Content uploaded by Daphna Oyserman
Author content
All content in this area was uploaded by Daphna Oyserman on Aug 05, 2015
Content may be subject to copyright.
Social Cognition, Vol. 33, 2015, pp. 1–18
1
© 2015 Guilford Publications, Inc.
Thanks go to NSF for funding (Smith), the Culture and Self Lab for feedback, to the USC Hispanic
Alumni Association for funding our research assistants Reinhold C. Schrader II and Liza Chavac
for data collection and data entry for Study 1a, to Cristina Aelenei for supervising data collection,
data entry, and conducting Study 1a data analysis, to Tess Levy, Jerri Bamberger, Wendy Cortes, and
Michael Joiner-Hill for data collection and entry for Studies 1b and 2.
Address correspondence to Daphna Oyserman, Department of Psychology, University of Southern
California, SGM 501 3620 South McClintock Ave., Los Angeles, CA 90089-1061; E-mail: Oyserman@
usc.edu.
SMITH AND OYSERMAN
JUST NOT WORTH MY TIME
JUST NOT WORTH MY TIME? EXPERIENCED DIFFICULTY
AND TIME INVESTMENT
George C. Smith
University of Michigan
Daphna Oyserman
University of Southern California
Students often fail to devote sufcient time to schoolwork even though
they value school success. One reason may be they (mis)interpret what
experienced difculty with schoolwork implies because they misgauge
their relative standing. To test this prediction we divided students into four
guided-recall groups. For half, the recall was a time that they interpreted
experienced difculty with schoolwork as meaning that it was important to
succeed and for half the recall was a time that it meant it was impossible
to succeed. Students were then led to believe that they had the guided
interpretation more or less frequently than others. Students in the difculty
means importance more for oneself than for others and in the difculty
means impossibility less for oneself than for others conditions were more
academically engaged (Study 1) and invested more time (Study 2). Invest-
ment mattered, inuencing performance on a test of uid intelligence
(Study 2).
American students aspire to get good grades and succeed in college but their
attainments often fall short (Rosenbaum, Deil-Amen, & Person 2006; Symonds,
Schwarz, & Ferguson, 2011; Trusty, 2000). While a variety of barriers related to so-
cial class, race-ethnicity, and gender have been identified as reasons for underper-
2 SMITH AND OYSERMAN
formance (e.g., Jackson, 2010; Orfield, Losen, Wald, & Swanson, 2004; Steele, 1997),
these factors do not address underperformance among non-stereotyped groups
and do not explicitly focus on time allocation. Yet time use analyses suggest that
time use matters. American students relegate studying to only about 14 hours a
week, less than half of the recommended amount (Babcock & Marks, 2010). Spend-
ing enough time on academics is critical for school success (Allensworth & Easton,
2007; Arum, Roksa, & Cho, 2011; Astin, 1993; Kuh, Kinzie, Buckley, Bridges, &
Hayek, 2006; Pascarella & Terenzini, 2005). Students who report investing more
time on academics earn higher salaries later on, even controlling for the effect of
time investment on their grades while in school (Babcock & Marks, 2010). Why
might students under-invest in academics? To address this question, in the current
studies we start with the assumption that fully engaging in schoolwork is often
experienced as difficult and that students can interpret their experienced difficulty
as implying that schoolwork is important to them but also that it is impossible
to attain. We build on prior research using identity-based motivation theory and
focus explicitly on the effect of implied social context in driving the motivational
consequences of students’ interpretation of their experienced difficulty.
IDENTITY-BASED MOTIVATION
Identity-based motivation theory (IBM) predicts that although people prefer to
act in ways that are congruent with their identities, the identity-to-behavior link
is often opaque because identities are situated (Oyserman, 2007, 2015). Situations
influence which identity comes to mind, what a salient identity means in the mo-
ment, whether strategies to work toward salient identities feel identity congruent,
and how difficulty engaging in these strategies is interpreted. The power of situa-
tions is a joint consequence of the richness of self-concept content, the looseness of
self-concept structure, and the pragmatic nature of human cognition. One’s self-
concept consists of an array of disjointed identities rather than as an integrated
unit (e.g., Markus & Kunda, 1986; Oyserman, Elmore, & Smith, 2012; Swann &
Bosson, 2010). Thinking is for doing, implying that how one considers an object,
including the self, is dependent on what seems possible in the situation (e.g., Oy-
serman et al., 2012; Smith & Semin, 2004).
Viewed in this way, what is often termed self-consistency really involves behav-
ing in a way that is consistent with a particular identity; it is not possible to act con-
sistently with all the identities included in one’s sense of self. Whether an action
appears consistent depends on the identity to which it is matched and what that
identity means in the situation. For example, Shih, Pittinsky, and Trahan (2006)
showed that verbal task performance of Asian heritage female students matches
stereotypes about women and Asians. Performance was higher if being female
was on their mind and lower if being Asian was on their mind. Going beyond
stereotype activation, IBM predicts that the effect of considering one’s gender is
dependent on implications of the situation. Indeed, Elmore and Oyserman (2012)
showed that students’ performance on a math task depended on what the imme-
JUST NOT WORTH MY TIME 3
diate context implied about their gender. Males performed worse, reported fewer
academic possible selves and lower long-term occupational and financial aspira-
tions when asked to interpret a census graph showing males underperforming
females (fewer graduate high school); the pattern reversed if the graph showed
males outperforming females (higher average salary). The effect is not due to a
pre-established stereotype, but to the implications of the situation for what being
male means.
IBM takes the next step, predicting that the same behavior may feel identity-
congruent, consistent with a salient identity, or identity-incongruent, inconsistent
with that identity, depending on how difficulty engaging in the behavior is inter-
preted in context. An initial test of this prediction was conducted as a school-based
intervention. Low-income students were randomly assigned to either a 7-week
(11-class period) intervention or a “school as usual” control group and their aca-
demic outcomes tracked over time (Oyserman, Bybee, & Terry, 2006). While all
students experienced the usual difficulties associated with being an eighth grader,
intervention group students participated in small-group activities designed to ac-
tivate an interpretation of this difficulty as a normative part of working toward
one’s academic possible identities. Control group students went to class as usual
and received no structured interpretation of their experienced difficulties. Content
of possible identities, school grades, attendance, homework time, and in-class be-
havior were obtained for both groups. The two groups did not differ on any mea-
sure prior to intervention. Post intervention, students in the intervention group
were more likely to report that being successful in school was a possible identity
and that they had strategies to achieve that identity. They also spent more time
on their homework, even a year after the intervention. Their high school teachers
reported that they were more engaged in classroom activities, and their schools
reported fewer skipped classes. The effect of the intervention on grades, atten-
dance, and academic engagement was mediated by an increase in school-focused
possible identities and strategies to attain them.
The intervention demonstrated that interpretation of experienced difficulty
influences content of identity and that this has consequences for behavioral en-
gagement. However, it did not directly test the effect of relative social standing
on the implications drawn from salient interpretations of experienced difficulty.
As detailed next, social comparisons can be an informational source for what an
interpretation of experienced difficulty means for one’s self.
INTERPRETING EXPERIENCE
Social context is a rich informational source, providing explanations for one’s ex-
periences (Festinger, 1954; Weiner, 1985). People routinely, automatically, and non-
consciously use others as standards of comparison to inform themselves about
their own abilities, interests, and desires (e.g., Cialdini & Trost, 1998; Smith & Col-
lins, 2009). Schwarz and colleagues have generated a large body of evidence rele-
vant to this point (Schwarz, Bless, Bohner, Harlacher, & Kellenbenz, 1991; Schwarz,
4 SMITH AND OYSERMAN
Bless, Strack, Klumpp, Rittenauer-Schatka, & Simons, 1991; Schwarz, Groves, &
Schuman, 1998). Their work shows that people are sensitive to subtle information
about the frequency with which they experience something relative to others and
infer from response scale options what the normal distribution of an experience
or behavior is. If the scale includes high-frequency options, the experience or be-
havior is common. If it includes low-frequency options, the experience or behavior
is uncommon. The scale influences people’s responses to the question and also
influences their subsequent interpretation of what their own behavior implies. As
detailed next, the same pattern of behavior can lead to different conclusions about
the self, depending on the perceived behavioral frequency of others.
For example, Schwarz and Scheuring (1988) led German adults to believe that
they masturbated either more or less often than others. Compared to those in the
less condition, those in the more condition later reported lower marital satisfac-
tion presumably because they inferred that they must not be satisfied with their
marital life given what their masturbation rate implied about their unmet sexual
needs compared to others. Similarly, intentions to use condoms increased for col-
lege students led to believe that others had had more sexual partners than they
had, presumably because they inferred they would otherwise be putting them-
selves at risk given others’ promiscuity (Rothman, Haddock, & Schwarz, 2001).
Across studies effects were found by having people rate the frequency of their own
behavior on a scale that manipulated their sense of typical behavioral frequency. If
scale frequency was high, they inferred that their personal frequency was less than
average and the reverse if scale frequency was low. In this way, biased frequency
scales provide information on what the experience of others tends to be, informing
what that experience means for the self (Deutsch & Gerard, 1955).
INTERPRETING EXPERIENCED DIFFICULTY
People are likely to seek an explanation for their experiences of difficulty because
difficulty implies that current energy investment is insufficient. Should investment
go up to overcome difficulty or go down so that energy can be used elsewhere?
From an evolutionary perspective, both interpretations of difficulty, as a sign of
importance or as a sign of impossibility, are logical (Charnov, 1976; Nesse, 2009).
If experienced difficulty is interpreted as a signal of identity-congruence and task
importance then effort should be sustained and even increased in the face of dif-
ficulty so that opportunities for success are not missed. If experienced difficulty is
interpreted as a signal of identity-incongruence and task impossibility then effort
should be channeled away from the unattainable goal so that resources are not
wasted and can be used to attain another goal. Thus, sensitivity to experienced
difficulty and sensitivity to social and nonsocial cues as to how to interpret expe-
rienced difficulty are likely to be rooted in evolutionary necessities to both engage
and disengage.
In psychology, the idea that difficulty can increase the intensity of motivation
has been discussed at least since William James (1890) and Ach, who discussed it
JUST NOT WORTH MY TIME 5
in terms of the will to overcome distraction (as discussed in Brehm & Self, 1989).
The role of difficulty in influencing belief in one’s abilities to succeed (see self-
efficacy theory; Bandura 1988, 1997), in altering expectations for the likelihood of
success (see expectancy-value theories; Atkinson, 1966, 1974; Eccles et al., 1983;
Feather 1982, 1992; Wigfield & Eccles, 1992), and in impacting motivation (see
Brehm & Self, 1989; Carver & Scheier, 1998) have all been studied. Difficulty in-
creases motivation and the desirability of a goal so long as difficulty is not so great
as to render tasks impossible; in this case, effort quickly declines (e.g., Brehm &
Self, 1989; Roese & Olson, 2007; Silvestrini & Gendolla, 2013). While difficulty is
typically assumed to be a feature inherent to the task or goal, a situated approach
focuses on the effects of context on how difficulty is interpreted. For example,
Oyserman, Destin, and Novin (2014) showed that interpretation of experienced
difficulty with schoolwork depended how the college context and the future self
were considered. Students led to consider their desired future self and to imagine
college as a success-likely context were more likely to endorse an interpretation of
difficulty with schoolwork as implying that schoolwork was for them and worth
the effort, so were students led to consider their undesired future self and to imag-
ine college as a failure-likely context.
THE CURRENT STUDIES
Research to date implies but does not specifically test that the experience of others
can be used as an interpretive cue to help understand what one’s own experiences
of difficulty with schoolwork imply. In the current studies we tested the predic-
tion that students will be more engaged and invested in their schoolwork in two
circumstances. First, if students are led to recall a time that they experienced diffi-
culty engaging in school tasks as a signal of importance and are led to believe that
they have this interpretation more frequently than their peers. Second, if they are
led to recall a time that they experienced difficulty engaging in school tasks as a
signal of impossibility and are led to believe that they have this interpretation less
frequently than their peers. We use a 2 × 2 design manipulating how experienced
difficulty is interpreted (importance, impossibility) and relative frequency (high,
low). The interpretation manipulation follows Oyserman and colleagues (e.g.,
Oyserman et al., 2014) and the frequency manipulation follows Schwarz and col-
leagues (e.g., Rothman et al., 2001). High-frequency scales in which one’s standing
is low relative to others should imply that the experience is identity-incongruent.
In contrast, low-frequency scales in which one’s standing is high relative to others
should imply that the experience is identity-congruent. Thus, students asked to re-
port on their experience of difficulty at school on a low-frequency scale should be
led on a biased memory search yielding the conclusion that they have the experi-
ence of difficulty more often than others. This should be motivating if their biased
search is for times difficulty implied importance, and demotivating otherwise. In
the same way, students asked to report on their experience of difficulty at school
on a high-frequency scale should be led on a biased memory search yielding the
6 SMITH AND OYSERMAN
conclusion that they have the experience of difficulty less often than others. This
should be motivating if their biased search is for times difficulty implied impos-
sibility, and demotivating otherwise.
STUDY 1A
SAMPLE AND METHOD
Students attending summer classes at area colleges (University of Southern Cali-
fornia, Glendale College, College of the Canyons, Santa Monica College, N = 121,
53 women, 68 men) were approached on campus and asked to participate in a
1-page study. Unbeknownst to participants, the front of the page was the experi-
mental manipulation and the questions on the back of the page were the depen-
dent measures plus two demographic controls. Questionnaires were randomized
prior to distribution. We planned to obtain 30 participants per cell and collected
data until that goal had been reached.
On the front of the page was the text: Experiencing difficulty working on a school
task can be thought of as signaling importance [impossibility], that what you are working
on is [not] worth your effort because it is important to [it is not for] you. This can be a
common occurrence for students. How often have you had the feeling of difficulty in the
past month? In the low frequency relative to others condition, the response scale
ranged from ≤ 10 times to ≥ 31. In the high frequency relative to others condition
the response scale ranged from 1–2 times to ≥ 11 times (see Figure 1). We chose this
range because experiencing difficulty working on a school task could happen from
once a month to more than daily.
On the back of the page were four questions about academic engagement and
identity, gender, and year in school. A factor analysis of the four items (standard-
ized with varimax or oblim rotation) yielded a single factor so we took the mean of
the four standardized items in our subsequent analysis of the identity congruence
of academics (α = 0.525). The exact wording of each item, response scales, item
means, and standard deviations are presented in Table 1.
Manipulation Check. Our frequency manipulation worked as expected. Partici-
pants who gave an answer reported more frequent experience of difficulty if they
FIGURE 1. Frequency manipulation used in Studies 1a, 1b, and 2. The scale used in the low
relative to others condition is on the left and the scale used in the high relative to others
condition is on the right.
JUST NOT WORTH MY TIME 7
were in the high relative to others rather than the low relative to others condition,
F(1, 118) = 9.03, p = .003.
Examination of Demographics. Preliminary analyses showed that gender, F(1, 119)
= 8.763, p = .004, but not of year in school (p = .606) mattered, women scored higher
on the identity congruence of academics. Gender is a covariate in the analyses
reported below.
RESULTS AND DISCUSSION
Analyses of covariance (ANCOVA) controlling for gender revealed the predicted
interaction between interpretation of experienced difficulty condition and relative
frequency condition, F(1, 116) = 3.141, p = .079, d = 0.32, CI [.091, .708], that modi-
fied the effect of frequency, F(1, 116) = 3.293, p = .072, d = 0.33, CI [-.025, .693], and
interpretation of experienced difficulty, F < 1, p = .456, conditions. As can be seen
in Figure 2, being reminded that difficulty can be interpreted as importance bol-
stered the identity congruence of academics for students led to believe that they
experienced this interpretation more frequently than others and undermined it
for students led to believe they experienced this interpretation less than others.
Academics is experienced as more identity congruent in the two motivating in-
terpretations of difficulty conditions (M = 0.105, SD = 0.86) compared to the two
undermining conditions (M = -.100, SD = 0.88), at trend level, F(1, 118) = 3.35, p =
.07, d = .34, CI [-.023, .696],1 controlling for gender.
Decomposing the interaction into simple effects and still controlling for gender,
we find an effect of frequency scale for participants in the importance condition,
F(1, 59) = 5.713, p = .02, d = .62, CI [.109, 1.130], but not the impossibility condition
TABLE 1. Items Used in Studies 1a and 1b as Academic Identity Engagement
Variable Scale Study 1a Study 1b
M SD M SD
Doing well in classes in my major is important
to me
1 = strongly
disagree, 6 =
strongly agree
5.39 .88 5.51 .67
Doing well in classes outside my major is
important to me
1 = strongly
disagree, 6 =
strongly agree
4.60 1.13 4.91 .89
How likely are you to skip going out/
socializing this weekend to prepare for class
1 = not at all likely,
6 = very likely
3.88 1.37 3.86 1.42
Realistically, how many hours do you plan to
study tonight
Open-endeda2.16 1.36 3.26 1.56
Note. aThe range of responses to the open-ended study question was 0 to 6 in Study 1a and 0 to 8 in Study 1b.
1. The reader will note that degrees of freedom increases in this analysis because rather than two
factors (interpretation of difficulty, relative frequency), this analysis uses a single factor coded at -1
difficulty means importance less often for the self than others, difficulty means impossibility more
often for the self than others, and +1 difficulty means importance more often for the self than others,
difficulty means impossibility less often for the self than for others.
8 SMITH AND OYSERMAN
(F < 1). Effects are in the predicted direction but not significant when the effect of
interpretation of difficulty in the low, F(1, 60) = 2.23, p = .14, d = .38, CI [-.114, .884],
and high, F(1, 55) = 1.01, p = .32, d = .27, CI [-.247, .787], frequency conditions are
examined separately.
Though providing initial support for the predicted effect of interpretation of
experienced difficulty, effects are weaker than expected and their interpretation
is made ambiguous by the fact that the probe simply asked how often students
had experienced difficulty in the past month. It did not direct their memory scan
to look for experiences of difficulty interpreted in a manner congruent with the
prime. Hence, it is not clear whether the response reflected the intended biased
scan of memory. To strengthen clarity of causal interpretation, in Study 1b we in-
cluded the biased memory search instruction as detailed below.
STUDY 1B
SAMPLE AND METHOD
University of Michigan undergraduates (N = 104, 63 female, 49 underclassmen)
were approached on campus and asked to participate in a 1-page study using the
same procedure as Study 1a, with the modification of the manipulation so that the
bias scan of memory was tested. The text read: Experiencing difficulty working on a
school task can be thought of as signaling importance [impossibility], that what you are
working on is [not] worth your effort because it is important to [it is not for] you. This can
be a common occurrence for students. How often have you had the feeling of difficulty as
importance [impossibility] in the past month? On the back of the page were the same
four questions as in Study 1a (see Table 1 for means and standard deviations) as
well as gender and year in school. We planned to obtain 25 participants per cell
FIGURE 2. Study 1a: Identity-congruence of academics as a function of interpretation of
experienced difculty with schoolwork and implied frequency this interpretation occurs for
others (controlling for gender effects).
JUST NOT WORTH MY TIME 9
and stopped data collection once that goal had been reached. We again took the
standardized mean of the four items as a rough indicator of identity congruence
of academics.
Manipulation Check. We verified that our manipulation of frequency worked as
expected. Participants reported more frequent experience of difficulty in the high
relative to others frequency condition, F(1, 96) = 3.90, p = .051.2
Examination of Demographics. As is Study 1a, gender (p = .004) but not year in
school (p = .339) mattered for academic identity-congruence so we included gen-
der in the final analyses reported below.
RESULTS AND DISCUSSION
Analyses of covariance (ANCOVA) controlling for gender revealed the predicted
interaction between interpretation of experienced difficulty condition and relative
frequency condition, F(1, 96) = 11.843, p = .001, d = .35, CI [1.267, .340]; there was
no main effect of frequency, F(1, 96) = 1.757, p = .188, d = .27, CI [-.125, .659], or
interpretation of experienced difficulty, F(1, 96) = 1.026, p = .314, d = .20, CI [-.188,
.595], conditions. As can be seen in Figure 3, being reminded that difficulty can
be interpreted as importance bolstered the identity congruence of academics for
students led to believe that they experienced this interpretation more frequently
than others and undermined it for students led to believe they experienced this
interpretation less than others. Academics is experienced as more identity congru-
ent in the two motivating interpretations of difficulty conditions (M = 0.191, SD =
FIGURE 3. Study 1b: Identity-congruence of academics as a function of interpretation of
experienced difculty with schoolwork and implied frequency this interpretation occurs for
others (controlling for gender effects).
2. Six participants did not mark a response on the frequency scale used in the manipulation check,
one each failed to answer the question about importance of classes outside one’s major and the
question about skipping socializing to study, three did not report their planned study hours, and
three did not report their gender, resulting in loss of sample size.
10 SMITH AND OYSERMAN
0.82) compared to the two undermining conditions (M = -.201, SD = 0.82), F(1, 98)
= 11.21, p = .001, d = .68,3 CI [.278, 1.081], controlling for gender.
Simple effects show that controlling for gender, the effect of frequency scale is
significant in the importance condition, F(1, 46) = 8.952, p = .004, d = 0.90, CI [.312,
1.487], and in the same direction but not significant in the impossibility condition,
F(1, 49) = 2.687, p = .108, d = .46, CI [-.086, 1.016]. The effect of interpretation of dif-
ficulty as importance rather than impossibility is significant in the high frequency
condition, F(1, 47) = 9.345, p = .004, d = .90, CI [.316, 1.48], and in the same direction
but not significant in the low frequency condition, F(1, 48) = 3.063, p = .086, d =
0.51, CI [-.052, 1.064].
Thus, taken together the results of Studies 1a and 1b support the prediction that
accessible interpretation of school difficulty influences the identity congruence of
academics if biased recall provides a sense that one’s interpretation is positively
distinctive. That is, academics are experienced as more identity congruent if bi-
ased recall implies that a productive interpretation of difficulty (difficulty means
the task is important) is more common for oneself than for others. In the same way,
academics are experienced as less identity congruent if biased recall implies that
an unproductive interpretation of difficulty (difficulty means the task is impossi-
ble) is more common for oneself than for others. Effects are in the expected pattern
though smaller for the predicted reversal, that recalling times when experienced
difficulty was experienced as impossibility can be motivating if one considers oth-
ers to have had that experience more frequently than others.
The recall question needed to direct attention to the primed interpretation of
difficulty as was done in Study 1b: How often have you had the feeling of difficulty
as importance [impossibility] in the past month? Effects were weaker when this was
omitted as in Study 1a (How often have you had the feeling of difficulty in the past
month?). In Study 2 we retain the wording of Study 1b’s manipulation and move
from self-report to behavior, examining time on task in the difficult problems in
the Raven’s Progressive Matrices test of fluid intelligence.
STUDY 2
Time investment is crucial for success on difficult tasks and we predict that our bi-
ased recall method should influence time investment. Students led to believe that
they experience difficulty in schoolwork as signaling the importance of school-
work more than others (or difficulty in schoolwork as signaling the impossibility
of schoolwork less than others) will actually invest more time on this task. As a
follow-up, we test the assumption that time on task reaps benefits of better per-
formance.
3. The reader will note that degrees of freedom increases in this analysis because rather than two
factors (interpretation of difficulty, relative frequency), this analysis uses a single factor coded at -1
difficulty means importance less often for the self than others, difficulty means impossibility more
often for the self than others, and +1 difficulty means importance more often for the self than others,
difficulty means impossibility less often for the self than for others.
JUST NOT WORTH MY TIME 11
SAMPLE AND METHOD
Introductory Psychology subject pool (N = 292, 169 female, 180 underclassmen4
students were randomized to the same conditions as in Study 1. Students received
course credit for participating in the Qualtrics programmed “difficulty during
the college years” study. Following the experimental manipulation, participants
completed the 12-item Bors and Stokes (1998) short form of Raven’s Progressive
Matrices (RPM; Raven, 1962). The short form RPM is a test of fluid intelligence
(see Conway, Kane, & Engle, 2003; Gray, Chabris, & Braver, 2003) that predicts
performance on the full set of Raven’s items. We recorded average time spent on
each item (M = 36.91 seconds, SD = 18.96 seconds) as well as the solution chosen.
Because time data can require transformation, we checked skewness (.94, SE = .14)
and kurtosis (1.58, SE = .28). Both were within acceptable limits for a normal dis-
tribution so analyses use untransformed (raw) time data. Average time spent and
average accuracy were highly positively correlated, r = .61, p < .01. Demographics
questions followed the dependent measures as in Study 1. We planned to obtain
a relatively large sample size of 70 per cell because our dependent variable was
time on a difficult task and we assumed that the effect of our prime would be small
given the prior results.
Manipulation Check. The manipulation of frequency worked; participants report-
ed more frequent experience of difficulty in the high relative to others frequency
condition compared to the low relative to others frequency condition, F(1, 290) =
6.58, p = .011.
Examination of Demographics. Preliminary analyses showed that both being an
advanced student (i.e., not a freshman), F(1, 289) = 2.72, p = .100, and being male,
F(1, 288) = 6.33, p = .012, were associated with more time spent on the Raven’s
items and that being male was also associated with better performance on the Ra-
ven’s items, F(1, 288) = 6.91, p = .009. Therefore, gender and freshman status were
included as controls in all analyses reported next.
RESULTS AND DISCUSSION
Analyses of covariance (ANCOVA) controlling for gender and freshman status
revealed a significant interaction effect, F(1, 284) = 4.22, p = .040, d = .24, CI [2.828,
15.347], which moderated the main effect of interpretation of difficulty, F(1, 284)
= 4.28, p = .040, d = .25, CI [.015, .477], and the nonsignificant effect of frequency
scale, F(1, 284) = .06, p = .81, d = .03, CI [-.202, .258]. This interaction is depicted
graphically in Figure 4. Time on task was higher in the two motivating interpreta-
tions of difficulty conditions (M = 38.872, SE = 1.52) compared to the two under-
4. One participant ran out of time and did not complete the demographics measures; another
participant did not wish to disclose his or her gender. These participants are not included in analyses
that involve these variables.
12 SMITH AND OYSERMAN
mining conditions (M = 34.570, SE = 1.58), F(1, 286) = 3.856, p = .051, d = .23, CI
[.001, .463]5 controlling for gender and freshman status.
Simple effects show that controlling for gender and freshman status, the effect
of interpretation of difficulty is significant in the high frequency relative to others
condition, F(1, 136) = 7.916, p = .006, d = 0.49, CI [.149, .822]. Students spent more
time on the Raven’s items if they recalled times that they interpreted difficulty
as importance more than others compared to if they recalled times that they in-
terpreted difficulty as impossibility more than others. No effect of interpretation
of difficulty was found in the low relative to others condition (F < 1). The pattern
of effects was in the predicted direction for both of the interpretation of difficulty
conditions. For interpretation of difficulty as impossibility, students in the high
relative to others condition spent less time than students in the low relative to oth-
ers condition, F(1, 141) = 2.865, p = .093, d = .28, CI [-.044, .612]. For interpretation
of difficulty as importance, students in the high relative to others condition spent
more time than students in the low relative to others condition, F(1, 141) = 1.522, p
= .219, d = .21, CI [-.120, .533].
Time on task correlated with task performance (percentage of items correctly
answered) at r = .61, p < .01. Performance was not a direct function of the condi-
tion assignment interaction, F(1, 284) = .04, p = .84. However, we predicted an
indirect effect of condition on performance via time. Specifically, our prediction
was that condition assignment would influence time on task and that time on task
would influence performance. To test this prediction, we compared performance
in the motivating conditions to performance non-motivating conditions as above.
Condition influenced time spent on the Raven’s items (unstandardized b = 4.301),
FIGURE 4. Study 2 Mean Time (seconds) spent on each Raven’s Progressive Matrix Item: Graph
depicts the interaction between interpretation of experienced difculty in schoolwork and
frequency of interpretation relative to others (controlling for gender and class standing).
5. The reader will note that degrees of freedom increases in this analysis because rather than two
factors (interpretation of difficulty, relative frequency), this analysis uses a single factor coded at -1
difficulty means importance less often for the self than others, difficulty means impossibility more
often for the self than others, and +1 difficulty means importance more often for the self than others,
difficulty means impossibility less often for the self than for others.
JUST NOT WORTH MY TIME 13
t(3, 286) = 1.96, p = .051, and time spent on the Raven’s items predicted accuracy
overall (unstandardized b = .078), t(3, 286) = 12.74, p < .001. Using the PROCESS
computational tool to examine indirect simple mediation (Hayes, 2012), we found
the posited indirect effect of condition on accuracy via time. The bootstrap confi-
dence interval for the indirect effect (CI: .0025, .6908) did not contain zero, indicat-
ing mediation for the Raven’s items. Increasing time spent on difficult tasks pays
off in terms of better performance, and whether time is spent on difficult academic
tasks is influenced by how students interpret their experienced difficulty relative
to their peers.
GENERAL DISCUSSION
American college students want to succeed academically, but invest too little time
pursuing academic success (e.g., Arum, Roksa, & Cho, 2011). Structural reasons for
academic disengagement, including stereotype threat (see Steele, 1997) and lack of
economic resources (see Jackson, 2010; Orfield, Losen, Wald, & Swanson, 2004) are
important but do not explain underinvestment in non-stigmatized and non-eco-
nomically disadvantaged students. To understand this larger issue, in the current
article we synthesized social cognition (e.g., Bless & Schwarz, 2010) and identity-
based motivation (Oyserman, 2007, 2009) approaches to predict that social context
influences interpretation of experienced difficulty in the academic domain. That is,
difficulty with schoolwork can be interpreted either as implying that one should
turn one’s attention elsewhere (“schoolwork is impossible for me”) or that one
should increase effort (“schoolwork is important to me”) and students are likely to
have had both interpretations at differing times in the past. Whether having had a
particular interpretation in the past is motivating should depend on what it implies
and the relative frequency one has had this interpretation compared to others.
Indeed, considering times in which experienced difficulty implied that school-
work is important and worth one’s time created a sense that schoolwork was iden-
tity congruent and increased time on task for students led to believe that they had
this interpretation more frequently than others. The reverse was also true. Consid-
ering times in which experienced difficulty implied that schoolwork is impossible
and not worth one’s time created a sense that schoolwork is identity congruent
and increased time on task for students led to believe that they had this interpreta-
tion less frequently than others. Time on task mediated task success. Given that the
task was an intelligence test, results provide a strong demonstration of the socially
contextualized motivational power of interpretation of experienced difficulty.
Effects were stronger if we guided students’ recall and also guided how they
interpreted their relative standing so that their own memory seemed to validate
the impression created by the priming task. Our results demonstrate that students
are sensitive to the experience of others in making sense of their own experiences
of difficulty with schoolwork and that even negative experiences can be positive if
one is doing relatively better than others.
14 SMITH AND OYSERMAN
Our manipulations showed small, significant effects in an important domain.
Like any set of studies, our studies have a number of limitations. First, it would be
useful to test our effects using a variety of operationalizations of academic engage-
ment. For example, we show an effect of interpretation of experienced difficulty
on engagement and investment which appears similar to other research show-
ing effects on executive functioning for students led to believe that school ability
is a malleable skill (Autin & Croizet, 2012). The interplay between these mind-
sets seems to be an important avenue for future research. That said, initial results
suggest that each of these mindsets or interpretations explains unique variance
in students’ efficacy and that endorsement of one set of beliefs is not correlated
with endorsement of the others (Oyserman, 2015; Oyserman et al., 2014). Second,
it would be useful to know the circumstances in which our manipulations yield
lasting or more ephemeral results. Our results show that effects can occur and
awake further examination of conditions in which they are likely to be consistently
brought to mind.
Our results complement a number of contemporary construal models of goal
pursuit (e.g., Kruglanski et al., 2002; Trope & Liberman, 2003; Vallacher & Weg-
ner, 1987). Of particular interest is the integration of our results with the predic-
tions from action identification (Vallacher & Wegner, 1987) and temporal construal
models (Trope & Liberman, 2003). Both models distinguish concrete, low-level
construal (how to do it) from abstract, high-level construal (why do it). Action
identification theory (Vallacher & Wegner, 1987) predicts that to take action, people
need to shift from high-level “why” goals such as doing well in school to lower-
level “how” goals such as spending time on school work. Construal level theory
(Trope & Liberman, 2003; Liberman & Trope, 2008) predicts that goals associated
with high-level “why” construal are perceived to be more important than goals
associated with low-level “how” construal. Goals construed at a high-level feel
meaningful but are temporally distal, so students do not feel constrained in plan-
ning how they will actually attain them (Liberman & Trope, 1998), leading to both
lack of preparation and to overly optimistic estimates about the likelihood of dis-
tal goal attainment (the planning fallacy, Buehler, Griffin, & Peetz, 2010; Gilovich,
Kerr, & Medvec, 1993; Kahneman & Tversky, 1979).
In our studies, experiencing academics as identity-congruent can been seen as
a “why” response and investing more time in schoolwork as a “how” response.
Interestingly, our participants seemed to be both more “why” and more “how” en-
gaged if they were made to believe that they experienced difficulty as importance
more often than others or if they were made to believe that they experienced dif-
ficulty as impossibility less than others. IBM theory predicts that one needs both
a “why” explanation for experienced difficulty (because it is identity-congruent)
and a “how” explanation (by engaging in identity-congruent behaviors) in order
to motivate action. Without a “why” explanation, one might know how to suc-
ceed but not see the importance of doing so for the self (it is not for me); without a
“how” explanation, one might know why to work hard (it is for me) but strategies
to do so will not be readily apparent.
JUST NOT WORTH MY TIME 15
From an evolutionary perspective, it makes sense that people have mechanisms
to encourage investing energy in pursuit of important goals as well as to trigger
turning away from pursuit of impossible goals that are simply out of reach, not
worth the time (Charnov, 1976; Nesse, 2009). In the current studies, we focused
on the first part, triggers that encourage investing energy. It is possible, though,
that when a student sees a task as out of his or her reach because of experienced
difficulty, it would be useful to switch to another task or to another goal entirely.
Quitting one goal could then facilitate moving onto another, though the evidence
suggests that quitting is difficult (Worsch, 2010). Students might use this method
to reframe, if a particular academic goal feels impossible, they could switch to
another or ask themselves what else they need to give up to make their academic
goal possible to attain. Of course it is critical to know when this switching involves
another academic goal and when it involves a goal that competes with academics.
While the current studies were not aimed to answer this question, it is an impor-
tant issue both theoretically and practically. Future work could address this issue,
for example, by priming participants with both academic and social goals, provid-
ing relative standing feedback on both, and seeing how participants respond.
While our studies involved experimental manipulation, they are likely to be ap-
plicable to everyday time investment of students. Our results imply that students
will be sensitive to contextual cues of relative standing. Even something as prosaic
as learning that test results are curved can matter. A curve implies that everyone
experienced difficulty so raw scores cannot be used. The interpretation for those
on the top of the curve is that they experienced difficulty as impossibility less
frequently than others, implying that school engagement is identity-congruent for
them. The interpretation for those at the bottom of the curve is that they experi-
enced difficulty as impossibility more frequently than others, implying that school
engagement may not be identity-congruent for them. These interpretations are
consequential; students at the bottom of the curve may invest less time studying
for the next exam, a proposition that deserves empirical test.
CONCLUSION
If experienced difficulty engaging in schoolwork is (mis)interpreted as signaling
that school success is impossible, then school engagement feels less like a “me”
thing to do and attention shifts. Schoolwork is not worth one’s time. In contrast,
framing difficulty as a signal that school success is important increases students’
sense that school engagement is a “me” thing to do. Then students not only priori-
tize schoolwork, but also spend more time on difficult tasks, increasing the likeli-
hood of success. Taken together, our results highlight that the interpretation of
experienced difficulty matters for school outcomes. Academic engagement and
ultimate success depends on how experienced difficulty is interpreted in light of
the (presumed) interpretive experiences of others.
16 SMITH AND OYSERMAN
REFERENCES
Allensworth, E. M., & Easton, J. Q. (2007). What
matters for staying on-track and graduation
in Chicago public high schools. Chicago:
Consortium on Chicago School Re-
search at the University of Chicago. Re-
trieved from http://ccsr.uchicago.edu/
sites/default/ files/publications/07%
20What%20Matters%20Final.pdf.
Arum, R., Roksa, J., & Cho, E. (2011). Improving
undergraduate learning: “Findings and pol-
icy recommendations from the SSRC-CLA
longitudinal project.” Social Science Re-
search Council. Obtained on-line April
11, 2013, http://www.eric.ed.gov/
PDFS/ED514983.pdf.
Astin, A. W. (1993). What matters in college?
Four critical years revisited. San Fran-
cisco: Jossey-Bass.
Atkinson, J. W. (1966). Mainsprings of achieve-
ment oriented activity. In J. D. Krum-
boltz (Ed.), Learning and the educational
process. Chicago: Rand McNally.
Atkinson, J. W. (1974). Strength of motivation
and efficiency of performance. In J. W.
Atkinson & J. O. Raynor (Eds.), Motiva-
tion and achievement. Washington, DC: V.
H. Winston.
Autin, F., & Croizet, J.-C. (2012). Improving
working memory efficiency by refram-
ing metacognitive interpretation of task
difficulty. Journal of Experimental Psycho-
logy: General, 141(4), 610-618.
Babcock, P., & Marks, M. (2010). Leisure colle-
ge, USA: The decline in student study time.
Washington, DC: American Enterprise
Institute for Public Policy Research.
Bandura, A. (1988). Self-regulation of motiva-
tion and action through goal systems. In
V. Hamilton, G. H. Bower, & N. H. Frijda
(Eds.), Cognitive perspectives on emotion
and motivation (pp. 37-61). New York:
Kluwer Academic.
Bandura, A. (1997). Self-efficacy: The exercise of
control. New York: W. H. Freeman.
Bless, H., & Schwarz, N. (2010). Mental con-
strual and the emergence of assimila-
tion and contrast effects: The inclusion/
exclusion model. In M. P. Zanna (Ed.),
Advances in experimental social psycholo-
gy, Vol. 42 (pp. 319-373). San Diego: Aca-
demic Press.
Bors, D. A., & Strokes, T. L. (1998). Raven’s
Advanced Progressive Matrices (RPM):
Norms for first-year university students
and the development of a short form.
Educational and Psychological Measure-
ment, 58, 382-399.
Brehm, J. W., & Self, E. A. (1989). The intensity
of motivation. Annual Review of Psycho-
logy, 40, 109-131.
Buehler, R., Griffin, D., & Peetz, J. (2010). The
planning fallacy: Cognitive, motivation-
al, and social origins. In M. P. Zanna & J.
M. Olson (Eds.), Advances in experimental
social psychology, Vol. 43 (pp. 1-62). San
Diego: Academic Press.
Carver, C. S., & Scheier, M. F. (1990). Principles
of self-regulation: Action and emotion.
In E. T. Higgins & R. M. Sorrentino
(Eds.), Handbook of motivation and cogni-
tion: Foundations of social behavior (Vol. 2,
pp. 3-52). New York: Guilford.
Carver, C. S., & Scheier, M. F. (1998). On the self-
regulation of behavior. New York: Cam-
bridge University Press.
Charnov, E. (1976). Optimal foraging, the mar-
ginal value theorum. Theoretical Popula-
tion Biology, 9, 129-136.
Cialdini, R. B., & Trost, M. R. (1998). Social in-
fluence: Social norms, conformity and com-
pliance. New York: McGraw-Hill.
Conway, A. R., Kane, M. J., & Engle, R. W.
(2003). Working memory capacity and
its relation to general intelligence. Trends
in Cognitive Sciences, 7(12), 547-552.
Deutsch, M., & Gerard, H. B. (1955). A study
of normative and informational social
influences upon individual judgment.
Journal of Personality and Social Psycholo-
gy, 51, 629-636.
Eccles J. S., Adler, T. F., Futterman, R., Goff, S.
B., Kaczala, C. M., Meece, J. L., & Midg-
ley, C. (1983). Expectancies, values, and
academic behaviors. In J. T. Spence
(Ed.), Achievement and achievement mo-
tivation: Psychological and sociological ap-
proaches (pp. 75-146). San Francisco: W.
H. Freeman.
Elmore, K. C., & Oyserman, D. (2012). If “we”
can succeed, “I” can too: Identity-based
motivation and gender in the classroom.
Contemporary Educational Psychology,
37(3), 176-185.
Feather, N. T. (1982). Expectancy-value ap-
proaches: Present status and future di-
rections. In N. T. Feather (Ed.), Expecta-
JUST NOT WORTH MY TIME 17
tions and actions: Expectancy-value models
in psychology. Hillsdale, NJ: Erlbaum.
Feather, N. T. (1992). Values, valences, expecta-
tions, and actions. Journal of Social Issues,
42(2), 109-124.
Festinger, L. (1954). A theory of social com-
parison processes. Human Relations, 7,
117-140.
Fishbach, A., & Dhar, R. (2005). Goals as ex-
cuses or guides: The liberating effect of
perceived goal progress on choice. Jour-
nal of Consumer Research, 32(3), 370-377.
Gilovich, T., Kerr, M., & Medvec, V. H. (1993).
The effect of temporal perspective on
subjective confidence. Journal of Perso-
nality and Social Psychology, 64, 552-560.
Gray, J. R., Chabris, C. F., & Braver, T. S. (2003).
Neural mechanisms of general fluid in-
telligence. Natural Neuroscience, 6, 316-
322.
Hayes, A. F. (2012). PROCESS: A versatile com-
putational tool for observed variable media-
tion, moderation, and conditional process
modeling [White paper]. Retrieved from
http://www.afhayes.com/public/pro-
cess2012.pdf.
Jackson, K. (2010). A stitch in time: The effects of a
novel incentive-based high-school interven-
tion on college outcomes (National Bureau
of Economic Research Working Paper
Series, Working Paper 15722). Retrieved
from http://www.nber.org/papers/
w15722.
James, W. (1890). The principles of psychology.
New York: Holt.
Kahneman, D., & Tversky, A. (1979). Prospect
theory: An analysis of decision under
risk. Econometrica, 47(2), 263-291.
Kruglanski, A. W., Shah, J. Y., Fishbach, A.,
Friedman, R., Chun, W. Y., & Sleeth-
Keppler, D. (2002). A theory of goal sys-
tems. In M. P. Zanna (Ed.), Advances in
experimental social psychology, Vol. 34 (pp.
331-378). San Diego: Academic Press.
Kuh, G. D., Kinzie, J., Buckley, J. A., Bridges,
B. K., & Hayek, J. C. (2006, July). What
matters to student success: A review of the
literature. National Postsecondary Edu-
cation Cooperative (NPEC) Commis-
sioned Paper.
Liberman, N., & Trope, Y. (1998). The role of
feasibility and desirability consider-
ations in near and distant future de-
cisions: A test of temporal construal
theory. Journal of Personality and Social
Psychology, 75, 5-18.
Liberman, N., & Trope, Y. (2008). The psychol-
ogy of transcending the here and now.
Science, 322, 1201-1205.
Markus, H., & Kunda, Z. (1986). Stability and
malleability in the self-concept in the
perception of others. Journal of Persona-
lity and Social Psychology, 51(4), 858-866.
Nesse, R. (2009). Explaining depression, neu-
roscience is not enough: Evolution is es-
sential. In C. Pariante, R. Nesse, D. Nutt,
& L. Wolpert (Eds.), Understanding de-
pression: A translational approach. Oxford,
UK: Oxford University Press.
Orfield, G., Losen, D., Wald, J., & Swanson, C.
(2004). Losing our future: How minority
youth are being left behind by the gradua-
tion rate crisis. Cambridge, MA: The
Civil Rights Project at Harvard Uni-
versity. Retrieved from http://www.
urban.org/UploadedPDF/410936_Los-
ingOurFuture.pdf.
Oyserman, D. (2007). Social identity and self-
regulation. In A. W. Kruglanski & E. T.
Higgins (Eds.), Social psychology: Hand-
book of basic principles (pp. 432-453). New
York: Guilford.
Oyserman, D. (2009). Identity-based motiva-
tion: Implications for action-readiness,
procedural-readiness, and consumer
behavior. Journal of Consumer Psychology,
19, 250-260.
Oyserman, D. (2015). Pathways to success
through identity-based motivation. New
York: Oxford University Press.
Oyserman, D., Bybee, D., & Terry, K. (2006).
Possible selves and academic outcomes:
How and when possible selves impel
action. Journal of Personality and Social
Psychology, 91, 188-204.
Oyserman, D., Destin, M., & Novin, S. (2014).
The context sensitive future self: Pos-
sible selves motivate in context, not oth-
erwise. Self and Identity (ahead-of-print).
DOI:10.1080/15298868.2014.965733
Oyserman, D., Elmore, K., & Smith, G. (2012).
Self, self-concept, and identity. In M.
Leary & J. Tangney (Eds.), Handbook of
self and identity, 2nd edition (pp. 69-104).
New York: Guilford.
Pascarella, E. T., & Terenzini, P. T. (2005). How
college affects students, Volume 2: A third
decade of research. San Francisco: Jossey-
Bass.
18 SMITH AND OYSERMAN
Raven, J. C. (1962). Advanced progressive matri-
ces, set II. London, UK: H. K. Lewis.
Roese, N. J., & Olson, J. M. (2007). Better, stron-
ger, faster: Self-serving judgment, affect
regulation, and the optimal vigilance
hypothesis. Perspectives on Psychological
Science, 2(2), 124-141.
Rosenbaum, J. E., Deil-Amen, R., & Person, A.
E. (2006). After admission: From college ac-
cess to college success. New York: Russell
Sage Foundation Publications.
Rothman, A. J., Haddock, G., & Schwarz,
N. (2001). “How many partners is too
many?” Shaping perceptions of person-
al vulnerability. Journal of Applied Social
Psychology, 31(10), 2195-2214.
Schwarz, N. (1999). Frequency of reports of
physical symptoms and health behav-
ior: How the questionnaire determines
the results. In D. C. Park (Ed.), Processing
medical information in aging patients: Co-
gnitive and human factors perspective (pp.
93-108). Hillsdale, NJ: Erlbaum.
Schwarz, N., Bless, H., Bohner, G., Harlacher,
U., & Kellenbenz, M. (1991). Response
scales as frames of reference: The impact
of frequency range on diagnostic judg-
ment. Applied Cognitive Psychology, 5(1),
37-49.
Schwarz, N., Bless, H., Strack, F., Klumpp, G.,
Rittenauer-Schatka, H., & Simons, A.
(1991). Ease of retrieval as information:
Another look at the availability heuris-
tic. Journal of Personality and Social Psy-
chology, 61(2), 195-202.
Schwarz, N., Groves, R. M., & Schuman, H.
(1998). Survey methods. In D. T. Gilbert
& S. T. Fiske (Eds.), The handbook of social
psychology 4th edition (pp. 143-179). New
York: McGraw-Hill.
Schwarz, N., & Scheuring, B. (1988). Judg-
ments of relationship satisfaction: Inter-
and intraindividual comparisons as a
function of questionnaire structure. Eu-
ropean Journal of Social Psychology, 18(6),
485-496.
Shih, M., Pittinsky, T., & Trahan, A. (2006). Do-
main-specific effects of stereotypes on
performance. Self and Identity, 5(1), 1-14.
Silvestrini, N., & Gendolla, G. H. E. (2013). Au-
tomatic effort mobilization and the prin-
ciple of resource conservation: One can
only prime the possible and justified.
Journal of Personality and Social Psycholo-
gy, 104(5), 803-816.
Smith, E. R., & Collins, E. C. (2009). Contex-
tualizing person perception: Distributed
social cognition. Psychological Review,
116(2), 343-364.
Smith, E. R., & Semin, G. R. (2004). Socially sit-
uated cognition: Cognition in its social
context. In M. P. Zanna (Ed.), Advances
in experimental social psychology (pp. 53-
117). San Diego: Elsevier Academic.
Steele, C. M. (1997). A threat in the air: How
stereotypes shape intellectual identity
and performance. American Psychologist,
52, 613-629.
Swann, W. B., & Bosson, J. K. (2010). Self and
identity. In S. T. Fiske, D. T. Gilbert, & G.
Lindzey (Eds.), Handbook of social psycho-
logy 5th edition (pp. 589-628). Hoboken,
NJ: Wiley.
Symonds, W. C., Schwartz, R. B., & Ferguson,
R. (2011). Pathways to prosperity: Meeting
the challenge of preparing young Americans
for the 21st century. Report issued by the
Pathways to Prosperity Project, Harvard
Graduate School of Education.
Trope, Y., & Liberman, N. (2003). Temporal con-
strual. Psychological Review, 110, 403-421.
Trusty, J. (2000). High educational expectations
and low achievement: Stability of edu-
cational goals across adolescence. Jour-
nal of Educational Research, 93(6), 356-395.
Vallacher, R. R., & Wegner, D. M. (1987). What
do people think they’re doing? Action
identification and human behavior. Psy-
chological Review, 94(1), 3-15.
Weiner, B. (1985). “Spontaneous” causal think-
ing. Psychological Bulletin, 97(1), 74-84.
Wigfield, A. E., & Eccles, J. S. (1992). The de-
velopment of achievement task values:
A theoretical analysis. Developmental Re-
view, 12(3), 265-310.
Wrosch, C. (2010). The self-regulation of unat-
tainable goals and pathways to quality
of life. In S. Folkman (Ed.), The Oxford
handbook of stress, health, and coping (pp.
319-333). New York: Oxford University
Press.
