10.

1177/0146167204271598
PERSONALITY
Lockwood et al. AND
/ ROLE
SOCIAL
MODELS
PSYCHOLOGY
ACROSS CUL
BULLETIN
TURES ARTICLE

Promoting Success or Preventing Failure:

Cultural Differences in Motivation
by Positive and Negative Role Models

Penelope Lockwood
Tara C. Marshall
University of Toronto
Pamela Sadler
Wilfrid Laurier University

In two studies, cross-cultural differences in reactions to positive
and negative role models were examined. The authors predicted
that individuals from collectivistic cultures, who have a stronger
prevention orientation, would be most motivated by negative role
models, who highlight a strategy of avoiding failure; individu-
als from individualistic cultures, who have a stronger promotion
focus, would be most motivated by positive role models, who high-
light a strategy of pursuing success. In Study 1, the authors
examined participants’ reported preferences for positive and neg-
ative role models. Asian Canadian participants reported find-
ing negative models more motivating than did European Cana-
dians; self-construals and regulatory focus mediated cultural
differences in reactions to role models. In Study 2, the authors
examined the impact of role models on the academic motivation
of Asian Canadian and European Canadian participants.
Asian Canadians were motivated only by a negative model, and
European Canadians were motivated only by a positive model.
Keywords: social comparison; culture; motivation; role models; regu-
latory focus
by highlighting the example of a student who has
dropped out of university, an HIV-positive patient, or a
person who has suffered injury as a result of drinking
and driving, one can motivate an audience to study
harder, practice safer sex, or avoid driving while drunk.
Successful others can inspire individuals by exempli-
fying the success that one can achieve, providing a guide
to achieving such success (cf. Buunk, Collins, Taylor, Van
Yperen, & Dakof, 1990; Collins, 1996; Taylor & Lobel,
1989; Taylor, Wayment, & Carillo, 1996; Wood, 1989).
Unsuccessful others, in contrast, can motivate individu-
als by exemplifying the problems that may lie ahead,
emphasizing the behaviors that should be avoided to
avert a similar disaster (cf. Aspinwall, 1997; Buunk et al.,
1990; Wood & VanderZee, 1997). It is not clear however
that positive and negative examples will be equally moti-
vating for all people. In past research, we found that
although participants were spontaneously inspired by
positive role models (Lockwood & Kunda, 1997, 1999),
they typically viewed negative role models as irrelevant
(Lockwood, 2002). When the positive models repre-
sented an attainable outcome, participants hoped to

It is commonly assumed that both positive and negative
role models can motivate individuals to change their
behavior. A positive role model, an individual who has
achieved outstanding success, can motivate others to
pursue similar excellence. A negative role model, an
individual who has experienced some kind of failure or
misfortune, can motivate others to avoid similar distress.
For example, it is expected that by highlighting the
example of a successful athlete, student, or employee,
one can inspire an audience to exercise more, spend
more time studying, or work harder at a job. Conversely,
Authors’ Note: Penelope Lockwood and Tara C. Marshall, Department
of Psychology, University of Toronto; Pamela Sadler, Wilfrid Laurier
University. This research was supported by a grant from the Social Sci-
ences and Humanities Research Council of Canada to the first author.
We are grateful to Erik Woody for his advice with statistical analyses and
to Susan Furs, Niki Capogiannis, and Stephanie Cassin for their assis-
tance with data collection. Correspondence concerning this article
should be addressed to Penelope Lockwood, Department of Psychol-
ogy, University of Toronto, 100 St. George Street, Toronto, Ontario,
Canada, M5S 3G3 (e-mail: lockwood@psych.utoronto.ca).
PSPB, Vol. 31 No. 3, March 2005 379-392
DOI: 10.1177/0146167204271598
© 2005 by the Society for Personality and Social Psychology, Inc.

379

Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015

380 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
become like the more successful other in the future and
so were motivated to work harder (Lockwood & Kunda,
1999). In contrast, even when a negative model repre-
sented an outcome to which participants were vulnera-
ble, they tended to see this outcome as implausible, and
consequently their motivation was unaffected. Only
when they were forced to draw parallels between them-
selves and the other and consider the model as a possible
future self were they motivated to avoid this outcome
(Lockwood, 2002). These results suggest that negative
role models typically lack the strong motivational impact
that positive role models have.
This pattern of findings, however, may be unique to
individuals in Western, individualistic cultures. Mem-
bers of individualistic cultures tend to have strong inde-
pendent self-construals and focus on the self as a distinct,
unique individual (for reviews see Heine, Lehman,
Markus, & Kitayama, 1999; Markus & Kitayama, 1991;
Triandis, 1989). They are motivated to stand out from
their group as someone who is exceptional and success-
ful (Markus & Kitayama, 1991). They tend to focus on
their personal achievements and aspirations and exhibit
an attributional bias aimed at enhancing their self-
esteem (Heine & Lehman, 1997). Indeed, they view
information about successes as more relevant to their
self-esteem than information about failures (Kitayama,
Markus, Matsumoto, & Norasakkunkit, 1997). It seems
likely that positive role models may be especially relevant
to people in individualistic cultures. The positive model
demonstrates the kinds of achievements that are possi-
ble and the strategies needed to attain such successes.
Indeed, positive role models highlight a strategy that
may be particularly relevant to the regulatory focus of
North Americans. Recent evidence suggests that mem-
bers of individualistic cultures tend to favor
promotion over prevention strategies, focusing on the
positive outcomes they hope to approach rather than
the negative outcomes they hope to avoid (Elliot,
Chirkov, Kim, & Sheldon, 2001; Lee, Aaker, & Gardner,
2000). Promotion-focused individuals are especially
attuned to information relating to the pursuit of success
(Higgins, 1997, 1998). They tend to notice and recall
information pertaining to success achieved by other peo-
ple (Higgins & Tykocinski, 1992) and show especially
high motivation and persistence on tasks that are framed
in terms of promoting successful outcomes (Shah, Hig-
gins, & Friedman, 1998). The concern with self-
enhancement and personal achievement that character-
izes individuals with strongly independent self-
construals appears to be positively associated with pro-
motion strategies: North Americans tend to view events
framed in terms of winning as more important than
events framed in terms of losing and exhibit stronger
emotional reactions to gains than losses (Lee et al.,
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
2000); they are also especially likely to report striving for
approach rather than avoidance goals (Elliot et al.,
2001). In addition, North Americans are more likely to
persist on a task following a success than a failure experi-
ence, suggesting that they are more motivated to pursue
success than they are to correct their mistakes (Heine
et al., 2001). Members of cultures that emphasize inde-
pendent self-construals may therefore find positive role
models to be especially motivating. These models illus-
trate desirable outcomes to be pursued and demonstrate
strategies that can be used to achieve a similar degree of
success. Negative role models, on the other hand, pro-
vide little information about the achievements to which
one can aspire or the means by which one may attain
such achievements.
In contrast, in Eastern, more collectivistic cultures,
individuals may be particularly motivated by negative
role models. Members of collectivistic cultures tend to
have strong interdependent self-construals, focusing on
the self as part of a web of interpersonal relationships
(Heine et al., 1999; Markus & Kitayama, 1991; Triandis,
1989). They are motivated to fit in with their group and
maintain social harmony (Markus & Kitayama, 1991).
Consequently, they tend to focus on their responsibili-
ties and obligations to others and try to avoid behaviors
that might cause social disruptions or disappoint signifi-
cant others in their lives (Heine et al., 1999). They are
especially prone to self-criticism, seeking information
about faults that need to be corrected (Heine, Takata, &
Lehman, 2000). Indeed, these individuals view informa-
tion about failures as more relevant to their self-esteem
than information about successes (Kitayama et al.,
1997). It therefore seems reasonable that negative role
models will be especially important to members of
collectivistic cultures. By paying attention to the exam-
ple set by the negative model, these individuals can glean
information about the personal faults that they need to
correct and the strategies that are needed to prevent
failures in the future.
Indeed, the avoidance strategy highlighted by nega-
tive role models may be especially compatible with the
regulatory concerns that are emphasized in East Asian
cultures. Whereas North Americans favor a promotion
focus, striving to achieve success, East Asians tend to
adopt a prevention focus, striving to avert failures (Lee
et al., 2000). Prevention-focused individuals are espe-
cially attuned to information relating to the avoidance of
failure (Higgins, 1997). They tend to notice and recall
information pertaining to the failures experienced by
other people (Higgins & Tykocinski, 1992) and show
especially high motivation and persistence on tasks that
are framed in terms of preventing unsuccessful out-
comes (Shah et al., 1998). The concern with duties and
obligations that characterizes individuals with strongly
 Lockwood et al. / ROLE MODELS ACROSS CULTURES
interdependent self-construals appears to be positively
associated with such prevention strategies: East Asians
view events framed in terms of losing as more important
than events framed in terms of winning and exhibit
stronger emotional reactions to losses than gains (Lee
et al., 2000); they also report striving for avoidance
rather than approaching goals (Elliot et al., 2001). In
addition, East Asians show greater task persistence fol-
lowing a failure than a success experience, suggesting
that they are particularly concerned with preventing sim-
ilar failures in the future (Heine et al., 2001). Members
of cultures that emphasize interdependent self-
construals may therefore find negative role models to be
especially motivating. These models illustrate undesir-
able outcomes to be avoided and highlight strategies
that can be used to prevent similar failures. Positive mod-
els, on the other hand, provide little information about
the shortcomings that one needs to correct or the means
by which one may avoid future failures.
Consistent with the possibility that cultural back-
ground may affect responses to role models, recent
research suggests that individuals are motivated by role
models only when the models are congruent with their
regulatory concerns (Lockwood, Jordan, & Kunda,
2002). In two studies, when a promotion focus was tem-
porarily activated through a priming task, students were
motivated by a successful recent graduate who had
achieved a plum job but not by an unsuccessful recent
graduate who had ended up working in a fast food res-
taurant. In contrast, when a prevention focus was
primed, participants were motivated by the unsuccessful
but not the successful recent graduate. Given that pro-
motion is emphasized in Western cultures and preven-
tion is emphasized in Eastern cultures, we might expect
corresponding differences in how members of these
cultures react to positive and negative role models.
In sum, past research has found that individuals from
Eastern cultures tend to focus on avoiding negative out-
comes, whereas those from Western cultures tend to
focus on promoting positive outcomes (Elliot et al.,
2001). These differences in regulatory focus appear to
be at least partially mediated by cultural differences in
self-construal: Interdependent self-construals are associ-
ated with prevention orientation, whereas independent
self-construals are associated with promotion orienta-
tion (Lee et al., 2000). Finally, past research indicates
that prevention-focused individuals tend to be more
motivated by negative models, whereas promotion-
focused individuals tend to be more motivated by posi-
tive models (Lockwood et al., 2002).
Figure 1 shows the resulting theoretical model, in
which self-construal partly mediates the relationship of
cultural background to regulatory focus and regulatory
focus in turn mediates the relationship of cultural back-
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
381
ground and self-construal to motivation by role models.
In the present research, we provide the first test of this
full model. The model specifies a separate chain of
effects for each of the two types of motivational out-
comes, reflecting the fact that the respective pairs of vari-
ables (interdependent and independent self-construal,
prevention and promotion focus, and negative-model
and positive-model motivation) are not conceptual
opposites but reasonably independent. The two parallel
processes, represented by the upper and lower limbs of
the model, are hypothesized not to affect each other
(indicated by the lack of crossing paths) and to share
only the common impact of cultural background.1
With this proposed theoretical model as a framework,
we examined the reactions of participants from East
Asian and Western European cultural backgrounds to
positive and negative role models. In Study 1, we exam-
ined participants’ beliefs about whether they would be
more motivated by positive or negative role models. In
Study 2, we exposed participants to positive and negative
role models and then examined the impact of the mod-
els on their motivation. We predicted that negative mod-
els would be more motivating to the participants with an
East Asian cultural background than to those with a
Western European background, whereas positive role
models would be more motivating to participants with a
Western European cultural background than to those
with an East Asian background. Study 1 also examined
self-construals and regulatory focus as hypothesized
mediators of these differences.
STUDY 1: CULTURAL DIFFERENCES IN BELIEFS
ABOUT THE MOTIVATING IMPACT OF
POSITIVE AND NEGATIVE ROLE MODELS
We first examined participants’ beliefs about whether
they would be more motivated by positive or negative
role models. Student participants from East Asian and
Western European cultural backgrounds were asked to
indicate how motivated they would be by a series of posi-
tive and negative academic role models. They also com-
pleted measures assessing their self-construals and their
regulatory focus. We expected that individuals from East
Asian and Western European cultural backgrounds
would differ in their responses to positive and negative
role models. More specifically, we expected that East
Asian Canadian participants would have stronger inter-
dependent self-construals, which would be associated
with a stronger prevention focus, which in turn would be
associated with greater motivation by negative role mod-
els. In contrast, we expected that European Canadian
participants would have stronger independent self-
construals, which would be associated with a stronger
promotion focus, which in turn would be associated with
greater motivation by positive role models.
382 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
Interdependent
Self-Construal
Cultural
Background
Independent
Self-Construal
Figure 1 Model for motivation by negative and positive role models as functions of prevention and promotion focus, interdependent and inde-
pendent self-construal, and cultural background.
Method
Participants. Participants were 58 Asian Canadian (28
men and 30 women) and 23 European Canadian (13
men and 10 women) students recruited from summer
classes at the University of Toronto. Participants were
each paid $10 for taking part in the study. There were no
gender effects on any of the variables, therefore gender
is not discussed further. Participants filled out an initial
one-page sign-up form in which they provided informa-
tion about their cultural background along with more
general contact and demographic information. Partici-
pants were selected for the study if they identified their
cultural background as East Asian or Western European.
Asian Canadian participants had lived in Canada for sig-
nificantly fewer years (M = 9.4) than had European
Canadian participants (M = 20.5), F(1, 78) = 70.15, p <
.0001. The mean age of Asian Canadian participants
(M = 20.21) did not differ significantly from that of Euro-
pean Canadian participants (M = 20.57), F(1, 79) = 1.02,
p = .31. Among Asian Canadian participants, cultural
backgrounds included Hong Kong (n = 23), Taiwan (n =
11), South Korea (n = 6), China (n = 6), and other (e.g.,
Singapore, Malaysia, the Philippines, etc.; n = 12).
Procedure. Participants were invited to take part in a
study on adjustment to university life. They were told
that the researchers were interested in finding out about
the factors that affect students’ experiences, adjustment,
and behaviors.
Participants first completed Singelis’s (1994) Self-
Construal Scale. Ratings on the 12 independence and 12
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
Prevention Motivation by
Focus Negative Model
Promotion Motivation by
Focus Positive Model
interdependence items were made on a 7-point scale
with endpoints ranging from 1 (strongly disagree) to 7
(strongly agree). Next, participants completed a measure
of regulatory focus (Lockwood et al., 2002); this scale
includes 9 items assessing prevention concerns (e.g., “I
frequently think about how I can prevent failures in my
life”) and 9 items assessing promotion concerns (e.g., “I
frequently imagine how I will achieve my hopes and aspi-
rations”). Ratings on the promotion and prevention
items were made using a 9-point scale with endpoints
ranging from 1 (not at all true of me) to 9 (very true of me).
Next, participants were told that the researchers
were interested in finding out how people motivate
themselves:
Sometimes, we are motivated to avoid becoming like a
person who is experiencing problems, and sometimes
we are motivated to strive to become like a person who is
experiencing success. Both positive and negative exam-
ples can motivate us to change our behaviors. We are
interested in the kinds of examples that might motivate
YOU to work harder.
Participants then rated the extent to which they would
be motivated by six positive models (e.g., “someone from
my major who is getting straight A+s,” “a person who
graduated from my program and was offered a fantastic
job”) and six negative models (e.g., “someone from my
major who has been placed on academic probation,” “a
person who graduated from my program and couldn’t
get the job he/she wanted”). The positive and negative
models were presented in random order. Ratings were
 Lockwood et al. / ROLE MODELS ACROSS CULTURES 383

made on a 7-point scale with endpoints ranging from 1 TABLE 1: Asian Canadian and European Canadian Participants’
Mean Ratings of Self-Construals, Regulatory Focus, and
(not at all motivated) to 7 (very motivated).
Motivation by Negative and Positive Role Models (Study 1)
Results and Discussion Cultural Background
Means and standard deviations on all variables for Asian European
Asian Canadian and European Canadian participants Canadian Canadian
are presented in Table 1. Variable M SD M SD

Self-construals. The independence (Cronbach’s α =
.60) and interdependence (Cronbach’s α = .63)
subscales were moderately reliable; the two subscales
were not correlated, r = –.04, p = .70. As predicted, Asian
Canadian participants had higher interdependent self-
construals than did European Canadians, F(1, 79) = 4.52,
p = .04. Unexpectedly, Asian Canadians’ independent
self-construals were only marginally lower than those of
European Canadians, F(1, 79) = 2.82, p = .10. Partici-
pants were selected based on their cultural background
rather than on their actual subjective identification with
a particular cultural group. Indeed, many of the East
Asian participants had spent a large portion of their lives
in Canada and so may have had considerable exposure
to the predominantly independence-oriented North
American culture; these individuals may have identified
almost as much with North American as with East Asian
culture. Thus, the Asian Canadian participants may have
developed self-construals that reflected aspects of both
their Asian cultural background and the North Ameri-
can cultural environment in which they were living.
Regulatory focus. The regulatory focus measure
includes a promotion and a prevention subscale; both
were reliable (promotion α = .73, prevention α = .82).
The two subscales were not correlated, r = .03, p = .82. As
predicted, Asian Canadian participants had a stronger
prevention focus than did European Canadian partici-
pants, F(1, 79) = 20.21, p < .0001. Unexpectedly, the pro-
motion focus of Asian Canadian and European Cana-
dian participants did not differ significantly, F(1, 79) =
1.74, p = .19. These individuals may have developed a
prevention focus through their family background but
may also have developed a promotion focus through
their contact with North American culture.
Consistent with previous research (Elliot et al., 2001),
independent self-construals were positively correlated
with promotion focus (r = .22, p = .05) but not with pre-
vention focus (r = –.12, p = .31); interdependent self-
construals were positively correlated with prevention
focus (r = .30, p = .007) but not with promotion focus (r =
–.08, p = .46).
Motivation by negative and positive role models. The nega-
tive role model items were combined into a single index
(Cronbach’s α = .76). The positive role model items were
also combined into a single index (Cronbach’s α = .84).
To examine cultural differences in perceived motivation
Interdependent self-construals 4.87 0.71 4.51 0.63
Independent self-construals 4.59 0.65 4.87 0.80
Prevention focus 5.84 1.04 4.57 1.37
Promotion focus 7.08 1.02 7.43 1.18
Motivation by negative models 4.60 0.97 3.67 1.17
Motivation by positive models 4.95 1.13 4.85 1.18
by role models, we conducted a mixed factorial ANOVA,
with cultural background (Asian Canadian or European
Canadian) as a between-participants variable and role
model type (positive or negative) as a within-participants
variable. As expected, the cultural background by role
model type interaction was significant, F(1, 79) = 5.93, p =
.02. Asian Canadians rated negative role models as signif-
icantly more motivating than did European Canadians,
F(1, 79) = 11.95, p < .001. However, Asian Canadians and
European Canadians did not differ significantly in their
ratings of the motivating impact of positive models, F(1,
79) = .12, p = .73. As noted earlier, the independent self-
construals and promotion focus of the Asian Canadian
participants in our sample were almost as high as those
of European Canadians; consequently, positive exam-
ples may have been motivating for both groups.
We next examined self-construals as predictors of
motivation by positive and negative role models. We first
regressed motivation by negative models on interdepen-
dent and independent self-construals simultaneously. As
we would expect, interdependent self-construals pre-
dicted motivation by negative role models, β = .55, p <
.002, but independent self-construals did not, β = –.15,
p = .38. Next, we regressed motivation by positive models
on both forms of self-construal. Independent self-
construals predicted motivation by positive models, β =
.36, p = .05, but interdependent self-construals did not,
β = .18, p = .32.
We then examined regulatory orientations as predic-
tors of motivation by positive and negative role models.
We regressed motivation by each kind of model on pre-
vention and promotion focus simultaneously. Preven-
tion focus predicted motivation by negative models, β =
.54, p < .0001, but promotion focus did not, β = –.03, p =
.76. In contrast, promotion focus predicted motivation
by positive models, β = .49, p < .0001, but prevention
focus did not, β = .14, p = .11.
Structural equation model to evaluate mediation hypotheses.
The regression analyses revealed that both self-

Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015

384 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
construals and regulatory focus demonstrate relation-
ships with motivation by role models that are consistent
with our hypotheses. To further evaluate the role of
these variables as hypothesized mediators, we applied
structural equation modeling to the data from Study 1 to
estimate the integrative model previously shown in Fig-
ure 1.2
The resulting parameter estimates, obtained using
Amos 5.0 (Arbuckle & Wothke, 1999), are shown in Fig-
ure 2. (By convention, all measured variables are shown
in rectangles, and all variables labeled Zi represent unex-
plained variance in the corresponding criterion variable
to which they point.) This model fit very well, χ2(13, N =
81) = 14.35, ns, Comparative Fit Index = 0.98, root mean
square error of approximation = 0.04, probability of
close fit = 0.53. Model fit is a reflection of the fixed
parameters (i.e., those paths that are set to zero) in the
proposed model, not the path estimates (McDonald &
Ho, 2002). To better understand the implications of the
excellent fit of the present model, note that there are two
kinds of omitted paths in this model. First, there are the
omitted crossing paths from one chain to the other (e.g.,
the omitted path from prevention focus to motivation by
positive models). These represent the hypothesis that
the two parallel processes, one for each type of motiva-
tion, do not affect each other. Second, within each chain
there are omitted direct-effect paths to motivation (e.g.,
for the upper chain, omitted direct effects of cultural
background and interdependent self-construal on moti-
vation by negative models). These represent the hypoth-
esis that any effects on motivation are fully mediated by
regulatory focus. The excellent fit of the model simulta-
neously supports both of these hypotheses.
The standardized path coefficients (as shown in Fig-
ure 2) likewise support our mediation hypotheses, par-
ticularly for the variables in the upper chain of the
model. More specifically, cultural background signifi-
cantly affected interdependent self-construal, which in
turn significantly affected prevention focus. In addition,
controlling for interdependent self-construal, cultural
background directly affected prevention focus. These
results suggest that interdependent self-construal is a
partial mediator of the relationship between cultural
background and prevention focus. In contrast, the excel-
lent fit of the model despite the lack of direct paths
between the other variables in the top part of the dia-
gram supports the hypothesis that prevention focus fully
mediates the relationship of cultural background and
interdependent self-construal to motivation by negative
role models.3
The results from the lower chain of the model were
generally weaker but still in the predicted direction.
Although the effect of cultural background on inde-
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
pendent self-construal was in the predicted direction, it
was marginally significant, as was the effect of independ-
ent self-construal on promotion focus; however, promo-
tion focus in turn significantly affected motivation by
positive role models, as predicted. Controlling for inde-
pendent self-construal, the direct effect of cultural back-
ground on promotion focus was not statistically signifi-
cant. These results provide tentative evidence that
independent self-construal may mediate the relation-
ship between cultural background and promotion focus
and somewhat stronger evidence that promotion focus
may mediate the relationship between independent self-
construal and motivation by positive role models.4
Summary and discussion. Overall, consistent with previ-
ous research, we found cultural differences in regulatory
focus and self-construals: Asian Canadians had stronger
prevention orientations and more interdependent self-
construals than did European Canadians. Also consis-
tent with previous research, differences in regulatory
focus were associated with motivation by role models:
Promotion focus was associated with motivation by posi-
tive models, and prevention focus was associated with
motivation by negative models. Moreover, Study 1 not
only replicated these previously studied effects but also
demonstrated how self-construal and regulatory focus
mediate cultural differences in motivation by positive
and negative role models. Asian Canadians’ stronger
interdependent self-construals were associated with pre-
vention focus, which in turn predicted motivation by
negative models. In contrast, European Canadians’
stronger independent self-construals were associated
with promotion focus, which predicted motivation by
positive models. This pattern of relationships is consis-
tent with our hypotheses; individuals from cultures that
place a strong emphasis on independence are likely to
be especially motivated by exemplars of success. Individ-
uals from cultures that place a strong emphasis on inter-
dependence are likely to be especially motivated by
exemplars of failure. Study 1 thus provides new evidence
indicating that culture and self-construals are connected
to motivation by positive and negative role models.
Contrary to our predictions however, European Cana-
dians were not more motivated by the positive exemplars
than were Asian Canadians. We surmise that the Asian
participants in this study, who had lived in Canada for an
average of 9 years, may have been exposed to sufficient
North American culture to develop a relatively strong
promotion focus. Individuals who have been exposed to
both collectivistic and individualistic cultures may be
motivated to some degree by both negative and positive
role models. To examine the possibility that regulatory
focus and self-construal can change with exposure to
 Lockwood et al. / ROLE MODELS ACROSS CULTURES 385

Z1 Z3 Z5

Interdependent .20* Prevention .61*** Motivation by

Self-Construal Focus Negative Model

.23**
.40***

Cultural
Background

†
.11
"
.19
"

Independent .20† Promotion .46*** Motivation by

Self-Construal Focus Positive Model

Z2 Z4 Z6

Figure 2 Motivation by negative and positive role models as functions of prevention and promotion focus, interdependent and independent self-
construal, and cultural background. Standardized estimates (betas) are presented.
NOTE: Cultural background was coded 0 for European Canadians and 1 for East Asian Canadians.
†p < .10; *p < .05; **p < .01; ***p < .001.

North American culture, we assessed the relationship
between these variables and the length of time that East
Asian individuals had lived in Canada. First, in a sample
of 316 introductory psychology students who identified
their ethnic background as East Asian, participants rated
themselves on a subset of five promotion and five pre-
vention items (Cronbach’s alphas = .59 and .70, respec-
tively); they also indicated how long they had lived in
Canada on a 10-point scale (1 = less than 1 year; 2 = 1 to 2
years; 3 = 2 to 3 years; 4 = 3 to 4 years; 5 = 4 to 5 years; 6 = 5
to 10 years; 7 = 10 to 15 years; 8 = 15 to 20 years; 9 = more
than 20 years but not my entire life; 10 = my entire life).
Promotion focus was weakly but positively correlated
with length of time in Canada, r = .14, p = .01; prevention
focus was not related to length of time in Canada, r = .05,
p = .39. Second, in a separate sample of 563 introductory
psychology students who identified their ethnic back-
ground as East Asian, participants rated themselves on a
subset of five independence and five interdependence
items (Cronbach’s alphas = .47 and .55, respectively);
they also indicated how long they had lived in Canada on
the 10-point scale. Independence was weakly but posi-
tively correlated with length of time in Canada, r = .09, p =
.03; interdependence was not significantly correlated
with length of time in Canada, r = –.06, p = .16. Thus, as
individuals of East Asian cultural origins spend more
time in Canada, their promotion orientation and inde-
pendent self-construals grow somewhat stronger. In
these samples, the correlations with length of time in
Canada were quite low, possibly due to the relatively
poor reliability of the reduced promotion and inde-
pendence scales. In addition, number of years in Canada
is an imperfect proxy for acculturation: Some East Asian
Canadians may have adopted North American cultural
values relatively quickly; others, living in close-knit fam-
ily environments in which East Asian cultural influences
remained strong, may have experienced less accultura-
tion despite longer sojourns in North America. Overall,
however, it seems likely that East Asians who have spent
more time in Western environments will be somewhat
more promotion focused and more independent and
consequently more likely to be motivated by positive role
models.
STUDY 2: CULTURAL DIFFERENCES IN THE MOTIVATING
IMPACT OF POSITIVE AND NEGATIVE ROLE MODELS
In Study 2, we sought additional evidence for our
hypothesis that Asian Canadians and European Canadi-

Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015

386 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
ans would differ in their reactions to positive and nega-
tive role models. To ensure that the Asian Canadian
participants would have had less exposure to North
American culture, we limited our sample to Asian Cana-
dians who had been living in North America for less than
10 years.
In addition, in Study 1, we examined participants’
beliefs about how their motivation levels would be
affected by positive and negative role models. Individu-
als’ theories about how they would respond, however,
may not accurately reflect the degree to which they
would actually be motivated by such models. Asian Cana-
dians may be inspired by positive models but due to a
bias toward modesty (Markus & Kitayama, 1991) might
be unwilling to admit that they believe they could
become like the outstanding other in the future. Alter-
natively, European Canadians, who tend to pay less atten-
tion to information about failures (Heine et al., 1999),
may be unaware that negative models can boost their
motivation levels. Accordingly, in a second study, we
examined participants’ actual reactions to role models.
We also used Study 2 to rule out an alternative expla-
nation for our findings. It is possible that new migrants to
Canada may be more motivated by negative role models
not because they have a stronger prevention focus but
rather because they have lower academic expectations
and more negative perceptions of their academic com-
petence. Indeed, the Asian Canadian participants in
Study 1 took part in the study in a context in which many
were studying in a second language and may therefore
have been experiencing academic difficulties; they may
simply have seen themselves as more likely to experience
problems like those of the negative role model than
would those students studying in their first language. In
general, individuals choose comparison to others who
have achieved a similar performance level (e.g.,
Hakmiller, 1966; Wheeler, 1966) or who are similar on
attributes related to performance (for a review, see
Goethals & Darley, 1977); for example, an aspiring
skater can better assess her athletic skill by comparing
herself to another athlete at the same level who has had
similar training and practice. To rule out the possibility
that similarity in performance level was responsible for
the obtained effects, we assessed participants’ past aca-
demic performance and their expected future perfor-
mance. We also assessed participants’ more global per-
ceptions of their self-competence by using a measure of
self-esteem that includes subscales assessing both self-
competence and self-liking (Tafarodi & Swann, 1995).
Participants read about a highly successful or highly
unsuccessful student and then rated their own academic
motivation. They also rated the degree to which they per-
ceived the model as relevant to themselves. We expected
that Asian Canadian participants would be more moti-
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
vated by the negative than the positive model and would
view the negative model as more relevant. In contrast, we
expected that European Canadian participants would be
more motivated by the positive than the negative model
and would view the positive model as more relevant.
Method
Pretesting. Participants completed a pretest question-
naire in which they indicated their academic major, their
cultural background, and the number of years they had
lived in Canada. Asian Canadian participants were
selected if they indicated that they were born in an East
Asian country (i.e., China, Hong Kong, Japan, Korea)
and had lived in Canada for less than 10 years (M = 3.69).
European Canadians were selected if they indicated that
they had been born in Canada and that their cultural
background was Western European. As part of this ques-
tionnaire, participants also completed a measure of self-
esteem, the Self-Liking/Self-Competence Scale (SLCS;
Tafarodi & Swann, 1995). Ratings for this measure were
made on a 5-point scale with endpoints ranging from 1
(strongly agree) to 5 (strongly disagree).5
Participants. Participants were 47 Asian Canadian (19
men and 28 women) and 45 European Canadian (20
men and 25 women) students who took part in the study
for course credit. To ensure that participants would be
able to view the models as potential future selves, we
selected participants at the beginning of the fall term of
their first year of university. All participants read about a
model who had completed his or her first year and thus
was sufficiently far advanced from the participants’ own
academic stage that the model’s experiences would
seem like a plausible future outcome.
Five participants were excluded because they had
changed their academic major since they were recruited
for the study and so were exposed to a nonrelevant role
model (targets were matched with participants on aca-
demic major because previous research found that par-
ticipants were unaffected by a role model in a domain
that they viewed as irrelevant; Lockwood & Kunda,
1997). Altogether, 44 Asian Canadian and 43 European
Canadian participants were included in the analyses.
Gender had no impact on any of the variables and there-
fore will not be discussed further.
Procedure. About 2 to 6 weeks after the prescreening,
participants were invited to take part in a study on adjust-
ment to university life. They were informed that the Psy-
chology Department had been gathering data to investi-
gate the factors associated with academic success and
failure; the researchers were interested in learning
about students’ impressions of other individuals’ experi-
ences and also in the participants’ own level of academic
adjustment.
 Lockwood et al. / ROLE MODELS ACROSS CULTURES
Positive and negative role model descriptions. Participants
were then exposed to a positive or negative role model.
First, they read a one-page sign-up sheet that included
blanks for name, phone number, gender, academic
major, and cultural background. The sign-up sheet had
been filled in by hand, ostensibly by a participant in a
previous study; the name and phone number were
blacked out in marker as though to preserve the ano-
nymity of this bogus student. Participants were always
exposed to a target who was matched with them on gen-
der, academic major, and cultural background. For
example, a female European Canadian chemistry stu-
dent would read about a female chemistry student with a
Western European cultural background. We matched
participants on cultural background because we rea-
soned that it was possible that Asian participants might
otherwise assume that the role model was non-Asian;
such an outgroup member might be viewed as irrelevant.
The page following the sign-up sheet was a self-
description, ostensibly written by the same person. In
the positive model condition, the target described suc-
cessful academic experiences and finished the self-
description by noting,
I found that my second term went even better than the
first. I kept my grades up, but balanced my studies with
extracurricular activities. I just found out that my final
marks exceeded my expectations. In fact, I won a U of T
award for outstanding first-year accomplishment. Over-
all, I would say that my first year has been fantastic.
In the negative model condition, the target described
experiencing academic difficulties and finished the self-
description by noting,
I fell behind again in my second term. My courses
seemed even harder and I couldn’t keep up. My marks
were pretty pitiful—I did so poorly that I was put on aca-
demic probation, and I may not be able to continue here
at U of T. Overall, I would say that my first year has been a
disaster.
Role model adjustment ratings. Next, participants rated
the target they read about on five items tapping aca-
demic adjustment (e.g., “How successful do you think
this person is?” “How well-adjusted do you think this per-
son is?”). Ratings were made on a 9-point scale with end-
points ranging from 1 (not at all) to 9 (very).
Motivation ratings. Participants then went on to “Part
2” of the study in which they were asked to provide infor-
mation about their own activities and adjustment. Partic-
ipants first rated themselves on 18 items designed to tap
into their academic motivation (e.g., “I plan to study
harder,” “I plan to spend more time at the library,” “I
plan to keep up with my reading assignments”). Ratings
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
387
were made on an 11-point scale with endpoints ranging
from 1 (not at all true) to 11 (very true).
Relevance ratings. Next, participants rated the degree
to which they viewed the target as relevant to themselves;
they read that
people often seek information about themselves by com-
paring themselves to other people. We tend to judge our
own abilities, skills, interests etc. by looking at how well
the people around us are doing. Some people are more
relevant to us than others for the purpose of compari-
son. How relevant is this person to you for the purpose of
comparison?
Participants then rated the target’s relevance on an 11-
point scale with endpoints ranging from 1 (not at all rele-
vant) to 11 (very relevant).
Comparative adjustment. Participants also completed a
manipulation check in which they rated how their own
level of adjustment compared to that of the person they
read about. Ratings were made on a 7-point scale with
endpoints ranging from –3 (I am doing much worse than
this person) to +3 (I am doing much better than this person).
Past and future academic performance. Finally, partici-
pants were asked to provide their final high school aver-
age and the average grade they expected to receive in
their final year of university.
A no target control condition was also included in
which participants rated themselves on the motivation
items without first reading about a role model.
In sum, the study had a 2 (cultural background: East
Asian or Western European) by 3 (role model: positive,
negative, or no model) between-participants design. At
the end of the experimental session, all participants
were probed for suspicion and then debriefed.
Results and Discussion
Self-liking and self-competence. The self-liking subscale
items were combined into a single index after first revers-
ing the negative items (Cronbach’s α = .87). Asian Cana-
dian (M = 3.39) and European Canadian (M = 3.34) par-
ticipants did not differ significantly in their self-liking,
F(1, 69) = .05, p = .82. The self-competence subscale
items were also combined into a single index after first
reversing the negative items (Cronbach’s α = .75). Asian
Canadian participants’ self-competence ratings (M =
3.03) were marginally lower than those of European
Canadian participants (M = 3.30), F(1, 69) = 3.22, p = .08.
Academic background and expectations. The graduating
high school averages of Asian Canadian (M = 85.07) and
European Canadian (M = 84.38) students did not differ
significantly, F(1, 84) = .33, p = .57, neither did the
expected university averages of Asian Canadians (M =
388 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
73.51) and European Canadians (M =75.17) also did not
differ, F(1, 83) = 1.08, p = .30; both groups expected to
graduate with a relatively strong academic record.
Role model adjustment. Ratings of the target’s adjust-
ment were combined to form a single index (Cronbach’s
α = .99). The main effect of condition was highly signifi-
cant, F(1, 53) = 333.92, p < .0001. Participants rated the
positive role model (M = 7.95) as significantly better
adjusted than the negative role model (M = 2.39). Unex-
pectedly, the main effect of cultural background was also
significant, F(1, 53) = 7.01, p = .01; Asian Canadian partic-
ipants rated the positive (M = 7.43) and negative (M =
2.10) models as somewhat less well adjusted than did
European Canadians (Ms = 8.46 and 2.69, respectively).
The model condition by cultural background interac-
tion was not significant, F(1, 53) = .52, p = .47.
Comparative adjustment. Participants rated their own
adjustment as better than that of the negative model
(M = 1.96) and worse than that of the positive model
(M = –1.62). This main effect of condition was highly sig-
nificant, F(1, 53) = 198.62, p < .0001. The main effect of
cultural background was not significant, F(1, 53) = .20,
p = .66, suggesting that although Asian Canadians rated
the models’ adjustment more negatively than did Euro-
pean Canadians, they did not view their own level of
adjustment relative to the model more positively; Asian
Canadians appeared to have more negative perceptions
of both their own and the models’ adjustment than did
European Canadians. The condition by cultural back-
ground interaction was not significant, F(1, 53) = .91, p =
.34. Thus, we were successful in creating positive and
negative models that were viewed about equally posi-
tively and negatively relative to the self by Asian Canadi-
ans and European Canadians.
Motivation. Motivation ratings were combined to form
a single index (Cronbach’s α = .85). As may be seen in
Figure 3, Asian Canadian participants were more moti-
vated by a negative than a positive model, whereas Euro-
pean Canadians were more motivated by a positive than
a negative model; this condition by cultural background
interaction was significant, F(2, 81) = 5.83, p < .004. This
interaction remained significant after controlling for
both the self-competence and self-liking subscales of the
SCSL Scale, F(2, 63) = 3.07, p = .05. The interaction also
remained significant after controlling for high school
average and expected graduating average, F(2, 77) =
5.07, p = .01. This suggests that the different reactions of
Asian Canadian and European Canadian participants
cannot be explained by differences in how these two
groups viewed their abilities or how well they were
expecting to perform academically.
We had predicted that participants would be moti-
vated only by models that were congruent with the regu-
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
No Model
Positive Model
9
Negative Model
8.5
8
Motivation
7.5
7
6.5
6
5.5
5
East Asian Western European
Cultural Background
Figure 3 Motivation ratings of Asian Canadian and European Cana-
dian participants exposed to a positive, negative, or no role
model (Study 2).
latory focus favored by their culture. As expected, Asian
Canadian participants exposed to a negative model were
more motivated than were Asian Canadians exposed to a
positive model, t(81) = 1.96, p = .053, or no model, t(81) =
2.47, p = .02. The motivation of Asian Canadians exposed
to a positive model did not differ significantly from the
motivation of those exposed to no model, t(81) = .14, p =
.89. In contrast, European Canadians exposed to a posi-
tive model were more motivated than were European
Canadians exposed to a negative model, t(81) = 2.79, p =
.007, or no model, t(81) = 2.80, p = .006. The motivation
of European Canadians exposed to a negative model did
not differ from the motivation of those exposed to no
model, t(81) = .10, p = .92. Thus, Asian Canadians were
motivated only by negative models, and European Cana-
dians were motivated only by positive models.
Relevance. The role model condition by cultural back-
ground interaction was significant, F(1, 53) = 5.10, p =
.03. As may be seen in Figure 4, Asian Canadians viewed
the negative model to be more relevant than the positive
model, F(1, 53) = 3.84, p = .06. In contrast, European
Canadians viewed the positive model to be more rele-
vant than the negative model; however, this contrast did
not reach significance, F(1, 53) = 1.57, p = .21. Thus,
Asian Canadians viewed a model to be most relevant
when that model fit their regulatory concerns.
Overall, the results of Study 2 support our hypothesis
that individuals will be most motivated by role models
that are congruent with their culturally relevant regula-
tory orientations. Asian Canadians, who are more likely
 Lockwood et al. / ROLE MODELS ACROSS CULTURES
8 ingroup member but not by a successful ingroup mem-
7.5 Positive Model
7 Negative Model
6.5
Relevance
6
5.5
5 it seems unlikely that the different reactions of Asian
4.5
4
3.5
3 hypothesis that Asian Canadians are more motivated by
East Asian Western European
Cultural Background
Figure 4 Asian Canadian and European Canadian participants’ rat-
ings of the role model’s relevance (Study 2).
to have strong prevention concerns, were motivated by
negative models, who highlighted their preferred strat-
egy of avoiding failure. European Canadians, who are
more likely to have strong promotion concerns, were
motivated by positive models, who emphasized their
favored strategy of pursuing success. Models incongru-
ent with an individual’s culturally relevant regulatory
focus had no impact on motivation.
Consistent with our hypotheses, neither self-esteem
nor academic performance accounted for different
reactions to role models by the different cultural groups.
Thus, it seems unlikely that Asian Canadians are moti-
vated only by negative models simply because they view
their own abilities and future prospects negatively or that
European Canadians are motivated only by positive
models because they view themselves highly positively
and expect to excel in the future. Instead, it appears that
cultural differences in regulatory strategies determine
the motivating impact of role models.
It could be argued that Asian Canadian participants
responded differently to the role models because of
their status as a minority group; they may have viewed the
success or failure of an ingroup member differently than
would European Canadians. Indeed, in one study, Black
female participants viewed their own performance more
negatively following exposure to an unsuccessful Black
woman than a successful Black woman; the performance
of the other confirmed or refuted the stereotype of Black
Americans as academically unsuccessful and conse-
quently had implications for participants’ own abilities
(Blanton, Crocker, & Miller, 2000). It seems unlikely,
however, that such ingroup effects could account for the
pattern of results obtained in Study 2; it is not clear why
Asian Canadians would be motivated by an unsuccessful
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
389
ber. Indeed, one might expect that to the extent the
target is viewed as particularly relevant because of shared
ethnic identity, both the positive and negative role mod-
els would exert a strong impact on motivation. This was
not the case. Moreover, in Study 1, in which no informa-
tion about the role models’ cultural background was
provided, we obtained a similar pattern of results. Thus,
Canadian and European Canadian participants to the
positive and negative role model can be explained by the
model’s status as a minority ingroup member among
Asian participants. Instead, these results support our
negative role models because they have especially strong
prevention concerns.
GENERAL DISCUSSION
Taken together, these studies provide strong evidence
that cultural differences in regulatory orientation play a
key role in determining the motivating impact positive
and negative role models. Individuals from East Asian
cultures that emphasize the goal of avoiding undesir-
able outcomes will be especially motivated by negative
role models. Individuals from Western European cul-
tures that emphasize the goal of promoting positive
outcomes will be especially motivated by positive role
models. Individuals with more interdependent self-
construals are concerned about their duties and obliga-
tions and are especially motivated to avoid situations that
might lead to a failure to live up to these responsibilities;
negative role models provide a cogent example of such a
failure and highlight strategies for avoiding a similarly
unpleasant outcome. Individuals with more independ-
ent self-construals are concerned with achieving success
and enhancing the self and consequently are especially
motivated to pursue activities that could assist them in
attaining positive outcomes; positive role models pro-
vide a vivid example of such success and highlight strate-
gies for achieving such an outcome.
In these studies, we examined the impact of role mod-
els who represented success or failure without providing
information about the source of the target’s outcome. In
his model of regulatory focus, Higgins (1997; Higgins
et al., 2001) argued that regulatory focus may be charac-
terized by specific strategic means: Promotion-focused
individuals’ goal pursuit will be characterized by eager-
ness; prevention-focused individuals’ goal pursuit will be
characterized by vigilance. It may be the case that indi-
viduals will be especially motivated by a role model when
the model’s success or failure arises from the presence or
absence of the specific strategic means that fits the indi-
viduals’ regulatory orientation (Higgins, 2000). For
example, East Asians may be particularly motivated by a
390 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
negative role model whose unfortunate circumstances
can be traced to carelessness, a failure to exercise cau-
tion. A worse-off student who fails because he or she
spends too much time partying will likely be more moti-
vating than a worse-off student who has failed due to a
lack of enthusiasm for his or her studies. The imprudent
model provides more information about the kinds of
behaviors that one should be vigilant to avoid to prevent
a similarly negative outcome. In contrast, North Ameri-
cans, who are more concerned with promotion, may be
especially motivated by a positive model who has
achieved success through eagerness. A go-getter who has
become rich through his or her enthusiastic pursuit of
success may thus be more motivating than someone who
has achieved financial security by carefully saving and
protecting his or her money. The go-getter provides
more useful information on the eagerness-related strate-
gies that should be adopted to promote a similarly posi-
tive outcome. Thus, the source of the negative or positive
role models’ outcome as well as the valence of that out-
come may be important in determining the degree to
which the model is motivating; individuals with more
interdependent self-construals may be especially moti-
vated by negative models who have failed due to insuffi-
cient vigilance, and individuals with more independent
self-construals may be especially motivated by positive
models who have achieved success through eagerness.
These studies have important implications for
research on social comparison. Traditionally, theorists
have argued that individuals use upward comparisons to
more successful others to serve self-improvement goals
but use downward comparisons to less successful others
to serve self-enhancement goals (Taylor & Lobel, 1989;
Wills, 1981; Wood, 1989). A more successful other pro-
vides information on the achievements for which one
can strive and can motivate one to adopt behaviors con-
ducive to attaining similar success in the future (e.g.,
Lockwood & Kunda, 1997). A worse-off other, in con-
trast, provides evidence of one’s relative success, allow-
ing one to bask in one’s superiority over another individ-
ual (Wills, 1981). The present research, however,
suggests that the functions of upward and downward
comparisons may vary across cultures. For example, indi-
viduals from Eastern cultures may use downward com-
parisons to serve self-improvement rather than self-
enhancement needs. Rather than enjoying their superi-
ority over the less successful individual, East Asians may
use the example of the other as a means of identifying
and correcting their own shortcomings. Recent evi-
dence suggests that downward comparisons can also be
motivating for North Americans but only when they are
forced to reflect on how they might become like the
other in the future (Lockwood, 2002). Moreover,
although such a downward comparison can enhance
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
motivation among North Americans, the process of
imagining a self like a worse-off other can be highly
threatening to the self (Wood & VanderZee, 1997);
indeed, the activation of avoidance motivation may
reduce subjective well-being among individuals with
strongly independent self-construals (Elliot et al., 2001).
Among Asians however, we might expect downward
comparisons to be motivating without threatening well-
being; because the process of avoiding the role model’s
negative fate is consistent with the culturally emphasized
prevention orientation, the subjective well-being of indi-
viduals with strongly interdependent self-construals will
likely be unaffected or possibly even enhanced by an
encounter with a negative role model (cf. Elliot et al.,
2001).
By demonstrating that East Asian and European Can-
adians differ in the degree to which they perceive posi-
tive and negative role models to be motivating and in the
extent to which positive and negative models activate
plans to engage in specific behaviors, these studies take
an important first step in identifying cultural differences
in responses to role models. Considerable evidence sug-
gests that specific behavioral intentions predict actual
planned behaviors (e.g., Ajzen, 1991, 1996; Ajzen &
Madden, 1986). Nevertheless, the relationship between
intentions and behaviors is far from perfect. In future
research, it will be important to examine the long-term
behavioral consequences of exposure to role models
among different cultural groups.
In this research, we have emphasized cultural differ-
ences in responses to positive and negative role models.
It is important to note, however, that situational fluctua-
tions in self-construals can also affect reactions to suc-
cessful and unsuccessful others (cf. Stapel & Koomen,
2001). At times, East Asians may become more inde-
pendent and North Americans may become more inter-
dependent. For example, when North Americans are in
a situation that emphasizes the interdependence rather
than independence of the self, such as participating in a
team sporting event, their prevention orientation is
enhanced (Lee et al., 2000); in such situations, they may
be especially motivated by negative models. Similarly,
when East Asians are asked to think about ways in which
they are different from friends and family members, they
exhibit stronger independent self-construals (Trafimow,
Triandis, & Goto, 1991); when independent selves are
highlighted, Asians may become more promotion
focused and consequently may be inspired by positive
models. To maximize the effectiveness of role models,
one should consider both the chronically accessible self-
construals of different cultural groups and the situa-
tional cues that can be used to activate promotion or
prevention orientations in a target audience.
 Lockwood et al. / ROLE MODELS ACROSS CULTURES
NOTES
1. In line with our hypotheses, we expected that self-construals
would at least partially mediate the relationship between culture and
regulatory focus. However, we also included direct paths from culture
to regulatory focus. To the extent that one’s culture emphasizes a par-
ticular regulatory orientation, one may learn to adopt this orientation
more generally, independent of one’s self-construals. For example, a
cultural focus on prevention may arise initially as a result of a cultural
emphasis on interdependence but may lead to an overall cultural con-
cern with self-criticism and the consideration of possible failure experi-
ences. As a result, individuals may become prevention oriented partly
as a result of their personal interdependence concerns and partly
because they have internalized more general, culturally emphasized
prevention values. To allow for this possibility, we included in the
model direct paths from culture to both prevention and promotion
focus as well as the predicted indirect paths mediated through self-
construals.
2. We note that Asian Canadian and European Canadian partici-
pants did not differ significantly in their ratings of the motivating
impact of positive role models; thus, for this variable, the first condition
in Baron and Kenny’s (1986) test for mediation is not met. How-
ever, recent analyses of mediation (i.e., MacKinnon, Lockwood,
Hoffman, West, & Sheets, 2002; Shrout & Bolger, 2002) suggest that
this step is not necessary when effect sizes are expected to be small. As
noted earlier, because the East Asians in our sample had been exposed
to promotion-oriented North American culture, it seemed likely that
they would perceive positive models to be at least somewhat motivating
and that differences between the two cultural groups on this measure
would therefore be small. Accordingly, we tested the lower as well as the
upper chain depicted in Figure 1.
3. In a comparison of several proposed tests of mediation,
MacKinnon et al. (2002) found that mediation can best be tested by
showing the statistical significance of both the path to and the path
from the hypothesized mediator. According to this test, both interde-
pendent self-construal and prevention focus qualify as mediators.
4. Promotion focus does not quite meet the test for mediation advo-
cated by MacKinnon et al. (2002) due to the marginal significance of
one of the paths.
5. Self-esteem scores were available for only 71 of the participants.
Accordingly, analyses involving self-competence and self-liking were
conducted using this reduced sample.
REFERENCES
Ajzen, I. (1991). The theory of planned behavior. Organizational
Behavior and Human Decision Processes, 50, 179-211.
Ajzen, I. (1996). The directive influence of attitudes on behavior. In P.
M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action: Linking
cognition and motivation to behavior (pp. 385-403). New York:
Guilford.
Ajzen, I., & Madden, T. J. (1986). Prediction of goal-directed behav-
ior: Attitudes, intentions, and perceived behavioral control. Jour-
nal of Experimental Social Psychology, 22, 453-474.
Arbuckle, J. L., & Wothke, W. (1999). Amos 4.0 user guide. Chicago:
SmallWaters Corporation.
Aspinwall, L. (1997). Future-oriented aspects of social comparison: A
framework for studying health-related comparison activity. In B. P.
Buunk & F. X. Gibbons (Eds.), Health, coping, and well-being: Per-
spectives from social comparison theory (pp. 125-165). Hillsdale, NJ:
Lawrence Erlbaum.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator vari-
able distinction in social psychological research: Conceptual, stra-
tegic, and statistical considerations. Journal of Personality and Social
Psychology, 51, 1173-1182.
Blanton, H., Crocker, J., & Miller, D. T. (2000). The effects of in-group
versus out-group social comparison on self-esteem in the context
of a negative stereotype. Journal of Experimental Social Psychology, 36,
519-530.
Buunk, B. P., Collins, R. L., Taylor, S. E., Van Yperen, N. W., & Dakof,
G. A. (1990). The affective consequences of social comparison:
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
391
Either direction has its ups and downs. Journal of Personality and
Social Psychology, 59, 1238-1249.
Collins, R. L. (1996). For better or worse: The impact of upward social
comparisons on self-evaluations. Psychological Bulletin, 119, 51-69.
Elliot, A. J., Chirkov, V. I., Kim, Y., & Sheldon, K. M. (2001). A cross
cultural analysis of avoidance (relative to approach) personal
goals. Psychological Science, 12, 505-510.
Goethals, G. R., & Darley, J. M. (1977). Social comparison theory. In J.
M. Suls & R. L. Miller (Eds.), Social comparison processes: Theoretical
and empirical perspectives (pp. 259-278). Washington, DC:
Hemisphere.
Hakmiller, K. L. (1966). Need for self-evaluation, perceived similarity
and comparison choice. Journal of Experimental Social Psychology,
(Suppl. 1), 49-54.
Heine, S. J., Kitayama, S., Lehman, D. R., Takata, T., Ide, E., Leung,
C., et al. (2001). Divergent consequences of success and failure in
Japan and North American: An investigation of self-improving
motivation and malleable selves. Journal of Personality and Social
Psychology, 81, 599-615.
Heine, S. J., & Lehman, D. R. (1997). The cultural construction of
self-enhancement: An examination of group-serving biases. Jour-
nal of Personality and Social Psychology, 72, 1268-1283.
Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is
there a universal need for positive self-regard? Psychological Review,
106, 766-794.
Heine, S. J., Takata, T. L., & Lehman, D. R. (2000). Beyond self-
presentation: Evidence for self-criticism among Japanese. Person-
ality and Social Psychology Bulletin, 26, 71-78.
Higgins, E. T. (1997). Beyond pleasure and pain. American Psycholo-
gist, 52, 1280-1300.
Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as
a motivational principle. Advances in Experimental Social Psychology,
30, 1-46.
Higgins, E. T. (2000). Making a good decision: Value from fit. Ameri-
can Psychologist, 55, 1217-1230.
Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O.
N., & Taylor, A. (2001). Achievement orientations from subjective
histories of success: Promotion pride versus prevention pride.
European Journal of Social Psychology, 31, 3-23.
Higgins, E. T., & Tykocinski, O. (1992). Self-discrepancies and bio-
graphical memory: Personality and cognition at the level of psy-
chological situation. Personality and Social Psychology Bulletin, 18,
527-535.
Kitayama, S., Markus, H. R., Matsumoto, H., & Norasakkunkit, V.
(1997). Individual and collective processes of self-esteem man-
agement: Self-enhancement in the United States and self-
depreciation in Japan. Journal of Personality and Social Psychology,
72, 1245-1267.
Lee, A. Y., Aaker, J. L., & Gardner, W. L. (2000). The pleasures and
pains of distinct self-construal: The role of interdependence in
regulatory focus. Journal of Personality and Social Psychology, 78,
1122-1134.
Lockwood, P. (2002). Could it happen to you? Predicting the impact
of downward comparisons on the self. Journal of Personality and
Social Psychology, 82, 343-358.
Lockwood, P., Jordan, C. H., & Kunda, Z. (2002). Motivation by posi-
tive or negative role models: Regulatory focus determines who will
best inspire us. Journal of Personality and Social Psychology, 83, 854-
864.
Lockwood, P., & Kunda, Z. (1997). Superstars and me: Predicting the
impact of role models on the self. Journal of Personality and Social
Psychology, 73, 91-103.
Lockwood, P., & Kunda, Z. (1999). Salience of best selves undermines
inspiration by outstanding role models. Journal of Personality and
Social Psychology, 76, 214-228.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., &
Sheets, V. (2002). A comparison of methods to test mediation and
other intervening variable effects. Psychological Methods, 7, 83-104.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implica-
tions for cognition, emotion, and motivation. Psychological Review,
98, 224-253.
392 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
McDonald, R. P., & Ho, M.-H. R. (2002). Principles and practice in
reporting structural equation analyses. Psychological Methods, 7, 64-
82.
Shah, J., Higgins, E. T., & Friedman, R. (1998). Performance incen-
tives and means: How regulatory focus influences goal attain-
ment. Journal of Personality and Social Psychology, 74, 285-293.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and
nonexperimental studies: New procedures and recommenda-
tions. Psychological Methods, 7, 422-445.
Singelis, T. M. (1994). The measurement of independent and inter-
dependent self-construals. Personality and Social Psychology Bulletin,
20, 580-591.
Stapel, D. A., & Koomen, W. (2001). I, we, and the effects of others on
me: How self-construal level moderates social comparison effects.
Journal of Personality and Social Psychology, 80, 766-781.
Tafarodi, R. W., & Swann, W. B. (1995). Self-liking and self-
competence as dimensions of global self-esteem: Initial validation
of a measure. Journal of Personality Assessment, 65, 322-342.
Taylor, S. E., & Lobel, M. (1989). Social comparison activity under
threat: Downward evaluation and upward contacts. Psychological
Review, 96, 569-575.
Taylor, S. E., Wayment, H. A., & Carillo, M. (1996). Social compari-
son, self-regulation, and motivation. In R. M. Sorrentino & E. T.
Downloaded from psp.sagepub.com at CARLETON UNIV on May 11, 2015
Higgins (Eds.), Handbook of motivation and cognition (pp. 3-27).
New York: Guilford.
Trafimow, D., Triandis, H. C., & Goto, S. G. (1991). Some tests of the
distinction between the private self and the collective self. Journal
of Personality and Social Psychology, 60, 649-655.
Triandis, H. C. (1989). The self and social behavior in differing cul-
tural contexts. Psychological Review, 96, 506-520.
Wheeler, L. (1966). Motivation as a determinant of upward compari-
son. Journal of Experimental Social Psychology, (Suppl. 1), 27-31.
Wills, T. A. (1981). Downward comparison principles in social psy-
chology. Psychological Bulletin, 90, 245-271.
Wood, J. V. (1989). Theory and research concerning social compari-
sons of personal attributes. Psychological Bulletin, 106, 231-248.
Wood, J. V., & VanderZee, K. (1997). Social comparisons among can-
cer patients: Under what conditions are comparisons upward and
downward? In B. P. Buunk & F. X. Gibbons (Eds.), Health, coping,
and well-being: Perspectives from social comparison theory (pp. 299-
328). Hillsdale, NJ: Lawrence Erlbaum.
Received June 10, 2002
Revision accepted June 7, 2004