Content uploaded by Todd Mcelroy
Author content
All content in this area was uploaded by Todd Mcelroy on Feb 26, 2015
Content may be subject to copyright.
Journal of Health Psychology
1 –8
© The Author(s) 2015
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1359105314565827
hpq.sagepub.com
It has been clear for some time that physical
activity level is an important contributor to human
health (e.g. Pate et al., 1995) and is especially
influential in overall quality of life (Bize et al.,
2007). The importance of physical activity has
recently garnered a great deal of attention because
of the pandemic rise in obesity (Wang et al.,
2011) and negative health effects (Kohl et al.,
2012) associated with low activity levels. Perhaps
more striking is that many people believe they
can overcome their lack of physical activity by
taking a quick run through the gym. While this is
likely helpful, its effects on overall human health
are minimal compared to the impact of a person’s
overall daily activity level (Levine et al., 1999).
Maintaining a proper balance of physical activity
to promote good health and quality of life is a
centerpiece of the American College of Sports
Medicine (ACSM) and has been shown to sup-
port mental well-being (Fox, 1999) and good
mental health (Paluska and Schwenk, 2000).
Aside from physical activity, another domain in
which people vary is how much they like to think,
and one of the most widely used measures of dif-
ferences in thinking propensity is “need for cog-
nition” (NFC). NFC is defined as a tendency to
engage in and enjoy effortful cognitive endeavors
(Cacioppo and Petty, 1982). NFC is an individual
The physical sacrifice of thinking:
Investigating the relationship
between thinking and physical
activity in everyday life
Todd McElroy1, David L Dickinson2, Nathan
Stroh2 and Christopher A Dickinson2
Abstract
Physical activity level is an important contributor to overall human health and obesity. Research has shown
that humans possess a number of traits that influence their physical activity level including social cognition.
We examined whether the trait of “need for cognition” was associated with daily physical activity levels.
We recruited individuals who were high or low in need for cognition and measured their physical activity
level in 30-second epochs over a 1-week period. The overall findings showed that low-need-for-cognition
individuals were more physically active, but this difference was most pronounced during the 5-day work
week and lessened during the weekend.
Keywords
cognition, decision, obesity, physical activity, risk
1Florida Gulf Coast University, USA
2Appalachian State University, USA
Corresponding author:
Todd McElroy, Department of Psychology, Florida Gulf
Coast University, 10501 FGCU Blvd, South Fort Myers,
FL 33965, USA.
Email: toddmcelroyfgcu@gmail.com
565827HPQ0010.1177/1359105314565827Journal of Health PsychologyMcElroy et al.
research-article2015
Article
by guest on February 2, 2015hpq.sagepub.comDownloaded from
2 Journal of Health Psychology
difference measure of thinking that allows
researchers to study thinking without placing
demands on cognitive resources. Furthermore,
NFC is driven by intrinsic motivation and is rela-
tively stable across a person’s lifetime (Cacioppo
et al., 1996). Researchers have used this measure
for over 30 years to examine the relationship
between enjoyment of effortful cognitive endeav-
ors and other variables related to cognition
(Cacioppo et al., 1996).
For example, research has shown that indi-
viduals high in NFC appear to perform better on
memory tasks (Boehm, 1994; Cacioppo et al.,
1983), are generally more positive toward cog-
nitively difficult tasks (Cacioppo et al., 1996),
spend more effort when making decisions
(Verplanken et al., 1992) and can make better
ones (Levin et al., 2000). Low-NFC individuals
have been shown to rely more on peripheral
information (Cacioppo and Petty, 1982) and
contextual cues such as attractiveness or a per-
son’s mood (Cacioppo et al., 1996) when think-
ing and forming attitudes.
Overall, these types of studies depict a psy-
chometric tool that reveals an important force
behind human cognition and its effects on eve-
ryday life. In the current investigation, we
explore how NFC may be associated with daily
physical activity levels. Although previous
research has not specifically examined such a
connection, related research suggests that it
may exist.
Cognition and physical activity
The relationship between cognition and physi-
cal activity is important for health concerns, but
it also speaks to a more fundamental question of
how cognition interacts with the physical body
across the human lifespan. Research looking at
children and adolescents has shown many cog-
nitive variables that are related to physical
activity, including preferences, intentions
(Sallis et al., 2000), and self-efficacy (Strauss
et al., 2001). There is also evidence that neuro-
anatomical and neurochemical differences are
linked to more pervasive behavioral disorders
such as attention deficit hyperactivity disorder
(ADHD) (Swanson et al., 1998) and anxiety
(Fride and Weinstock, 1988). Similarly, research
looking at older adults has also consistently
shown a relationship between physical activity
and cognitive decline (Laurin et al., 2001;
Weuve et al., 2004). Thus, the relationship
between cognition and physical activity is an
important question for the human experience,
and the interaction likely extends across the
lifespan (e.g. Heyn et al., 2004; Kramer and
Erickson, 2007).
Research has also revealed a number of indi-
vidual difference variables that appear to be
associated with physical activity levels. For
example, intention to perform a physical activ-
ity (e.g. Petty et al., 2013) is associated with
physical behavior. Individual difference varia-
bles such as approach/avoidance motivation
(Hevey and Dolan, 2014), “Health Types”
(McGinty et al., 2012), sensation seeking in
adolescents (Sallis et al., 2000) and some com-
ponents of the Big Five personality traits
(Rhodes and Smith, 2006) influence the likeli-
hood of a person performing certain types of
physical activities. Second, there is strong evi-
dence that cognition is related to physical activ-
ity in daily life. This idea is highlighted in a
large-scale study by Godin et al., (2010). In this
study, Godin et al. tested the effects of several
variables on physical activity. These variables
were grouped by either social structure, which
represents a person’s hierarchical status, or
social cognition, which involves the processing
of social information. The findings revealed
that social structure had only a small effect on
physical activity, whereas social cognition was
determined to be the key factor in predicting
physical activity level.
Thus, existing research supports the view
that cognition and physical activity level are
associated, yet an important question is whether
individual preference toward cognitive endeav-
ors is associated with more or less physical
activity. While research has not directly exam-
ined this question, our search identified studies
that provide differing clues for how this rela-
tionship may unfold, and they provide a basis
for our investigation.
by guest on February 2, 2015hpq.sagepub.comDownloaded from
McElroy et al. 3
First, it is possible that high- and low-NFC
individuals may be engaging in different strate-
gies that ultimately affect their physical behavior.
Specifically, it could be that an associative–
dissociative attentional strategy (Masters and
Ogles, 1998) dichotomy may be at play such that
low-NFC individuals engaged in more physical
activity because they are better able to dissociate
themselves from cues related to physical exertion
(e.g. Stanley et al., 2007). In other words, this dis-
sociation would make physical activity seem
easier to these individuals. Some support for this
can be found in a study by Watt and Blanchard
(1994). In this study, low-NFC individuals dem-
onstrated a greater propensity toward boredom
and more strongly experienced its associated
negative effect. High-NFC individuals appear to
avoid this because of their ability to provide their
own mental stimulation. Thus, high-NFC indi-
viduals seem more content to “entertain them-
selves” mentally, whereas low-NFC individuals
quickly experience boredom and experience it
more negatively.
In another study that involved a limited
behavioral task, participants were charged with
observing directionality of dots on a computer
monitor. The findings showed that low-NFC
individuals performed better individually than
collectively, and they tended to outperform
high-NFC persons during individual perfor-
mance (Smith et al., 2001). This suggests that
low-NFC individuals may “loaf” more in groups
but may be more active at the individual level.
While these studies seem to suggest that low
NFC will be associated with more physical
activity, another set of findings seems to sug-
gest an opposite relationship. For example, a
study by Hess et al. (2011) looked at longitudi-
nal effects of cognitive motivation across a
wide age range. They combined Personal Need
for Structure (PNS) with NFC scores to create a
composite measure of cognitive motivation.
Their results showed that this cognitive motiva-
tion measure was positively associated with
social activities and interactions. Thus, this
study would seem to suggest that high-NFC
individuals may be more physically active in
their daily lives. This finding is consistent with
research showing that high-NFC individuals
have a stronger tendency to seek out informa-
tion (Verplanken et al., 1992), and they appear
more motivated (Cacioppo et al., 1983).
Summary and predictions
Our review of the literature reveals good evi-
dence that individual differences as well as cog-
nition appear to be associated with physical
activity. However, the direction of this relation-
ship is not clear. One set of findings seems to
suggest a tradeoff of sorts between cognitive and
physical activity. Because high-NFC individuals
are more content and eager to be involved in cog-
nitive activities, the natural outcome is that they
may be less physically active. On the other hand,
another set of findings seems to suggest that high
NFC may reflect an overall increase in motiva-
tion level that could lead to greater exploration of
the environment and social activities. Thus,
because our assessment of findings in the litera-
ture appears to present a contradictory picture,
we designed the current investigation as a way to
test this relationship and determine whether a
person’s level of cognitive activity is associated
with more or less physical activity.
Method
Participants and design
The participants in this study were 30 high- and
30 low-NFC individuals; 45 of the participants
were female.1 The conditions were roughly equal
in regard to gender; 20 females were in the high-
NFC condition and 25 in the low-NFC condition.
All participants were undergraduate students at
Appalachian State University. The experiment
utilized a one-way factorial design. The inde-
pendent variable in this study was NFC level
(high or low), and the dependent variable was the
participant’s activity levels across 1 week.
Procedure
The primary screening procedure was conducted
through an online survey using the SONA
by guest on February 2, 2015hpq.sagepub.comDownloaded from
4 Journal of Health Psychology
software system. This selection method was
necessary for several reasons: the relative scar-
city of low-NFC individuals in our sample pop-
ulation, the week-long sampling period, and the
monetary expense associated with compensat-
ing each participant. In this initial session, par-
ticipants were informed about the nature of the
study, including the potential for participation in
the second stage. They were then asked to com-
plete the NFC scale (Cacioppo et al., 1984).
After completing the NFC scale, participants
were awarded credit for their participation in
this initial screening stage.
Next, we established criteria for discerning
individuals who were high or low in NFC.
Relying on our initial NFC screening as a sample
population, we used the upper and lower 10 per-
cent of our distribution as the criteria for deter-
mining our maximum and minimum scores for
categorizing high and low NFC. Participants who
were eligible to take part in the study were con-
tacted via recruitment email; those who responded
affirmatively were scheduled for an initial lab
meeting. As a result of this recruitment classifica-
tion method, high-NFC participants had NFC
scores in the range of (4–62), whereas the range
for low-NFC participants was (−11 to −39).
The observation weeks were arranged ahead
of time so that they occurred during the semester
and did not include holidays. Participants were
contacted via email to set up the initial lab meet-
ing. Prior to the initial lab meeting, actigraphy
devices (described below) were configured and
assigned to each participant. During the initial
lab meeting, participants were informed about
the study and how to wear the device. Participants
were instructed to carry out their typical daily
routines. They were then assigned a follow-up
lab meeting time. The follow-up lab meeting
took place approximately 1 week later; schedul-
ing was based on participants’ availability. We
chose a 1-week observational period because
prior research has shown this to be the desirable
time period for assessing variability in activity
patterns (Matthews et al., 2002).
In the final lab meeting, participants returned
their actigraphy devices, and they were compen-
sated US$10.00 each for participation and return
of the device. They were given an overview of
the nature of the study and also offered an output
of their daily activity level data, which we prom-
ised to send them after the study concluded, and
the data were scored. At the end of the final lab
meeting, participants were asked several sets of
questions unrelated to this study. Data from the
actigraphy devices were downloaded using the
manufacturer’s software. Time periods when the
device was removed, which were rare, were
cleared from the dataset to avoid miscounting
them as periods of zero activity. Sleep episodes
were also removed from the data. The 1-week
period of measurement yielded ~20,000 activity
measurements per participant.
Materials
To assess participants’ level of NFC, we used
the NFC scale (Cacioppo et al., 1984). This
scale consists of 18 items; half have positive
orientations and half contain negative orienta-
tions. Participants indicated how much they
agreed or disagreed with each item on a 9-point
scale ranging from very strong disagreement
(−4) to very strong agreement (+4). Total scores
on this scale range from 72 to −72.
To measure participants’ activity levels, we
used an actigraphy device. The device is an
accelerometer worn on the non-dominant wrist
as a common means for measuring gross motor
activity (Ancoli-Israel et al., 2003). This device
resembles a common wrist watch and can be
conveniently worn by participants. Measurement
is made by internal accelerometers with sensi-
tivity of .05 g-force. This sensitivity generates
“activity counts” of varying strength and fre-
quency during each time epoch used for data
collection. We set the data sampling for the
device to occur at epoch lengths of 30 seconds.
The device is impact resistant, waterproof to
1-m depth for 30 minutes, and can be worn
24 hours a day with few exceptions.
Results
After completion of the study, daily activity
counts were obtained by averaging the 30-second
by guest on February 2, 2015hpq.sagepub.comDownloaded from
McElroy et al. 5
epoch readings across all waking hours for each
individual participant. There was a malfunction
with the actigraphy for 1 day of a low-NFC par-
ticipant’s data, and so that day was not obtainable
and excluded from the analysis. With this excep-
tion, the daily activity counts for each participant
were combined within each of the 7 days; subse-
quent analyses relied on the entire daily activity
counts within each individual day.
Because the one-way analysis of variance
(ANOVA) F-test and t-test produce statistically
identical outcomes, we relied on the ANOVA
because the analysis yields more variance infor-
mation. First, to test whether a person’s level of
NFC was associated with his or her physical
activity levels, we performed an ANOVA with
NFC level as our independent variable and over-
all daily activity level as our dependent variable.
The results from this analysis revealed that think-
ing does seem to be associated with less physical
activity. As shown in Figure 1, the difference
between high- and low-NFC individuals in over-
all weekly physical activity level was highly sig-
nificant (F(1, 58) = 7.4, p < .009, η2 = .113) such
that high-NFC individuals were far less active
overall than low-NFC participants.
Prior research directed toward measuring
daily differences in physical activity levels has
shown that weekday activity levels (Monday–
Friday) differ substantially from weekend lev-
els (Matthews et al., 2002). To test whether this
weekend effect might be present in this study,
we first performed an analysis of the weekday
activity levels comparing high- and low-NFC
individuals for Monday–Friday activity levels
(see Figure 1), and, as suspected, they differed
greatly across the 5-day typical work week
(F(1, 58) = 9.94, p < .003, η2 = .146). Next, we
tested whether this effect remained for the
weekend days. Collapsing across the weekend
days, we see that activity levels for high- and
low-NFC individuals did not differ significantly
(F(1, 58) = 2.53, p < .117, η2 = .042) on the
weekend. The results revealed that this lack of a
statistical difference in activity levels is true for
Saturday data (F(1, 58) = 2.4, p < .127, η2 = .04)
and even more so for Sunday data (F(1,58) = .21,
p > .65, η2 = .004).2
Figure 1. Average daily physical activity level for high-NFC individuals and low-NFC individuals as
measured in 30-second epochs and based on .05 g-force sensitivity. These data include the average daily
physical activity level for each group across the 1-week period. Error bars indicate the standard error of
the mean.
by guest on February 2, 2015hpq.sagepub.comDownloaded from
6 Journal of Health Psychology
Discussion
In this study, we tested whether people who pre-
fer to think more will be less physically active
in their daily lives than people who do not pre-
fer to think. Our findings build upon prior
research (e.g. Godin et al., 2010) and provide
support for our hypothesis by revealing robust
physical activity level differences during the
5-day work week and attenuated differences
over the weekend. Furthermore, it is important
to note that these differences were found using
a robust measure of physical activity over a
1-week period. This type of objective measure
has been called for to help validate other types
of self-report measures (see Godin et al., 2010).
However, the sampling method used in our
study created potential limitations that should
be noted.
First, it is important to note that part of the
“weekend effect” in our study may be due to
our sample population, which consisted of col-
lege students. Although college students are a
standard participant pool in the vast majority of
experimental psychology studies, their behav-
ior and habits may be more indicative of young
adult behavior than adult behavior in general. It
is reasonable to assume that this “weekend
effect” may change as people progress through
different life stages, which is a question that
future researchers may want to consider. A sim-
ilar limitation with our methodology was that
the participants were all involved in course-
work, a time in their lives that should revolve
around cognitively focused events. While this
was true for both high- and low-NFC partici-
pants, it may limit the external validity of this
study to cognitively oriented life situations. In
conclusion, it seems noteworthy to point out
that if this association between physical activ-
ity and cognition leads to health issues such as
obesity, it may be prudent for more thoughtful
individuals to consider lifestyle changes as
countermeasures to the negative health out-
comes associated with their lower activity lev-
els. For example, research has shown that
simply being active in mundane behaviors such
as moving about, fidgeting, or even walking to
the bathroom increases non-exercise activity
thermogenesis (NEAT). These types of activi-
ties have been shown to expend excess energy
the body has taken in, which will help avoid fat
storage and promote leanness (Levine et al.,
1999). An example of a more dramatic counter-
measure would be to replace one’s workstation
with a walking treadmill desk. These have been
shown to increase energy expenditure of
100 kcal/hour in the neighborhood, which can
result in substantial benefits (Levine and
Miller, 2007). Ultimately, an important factor
that may help more thoughtful individuals
combat their lower average activity levels is
awareness. Awareness of their tendency to be
less active, coupled with an awareness of the
cost associated with inactivity, more thoughtful
individuals may then choose to become more
active throughout the day.
Acknowledgements
Special thanks to Sarah Pollard for her devoted assis-
tance with the early data gathering in the project. The
data set of daily averages as well as the entire set of
daily activity measurements is available from the
first author.
Funding
Partial support for this research was provided by the
National Science Foundation (NSF Grant number:
1229067) and the Division of Research and
Sponsored Programs, Appalachian State University.
Some subject payments were funded by a grant from
the Office of Student Research, Appalachian State
University.
Notes
1. The study was approved by the University’s
Institutional Review Board (IRB No. 11-0067),
which is governed by the Office of Research
Protections. Written consent was obtained from
all participants.
2. We performed an additional analysis by collaps-
ing weekdays and weekends into two separate
variables. We then performed a repeated meas-
ures analysis with these two new variables as a
within factor and NFC level as a between factor.
This approach to treat our data as a mixed design
yielded a marginally significant main effect for
by guest on February 2, 2015hpq.sagepub.comDownloaded from
McElroy et al. 7
NFC; F(1, 58) = 3.59, p < .07, a non-significant
main effect for the weekday/weekend variable;
F < 1 and a non-significant interaction F < 1.
References
Ancoli-Israel S, Cole R, Alessi C, et al. (2003) The
role of actigraphy in the study of sleep and cir-
cadian rhythms. American Academy of Sleep
Medicine Review Paper. Sleep 26: 342–392.
Bize R, Johnson JA and Plotnikoff RC (2007)
Physical activity level and health related quality
of life in the general adult population: A system-
atic review. Preventive Medicine 45: 401–415.
Boehm LE (1994) The validity effect: A search for
mediating variables. Personality and Social
Psychology Bulletin 20: 285–293.
Cacioppo JT and Petty RE (1982) The need for
cognition. Journal of Personality and Social
Psychology 42(1): 116–131.
Cacioppo JT, Petty RE and Kao CF (1984) The effi-
cient assessment of need for cognition. Journal
of Personality Assessment 48(3): 306–307.
Cacioppo JT, Petty RE and Morris KJ (1983) Effects
of need for cognition on message evaluation,
recall, and persuasion. Journal of Personality
and Social Psychology 45: 805–818.
Cacioppo JT, Petty RE, Feinstein JA, et al. (1996)
Dispositional differences in cognitive motiva-
tion: The life and times of individuals vary-
ing in NFC. Psychological Bulletin 119(2):
197–253.
Fox KR (1999) The influence of physical activity on
mental well-being. Public Health Nutrition 2:
411–418.
Fride E and Weinstock M (1988). Prenatal stress
increases anxiety related behavior and alters
cerebral lateralization of dopamine activity. Life
Science 42: 1059–1065.
Godin G, Sheeran P, Conner M, et al. (2010) Social
structure, social cognition, and physical activ-
ity: A test of four models. British Journal of
Health Psychology 15: 79–95.
Hess TM, Emery L and Neupert SD (2011)
Longitudinal relationships between resources,
motivation, and functioning. The Journals of
Gerontology, Series B: Psychological Sciences
and Social Sciences 67: 299–308.
Hevey D and Dolan M (2014) Approach/avoidance
motivation, message framing and skin cancer
prevention: A test of the congruency hypothesis.
Journal of Health Psychology 19: 1003–1012.
Heyn P, Abreu BC and Ottenbacher KJ (2004) The
effects of exercise training on elderly persons
with cognitive impairment and dementia: A
meta-analysis. Archives of Physical Medicine
and Rehabilitation 85: 1694–1704.
Kohl HW, Craig CL, Lambert EV, et al. (2012) The
pandemic of physical inactivity: Global action
for public health. The Lancet 380: 294–305.
Kramer AF and Erickson KI (2007) Capitalizing on
cortical plasticity: Influence of physical activ-
ity on cognition and brain function. Trends in
Cognitive Sciences 11: 342–348.
Laurin D, Verreault R, Lindsay J, et al. (2001)
Physical activity and risk of cognitive impair-
ment and dementia in elderly persons. Archives
of Neurology 58: 498–504.
Levin IP, Huneke ME and Jasper JD (2000)
Information processing at successive stages
of decision making: Need for cognition and
inclusion-exclusion effects. Organizational
Behavior and Human Decision Processes 82:
171–193.
Levine JA, Eberhardt NL and Jensen MD (1999)
Role of nonexercise activity thermogenesis in
resistance to fat gain in humans. Science 283:
212–214.
Levine JA and Miller JM (2007) The energy expend-
iture of using a “walk-and-work” desk for office
workers with obesity. British Journal of Sports
Medicine 41: 558–561.
McGinty H, Dark-Freudeman A and West RL (2012)
Health hopes and fears for the future in relation
to health behavior and current health status.
Journal of Health Psychology 18: 1509–1518.
Masters KS and Ogles BM (1998) Associative and
dissociative cognitive strategies in exercise and
running: 20 years later, what do we know? Sport
Psychologist 12: 253–270.
Matthews CE, Ainsworth BE, Thompson RW, et al.
(2002) Sources of variance in daily physical
activity levels as measured by an accelerometer.
Medicine and Science in Sports and Exercise
34: 1376–1381.
Paluska SA and Schwenk TL (2000) Physical activ-
ity and mental health: Current concepts. Sports
Medicine 29: 167–180.
Pate RR, Pratt M, Blair SN, et al. (1995) Physical
activity and public health: A recommenda-
tion from the Centers for Disease Control and
Prevention and the American College of Sports
Medicine. JAMA: The Journal of the American
Medical Association 273: 402–407.
by guest on February 2, 2015hpq.sagepub.comDownloaded from
8 Journal of Health Psychology
Petty KN, Knee CR and Joseph AK (2013) Sunscreen
use among recreational cyclists: How intentions
predict reported behavior. Journal of Health
Psychology 18(3): 439–447.
Rhodes RE and Smith NEI (2006) Personality cor-
relates of physical activity: A review and meta-
analysis. British Journal of Sports Medicine 40:
958–965.
Sallis J, Prochaska JJ and Taylor WC (2000) A
review of correlates of physical activity of chil-
dren and adolescents. Medicine and Science in
Sports and Exercise 32: 963–975.
Smith BN, Kerr NA, Markus MJ, et al. (2001)
Individual differences in social loafing: Need
for cognition as a motivator in collective perfor-
mance. Group Dynamics: Theory, Research, &
Practice 5: 150–158.
Stanley C, Pargman D and Tenenbaum G (2007)
The effect of attentional coping strategies on
perceived exertion in a cycling task. Journal of
Applied Sport Psychology 19: 352–363.
Strauss RS, Rodzilsky D, Burack G, et al. (2001)
Psychosocial correlates of physical activity
in healthy children. Archives of Pediatrics &
Adolescent Medicine 155: 897–902.
Swanson J, Castellanos FX, Murias M, et al. (1998)
Cognitive neuroscience of attention deficit
hyperactivity disorder and hyperkinetic disor-
der. Current Opinion in Neurobiology 8: 263–
271.
Verplanken B, Hazenberg PT and Palenéwen GR
(1992) Need for cognition and external infor-
mation search effort. Journal of Research in
Personality 26: 128–136.
Wang YC, McPherson K, Marsh T, et al. (2011)
Health and economic burden of the projected
obesity trends in the USA and the UK. Lancet
378: 815–825.
Watt JD and Blanchard MJ (1994) Boredom prone-
ness and the need for cognition. Journal of
Research in Personality 28: 44–51.
Weuve J, Kang JH, Manson JE, et al. (2004)
Physical activity including walking and cog-
nitive function in older women. The Journal
of the American Medical Association 292:
1454–1461.
by guest on February 2, 2015hpq.sagepub.comDownloaded from
