ArticlePDF Available

The mutual constitution of culture and psyche: The bidirectional relationship between individuals' perceived control and cultural tightness-looseness

Authors:

Abstract

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).
Journal of Personality and Social Psychology
The Mutual Constitution of Culture and Psyche: The Bidirectional
Relationship Between Individuals’ Perceived Control and Cultural
Tightness–Looseness
Anyi Ma, Krishna Savani, Fangzhou Liu, Kenneth Tai, and Aaron C. Kay
Online First Publication, October 31, 2022. https://dx.doi.org/10.1037/pspa0000327
CITATION
Ma, A., Savani, K., Liu, F., Tai, K., & Kay, A. C. (2022, October 31). The Mutual Constitution of Culture and Psyche: The
Bidirectional Relationship Between Individuals’ Perceived Control and Cultural Tightness–Looseness. Journal of Personality
and Social Psychology. Advance online publication. https://dx.doi.org/10.1037/pspa0000327
The Mutual Constitution of Culture and Psyche: The Bidirectional
Relationship Between IndividualsPerceived Control and
Cultural TightnessLooseness
Anyi Ma
1
, Krishna Savani
2
, Fangzhou Liu
3
, Kenneth Tai
4
, and Aaron C. Kay
5
1
Management Area, Freeman School of Business, Tulane University
2
Department of Management and Marketing, Faculty of Business, The Hong Kong Polytechnic University
3
School of Management, Huazhong University of Science and Technology
4
Department of Organisational Behaviour and Human Resources, Lee Kong Chian School of Business,
Singapore Management University
5
Department of Psychology and Neuroscience, Department of Management and Organization,
Fuqua School of Business, Duke University
According to the theory of mutual constitution of culture and psyche, just as culture shapes people,
individualspsychological states can inuence 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. Specically, we tested whether
individualslack of personal control increases their preference for tighter cultures as a means of restoring
order and predictability, and whether tighter cultures in turn reduce peoples feelings of personal control.
Seven studies (ve preregistered) with participants from the United States, Singapore, and China examine
this cycle of mutual constitution. Specically, 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 35 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 peoples sense of collective
control. Overall, the ndings 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 peoples sense of personal control but increase their sense of collective control.
Keywords: perceived control, tightnesslooseness, need for structure, mutual constitution, culture
Supplemental materials: https://doi.org/10.1037/pspa0000327.supp
The theory of mutual constitution of culture and psychology
argues that whereas culture shapes individualscognition, emotion,
motivation, and behavior, individualspsychological states can also
shape cultural patterns and characteristics (Adams & Markus, 2004;
Fiske et al., 1998). Although extensive research has examined the
rst half of this proposition (Cohen & Kitayama, 2007), less
research has empirically investigated the inuence of individuals
psychological states on broader cultural patterns (for notable ex-
ceptions, see Kashima, 2000;Kitayama et al., 1997;Morling &
Lamoreaux, 2008;Salter et al., 2018). More importantly, virtually
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Editors Note. Yoshi Kashima served as the handling editor for this
article.SK
Anyi Ma https://orcid.org/0000-0003-1220-6949
Krishna Savani https://orcid.org/0000-0002-6934-1917
Fangzhou Liu https://orcid.org/0000-0002-4133-1151
Kenneth Tai https://orcid.org/0000-0002-7817-3599
Aaron C. Kay https://orcid.org/0000-0001-7895-0987
The authors thank Dayana Bulchand, Sylvia Chin, Ee-Hwee Lau, and
Andrea Low, as well as for invaluable research assistance. Parts of this
research were conducted while Krishna Savani was at Nanyang Technologi-
cal University.
Anyi Ma played lead role in conceptualization, data curation, formal
analysis, investigation, methodology, project administration and writing of
original draft and equal role in writing of review and editing. Krishna Savani
played lead role in resources, supervision and validation, supporting role in
conceptualization, investigation and methodology and equal role in writing
of review and editing. Fangzhou Liu played supporting role in investigation.
Kenneth Tai played supporting role in investigation. Aaron C. Kay played
supporting role in conceptualization, investigation, methodology and super-
vision and equal role in writing of review and editing.
All survey materials, data, and analysis code are available at https://
osf.io/f8wzh.
Correspondence concerning this article should be addressed to Krishna
Savani, Department of Management and Marketing, Faculty of Business,
The Hong Kong Polytechnic University, Li Ka Shing Tower M857, Hung
Hom, Kowloon, Hong Kong or Fangzhou Liu, School of Management,
Huazhong University of Science and Technology, 1037 Luoyu Road,
Hongshan, Wuhan 430074, China. Email: krishna.savani@polyu.edu.hk
or liufangzhou@hust.edu.cn
Journal of Personality and Social Psychology:
Attitudes and Social Cognition
© 2022 American Psychological Association
ISSN: 0022-3514 https://doi.org/10.1037/pspa0000327
1
no research to our knowledge has examined the full cycle of the
mutual constitution of culture and psyche in the same article.
The rst phase of research in cross-cultural psychology focused on
a key dimension across which cultures varyindependent versus
interdependent self-construal (Markus & Kitayama, 1991) and indi-
vidualism versus collectivism (Triandis, 1989). Another dimension
that has attracted much recent attention is tightnesslooseness
(Gelfand et al., 2011;Triandis, 1996). Whereas tight societies
(e.g., Pakistan, Singapore) have strong norms and a low tolerance
for deviance from the norm, loose societies (e.g., Ukraine, USA) have
weak norms and are more likely to tolerate norm-violating behaviors
(Gelfand et al., 2011). An extensive program of research by Gelfand
et al. has demonstrated that cross-cultural tightness predicts a range
of psychological outcomes, such as prevention focus, self-regulation
strength, self-monitoring (Gelfand et al., 2011), preferences for
autonomous leadership (Aktas et al., 2016), propensity to engage
in convergent thinking (Chua et al., 2015), religious beliefs and
practices (Caluori et al., 2020;Jackson, Caluori, Abrams, et al., 2021,
Jackson, Caluori, Gray, et al., 2021), and strength of electroencepha-
lography (EEG) reactions to norm violations (Mu et al., 2015).
Differences in cultural tightnesslooseness are theorized to be
caused by the presence of external threats (e.g., war, natural disasters,
pathogen load, or top-downprocesses) and reinforced by variation
in psychological tendencies within cultures (i.e., bottom-up,
processes; Gelfand et al., 2011;Triandis, 1972). However, although
the role of such bottom-upprocesses in which individual-level
characteristics shape cultural differences has been theorized, they
have not yet been examined empirically.
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: individualsperceptions of personal control. Speci-
cally, we seek to uncover the bidirectional relationship between the
psyche and culture in the domain of personal control and cultural
tightness. We test whether individualslack of personal control can
increase their preference for tighter cultures as a means of restoring
order and predictability, and whether being in a tight culture can then
further constrain an individuals personal control. We also test if this
bidirectional relationship also comes with some psychological ben-
ets, as people in tighter cultures may experience an increased sense
of collective control. Interdependent collective action can often help
groups achieve goals that individuals may not otherwise be capable of
attaining alone, and due to the increased social coordination afforded
by tight cultural groups, we test whether tight cultures can bolster
peoples sense of collective control (Bandura, 2000;Fritsche et al.,
2008,2013;Greenaway et al., 2015).
Taken together, these relationships suggest a mutually reinforcing
relationship between personal control and cultural tightness, such
that people experience less personal control and more collective
control as cultures become tighter over time. Our model can thus
provide a theoretical account for the persistence of cultural differ-
ences in personal and collective control and tightnesslooseness
over time.
The Cycle of Mutual Constitution and
Cultural TightnessLooseness
Thus far, one view of culture is that it is a relatively independent,
stable system of shared meanings that structures human experience
(for a review, see Kashima, 2000). In contrast, inuenced by
sociologists and anthropologists such as Bourdieu (1977) and
Giddens (1979), another school of thought argues that macrolevel
cultural characteristics and structure inuence peoples cognitive,
emotional, motivational, and behavioral functioning, which in turn
produce and reproduce these sociocultural environments (Adams
& Markus, 2001;Fiske et al., 1998;Kashima, 2000). This idea is
described in Shweders (1991, p. 24) memorable phrase: culture
and psyche make each other up.
Several lines of research have demonstrated this mutual consti-
tution process by documenting the interrelationship between psy-
chological tendencies and cultural products (Morling & Lamoreaux,
2008), interpersonal situations (Kitayama et al., 1997;Morling
et al., 2002;Savani et al., 2011), and shared narratives (Kashima,
2000,2001;Kashima et al., 2018,2019;Kashima & Kashima,
1998). Recent research has documented the role of mutual consti-
tution processes in perpetuating racial disparities (Salter et al.,
2018). For example, White Americans prefer relatively less critical
representations of Black history from majority White schools over
more critical representations of Black history from predominantly
Black schools. This suggests that White Americans select certain
historical narratives over others, thereby perpetuating racial inequal-
ity (Salter & Adams, 2016).
Despite the above exceptions, most research in cultural psychology
separates the mutual constitution process. Researchers tend to focus
either on how psychological tendencies vary across cultures, thereby
examining the inuence of culture on the psyche (e.g., Nisbett et al.,
2001), or on how products vary by culture, thereby examining
the inuence of the psyche on culture (Morling & Lamoreaux,
2008), but rarely examining both processes simultaneously.
Cultural Tightness
In the current research, we sought to investigate the bidirectional
relationship between culture and psyche in a new domainperceived
control and cultural tightness. The concept of tight cultures was rst
developed in the 60s in a study of traditional societies (Pelto, 1968).
Some societies (e.g., the Azande of Central Africa) possessed strong
norms and imposed severe punishments on those who deviated from
norms. In contrast, other societies (e.g., the Kung Bushman of South
Africa) had weak norms and allowed people to violate norms without
punishment. Since then, this construct has also been applied to
modern societies (Gelfand et al., 2006,2011;Triandis, 1989).
Researchers have suggested that tighter cultures evolved to allow
people to effectively overcome societal threats. For example, Pelto
(1968) argued that tightnesslooseness originated as an ecological
adaptation to the environment, such as population density and
dependency on uncertain harvests. Triandis (1989) suggested that
geographically isolated cultures (e.g., Japan) tend to be homogenous
and tight; in contrast, cultures that are centrally located (e.g.,
Thailand) are more likely to be exposed to different practices and
behaviors from neighboring locales, and thus tend to be loose so that
they can mitigate the threat of social conict and instability arising
from cultural diversity. The notion that tighter cultures developed as
an adaptative response toward mitigating ecological and historical
threats received empirical supporttighter nations tended to have
higher population density, scarce natural resources, and suffer more
natural disasters (Gelfand et al., 2011).
Scholars have also argued that tightness might also be reinforced
by bottom-up processes through which individuals in a culture can
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2MA, SAVANI, LIU, TAI, AND KAY
shape cultural patterns and practices at the societal level (Gelfand et
al., 2006,2011). For example, people who chronically experience a
sense of uncertainty might prefer cultures with strong norms so that
they can reduce their sense of uncertainty (Aktas et al., 2016).
However, limited research has tested for the existence of such
bottom-up processes. We sought to address this gap by providing
evidence for a bidirectional relationship between personal control at
the individual level and tightnesslooseness at the societal level.
Personal Control Shapes Cultural TightnessLooseness
We rst consider how the personal control dimension of the
psyche inuences 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 compen-
satory control theory suggest that the search for nonepistemic
structure, or interpretations of ones social and physical environ-
ments as simple (vs. complex), clear (discernable; not hidden or
obscure, vague or ambiguous), and consistent (stable as opposed to
erratic; marked by a coherent relation of parts vs. disordered)
(Landau et al., 2015, p. 694) might be appealing when people lack
control. A lack of personal control increases peoples preference for
various forms of nonspecic epistemic structure, including social
hierarchy (Friesen et al., 2014), and leads them to perceive con-
nections between unrelated events (Ma et al., 2017;Whitson &
Galinsky, 2008; for other examples, see Landau et al., 2015, for a
review). Thus, consistent with compensatory control theory, we
predict that those with lower personal control would desire greater
structure.
Although most research on compensatory control theory has been
conducted in North American cultural settings, there is reason to
believe that key tenets of the theory generalize across cultures. For
instance, predictions of compensatory control theory have been
validated in the context of the 2008 Malaysian general election (Kay
et al., 2010). Similarly, across countries, lower personal control was
associated with greater condence in the government (Kay et al.,
2008, Study 3), as well as a greater tendency to seek structure and
endorse simple characterizations of various social groups (i.e., to
stereotype; Ma et al., 2019, Study 1).
Thus far, the theory of cultural tightnesslooseness and compen-
satory control theory have been studied in separate literature.
However, we propose that the two theories are related in that the
strength of social normsa core feature of tight culturesgives
people a simple and coherent interpretation of the world and thus
could be considered a form of epistemic structure. As strong norms
both guide peoples own behaviors and allow them to predict others
behaviors, they can provide a signicant source of order and
predictability in everyday social life (Gelfand et al., 2011). Thus,
we predict that to the degree tight cultures are seen as conferring
order and predictability, when people lack control and desire
structure, they may come to prefer tighter cultures.
Indeed, consistent with our idea, past research has found a
positive association between the need for structure and tightness
at the country level (Gelfand et al., 2011). However, this evidence is
largely correlational in nature, and thus cannot speak to the causal
directionality between need for structure and preference for tight
cultures. We propose that the need for structure is a mediator that
captures the compensatory control process linking low perceived
control to the preference for tighter cultures. Put formally, we
predict that:
Hypothesis 1: People who perceive low personal control prefer
tighter cultures.
Hypothesis 2: People who perceive low personal control have a
higher need for structure.
Hypothesis 3: People who have a higher need for structure
prefer tighter cultures.
Hypothesis 4: The relationship between perceived personal
control and preference for tighter cultures is mediated by a
higher need for structure.
TightnessLooseness Shapes Personal and
Collective Control
We next consider how tighter cultures might inuence peoples
sense of control. Control theorists have long noted that there are
multiple ways of attaining a sense of control (Rothbaum et al., 1982).
One notable distinction is between personal control and collective
control (Morling & Evered, 2006). Whereas personal control is
dened as whether individuals perceive that they have control
over the means needed to achieve valued goals, collective control
is dened as whether people believe that they can achieve shared
goals via collective action (also referred to as communal mastery
or collective efcacy;Bandura, 1997;Morling & Evered, 2006;
Yamaguchi, 2001). We theorize that tight cultures have divergent
inuences on peoples perceptions of personal and collective control.
We rst consider how tight cultures inuence personal control.
By denition, tight cultures limit how individuals can act in various
social situationsindividualsbehaviors are circumscribed by prev-
alent norms, and deviations from norms are punished. For example,
Gelfand et al. (2011) found that, in tighter cultures, a narrower range
of behaviors is deemed appropriate in a wide range of social
situations. The constraints on individual behavior associated with
tight cultures can reduce peoples sense of personal control, which
would suggest a cyclical bidirectional relationshiplower personal
control increases peoples preference for tighter cultures, and tighter
cultures reduce peoples sense of personal control. Therefore, we
predict that:
Hypothesis 5: People in tighter cultures perceive lower personal
control.
Further, we predict that, despite impeding personal control, tight
cultures will increase individualssense of collective control.
Greater compliance with norms in tight cultures can help groups
achieve goals that individuals may not otherwise be capable of
attaining alone (Greenaway et al., 2015). Given the increased social
coordination afforded by tight cultures (Gelfand et al., 2011), we test
if people in tight cultures might feel that they are capable of
achieving collective goals. Indeed, according to the group-based
control model, groups are an important source of control as they can
pursue the goals and uphold the values they share with their
members(Fritsche et al., 2008, p. 598; Triandis, 1989). Consistent
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
PERSONAL CONTROL AND CULTURAL TIGHTNESSLOOSENESS 3
with this idea, tighter cultures were better able to control the
COVID-19 pandemic as they could more easily regulate indivi-
dualsactions to achieve the collective goal of containing the virus
(Gelfand et al., 2021). This positive relationship between tight
cultures and collective control can perhaps balance the negative
bidirectional relationship between tighter cultures and low personal
control.
1
Hypothesis 6: People in tighter cultures perceive higher collec-
tive control.
In sum, we propose a full cycle of mutual constitution of the
psyche and culture in the current research. We suggest that societal-
level differences in tightnesslooseness and individual-level per-
ceptions of personal control and collective control can be mutually
reinforced. Our model suggests an unstable equilibrium in which
people in tighter cultures perceive less individual personal control
and greater collective control over time. An instigating instance of
lowered personal control (e.g., COVID-19 pandemic) can increase
peoples desire for structure and, consequently, their preference for a
tighter culture. Being in a tighter culture then reduces peoples sense
of personal control but increases their sense of collective control
over time (Figure 1).
Overview of Studies
To test our hypotheses, we conducted an archival study and ve
correlational and experimental studies across multiple cultures.
Studies 13 examine the correlational (Studies 1 and 2) and causal
(Study 3) relation between personal control on preferences for
tight cultures. Study 1 investigated whether individuals with lower
levels of perceived control prefer to live in tighter U.S. states
(Hypothesis 1). Employing a two-wave longitudinal survey design
with employees in China, Study 2 examined whether employees
who experience lower levels of perceived control have a greater
need for structure and, consequently, prefer tighter cultures
(Hypotheses 14). Study 3 tested whether experimentally induced
feelings of low personal controlled Singaporeans to punish norm
violators in an economic game (Hypothesis 1). Together with
Study 3, Studies 4 and 5 employed an experimental chain design
and tested the causal relationships between personal control, need
for structure, and preference for a tighter culture. Specically,
Study 4 tested whether experimentally manipulated feelings of low
perceived control increase peoples need for structure (Hypothesis
2); Study 5 examined whether experimentally induced need for
structure increases peoples preference for tighter organizational
cultures (Hypothesis 3). Finally, Studies 6a and 6b manipulated
cultural tightness and examined whether being in a tighter culture
reduces peoples perceptions of personal control but increases
their feelings of collective control (Hypotheses 5 & 6).
Across all studies, we report all participants, measures, and
conditions. The present research was approved by the institutional
review board at the authorsuniversities. All scale items used in
the present research can be found in the Supplemental Materials.
All survey materials, data, and analysis code are available at
https://osf.io/f8wzh. This study was approved by Tulane Univer-
sitys institutional review board protocol IRB-2020-1432, titled
Attitudes and Beliefs in the Workplace,and Nanyang Techno-
logical Universitys institutional review board protocol IRB-
2015-07-018-06, titled The Role of Implicit Processes in Cultural
Learning.
Study 1: Preferences to Live in Tighter or
Looser U.S. States
Study 1 tested Hypothesis 1, whether people who perceive low
personal control prefer tighter cultures. We assessed preference for
tighter cultures indirectly based on which U.S. state participants
wanted to move to. Specically, we predicted that people with lower
perceived control would prefer to live in culturally tighter states.
Method
Participants
We analyzed individuals who completed Wave 2 (N=3,422;
1,863 women, 1,559 men, M
age
=55.66, SD =12.26) and Wave 3
(N=2,295; 1,244 women, 1,051 men; M
age
=63.96, SD =11.02) of
a three-wave Midlife in the United States (MIDUS) longitudinal
survey (Brim et al., 2004). We did not use the Wave 1 data because it
did not contain the dependent variable of interest.
Procedure
Perceived personal control was assessed using an established
12-item scale (Ettner & Grzywacz, 2001;Lachman & Weaver,
1998a,1998b, e.g., There is little I can do to change the important
things in my life,α
Wave 2
=.87, α
Wave 3
=.87). These items were
measured on a 7-point scale (1 =agree strongly,7=disagree
strongly). We coded participantsresponses so that higher
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Figure 1
Theorized Model and Overview of Studies
Note. See the online article for the color version of this gure.
1
Our prediction that tighter cultures would decrease personal control but
increase collective control is consistent with the literature on secondary
control, which has found that people accept a loss of personal control if they
can gain a sense of secondary control through third parties (Rothbaum et al.,
1982). For example, although people forfeit their sense of personal control
when they submit to powerful others, doing so can also increase their sense of
vicarious secondary control through the powerful agent (Rothbaum et al.,
1982).
4MA, SAVANI, LIU, TAI, AND KAY
scores indicated greater personal control (M
wave 2
=5.56, 95%
CI [5.52, 5.59], SD
wave 2
=.98, M
wave 3
=5.48, 95% CI [5.44, 5.53],
SD
wave 3
=1.00).
Although the survey did not collect information about which state
each participant was currently living in, participants in both waves
were asked, Thinking back over all the places youve lived during
your lifetime, including where you live now, which state would you
most like to live in for the next 10 years if you could easily move
there now?We computed the tightnesslooseness score of parti-
cipantspreferred state based on Harrington and Gelfands (2014)
work. Harrington and Gelfand (2014) developed a measure of state-
level tightnesslooseness using a wide variety of indicators of
tightnesslooseness. These indicators included strength of punish-
ment (e.g., the legality of corporal punishment, punitiveness of
laws), latitude/permissiveness (access to alcohol), diversity (as
measured by the percentage of total population that is foreign),
and prevalence and strength of institutions (e.g., state-level religi-
osity). Higher scores on this measure indicated greater preference
for tighter states (M
wave 2
=48.26, 95% CI [47.84, 48.67], SD
wave 2
=
12.42, M
wave 3
=48.47, 95% CI [47.95, 48.99], SD
wave 3
=12.71).
We included participantsgender, age, income, and education as
covariates.
2
We also included state-level variables that were corre-
lated with tightnesslooseness, including political orientation of
the state (Jones, 2005;Saad, 2014), individualismcollectivism
(Vandello & Cohen, 1999), and gross product of the state
(Bureau of Economic Analysis, 2006,2014).
Results
See Table 1 for means, standard deviations, and correlations
between variables. As this was a longitudinal study, most partici-
pants completed both Wave 2 and Wave 3, which led to noninde-
pendence of observations. Therefore, we conducted a multilevel
linear regression model with random intercepts. Responses at Wave
2 and Wave 3 nested within participants. We regressed participants
preferred states tightnesslooseness score (the dependent variable)
on perceived control (the independent variable). Consistent with
Hypothesis 1, lower perceived personal control was associated with
a greater preference for tighter states (Table 2, Model 1). This
persisted and even strengthened after controlling for participants
gender, age, income, education, state-level political party afliation,
state-level individualismcollectivism, and state-level gross national
product (Table 2,Model2).
Discussion
Study 1 found that people with lower personal control preferred to
move to tighter U.S. states. As people may desire to move to various
states for reasons other than personal control, we ensured that this
relationship was robust after controlling for potential confounds,
such as the collectivism, political orientation, and per capita gross
domestic product (GDP) of the state.
An important limitation of this study is that it did not assess the
state participants were currently living in. As it is possible that
participants may have reported the state in which they currently
reside as their preferred state, an alternative explanation is that this
study merely shows that people from tighter states experience
lower personal control. To remedy this limitation, Study 3 tests
the causal effect of personal control on preference for tighter
cultures.
Study 2: Evidence From a Longitudinal
Two-Wave Study in China
Study 2 employed a longitudinal study design to provide further
evidence for the relationships between perceived control, need for
structure, and preference for tighter cultures (Hypotheses 14). With
the assistance of the companys human resource management
department, we invited all employees of an apparel retailer located
in China to participate in our study. We tested whether lower
personal control measured at Time 1 will be associated with a
greater preference for a tighter organizational culture (Hypothesis
1), as well as a greater need for structure (Hypothesis 2) measured at
both Time 1 and Time 2. We also tested whether a greater need for
structure measured at Time 1 will be associated with a greater
preference for a tighter organizational culture measured at both Time
1 and Time 2 (Hypothesis 3). We then tested whether the need for
structure measured at Time 2 will mediate the predictive effect of
personal control assessed at Time 1 on preference for tighter cultures
measured at Time 2 (Hypothesis 4). Finally, to provide converging
mediation evidence, we also examined if the effect of personal
control (Time 1) on preference for a tighter organization (Time 2)
was mediated by a greater need for structure (Time 1).
Method
The hypotheses, power analysis, method, sample size, and pre-
selection criteria for this study were preregistered at http://aspredicte
d.org/blind.php?x=vf27gn.
Participants
As part of a larger study, participants completed measures of
perceived personal control, need for structure, and preference for a
tighter organizational culture. We sought to recruit as many parti-
cipants as possible from an apparel retailer in southern China. To
ensure condentiality, identication codes were used to match
subordinatesupervisor dyads across different times. All employees
were assured that their responses were completely condential to
alleviate evaluation apprehension and increase the candidness of
responses. A total of 323 employees were invited to participate in the
present survey, and 225 participants provided complete responses
(response rate of 70%). No participants were excluded from this
study. Our sample included 225 employees (148 women, 77 men;
M
age
=31.18, SD =2.34).
Procedure
Perceived control was assessed using a 10-item scale (taken from
Ma & Kay, 2017,I am in control of my life,M
wave 1
=4.59, 95%
CI [4.48, 4.71], SD
wave 1
=.86, M
wave 2
=4.51, 95% CI [4.40, 4.61],
SD
wave 2
=.81). We measured need for structure using a 12-item
scale (taken from Neuberg & Newsom, 1993,I hate to change my
plans at the last minute,M
wave 1
=4.33, 95% CI [4.24, 4.42],
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2
Individualspolitical orientation was not measured in the MIDUS
survey. We also sought to include the race/ethnicity of the participant as
a covariate, but this information was not available.
PERSONAL CONTROL AND CULTURAL TIGHTNESSLOOSENESS 5
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 1
Variable Intercorrelations, Means, and Standard Deviations Among Variables (Study 1)
Variable MSD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1. Perceived control (Wave 2) 5.56 0.98 (0.87)
2. Perceived control (Wave 3) 5.48 1.00 0.64*** (0.87)
3. Gender (1 =male,
2=female)
1.54 0.50 0.07*** 0.06*
4. Age (Wave 2) 55.66 12.26 0.04*0.06** 0.03
5. Age (Wave 3) 63.96 11.01 0.06** 0.09** 0.03 1.00***
6. Education (Wave 2) 7.37 2.52 0.16*** 0.17*** 0.11*** 0.14*** 0.11***
7. Education (Wave 3) 7.63 2.49 0.13*** 0.18*** 0.12*** 0.13*** 0.13*** 0.90***
8. Income (Wave 2) 42458.94 40543.80 0.18*** 0.18*** 0.31*** 0.18*** 0.12*** 0.35*** 0.35***
9. Income (Wave 3) 57078.54 58991.30 0.15*** 0.19*** 0.31*** 0.14*** 0.15*** 0.36*** 0.36*** 0.62***
10. % of republicans (Wave 2) 45.62 6.46 0.01 0.03 0.01 0.02 0.01 0.07** 0.06** 0.04*0.02
11. % of republicans (Wave 3) 40.13 6.22 0.04 0.03 0.01 0.01 0.02 0.14*** 0.12*** 0.07** 0.07*** 0.55***
12. Collectivism (Wave 2) 50.36 8.94 0.01 0.01 0.01 0.03 0.03 0.01 0.04 0.03 0.06** 0.08*** 0.18***
13. Collectivism (Wave 3) 50.17 9.11 0.02 0.01 0.01 0.02 0.02 0.01 0.01 0.03 0.04 0.04 0.20*** 0.74***
14. Tightnesslooseness
(Wave 2)
48.26 12.42 0.05** 0.04 0.03 0.02 0.02 0.11*** 0.08*** 0.09*** 0.06** 0.48*** 0.44*** 0.24*** 0.19***
15. Tightnesslooseness
(Wave 3)
48.47 12.71 0.04 0.03 0.03 0.01 0.02 0.11*** 0.10*** 0.06** 0.08*** 0.36*** 0.60*** 0.19*** 0.27*** 0.77***
16. Gross state product
(logged, Wave 2)
12.57 0.99 0.01 0.02 0.01 0.02 0.02 0.08*** 0.09*** 0.04** 0.05*0.32*** 0.43*** 0.50*** 0.39*** 0.23*** 0.15***
17. Gross state product
(logged, Wave 3)
12.94 0.99 0.02 0.03 0.02 0.01 0.01 0.08*** 0.08*** 0.06*0.06** 0.20*** 0.59*** 0.41*** 0.50*** 0.14*** 0.20*** 0.70***
*p<.05. ** p<.01. *** p<.001.
6MA, SAVANI, LIU, TAI, AND KAY
SD
wave 1
=.66, M
wave 2
=4.49, 95% CI [4.39, 4.59], SD
wave 2
=.74).
Finally, we adapted Gelfand et al.s (2011) measure of cultural
tightness and developed a ve-item
3
scale assessing preference for a
tighter organization (e.g., My company should have more social
norms that people should abide by,M
wave 1
=4.64, SD
wave 1
=.84,
95% CI [4.53, 4.75], M
wave 2
=4.72, 95% CI [4.61, 4.84], SD
wave 2
=.86). All items were measured on a 7-point scale (1 =strongly
disagree,7=strongly agree). Higher values on these three scales
indicated greater perceived personal control, need for structure, and
preference for a tighter organization. All scale items used can be
found in the Supplemental Materials.
Results
Variable means, standard deviations, and intercorrelations are
provided in Table 3. Conrmatory factor analyses with personal
control, need for structure, and tightnesslooseness measures at
Time 1 indicated a well-tting correlated three-factor structure,
χ
2
(321) =495.47, p<.001, comparative t index (CFI) =.95,
TuckerLewis index (TLI) =.94, root-mean-square error of approx-
imation (RMSEA) =.049.
Supporting Hypothesis 1, lower perceived personal control mea-
sured at Time 1 was associated with a preference for a tighter
organization at both Time 1, b=.17, 95% CI [.298, .045], β=
.18, SE =.06, t(223) =2.68, p=.008, η
2
=.03, and Time 2, b=
.25, 95% CI [.378, .120], β=.25, SE =.07, t(223) =3.81,
p<.001, η
2
=.06. Consistent with Hypothesis 2, lower perceived
personal control measured at Time 1 was also associated with a
greater need for structure measured at both Time 1, b=.12, 95%
CI [.223, .022], β=.16, SE =.05, t(223) =2.39, p=.018,
η
2
=.03, and at Time 2, b=.29, 95% CI [.401, .186], β=.34,
SE =.05, t(223) =5.37, p<.001, η
2
=.11. Supporting Hypothesis 3,
greater need for structure measured at Time 1 was associated with a
preference for a tighter organization measured at both Time 1, b=
.21, 95% CI [.042, .371], β=.16, SE =.08, t(223) =2.48, p=.014,
η
2
=.03, and Time 2, b=.27, β=.21, 95% CI [.105, .442], SE =.09,
t(223) =3.20, p=.002, η
2
=.04.
We tested a two-wave longitudinal mediation model recom-
mended by Little et al. (2007). We bootstrapped the product term
of the perceived personal control (Time 1) to need for structure
(Time 2) and the need for structure (Time 1) to preference for a
tighter organization (Time 2) paths. Supporting Hypothesis 4, we
found that the 95% condence intervals of bootstrapped indirect
effect did not include zero, suggesting signicant mediation, Coeff
=.02, SE =.01, 95% CI [.057, .003] (Figure 2). The indirect
effect was signicant after controlling for individualsgender, age,
and education level, Coeff =.02, SE =.01, 95% CI [.059,
.003]. Finally, we tested reverse mediation and bootstrapped the
product term of the preference for a tighter organization (Time 1) to
need for structure (Time 2) and the need for structure (Time 1) and
personal control (Time 2) paths. We found that the 95% condence
intervals of bootstrapped indirect effect included zero, ruling out
reverse mediation, Coeff =.03, SE =.03, 95% CI [.017, .091].
As an additional test of mediation, we also examined if the effect
of personal control (Time 1) on preference for a tighter organization
(Time 2) was mediated by a greater need for structure (Time 1).
Again, we found that the 95% condence intervals of bootstrapped
indirect effect did not include zero, suggesting signicant mediation,
Coeff =.03, SE =.02, 95% CI [.065, .002]. The indirect effect
was signicant after controlling for individualsgender, age, and
education level, Coeff =.03, SE =.02, 95% CI [.064, .003].
The reverse mediation with the preference for a tighter organization
(Time 2) as the independent variable, need for structure (Time 1) as
the mediator, and perceived personal control (Time 1) as the
dependent variable did not yield a signicant indirect effect, Coeff
=.02, SE =.02, 95% CI [.06, .004].
Discussion
Employing a longitudinal design, Study 2 provided further evi-
dence for the hypothesized relationships in a eld setting. We found
that employees with lower personal control had a greater need for
structure and, therefore, preferred their organization to have a tighter
culture. Further, consistent with our hypothesized model, need for
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 2
Preference for Tight (vs. Loose) States as a Function of Perceived Personal Control (Study 1)
Variable
Model 1 Model 2
b95% CI SE z b SE 95% CI z
Intercept 49.87** [48.25, 51.48] 0.82 60.48 32.55*** 2.80 [27.06, 38.05] 11.61
Perceived control 0.29*[0.57, 0.01] 0.15 1.97 0.28*0.13 [0.55, 0.02] 2.08
Gender (1 =male, 2 =female) 0.83*0.34 [0.15, 1.50] 2.41
Age 0.16*** 0.01 [0.13, 0.18] 13.10
Income 0.01 0.01 [0.01, 0.01] 1.57
Education 0.21*** 0.06 [0.34, 0.09] 3.30
Gross product of state that Pwants to live (log) 2.92*** 0.17 [3.26, 2.59] 16.93
Political conservatism of state that Pwants to live 0.46*** 0.02 [0.42, 0.50] 22.75
Collectivism of state that Pwants to live 0.50*** 0.02 [0.47, 0.54] 28.22
N (Level 1) 6,320 5,717
N (Level 2) 4,036 3,750
Note.Nfor Model 2 is smaller due to missing demographic information. CI =condence interval; SE =standard error.
*p<.05. ** p<.01. *** p<.001.
3
We did not include an adapted version of the item In this country, if
someone acts in an inappropriate way, others will strongly disapprove
because we received feedback from the company representative that the
adapted version of this item, In this company, if someone acts in an
inappropriate way, others should more strongly disapprovewas unclear
in Chinese and might be difcult for employees to respond to.
PERSONAL CONTROL AND CULTURAL TIGHTNESSLOOSENESS 7
structure mediated the link between perceived control and prefer-
ence for tighter cultures (as hypothesized), but not the reverse link
between preference for tighter cultures and perceived control.
Study 3: Perceived Personal Control
Cultivating a Culture of Tightness
Employing an experimental chain design (Spencer et al., 2005),
Studies 36 aimed to provide causal evidence for our model.
Specically, Study 3 tested whether experimentally induced lack
of personal control increases peoples likelihood of enforcing norms
(a key component of tightness) in an economic game. Participants
had to decide whether to punish or reward a person who acted in
either a self-interested or a prosocial manner. We framed coopera-
tion as the more socially desirable behavior and posttested the game
instructions to ensure that participants perceived that this was the
case. We predicted that participants with low personal control would
be more likely to uphold this norm of cooperation by rewarding
prosocial behavior and punishing self-interested behavior even at a
monetary cost to themselves (Fehr & Gachter, 2000).
Further, we sought to address the role of perceived power as an
alternative explanation. Control and power are conceptually linked
(Inesi et al., 2011), and people with higher power (who also tend to
have higher personal control) prefer situations with fewer situational
constraints (Fiske, 1993;Galinsky et al., 2008;Whitson et al.,
2013). Therefore, we assessed whether personal control drives
preference for tighter cultures above and beyond the sense of power.
Method
Participants
As this was the rst study that employed this manipulation and
assessed this dependent variable, we assumed a moderate effect size
of d=.50 (Richard et al., 2003). A power analysis with an effect size
of d=.50, p=.05 (two-tailed), and power =80% indicated that we
needed to recruit 128 participants. We posted a study seeking 128
participants in the lab at a university in Singapore, but only 126
participants showed up (84 women, 40 men; 2 unreported, mean age
of 22.76 years, SD =2.06).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 3
Variable Intercorrelations, Means, and Standard Deviations Among Variables (Study 3)
Variable MSD 123456
1. Perceived control (Wave 1) 4.59 0.86 (.93)
2. Perceived control (Wave 2) 4.51 0.81 0.38*** (.90)
3. Need for structure (Wave 1) 4.33 0.66 0.16*0.02 (.90)
4. Need for structure (Wave 2) 4.49 0.74 0.34*** 0.21** 0.28*** (.88)
5. Desire for organization greater tightnesslooseness (Wave 1) 4.64 0.84 0.18** 0.09 0.16*0.41*** (.90)
6. Desire for organization greater tightnesslooseness (Wave 2) 4.72 0.86 0.25*** 0.10 0.21** 0.46*** 0.89*** (.85)
Note.N=225.
*p<.05. ** p<.01. *** p<.001.
Figure 2
Need for Structure Mediates the Relationship Between Personal Control and Desire
for a Tighter Organizational Culture
Note. Bolded paths were used in the computation of the indirect effect. Unstandardized
coefcients. Study 3.
*p<.05. ** p<.01. *** p<.001.
8MA, SAVANI, LIU, TAI, AND KAY
Procedure
All participants were asked to complete a concept identication
task (Whitson & Galinsky, 2008). Participants were randomly
assigned to either the high or the low-control condition. All parti-
cipants were informed that they would complete a concept identi-
cation task, in which they would be presented with pairs of symbols,
and in each pair, one symbol represents the concept while the other
does not. They were told that they would be given feedback about
whether they had chosen the correct symbol whenever they selected
a symbol.
Participants rst completed a practice session and then four
sessions with four different concepts. Each session had 10 rounds.
Participants in the control condition did not receive any feedback for
each round. In contrast, participants in the low-control condition
received random feedback not contingent on their responses. Past
research has found that feedback that is unrelated to performance
reduces peoples sense of control (Pittman & DAgostino, 1989;
Pittman & Pittman, 1979;Whitson & Galinsky, 2008).
After the personal control manipulation, participants were
informed that they would be playing a points allocation game
and that they would receive a nal bonus depending on the decisions
that they and two other participants made. They were told that the
identity of the other participants was anonymous, but the bonus
payments that they would receive were real. Participants were then
asked to complete a three-person strategic game with three different
roles: the Decider, the Receiver, and the Judge (adapted from Ding
& Savani, 2020, Study 4; see also Peysakhovich et al., 2014;Strang
et al., 2017). The game comprised 20 rounds. For each round, the
Decider received 10 points whereas the Receiver received 0 points.
The Decider would have to decide whether to keep all 10 points for
themselves and give 0 points to the Receiver (i.e., to act in a self-
serving manner), or to give 5 points to the Receiver and keep 5
points for themselves (i.e., to act in an altruistic manner). The
Receiver had no choice but to accept what the Decider gave them.
The Judge received 20 points in each round and was informed
about the Deciders decision in each round. If the Decider acted in a
self-serving manner, the Judge could choose to give up 5 of their
points to punish the Decider, who would receive a deduction of 10
points and thus end up with nothing. If the Decider acted altruisti-
cally, the Judge could choose to give up 5 of their points to reward
the Decider, who would receive 10 additional points (to receive a
total of 15 points). Because Judges were asked if they wanted to
rewardDeciders when Deciders chose to split their points evenly
or to punishDeciders when Deciders chose to split their points
unevenly, the wording of the instructions implied that acting
prosocially was the socially desirable behavior. A posttest con-
rmed our assumption (see Supplemental Materials). Participants
were also informed that they would receive their cumulative points
in the form of a monetary bonus, with 1,000 points equaling one
Singapore dollar.
Unbeknownst to participants, all of them were assigned to the role
of a Judge. All participants played 10 rounds in which they were told
that the Decider acted self-servingly, and 10 rounds in which they were
told that the Decider acted altruistically. We randomized the order of
these 20 rounds to address the possibility that the effects are explained
by a particular sequence of choices on the part of the Decider.
4
After completing the 20 trials, participants were asked to com-
plete an eight-item scale that assessed their subjective feelings of
power when they played the role of the Judge in the game (e.g.,
During the experimental game, I can get the other participants to
listen to what I say;During the experimental game, my wishes did
not carry much weight;During the experimental game, I could get
the other participants to do what I wanted; adapted from Anderson
et al., 2012;α=.79; see Supplemental Materials, for the full scale).
Results
Across the 20 rounds, we summed the number of times that
participants chose to uphold the norm of cooperation by rewarding
Deciders when they acted altruistically and punishing Deciders
when they acted self-servingly. As the dependent variable was a
count variable, we analyzed it using a Poisson regression. As
predicted, participants in the low-control condition (M=8.83,
95% CI [7.08, 10.57], SD =6.93) chose to uphold the norm of
cooperation more so than participants in the high-control condition
(M=6.73, 95% CI [5.07, 8.39], SD =6.58), b=.27, 95% CI
[.40, .14], SE =.06, incidence rate ratio =.76, z=4.20, p<
001. Another Poisson regression found that this nding held even
after controlling for perceived power,
5
b=.32, 95% CI [.45,
.19], SE =.07, incidence rate ratio =.73, z=4.87, p<.001.
We also examined the effect of manipulated control on partici-
pantstendency to punish selsh behaviors separately from the
tendency to reward cooperative behaviors. We found that partici-
pants in the low-control condition (M=4.27, 95% CI [3.27, 5.27],
SD =3.96) chose to reward cooperative behaviors more so than
participants in the high-control condition (M=2.89, 95% CI [2.00,
3.77], SD =3.52), b =.39, 95% CI [.58, .20], SE =.10,
incidence rate ratio =.68, z=4.07, p<001. Participants in the
low-control condition (M=4.56, 95% CI [3.54, 5.57], SD =4.03)
chose to punish selsh behaviors more so than participants in the
high-control condition (M=3.84, 95% CI [2.85, 4.84], SD =3.95),
b=.17, 95% CI [.34, .01], SE =.09, incidence rate ratio =.84,
z=1.95, p=.051. Finally, the effect of control on rewarding and
punishing behaviors held after controlling for participantssense of
power. A Poisson regression found that compared to those in the
high-control condition, participants in the low-control condition
were more likely to punish deciders who acted selshly, b=.20,
95% CI [.37, .02], SE =.09, incidence rate ratio =.82, z=2.24,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4
It is possible that different sequences of altruistic versus self-serving
decisions by the Decider might have inuenced participantsdecision to
reward or punish the Decider. For example, Judges in the low-control
condition might be more likely to punish Deciders when Decidersself-
serving choices were clustered mostly in the rst 10 decisions instead of
evenly distributed across the 20 trials. To address this concern, we random-
ized the order of the 20 trials independent of the experimental manipulation.
5
We further examined whether a higher sense of power can explain the
effect of the experimental manipulation in the high-control condition. We
found power varied in an opposite direction as expectedparticipants in the
high-control condition felt a lower sense of power (M=4.41, 95% CI [4.14,
4.67], SD =1.06) than those in the low-control condition (M=4.79, 95% CI
[4.52, 5.05], SD =1.05), t(124) =2.02, p=.046, d=0.36. Importantly,
participantssense of power was also not signicantly correlated with the
number of times participants chose to uphold social norms by punishing self-
serving deciders and rewarding altruistic deciders (r=.12, p=.19).
Breaking down the two types of trials, we found that perceived power was
uncorrelated with the number of times participants rewarded altruistic
behaviors (r=.14, p=.11) or punished self-serving behaviors (r=
.06, p=.48). These ndings suggest that power cannot serve as an
alternative explanation of the present ndings.
PERSONAL CONTROL AND CULTURAL TIGHTNESSLOOSENESS 9
p=.025; they were also more likely to reward deciders who acted
cooperatively, b=.46, 95% CI [.65, .27], SE =.10, incidence
rate ratio =.63, z=4.74, p<.001.
Discussion
In sum, Study 3 demonstrated that participants experiencing low
personal control were more willing to uphold the norm of coopera-
tion even at a nancial cost to themselves. Importantly, this effect
was robust even after controlling for participantssense of power.
Study 4: Perceived Personal Control
Need for Structure
In this study, we manipulated perceived personal control and
tested whether it inuenced participantsneed for structure (Hypoth-
esis 2). Further, as in Study 3, we addressed feelings of power as an
alternative explanation. In addition, because participants might feel
more negative affect and less positive affect in the low personal
control condition than in the high personal control condition, we
also controlled for positive and negative affect in our analyses.
Method
The hypotheses, power analysis, method, sample size, and pre-
selection criteria for this study were preregistered at https://osf.io/
bzeq9.
Participants
Landau et al. (2015) estimated that the average effect of the recall
control manipulation that we sought to employ in this study on
structure was r=.23 or d=.47. Given an effect size of Cohen d=
.47, a=.05 (two-tailed), and power =.90, we need to recruit 194
participants. We rounded up this number to 200 participants and
posted a survey seeking 200 U.S. residents on Amazon Mechanical
Turk. In response, 201 participants completed the survey. Consistent
with our preregistration, we excluded eight participants with dupli-
cate IP addresses or geolocations. The nal sample included 193
participants (98 women, 95 men; M
age
=44.75, SD =12.38).
Procedure
Participants were randomly assigned to either the high-control or
the low-control condition. Participants who were assigned to the low
personal control condition were asked:
Please recall a time in the past when you had low control. For the next 2
minutes, please write about the experience in which you felt a complete
lack of control over the situation. In other words, write about a time
when you experienced very low or no feelings of control. Describe the
situation, what exactly happened, and how you felt at that moment in
detail. Please use at least 60 words to describe the situation.
Participants assigned to the high-control condition were asked:
Please recall a time in the past when you had high control. For the next 2
minutes, please write about the experience in which you felt complete
control over the situation. In other words, write about a time when you
experienced a great sense of control. Describe the situation, what
exactly happened, and how you felt at that moment in detail. Please
use at least 60 words to describe the situation.
All participants then responded to a four-item state need for
structure scale adapted from Neuberg and Newsom (1993):Right
now, I would not like to be in situations where the rules are not
clear,”“Right now, I would hate to be in unpredictable situations,
Right now, I would not like to work with people who are
unpredictable,and Right now, I would feel upset if I go into a
situation without knowing what I can expect from it(α=.82).
Finally, we assessed participantspositive and negative affect using
the 20-item Positive and Negative Affect Schedule (PANAS) scale
(α
positive
=.93; α
negative
=.94, Watson et al., 1988) and sense of
power using an eight-item scale (α=.94, Anderson et al., 2012).
Results
As preregistered, we excluded three participants who provided
gibberish responses (e.g., good,”“yes) to the open-ended prompt
after the experimental manipulation. Consistent with Hypothesis 2,
we found that participants in the low personal control condition
(M=5.43, 95% CI [5.22, 5.65], SD =1.07) reported a signicantly
higher need for structure than those in the high personal control
condition (M=4.64, 95% CI [4.36, 4.92], SD =1.37), t(188) =4.49,
p<.001, d=.64. A regression with need for structure as the
dependent variable and personal control (0 =low control, 1 =high
control), sense of power, and positive and negative affect as pre-
dictors indicated a signicant effect of condition, b=.79, 95%
CI [1.13, .44], β=.31, SE =.18, t(185) =4.50, p<.001, η
2
=
.10. Perceived power, b=.16, 95% CI [.03, .34], β=.14, SE =.09,
t(185) =1.70, p=.091, η
2
=.02, and positive affect, b=.05, 95%
CI [.10, .19], β=.05, SE =.07, t(185) =.66, p=.511, η
2
=.01,
were not signicantly associated with greater need for structure.
Greater negative affect was associated with greater need for struc-
ture, b=.43, 95% CI [.09, .76], β=.18, SE =.17, t(185) =2.50,
p=.013, η
2
=.03.
Discussion
Study 4 provided support for the causal effect of personal
control on the need for structure. Specically, we found that
participants who recalled a time when they had low personal
control wanted more structure than those who recalled a time
when they had high personal control. Importantly, the causal effect
of personal control on the need for structure remained signicant
after controlling for participantssense of power, positive affect,
and negative affect.
Study 5: Need for Structure TightnessLooseness
Study 5 tested whether a high need for structure increases
peoples desire for tighter cultures (Hypothesis 3). Once again,
we measured perceived power and positive and negative affect
measures as potential confounds.
Further, in this study, we employed a different measure of tight
cultures to provide converging evidence for our predictions. As we
recruited respondents from a variety of different organizations in
this study. To facilitate comparability, we opted to use a measure of
tightnesslooseness that assessed the appropriateness of various
concrete behaviors in the workplace (e.g., swearing, arguing).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
10 MA, SAVANI, LIU, TAI, AND KAY
Method
Participants
As this was the rst experiment in which we manipulated the
need for structure, we could not use a prior effect size to conduct a
power analysis. Nevertheless, we conducted a power analysis
assuming a medium effect size, d=.40, p=.05 (two-tailed), and
power =80%, indicating that we needed 200 participants. We
posted a survey seeking 200 U.S. residents employed full-time
and working from their ofce (as opposed to from their homes or
commercial locations) from Prolic Academic. We received
responses from 202 participants. Consistent with the other online
studies, we excluded two participants with duplicate IP ad-
dresses/geolocations. The nal sample included data from 200
participants (174 women, 25 men, and 1 other gender; M
age
=27.
20, SD =7.66).
Procedure
Participants were randomly assigned to either a high or a low need
for structure condition. In the low need for structure condition,
participants were told:
Extensive research has found that to be happy and successful at work,
employees need to be spontaneous, to avoid being creatures of habits
and routines. Successful workers are those who dont have a single,
xed work schedule.
Participants were then given three open-ended boxes and asked to
indicate three reasons why it is important to work in a free,
spontaneous manner at your current workplace.
In the high need for structure condition, participants were told:
Extensive research has found that to be happy and successful at work,
employees need to be organized, and to have regular habits and routines.
Successful workers are those who have a clear and structured work
schedule.
They were then given three open-ended textboxes and asked to
indicate three reasons why it is important to work in a structured,
organized manner at your current workplace.
After completing this task, all participants were informed:
From everyday experiences interacting with our coworkers, we have
all developed a subjective impressionor feelingfor the appro-
priateness of any given behavior in the workplace. We are interested
in your judgment of the appropriateness of a number of behaviors at
your current workplace. Please rate the extent to which you think
each of the following behaviors should be appropriate at your current
workplace.
To assess the desire for a tighter culture, we presented all
participants with 12 behaviors (e.g., arguing, laughing, kissing;
taken from Gelfand et al., 2011) and asked them to rate whether
each behavior was appropriate at their current workplace (1 =
extremely inappropriate,7=extremely appropriate;α=.74).
We recoded these items such that higher values indicated greater
inappropriateness, indicating tighter cultures, and averaged them to
form a composite score. Finally, as in Study 4, we assessed
participantspositive and negative affect using the PANAS scale
(α
positive
=.93, α
negative
=.90), and their sense of power (α=.87).
Results
No participants provided gibberish or nonsensical answers in the
open-ended prompts in the experimental manipulation. An inde-
pendent samples ttest found that participants in the high need for
structure condition indicated that the various behaviors were less
appropriate in their workplace (M=4.38, 95% CI [4.22, 4.54], SD =
0.80) than those in the low need for structure condition (M=4.14,
95% CI [4.02, 4.26], SD =0.62), t(198) =2.43, p=.016, d=0.34,
consistent with Hypothesis 3.
Next, we regressed participantsdesire for a tight organization on
experimental conditions (0 =low need for structure,1=high need
for structure), perceived power, and positive and negative affect.
The effect of the experimental manipulation was signicant, b=.22,
95% CI [.03, .42], β=.15, SE =.10, t(195) =2.23, p=.027, η
2
=
.02. Perceived power, b=.08, 95% CI [.19, .04], β=.10, SE =
.06, t(195) =1.32, p=.187, η
2
=.01, and positive affect, b=.03,
95% CI [.05, .10], β=.05, SE =.04, t(195) =.71, p=.479, η
2
=
.01, were not signicantly associated with a desire for a tight
organization. Greater negative affect was associated with greater
need for structure, b=.15, 95% CI [.26, .05], β=.21, SE =
.05, t(195) =2.82, p=.005, η
2
=.04.
Discussion
Study 5 offered causal support for the relationship between the
need for structure and desire for organizational tightness. Partici-
pants who thought about the benets of having more structure and
order in the workplace perceived those various behaviors as less
appropriate in their workplace than those who thought about the
benets of having more exibility. Although we did not directly
measure desire or preference for a tighter culture in this study, our
dependent measure is a direct measure of tightness as fewer
behaviors are appropriate in each situation in tight cultures. Given
that the objective tightness of participantsorganization was likely
randomly distributed across conditions, the ndings are consistent
with our hypothesis that a high need for structure increases peoples
preference for a tighter culture.
Study 6a: TightnessLooseness Personal Control
Study 6a tested whether being in a tight culture reduces peoples
perceptions of personal control (Hypothesis 5).
Method
The hypotheses, power analysis, method, and preselection criteria
for this study were preregistered at https://osf.io/upv3q.
Participants
A previous study employing a similar design yielded an effect
size of d=1.89 for personal control. A power analysis using
Cohensd=1.89, α=.05 (two-tailed), and power =95%, indicated
that we need to recruit 18 participants. However, to keep up with
conventional sample size levels, we decided to recruit 100 partici-
pants. We posted a survey seeking 100 U.S. residents on Amazon
Mechanical Turk and received 100 responses. We sought to exclude
participants with duplicate IP addresses/geolocations. Data from
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
PERSONAL CONTROL AND CULTURAL TIGHTNESSLOOSENESS 11
two participants were excluded. The nal sample thus included 98
participants (44 women, 54 men, M
age
=40.89, SD =11.68).
Procedure
Participants were randomly assigned to read about a company that
either had a tight culture or a loose culture (see Supplemental
Materials). After they read the description, participants were asked
to summarize the companys beliefs and to provide one or two
examples from their lives that were consistent with the companys
beliefs. Next, as a manipulation check, participants indicated the
extent to which they shared the companys belief that it is important
(vs. not important) to have many social rules and norms in the
workplace (1 =not at all, 100 =a great deal). We reverse-coded the
manipulation check item in the culturally loose company so that
higher numbers indicated greater perceived importance of social
rules and norms in both conditions.
Thereafter, participants were asked to imagine that they accepted
a job in the company and were asked to complete a three-item
measure of personal control (e.g., I would feel in control of my
life,α=.91; adapted from Greenaway et al., 2015). All items were
measured on 7-point scales (1 =strongly disagree,7=
strongly agree).
Results
As preregistered, we excluded one participant who provided
gibberish responses to the open-ended prompts used after the
manipulation. The manipulation check was successfulparticipants
in the culturally loose condition believed that social rules and norms
were less important (M=45.25, 95% CI [36.13, 54.38], SD =32.46)
than those in the culturally tight condition (M=75.11, 95% CI
[68.96, 81.26], SD =20.70), t(95) =5.33, p<.001, d=1.10.
Consistent with Hypothesis 5, participants who were asked to
imagine working for a company with a tight culture perceived
signicantly lower personal control (M=4.43, 95% CI [4.02,
4.85], SD =1.41) than those asked to imagine working for a
company with a loose culture (M=5.83, 95% CI [5.58, 6.08],
SD =.90), t(95) =5.87, p<.001, d=1.19.
Consistent with our hypothesis, Study 6a provided causal evi-
dence for the idea that tight cultures lower peoples sense of personal
control.
Study 6b: TightnessLooseness Collective Control
Finally, Study 6b tested the last link in the cycle of mutual
constitution: whether tighter cultures would increase peoples per-
ceptions of collective control (Hypothesis 6).
Method
The hypotheses, power analysis, method, and preselection criteria
for this study were preregistered at https://osf.io/q7ncb.
Participants
A previous study employing a similar design yielded an effect
size of d=.98 for a slightly different dependent measure (perceived
predictability of social interactions). A power analysis with Cohens
d=.98, a=.05 (two-tailed), and power =95% indicated that we
needed to recruit 36 participants. However, we sought to recruit 100
participants to have at least 50 participants per condition. We posted
a survey seeking 100 U.S. residents on Mechanical Turk and
received 100 responses. Per our preregistered exclusion criteria,
we excluded four participants with duplicate IP addresses/geoloca-
tions. The nal sample thus included 96 participants (49 women,
46 men, 1 unreported, M
age
=43.39, SD =10.66).
Procedure
As in Study 6a, participants were randomly assigned to read about
a company that either had a tight culture or a loose culture (see
Supplemental Materials). After they read the description, partici-
pants were asked to summarize the companys beliefs and to provide
one or two examples from their lives that were consistent with the
companys beliefs. Next, as a manipulation check, participants
indicated the extent to which they shared the companys belief
that it is important (vs. not important) to have many social rules and
norms in the workplace (1 =not at all, 100 =a great deal). We
recoded participantsresponses such that higher numbers indicated
greater perceived importance of social rules and norms in both
conditions.
Thereafter, participants were asked to imagine that they accepted
a job in the company, and were asked to complete a three-item
measure of collective control (e.g., I would feel that employees in
the company can work together to control the fate of the company,
α=.94, adapted from Ding & Savani, 2020). All items were
measured on 7-point scales (1 =strongly disagree,7=
strongly agree).
Results
No participants provided gibberish responses to the open-ended
prompts used in the manipulation. The manipulation check was
successfulparticipants in the culturally loose condition believed
that social rules and norms were less important (M=46.00, 95% CI
[37.32, 54.68], SD =31.20) than those in the culturally tight
condition (M=74.77, 95% CI [67.17, 82.36], SD =24.67),
t(93) =4.91, p<.001, d=1.02.
Consistent with Hypothesis 6, participants asked to imagine
working for the tight company perceived signicantly higher col-
lective control (M=5.84, 95% CI [5.61, 6.08], SD =.78) than those
asked to imagine working for a loose company (M=4.99, 95% CI
[4.64, 5.35], SD =1.29), t(94) =3.80, p<.001, d=.878.
Consistent with our hypothesis, Study 6b provided causal evi-
dence for the idea that tight cultures lower peoples feelings of
personal control but increase their sense of collective control.
General Discussion
Six studies documented the cycle of mutual constitution between
individualssense of control and need for structure and cultural
tightnesslooseness. American adults experiencing low personal
control preferred to live in tighter states (Study 1). Chinese employ-
ees perceiving a low sense of control had a higher need for structure,
and thus preferred a tighter organizational culture (Study 2). Sin-
gaporean students led to experience low personal control were more
willing to uphold the norm of cooperation in an economic game at a
personal cost (Study 3), and American adults led to experience low
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
12 MA, SAVANI, LIU, TAI, AND KAY
personal control had a higher need for structure (Study 4). Addi-
tionally, American adults who considered the benets of high
structure preferred a tighter culture (Study 5). Finally, when asked
to imagine themselves working in a tight culture, American adults
had a lower sense of personal control (Study 6a) but a higher sense
of collective control (Study 6b). Overall, the ndings illustrate the
bidirectional relationship between tight cultures and personal
control.
Theoretical Implications
The present research contributes to the eld of cultural psychol-
ogy by demonstrating the mutual constitution of psyche and culture.
An important tenet of cultural psychology is that cultural patterns
and individual psychological states mutually shape each other
(Fiske et al., 1998;Markus & Kitayama, 1994). However, limited
research has documented the bidirectional relationship between the
culture and psyche within the same article (although see Kashima,
2000;Kitayama et al., 1997;Morling & Lamoreaux, 2008;Salter et
al., 2018, for exceptions). By examining how the desire for personal
control, a fundamental human motivation, inuences preference for
tight cultures (and vice versa), our research contributes to the theory
of mutual constitution by documenting a motivational individual-
level pathway that could lead to cultural change.
The present research also represents an important theoretical
advancement to compensatory control theory. By focusing on cultural
tightness as a means of compensatory control, it makes clear a path in
which control-motivated structure-seeking might be mutuallyreinfor-
cing, a possibility that has so far not been considered before (Landau
et al., 2015). That is, experiencing low personal control could lead to
greater preference for the structure prevalent in tight cultures, and
possessing structure can then reduce individual personal control.
Further, by showing that the desire for control can also be satised
by the increased collective control afforded by tight cultures, we offer
a path in which the gradual decrease in personal control in tight
cultures can unfold in an adaptive way. Through these mutually
reinforcing relationships over time,people in tighter cultures lean less
on personal agency and more on collective control as a means of
perceiving an orderly and controllable world. Thus these pathways
might help explain cultural differences in the relative emphasis on
personal and collective control (Morling & Evered, 2006).
Our work also contributes to cultural tightnesslooseness
research by providing evidence for key assumptions of the theory.
A core prediction of tightnesslooseness theory is that psychologi-
cal affordances, such as peoples feelings of personal control,
reect and supporttightnesslooseness (Gelfand et al., 2011,
p. 1101). In other words, tightnesslooseness theory predicts that
personal control and tightnesslooseness possess a causal bidirec-
tional relationship. However, due to their correlational nature,
research that has tested the relation between personal control and
tightnesslooseness is unable to reveal the causality between these
two constructs. For example, personal control was found to be
correlated with tightness at the state level in the United States
(Harrington & Gelfand, 2014;Uz, 2015). By providing causal
evidence for a bottom-up pathway in which psychological states
(i.e., peoples personal control) inuence preference for tight cul-
tures, and a top-down pathway in which being in tight cultures
inuences peoples personal control, we extend existing work on
tightnesslooseness.
In addition, scholars have argued that tight cultures evolved as a
way for people to collectively mitigate societal threats (Gelfand et
al., 2011;Pelto, 1968;Triandis, 1989). We support this idea by
showing that being in a tight culture increases peoples perceptions
of collective control, which makes them feel more condent in
overcoming external threats as a group. The increased collective
control afforded by tight cultures might be especially important in
the current COVID-19 pandemic, in which a coordinated collective
response is vital for survival. Indeed, in support of the idea that
tightness can help cultural groups respond to collective threats,
Gelfand et al. (2021) recently found fewer COVID-19-related deaths
in tighter cultures.
Strengths, Limitations, and Future Directions
Notable strengths of our work include a test of our hypotheses in
culturally diverse contexts that also vary in tightnesslooseness,
including America, China, and Singapore. We also provide con-
verging tests of our hypotheses using multiple methods, including
experiments as well as longitudinal designs. Despite these strengths,
our research also has limitations which we discuss below.
One reason there has been limited research examining the mutual
constitution of the psyche and culture is that the process itself
unfolds over a long interval of time, rendering it difcult to capture
empirically. Although we have obtained support for some aspects of
this process using different manipulations of control while employ-
ing eld, experimental, and archival study designs, our studies are
limited in that they examine these processes in a static and piecemeal
fashion, which may risk oversimplifying the dynamic and interwo-
ven nature of the mutual constitution process.
Further, when demonstrating the effect of personal control on
cultural tightness, we have largely assessed peoples preference for
tighter cultures or their tendency to punish/reward others to uphold
norms, not actual changes in cultural tightness. Indeed, an important
assumption of our cyclical model is that preference for tighter
organizations would lead to objective changes in cultural tightness,
which would then reduce personal control but bolster collective
control, and so forth. One way of addressing this limitation could be
to conduct eld experiments in which some groups could be
randomly assigned to have high personal control and others to
have low personal control. Researchers can then track differences in
preferences for tightness and actual tightness of these groups, as well
as psychological and behavioral indicators of need for structure,
personal control, and collective control over time.
Although we sought to address the role of power as an alternative
explanation in Study 3 by showing that the effect of perceived
control held after controlling for power as a covariate, our measure-
ment of power in Study 3 may not capture all the ways in which
power is manifested in everyday life. Indeed, the measure of power
that we used (Anderson et al., 2012) taps peoples subjective sense
of power, as opposed to objective power that stems from structural
factors such as socioeconomic status, occupation, income, gender,
ethnicity, and so on. It is possible, for instance, that people with a
lower chronic sense of power (e.g., women) may prefer stronger
norms (e.g., against sexual harassment). Those with higher levels of
power may prefer looser cultures as they would prefer not to be
shackled by rules and norms (Fiske, 1993). Future research can
investigate the relationship between structural power and tightness.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
PERSONAL CONTROL AND CULTURAL TIGHTNESSLOOSENESS 13
Finally, cultural characteristics are multiply determined, and
various forces may countervail and even amplify the effect of
personal control on cultural tightness and the effects of tightness
on personal agency and collective control (Cohen, 2001). Although
we documented that the mutual constitution process generalizes
across countries varying in cultural tightness, other dimensions of
culture are likely to shape the bidirectional relationship between
personal control and tight cultures. For example, individualism
collectivism might inuence the degree to which people prefer tight
cultures when they feel that they lack personal control (Kay &
Sullivan, 2013;Ma et al., 2019). Specically, whereas people in
individualistic cultures are more likely to cope with control threats
by directly bolstering their sense of personal control, those in
collectivistic cultures may opt to regain their sense of control
through compensatory control processes (e.g., preferring tighter
cultures). Further, individualismcollectivism might also shape
the degree to which being in a tight culture inuences peoples
feelings of personal and collective control. Specically, collectivism
might strengthen the effect of tight cultures on peoples sense of
personal and collective control because people in these cultures are
more responsive to social norms (Savani et al., 2015). Future
research can examine how other cultural values can strengthen or
weaken the observed effects.
Tightnesslooseness is a signicant dimension along which
cultures vary (Gelfand et al., 2011;Triandis, 1972). By documenting
a bidirectional relationship between personal control and cultural
tightness, the present research documents that culture and psyche
can indeed make each other up(Shweder, 1991) and that the basic
human desire for control can have profound societal implications.
References
Adams, G., & Markus, H. R. (2001). Culture as patterns: An alternative
approach to the problem of reication. Culture and Psychology,7(3), 283
296. https://doi.org/10.1177/1354067X0173002
Adams, G., & Markus, H. R. (2004). Toward a conception of culture suitable
for a social psychology of culture.In M. Schaller & C. S.Crandall (Eds.), The
psychological foundations of culture (pp. 335360). Lawrence Erlbaum.
Aktas, M.,Gelfand, M. J., & Hanges, P. J. (2016). Cultural tightnesslooseness
and perceptions of effective leadership. Journal of Cross-Cultural Psychol-
ogy,47(2), 294309. https://doi.org/10.1177/0022022115606802
Anderson, C., John, O. P., & Keltner, D. (2012). The personal sense of
power. Journal of Personality,80(2), 313344. https://doi.org/10.1111/j
.1467-6494.2011.00734.x
Bandura, A. (1997). Self-efcacy: The exercise of control. Freeman.
Bandura, A. (2000). Exercise of human agency through collective efcacy.
Current Directions in Psychological Science,9(3), 7578. https://doi.org/
10.1111/1467-8721.00064
Bourdieu, P. (1977). Outline of a theory of practice. Cambridge University
Press. https://doi.org/10.1017/CBO9780511812507
Brim, O. G., Ryff, C. D., & Kessler, R. C. (2004). The MIDUS national
survey: An overview. In O. G. Brim, C. D. Ryff, & R. C. Kessler (Eds.),
How healthy are we?: A national study of well-being at midlife (pp. 134).
The John D. and Catherine T. MacArthur foundation series on mental
health and development. Studies on successful midlife development. The
University of Chicago Press.
Bureau of Economic Analysis. (2006). Western states led economic growth
in 2005.https://apps.bea.gov/newsreleases/regional/gdp_state/2006/pdf/
gsp0606.pdf
Bureau of Economic Analysis. (2014). Broad growth across states in 2014.
https://apps.bea.gov/newsreleases/regional/gdp_state/2015/pdf/gsp0615.pdf
Caluori, N., Jackson, J. C., Gray, K., & Gelfand, M. (2020). Conict changes
how people view God. Psychological Science,31(3), 280292. https://
doi.org/10.1177/0956797619895286
Chua, R. Y., Roth, Y., & Lemoine, J. F. (2015). The impact of culture on
creativity: How cultural tightness and cultural distance affect global
innovation crowdsourcing work. Administrative Science Quarterly,
60(2), 189227. https://doi.org/10.1177/0001839214563595
Cohen, D. (2001). Cultural variation: Considerations and implications. Psy-
chological Bulletin,127(4), 451471. https://doi.org/10.1037/0033-2909
.127.4.451
Cohen, D., & Kitayama, S. (2007). Cultural psychology: This stanza and the
next. In S. Kitayama & D. Cohen (Eds.), Handbook of cultural psychology
(pp. 847851). Guilford Press.
Ding, Y., & Savani, K. (2020). From variability to vulnerability: People
exposed to greater variability judge wrongdoers more harshly. Journal of
Personality and Social Psychology,118(6), 11011117. https://doi.org/10
.1037/pspa0000185
Ettner, S. L., & Grzywacz, J. G. (2001). Workersperceptions of how jobs affect
health: A social ecological perspective. Journal of Occupational Health
Psychology,6(2), 101113. https://doi.org/10.1037/1076-8998.6.2.101
Fehr, E., & Gachter, S. (2000). Cooperation and punishment in public goods
experiments. The American Economic Review,90(4), 980994. https://
doi.org/10.1257/aer.90.4.980
Fiske, A., Kitayama, S., Markus, H. R., & Nisbett, R. E. (1998). The cultural
matrix of social psychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey
(Eds.), The handbook of social psychology (pp. 915981). McGraw-Hill.
Fiske, S. T. (1993). Controlling other people: The impact of power on