Taking Stock of Self-Control A Meta-Analysis of How Trait Self-Control Relates to a Wide Range of Behaviors

Abstract

Given assertions of the theoretical, empirical, and practical importance of self-control, this meta-analytic study sought to review evidence concerning the relationship between dispositional self-control and behavior. The authors provide a brief overview over prominent theories of self-control, identifying implicit assumptions surrounding the effects of self-control that warrant empirical testing. They report the results of a meta-analysis of 102 studies (total N = 32,648) investigating the behavioral effects of self-control using the Self-Control Scale, the Barratt Impulsiveness Scale, and the Low Self-Control Scale. A small to medium positive effect of self-control on behavior was found for the three scales. Only the Self-Control Scale allowed for a fine-grained analysis of conceptual moderators of the self-control behavior relation. Specifically, self-control (measured by the Self-Control Scale) related similarly to the performance of desired behaviors and the inhibition of undesired behaviors, but its effects varied dramatically across life domains (e.g., achievement, adjustment). In addition, the associations between self-control and behavior were significantly stronger for automatic (as compared to controlled) behavior and for imagined (as compared to actual) behavior.

Full text

http://psr.sagepub.com/
Review
Personality and Social Psychology
http://psr.sagepub.com/content/16/1/76
The online version of this article can be found at:
DOI: 10.1177/1088868311418749
2012 16: 76 originally published online 30 August 2011Pers Soc Psychol Rev
Denise T. D. de Ridder, Gerty Lensvelt-Mulders, Catrin Finkenauer, F. Marijn Stok and Roy F. Baumeister
Behaviors
Taking Stock of Self-Control : A Meta-Analysis of How Trait Self-Control Relates to a Wide Range of
Published by:
http://www.sagepublications.com
On behalf of:
Society for Personality and Social Psychology
can be found at:Personality and Social Psychology ReviewAdditional services and information for
http://psr.sagepub.com/cgi/alertsEmail Alerts:
http://psr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://psr.sagepub.com/content/16/1/76.refs.htmlCitations:
What is This?
- Aug 30, 2011 OnlineFirst Version of Record
- Dec 21, 2011Version of Record >>
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

Personality and Social Psychology Review
16(1) 76 –99
© 2012 by the Society for Personality
and Social Psychology, Inc.
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1088868311418749
http://pspr.sagepub.com
Self-control is related to a wide range of behaviors. Empirical
research shows that people with high self-control are better
able to control their thoughts, regulate their emotions, and
inhibit their impulses than people with low self-control
(Baumeister, Bratslavsky, Muraven, & Tice, 1998). They enjoy
greater psychological well-being, more academic success, and
better interpersonal relations (W. Mischel, Shoda, & Peake,
1988; Shoda, Mischel, & Peake, 1990; Tangney, Baumeister,
& Boone, 2004). High self-control is relevant to nearly all
forms of behavior conducive to a successful and healthy life.
Conversely, low self-control is assumed to be at the heart of
many societal problems, including obesity, substance
abuse, criminality, impulsive buying, and procrastination
(Baumeister & Heatherton, 1996; Gottfredson & Hirschi,
1990; Patton, Stanford, & Barratt, 1995; Vohs & Faber, 2007).
In view of its beneficial effects for human functioning, self-
control is considered a hallmark of adaptation (W. Mischel,
Cantor, & Feldman, 1996; Rothbaum, Weisz, & Snyder, 1982;
Vohs & Baumeister, 2004) and has become a prominent con-
cept in different areas of research in psychology and other dis-
ciplines, including social psychology, clinical psychology,
developmental psychology, health psychology, criminology,
sociology, and medical sciences.
Given the frequent assertions of the theoretical, empirical,
and practical importance of self-control, the present inves-
tigation undertook to review the evidence concerning the
behavioral concomitants of trait self-control. We sought to learn
whether trait self-control has been shown to be reliably related
to behavior and, if so, how large these effects are. We tested a
series of hypotheses about possible moderators of the rela-
tionship between self-control and behavior, such as whether it
is more strongly related to inhibiting unwanted behaviors or
promoting desired ones, and whether it is more relevant for
habitual, automatic behaviors or for controlled actions.
418749PSRXXX10.1177/1088868311418749de
Ridder et al.Personality and Social Psychology Review
1Utrecht University, Utrecht, Netherlands
2VU University Amsterdam, Amsterdam, Netherlands
3Florida State University, Tallahassee, FL, USA
Corresponding Author:
Denise T. D. de Ridder, Utrecht University, Department of Clinical &
Health Psychology, PO Box 80140, 3508 TC Utrecht, Netherlands
Email: D.T.D.deRidder@uu.nl
Taking Stock of Self-Control:
A Meta-Analysis of How Trait
Self-Control Relates to a Wide
Range of Behaviors
Denise T. D. de Ridder1, Gerty Lensvelt-Mulders1,
Catrin Finkenauer2, F. Marijn Stok1, and Roy F. Baumeister3
Abstract
Given assertions of the theoretical, empirical, and practical importance of self-control, this meta-analytic study sought to review
evidence concerning the relationship between dispositional self-control and behavior. The authors provide a brief overview
over prominent theories of self-control, identifying implicit assumptions surrounding the effects of self-control that warrant
empirical testing. They report the results of a meta-analysis of 102 studies (total N = 32,648) investigating the behavioral effects
of self-control using the Self-Control Scale, the Barratt Impulsiveness Scale, and the Low Self-Control Scale. A small to medium
positive effect of self-control on behavior was found for the three scales. Only the Self-Control Scale allowed for a fine-grained
analysis of conceptual moderators of the self-control behavior relation. Specifically, self-control (measured by the Self-Control
Scale) related similarly to the performance of desired behaviors and the inhibition of undesired behaviors, but its effects varied
dramatically across life domains (e.g., achievement, adjustment). In addition, the associations between self-control and behavior
were significantly stronger for automatic (as compared to controlled) behavior and for imagined (as compared to actual)
behavior.
Keywords
self-control, impulsiveness, self-regulation, adaptive behavior, meta-analysis
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 77
The present article is organized as follows. First, it defines
self-control and provides a brief overview of the most promi-
nent theories on self-control, identifying implicit assumptions
surrounding the effects of self-control that warrant empirical
testing. Second, it reports the results of a meta-analysis on
studies investigating the behavioral correlates of trait self-
control as measured by the Self-Control Scale (Tangney et al.,
2004), the Barratt Impulsiveness Scale (Patton et al., 1995),
and the Low Self-Control Scale (Grasmick, Tittle, Bursik, &
Arneklev, 1993). It includes all published and unpublished
studies since 2004. Third, based on the results of the meta-
analysis, it evaluates the three scales and what the meta-
analytic results have to say about trait self-control and
self-control theory.
What Is Self-Control? Although there is considerable dissent in
the literature over how to name, define, and measure the con-
struct of self-control (Duckworth & Kern, 2011), existing
theories generally agree that self-control can be defined as
the capacity to alter or override dominant response tenden-
cies and to regulate behavior, thoughts, and emotions (Ban-
dura, 1989; Carver & Scheier, 1981, 1982; Metcalfe &
Mischel, 1999; Rothbaum et al., 1982; Vohs & Baumeister,
2004). Because self-control includes the successful regula-
tion of impulses, researchers often equate low trait self-con-
trol with trait impulsiveness, though in principle impulse
strength and self-control or restraint contribute indepen-
dently to whether a behavior is enacted (Duckworth & Kern,
2011; Duckworth & Seligman, 2005; Tangney et al., 2004).
In addition, researchers agree that self-control focuses on the
efforts people exert to stimulate desirable responses and
inhibit undesirable responses and that self-control thereby con-
stitutes an important prerequisite for self-regulation (Bau-
meister, Heatherton, & Tice, 1994; Carver & Scheier, 1998;
Muraven & Baumeister, 2000; Tangney et al., 2004).
Research distinguishes between state self-control and
dispositional self-control (Tangney et al., 2004). State self-
control varies across situations and time. Ample empirical
evidence confirms that people’s capacity to exert self-control
is susceptible to situational influences, including previous
attempts at self-control (Baumeister et al., 1998; Muraven &
Baumeister, 2000), mood (Fishbach & Labroo, 2007; Tice,
Baumeister, Shmueli, & Muraven, 2007), working memory
capacity (Hofmann, Gschwendner, Friese, Wiers, & Schmitt,
2008; Schmeichel, 2007), and motivation (Muraven, 2007).
Dispositional self-control is assumed to be relatively sta-
ble across situations and over time; people with high self-
control are better than others at controlling their impulses
(Gottfredson & Hirschi, 1990; W. Mischel et al., 1996;
Rothbart, Ellis, Rueda, & Posner, 2003). Similarly, as com-
pared to people with low self-control, people with high self-
control report less substance abuse, psychopathology, eating
disorders, physical and verbal aggression (Tangney et al.,
2004), show greater inhibition of a negative emotional
response (Kieras, Tobin, Graziano, & Rothbart, 2005), and
make greater accommodations in close relationships (Finkel
& Campbell, 2001). Conversely, children with low self-con-
trol, as indicated by poor performance on a delay of gratifi-
cation measure, had poorer academic performance 10 years
later than those with high self-control (W. Mischel et al.,
1988). Adolescents with low self-control engage in more
health risk behaviors, such as increased use of alcohol,
tobacco, and marijuana as well as increased saturated fat
intake than adolescents with high self-control (Wills et al.,
2001; Wills, Isasi, Mendoza, & Ainette, 2007; Wills, Walker,
Mendoza, & Ainette, 2006). Adults low in self-control
engage more often in deviant behavior, including risky driv-
ing, not wearing seatbelts, using force, and committing fraud
(Pratt & Cullen, 2000; Vazsonyi, Pickering, Junger, &
Hessing, 2001). The present article is focused on the behav-
ioral implications of dispositional self-control.
Theories of Self-Control. In this section we briefly describe the
most prominent theories on self-control and identify implicit
assumptions regarding the effects of self-control that have
remained untested. Moreover, we highlight how the differ-
ent theories converge to suggest that self-control is a quintes-
sential feature of self-regulatory behavior.
The discounting model of impulsiveness (Ainslie, 1975)
considers self-control as the choice of a delayed but more valu-
able outcome over a more immediate outcome that is ultimately
of less value. This perspective on self-control is similar to the
delay of gratification concept (W. Mischel, 1974) and equally
emphasizes the importance of controlling immediate impulses
and responses. Similarly, other approaches in this tradition
highlight that self-control requires one to make decisions
and to act in accordance with long-term rather than short-
term outcomes (Gottfredson & Hirschi, 1990; Logue, 1988;
Rachlin, 2000). Specifically, Gottfredson and Hirschi’s
(1990) self-control theory contends that the ability to exercise
self-control in the face of temptation accounts for individual
differences in criminal and deviant behavior. Individuals with
low self-control are likely to give in to temptations for misbe-
havior because they have trouble anticipating the long-term
costs of their behavior. Individuals with high self-control, on
the contrary, can resist temptation because they recognize that
in the long run misbehavior comes with costs. Self-control in
these models thus concerns decisions in which people sacri-
fice short-term outcomes in favor of long-term interests, deci-
sions in which immediate (and thus more certain) options are
preferred over delayed (and thus more uncertain) outcomes
(i.e., delay discounting; cf. Frederick, Loewenstein, &
O’Donoghue, 2003).
In hot/cool system approaches to self-regulation
(Loewenstein, 1996; Metcalfe & Mischel, 1999; W. Mischel,
Shoda, & Rodriguez, 1989), self-control is typically concep-
tualized as part of the cool-cognitive or reflective system that
guides goal-directed behavior and requires a person’s voli-
tional control or willpower to be effective. The cool system is
seen as having evolved to serve long-term self-regulatory
purposes that, by means of executive functions (e.g., rea-
soned judgments, strategic action plans), are able to override
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

78 Personality and Social Psychology Review 16(1)
prepotent impulses and habits. The cool system operates by
a pragmatic principle (“do it if it makes sense”) and is associ-
ated with high self-control, rational self-interest, and lack of
impulsive decision making. In contrast, the hot system oper-
ates by a feeling principle (“do it if it feels good”) and is
associated with low self-control and the potential for impul-
sive action.
The self-regulatory strength model of self-control (Baumeister
et al., 1994; Baumeister & Heatherton, 1996) theorizes that
exerting self-control to change or alter behavior or emotions
requires effort and some form of energy or willpower. Self-
control is considered a strength (rather than a skill or a cogni-
tive schema). By exerting self-control to resist temptations or
engage in desirable behavior, for example, people deplete a
reservoir of limited resources. When self-regulatory resources
have been expended, a state of ego depletion results and failure
on a subsequent, unrelated task requiring self-control is more
likely (Baumeister et al., 1998; Muraven, Tice, & Baumeister,
1998). Importantly, the model and empirical evidence suggest
that different types of self-control (e.g., temptation resistance,
impulse overcoming, task persistence, emotion regulation,
choice making) tap into a common, limited resource. The
important implication is that exerting self-control temporar-
ily depletes resources necessary for a large variety of self-
regulatory behavior across a variety of behavioral domains,
making subsequent self-control failure more likely.
As becomes evident, all models share our definition of self-
control as the capacity of the self to alter dominant responses
and to regulate behavior, thoughts, and emotions. They gener-
ally assume (a) that self-control helps to promote desirable
behavior and inhibit undesirable behavior, (b) that it is ben-
eficial for a large range of behaviors, (c) that it is a conscious
and effortful form of regulating behavior, and (d) that it
affects actual behavior (rather than imagined behavior). In
light of the abundant research on self-control, these assump-
tions seem robust. Nevertheless, as we show in the following,
many of them have not yet been put to an empirical test.
Self-Control Promotes Desirable Behavior and Inhibits Undesirable
Behavior. Most theories and definitions agree that self-control
facilitates both the inhibition of undesirable behavior and the
promotion of desirable behavior to the same extent (although
some theories deny the existence of a behavioral promotion
system and argue that desired behavior comes naturally once
an individual has successfully inhibited an undesired
response; cf. Norman & Shallice, 1986). Nevertheless, sound
empirical evidence for the assumption that self-control has
similar effects on both is lacking. Most research focuses on
the influence of self-control on either undesirable behavior
(e.g., impaired reasoning; Schmeichel, Vohs, & Baumeister,
2003) or desirable behavior (e.g., academic performance;
Duckworth & Seligman, 2005). Even studies that included
both types of behavior assessed many more measures of
undesirable behavior than desirable behavior (Tangney et al.,
2004). Moreover, researchers often seem to assume that
when self-control affects undesirable behavior (less binge
eating; Tangney et al., 2004), this also implies that it affects
desirable behaviors (e.g., healthy eating), and vice versa.
Although this assumption may be valid, it has not yet been
empirically tested. Importantly, the literature suggests rea-
sons to argue that self-control may have differential effects on
desirable and undesirable behavior.
Research on the positive–negative asymmetry consistently
shows that negative events have stronger effects than positive
events for virtually all dimensions of people’s lives, including
their thoughts, their feelings, their behavior, and their relation-
ships (for a review, see Baumeister, Bratlavsky, Finkenauer, &
Vohs, 2001). For example, people are more distressed by the
loss of a certain amount of money than they are made happy
by finding the same amount of money (Kahneman & Tversky,
1984). Some researchers suggest that for positive events to be
stronger than negative events, they need to outnumber them.
For example, Gottman (1994) proposed that positive and
good interactions between partners must outnumber the nega-
tive and bad ones by at least 5 to 1 for close relationships to
succeed. Thus, many good interactions can override the nega-
tive effects of one bad interaction. Given equal numbers of
positive and negative interactions, however, the effects of
negative ones are generally stronger than those of the positive
ones.
What are the implications of the positive–negative asymme-
try for the effect of self-control on desirable versus undesirable
behavior? Theoretically, the hypothesis can go both ways. On
one hand, one could argue that self-control is less effective for
the inhibition of undesirable behavior than for the promotion
of desirable behavior. If undesirable behavior weighs stron-
ger than desirable behavior, then people should need much
more self-control to inhibit undesirable behavior (e.g., yell-
ing back at one’s partner) than to engage in desirable behav-
ior (e.g., engage in accommodation; Rusbult, Verette,
Whitney, Slovik, & Lipkus, 1991). Conversely, one could
argue that self-control is less effective for the promotion of
desirable behavior than it is for the inhibition of undesirable
behavior. Indeed, if self-control is needed to replace undesir-
able behavior (e.g., yelling back at one’s partner) with desir-
able behavior (engage in accommodation; Finkel & Campbell,
2001), then people should need much more self-control to
approach the desirable behavior because they need to over-
come the pull of the undesirable behavior, which is much
stronger.
These predictions become even more complex when one
considers the great variety of behavior that is affected by self-
control. For example, self-control is assumed to help people
to inhibit an impulse toward a desired outcome (foregoing an
enjoyable evening with friends) in the service of attaining
another desired outcome (a high grade for an exam). In this
example, the undesired behavior is actually a desired out-
come, yet this outcome is in conflict with a delayed, even
more desirable outcome. Taking one more step, some undesir-
able behaviors that at first glance appear to be self-control fail-
ures (e.g., smoking or alcohol consumption) may in fact be acts
of self-control because they are performed in the service of a
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 79
valued long-term goal (e.g., acceptance by significant others;
Rawn & Vohs, 2011). Whether behavior is regarded as desir-
able or undesirable is thus highly influenced by contextual
factors and may even be idiosyncratic as it relates to the
personal goals an individual holds. To avoid confusion with
respect to the ambiguity of desirability in the long versus
short term, we conceptualize desirable behavior as all
behaviors that are associated with people’s goal to meet
their obligations, duties, and responsibilities and adjust to
social norms to live happy, successful, and healthy lives,
including psychosocial adjustment, adequate and appropri-
ate expression of emotions, physical exercise, and academic
success. Undesirable behaviors, on the contrary, are behav-
iors that interfere with this goal, including antisocial and
destructive impulses, absenteeism, overeating, and interper-
sonal conflict.
In short, although theories on self-control generally agree
that self-control is necessary to inhibit undesirable behavior
and stimulate desirable behavior, studies have not directly
compared the influence of self-control on desirable and unde-
sirable behaviors. So the first aim of this meta-analysis is to
examine whether self-control relates differently to desirable
and undesirable behaviors.
Self-Control Is Beneficial for a Large Range of Behaviors. We con-
ceptualize self-control as people’s capacity to override or
change their inner responses, to inhibit undesired behavioral
tendencies, and to facilitate desired behavior tendencies. This
conceptualization suggests that self-control should be rele-
vant to various behavioral domains. In line with this sugges-
tion, Tangney and her colleagues (2004) identified five
behavioral domains for which dispositional self-control
should be particularly relevant: achievement and task perfor-
mance (e.g., grades, SAT scores), impulse control, psychoso-
cial adjustment (e.g., depression, anxiety), interpersonal
functioning (e.g., accommodation, relationship satisfaction),
and moral emotions (e.g., shame, guilt). Consistent with
their predictions, people with high self-control had more
positive outcomes in all five domains than people with low
self-control. Given that self-control has been proposed to
play a crucial role in the control and inhibition of impulses,
research has increasingly investigated the role of self-control
for academic performance (Duckworth & Seligman, 2005),
health-related behaviors (e.g., physical exercise, condom
use, dieting; cf. Kuijer, De Ridder, Ouwehand, Houx, & Van
den Bos, 2008; Wills et al., 2007), and affect regulation (e.g.,
anger control). To capture the broad variety of behavioral
domains covered in the existing literature on self-control
more effectively, we integrated the different behavioral
domains into nine categories, namely (a) school and work
achievement, (b) eating and weight-related behavior, (c) sex-
ual behavior, (d) addictive behavior, (e) interpersonal func-
tioning, (f) affect regulation, (g) well-being and adjustment,
(h) deviant behavior, and (i) planning and decision making.
The second aim of the present meta-analysis is to examine
whether self-control relates similarly to behavior across the
nine domains.
Self-Control Is Effortful and Conscious: Does It Equally Affect
Controlled and Automatic Behavior? As discussed previously,
virtually all theoretical approaches to self-control highlight
the role of willpower and an active self in the exertion of
self-control (Baumeister et al., 1998; W. Mischel et al.,
1996). The prevailing assumption, and the favored hypoth-
esis in this investigation also, is that self-control is relevant
mainly to behaviors that are under conscious control,
whereas behaviors that are performed without conscious
effort (such as habitual behaviors) are resistant if not
immune to self-control. Still, alternative predictions could
be put forward.
It has been suggested that the exertion of self-control may
not necessarily be related only to conscious or effortful
behavioral processes (Alberts, Martijn, Greb, Merkelbach, &
De Vries, 2007; Ferguson, 2008; Fishbach, Friedman, &
Kruglanski, 2003; Fitzsimons & Bargh, 2004). Whether self-
control is exerted in an automatic or controlled fashion is not
an issue we want to debate in this article. Nevertheless, it is
possible that many automatic behavior patterns are potentially
subject to being overridden or altered by self-control and that
self-control might therefore exert its impact mainly by its
influence on such automatic responses. Research on the reg-
ulatory strength model generally assumes that behaviors
that are more effortful also consume more self-regulatory
resources (self-control) than automatic behaviors, such as
habits (Baumeister et al., 1994). For that reason, as Baumeister
and Alquist (2009) point out, people who are high on self-
control are probably good at automatizing behavior.
To illustrate, when first starting to exercise, Mary may
need to exert a great deal of self-control to do her five miles
of running after a long day at work and taking care of the
children and the household chores. After a couple of weeks
and continued exertion of self-control, the exercise becomes
part of her daily routine, and Mary may need to exert less
self-control to do her running at the end of the day. In this
case, Mary’s exercise routine becomes so engrained in her
daily schedule that she does it almost automatically. Thus
over time, Mary needs to exert less self-control to maintain
her exercising behavior, although self-control may still be
active to monitor her efforts and ensure that Mary continues
to behave in ways that help her to attain her goals (Carver &
Scheier, 1998). In a sense, the main value of self-control may
lie more in creating the healthy habit than in regulating
behavior each day anew.
When self-control operates in such a way that it eventually
does not consume resources, such as when the behavior
becomes habitual (Baumeister & Alquist, 2009), it may simi-
larly affect responses that are automatic as it affects behaviors
that are regulated by conscious control. Evidence examining
whether dispositional self-control affects controlled and auto-
matic behavior in the same fashion is lacking, however. The
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

80 Personality and Social Psychology Review 16(1)
third aim of this meta-analysis therefore is to examine whether
the effects of self-control differ for effortful and automatic
behaviors.
Is Self-Control Related to Actual Behavior, or Do People With
High Self-Control Merely Imagine That They Are Doing Better?
An impressive number of studies have provided convincing
evidence that intended behavior does not necessarily trans-
late into actual behavior (Gollwitzer, 1990; Gollwitzer &
Sheeran, 2006; Heckhausen & Gollwitzer, 1987). In a similar
vein, people’s reports about what think they can do (e.g.,
expectations of behavior or behavior-specific self-efficacy)
and what they should do (e.g., subjective norms or attitudes) do
not necessarily reflect what they actually do (Nordgren, Van
der Pligt, & Harreveld, 2010). Therefore, the distinction
between actual behavior and imagined behavior (i.e., behavior
that one intends to do, thinks one can do, or thinks one should
do) is relevant for examining the link between self-control
and behavior. As a fourth aim of this meta-analysis we
investigated whether self-control equally affects actual
behavior and imagined behavior. Imagined behaviors may
be more vulnerable to wishful thinking and may therefore
reflect biased beliefs about one’s capacity for self-control,
resulting in stronger associations between self-control and
behavior.
Assessing Dispositional Self-Control. Self-control is at the heart
of many desirable behavioral responses, whereas its lack is
associated with many undesirable behavioral responses.
Given the important implications of self-control for psycho-
social adjustment and well-being, it is crucial to assess dis-
positional self-control with a reliable and valid scale.
Moreover, researchers, practitioners, and laypeople need to
know whether the scale is able to detect self-control on a
sound and solid basis that is not vulnerable to variations in the
particular sample that is investigated (e.g., age, gender distri-
bution) or methodological variables (e.g., lab study vs. field
study).
A variety of scales have been developed to assess
self-control, including the Self-Control Behavior Inventory
(Fagen, Long, & Stevens, 1975), the Self-Control Schedule
(Rosenbaum, 1980), the Self-Control subscale of the Californ ia
Personality Inventory (Gough, 1987), the Self-Control
Questionnaire (Brandon, Oescher, & Loftin, 1990), the
adapted Kendall-Wilcox Inventory for self-management
(Kendall & Williams, 1982; Wills, Vaccaro, & McNamara,
1994), and the Ego-Undercontrol Scale (Letzring, Block, &
Funder, 2005). In fact, a recent meta-analysis of self-control
measures identified more than 100 self-report questionnaires
on self-control, most of which have been used only spo-
radically (Duckworth & Kern, 2011). Rather than assess-
ing individual differences in self-control across broad
behavioral domains in general populations (Baumeister et al.,
1994), most scales target specific behaviors (e.g., health
behavior; Brandon et al., 1990) in specific populations
(e.g., adolescents—Kendall & Williams, 1982; clinical
samples—Rosenbaum, 1980). Other scales are outdated
and have not been used recently (Fagen et al., 1975; Gough,
1987) or focus on a specific aspect of self-control such as ego
undercontrol (Letzring et al., 2005). In sum, none of these
scales have been used frequently in general populations.
Neither were they developed to examine the impact of self-
control on a wide range of behaviors, including thoughts and
emotions, across different life domains.
The present analysis examined three self-control scales
that have been used relatively frequently in a variety of pop-
ulations and with different types of behavioral outcomes: the
Self-Control Scale (Tangney et al., 2004), the Barratt
Impulsiveness Scale (Patton et al., 1995), and the Low-Self-
Control Scale (Grasmick et al., 1993).1 In line with the defin-
ing features of self-control, the Self-Control Scale (Tangney
et al., 2004) assesses people’s ability to override or change
inner responses (e.g., “I get carried away by my feelings”;
reversed) and to interrupt undesired behavioral tendencies
and refrain from acting on them (e.g., “I am good at resisting
temptations”). In two large studies, Tangney et al. (2004)
demonstrated that the scale has good reliability (Cronbach’s
α = .89) and good test–retest reliability (r = .89 over 3 weeks).
In addition to the 36-item full scale, Tangney and her col-
leagues developed a 13-item brief scale, which showed a
strong correlation (r = .93) with the full scale and good
psychometric properties. Since its publication in 2004, the
scale has been used among different populations (young
adolescents—Finkenauer, Engels, & Baumeister, 2005;
adult romantic partners—Finkel & Campbell, 2001; stu-
dent samples—Gailliot, 2007b).
The Barratt Impulsiveness Scale (Patton et al., 1995)
assesses lack of planning, spontaneous decision making,
and acting without thinking (sample items are “I am more
interested in the present than in the future” and “I do things
without thinking”). Although trait self-control focuses on
overriding an impulse, trait impulsiveness highlights low
self-control. This scale thus seemingly assumes that impul-
siveness and (low) self-control are equivalent constructs
because they represent the two end points of the same
dimension (Duckworth & Kern, 2011; Tangney et al., 2004).
Although there is some debate about the separate dimensions
that constitute impulsiveness (Patton et al., 1995), the Barratt
Impulsiveness Scale is often used as a generic measure of
impulsiveness and is among the most widely used measures
of self-control (Duckworth & Kern, 2011). The 30-item scale
has good reliability (Cronbach’s α > .80) and discriminates
between populations known to be high or low in impulsive-
ness (e.g., substance-abuse patients vs. undergraduates; Patton
et al., 1995).
Another widely used measure is the Low Self-Control
Scale (Grasmick et al., 1993), derived from Gottfredson and
Hirschi’s (1990) self-control theory. As mentioned above, this
theory contends that variation among individuals in their abil-
ity to exercise self-control in the face of temptation accounts
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 81
for individual differences in deviant behavior. The 24-item
Low Self-Control Scale intends to capture six components of
low self-control: impulsivity, preference for simple rather than
complex tasks, risk seeking, preference for physical rather
than cerebral activities, self-centered orientation, and low
tolerance for frustration (sample items are “I often act on the
spur of the moment without stopping to think” and “I lose my
temper pretty easily”). The scale has shown good reliability
(Cronbach’s α > .80) and is often used in studies on deviant
behavior in both student samples and community samples
(Pratt & Cullen, 2000).
The present analysis focused on these three scales as mea-
sures of dispositional self-control. There are two reasons
for doing so. First, compared to other measures, they better
match the most widely accepted conceptualization of the
self-control construct in the literature. Second, because they
have been used relatively frequently in a variety of popula-
tions and with different types of behavioral outcomes, they
allowed us to investigate whether self-control is equally ben-
eficial in different behavioral domains.
In addition to the aims of this meta-analysis already
described, another aim was to explore two types of modera-
tors, study moderators (e.g., study design) and sample charac-
teristics (e.g., gender distribution).
Study characteristics. The first characteristic that warrants
consideration is the study design. As compared to survey stud-
ies, experimental studies may detect stronger associations
between self-control and behavior because they control for
confounding contextual influences (e.g., distractors, noise).
The second characteristic is the publication status of studies.
As compared to published studies, unpublished studies are
likely to have smaller or nonsignificant effects. As a third
characteristic, our analysis considered whether the impact of
self-control on behavior depends on whether that behavior is
self-reported or objectively measured. Self-reported behav-
iors may overestimate the association between self-control
and behavior because of social desirability or memory biases.
Fourth, we considered the time interval between the assess-
ment of self-control and the assessment of the behavioral
outcome. Because this meta-analysis is concerned with self-
control as a dispositional variable, we consider relations
between self-control and behavior to be more robust if such
associations are maintained when a longer time frame is
employed. Finally, and applicable only to the Self-Control
Scale (Tangney et al., 2004), we considered the scale
version (full or brief) as a potential moderator of the
self-control-behavior link.
Sample characteristics. To establish the link between dispo-
sitional self-control and behavior and minimize the influence
of potential confounds, our analysis considered sample
types, age, gender, and country. For all four characteristics
mean-level differences have been found. To illustrate, self-
control may be higher among older than younger people
(H. N. Mischel & Mischel, 1983; Steinberg et al., 2009; Wills
et al., 2006; cf. Roberts, Walton, & Bogg, 2005), and women
have been found to have higher levels of self-control than
men (Gibson, Ward, Wright, Beaver, & Delisi, 2010; McCa be,
Cunnington, & Brooks-Gunn, 2004; Silverman, 2003).
Although these mean differences do not necessarily affect the
relation between self-control and behavior, they may have
implications for the general use and validity of various self-
control scales.
The Present Research
The present research aimed to take stock of the relationship
between dispositional self-control and behavior. It investi-
gated a number of assumptions regarding self-control that
have largely remained untested by empirical studies. To put
the effect of self-control on behavior to a test, we adopted a
broad view of the kinds of behaviors that may be related to
self-control. Specifically, we considered any cognition, emo-
tion, or overt behavior potentially susceptible to the influence
of self-control, regardless of whether the behavior was
assessed in the lab or in survey studies and of whether it was
observed or self-reported. This choice reflects the enormous
variety of behaviors that have been linked to self-control,
ranging from the self-rated likelihood of engaging in sexual
infidelity (Gailliot & Baumeister, 2007) to refraining from
eye blinking (Schmeichel & Zell, 2007) and from consuming
potato chips (Friese & Hofmann, 2009) to the expression of
affect (Zabelina, Robinson, & Anicha, 2007) and music
piracy (Wolfe, Higgins, & Marcum, 2008). Our analysis
excluded only dependent variables that are dispositional or
trait-like characteristics that are by definition invariant
(e.g., personality traits) and some very specific outcomes
(e.g., MRI scans).
To examine the association between self-control and behav-
ior, we report on the three self-control scales separately. Our
initial aim was to directly compare the three scales, but, unfor-
tunately, the types of moderator variables that were included
in studies with each of the three scales differed dramatically
(with most of the conceptual moderators that guide the pres-
ent meta-analysis lacking from studies with the Barratt
Impulsivity Scale and the Low Self-Control Scale), making
it impossible to undertake such a direct comparison. For each
scale, we first quantify the overall impact of self-control on
behavior. Second, we use meta-analysis to examine the four
implicit assumptions we identified in the existing literature,
that is (a) whether self-control promotes desirable behavior
and inhibits undesirable behavior to the same extent, (b)
whether self-control is equally beneficial across behavioral
domains, (c) whether self-control equally affects controlled
and automatic behavior, and (d) whether self-control equally
affects actual and imagined behavior. Because the distinction
between desired and undesired behavior is considered to be
a central element in theoretical models of self-control, we
report all analyses for both types of behavior separately to
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

82 Personality and Social Psychology Review 16(1)
search for differential effects of self-control on both types
of behavior (De Boer, Van Hooft, & Bakker, in press; De
Ridder, De Boer, Lugtig, Bakker, & Van Hooft, 2011).
Third, we examine the influence of study and sample
characteristics.
Method
Selection of Studies. The following methods were used to gen-
erate the sample of studies (cf. Lipsey & Wilson, 2001): (a)
computerized searches of social scientific databases were
performed (Web of Science, PsycINFO, and Dissertation
Abstracts International) for the years 2004–2009 on the
search term self-control (studies had to include the term in
either the title or the abstract), (b) reference lists in each
article were evaluated for inclusion of relevant studies, and
(c) researchers in the field of self-control were contacted
(via networks) and asked for copies of studies that were
unpublished or in press. Two authors performed indepen-
dent searches to increase the odds that all relevant articles
would be retrieved.
Studies were considered eligible for this meta-analysis
when they met the following criteria. First, they had to employ
a version of the Tangney et al. (2004) Self-Control Scale, either
the full 36-item scale or the brief 13-item scale, and adapted
versions were also considered (e.g., Duckworth & Seligman,
2005); the Low Self-Control Scale (Grasmick et al., 1993); or
the Barratt Impulsiveness Scale (Patton et al., 1995). Second,
they had to include a measure of behavior to examine associa-
tions with self-control. We employed a broad definition of
behavior, including overt behaviors, cognitions, and emo-
tions. Our focus on types of behavior was strongly associated
with the behavioral categories employed in previous studies
on the three scales, such as adjustment, interpersonal func-
tioning, and performance (Tangney et al., 2004), deviant and
addictive behavior (Pratt & Cullen, 2000), and planning and
decision making. Third, to be included in the database, studies
had to report sufficient statistical information to enable the
computation of a standardized effect size ρ from correlations,
t values, or F values, accompanied by their standard devia-
tions or variances as well as the number of participants
(Cooper & Hedges, 1994; Lipsey & Wilson, 2001). We con-
tacted authors for additional information if insufficient details
were reported.
Self-Control Scale. The literature search identified 53 studies
that could be potentially included in the review. Of these, 3
were rejected because they did not include a measure of
behavior. The majority of the remaining 50 studies reported
several outcomes. The final database contained 312 tests of the
association between self-control and behavior and a combined
sample of 15,455 respondents (an average sample size of 309
participants per study with a range of 20 to 1,828).
Barratt Impulsiveness Scale. The literature search identified
58 published studies that could be potentially included in the
review. A total of 27 studies were rejected because they
reported insufficient statistical details (n = 17),2 employed a
dependent measure that was not relevant for the present
meta-analysis (n = 7), or had a within-subjects design (n = 3),
resulting in a sample of 31 studies that met the inclusion crite-
ria. Most studies reported several outcomes. The final database
included 97 tests of the association between impulsiveness
and behavior and a combined sample of 4,791 respondents
(an average sample size of 154 participants per study ranging
from 14 to 617).
Low Self-Control Scale. We found 26 published studies that
could be potentially included in the review of which 21 met
the inclusion criteria. Five studies were rejected because they
reported insufficient statistical details. Most studies reported
several outcomes, resulting in a database that included 40
tests of the relation between low self-control and behavior.
The combined sample consisted of 14,402 participants (an
average sample size of 591 respondents per study, ranging
from 64 to 2,437).
Data Coding. A detailed coding format was developed (cf.
Lensvelt-Mulders, Hox, Van der Heijden, & Maas, 2005),
comprising information about (a) statistical details required
to compute standardized effect sizes, (b) information about
the study, the sample, and measurement of relevant variables
that was used either to determine study quality or to provide
information about potential moderator effects, and (c) con-
ceptual variables that are of theoretical interest to explain the
relation between self-control and behavior. More specifically,
the following characteristics were coded:
Statistical details included (a) sample size at baseline and,
if applicable, at follow-up and (b) statistical information to
enable the computation of a standardized effect size (e.g., F
value, correlation).
Study characteristics included (c) study design (experimen-
tal vs. survey), (d) publication status (peer-reviewed published
or in-press article, unpublished manuscript, report, or book
chapter), (e) in case of the Self-Control Scale only, the ver-
sion of the self-control scale (full, brief, or adapted version),
(f) measurement of dependent variable, self-reported behav-
ior versus objectively assessed (e.g., food consumption, grades,
performance at lab tasks such as time spent on puzzle solving),
and (g) the time interval between assessment of self-control
and the behavior under study.
Sample characteristics included (h) sample type (student,
community, or clinical), (i) the mean age of the sample,
(j) the gender distribution of the sample (male vs. female),
and (k) the country where the study was conducted.
Conceptual characteristics of the behavioral measure
included (l) whether the behavior involved the inhibition of an
undesired response or the performance of a desired response.
As explained in the introduction, desirable behavior is concep-
tualized as any behavior that contributes to people’s goals to
meet their obligations, duties, and responsibilities and adjust to
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 83
social norms of living happy, successful, and healthy lives.
Typical examples of such behaviors are homework hours,
physical exercise, eating healthy foods, condom use, marital
satisfaction, health motivation, loyalty, and self-disclosure.
Undesirable behaviors, in contrast, are behaviors that interfere
with this goal, including, for example, delinquency, aggres-
sive behavior, health risk taking, worrying, sexual infidelity,
lying, drug use, absenteeism, overeating, and marital conflict.
To illustrate, if a behavior involves eating fatty foods (an
undesired behavior), people could score either low (they
don’t eat fatty foods) or high (they do) on this dimension.
Alternatively, if the behavior involves eating fruits (a desired
behavior) it would be coded low if people do not eat fruits and
high if they do.
(m) Whether the behavior was controlled or automatic
was coded. Controlled behaviors are defined as any behavior
requiring conscious attention or deliberation, for example,
making coping plans, expressing intentions, quitting smok-
ing, and the number of anagrams solved. Automatic behav-
iors are defined as behaviors that are performed efficiently,
unintentionally, without awareness and without conscious
control (Bargh, 1994). Examples derived from the studies
included in this meta-analysis involve addictive behaviors
(smoking and alcohol) and habitual behaviors (e.g., habitual
condom use, habitual snacking). To illustrate, whereas smok-
ing is a habitual or addictive behavior that is performed with-
out conscious attention that would therefore qualify as an
“automatic behavior,” quitting smoking qualifies as a con-
trolled behavior because breaking a bad habit typically
requires conscious effort.
(n) Behavioral domain was coded. Because our aim was not
to design an exhaustive categorization of behavioral domains,
we categorized the measures of behavior that were available
from the studies into nine comprehensive clusters: (1) school
and work performance (e.g., GPA, homework hours, persis-
tence at solving task), (2) eating and weight-related behavior
(e.g., emotional eating, dieting), (3) sexual behavior (e.g., atti-
tudes and subjective norms about condom use, sexual restraint),
(4) addictive behavior (smoking, alcohol use), (5) interpersonal
functioning (e.g., commitment to relationship, loyalty tenden-
cies, perceived parental supportiveness), (6) affect regulation
(e.g., difficulty describing emotions, positive emotion words
used), (7) well-being and adjustment (e.g., self-esteem, happi-
ness, depressed mood), (8) deviant behavior (e.g., cheating,
stealing), and (9) planning and decision making (e.g., Iowa
Gambling Task, Stroop Task, Tower of Hanoi).
(o) Whether the behavior was imagined and involved
thoughts and feelings about a behavior or actual behavior was
coded. Typical examples of imagined behavior are perceived
social norms about behavior, behavioral expectancies, imagin-
ing how one would act in fictitious scenarios, and action plans.
Of course, imagined behaviors do not necessarily translate into
actual behavior that may be assessed independently from what
is going on in a person’s mind. Typical examples of actual,
observable behavior are absence of work, number of hours
in the gym, calories consumed from snacks, errors made in
a Stroop Task, and persistence at solving a task.
The first 15 studies were coded by four independent coders.
The independent codings showed marginal differences that
were resolved by considering the original study. Interrater
agreement was very good, with Cohen’s kappas (categorical
variables) or correlations (continuous variables) ranging from
80% (life domain) to 100% (all other variables). The remainder
of the studies were coded by one of the authors (F.M.S.); when
the information in the research was unclear, the study was
discussed by the four original coders, and disagreements
were jointly resolved.
Analytic Strategy. Most studies reported the correlation
between self-control and behavior as an outcome measure.
We therefore recomputed all other outcome measures into
correlation coefficients, using the transformation procedures
provided by Cooper and Hedges (1994) and Lipsey and
Wilson (2001). Effect sizes were computed in standardized,
sample weighted correlation coefficients ρ.3 For convenience
of interpretation, we report effect sizes in simple rs. Cohen’s
(1992) guidelines for interpreting average effect size values
were used. According to Cohen’s power primer, r = .10 should
be considered a small effect size, r = .30 is a medium effect
size, and r = .50 is a large effect size.
Computations were undertaken using standard meta-
analysis procedures. First, a total absolute effect size |ρ| was
computed for each of the three self-control scales, using
SPSS macros originally developed by Wilson (2000). The
overall effect sizes were significant but showed a significant
variability, which could not be explained by mere sample
variance. Thus, a random effects model was chosen because
not all variance could be explained by the predetermined
moderating factors (Cooper, 1986).
Because the distinction between desired and undesired
behavior is central in most models of self-control, we report
results from moderator analyses for both types of behavior
separately.4 When possible, each potential moderator was
treated as a dichotomous variable and the effect sizes from
each study were coded into one of two levels of the modera-
tor. For example, studies that examined effects of self-control
on controlled behavior were compared with studies that inves-
tigated effects of self-control on automatic behavior. Next,
the effect size (r) and homogeneity statistic (Q) were calcu-
lated separately for the two groups of studies. As the number
of tests (k) varies across studies, the Q statistic cannot be
compared across analyses, so we also calculated the I2 statistic
as a measure of true heterogeneity expressed as a percentage
(J. P. Higgins, Thompson, Deeks, & Altman, 2003), with levels
of 25%, 50%, and 75% representing low, medium, and high
levels of heterogeneity, respectively (J. P. Higgins & Tho mpson,
2002). The d statistic (ρ – ρ/SEpooled) was used to compare the
coefficients. When dichotomization was impossible (i.e., in
case of multiple behavioral domains), separate rs were calcu-
lated for each relevant category to compare effects.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

84 Personality and Social Psychology Review 16(1)
We report results for each self-control scale separately
because the information about moderator variables that could
be derived from the studies differed dramatically from scale to
scale, making a direct comparison of the three scales impossi-
ble. For each scale, we first present descriptive data of the stud-
ies included in the analysis. Second, we report the overall effect
size of self-control on all behaviors and effect sizes for desired
and undesired behaviors separately. In the third section, we
report results from the analyses of sample and study modera-
tors to rule out any systematic biases relating to these charac-
teristics. Finally, we discuss results relating to the conceptual
qualifiers of the self-control–behavior association.
Results
Self-Control Scale
Descriptive Data. Of the studies using the Self-Control Scale,
34 were descriptive and 12 had an experimental design; 4
studies combined descriptive and experimental designs. In
all, 43 studies were cross-sectional, and a minority of 7 had
a prospective design (ranging from 3 to 365 days). Also,
20 studies were published or in-press reports in peer-reviewed
journals; the others were unpublished papers or reports. We
therefore dichotomized this category into published (includ-
ing papers that were in press) versus unpublished papers. In
all, 22 studies were conducted in the United States, 27 were
conducted in Europe, and 1 study reported data on samples
from different countries but included predominantly European
participants. We therefore dichotomized this variable into
European (including the mixed sample) vs. American sam-
ples. The majority of studies focused on student samples (n =
32), 16 focused on community samples, and 2 focused on
clinical samples. We decided to compare student to nonstu-
dent samples. Of the studies, 13% comprised samples that
were predominantly male (i.e., including > 67% males) and
19% comprised samples that were predominantly female
(i.e., including > 67% females); the remainder of studies
examined samples that were about equal in gender distribu-
tion. The mean age of the total sample was 21.8 years, and
67% of the studies comprised samples that included adults
only, whereas 33% pertained to predominantly adolescent
samples. Fewer than a quarter (20%) of the studies employed
the full version of the scale, 61% used the brief version, and the
remainder (19%) used adapted versions. We dichotomized this
variable into full version versus other versions of the scale. To
control for potential dependencies between moderators, we
examined correlations between moderator variables.5 Because
of the large sample size, only correlations greater than .35
(thus accounting for more than 10% shared variance) were
considered, showing that sample type (student samples vs.
other samples) was associated with study design (87% of
students participated in experimental designs whereas 70%
of nonstudents participated in surveys), country of origin
(70% of U.S. samples were students, whereas 27% of other
samples were students), sample age (100% of student
samples were adults, whereas other samples included both
adolescents and adults), and Self-Control Scale version
(77% of studies with student samples employed the full
version of the scale, whereas 23% of studies with other
samples used the full-scale version). This pattern of cor-
relations shows that all associations are inherent to study
characteristics (e.g., experimental designs are most of the
time conducted in student samples). There were no correla-
tions greater than .35 for the conceptual moderator variables.
The mean level of self-control was 3.26 (SD = 0.58),
varying from 2.87 to 4.26 (on a scale ranging from 1 to 5),
with higher scores reflecting more self-control. Brief descrip-
tions of the samples and selected study characteristics are
provided in the appendix (available at http://pspr.sagepub
.com/supplemental).
Overall Effect Size of Self-Control. We began by computing the
overall effect size for the association between the Self-Con-
trol Scale and behavior. The average absolute (with recoded
effects for undesired behavior) effect size |ρ| derived from
these studies was .26 (p < .001), with a 95% confidence inter-
val from .23 to .28, based on 50 studies and a total sample size
of 15,455. This means that self-control measured by the Self-
Control Scale had, on average, a beneficial small to medium
effect on behavior, regardless the type of behavior involved.
The forest plot for all studies, including the mean standard-
ized effect size per study and its confidence interval, showed
that there were no outliers. Neither was there a difference
relating to sample size of the study.
The homogeneity test of the overall effect size was signifi-
cant (Q = 375.95, df = 311, p = .009), indicating that the data
set was heterogeneous and that the observed variation in the
effect sizes derived from the primary studies was much larger
than could be expected from mere sampling error, although
the percentage of between-study variance (in terms of the
I2 index) was quite low (17%). The observed between-study
variance encouraged a search for moderators of the relation
between self-control and behavior.
Study and Sample Moderators of the Association Between Self-
Control and Behavior. We first computed effect sizes of the
association between self-control and desired and undesired
behavior, respectively, but did not find a significant differ-
ence (ESdesired = .21, ESundesired = –.23, Qbetween = .212, df = 1,
p = .65). Because effect sizes of self-control may be differ-
ently affected by the potential moderating variables, we
report on moderator analyses for the performance of desired
behavior and the inhibition of undesired behavior separately
(see Tables 1 and 2, respectively).
Study Moderators. We began by examining moderation by
study characteristics. Five factors were considered: study
design, publication status, version of the Self-Control Scale,
type of behavioral measure, and time interval. With regard to
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 85
Table 1. Moderators of the Self-Control Behavior Relation for Desired Behaviors (as Assessed by the Self-Control Scale)
Moderator Level 1 N k r Q I2Level 2 N k r Q I2δ
Study characteristics
Design#Survey 7,110 23 .26*** 123.25 Experimental 4,637 15 .11** 33.22†39.7% .19
Publication status### Published 3,091 10 .32*** 58.8 Unpublished 8,655 28 .19*** 80.89 .43
Scale version Full scale 2,781 9 .24*** 13.05 Other versions 8,964 29 .22*** 126.12
Self-report vs. observed Self-report 8,346 27 .21*** 119.43 Observed 3,400 11 .21** 42.73
Time interval Cross-sect 10,819 35 .21*** 117.94 Prospective 2,473 8 .21 41.63
Sample characteristics
Sample## Student 7,110 23 .12*** 46.96 Nonstudent 4,637 15 .24*** 103.0 .21
Age### Adolescent 3,709 12 .31** 50.38 Adult 8,036 26 .10*** 52.90 .58
Gender Male 4,018 13 .25*** 64.23 Female 8,964 29 .18*** 86.66
Country United States 6,800 22 .22*** 64.35†† 37.8% Europe 4,946 16 .21*** 97.75
Conceptual factors
Controlled vs.
automatic###
Controlled 10,818 35 .15*** 93.88 Automatic 6,800 22 .36*** 28.64 .22
Imagined vs. actual### Imagined 9,270 30 .26*** 140.53 Actual 6,180 20 .14*** 46.64 .48
N = average N per study (309) × k; k = tests; r = correlation coefficient; Q = heterogeneity; I2 = proportion unexplained variance (Q – df/Q).
*Significant ES: **p < .01. ***p < .001.
#Significant difference between groups: #p < .05. ##p < .01. ###p < .001.
†Significant Q = heterogeneity (per group, after meta ANOVA random model): †p < .05. ††p < .01.
Table 2. Moderators of the Self-Control Behavior Relation for Undesired Behaviors (as Assessed by the Self-Control Scale)
Moderator Level 1 N k r Q I2Level 2 N k r Q I2δ
Study characteristics
Design Survey 9,891 32 -.22*** 321.22††† 65% Experimental 6,800 22 -.14*** 13.63
Publication status### Published 4,018 13 -.27*** 187.35††† 80% Unpublished 6,800 22 -.12*** 118.29†82% .27
Scale version Full scale 2,473 8 -.37*** 18.50 Other versions 7,110 23 -.20*** 304.36††† 70%
Self-report vs. observed Self-report 8,964 29 -.21** 328.54††† 65% Observed 2,473 8 -.16* 8.44
Time interval Cross-sect 9,892 32 -.23*** 322.78††† 68% Prospective 2,473 8 -.14*** 6.83
Sample characteristics
Sample Student 5,564 18 -.24*** 33.41 Nonstudent 49,467 16 -.21*** 303.55††† 75%
Age### Adolescent 3,091 10 -.26*** 200.47††† 75% Adult 5,873 19 -.11*** 94.61†† 81% .20
Gender### Male 3,709 12 -.26*** 183.65††† 72% Female 8,036 26 -.14*** 129.08††† 39% .11
Country United States 3,709 12 -.20*** 16.23 Europe 7,110 23 -.21*** 321.34††† 68%
Conceptual factors
Controlled vs. automatic### Controlled 6,800 22 -.16*** 204.27††† 65% Automatic 5,255 17 -.40*** 66.05†† 76% .23
Imagined vs. actual### Imagined 6,180 20 -.30*** 143.88††† 59% Actual 7,725 25 -.17*** 169.70††† 60% .42
N = average N per study (309) × k; k = number of tests; r = correlation coefficient; Q = heterogeneity; I2 = proportion unexplained variance (Q – df/Q); δ =
Cohen’s δ = ρ – ρ/SEpooled.
*Significant ES: *p < .05. **p < .01. *** p < .001.
#Significant difference between groups: ###p < .001.
†Significant Q = heterogeneity (per group, after meta ANOVA random model): †p < .05, ††p < .01, †††p < .001
study design, more rigorous experimental studies showed a
smaller (but still significant) effect size than survey studies,
but only for desired behavior; a similar nonsignificant trend
was observed for undesired behavior. There was also a differ-
ence with regard to publication status: Associations between
self-control and desired (Table 1) and undesired behavior
(Table 2) were stronger in published than in unpublished
studies. This finding confirms the presence of a publication
bias with smaller effects having a lower chance of being pub-
lished.6 We also examined whether the scale version had an
effect on the association between self-control and behavior.
The full scale resulted in significantly stronger effects in the
case of undesired behaviors, suggesting that the full scale
assesses inhibition of undesired behavior better than other
versions of the scale.7 Observed behaviors (either desired or
undesired) and self-reported behaviors were equally related
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

86 Personality and Social Psychology Review 16(1)
to self-control, thus indicating absence of flawed or overesti-
mated effects in case of self-report.
Finally, with regard to time interval between the assess-
ment of self-control and the behavior under study, cross-
sectional designs measuring self-control and the inhibition of
undesired behavior at the same moment resulted in signifi-
cantly stronger effect sizes than prospective designs with a
longer time interval between the measurement of self-
control and behavior. Such a difference was not observed
in studies examining prospective effects of self-control on the
performance of desired behavior, which was similar to the
overall effect found in cross-sectional studies albeit not sig-
nificant (probably related to the small number of studies).
Taken together, examination of study moderators warrants
some caution in interpreting effects of self-control as mea-
sured by the Self-Control Scale on behavior as studies with
more rigorous designs (experimental and/or longitudinal)
result in smaller effect sizes. In addition, “true” effects of
self-control may be somewhat lower than published studies
suggest because unpublished studies report significantly
lower effect sizes. Finally, when examining the effects of
self-control on undesired behavior, the version of the Self-
Control Scale should be taken into account.
Sample Moderators. Next, we considered four potential mod-
erating factors that related to sample characteristics: sample
type, age, gender distribution of the sample, and country
where the study was conducted. There were significant dif-
ferences with regard to the type of sample that was studied.
Effect sizes in student samples were smaller than those
reported in community samples, but only for desired behav-
iors. This finding suggests that community samples may
experience more benefit from self-control, regardless of
whether their trait self-control scores are high or low. There
was a significant effect of age on the association between
self-control and behavior with stronger effects of self-control
on behavior in younger samples, in case of both desired and
undesired behavior, suggesting that relatively younger sam-
ples experience more benefit from self-control than older
samples. With regard to gender, the effect of self-control
proved equally strong in females and males for the perfor-
mance of desired behavior. For the inhibition of undesired
behavior, the effects of self-control in predominantly female
samples were much smaller than the effects found in males.8
With regard to the country where the study was conducted,
studies of American and European samples showed equally
small to medium effect sizes for self-control, for both the
performance of desired behaviors and the inhibition of unde-
sired behaviors. Taken together, our analysis of sample mod-
erators suggests that samples of people with relatively
stronger impulses (males, adolescents) benefit more from
having higher self-control than other categories of people.
Conceptual Moderators of the Association Between Self-Control
and Behavior
Behavioral domains. We distinguished among nine domains
of behavior (school and work performance, eating and weight
behavior, sexual behavior, addictive behavior, interpersonal
functioning, affect regulation, well-being and adjustment,
deviant behavior, and planning and decision making), but
because of an insufficient number of tests (k < 4), we were
unable to calculate separate effect sizes for the domains of
sexual behavior, addictive behaviors, affect regulation, devi-
ant behavior, and planning and decision making. For the
remaining four categories absolute effect sizes composing
both desired and undesired behaviors (with recoded effects for
undesired behavior) were computed because the relatively low
number of studies addressing each of these behavioral domains
did not allow for a distinction between desired and undesired
behavior. Table 3 shows that the effect sizes of self-control
vary across behavioral domains, ranging from a relatively
small effect size of .17 for eating behavior and weight control
to a medium to strong effect size of .36 for school and work
performance.9 Effect sizes for the impact of self-control on
prosocial behavior (r = .25) and well-being (r = .32) were in
the medium range. For most behavioral domains, effects were
homogeneous with the exception of studies in the domain of
well-being. These findings suggest that the effects of self-
control generalize across life domains but that behavioral
domains that are (partly) regulated by biological regulatory
mechanisms (e.g., eating) may be less susceptible to the
influence of self-control than behavior involving (in part)
external or social regulation (such as school and work).
Controlled versus automatic behavior. There were significant
differences between the effect sizes for controlled versus auto-
matic behavior, in case of both desired and undesired behavior
(see Tables 1 and 2). Although the overall effect sizes for con-
trolled behaviors (both desired and undesired) were small,
those established for automatic behaviors were medium to
strong and in fact comprised the largest effect sizes found
in this meta-analysis. This somewhat unexpected finding
shows that the benefits of self-control are most manifest in
Table 3. Effects of Self-Control in Different Behavioral Domains
(as Assessed by the Self-Control Scale)
N k r SD Q I2
Behavioral domains
School and work 1,546 5 .36*** .048 8.87
Eating and weight 4,328 14 .17*** .029 14.40
Interpersonal
functioning
5,255 17 .25*** .018 75.71
Well-being and
adjustment
4,946 16 .33*** .022 114.22††† 51.8%
N = average N per study (309) × k; k = number of tests; |r| = correlation
coefficient; Q = heterogeneity; I2 = proportion unexplained variance (Q
– df/Q).
*Significant ES: ***p < .001.
†Significant Q = heterogeneity (per group, after meta ANOVA random
model): †††p < .001.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 87
behavior that is performed relatively effortlessly and without
conscious attention or conscious control, suggesting that
people with high self-control are good at automatizing their
behavior (Baumeister & Alquist, 2009) regardless of whether
it relates to doing what they want to do or not doing what they
are supposed to inhibit.
Imagined versus actual behavior. Effect sizes for imagined
behaviors—things people want to do or think they should
do—were significantly larger than those for actual behavior.
Regardless of whether desired or undesired behavior was
involved, thoughts and feelings about behavior were more
strongly associated with self-control (small to medium effects)
than reports of actual behavior (small effects). These differen-
tial effects of self-control for thinking and doing suggest that
high levels of self-control are associated with inflated beliefs
about what one is capable of doing without respect to whether
people can actually enact those imagined behaviors when
action is required.
Heterogeneity in effect sizes for undesired behavior. Overall,
the results reported in Table 1 show that effect sizes for desired
behaviors are homogeneous with few exceptions, indicating
there is no additional variance to explain that cannot be attrib-
uted to the moderator variables that were included in the pres-
ent analysis. This implies that self-control predicts desired
behavior in a relatively straightforward manner that does not
depend on other factors that were not considered. In contrast,
effect sizes for undesired behavior (reported in Table 2) still
show considerable heterogeneity that could not be explained
by the moderator variables. Given that the overall relation-
ship of self-control with the inhibition of undesired behavior
was just as strong as the overall relationship with the promo-
tion of desired behavior, the higher variability in self-control
effects on undesired behavior suggests that other factors that
were not examined may qualify these effects. At this point,
we can only speculate about these other factors. It may be
that the category of undesired behaviors is more heteroge-
neous than we initially thought, comprising behaviors that
are undesired because they violate social norms about what
is appropriate (being violent to significant others, being
absent at work) as well as behaviors that are undesired because
they pose a personal long-term risk (drinking alcohol, eating
fatty foods).
Barratt Impulsiveness Scale
Descriptive Data. Of the 31 studies employing the Barratt
Impulsiveness Scale, 20 were descriptive and 11 had an
experimental design (see the appendix for all descriptive
data). Most studies (n = 29) were cross-sectional, leaving too
few tests (< 4) of prospective designs to include in the mod-
erator analysis. The majority of studies were conducted in the
United States (n = 19), 6 took place in Europe, and the
remainder in a variety of other countries. We recoded this
variable into United States versus other countries. In all,
20 studies focused on student samples, 8 focused on clinical
samples, 2 focused on community samples, and 1 included
a mixed sample. We recoded sample type into student ver-
sus nonstudent samples. The majority of studies (64%) used
samples that were about equal in the distribution of males and
females, whereas about one third studied samples that were
either predominantly male (13%) or female (23%). The mean
age of the total sample was 27.65 years (SD = 8.60), ranging
from 18 to 48 years, thus precluding a comparison of adoles-
cent and adult samples. The mean level of impulsiveness was
63.80 (SD = 5.78; theoretical range = 30–120), with higher
scores reflecting higher levels of impulsiveness. The major-
ity of studies examined the relation between impulsiveness
and undesired behavior (86%); the remaining studies that
examined desired behaviors (e.g., number of advantageous
choices in the Iowa Gambling Task) were recoded. Positive
correlations thus represent a relation between impulsiveness
and undesired behavior. The behavioral domains under study
varied, including planning and decision making (58%; e.g.,
actual or hypothetical reward choice), addictive behavior
(31%; e.g., cocaine abuse, Internet addiction), deviant behav-
ior (6%; e.g., speed deviations), and some other behaviors
(5%; e.g., binge eating, symptoms of psychopathology). All
dependent variables related to controlled behavior, making it
impossible to conduct moderator analyses for controlled ver-
sus automatic behavior. In addition, there were fewer than
four tests of imagined behavior, making it impossible to
compare the effect of self-control on imagined versus actual
behavior. To control for potential dependencies between
moderators, we examined correlations between moderator
variables.10 The sole correlation greater than .35 between
moderator variables related to study design and type of
dependent variable, showing that experimental studies more
often examined observed behaviors as the dependent
variable.
Overall Effect Size of Barratt Impulsiveness Scale. The average
absolute effect size |ρ| was .19 (p < .001), with a 95% confi-
dence interval ranging from .16 to .22, based on 31 studies and
a total sample size of 4,791. This finding shows that low self-
control (impulsiveness) had on average a significant but rela-
tively modest effect on undesired behavior that was slightly
lower than the overall effect size found for studies using the
Self-Control Scale. No significant differences relating to
sample size of the study were found. The homogeneity test
of the overall absolute effect size was not significant (Q =
53.01, df = 90, p = .90), which may be the result of using the
absolute r, which does not express all the variance between
studies. We therefore also computed the homogeneity test of
the simple r, which proved significant (Q = 127.96, df = 90,
p = .005) and larger than the heterogeneity found in the
studies that employed the Self-Control Scale. The percent-
age of between-study variance was low to medium (30%; J.
P. Higgins & Thompson, 2002), which is higher than the I2 of
the studies using the Self-Control Scale. We therefore exam-
ined the potential impact of study and sample moderators.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

88 Personality and Social Psychology Review 16(1)
Table 4 displays the results from these moderator analyses,
showing that effects of impulsiveness on undesired behav-
ior were significantly larger when a survey design (as com-
pared to an experimental design) was employed and when the
dependent variable was self-reported (as compared to
observed). Considerable unexplained variance in the effect of
self-reported behavior remained, however, probably relating
to the wide variety of behaviors that were assessed. There
were no moderator effects of sample type (students vs. non-
students), although there was still considerable unexplained
variance in case of the nonstudent samples, which may
relate to the fact that we combined community samples and
clinical samples and/or to the diversity of clinical samples
that were studied. The moderator analysis for gender
revealed no significant effects. There were larger effects of
impulsiveness on behavior in American samples compared
to non-American samples. We also examined potential dif-
ferential effects of impulsiveness across behavioral
domains and found that effect sizes for addictive and devi-
ant behavior were about the same as the generic effect of
impulsiveness (see Table 6). Ironically, the effects of
impulsiveness were weakest in the domain in which it is
most studied, namely, planning and decision-making tasks.
Low Self-Control Scale
Descriptive Data. All 21 studies using the Grasmick Low Self-
Control Scale employed a descriptive design (see the appen-
dix for all descriptive data). Most studies (n = 18) were
cross-sectional, precluding a comparison of cross-sectional
and prospective designs. The majority of studies were con-
ducted in the United States (n = 19), again precluding a com-
parison of effects across countries. In all, 17 studies focused
on student samples and 4 on community samples. All studies
examined samples that were about equal in male–female
ratio, thus not allowing for a gender comparison of effects
of self-control. The mean age of the total sample was
21.10 years (SD = 5.26). The mean level of self-control was
57.45 (SD = 11.83; theoretical range = 24–96), with higher
scores reflecting lower levels of self-control. Almost all stud-
ies examined the relation between low self-control and unde-
sired behavior (97%). Therefore, the remaining studies that
examined desired behaviors (e.g., positive discipline) were
recoded. Positive correlations thus represent a relation
between low self-control and undesired behavior. The behav-
ioral domains under study varied, including deviant behavior
(42%; e.g., cheating, [non]violent crime, driving above speed
Table 4. Moderators of the Self-Control Behavior Relation for Undesired Behaviors (as Assessed by the Barratt Impulsiveness Scale)
Moderator Level 1 N k r Q I2Level 2 N k r Q I2δ
Study characteristics
Design### Survey 2,664 24 .23*** 34.85 Experimental 7,437 67 .05* 61.54 .81
Self-report vs. observed### Self-report 3,663 33 .20*** 62.77†† 49% Observed 6,438 58 .05* 43.75 .68
Sample characteristics
Sample Student 6,660 60 .14*** 73.75 Nonstudent 3,441 31 .13*** 54.12†† 45%
Gender Male 1,998 18 .10* 1.20 Female 1,776 16 .10** 20.39
Country#United States 6,660 60 .14*** 60.17 Non–United States 1,332 12 .05 15.03 .31
N = average N per study (111) × k; k = number of tests; r = correlation coefficient; Q = heterogeneity; I2 = proportion unexplained variance (Q – df/Q);
δ = Cohen’s δ = ρ – ρ/SEpooled.
*Significant ES: *p < .05. ** p < .01. *** p < .001.
#Significant difference between groups: #p < .05. ###p < .001.
†Significant Q = heterogeneity (per group, after meta ANOVA random model): ††p < .01.
Table 5. Moderators of the Self-Control Behavior Relation for Undesired Behaviors (as Assessed by the Low Self-Control Scale)
Moderator Level 1 N k r Q I2Level 2 N k r Q I2δ
Sample moderators
Sample### Student 15,372 28 .07*** 430.20††† 93% Nonstudent x 12 -.01 135.46††† 92% .62
Age Adolescent 4,392 8 .01 148.12††† 95% Adult x 32 .05*** 428.78††† 92%
Conceptual factors
Imagined vs. actual Imagined 8,235 15 .05* 168.23††† 92% Actual x 25 .04** 412.48††† 94%
N = average N per study (549) × k; k = number of tests; r = correlation coefficient; Q = heterogeneity; I2 = proportion unexplained variance (Q – df/Q);
δ = Cohen’s δ = ρ – ρ/SEpooled.
*Significant ES: *p < .05. **p < .01. ***p < .001.
#Significant difference between groups: ###p < .001.
†Significant Q = heterogeneity (per group, after meta ANOVA random model): †††p < .001.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 89
limit), addictive behavior (30%; e.g., smoking, marijuana use),
and a variety of other behaviors that were too heterogeneous
to be categorized (28%; e.g., unsafe sexual behavior, eating
disorder symptoms). All dependent variables related to con-
trolled behavior, making it impossible to conduct moderator
analyses for controlled versus automatic behavior. We exam-
ined correlations between moderator variables to control for
potential dependencies between moderators, which showed
that student samples were significantly younger than com-
munity samples.11
Overall Effect Size of Low Self-Control Scale. The average abso-
lute effect size |ρ| was .22 (p < .001), with a 95% confidence
interval ranging from .17 to .26, based on 21 studies (40 con-
ditions) and a total sample size of 12,402, indicating that self-
control had on average a significant but modest relationship
with the prevention of undesired behavior. No significant dif-
ferences relating to sample size of the study were obtained.
The homogeneity test of the overall effect size was significant
(Q = 206.14, df = 39, p < .001), indicating that the observed
variation in the effect sizes derived from the primary studies
was larger than could be expected from mere sampling error.
The percentage of between-study variance was extremely
high (81%; J. P. Higgins & Thompson, 2002). Examination of
moderator variables relating to sample characteristics did not
improve the model, as heterogeneity continued to be
extremely high when comparing student and nonstudent sam-
ples or adolescent and adult respondents (see Table 5 for
details). Comparing imagined (e.g., intention to cheat or
steal) to actual behaviors (e.g., actual cheating of stealing)
also did not decrease heterogeneity of variance. Examination
of the two life domains that were studied most with the Low
Self-Control Scale revealed a larger effect size for addictive
behaviors than for deviant behaviors (see Table 6 for details),
but again the variance remained heterogeneous. We there-
fore conclude that none of the moderators included in the
present meta-analysis explains the heterogeneity in vari-
ance of effects of the Low Self-Control Scale on behavior.
Discussion
Many theories have characterized self-control as an impor-
tant capability that contributes to effective functioning,
both of society as a whole and of individuals within it. Our
meta-analysis is a first attempt to integrate the findings
from empirical studies that employ different designs and
different populations. It examined the association of dispo-
sitional self-control with a variety of behavioral outcomes.
In line with the literature arguing that self-control is an
important influence on a broad range of behaviors, our
review showed that dispositional self-control is related to a
wide spectrum of human functioning, including love, hap-
piness, binge eating, alcohol use, getting good grades, com-
mitment in a relationship, occasional speeding, and lifetime
delinquency. Despite this variety, our review found a small
to medium relationship between self-control and such out-
comes, regardless of the scale that was used to assess self-
control. Thus, as many theories have asserted, self-control
is associated with benefits in many spheres of human life.
That said, the Self-Control Scale had stronger relationships
than the Barratt Impulsiveness Scale and the Low Self-Control
Scale to behavior overall, and it also allowed for a more fine-
grained analysis of its effects across different life domains and
different types of behavior. Many of the hypotheses that
guided this meta-analysis could not be tested with results
obtained from studies using the Barratt Impulsiveness Scale or
the Low Self-Control Scale because of missing information on
desired, automatic, or imagined behaviors. Moreover, the
behavioral domains addressed with the Barratt Impulsiveness
Scale (planning and decision making, deviant and addictive
behavior) and the Low Self-Control Scale (deviant and addic-
tive behavior) were different from those studied with the Self-
Control Scale, making a comparison of effects of self-control
obtained with the different scales impossible.
The relatively weaker performance of the Low Self-Control
Scale and the Barratt Impulsiveness Scale may be the result, in
part, of the selection of target variables by researchers who use
those scales. The most commonly studied behavioral domains
that are assessed with the Barratt Impulsiveness Scale (plan-
ning and decision making) and the Low Self-Control Scale
(deviant behavior) produced the lowest effect sizes. Thus the
lack of information on conceptual moderators in studies with
the Barratt Impulsiveness Scale and the Low Self-Control
Scale compromised the possibility of finding convergent results
across scales of self-control.
Table 6. Effects of Self-Control in Different Behavioral Domains
(as Assessed by the Barratt Impulsiveness Scale and the Low Self-
Control Scale)
Barratt Impulsiveness
Scale Nak r SD Q I2
Behavioral domain
Addictive behavior 3,219 29 .23*** .022 33.54
Deviant behavior 666 6 .25*** .046 0.97
Planning and decision
making
6,105 55 .14*** .026 10.47
Low Self-Control Scale Nbk r SD Q I2
Behavioral domain
Addictive behavior 7,605 13 .25*** .018 41.58 71%
Deviant behavior 12,870 22 .15*** .011 139.87 85%
k = number of tests; r = correlation coefficient; Q = heterogeneity; I2 =
proportion unexplained variance (Q – df/Q).
a.N = average N per study (111) × k.
b.N = average N per study (585) × k.
*Significant ES: ***p < .001.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

90 Personality and Social Psychology Review 16(1)
In addition, all analyses with the Low Self-Control Scale
produced exceptionally high levels of unexplained variance
(which were much higher than those using the Self-Control
Scale and the Barratt Impulsiveness Scale), indicating that
other factors unaccounted for in the present meta-analysis
influenced the effects of self-control obtained with this scale.
Research has suggested that better specification of the condi-
tions under which the Low Self-Control Scale is likely to have
more or less effect on deviant behavior should be undertaken
(Tittle, Ward, & Grasmick, 2003), and our research supports
this recommendation. Gender and age effects of self-control
that were revealed by the Self-Control Scale could not be rep-
licated with the Barratt Impulsiveness Scale (gender) or the
Low Self-Control Scale (age), suggesting that the Self-Control
Scale is more sensitive to such differences.
In summary, despite its more recent publication date, the
Self-Control Scale has been used more often to study a broader
variety of behavioral categories than the two other self-control
scales that were included in this meta-analysis. Despite the
relatively large number of unpublished studies reporting lower
effect sizes, studies employing the Self-Control Scale detected
larger and more homogeneous effects of self-control on behav-
ior. In line with the hypotheses that guided our meta-analysis,
studies with the Self-Control Scale show that trait differences
in self-control are significantly more relevant to some behav-
iors than others. Our analysis addressed a number of factors
that may contribute to the explanation of variance in the
strength of the relationship between self-control and behav-
ior, including the usual suspects relating to study and sample
characteristics but also encompassing conceptually important
moderators that have implications for further research and
theorizing about self-control. We turn now to consider the
major findings and their implications based primarily on
studies employing the Self-Control Scale.
Desired Versus Undesired Behaviors. Much theorizing has
emphasized that self-control is aimed more at inhibiting
undesirable behaviors than at promoting desirable behaviors.
(Indeed, even the items on the Self-Control Scale refer more
to avoiding undesirable behaviors than to promoting desir-
able ones.) Therefore, we predicted that self-control effects
would be larger and more consistent with undesired than
desired behaviors. This hypothesis was not supported. The
average effect size estimates for undesired behaviors were
no different from the estimates for desired behaviors. More-
over, the effects of self-control on undesirable behaviors
were significantly heterogeneous (for all three scales), unlike
the effects on desirable behaviors. Thus, the effects of self-
control on undesirable behaviors were less, rather than more,
consistent than effects with desirable behavior.
These findings disconfirm the view of the self-control pro-
cess as a general, all-purpose inhibiting mechanism. To be
sure, it is still possible that self-control developed or evolved
to facilitate the inhibition of some behaviors, and that its uses
for fostering desirable behaviors were a fortunate side effect.
Even so, self-control is apparently more effective at inhibiting
some behaviors than others.
Our review was unable to explain the heterogeneity of
effects on inhibiting bad behaviors, and this remains an impor-
tant question for further research. One likely possibility is
that some behaviors are far more amenable than others to
self-control. Among other theorists, Seligman (1994) has
written extensively about how adjustment depends on ascer-
taining which aspects of oneself can and cannot be changed.
People may strive to change both changeable and relatively
unchangeable undesired behaviors, and so their success would
inevitably be mixed, thereby producing the heterogeneity we
found. In particular, we found relatively small effects with
eating and dieting behaviors, which are seen by many as the
main spheres in which self-control is used. There is a fair
amount of evidence that long-term success in dieting is rare
(e.g., Seligman, 1994), and of course complete abstinence
from eating (unlike smoking, drinking, and unprotected sex)
is impossible. Hence, it is conceivable that some of the vari-
ability in self-control’s links to undesired behaviors arises
from people seeking unsuccessfully to lose weight.
Controlled Versus Automatic Behaviors. If some behaviors are
more easily controlled than others, then the degree to which a
particular behavior is automatic may be one highly relevant
consideration. Hence, we hypothesized that self-control
would be more effective with controlled than with automatic
behaviors. Surprisingly, the analyses clearly indicated the oppo-
site conclusion. Although the association between self-con-
trol and automatic behaviors proved relatively strong,
associations with controlled behaviors were small. The effects
of self-control on automatic behaviors were consistent across
both desired and undesired behaviors and were overall the
largest effect sizes in our entire meta-analysis.
To be sure, it would be nonsensical to conclude that con-
trollable behaviors are not controllable whereas automatic
behaviors are. A more sophisticated interpretation is needed.
We also note that if a behavior were fully and easily control-
lable by everyone in all cases, then the effect of individual
differences in self-control might well be zero—which would
produce a result consistent with what we found. Hence, one
possible explanation for the stronger relationship between
self-control and automatic behaviors is that controllable
behaviors are in general more easily controllable, whereas
changing automatic behaviors is more difficult, so that indi-
vidual differences in self-control have greater relevance with
the latter. But that explanation seems unlikely, not least
because researchers probably would not waste much time
studying the easiest behaviors to control.
In our sample of studies, the behaviors classified as auto-
matic consisted of acts that are normally performed effort-
lessly and without conscious attention, especially habits. The
relatively large relationship between trait self-control and
such behaviors thus suggests that people with good self-
control are especially effective at forming and breaking
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 91
habits. This suggests a change in emphasis for self-control
theory. Although most theorizing about self-control has
focused on the specific act of resisting temptation in a par-
ticular setting, self-control may in general operate more by
forming and breaking habits. It is thus mainly by establish-
ing and maintaining stable patterns of behavior rather than
by performing single acts of self-denial that self-control may
be most effective.
Working, Playing, Eating, Relating. Our findings showed dramat-
ically differential effects of self-control across life domains. The
Barratt Impulsiveness Scale, for example, showed much stron-
ger correlations with addictive and deviant behavior than with
planning and decision making. The Self-Control Scale showed
relatively strong effects on performance at work and school,
whereas the effects on regulating eating and weight were rela-
tively small. The impacts on interpersonal functioning and
adjustment were in between those extremes. This pattern again
turns conventional wisdom on its head, especially insofar as
dieting is probably the single most commonly used source of
examples in writings and talks about self-control.
The idea that self-control differences are largest on work
and school behavior may run counter to some theoretical
assumptions that self-control would be especially relevant for
regulating impulsive behavior (Carver & Scheier, 1998;
Metcalfe & Mischel, 1999; Vohs & Baumeister, 2004), but it
does mesh well with the other results we have reported.
Effective performance at work and school is rarely a matter of
single, prodigious acts of willpower. Instead, it probably
depends on forming and maintaining habits and routines that
foster efficient, steady performance in a regular and disci-
plined manner. Some students may look back on memorable
all-nighters as decisive feats of self-control, but the very need
to study all night may often arise because the person has pro-
crastinated, which can indicate a low self-control and a lack of
regular study habits, and which moreover tends to produce
significantly poorer performance overall than keeping on
schedule and ahead of deadlines (e.g., Tice & Baumeister,
1997). In contrast, eating is partly under control of visceral and
impulsive processes (e.g., Ditto, Pizarro, Epstein, Jacobson, &
MacDonald, 2006; Hofmann, Friese, & Strack, 2009). Weight,
moreover, depends on not only eating but also genetic predispo-
sitions and other factors, and so its amenability to conscious
control may be relatively minimal (e.g., Seligman, 1994).
The effects of self-control within behavioral domains were
generally homogeneous, with one exception: the domain of
well-being and adjustment. The category of adjustment and
well-being included a variety of concepts that have been cat-
egorized in the same way in previous research (Tangney
et al., 2004) and comprised, for example, self-esteem and
absence of depression. However, it is possible that variability
was introduced by multiple factors. Low self-control may
contribute to emotional lability, so that measuring happiness at
different times and in different ways will produce different
results. Low self-control may produce short-term gains but
long-term costs (e.g., W. Mischel, 1974), which again would
contribute to heterogeneity of effects. However, it is also pos-
sible that concepts of adjustment vary in the extent that bio-
logical factors are relevant, which may explain why the
impact of self-control differs across these concepts. In any
case, future research may find it useful to break this category
down into subcategories and examine method variance instead
of making broad generalizations about the contribution of self-
control to well-being and adjustment.
How and When Behavior Is Measured. Several findings indi-
cated that the way in which behavior is measured relative to
self-control has significant implications for the size of
effects. Effect size was also related to publication status, with
unpublished studies having smaller effects than published
ones. This difference could arise if studies with weak, unreli-
able, or confounded measures remain unpublished because
such measurement problems would also reduce the size of
effects of self-control. If published and unpublished studies
are both equally valid, however, then a reliance on published
studies will furnish an inflated estimate of the size of effects
of self-control.
Apart from publication status, multiple aspects of measure-
ment were relevant. First, larger effects of self-control were
obtained when behavior was measured by questionnaire self-
report than by direct observation of actual behavior. Second,
the effects were stronger when the (undesired) behavior was
measured at the same time as the trait self-control measure, as
opposed to measured after delay. Nevertheless, prospective
studies with longitudinal designs that measure self-control ini-
tially and assess delayed behavioral consequences still pro-
duced significant effects. Third, self-control had significantly
stronger relationships to imagined and hypothetical behaviors
than to actual ones. That is, self-control was strongly related to
what people say they would or should do, but the relationships
to what people really do, though still genuine and significant,
are weaker. Of course, it is easier to ask people about self-
control than to actually observe them; asking how much they
would eat or drink or whether they would have sex under cer-
tain circumstances, for example, is certainly more feasible
than measuring what they actually do under those circum-
stances. Yet the present findings clearly illustrate that it may
be important to include measures of actual behavior.
Taken together, this set of findings suggests that the effects
of self-control are subject to dilution in the real world, where
multiple factors come into play. The closer the measure of
behavior was in kind and style to the measure of self-control,
the stronger the effects were. When both trait and behavior are
measured by having the person go straight from one question-
naire to another on the same occasion, results tend to be larger
than if the trait is measured by questionnaire whereas behavior
is measured by direct observation or on another occasion.
Studies that rely purely on questionnaire self-reports to mea-
sure behavior may overestimate the true influence of
self-control.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

92 Personality and Social Psychology Review 16(1)
One additional finding was that studies using the full Self-
Control Scale found larger effects on the control and inhibi-
tion of undesired behaviors than studies using the brief or
adapted version of the scale. Most likely this relates to the
well-established principle of basic measurement theory that
shorter versions of scales tend to be less reliable than longer
versions (Emons, Sijtsma, & Meijer, 2007). Regardless of
the reason, future researchers interested in studying the self-
control of undesired behaviors may find it useful to employ
the full version of the scale.
Sample Moderators: Age and Gender. In general, self-control
scores were fairly similar across different categories of peo-
ple, but effects of self-control varied substantially with gen-
der and age. Regarding gender, the relationship between trait
self-control and undesirable behaviors was greater for
males than for females. Most plausibly, males and females
have similar psychological structures and capabilities for self-
control (as indicated by having similar mean scores on self-
control measures), but males may have stronger antisocial or
problematic impulses than females. For example, men may
be more attracted than women to drug and alcohol abuse,
and they have stronger sexual and aggressive impulses (e.g.,
Baumeister, Catanese, & Vohs, 2001; Eagly, 1987). Hence,
individual variations in strength of self-control will produce
wider variations in behavioral outcomes among males than
among females, insofar as low self-control is more likely to
allow problematic impulses to manifest in behavior.
Although adults scored similar to adolescents on the self-
control measures, the behavioral effects of trait self-control
were larger with the younger samples. The same reasoning may
apply as with gender: Antisocial and problematic impulses
(e.g., sex, aggression, alcohol, drugs) are likely stronger and
more frequent among younger than older people, and so weak
self-control is more likely to lead to problematic behavior
among younger than older people.
Future work should seek to establish separate measures of
impulse strength and self-regulatory capability, though we
recognize that teasing those constructs apart is difficult. Then
it would be possible to test the hypothesis that individual dif-
ferences in self-control strength are more strongly related to
behavior when impulses are strong rather than weak. If that
hypothesis turns out to be false, then another explanation of
the age and gender differences may be needed. For the pres-
ent, however, the evidence seems to fit this conclusion:
Capabilities for self-control are broadly similar in different
sociodemographic groups, but differences in the strength of
undesirable, antisocial impulses produce different behavioral
outcomes and also make individual differences in trait self-
control more powerful predictors of behavior in some groups
than in others.
Limitations and Future Directions. This study is the first system-
atic review of the relationship between dispositional self-con-
trol and a host of behaviors. Moreover, it is the first study that
explicitly introduces a number of dimensions of behavior that
are relevant to understanding the impact of self-control.
Despite these strengths, several limitations should be acknowl-
edged. First, we included only studies that employed a version
of the Self-Control Scale (Tangney et al., 2004), the Barratt
Impulsiveness Scale (Patton et al., 1995), and the Low Self-
Control Scale (Grasmick et al., 1993). These scales are among
the most widely used instruments to assess dispositional self-
control in the way it is typically conceptualized in the litera-
ture. Although it was not our primary aim to compare scales,
we found considerable differences in the way the three scales
have been used to establish connections between trait self-
control and behavior. Unfortunately, only the Self-Control
Scale allowed for a test of our main hypotheses. Therefore, it
remains to be determined whether our results can be replicated
when other scales assessing dispositional self-control are
employed. However, a recent meta-analysis that directly com-
pared a number of self-report trait self-control scales con-
cluded that there was convergence between constructs
(r = .46; Duckworth & Kern, 2011), suggesting that other
measures might generally be assumed to result in similar find-
ings. A second limitation relates to the relatively high number
of unpublished studies with the Self-Control Scale. Unpub-
lished studies do not allow for a full appreciation of study
characteristics, but we chose to include unpublished studies
to avoid potential publication bias. Importantly, all findings
reporting on the conceptual moderators that guided this meta-
analysis were replicated when analyses comprised published
studies only. Another limitation is the relative lack of behav-
ioral domains that could be included in our analyses. Future
research should examine whether self-control produces simi-
lar effects in behavioral domains that were not included in the
present meta-analysis because of a lack of empirical studies,
most prominently sexual behavior, risk behavior, and affec-
tive behaviors. Finally, as our study produced heterogeneous
findings for the relationship between trait self-control and
adjustment and well-being, future studies should employ
more fine-grained analyses of how different components of
adjustment and well-being relate to self-control.
Concluding Remarks
The topic of self-control has attracted extensive theorizing
and empirical study, presumably because of its widespread
potential relevance. The present findings confirm some com-
mon themes of self-control theory but suggest that others need
serious reconsideration. Our results confirm the view that
having high trait self-control is relevant to a rich assortment
of behaviors and outcomes. Furthermore, our findings con-
firm that these effects of self-control are generally beneficial
and adaptive. Self-control is thus one of the most beneficial
traits in personality.
However, contrary to the view that self-control is mainly
aimed at inhibiting undesirable behaviors, we found
that its effects on desirable and undesirable effects were
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 93
approximately equal in size. There was however greater
heterogeneity with the undesirable effects, possibly because
some problem behaviors are far more controllable than oth-
ers. Trait self-control may be most important and most
effective among individuals who grapple with relatively
strong and problematic impulses, such as young males.
Contrary to some assumptions about self-control, our meta-
analysis suggests that the trait differences have their strongest
effects neither in the dieting sphere nor via single feats of will-
power. Rather, some of the strongest effects obtained were in
connection with automatic behaviors, such as forming and
breaking habits. Other strong effects were found in school and
work performance. Possibly, those two large effects overlap
insofar as effective work depends on steady and regular perfor-
mance and good work habits.
In sum, the benefits of self-control appear to justify the
amounts of research and theory that have been devoted to it,
even if that work has yielded some surprises and some changes
in direction are indicated.
Acknowledgments
The authors thank Jannette van Beek for her help with preparing the
data set.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article:
Denise de Ridder and F. Marijn Stok were supported by a grant
from the European Community FP7 Research Program
(Health-F2-2008-223488). Catrin Finkenauer’s contribution was
supported by a grant from the Netherlands Organization for
Scientific Research, No. 452.05.322. Roy Baumeister’s effort was
supported by the National Institutes of Health Roadmap for Medical
Research Common Fund with Grants UL1-DE019586 and
1RL1AA017541.
Notes
1. We also searched for studies using the Rosenbaum Self-Control
Schedule (Rosenbaum, 1980), but there were not enough studies
that met the inclusion criteria (< 10) to be included in the
meta-analysis.
2. The high number of studies reporting insufficient statistical details
was related primarily to reporting on the Barratt Impulsiveness
Scale-13 (BIS-13) for the whole sample including both experi-
mental and control conditions instead of for the specific condi-
tions that were included in the meta-analysis.
3. Our data comprise a hierarchical structure with tests nested
within studies. Such a data structure warrants a meta-analytical
multilevel approach, which not only has the advantage of allow-
ing for the calculation of average effect sizes across studies but
also has the possibility of explaining variance at the study level
(Lensvelt-Mulders, Hox, Van der Heijden, & Maas, 2005; Van
den Noortgate & Onghena, 2003). In the present study we did
not employ this approach, however, because of the variability in
behavioral measures. To do justice to the fact that tests were
nested within studies, and therefore add extra weight to studies
including many tests, we corrected the inverse weight by a factor
equal to the number of tests in a study. We dealt with the hetero-
geneous variability of the studies by employing a random model,
using maximum likelihood estimation.
4. A direct comparison of the three self-control scales could be
tested only for the overall effect on undesired behavior, reveal-
ing similar effect sizes as with analyses per scale. However, it
should be kept in mind that the overall effect sizes are difficult
to compare because they relate to different types of behavior per
scale.
5. Correlations were computed for dichotomized variables. We
also examined associations with nonparametric tests (Mann–
Whitney), but this resulted in similar findings.
6. We reran all analyses (including the overall effect, desired vs.
undesired behavior, and moderator analyses for desired and
undesired behaviors separately) with published studies only and
found that the analyses confirmed the pattern of results obtained
with the analyses on both published an unpublished studies.
7. We also compared the brief and adapted versions of the Self-
Control Scale but found similar effect sizes.
8. See Note 6.
9. As research on eating behavior almost exclusively employs
female samples, we examined whether this differential gender
effect could be explained by studies in the eating domain.
However, the effect was similar when studies on eating were
excluded.
10. See Note 5.
11. See Note 5.
References
References marked with an asterisk indicate references used in
meta-analysis.
*Adriaanse, M. A., De Ridder, D. T. D., & De Wit, J. B. F. (2007).
[Self-control, spontaneous planning and product choice].
Unpublished raw data.
Ainslie, G. W. (1975). Specious reward: A behavioral theory of
impulsiveness and impulse control. Psychological Bulletin, 82,
463-496.
*Alberts, H., Maroofi, S., & Martijn, C. (2005). De relatie tussen
verwachtingen en interpersoonlijke verschillen in zelfcontrole
[The relation between expectations and interpersonal differ-
ences in self-control]. In R. W. Holland, J. Ouwerkerk, C. van
Laar, R. Ruiter, & J. Ham (Eds.), Jaarboek Sociale Psychologie
2005 (pp. 13-21). Groningen, Netherlands: Aspo Pers.
*Alberts, H., & Martijn, C. (2007a). [Self-control and BIS/BAS].
Unpublished raw data.
*Alberts, H., & Martijn, C. (2007b). [Self-control and promotion/
prevention focus]. Unpublished raw data.
Alberts, H., Martijn, C., Greb, J., Merkelbach, H., & De Vries,
N. (2007). Carrying on or giving in: The role of automatic
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

94 Personality and Social Psychology Review 16(1)
processes in overcoming ego depletion. British Journal of
Social Psychology, 46, 383-399.
*Anestis, M. D., Peterson, C. B., Bardone-Cone, A. M., Klein, M. H.,
Mitchell, J. E., Crosby, R. D., & Joiner, T. E. (2009). Affective
lability and impulsivity in a clinical sample of women with buli-
mia nervosa: The role of affect in severely dysregulated behavior.
International Journal of Eating Disorders, 42, 259-266.
*Bailey, C. A., & Ostrov, J. M. (2008). Differentiating forms and
functions of aggression in emerging adults: Associations with
hostile attribution biases and normative beliefs. Journal of Youth
and Adolescence, 37, 713-722.
*Bakhuys Roozeboom, M. C. (2004). De relatie tussen sekse en
State/Action-oriëntatie: Een onderzoek naar de discrepantie
tussen de daadwerkelijke en de ideale State/Action- oriëntatie bij
mannen en vrouwen [The relation between gender and state/
action orientation: A study of the discrepancy between actual
and ideal state/action orientation in males and females] (Unpub-
lished master’s thesis). VU University Amsterdam, Amsterdam,
Netherlands.
*Balodis, I. M., Potenza, M. N., & Olmstead, M. C. (2009). Binge
drinking in undergraduates: Relationships with sex, drinking
behaviors, impulsivity, and the perceived effects of alcohol.
Behavioural Pharmacology, 20, 518-526.
Bandura, A. (1989). Human agency in social cognitive theory.
American Psychologist, 44, 1175-1184.
Bargh, J. A. (1994). The four horsemen of automaticity: Aware-
ness, intention, efficiency, and control in social cognition. In
R. S. Wyer, Jr., & T. K. Srull (Eds.), Handbook of social cogni-
tion (2nd ed., pp. 1-40). Hillsdale, NJ: Lawrence Erlbaum.
Baumeister, R. F., & Alquist, J. L. (2009). Is there a downside to
good self-control? Self and Identity, 8, 115-130.
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D.
(2001). Bad is stronger than good. Review of General Psychol-
ogy, 5, 323-370.
Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M.
(1998). Ego depletion: Is the active self a limited resource?
Journal of Personality and Social Psychology, 74, 1252-1265.
Baumeister, R. F., Catanese, K. R., & Vohs, K. D. (2001). Is there
a gender difference in strength of sex drive? Theoretical views,
conceptual distinctions, and a review of relevant evidence. Per-
sonality and Social Psychology Review, 5, 242-273.
Baumeister, R. F., & Heatherton, T. F. (1996). Self-regulation failure:
An overview. Psychological Inquiry, 7, 1-15.
Baumeister, R. F., Heatherton, T. F., & Tice, D. M. (1994). Losing
control: How and why people fail at self-regulation. San Diego,
CA: Academic Press.
*Bjork, J. M., Hommer, D. W., Grant, S. J., & Danube, C. (2004).
Impulsivity in abstinent alcohol-dependent patients: Relation to
control subjects and type 1–/type 2–like traits. Alcohol, 34, 133-150.
*Brady, S. S. (2006). Lifetime community violence exposure and
health risk behavior among young adults in college. Journal of
Adolescent Health, 39, 610-613.
Brandon, J. E., Oescher, J., & Loftin, J. M. (1990). The self-control
questionnaire: An assessment. Health Values, 14, 3-9.
*Carmichael, S., & Piquero, A. R. (2004). Sanctions, perceived anger,
and criminal offending. Journal of Quantitative Criminology, 20,
371-393.
Carver, C. S., & Scheier, M. F. (1981). Attention and self-regulation:
A control-theory approach to human behavior. New York, NY:
Springer-Verlag.
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful con-
ceptual framework for personality—Social, clinical, and health
psychology. Psychological Bulletin, 92, 111-135.
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of
behavior. New York, NY: Cambridge University Press.
Cohen, J. (1992). A power primer. Psychological Bulletin, 112,
155-159.
Cooper, H. M. (1986). On the social psychology of using research
reviews: The case of desegregation and black achievement.
In R. Feldman (Ed.), The social psychology of education
(pp. 341-365). Cambridge, UK: Cambridge University Press.
Cooper, H., & Hedges, L. V. (1994). The handbook of research syn-
thesis. New York, NY: Russell Sage.
*Dahlen, E. R., Martin, R. C., Ragan, K., & Kuhlman, M. M. (2005).
Driving anger, sensation seeking, impulsiveness, and boredom
proneness in the prediction of unsafe driving. Accident Analysis
and Prevention, 37, 341-348.
*Danner, U. N., Ouwehand, C., Van Haastert, N., Hornsveld, H.,
& De Ridder, D. (in press). Decision making impairments in
women with binge eating disorders compared to obese and nor-
mal weight women. European Eating Disorders Review.
De Boer, B., Van Hooft, E. A. J., & Bakker, A. (in press). Stop and
start control: A distinction within self-control. European Journal
of Personality.
*De Hoog, S. R. (2005). Don’t worry be action!! De invloed van
verschillen in de State/Action-oriëntatie op relatietevredenheid
[Don’t worry be action!! The influence of differences in the
State/Action orientation on relationship satisfaction] (Unpub-
lished master’s thesis). VU University Amsterdam, Amsterdam,
Netherlands.
*De Kemp, R. A. T., Vermulst, A. A., Finkenauer, C., Scholte, R. H. J.,
Overbeek, G., Rommes, E. W. M., & Engels, R. M. C. E. (2009).
Self-control and early adolescent antisocial behavior: A longitudi-
nal analysis. Journal of Early Adolescence, 29, 497-517.
*Demaree, H. A., DeDonno, M. A., Burns, K. J., & Everhart, D. E.
(2008). You bet: How personality differences affect risk-
taking preferences. Personality and Individual Differences, 44,
1484-1494.
*Denissen, J. J. A., Penke, L., & van Aken, M. A. G. (2007). [Per-
sonality and self-control in a German adult sample]. Unpublished
raw data.
De Ridder, D. T. D., De Boer, B. J., Lugtig, P., Bakker, A. B., & Van
Hooft, E. A. J. (2011). Not doing bad things is not equivalent
to doing the right thing: Distinguishing between inhibitory and
initiatory self-control. Personality and Individual Differences,
50, 1006-1011.
*DeWall, C. N., Baumeister, R. F., Gailliot, M. T., & Maner, J. K.
(2008). Depletion makes the heart grow less helpful: Helping
as a function of self-regulatory energy and genetic relatedness.
Personality and Social Psychology Bulletin, 34, 1653-1662.
*DeWall, C. N., Baumeister, R. F., Stillman, T. F., & Gailliot, M.
T. (2007). Violence restrained: Effects of self-regulation and its
depletion on aggression. Journal of Experimental Social Psy-
chology, 43, 62-76.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 95
*De Wit, J. B. F., & Adam, P. C. G. (2007). Sexual risk-taking in gay
men: A focus on non-intentional processes. Utrecht, Netherlands:
Institute for Prevention & Social Research.
Ditto, P. H., Pizarro, D. A., Epstein, E. B., Jacobson, J. A., &
MacDonald, T. K. (2006). Visceral influences on risk-taking
behavior. Journal of Behavioral Decision Making, 19, 99-113.
*Dom, G., De Wilde, B., Hulstijn, W., & Sabbe, B. (2007). Dimen-
sions of impulsive behaviour in abstinent alcoholics. Personal-
ity and Individual Differences, 42, 465-476.
*Dom, G., De Wilde, B., Hulstijn, W., Van den Brink, W., & Sabbe,
B. (2006). Behavioural aspects of impulsivity in alcoholics with
and without a cluster-B personality disorder. Alcohol & Alco-
holism, 41, 412-420.
*Doran, N., McChargue, D., & Cohen, L. (2007). Impulsivity and
the reinforcing value of cigarette smoking. Addictive Behaviors,
32, 90-98.
Duckworth, A. L., & Kern, M. L. (2011). A meta-analysis of the con-
vergent validity of self-control measures. Journal of Research in
Personality, 45, 259-268.
*Duckworth, A. L., & Seligman, M. E. P. (2005). Self-discipline
outdoes IQ in predicting academic performance of adolescents.
Psychological Science, 16, 939-944.
Eagly, A. H. (1987). Sex differences in social behavior: A social-role
interpretation. Hillsdale, NJ: Lawrence Erlbaum.
Emons, W. H. M., Sijtsma, K., & Meijer, R. R. (2007). On the con-
sistency of individual classification using short scales. Psycho-
logical Methods, 12, 105-120.
*Engels, R. C. M. E., Finkenauer, C., & Van Kooten, D. C. (2006).
Lying behavior, family functioning and adjustment in early ado-
lescence. Journal of Youth and Adolescence, 35, 949-958.
*Enticott, P. G., Ogloff, J. R. P., & Bradshaw, A. (2006). Associations
between laboratory measures of executive inhibitory control and
self-reported impulsivity. Personality and Individual Differences,
41, 285-294.
*Evers, C., Adriaanse, M., & De Ridder, D. T. D. (2007). [Eating
styles and affect regulation]. Unpublished raw data.
*Evers, C., & De Ridder, D. T. D. (2007). [Increased food intake
after resource depletion in order to replenish the self-regulatory
resource?]. Unpublished raw data.
Fagen, S. A., Long, N. J., & Stevens, D. J. (1975). Teaching children
self-control: Preventing emotional and learning problems in the
elementary school. Columbus, OH: Charles E. Merrill.
*Fennis, B. M., Janssen, L., & Vohs, K. D. (2009). Acts of benevo-
lence: A limited resource account of compliance with charitable
requests. Journal of Consumer Research, 35, 906-924.
Ferguson, M. J. (2008). On becoming ready to pursue a goal you
don’t know you have: Effects of nonconscious goals on evalu-
ative readiness. Journal of Personality and Social Psychology,
95, 1268-1294.
*Fillmore, M. T., Blackburn, J. S., & Harrison, E. L. R. (2008).
Acute disinhibiting effects of alcohol as a factor in risky driving
behavior. Drug and Alcohol Dependence, 95, 97-106.
*Finkel, E. J., & Campbell, W. K. (2001). Self-control and accom-
modation in close relationships: An interdependence analysis.
Journal of Personality and Social Psychology, 81, 263-277.
*Finkenauer, C. (2006–2008). [The Search for Inter-Personal Accu-
racy (SIPA) Project: A longitudinal study among newlywed
couples]. Unpublished raw data.
*Finkenauer, C., Engels, R. C. M. E., & Baumeister, R. F. (2005). Par-
enting and adolescent eternalizing and internalizing problems: The
role of self-control. International Journal of Behavioral Devel-
opment, 29, 58-69.
Fishbach, A., Friedman, R., & Kruglanski, A. (2003). Leading us not
unto temptation: Momentary allurements elicit automatic goal
activation. Journal of Personality and Social Psychology, 84,
296-309.
Fishbach, A., & Labroo, A. A. (2007). Be better or be merry: How
mood affects self-control. Journal of Personality and Social
Psychology, 93, 158-173.
Fitzsimons, G. M., & Bargh, J. A. (2004). Automatic self-regulation.
In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-
regulation: Research, theory, and applications (pp. 151-170).
New York, NY: Guilford.
Frederick, S., Loewenstein, G., & O’Donoghue, T. (2003).
Time discounting and time preference: A critical review. In
G. Loewenstein, D. Read, & R. F. Baumeister (Eds.), Time and
decision: Economic and psychological perspectives on inter-
temporal choice (pp. 13-86). New York, NY: Russell Sage.
*Friese, M., & Hofmann, W. (2009). Control me or I will control
you: Impulses, trait self-control, and the guidance of behavior.
Journal of Research in Personality, 43, 795-805.
*Frijns, T., & Finkenauer, C. (2005). [Psychosocial correlates of
keeping a secret: Longitudinal contribution to psychosocial
problems]. Unpublished raw data.
*Frijns, T., & Finkenauer, C. (2009). Longitudinal associations
between keeping a secret and psychosocial adjustment in ado-
lescence. International Journal of Behavioral Development, 33,
145-154.
*Frijns, T., Finkenauer, C., Vermulst, A., & Engels, R. C. M. E.
(2005). Keeping secrets from parents: Longitudinal associations
of secrecy in adolescence. Journal of Youth and Adolescence,
34, 137-148.
*Gailliot, M. T. (2007a). [Depletion basic study]. Unpublished raw
data.
*Gailliot, M. T. (2007b). [Depletion and controlled attention].
Unpublished raw data.
*Gailliot, M. T. (2007c). [Depletion, emotion regulation and help-
ing]. Unpublished raw data.
*Gailliot, M. T. (2007d). [Depletion and helping: Katie Banks].
Unpublished raw data.
*Gailliot, M. T. (2007e). [Depletion and memory]. Unpublished raw
data.
*Gailliot, M. T. (2007f). [Depletion and passivity]. Unpublished raw
data.
*Gailliot, M. T. (2007g). [Depletion and persistence]. Unpublished
raw data.
*Gailliot, M. T. (2007h). [Depletion and process dissociation].
Unpublished raw data.
*Gailliot, M. T. (2007i). [Depletion and public speaking]. Unpub-
lished raw data.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

96 Personality and Social Psychology Review 16(1)
*Gailliot, M. T. (2007j). [Depletion and sexual harassment]. Unpub-
lished raw data.
*Gailliot, M. T. (2007k). [Depletion and sexual jealousy]. Unpub-
lished raw data.
*Gailliot, M. T., & Baumeister, R. F. (2007). Self-regulation and sex-
ual restraint: Dispositionally and temporarily poor self-regulatory
abilities contribute to failures at restraining sexual behavior.
Personality and Social Psychology Bulletin, 33, 173-186.
*Gerrits, J. H. (2007). [Self-control and eating behavior of adolescents].
Unpublished raw data.
*Gerrits, J. H., O’Hara, R., Piko, B., Gibbons, F., De Ridder, D.,
De Wit, J., . . . Kamagh, S. (in press). Self-control, diet concerns,
and prototypes: Shared predictors of adolescents unhealthy eat-
ing across cultures. Health Education Research.
*Gibson, C., Schreck, C. J., & Miller, J. M. (2004). Binge drink-
ing and negative alcohol-related behaviors: A test of self-control
theory. Journal of Criminal Justice, 32, 411-420.
Gibson, C. L., Ward, J. T., Wright, J. P., Beaver, K. M., & Delisi, M.
(2010). Where does gender fit in the measurement of self-control?
Criminal Justice and Behavior, 37, 883-903.
*Glicksohn, J., Leshem, R., & Aharoni, R. (2006). Impulsivity and
time estimation: Casting a net to catch a fish. Personality and
Individual Differences, 40, 261-271.
*Glicksohn, J., & Nahari, G. (2007). Interacting personality traits?
Smoking as a test case. European Journal of Personality, 21,
225-234.
Gollwitzer, P. M. (1990). Action phases and mindsets. In E. T. Higgins
& R. M. Sorrentino (Eds.), Handbook of motivation and cognition
(Vol. 2, pp. 53-92). New York, NY: Guilford.
Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions
and goal achievement: A meta-analysis of effects and processes.
Advances in Experimental Social Psychology, 38, 69-119.
Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime.
Stanford, CA: Stanford University Press.
Gottman, J. (1994). Why marriages succeed or fail? New York, NY:
Simon & Schuster.
*Goudriaan, A. E., Grekin, E. R., & Sher, K. J. (2007). Decision
making and binge drinking: A longitudinal study. Alcoholism:
Clinical and Experimental Research, 31, 928-938.
Gough, H. G. (1987). California Psychological Inventory—
Administrator’s guide. Palo Alto, CA: Consulting Psychologists
Press.
*Gover, A. R., Kaukinen, C., & Fox, K. A. (2008). The relationship
between violence in the family of origin and dating violence
among college students. Journal of Interpersonal Violence, 23,
1667-1693.
Grasmick, H. G., Tittle, C. R., Bursik, R. J. J., & Arneklev, B. J.
(1993). Testing the core empirical implications of Gottfredson
and Hirschi’s general theory of crime. Journal of Research in
Crime and Delinquency, 30, 5-29.
*Ha, R., Namkoong, K., Kang, J., Kim, Y., & Kim, S. J. (2009).
Interaction between serotonin transporter promoter and dopa-
mine receptor D4 polymorphisms on decision making. Prog-
ress in Neuro-Psychopharmacology & Biological Psychiatry,
33, 1217-1222.
Heckhausen, H., & Gollwitzer, P. M. (1987). Thought contents and
cognitive functioning in motivational and volitional states of
mind. Motivation and Emotion, 11, 101-120.
*Heideheuvel Hospital. (2005, June). Self-control in obese adolescents
undergoing treatment. Hilversum, Netherlands: V. T. Colland &
D. T. D. De Ridder.
*Heyman, G. M., & Gibb, S. P. (2006). Delay discounting in college
cigarette chippers. Behavioural Pharmacology, 17, 669-679.
*Higgins, G. E. (2005). Can low self-control help with the understand-
ing of the software piracy problem? Deviant Behavior, 26, 1-24.
*Higgins, G. E. (2007). Digital piracy: An examination of low
self-control and motivation using short-term longitudinal data.
CyberPsychology & Behavior, 10, 523-529.
*Higgins, G. E., Fell, B. D., & Wilson, A. L. (2007). Low self-
control and social learning in understanding students’ intentions
to pirate movies in the United States. Social Science Computer
Review, 25, 339-357.
*Higgins, G. E., & Marcum, C. D. (2005). Can the theory of planned
behavior mediate the effects of low self-control on alcohol use?
College Student Journal, 39, 90-103.
*Higgins, G. E., Wolfe, S. E., & Marcum, C. D. (2008). Digital
piracy: An examination of three measurements of self-control.
Deviant Behavior, 29, 440-460.
Higgins, J. P., & Thompson, S. G. (2002). Quantifying heterogeneity
in a meta-analysis. Statistics in Medicine, 21, 1539-1558.
Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G.
(2003). Measuring inconsistency in meta-analyses. British Med-
ical Journal, 327, 557-560.
Hofmann, W., Friese, M., & Strack, F. (2009). Impulse and self-con-
trol from a dual-systems perspective. Psychological Science, 4,
162-176.
Hofmann, W., Gschwendner, T., Friese, M., Wiers, R. W., & Schmitt, M.
(2008). Working memory capacity and self-regulation: Toward
an individual difference perspective on behaviour determination
by automatic versus controlled processes. Journal of Personality
and Social Psychology, 95, 962-977.
*James, L. M., & Taylor, J. (2007). Impulsivity and negative emo-
tionality associated with substance use problems and Cluster B
personality in college students. Addictive Behaviors, 32, 714-727.
*Jollant, F., Guillaume, S., Jaussent, I., Bellivier, F., Leboyer, M.,
Castelnau, D., & Courtet, P. (2007). Psychiatric diagnoses and
personality traits associated with disadvantageous decision-
making. European Psychiatry, 22, 455-461.
*Jones, S., & Quisenberry, N. (2004). The general theory of crime:
How general is it? Deviant Behavior, 25, 401-426.
Kahneman, D., & Tversky, A. (1984). Choices, values and frames.
American Psychologist, 39, 341-350.
Kendall, P. C., & Williams, C. L. (1982). Assessing the cognitive
and behavioral components of children’s self-management. In
P. Karoly & F. H. Kanfer (Eds.), Self-management and behavior
change (pp. 240-284). New York, NY: Pergamon.
Kieras, J. E., Tobin, R. M., Graziano, W. G., & Rothbart, M. K.
(2005). You can’t always get what you want: Effortful control
and children’s response to undesired gifts. Psychological Sci-
ence, 16, 391-396.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 97
*Kuijer, R. G., & De Ridder, D. T. D. (2005). [Self-control and per-
formance on the Iowa Gambling Task]. Unpublished raw data.
*Kuijer, R. G., De Ridder, D. T. D., Ouwehand, C., Houx, B., &
Van den Bos, R. (2008). Dieting as a case of behavioral decision
making: Does self-control matter? Appetite, 51, 506-511.
*Langton, L., Piquero, N. L., & Hollinger, R. C. (2006). An empiri-
cal test of the relationship between employee theft and low self-
control. Deviant Behavior, 27, 537-565.
*Lejuez, C. W., Bornovalova, M. A., Reynolds, E. K.,
Daughters, S. B., & Curtin, J. J. (2007). Risk factors in the rela-
tionship between gender and crack/cocaine. Experimental and
Clinical Psychopharmacology, 15, 165-175.
*Leland, D. S., & Paulus, M. P. (2005). Increased risk-taking
decision-making but not altered response to punishment in
stimulant-using young adults. Drug and Alcohol Dependence,
78, 83-90.
Lensvelt-Mulders, G. J. L. M., Hox, J. J., Van der Heijden, P. G. M.,
& Maas, C. J. M. (2005). Meta-analysis of randomized response
research: 35 years of validation studies. Sociological Methods
and Research, 33, 319-348.
Letzring, T. D., Block, J., & Funder, D. C. (2005). Ego-control and
ego-resiliency: Generalization of self-report scales based on
personality descriptions from acquaintances, clinicians, and the
self. Journal of Research in Personality, 39, 395-422.
*Liao, P.-C., Uher, R., Lawrence, N., Treasure, J., Schmidt, U.,
Campbell, I. C., & Tchanturia, K. (2009). An examination of
decision making in bulimia nervosa. Journal of Clinical and
Experimental Neuropsychology, 31, 455-461.
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis:
Vol. 49. Applied social research methods series. Thousand Oaks,
CA: Sage.
Loewenstein, G. (1996). Out of control: Visceral influences on
behavior. Organizational Behavior and Human Decision Pro-
cesses, 65, 272-292.
Logue, A. W. (1988). Research on self-control: An integrating
framework. Behavioral and Brain Sciences, 11, 665-709.
*Loxton, N. J., Nguyen, D., Casey, L., & Dawe, S. (2008). Reward
drive, rash impulsivity and punishment sensitivity in problem
gamblers. Personality and Individual Differences, 45, 167-173.
*Malin, J., & Fowers, B. J. (2009). Adolescent self-control and
music and movie piracy. Computers in Human Behavior, 25,
718-722.
McCabe, L. A., Cunnington, M., & Brooks-Gunn, J. (2004). The
development of self regulation in young children: Individual char-
acteristics and environmental contexts. In R. F. Baumeister &
K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory,
and applications (pp. 340-356). New York, NY: Guilford.
Metcalfe, J., & Mischel, W. (1999). A hot/cool system analysis of
delay of gratification: Dynamics of willpower. Psychological
Review, 106, 3-19.
*Miller, B. L., Griffin, O. H., III, Gibson, C. L., & Khey, D. N. (2009).
Trippin’ on Sally D: Exploring predictors of Salvia divinorum
experimentation. Journal of Criminal Justice, 37, 396-403.
*Miller, J. D., Campbell, W. K., Young, D. L., Lakey, C. E.,
Reidy, D. E., Zeichner, A., & Goodie, A. S. (2009). Examining
the relations among narcissism, impulsivity, and self-defeating
behaviors. Journal of Personality, 77, 761-193.
Mischel, H. N., & Mischel, W. (1983). The development of children’s
knowledge of self-control strategies. Child Development, 54,
603-619.
Mischel, W. (1974). Processes in delay of gratification. In
L. Berkowitz (Ed.), Advances in experimental social psychology
(Vol. 7, pp. 249-292). San Diego, CA: Academic Press.
Mischel, W., Cantor, N., & Feldman, S. (1996). Principles of
self-regulation: The nature of willpower and self-control. In
E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology:
Handbook of basic principles (pp. 329-360). New York, NY:
Guilford.
Mischel, W., Shoda, Y., & Peake, P. K. (1988). The nature of adoles-
cent competencies predicted by preschool delay of gratification.
Journal of Personality and Social Psychology, 54, 687-696.
Mischel, W., Shoda, Y., & Rodriguez, M. L. (1989). Delay of grati-
fication in children. Science, 244, 933-938.
*Mobini, S., Grant, A., Kass, A. E., & Yeomans, M. R. (2007).
Relationships between functional and dysfunctional impulsiv-
ity, delay discounting and cognitive distortions. Personality and
Individual Differences, 43, 1517-1528.
*Morris, G. D., Wood, P. B., & Dunaway, R. G. (2006). Self-control,
native traditionalism, and native American substance use: Testing
the cultural invariance of a general theory of crime. Crime &
Delinquency, 52, 572-598.
Muraven, M. (2007). Autonomous self-control is less depleting.
Journal of Research in Personality, 42, 763-770.
Muraven, M., & Baumeister, R. F. (2000). Self-regulation and deple-
tion of limited resources: Does self-control resemble a muscle?
Psychological Bulletin, 126, 247-259.
Muraven, M., Tice, D. M., & Baumeister, R. F. (1998). Self-control
as a limited resource: Regulatory depletion patterns. Journal of
Personality and Social Psychology, 74, 774-789.
Nordgren, L., Van der Pligt, J., & Harreveld, F. (2010). The restraint
bias: How the illusion of self-restraint promotes impulsive
behavior. Psychological Science, 20, 1523-1528.
Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and
automatic control of behaviour. In R. J. Davidson, G. E. Schwartz,
& D. Shapiro (Eds.), Consciousness and self-regulation: Advances
in research and theory (pp. 1-18). New York, NY: Plenum.
*Ouwehand, C., & De Ridder, D. T. D. (2007). [Liking and wanting
in people with overweight]. Unpublished raw data.
*Ouwehand, C., & De Ridder, D. T. D. (2008). [Effect of tempting
food cues on the motivation to eat chocolate]. Unpublished raw
data.
*Özbay, Ö. (2008). Self-control, gender, and deviance among Turk-
ish university students. Journal of Criminal Justice, 36, 72-80.
Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor struc-
ture of the Barratt Impulsiveness Scale. Journal of Clinical Psy-
chology, 51, 768-774.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

98 Personality and Social Psychology Review 16(1)
*Pietrzak, R. H., Sprague, A., & Snyder, P. J. (2008). Trait impul-
siveness and executive function in healthy young adults. Journal
of Research in Personality, 42, 1347-1351.
Pratt, T. C., & Cullen, F. T. (2000). The empirical status of Gott-
fredson and Hirschi’s general theory of crime: A meta-analysis.
Criminology, 38, 931-964.
Rachlin, H. (2000). The science of self-control. Cambridge, MA:
Harvard University Press.
Rawn, C. D., & Vohs, K. D. (in press). People use self-control to risk
personal harm: An intra-interpersonal dilemma. Personality and
Social Psychology Review.
*Rebellon, C. J., Piquero, N. L., Piquero, A. R., & Thaxton, S.
(2009). Do frustrated economic expectations and objective
economic inequity promote crime? A randomized experiment
testing Agnew’s general strain theory. European Journal of
Criminology, 6, 47-71.
Roberts, B. W., Walton, K. E., & Bogg, T. (2005). Conscientiousness
and health across the life course. Review of General Psychology,
9, 156-168.
Rosenbaum, M. (1980). A schedule for assessing self-control behav-
iors: Preliminary findings. Behavior Therapy, 11, 109-121.
Rothbart, M. K., Ellis, L. K., Rueda, M. R., & Posner, M. I. (2003).
Developing mechanisms of temperamental effortful control.
Journal of Personality, 71, 1113-1143.
Rothbaum, F., 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.
Rusbult, C. E., Verette, J., Whitney, G. A., Slovik, L. F., & Lip-
kus, I. (1991). Accommodation processes in close relationships:
Theory and preliminary empirical evidence. Journal of Person-
ality and Social Psychology, 60, 53-78.
Schmeichel, B. J. (2007). Attention control, memory updating, and
emotion regulation temporarily reduce the capacity for execu-
tive control. Journal of Experimental Psychology: General,
136, 241-255.
*Schmeichel, B. J., Harmon-Jones, C., & Smith, S. M. (2008).
Cognitive reappraisal improves memory for emotional stimuli
among individuals high in trait self-control. Unpublished man-
uscript, Texas A&M University, College Station.
Schmeichel, B. J., Vohs, K. D., & Baumeister, R. F. (2003). Intellec-
tual performance and ego depletion: Role of the self in logical
reasoning and other information processing. Journal of Person-
ality and Social Psychology, 85, 33-46.
*Schmeichel, B. J., & Zell, A. (2007). Trait self-control predicts
performance on behavioral tests of self-control. Journal of Per-
sonality, 75, 1-13.
*Schoepfer, A., & Piquero, A. R. (2006). Self-control, moral beliefs,
and criminal activity. Deviant Behavior, 27, 51-71.
Seligman, M. (1994). What you can change and what you can’t.
New York, NY: Knopf.
Shoda, Y., Mischel, W., & Peake, P. K. (1990). Predicting adoles-
cent cognitive and self-regulatory competencies from preschool
delay of gratification: Identifying diagnostic conditions. Devel-
opmental Psychology, 26, 978-986.
Silverman, I. W. (2003). Gender differences in delay of gratifica-
tion: A meta-analysis. Sex Roles, 49, 451-463.
*Smith, T. R. (2004). Low self-control, staged opportunity, and
subsequent fraudulent behavior. Criminal Justice and Behavior,
31, 542-563.
Steinberg, L., Graham, S., O’Brien, L., Woolard, J., Cauffman, J., &
Banich, M. (2009). Age differences in future orientation and
delay discounting. Child Development, 80, 28-44.
*Stolk, R. P., Rosmalen, J. G., Postma, D. S., De Boer, R. A.,
Navis, G., Slaets, J. P., . . . Wolffenbuttel, B. H. R. (2008).
Universal risk for multifactorial diseases: LifeLines, a three-
generation population-based study. European Journal of Epi-
demiology, 23, 67-74.
*Stoltenberg, S. F., Batien, B. D., & Birgenheir, D. G. (2008). Does
gender moderate associations among impulsivity and health-
risk behaviors? Addictive Behaviors, 33, 252-265.
*Sun, I. Y., & Longazel, J. G. (2008). College students’ alcohol-
related problems: A test of competing theories. Journal of
Criminal Justice, 36, 554-562.
Tangney, J., Baumeister, R. F., & Boone, A. L. (2004). High self-
control predicts good adjustment, less pathology, better grades,
and interpersonal success. Journal of Personality, 72, 271-324.
*Thoolen, B. J. (2007). [Self-control and exercise]. Unpublished
raw data.
Tice, D. M., & Baumeister, R. F. (1997). Longitudinal study of pro-
crastination, performance, stress, and health: The costs and ben-
efits of dawdling. Psychological Science, 8, 454-458.
Tice, D. M., Baumeister, R. F., Shmueli, D., & Muraven, M. (2007).
Restoring the self: Positive affect helps improve self-regulation
following ego depletion. Journal of Experimental Social Psy-
chology, 43, 379-384.
Tittle, C. R., Ward, D. A., & Grasmick, H. G. (2003). Self-control
and crime/deviance: Cognitive vs. behavioral measures. Journal
of Quantitative Criminology, 19, 333-365.
*Unnever, J. D., Cullen, F. T., & Agnew, R. (2006). Why is “bad”
parenting criminogenic? Implications from rival theories. Youth
Violence and Juvenile Justice, 4, 3-33.
Van den Noortgate, W., & Onghena, P. (2003). Multilevel meta-
analysis: A comparison with traditional meta-analytical pro-
cedures. Educational and Psychological Measurement, 63,
765-790.
*VanderVeen, J. W., Cohen, L. M., Cukrowicz, K. C., & Trotter, D. R. M.
(2008). The role of impulsivity on smoking maintenance. Nic-
otine & Tobacco Research, 10, 1397-1404.
*VanderVeen, J. W., Cohen, L. M., Trotter, D. R. M., & Collins, F. L., Jr.
(2008). Impulsivity and the role of smoking-related outcome
expectancies among dependent college-aged cigarette smokers.
Addictive Behaviors, 33, 1006-1011.
Vazsonyi, A. T., Pickering, L. E., Junger, M., & Hessing, D.
(2001). An empirical test of a general theory of crime: A four-
nation comparative study of self-control and the prediction of
deviance. Journal of Research in Crime and Delinquency, 38,
91-131.
*Verhoeven, M., Junger, M., Van Aken, C., Dekovic, M., &
Van Aken, M. A. G. (2007). Parenting during toddlerhood:
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from

de Ridder et al. 99
Contributions of parental, contextual, and child characteristics.
Journal of Family Issues, 28, 1663-1691.
Vohs, K. D., & Baumeister, R. F. (2004). Understanding self-
regulation. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook
of self-regulation: Research, theory, and applications (pp. 1-9).
New York, NY: Guilford.
Vohs, K. D., & Faber, R. J. (2007). Spent resources: Self-regulatory
resource availability affects impulse buying. Journal of Con-
sumer Research, 33, 537-547.
*Vohs, K. D., Finkenauer, C., & Baumeister, R. F. (in press). The
sum of friends’ and lovers’ self-control scores predicts relation-
ship quality. Social Psychological and Personality Science.
*Wallace, H. M., Ready, C. B., & Weitenhagen, E. (2009). Narcissism
and task persistence. Self and Identity, 8, 78-93.
*Weller, R. E., Cook, E. W., III, Avsar, K. B., & Cox, J. E. (2008).
Obese women show greater delay discounting than healthy-
weight women. Appetite, 51, 563-569.
Wills, T. A., Cleary, S., Filer, M., Shinar, O., Mariani, J., & Spera, K.
(2001). Temperament related to early-onset substance use: Test
of a developmental model. Prevention Science, 2, 145-163.
Wills, T. A., Isasi, C. R., Mendoza, D., & Ainette, M. G. (2007).
Self-control constructs related to measures of dietary intake and
physical activity in adolescents. Journal of Adolescent Health,
41, 551-558.
Wills, T. A., Vaccaro, D., & McNamara, G. (1994). Novelty seeking
and related constructs as predictors of adolescent substance use.
Journal of Substance Abuse, 6, 1-20.
Wills, T. A., Walker, C., Mendoza, D., & Ainette, M. G. (2006).
Behavioral and emotional self-control: Relations to substance
use in samples of middle and high school students. Psychology
of Addictive Behaviors, 20, 265-278.
Wilson, D. B. (2000). Meta-analysis macros for SAS, SPSS, and
Stata [Computer software]. Retrieved from http://mason.gmu
.edu/~dwilsonb/ma.html
*Wojnar, M., Ilgen, M. A., Jakubczyk, A., Wnorowska, A.,
Klimkiewicz, A., & Brower, K. J. (2008). Impulsive suicide
attempts predict post-treatment relapse in alcohol-dependent
patients. Drug and Alcohol Dependence, 97, 268-275.
*Wolfe, S. E., Higgins, G. E., & Marcum, C. D. (2008). Deterrence
and digital piracy: A preliminary examination of the role of
viruses. Social Science Computer Review, 26, 317-333.
*Zabelina, D. L., Robinson, M. D., & Anicha, C. L. (2007). The
psychological tradeoffs of self-control: A multi-method inves-
tigation. Personality and Individual Differences, 43, 463-473.
at University Library Utrecht on January 30, 2012psr.sagepub.comDownloaded from