In a fourth experiment, we
demonstrated that the converse is also true: A manipulation de-
signed to challenge the usefulness of external systems of control
led to increased illusory perceptions of personal control.
In addition, in three of the experimental studies, we also dem-
onstrated that these effects were (a) moderated by the extent to
which the relevant external system was framed as offering control
and order (Study 1), (b) moderated by the extent to which the
participants perceived the external system as benevolent or con-
cerned with the best interests of the system’s constituents (Study
4), and (c) mediated by the extent to which the personal control
manipulation induced defensive reactions surrounding overarching
beliefs regarding the existence of randomness and chance in the
participants’ lives (Study 2).
Last, in a cross-cultural, nationally representative correlational
study, we demonstrated that those who chronically perceive lower
levels of personal control over their lives evince more support for
governmental control in their personal and professional lives, a
relationship that was especially strong for those living in countries
with more benevolent (less corrupt) governments.
Implications and Future Directions
This psychological process may have important implications in
helping us to understand several central sociopsychological phe-
nomena. The implications of our analyses for the widespread
adoption of basic religious ideologies and phenomena of system
justification should be clear from our five studies, but similar
processes may also play a role in more dramatic circumstances that
are not easily bottled within the laboratory. In situations in which
personal control is threatened to an extreme level, for example,
people may respond with equally extreme levels of endorsement of
external sources of control. The noted trend of increased rates of
religious extremism, and even terrorism in the name of extreme
religious ideology, among members of populations in which per-
sonal freedom is particularly restricted (Krueger & Laitin, 2004;
Krueger & Maleckova, 2003; Kruglanski & Golec, 2004) may be
one such example of this process. Future research that fleshes out
this possibility more definitively could prove very fruitful.
It is also worth noting the particular cultural context in which
our results and general theoretical position have been embedded.
Although we think it likely that most people, in most cultures, do
need to maintain a comfortable level of perceived order and
structure in their environments, the means by which they reach this
level is most certain to differ across cultures. In cultural contexts
in which personal choice and the ability to influence the environ-
ment are less inherent to positive views of the self, such as, for
example, certain East Asian cultures (e.g., Heine & Lehman, 1997;
Iyengar & Lepper, 1999; Kim & Markus, 1999), endorsing exter-
nal systems of control may play a much more significant role than
feelings of personal control. Moreover, in some cultural contexts
the meaning of personal control may be quite different than how
we have operationalized it in our studies. For example, members of
low socioeconomic status (SES) and high SES groups, because of
their very disparate experiences with the material and social world,
tend to view control and personal agency quite differently from
one another (Snibbe & Markus, 2005), with members of lower
SES groups placing more emphasis on their ability to control
integrity-based attributes, such as honesty, reliability, consistency,
and loyalty, than on constructs such as free choice and environ-
mental influence, which tend to be favored by members of higher
SES groups. Although the cross-cultural data from Study 4 suggest
that certain processes of compensatory control may act similarly
across very disparate contexts, research examining the different
manifestations of compensatory control among groups who have
had very different experiences with, and views of, the material
world would be a worthwhile avenue of research.
Another question worth considering involves the extent to
which external systems of control are themselves substitutable for
one another. According to our model of compensatory control,
systems of external control should not only be substitutable with
reductions in personal control, but also with one another. Indeed,
it may be the case that a particularly high level of endorsement of
one system of control (e.g., religion) can “free people up” to
criticize and lash out at other such sources (e.g., the scientific
community, the police, and the government). Although it may
seem odd to suggest the psychology of governmental support can
serve similar existential needs to the psychology of religion, such
an approach is consistent with recent theoretical positions regard-
ing the function of the system justification motive (Jost & Huny-
ady, 2005). A refined analysis of which external systems of control
are substitutable for one another, and which serve as best substi-
tutes for reductions in personal control, would be a logical next
step to pursue.
Similarly, one could imagine that not all governmental systems
or all religious ideologies would be equally effective channels of
compensatory control. As we observed in Study 3, members of
countries known to have more corrupt governments appear less
likely to compensate for lowered levels of personal control with
increased support for their government. Perhaps in regions such as
these, people are more likely to rely on other external sources, such
as religious beliefs or ethnic and cultural norms, to compensate for
lowered levels of personal control. Even among religious beliefs,
though, not all religious sources of external control will be equally
effective. The extent to which a given religion imbues more or less
order to the universe will have obvious implications for its use-
fulness as a means of compensatory control. It is also possible that
certain individuals, such as those with especially high levels of
self-esteem or defensiveness, may not rely on external systems at
all following a threat to personal control. Rather than opting to
preserve beliefs in order through a separate (but equifinal) means
(see Kruglanski, 1996), such as God or the government, following
a threat to personal control such individuals may cope by stead-
fastly and straightforwardly reasserting feelings of personal con-
trol (McGregor, 2006; McGregor et al., 2001).
In addition, we have argued here that those external systems
perceived as benevolent are especially likely to serve as means of
compensatory control. This, however, may not be the only such
factor. When compensating for lowered levels of personal control,
external systems perceived as competent or able to effectively
meet their goals may also be preferred. Indeed, it is possible to
perceive the moderators in both Studies 3 and 4 as assessing
governmental competence and governmental benevolence. Other
factors, such as stability, may also play a moderating role. More-
31
GOD AND THE GOVERNMENT
over, given the strength of the need to perceive order and structure
in the world, under circumstances of extreme loss of personal
control, or when no other feasible external systems exist, people
may be willing to adopt any external system of control, whatever
its features and characteristics. Research that more precisely as-
sesses the other features that make an external system more and
less likely to serve this purpose, as well as when people will and
will not care about these features, is needed.
Last, we interpret our findings as indicating that when one’s
perception of personal control (i.e., the extent to which they have
control over their outcomes) is lowered, their overall motivation to
believe in order becomes threatened, and they compensate for this
threat by increasingly defending external sources of control, such
as God and government. It is also possible, however, that the
effects we have observed could be an instantiation of an even
broader need for a sustained sense of certainty, self-integrity, or
broad-based meaning system that beliefs in order and nonrandom-
ness themselves may feed into (Heine, Proulx, & Vohs, 2006;
Kruglanski & Webster, 1996; McGregor, 2006; Solomon, Green-
berg, & Pyszczynksi, 2004; van den Bos, van Ameijde, & van
Gorp, 2006). Our manipulation check data, mediational study, and
triangulation across multiple manipulations and measurements
suggests that personal control and the defense of external systems
of control are clearly related, but this does not exclude the possible
involvement of an even larger motivational system. Regardless of
which motivational system the need for order is best subsumed
within—if it is subsumed within a broader motivational system,
such as mortality-based worldview defense or general uncertainty
avoidance, at all—the specific findings we have demonstrated are
interesting in their own right because they make predictions about
important social phenomena that have not been illuminated before
in other research programs and identify novel and important causal
factors that contribute to religious and political defense.
Relation to Other Theories of Compensatory Control
As described in the introduction, the model of compensatory
control we have put forth is similar to Rothbaum et al.’s (1982)
two-process model of perceived control and the uncertainty hy-
pothesis recently tested by Burger and Lynn (2005; also see
Malinowski, 1954) It is not, however, identical to either of these.
Proponents of the uncertainty hypothesis suggest that as a means
of preserving the belief in personal control during times of uncon-
trollability, people will engage in superstitious behavior. Accord-
ing to the two-process model of perceived control (Rothbaum et
al., 1982), when people’s means of primary control—people’s
ability to directly influence their environment in line with their
personal desires—is thwarted, they may engage in any number of
secondary control strategies (for a review, see Morling & Evered,
2006; Rothbaum et al., 1982). Most relevant to this line of re-
search, however, is Rothbaum et al.’s notion of vicarious control,
a secondary control strategy in which people are presumed to gain
a sense of personal control by identifying with powerful others.
Although both the uncertainty hypothesis and the two-process
model of perceived control, like the one presented here, begin with
the assumption that people will engage in motivated processes of
coping following experiences that threaten personal control, nei-
ther is redundant with what we are proposing. According to the
uncertainty hypothesis, people are said to rely on superstition as
way to regain personal control (i.e., by enacting superstitious
behavior, people find a way to feel they personally, but indirectly,
are controlling their outcomes even in uncontrollable situations).
Likewise, Rothbaum et al. (1982) proposed that people may iden-
tify with powerful others to psychologically share in the powerful
others’ control and thereby regain personal control. Both of these
notions are qualitatively different from what we are proposing
regarding the endorsement of external systems. In particular, we
are not suggesting that by endorsing external systems of control
people are trying to regain personal control, but that they are
instead trying to affirm the more general and overarching belief
that things are under control, even if not by their own means (see
Antonovsky, 1979).
The data presented here are the first to experimentally test any
of these theoretical approaches (in this context), and the modera-
tional and mediational evidence presented in Studies 2, 3, and 4 are
supportive of our specific presumed mechanism. In addition, these
alternative formulations do not explicitly predict the findings of
Study 5, in which a threat to an external system of control led to
increased perceptions of personal control. Nonetheless, future
work is needed to better distinguish between these theoretical
perspectives at an empirical level.
Other previous models of compensatory control, however, bear
less of a similarity to ours. Within the health and coping literature,
a model of compensatory control is outlined to explain how people
cope with decreasing senses of control during periods of illness
(see Morling & Evered, 2006). In such models, people compensate
for their reductions in control in a given domain by focusing their
control on other areas that they can control. Thus, they are com-
pensating for their lack of personal control in one domain by
increasing their personal control in other domains. This is different
from the model proposed here, which suggests that people com-
pensate for decreases in personal control via increases in the
endorsement of broad external systems that impose a particular
social order.
Concluding Remarks
Despite the dearth of evidence, an overwhelming majority of
people believe in the existence of God (Gallup & Castelli, 1989).
Likewise, despite the objective unfairness of many aspects of
sociopolitical systems, people tend to defend and support the
continued existence of these institutions at a surprising rate (for a
review, see Jost et al., 2004). Although beliefs in God and the
government obviously differ in many important ways, the high
levels of support they both receive may be due to one particular
aspect they share: their potential to serve as compensatory systems
of control and order.
References
Alloy, L. B., & Abramson, L. Y. (1979). Judgment of contingency in
depressed and non-depressed students: Sadder but wiser? Journal of
Experimental Psychology: General, 108, 441– 485.
Alloy, L. B., & Abramson, L. Y. (1982). Learned helplessness, depression,
and the illusion of control. Journal of Personality and Social Psychol-
ogy, 42, 1114–1126.
Alloy, L. B., Abramson, L. Y., & Viscusi, D. (1981). Induced mood and
illusion of control. Journal of Personality and Social Psychology, 41,
1129 –1140.
32
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
Allport, G. W., & Ross, J. M. (1967). Personal religious orientation and
prejudice. Journal of Personality and Social Psychology, 5, 432–443.
Antonovsky, A. (1979). Health, stress and coping. San Francisco: Jossey-
Bass.
Argyle, M., & Beit-Hallahmi, B. (1975). The social psychology of religion.
Oxford, England: Routledge & Kegan Paul.
Baumeister, R. F. (2002). Religion and psychology: Introduction to the
special issue. Psychological Inquiry, 13, 165–167.
Burger, J. M. (1989). Negative reactions to increases in perceived personal
control. Journal of Personality and Social Psychology, 56, 246–256.
Burger, J. M. (1992). Desire for control: Personality, social, and clinical
perspectives. New York: Plenum Press.
Burger, J. M., & Cooper, H. M. (1979). The desirability of control.
Motivation and Emotion, 3, 381–393.
Burger, J. M., & Lynn, A. L. (2005). Superstitious behavior among
American and Japanese professional baseball players. Basic and Applied
Social Psychology, 27, 71–76.
Callan, M. J., Ellard, J. H., & Nicol, J. E. (2006). The belief in a just world
and immanent justice reasoning in adults. Personality and Social Psy-
chology Bulletin, 32, 1646 –1658.
Crocker, J., & Luhtanen, R. (1990). Collective self-esteem and ingroup
bias. Journal of Personality and Social Psychology, 58, 60 67.
de Bres, G. (1984). The Belgic confession. In Canadian Reformed
Churches Standing Committee for the Publication of the Book of Praise
(Eds.), The book of praise. Winnipeg, Manitoba, Canada: Premier Print-
ing. (Original work published 1561)
Dechesne, M., Janssen, J., & van Knippenberg, A. (2000). Derogation and
distancing as terror management strategies: The moderating role of need
for closure and permeability of group boundaries. Journal of Personality
and Social Psychology, 79, 923–932.
Dunning, D. (2003). The zealous self-affirmer: How and why the self lurks
so pervasively behind social judgment. In S. J. Spencer, S. Fein, M. P.
Zanna, & J. M. Olson (Eds.), Ontario Symposium on Personality and
Social Psychology (pp. 45–72). Mahwah, NJ: Erlbaum.
Dweck, C. S., & Leggett, E. L. (1988). A social cognitive approach to
motivation and personality. Psychological Review, 95, 256 –273.
Fisher, J. D., & Fisher, W. A. (1992). People like us [Videotape]. Storrs:
University of Connecticut, Department of Psychology.
Gallup, G., Jr., & Castelli, J. (1989). The people’s religion: American faith
in the Nineties. New York: MacMillan.
Greenberg, J., Porteus, J., Simon, L., & Pyszczynski, T. (1995). Evidence
of a terror management function of cultural icons: The effects of mor-
tality salience on the inappropriate use of cherished cultural symbols.
Personality and Social Psychology Bulletin, 21, 1221–1228.
Gurin, G., Veroff, J., & Feld, S. (1960). Americans view their mental
health: A nationwide interview survey. Oxford, England: Basic Books.
Hafer, C. L. (2000). Do innocent victims threaten the belief in a just world?
Evidence from a modified Stroop task. Journal of Personality and Social
Psychology, 79, 165–173.
Hafer, C. L., & Be`gue, L. (2005). Experimental research on just-world
theory: Problems, developments, and future challenges. Psychological
Bulletin, 131, 128–167.
Heatherton, T. F., & Polivey, J. (1991). Development and validation of a
scale for measuring state self-esteem. Journal of Personality and Social
Psychology, 60, 895–910.
Heine, S. J., & Lehman, D. R. (1997). The cultural construction of
self-enhancement: An examination of group-serving biases. Journal of
Personality and Social Psychology, 72, 1268 –1283.
Heine, S. J., Proulx, T., & Vohs, K. D. (2006). The meaning maintenance
model: On the coherence of social motivations. Personality and Social
Psychology Review, 10, 88 –110.
Inglehart, R., Basan˜ez, M., Diez-Medrano, J., Halman, L, & Luijkx, R.
(2004). Human beliefs and values: A cross-cultural sourcebook based on
the 1999–2002 Value Surveys. Mexico City, Mexico: Siglo XXI.
Iyengar, S. S., & Lepper, M. R. (1999). Rethinking the value of choice: A
cultural perspective on intrinsic motivation. Journal of Personality and
Social Psychology, 76, 349 –366.
Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychol-
ogy of trauma. New York: Free Press.
Ji, L., Peng, K., & Nisbett, R. E. (2000). Culture, control, and perception
of relationships in the environment. Journal of Personality and Social
Psychology, 78, 943–955.
Jonas, E., & Fischer, P. (2006). Terror management and religion: Evidence
that intrinsic religiousness mitigates worldview defense following mor-
tality salience. Journal of Personality and Social Psychology, 91, 553–
567.
Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system-
justification and the production of false consciousness. British Journal of
Social Psychology, 33, 1–27.
Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system
justification theory: Accumulated evidence of conscious and uncon-
scious bolstering of the status quo. Political Psychology, 25, 881–920.
Jost, J. T., & Hunyady, O. (2002). The psychology of system justification
and the palliative function of ideology. European Review of Social
Psychology, 13, 111–153.
Jost, J. T., & Hunyady, O. (2005). Antecedents and consequences of
system-justifying ideologies. Current Directions in Psychological Sci-
ence, 14, 260–265.
Jost, J. T., & Kay, A. C. (2005). Exposure to benevolent sexism and
complementary gender stereotypes: Consequences for specific and dif-
fuse forms of system justification. Journal of Personality and Social
Psychology, 88, 498–509.
Kay, A. C., Jimenez, M. C., & Jost, J. T. (2002). Sour grapes, sweet
lemons, and the anticipatory rationalization of the status quo. Person-
ality and Social Psychology Bulletin, 28, 1300 –1312.
Kay, A. C., & Jost, J. T. (2003). Complementary justice: Effects of “poor
but happy” and “poor but honest” stereotype exemplars on system
justification and implicit activation of the justice motive. Journal of
Personality and Social Psychology, 85, 823– 837.
Kay, A. C., Jost, J. T., Mandisodza, A. N., Sherman, S. J., Petrocelli, J. V.,
& Johnson, A. L. (2007). Panglossian ideology in the service of system
justification: How complementary stereotypes help us to rationalize
inequality. In M. P. Zanna (Ed.), Advances in experimental social
psychology (Vol. 38, pp. 305–358). San Diego, CA: Academic Press.
Kay, A. C., Jost, J. T., & Young, S. (2005). Victim derogation and victim
enhancement as alternate routes to system justification. Psychological
Science, 16, 240–246.
Kay, A. C., & Zanna, M. P. (in press). A contextual analysis of the system
justification motive and its societal consequences. In J. T. Jost, A. C.
Kay, & H. Thorisdottir (Eds.), Social and psychological bases of ideol-
ogy and system justification. New York: Oxford University Press.
Keinan, G. (1994). Effects of stress and tolerance of ambiguity on magical
thinking. Journal of Personality and Social Psychology, 67,48–55.
Keinan, G. (2002). The effects of stress and desire for control on super-
stitious behavior. Personality and Social Psychology Bulletin, 28, 102–
108.
Kelly, G. A. (1955). The psychology of personal constructs. New York:
Norton.
Kim, H., & Markus, H. R. (1999). Deviance or uniqueness, harmony or
conformity? A cultural analysis. Journal of Personality and Social
Psychology, 77, 785–800.
Koenig, L. J., Clements, C. M., & Alloy, L. B. (1992). Depression and the
illusion of control: The role of esteem maintenance and impression
management. Canadian Journal of Behavioral Science, 24, 233–252.
Krause, N. (2005). God-mediated control and psychological well-being in
late life. Research on Aging, 27, 136 –164.
Krueger, A. B., & Laitin, D. D. (2004). Misunderstanding terrorism.
Foreign Affairs, 83(5), 8 –13.
33
GOD AND THE GOVERNMENT
Krueger, A. B., & Maleckova, J. (2003). Education, poverty, and terrorism:
Is there a causal connection? Journal of Economic Perspectives, 18,
201–214.
Kruglanski, A. W. (1989). Lay epistemics and human knowledge: Cogni-
tive and motivational bases. New York: Plenum Press.
Kruglanski, A. W. (1996). Goals as knowledge structures. In P. M. Goll-
witzer & J. A. Bargh (Eds.), The psychology of action: Linking cognition
and motivation to behavior (pp. 599 618). New York: Guilford Press.
Kruglanski, A. W., & Golec, A. (2004). Individual motivations, the group
process and organizational strategies in suicide terrorism. In E. M.
Meyersson Milgrom (Ed.), Suicide missions and the market for martyrs:
A multidisciplinary approach. Princeton, NJ: Princeton University Press.
Kruglanski, A. W., & Webster, D. M. (1996). Motivated closing of the
mind: “seizing” and “freezing.” Psychological Review, 103, 263–283.
Landau, M. J., Greenberg, J., Solomon, S., Pyszczynski, T., & Martens, A.
(2006). Windows into nothingness: Terror management, meaningless-
ness, and negative reactions to modern art. Journal of Personality and
Social Psychology, 90, 879 892.
Landau, M. J., Johns, M., Greenberg, J., Pyszczynski, T., Martens, A.,
Goldenberg, J. L., & Solomon, S. (2004). A function of form: Terror
management and structuring the social world. Journal of Personality and
Social Psychology, 87, 190 –210.
Lau, G. P., Kay, A. C., & Spencer, S. J. (2008). Loving those who justify
inequality: The effects of system threat on attraction to women who
embody benevolent sexist ideals. Psychological Science, 19, 20–21.
Lerner, M. J. (1980). The belief in a just world: A fundamental delusion.
New York: Plenum.
Malinowski, B. (1954). Magic, science, and religion. Garden City, NY:
Doubleday.
McGregor, I. (2006). Offensive defensiveness: Toward an integrative neu-
roscience of compensatory zeal after mortality salience, personal uncer-
tainty, and other poignant self-threats. Psychological Inquiry, 17, 299
308.
McGregor, I., & Marigold, D. C. (2003). Defensive zeal and the uncertain
self: What makes you so sure? Journal of Personality and Social
Psychology, 85, 838 852.
McGregor, I., Zanna, M. P., Holmes, J. G., & Spencer, S. J. (2001).
Compensatory conviction in the face of personal uncertainty: Going to
extremes and being oneself. Journal of Personality and Social Psychol-
ogy, 80, 472–488.
Mikulincer, M., Gerber, H., & Weisenberg, M. (1990). Judgment of control
and depression: The role of self-esteem threat and self-focused attention.
Cognitive Therapy and Research, 14, 589 608.
Morling, B., & Evered, S. (2006). Secondary control reviewed and defined.
Psychological Bulletin, 132, 269 –296.
Msetfi, R. M., Murphy, R. A., & Simpson, J. (2007). Depressive realism
and the effect of intertrial interval on judgments of zero, positive, and
negative contingencies. Quarterly Journal of Experimental Psychology,
60, 461–481.
Mullen, B., Brown, R., & Smith, C. (1992). Ingroup bias as a function of
salience, relevance, and status: An integration. European Journal of
Social Psychology, 22, 103–122.
Norenzayan, A., & Hansen, I. G. (2006). Belief in supernatural agents in
the face of death. Personality and Social Psychology Bulletin, 32,
174 –187.
Osarchuk, M., & Tatz, S. J. (1973). Effect of induced fear of death on belief
in afterlife. Journal of Personality and Social Psychology, 27, 256 –260.
Pargament, K. I. (1997). The psychology of religion and coping: Theory,
research, practice. New York: Guilford Press.
Pargament, K. I. (2000). Interventions based in religious congregations.
Washington, DC: American Psychological Association.
Pargament, K. I. (2002). The bitter and the sweet: An evaluation of the
costs and benefits of religiousness. Psychological Inquiry, 13, 168 –181.
Pargament, K. I., Smith, B. W., Koenig, H. G., & Perez, L. (1998). Patterns
of positive and negative religious coping with major life stressors.
Journal for the Scientific Study of Religion, 37, 710–724.
Pennebaker, J. W., & Stone, L. D. (2004). Translating traumatic experi-
ences into language: Implications for child abuse and long-term health.
In L. J. Koenig, L. S. Doll, A. O’Leary, & W. Pequegnat (Eds.), From
child sexual abuse to adult sexual risk: Trauma, revictimization, and
intervention. (pp. 201–216). Washington, DC: American Psychological
Association.
Pepitone, A., & Saffiotti, L. (1997). The selectivity of nonmaterial beliefs
in interpreting life events. European Journal of Social Psychology, 27,
23–35.
Perkins, C. C. (1968). An analysis of the concept of reinforcement. Psy-
chological Review, 75, 155–172.
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools
for probing interactions in multiple linear regression, multilevel model-
ing, and latent curve analysis. Journal of Education and Behavioral
Statistics, 31, 437–448.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for
estimating indirect effects in simple mediation models. Behavior Re-
search Methods, Instruments & Computers, 36, 717–731.
Presson, P. K., & Benassi, V. A. (1996). Illusion of control: A meta-
analytic review. Journal of Social Behavior and Personality, 11, 493–
510.
Presson, P. K., & Benassi, V. A. (2003). Are depressive symptoms posi-
tively or negatively associated with the illusion of control? Social
Behavior and Personality: An International Journal, 31, 483–495.
Raudenbush, S. W., Bryk, A. S., & Congdon, R. T. (2001). HLM for
Windows. Chicago: Scientific Software International.
Rodin, J., Rennert, K., & Solomon, S. K. (1980). Intrinsic motivation for
control: Fact or fiction. In A. Baum & J. E. Singer (Eds.), Advances in
environmental psychology: Vol. 2. Applications of personal control (pp.
131–151). Hillsdale, NJ: Erlbaum.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rothbaum, R., Weisz, J. R., Snyder, S. S. (1982) Changing the world and
changing the self: A two-process model of perceived control. Journal of
Personality and Social Psychology, 42, 5–37.
Ryan, R. M., Rigby, S., & King, K. (1993). Two types of religious
internalization and their relations to religious orientations and mental
health. Journal of Personality and Social Psychology, 65, 586–596.
Sales, S. M. (1972). Economic threat as a determinant of conversion rates
in authoritarian and nonauthoritarian churches. Journal of Personality
and Social Psychology, 23, 420 428.
Sales, S. M. (1973). Threat as a factor in authoritarianism: An analysis of
archival data. Journal of Personality and Social Psychology, 28, 44 –57.
Seligman, M. E. P. (1975). Helplessness: On depression, development and
death. San Francisco: Freeman.
Seligman, M. E. P. (1976). Learned helplessness and depression in animals
and men. Morristown, NJ: General Learning Press.
Sethis, S., & Lepper, M. R. (1998). Rethinking the role of choice in
intrinsic motivation: A cultural perspective. Unpublished manuscript,
Stanford University.
Shapiro, D. H., Sandman, C. A., Grossman, M., & Grossman, B. (1995).
Aging and sense of control. Psychological Reports, 77, 616 618.
Skinner, E. A. (1995). Perceived control, motivation, and coping. Thou-
sand Oaks, CA: Sage.
Snibbe, A. C., & Markus, H. R. (2005). You can’t always get what you
want: Educational attainment, agency, and choice. Journal of Personal-
ity and Social Psychology, 88, 703–720.
Solomon, S., Greenberg, J., & Pyszczynski, T. (2004). The cultural animal:
Twenty years of terror management theory. In J. Greenberg, S. Koole, &
T. Pyszczynski (Eds.), Handbook of experimental existential psychology
(pp. 13–34). New York: Guilford Press.
Tennen, H., & Sharp, J. P. (1983). Control orientation and the illusion of
control. Journal of Personality Assessment, 47, 369 –374.
34
KAY, GAUCHER, NAPIER, CALLAN, AND LAURIN
Tesser, A., Crepaz, N., Beach, S. R. H., Cornell, D., & Collins, J. C.
(2000). Confluence of self-esteem regulation mechanisms: On integrat-
ing the self-zoo. Personality and Social Psychology Bulletin, 26, 1476
1489.
Thompson, S. C., Kyle, D., Osgood, A., Quist, R. M., Phillips, D. J., &
McClure, M. (2004). Illusory control and motives for control: The role
of connection and intentionality. Motivation and Emotion, 28, 315–330.
Transparency International. (n.d.). Corruption Perceptions Index. Re-
trieved March 6, 2007, from http://www.transparency.org/policy_re-
search/surveys_indices/global/cpi
van den Bos, K., van Ameijde, J., & van Gorp, H. (2006). On the
psychology of religion: The role of personal uncertainty in religious
worldview defense. Basic and Applied Social Psychology, 28, 333–341.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
dation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Weisz, J. R., Rothbaum, F. M., & Blackburn, T. C. (1984). Standing out
and standing in: The psychology of control in America and Japan.
American Psychologist, 39, 955–969.
White, R. (1959). Motivation reconsidered: The concept of competence.
Psychological Review, 66, 297–330.
Wohl, M. J. A., & Enzle, M. E. (2003). The effects of near wins and near
losses on losses on self-perceived personal luck and subsequent gambling
behavior. Journal of Experimental Social Psychology, 39, 184 –191.
World Values Survey Association. (n.d.). European and World Values
Surveys four-wave integrated data file, 1981–2004, v. 20060423 [data
set]. Stockholm, Sweden: Author. Retrieved April 1, 2007, from http://
www.worldvaluessurvey.org
You, J. S., & Khagram, S. (2005). A comparative study of inequality and
corruption. American Sociological Review, 70, 136 –157.
Received March 6, 2007
Revision received February 4, 2008
Accepted February 5, 2008
Members of Underrepresented Groups:
Reviewers for Journal Manuscripts Wanted
If you are interested in reviewing manuscripts for APA journals, the APA Publications and
Communications Board would like to invite your participation. Manuscript reviewers are vital to the
publications process. As a reviewer, you would gain valuable experience in publishing. The P&C
Board is particularly interested in encouraging members of underrepresented groups to participate
more in this process.
If you are interested in reviewing manuscripts, please write to the address below. Please note the
following important points:
To be selected as a reviewer, you must have published articles in peer-reviewed journals. The
experience of publishing provides a reviewer with the basis for preparing a thorough, objective
review.
To be selected, it is critical to be a regular reader of the five to six empirical journals that are most
central to the area or journal for which you would like to review. Current knowledge of recently
published research provides a reviewer with the knowledge base to evaluate a new submission
within the context of existing research.
To select the appropriate reviewers for each manuscript, the editor needs detailed information.
Please include with your letter your vita. In the letter, please identify which APA journal(s) you
are interested in, and describe your area of expertise. Be as specific as possible. For example,
“social psychology” is not sufficient—you would need to specify “social cognition” or “attitude
change” as well.
Reviewing a manuscript takes time (1– 4 hours per manuscript reviewed). If you are selected to
review a manuscript, be prepared to invest the necessary time to evaluate the manuscript
thoroughly.
Write to Journals Office, American Psychological Association, 750 First Street, NE, Washington,
DC 20002-4242.
35
GOD AND THE GOVERNMENT
... Enfin, une faible perception du contrôle personnel entraînerait l'augmentation de la justification du système (Kay et al., 2008 ;Rutjens & Loseman, 2010). L'explication de cet effet est basée sur l'hypothèse du contrôle compensatoire. ...
... Les auteurs proposent que lorsqu'il est impossible pour une personne d'affirmer « qu'elle contrôle » les choses (i.e., primary control desires), elle sera alors motivée à percevoir ces choses comme étant « quand même sous contrôle » (i.e., secondary control strategies ; Kay et al., 2008). ...
... Cette hypothèse est opposée à celle de la théorie du contrôle compensatoire. En effet, comme discuté dans le chapitre 3, cette théorie propose que l'adhésion à un système externe (e.g., ordre politique ou religieux) constitue une manière de compenser une diminution de CPP (Kay et al., 2008). Autrement dit, la théorie du contrôle compensatoire prévoit que la diminution du CPP engendre l'augmentation de la justification du système. ...
Thesis
L'idéologie néolibérale est fréquemment théorisée comme un facteur de dépolitisation des citoyens. Néanmoins, les travaux explorant empiriquement ses effets sur des attitudes et des comportements politiques sont rares. Cette thèse a donc pour objet l'étude des implications psychosociales de l'idéologie néolibérale, comprise comme un ensemble de valeurs (i.e., idéographie) conditionnant une conception particulière de la personne (i.e., conception néolibérale du sujet). Afin d'éclairer la manière dont cette idéologie peut influencer les attitudes des personnes (e.g., justification du système) et leurs comportements politiques, 9 études ont été menées. Les deux premières études ont mis en évidence l'association entre l'adhésion aux valeurs néolibérales, la justification du système et les comportements politiques (i.e., manifestation et vote). Consécutivement, 5 études expérimentales ont été réalisées afin de vérifier la nature causale de la relation entre l'idéologie néolibérale et la justification du système. Plus encore, ces études visaient à éclairer le rôle du contrôle personnel perçu comme mécanisme explicatif de cette relation. Les résultats ne permettent pas de conclure sur l'ensemble de la médiation mais étayent l'hypothèse selon laquelle le contrôle personnel perçu constitue un antécédent à la justification du système. Enfin, les deux dernières études expérimentales explorent les effets de l'idéologie néolibérale et de la justification du système sur les intentions comportementales des sujets face à des problématiques systémiques (i.e., inégalités de genre et crise climatique). Les résultats indiquent que l'idéologie néolibérale, en tant qu'idéologie justificatrice, favorise des réponses individuelles, normatives et non-disruptives. Dans son ensemble, cette thèse fait apparaitre que l'idéologie néolibérale favorise un « citoyen minimal », figure individualisée polarisée autour de la liberté individuelle, à l'opposé d'un « citoyen agent social » polarisé autour de la liberté politique.
... To address this gap, we draw on compensatory control theory (Kay et al., 2008;Landau et al., 2015) to investigate a novel antecedent of tight cultures: individuals' perceptions of personal control. Specifically, we seek to uncover the bidirectional relationship between the psyche and culture in the domain of personal control and cultural tightness. ...
... We first consider how the personal control dimension of the psyche influences the tightness dimension of culture. To do so, we draw on compensatory control theory, which was conceived partly to explain the existence of variations in personal control both within as well as between cultures, even though control has been posited as a fundamental human need (Kay et al., 2008;Presson & Benassi, 1996;Seligman, 1975). Recent theoretical innovations in compensatory control theory suggest that the search for nonepistemic structure, or "interpretations of one's social and physical environments as simple (vs. ...
Article
Full-text available
According to the theory of mutual constitution of culture and psyche, just as culture shapes people, individuals' psychological states can influence culture. We build on compensatory control theory, which suggests that low personal control can lead people to prefer societal systems that impose order, to examine the mutual constitution of personal control and cultural tightness. Specifically, we tested whether individuals' lack of personal control increases their preference for tighter cultures as a means of restoring order and predictability, and whether tighter cultures in turn reduce people's feelings of personal control. Seven studies (five preregistered) with participants from the United States, Singapore, and China examine this cycle of mutual constitution. Specifically, documenting the correlational link between person and culture, we found that Americans lower on personal control preferred to live in tighter states (Study 1). Chinese employees lower on personal control also desired more structure and preferred a tighter organizational culture (Study 2). Employing an experimental causal chain design, Studies 3-5 provided causal evidence for our claim that lack of control increases desire for tighter cultures via the need for structure. Finally, tracing the link back from culture to person, Studies 6a and 6b found that whereas tighter cultures decreased perceptions of individual personal control, they increased people's sense of collective control. Overall, the findings document the process of mutual constitution of culture and psyche: lack of personal control leads people to seek more structured, tighter cultures, and that tighter cultures, in turn, decrease people's sense of personal control but increase their sense of collective control. (PsycInfo Database Record (c) 2022 APA, all rights reserved).