International Journal of Health Promotion and Education
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rhpe20
Effects of a single message exposure on exercise
motivation and behavior among adults aged 30-45
Emily L. Mailey, Deirdre Dlugonski, Gina M. Besenyi, Rebecca Gasper &
Stacey Slone
To cite this article: Emily L. Mailey, Deirdre Dlugonski, Gina M. Besenyi, Rebecca Gasper &
Stacey Slone (2022): Effects of a single message exposure on exercise motivation and behavior
among adults aged 30-45, International Journal of Health Promotion and Education, DOI:
10.1080/14635240.2022.2031250
To link to this article: https://doi.org/10.1080/14635240.2022.2031250
Published online: 15 Feb 2022.
Submit your article to this journal
Article views: 49
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rhpe20
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
https://doi.org/10.1080/14635240.2022.2031250
Effects of a single message exposure on exercise motivation
and behavior among adults aged 30-45
Emily L. Maileya, Deirdre Dlugonskib, Gina M. Besenyia, Rebecca Gaspera
and Stacey Slonec
a
Department of Kinesiology, Kansas State University, Manhattan, KS, USA; bDepartment of Athletic Training
and Clinical Nutrition, University of Kentucky, Lexington, KY, USA; cDepartment of Statistics, University of
Kentucky, Lexington, KY, USA
ABSTRACT
ARTICLE HISTORY
Promoting intrinsic affective exercise benefits may facilitate
autonomous motivation and exercise behavior. However, most
media messages primarily emphasize extrinsic weight and
health-related benefits. This study examined whether altering
exercise messages would influence individuals’ exercise motivation and behavior. Using a 2 × 2 factorial design, message
frames were manipulated on the following dimensions:
Intrinsic/Extrinsic exercise goal and Easy/Hard exercise routine.
A representative sample of adults aged 30–45 (N = 505) completed measures of leisure-time exercise and motivation and
then were randomly assigned to view one of the four messages
or a control message. Immediately after viewing the message,
participants indicated their primary exercise goal. One week
later, they completed the exercise and motivation measures
again. An ANCOVA revealed no between-group differences in
motivation or exercise behavior. Relative to the extrinsic and
control conditions, participants in the intrinsic message conditions were more likely to identify an effective primary exercise
goal immediately after viewing the message. Overall, a single
message exposure did not significantly affect individuals’ exercise motivation or behavior in this sample. Future studies
should consider honing optimal message content, assessing
message understanding and determining the minimum dose
of message exposure needed for a meaningful impact.
Received 7 June 2021
Accepted 15 January 2022
KEYWORDS
Physical activity; intrinsic;
extrinsic; self-determination
theory; message framing
Introduction
Over the past several decades, the field of exercise science has accumulated an abundance
of knowledge regarding the widespread benefits of physical activity. As the evidence base
supporting the relationships between physical activity and chronic disease has grown,
physical activity guidelines have been developed to encourage the public to engage in
sufficient activity to reap these well-established health benefits (USDHHS 2018). In spite of
these major advances in our field and the proliferation of knowledge about the benefits of
physical activity, physical activity levels have remained discouragingly low in the United
CONTACT Emily L. Mailey
Ave. Manhattan, KS 66506
emailey@ksu.edu
© 2022 Institute of Health Promotion and Education
Department of Kinesiology, Kansas State University, 920 Denison
2
E. L. MAILEY ET AL.
States and other developed countries. In 2018, about 54% of American adults engaged in
recommended amounts of aerobic activity, and only about 24% of American adults met
the guidelines for both aerobic and muscle-strengthening activity (CDC 2018).
Although the causes of this widespread inactivity are complex, one potential contributor may be the way in which we have typically communicated about physical
activity, and in particular exercise, to the public. In general, exercise promotion messages
in the popular media, and even those delivered by reputable health organizations, have
not targeted theoretical constructs or behavior change techniques known to be associated
with exercise behavior (Gainforth et al. 2011). In the popular media and the fitness
industry, exercise is most often portrayed as a means to lose weight and/or achieve rapid
changes in appearance (Berry and Latimer-Cheung 2013; Brown, Miller, and Palmer
2017). Thus, individuals have been socialized to think about exercise in terms of these
extrinsic benefits (Smith and Bonfiglioli 2015). Unfortunately, evidence is mounting that
appearance and health-related reasons for exercise do not motivate sustained behavior as
well as affective benefits (e.g. reduced stress, improved energy) that may contribute more
immediately to one’s daily well-being (Gellert, Ziegelmann, and Schwarzer 2012; Mailey
et al. 2018; Segar, Eccles, and Richardson 2011).
From a Self-Determination Theory perspective, outcomes such as weight loss and
appearance are considered extrinsic goals (i.e. outwardly focused on external indicators
of success), and focusing on such goals may elicit controlled motivation, or a sense that
one should be exercising to meet others’ expectations (i.e. external regulation) or to avoid
self-induced guilt or shame (i.e. introjected regulation) (Deci & Ryan, 2000). On the
other hand, affective outcomes generally represent intrinsic goals (i.e. inwardly focused
on personal growth), which tend to be more strongly associated with autonomous
motivation, or a genuine sense that one wants to be active, because they value the benefits
(i.e. identified regulation) or derive enjoyment and satisfaction from the experience itself
(i.e. intrinsic regulation) (Deci and M 2000). Thus, shifting the exercise promotion
narrative to better emphasize affective benefits has been identified as an important public
health priority (Segar et al. 2016).
In line with these efforts, several studies have examined whether exercise messages
that highlight affective benefits have a positive effect on the motivation and behavior of
those receiving the messages. Segar and colleagues (2012) found overweight women aged
40–60 who viewed an advertisement emphasizing daily well-being reported greater
autonomous motivation than those who viewed an advertisement emphasizing weight
loss, but overweight men exhibited the opposite response and reported higher autonomous motivation after viewing the weight loss advertisement. Conner et al. (2011) found
college students who received a message highlighting affective outcomes of exercise (e.g.
reduced stress) reported significantly more exercise 3 weeks later than participants who
received a message highlighting instrumental outcomes (e.g. reduced risk of chronic
disease). Recently, Hevel et al. (2019) compared the effects of messages highlighting
affective versus physical health benefits on motivation to join an exercise class among
adults in a university community and found that the affective messages were associated
with greater motivation among active, but not inactive individuals. Together, these
studies show some promise for the positive effects of messages emphasizing affective
benefits on exercise motivation and behavior, but also reveal some inconsistent effects
that warrant further investigation.
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
3
There have also been calls to revise our exercise messages to emphasize that any
movement is beneficial (Arena et al. 2018). Women, in particular, have reported negative
reactions to ‘typical’ messages that portray exercise as grueling and time-consuming
because they do not feel that they represent what they can realistically do (Thai et al.
2019). Evidence is emerging that moving away from ‘no pain, no gain’ type messages may
have a positive impact on individuals’ exercise perceptions and behavior. In a sample of
UK adults, Knox et al. (2015) compared messages that emphasized physiological aspects
of activity (e.g. increased breathing and heart rates) to messages that emphasized an
‘easier’ approach (i.e. walking), and found participants reported greater intentions to be
active in the walking message group as a function of more positive affective attitudes
towards physical activity. Similarly, a review by Latimer, Brawley, and Bassett (2010)
found preliminary evidence that participants report greater confidence to engage in
physical activity if the behavior is framed as easy (e.g. moderate intensity, manageable
time commitment) as opposed to difficult of complex (e.g. vigorous intensity, substantial
time commitment).
Willis and Knobloch-Westerwick (2014) recently conducted a scoping review of
physical activity messaging, and reported that the existing evidence generally supports
the use of gain-framed messages (i.e. emphasizing benefits of physical activity) over
loss-framed messages, as well as messages that emphasize short-term mental and
emotional benefits. Other recommendations included ground messages in theory,
target-specific demographic groups, incorporate ‘how to’ information in messages
and use ‘commercial style messages’. The present study sought to incorporate these
recommendations and add to the message framing literature in several ways. First,
although adults in general have been a common target of message framing studies, this
study aimed to focus on adults aged 30–45 because they are likely to be balancing
multiple demands, including work and caring for children and/or aging parents, and
thus may find it difficult to prioritize exercise amidst all of their other obligations
(Mailey et al. 2014; Welch et al. 2009). Additionally, although previous studies have
examined messages promoting intrinsic vs. extrinsic physical activity benefits and
depicting exercise as easy vs. hard, to our knowledge, no previous studies have
examined the potential interaction between these two approaches. Previous studies
have examined the effects of single and multiple message exposures; this study utilized
a single exposure to examine the impact, if any, of fleeting encounters with exerciserelated messages that are likely among individuals who seek health and exercise
information online.
The overall purpose of the present study was to examine whether a single exposure to
an exercise message could impact motivation, exercise behavior, or exercise goals among
adults aged 30–45. Messages highlighted affective (intrinsic) or instrumental (extrinsic)
exercise goals and described exercise as easy or hard. Our primary hypothesis was that
individuals exposed to messages depicting intrinsic goals and portraying exercise as easy
would report greater autonomous motivation and exercise behavior at one-week followup relative to individuals exposed to messages depicting extrinsic goals and portraying
exercise as hard. Our secondary hypothesis was that individuals who viewed an intrinsic
message would be more likely to identify an affective primary exercise goal, whereas
individuals who viewed an extrinsic message would be more likely to identify a weight/
appearance-related primary exercise goal immediately after viewing the message. Finally,
4
E. L. MAILEY ET AL.
we explored the overall frequency of primary exercise goals reported and the extent to
which motivation and exercise behavior differed among participants endorsing different
goals.
Methods
Study design
This study employed a 2 × 2 factorial design, whereby message frames were manipulated
on the following dimensions: Intrinsic/Extrinsic exercise goal and Easy/Hard exercise
routine. All participants viewed the exact same image and recommended exercises but
were randomly assigned to receive one of the four messages or to a control condition that
did not view any image or recommended exercises. The control condition was included
to determine whether exposure to any exercise message (regardless of content) would
influence exercise motivation or behavior. Participants viewed the exercise ‘advertisement’ immediately following completion of the baseline questionnaires and completed
follow-up measures 1 week later. All procedures were approved by the university's
Institutional Review Board, and all measures were completed online via Qualtrics.
Message development
In developing the advertisement and messages, we chose to emphasize weight loss as the
extrinsic exercise goal because of the prevalence of weight-related messages in the
popular media (Bazzini et al. 2015; Willis and Knobloch-Westerwick 2014). All messages
were gain-framed and highlighted the potential benefits of exercise. The specific language
used to describe Intrinsic/Extrinsic goals and an Easy/Hard exercise routine was adapted
from language commonly observed in popular magazines (Mailey, Gasper, and
Dlugonski 2019). The final terms selected were as follows: An immediate boost in your
energy and mood (Intrinsic Goal), To blast fat and lose weight fast (Extrinsic Goal), Quick,
Easy (Easy Routine) and High-intensity, hard-core (Hard Routine). The advertisement
recommended an at-home circuit because it would be feasible for individuals to complete
without access to a gym or expensive equipment, and individuals could tailor the workout to their preferred intensity and duration. The suggested exercises included a mix of
aerobic and strength-based activities that could, if completed in quick succession, contribute to both the aerobic and muscle strengthening physical activity recommendations.
We included photos to make the advertisement visually appealing. The photos selected
were intended to be neutral and represented both a man and a woman in exercise attire
(Figure 1).
Participants
Participants were recruited in July 2018 using the Qualtrics Panels service, which facilitates sample recruitment from a national pool of volunteers. The Qualtrics team contacted members of their panels within the designated age range (30–45) and informed
them that there was a survey they may be eligible to complete. Individuals who accessed
the survey completed a screening question to indicate that they could safely participate in
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
5
Figure 1. Exercise ‘advertisement’ (easy, intrinsic condition).
exercise, then provided demographic information, which Qualtrics used to ensure the
race/ethnicity of final sample was representative of the US population. Qualtrics
employed pre-set gender quotas for each of the conditions, so that the final sample had
equal distributions of male and female participants across all conditions.
Procedures
All participants provided informed consent online before proceeding to the
questionnaires.
Exercise advertisements were embedded into the baseline Qualtrics survey, and
a Randomizer feature was used to randomly assign participants to one of the five
conditions: Intrinsic Easy, Intrinsic Hard, Extrinsic Easy, Extrinsic Hard, or control.
Qualtrics utilized a number of procedures to screen out invalid responses. Specifically,
responses were discarded if response time was faster than a set minimum (determined by
the length and complexity of the survey), if they included gibberish responses to openended items or if they selected the same response choice (e.g. strongly agree) on a set of
multiple-choice items where different responses would be expected. After these screening
procedures had been completed and 500 initial responses were obtained, the primary
6
E. L. MAILEY ET AL.
investigator conducted a second inspection of all data to identify other potential issues
that may have been missed. An additional 87 responses were discarded because they met
at least one of the following criteria: 1) implausible values on the measure of exercise
behavior (e.g. single-digit responses to exercise duration questions) or 2) incongruent
response patterns on the exercise motivation questionnaire (e.g. selecting ‘strongly agree’
and ‘strongly disagree’ for similarly worded items). These participants were subsequently
replaced by Qualtrics with new participants until a total of 500 valid responses were
obtained.
On the baseline survey, participants completed measures of demographics, exercise
behavior, and motivation prior to viewing the exercise advertisement. For all groups, the
image was preceded by the following message: We know fitting exercise into your busy life
can be challenging! We hope this brief message of motivation will encourage you to try to
make exercise a priority. It will be worth it! Immediately after viewing the advertisement,
participants responded to one additional question that assessed their primary exercise
goal. At the end of the baseline survey, participants were reminded that they would
receive an invitation to complete a follow-up survey in 1 week.
One week after participants completed the baseline survey, Qualtrics sent them a link
to the follow-up survey, which included the same measures of exercise behavior, motivation and primary exercise goal. Individuals who had not completed the follow-up survey
after 48 hours received a reminder email. In addition to completing the same measures of
exercise behavior, motivation and primary exercise goal as the baseline surveys, participants reported whether they remembered seeing a brief message of motivation related to
exercise, and if so, what type of exercise the message recommended and why the message
recommended exercising.
Measures
Motivation
The Exercise Self-Regulation Questionnaire is a 12-item measure that assesses quality of
motivation and consists of four subscales: external regulation (e.g. Because others like me
better when I am in shape), introjected regulation (e.g. Because I’d be afraid of falling too
far out of shape), identified regulation (e.g. Because I have a strong value for being active
and healthy), and intrinsic regulation (e.g. Because it is fun and interesting). It was
originally developed based on guidelines for assessing behavioral regulations from
Ryan and Connell (1989). Each item begins with the stem: Why do you (or would you)
exercise? Participants reported the extent to which each statement was true for them on
a scale from 1 (not at all true) to 7 (very true). For each subscale, responses were averaged
to yield a total score.
Exercise behavior
Exercise was assessed using the Godin Leisure-Time Exercise Questionnaire (GLTEQ),
on which participants reported the current frequency of engaging in strenuous, moderate and light intensity exercise for at least 15 minutes per session in the past seven
days (Godin and Shephard 1985). A total weekly exercise score was then calculated by
multiplying the frequencies of strenuous, moderate and light activities by corresponding metabolic equivalent (MET) intensity values of nine, five and three, respectively,
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
7
and then summing the products. Thus, the GLTEQ total score estimates total leisuretime exercise energy expenditure during the past week in METs. Based on guidelines
provided by Godin (2011), participants were classified as ‘active’ if the sum of their
strenuous and moderate activities was ≥ 24 METs, and ‘inactive’ if the sum of their
strenuous and moderate activities was < 24 METs. Based on the descriptions and
examples of strenuous and moderate exercise included in the questionnaire (e.g.
strenuous exercise = heart beats rapidly), the GLTEQ is most likely to capture activities
that are aerobic in nature. Thus, these activity classifications likely reflect whether or
not individuals were meeting the aerobic activity guidelines at the time they participated in the study.
Primary exercise goal
Primary exercise goal was assessed by a single question: If you were to engage in
exercise during the next week, what is the primary motive or goal that would drive you
to be active? Participants selected their primary exercise goal from a list of ten options,
which were drawn from the Exercise Motivations Inventory-2 (Markland and
Ingledew 1997) to reflect common intrinsic and extrinsic goals. For analyses, responses
were combined to create four general categories: weight/appearance goal (to lose
weight or manage my weight, to enhance my appearance), affective goal (to manage
stress, to feel good, to have fun), health goal (to live a healthy lifestyle, to prevent health
problems) and other goal (to spend time with friends, to enhance fitness, to accomplish
a personal goal). Participants also had the option to write in an ‘other’ goal, which were
subsequently classified into one of the four general categories by the primary
investigator.
Data analysis
The raw baseline and follow-up motivation and exercise scores were highly skewed,
so univariate comparisons across the groups were conducted using the KruskalWallis test to investigate baseline differences, and change scores were calculated for
the primary analyses. Chi-square tests and independent samples t-tests were used to
determine whether those who dropped out differed from those who completed the
study on any baseline or demographic variables. To determine whether motivation
or exercise scores differed by group (primary analyses) after exposure to the
messages, analysis of covariance (ANCOVA) was conducted on the change from
baseline to follow-up. Baseline scores were added to the models as covariates in
addition to gender, parent status and baseline activity level (dichotomous variable
based on GLTEQ scores). Each model included three pre-planned comparisons to
examine contrasts between treatment vs. control groups, intrinsic vs. extrinsic
groups and easy vs. hard groups. To test the secondary hypothesis, chi-square
tests were conducted to determine whether the distribution of primary exercise
goal differed by group. Finally, we conducted Kruskal–Wallis tests to determine
whether motivation or exercise behavior differed according to primary exercise goal
at baseline (i.e. immediately after the message exposure). Changes from baseline
based on primary exercise goal were tested using ANCOVA adjusting for baseline
score, gender, parental status and baseline activity level as in the primary analyses.
8
E. L. MAILEY ET AL.
Table 1. Demographic characteristics of the sample by group at baseline (N = 505).
Mean (SD)/Frequency (%)
Age
Gender
Female
Male
Race
White
Asian
Black
Native American
Other
Employment status
Full-time
Part-time
Unemployed/homemaker
Student/other
Education
No high school diploma
High school graduate
Some college/2-year degree
College graduate
Graduate/professional degree
Annual household income
<$35,000
$35,000-$49,999
$50,000-$74,999
$75,000-$99,999
>$100,000
Marital status
Married
Single/Divorced/Separated
Number of children
0
1
2
3+
Physical activity level
Active
Inactive
Control
N = 101
37.3 (4.1)
Extrinsic Easy
N = 101
37.8 (4.7)
Intrinsic Easy
N = 100
36.6 (4.4)
Extrinsic Hard
N = 101
37.0 (4.6)
Intrinsic Hard
N = 102
37.6 (4.2)
51 (50.5%)
50 (49.5%)
50 (49.5%)
51 (50.5%)
50 (50.0%)
50 (50.0%)
50 (49.5%)
51 (50.5%)
51 (50.0%)
51 (50.0%)
59 (58.4%)
19 (18.8%)
18 (17.8%)
1 (1.0%)
5 (5.0%)
56 (55.4%)
14 (13.9%)
20 (19.8%)
5 (5.0%)
11 (10.9%)
56 (56.0%)
17 (17.0%)
22 (22.0%)
2 (2.0%)
6 (6.0%)
57 (56.4%)
11 (10.9%)
21 (20.8%)
3 (3.0%)
13 (12.9%)
59 (57.8%)
15 (14.7%)
22 (21.6%)
3 (2.9%)
7 (6.9%)
74 (73.3%)
8 (7.9%)
13 (12.9%)
6 (5.9%)
75 (74.3%)
9 (8.9%)
12 (11.9%)
5 (5.0%)
71 (71.0%)
10 (10.0%)
14 (14.0%)
5 (5.0%)
78 (77.2%)
5 (5.0%)
13 (12.9%)
5 (5.0%)
72 (70.6%)
12 (11.8%)
16 (15.7%)
2 (2.0%)
0 (0.0%)
13 (12.9%)
34 (33.7%)
22 (21.8%)
32 (31.7%)
0 (0.0%)
14 (13.9%)
33 (32.7%)
26 (25.7%)
28 (27.7%)
1 (1.0%)
13 (13.0%)
31 (31.0%)
32 (32.0%)
23 (23.0%)
4 (4.0%)
11 (10.9%)
33 (32.7%)
30 (29.7%)
23 (22.8%)
0 (0.0%)
13 (12.7%)
36 (35.3%)
28 (27.5%)
25 (24.5%)
22 (21.8%)
16 (15.8%)
18 (17.8%)
10 (9.9%)
33 (32.7%)
20 (19.8%)
14 (13.9%)
17 (16.8%)
13 (12.9%)
35 (34.7%)
23 (23.0%)
12 (12.0%)
28 (28.0%)
16 (16.0%)
21 (21.0%)
27 (26.7%)
14 (13.9%)
22 (21.8%)
13 (12.9%)
24 (23.8%)
24 (23.5%)
19 (18.6%)
21 (20.6%)
8 (7.8%)
30 (29.4%)
53 (52.5%)
48 (47.5%)
57 (56.4%)
44 (43.6%)
49 (49.0%)
51 (51.0%)
43 (42.6%)
58 (57.4%)
59 (57.8%)
43 (42.2%)
52 (51.5%)
21 (20.8%)
20 (19.8%)
8 (7.9%)
45 (44.6%)
19 (18.8%)
24 (23.8%)
13 (12.9%)
47 (47.0%)
17 (17.0%)
17 (17.0%)
19 (19.0%)
56 (55.4%)
20 (19.8%)
16 (15.8%)
9 (8.9%)
42 (41.2%)
17 (16.7%)
27 (26.5%)
16 (15.7%)
56 (55.4%)
45 (44.6%)
68 (67.3%)
33 (32.7%)
58 (58.0%)
42 (42.0%)
63 (62.4%)
38 (37.6%)
59 (57.8%)
43 (42.2%)
Results
Participant characteristics and retention
The final sample included 505 individuals that provided valid baseline data. The
demographic characteristics of the sample at baseline are presented in Table 1.
There were no significant demographic differences between groups. On average,
participants were 37.2 years old, 50% were male, 56.8% were White, 73.3% were
working full time, 51.7% were married and 52.1% had children. About 60% of the
participants were classified as ‘active’ (i.e. meeting physical activity guidelines) based
on their self-reported exercise behavior.
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
9
Of the 505 randomized participants, 382 (75.6%) provided valid follow-up data 1 week
later. Completion rates were similar across groups. However, individuals with children
were more likely to drop out than those without children (p = .01). Additionally,
participants who dropped out reported higher leisure-time exercise at baseline (M
(SD) = 46.8 (33.2)) than those who completed the study (M(SD) = 38.7 (28.0); p = .008).
Primary analysis
Descriptive statistics for the motivation subscales and exercise behavior are presented by the
group in Table 2. Exercise behavior differed significantly between groups at baseline
(p = 0.004); there were no baseline group differences in the motivation subscales. Figure 2
displays the mean change scores with 95% confidence intervals by group for motivation and
exercise behavior. ANCOVA analyses revealed no significant overall group effects for any of
the motivation subscales or for exercise behavior, indicating that changes in motivation and
exercise behavior were similar across all groups, including the control group. In addition, all
pre-planned comparisons yielded non-significant effects. When adding gender, parent status
and baseline activity level as covariates, all models remained non-significant. Baseline activity
level (active vs. inactive) was significantly associated with changes in all motivation subscales,
such that active individuals reported greater increases in intrinsic (B = 0.35, p = 0.001),
identified (B = 0.29, p = 0.004), introjected (B = 0.38, p = 0.001) and external regulation
(B = 0.28, p = 0.02) than inactive individuals from baseline to follow-up.
Secondary analysis
Immediately after viewing the advertisement at baseline, the distribution of primary
exercise goal in the full sample was as follows: weight/appearance (46%), health (30%),
affective (17%), other (7%). We examined the extent to which the distributions of
primary goals differed by group. An initial chi-square analysis including all groups and
all goals approached significance (X2 = 19.27, p = .08). Among those that viewed an
intrinsic message, 22% endorsed an affective primary exercise goal, compared to 15% of
those that viewed an extrinsic message and 12% in the control condition. Furthermore,
among those that viewed an extrinsic message, 48% endorsed a primary exercise goal
related to weight or appearance, compared to 39% of those that viewed an intrinsic
message and 54% in the control condition (Figure 3). Because the exercise messages
emphasized either weight-related or affective benefits, a secondary chi-square analysis
comparing just these two goals revealed a significant immediate difference between
groups (X2 = 10.64, p = .03). However, at one-week follow-up, the distribution of primary
exercise goal no longer differed by group (X2 = 7.70, p = .10).
Next, we examined whether the motivation subscales or exercise behavior differed
according to primary exercise goal, regardless of group assignment. Descriptive statistics
(median (IQR)) are presented in Table 3. The results revealed significant differences in
intrinsic regulation (p < 0.001), identified regulation (p < 0.001), and exercise behavior
(p = 0.004) by primary exercise goal at baseline. Specifically, participants with a primary
goal related to weight or appearance reported lower intrinsic regulation and exercise behavior
than participants with primary affective, health, or other goals. However, changes in motivation and exercise from baseline to follow-up did not differ according to primary exercise goal.
10
E. L. MAILEY ET AL.
Table 2. Descriptive statistics [median (IQR)] by group and time point for motivation and exercise variables (completers only).
Intrinsic
Regulation
Identified
Regulation
Introjected
Regulation
External
Regulation
Exercise
Behavior
a
Time point
Baseline
Follow-up
Baseline
Follow-up
Baseline
Follow-up
Baseline
Follow-up
Baseline
Follow-up
Intrinsic Easy
(n = 79)
4.0 (2.3–5.3)
4.3 (3.0–5.0)
5.2 (4.0–6.5)
5.0 (4.0–6.3)
4.2 (3.3–5.3)
4.0 (3.3–5.0)
4.3 (3.0–5.3)
4.3 (3.0–5.7)
35.0 (12.5–59.0)
31.0 (15.0–53.0)
P-values are from the Kruskal–Wallis test at baseline.
Extrinsic Easy
(n = 70)
4.3 (3.0–5.7)
4.3 (3.3–5.7)
5.7 (4.7–6.7)
5.7 (4.7–6.7)
4.3 (3.3–5.0)
4.3 (3.0–5.0)
4.3 (3.3–5.7)
4.3 (3.7–5.7)
43.0(26.0–61.0)
39.5(23.0–58.0)
Intrinsic Hard
(n = 79)
4.0 (3.0–5.0)
4.0 (3.0–5.0)
5.3 (4.3–6.3)
5.0 (4.3–6.0)
4.0 (2.7–5.0)
4.0 (3.0–4.7)
4.0 (2.7–5.0)
4.3 (3.0–5.3)
35.0 (17.0–52.0)
30.0 (17.0–48.0)
Extrinsic Hard
(n = 76)
4.0 (2.7–5.3)
4.3 (2.7–5.7)
5.7 (4.3–6.5)
6.0 (5.0–7.0)
4.3 (3.3–5.3)
4.8 (3.7–5.3)
4.7 (3.7–5.7)
5.0 (3.8–6.0)
43.0 (23.0–63.0)
39.5 (25.0–62.0)
Control
(n = 78)
3.7 (2.7–5.3)
4.0 (2.3–5.0)
5.3 (4.0–6.3)
5.3 (4.3–6.3)
4.0 (3.0–5.0)
4.3 (3.0–5.0)
4.0 (2.7–5.3)
4.7 (3.0–5.7)
31.0 (17.0–48.0)
34.0 (14.0–49.0)
pa
0.22
0.16
0.12
0.08
0.004
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
11
Figure 2. Mean changes in motivation (2a) and exercise (2b) outcomes by group, with 95% confidence
intervals.
Figure 3. Distribution of primary exercise goals immediately after viewing the advertisement.
Table 3. Descriptive statistics [median (IQR)] for motivation and exercise behavior by primary exercise
goal in the full sample.
Intrinsic
Regulation
Identified
Regulation
Introjected
Regulation
External
Regulation
Exercise Behavior
a
Time
point
Baseline
Followup
Baseline
Followup
Baseline
Followup
Baseline
Followup
Baseline
Followup
Weight/ appearance
(N = 231)
3.7 (2.3–5.0)
3.7 (2.3–5.0)
Affective
(N = 86)
4.3 (3.0–5.3)
4.3 (3.0–5.0)
Health
(N = 153)
4.7 (3.0–5.5)
4.3 (3.0–5.3)
5.0 (4.0–6.3)
5.3 (4.3–6.3)
5.0 (4.0–6.3)
5.0 (4.0–6.3)
5.7 (5.0–6.7)
6.0 (5.0–7.0)
6.3 (5.0–7.0)
6.3 (4.5–7.0)
<0.001
4.3 (3.3–5.0)
4.3 (3.3–5.3)
4.0 (2.7–5.0)
3.7 (2.7–4.3)
4.0 (3.3–5.0)
4.5 (3.3–5.3)
4.3 (3.0–5.3)
4.3 (2.8–5.2)
0.52
4.3 (3.3–5.3)
4.7 (3.7–5.7)
4.0 (3.0–5.0)
4.0 (3.0–5.3)
4.0 (3.0–5.3)
4.3 (3.3–5.7)
4.7 (3.3–5.3)
4.7 (3.3–5.3)
0.55
32.0 (15–52)
31.5 (16–50)
39.5 (23–60)
33.5 (24–53)
42.0 (22–59)
39.0 (23–57)
38.0 (19–56)
37.5 (17–50.5)
0.004
P-values are from the Kruskal–Wallis test at baseline.
Other (N = 35)
pa
5.0 (3.7–5.7) <0.001
4.7 (3.2–6.2)
12
E. L. MAILEY ET AL.
Message recall
Of the 304 participants who completed the follow-up survey and viewed one of the
advertisements (i.e. not in the control group), 55% said they remembered seeing
a motivational message, 17% said they did not remember and 28% were unsure. In
terms of the type of exercise recommended, some participants mentioned specific
exercises included in the advertisement (e.g. at-home circuit training, bodyweight exercises, push ups), but others referenced more general exercise modes (e.g. cardio, moderate activity). When asked about why the message recommended exercise, a small
number of participants recalled specific outcomes from the ads (i.e. energy, mood, weight
loss), but a majority said ‘to be healthy’ (or something similar).
Discussion
The purpose of this study was to determine whether a single exposure to an exercise
message depicting intrinsic vs. extrinsic goals and portraying exercise as easy vs. hard
would differentially impact autonomous motivation, exercise behavior, or exercise goals
among adults aged 30–45. The primary hypothesis was not supported, as changes in
exercise motivation and behavior were similar across groups. Results did reveal an
immediate effect of message frame on individuals’ primary exercise goal, such that
participants were more likely to identify an affective primary exercise goal if they had
viewed an advertisement highlighting affective benefits. However, this effect was not
evident 1 week later.
Overall, there were very few effects of message frame on the outcomes assessed. One
potential explanation for these findings was that the exposure to the message was too
brief. Latimer, Brawley, and Bassett (2010) found message effects were more consistent in
studies with repeated exposures relative to a single exposure. In the present study,
participants may have viewed the message for only a few seconds before continuing
with the surveys, and likely spent little time processing the content of the message (Berry
and Latimer-Cheung 2013). This fleeting exposure was competing against existing automatic associations about exercise that may have accumulated over decades of personal
experience and media influence (Berry, McCarville, and Rhodes 2008). Future studies
should attempt to determine the minimum dose of message exposure needed to have
a meaningful impact on individuals’ motivation and behavior. The lack of message
tailoring in this study may also have contributed to the lack of differences between
groups. When messages are tailored to match the characteristics of the recipients,
individuals may attend to the messages more carefully, ultimately increasing the likelihood that they enact the recommended behavior (Berry and Latimer-Cheung 2013; op
den Akker et al. 2015; Pope, Pelletier, and Guertin 2017). In the present study, the
motivational message may have been more compelling to individuals if it was tailored
to their current activity level, preferred mode of exercise, or gender.
The results of this study indicated that nearly half of all participants reported primary
exercise goals related to weight or appearance, whereas only 17% of the participants
identified an affective outcome as their primary exercise goal. This is concerning considering weight/appearance goals were associated with less intrinsic motivation and
exercise behavior, regardless of group. Our secondary analyses did provide some
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
13
evidence that messages highlighting immediate affective benefits may increase the likelihood that individuals endorse these exercise goals. The novelty of these messages
relative to ‘typical’ messages that emphasize weight and appearance outcomes may
have captured participants’ attention, albeit briefly (Berry and Latimer-Cheung 2013).
Together, these results provide some support for calls to ‘rebrand exercise’ and highlight
affective benefits related to daily quality of life (Segar, Eccles, and Richardson 2011).
Implementing these communication strategies in contexts beyond the media (e.g. healthcare, worksite wellness) may provide a more sustained and meaningful exposure to the
messages.
Limitations
This study has a number of limitations, which are important to acknowledge. First, the
study duration was short, and the follow-up data were collected only 1 week after
baseline. However, given the brief nature of the message exposure, it is unlikely that
a longer follow-up would have yielded different results, as any immediate effects observed
were no longer present even 1 week later. Additionally, although the recruitment strategy
yielded a large, demographically representative sample, the study may have been underpowered to detect significant effects given the brief nature of the manipulation. In
addition, the extent to which panelists took the time to view/read the advertisement
carefully or answer questions thoughtfully is unknown. The follow-up survey demonstrated that though many participants recalled seeing an exercise message on the initial
survey, few remembered specific details of the image or message. This may be a function
of the brief message exposure or of the sample (i.e. members of survey panels may
complete multiple surveys each week). A manipulation check with specific response
options that matched the message characteristics may have provided a more reliable
and conclusive assessment than the open-ended approach we utilized to assess the extent
to which individuals attended to and processed the motivational messages. Furthermore,
future studies should pretest the messages to evaluate how they are received by the target
audience. In this study, other aspects of the advertisements (e.g. photos, exercises
recommended) may have influenced participants’ perceptions of whether the exercises
were ‘easy’ or ‘hard’. For example, individuals who have limited exercise experience or
are not currently active may perceive exercises such as triceps dips or high knees running
to be intimidating or difficult, regardless of whether they are described as ‘easy’ or ‘hard’.
Other participant characteristics not assessed in this study could also be explored. For
example, individuals of different weight statuses may have different perceptions of
whether the proposed exercises are easy or hard and react differently to messages such
as ‘blast fat and lose weight fast’.
Conclusion
In conclusion, the results of this study suggest a single exposure to a motivational
message did not have a significant impact on individuals’ exercise motivation or behavior. Because messages emphasizing weight and appearance as primary exercise goals are
so pervasive in our society, a more concerted effort that includes multiple message
exposures across a variety of contexts is likely needed to effectively ‘rebrand exercise’
14
E. L. MAILEY ET AL.
such that individuals are more cognizant of the immediate affective benefits exercise can
provide.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
This work was supported by a Kansas State University. The award is a Small Research Grant
(USRF) #3574.
References
Arena, R., A. McNeil, S. Street, S. Bond, D. R. Laddu, C. J. Lavie, and A. P. Hills. 2018. “Let Us Talk
about Moving: Reframing the Exercise and Physical Activity Discussion.” Current Problems in
Cardiology 43 (4): 154–179. doi:10.1016/j.cpcardiol.2017.06.002.
Bazzini, D. G., A. Pepper, R. Swofford, and K. Cochran. 2015. “How Healthy are Health
Magazines? A Comparative Content Analysis of Cover Captions and Images of Women’s and
Men’s Health Magazine.” Sex Roles 72 (5–6): 198–210. doi:10.1007/s11199-015-0456-2.
Berry, T. R., and A. E. Latimer-Cheung. 2013. “Overcoming Challenges to Build Strong Physical
Activity Promotion Messages.” American Journal of Lifestyle Medicine 7 (6): 371–378.
doi:10.1177/1559827613499289.
Berry, T. R., R. E. McCarville, and R. E. Rhodes. 2008. “Getting to Know the Competition:
A Content Analysis of Publicly and Corporate Funded Physical Activity Advertisements.”
Journal of Health Communication 13 (2): 169–180. doi:10.1080/10810730701854086.
Brown, T. C., B. M. Miller, and B. M. Palmer. 2017. “What’s in a Name? Group Fitness Class
Names and Women’s Reasons for Exercising.” Health Marketing Quarterly 34 (2): 142–155.
doi:10.1080/07359683.2017.1309212.
Centers for Disease Control and Prevention. 2018. “Trends in Meeting the 2008 Physical Activity
Guidelines, 2008–2018.“ Accessed 6 Oct 2021. Available at: https://www.cdc.gov/physicalactiv
ity/downloads/trends-in-the-prevalence-of-physical-activity-508.pdf
Conner, M., R. E. Rhodes, B. Morris, R. McEachan, and R. Lawton. 2011. “Changing Exercise
through Targeting Affective or Cognitive Attitudes.” Psychology and Health 26 (2): 133–149.
doi:10.1080/08870446.2011.531570.
Deci, E. L., and R. M. 2000. “The ‘What’ and ‘Why’ of Goal Pursuits: Human Needs and the Selfdetermination of Behavior.” Psychological Inquiry 11 (4): 227–268. doi:10.1207/
S15327965PLI1104_01.
Gainforth, H. L., C. J. Barg, A. E. Latimer, K. L. Schmid, D. O’Malley, and P. Salovey. 2011. “An
Investigation of the Theoretical Content of Physical Activity Brochures.” Psychology of Sport
and Exercise 12 (6): 615–620. doi:10.1016/j.psychsport.2011.06.002.
Gellert, P., J. P. Ziegelmann, and R. Schwarzer. 2012. “Affective and Health-related Outcome
Expectancies for Physical Activity in Older Adults.” Psychology & Health 27 (7): 816–828.
doi:10.1080/08870446.2011.607236.
Godin, G., and R. J. Shephard. 1985. “A Simple Method to Assess Exercise Behavior in the
Community.” Canadian Journal of Applied Sport Sciences 10 (3): 141.
Godin, G. 2011. “The Godin-Shephard Leisure-time Physical Activity Questionnaire.” The Health
& Fitness Journal of Canada 4 (1): 18–22.
Hevel, D. J., A. J. Amorose, K. M. Lagally, A. Rinaldi-Miles, and S. Pierce. 2019. “Testing the Effects
of Messaging on Physical Activity Motivation in Active and Non-active Adults.” Psychology of
Sport & Exercise 43: 333–342. doi:10.1016/j.psychsport.2019.04.010.
INTERNATIONAL JOURNAL OF HEALTH PROMOTION AND EDUCATION
15
Knox, E. C. L., S. J. H. Biddle, I. M. Taylor, A. E. Latimer-Cheung, O. J. Webb, and L. B. Sherar.
2015. “Messages to Promote Physical Activity: Are Descriptors of Required Duration and
Intensity Related to Intentions to Be More Active?” Journal of Education and Health
Promotion 4: 77. doi:10.4103/2277-9531.171790.
Latimer, A. E., L. R. Brawley, and R. L. Bassett. 2010. “A Systematic Review of Three Approaches
for Constructing Physical Activity Messages: What Messages Work and What Improvements
are Needed?” International Journal of Behavioral Nutrition and Physical Activity 7 (1): 36.
doi:10.1186/1479-5868-7-36.
Mailey, E. L., D. Dlugonski, W. Hsu, and M. L. Segar. 2018. “Goals Matter: Exercising for
Well-being but Not Health or Appearance Predicts Future Exercise among Parents.” Journal
of Physical Activity and Health 15 (11): 857–865. doi:10.1123/jpah.2017-0469.
Mailey, E. L., J. L. Huberty, D. Dinkel, and E. McAuley. 2014. “Physical Activity Barriers and
Facilitators among Working Mothers and Fathers.” BMC Public Health 14 (1): 657. doi:10.1186/
1471-2458-14-657.
Mailey, E. L., R. Gasper, and D. Dlugonski. 2019. “Why and How Should I Exercise? A Content
Analysis of Popular Magazines.” American Journal of Health Behavior 43 (2): 349–360.
doi:10.5993/AJHB.43.2.11.
Markland, D., and D. K. Ingledew. 1997. “The Measurement of Exercise Motives: Factorial Validity
and Invariance across Gender of a Revised Exercise Motivations Inventory.” British Journal of
Health Psychology 2 (4): 361–376. doi:10.1111/j.2044-8287.1997.tb00549.x.
op den Akker, H., M. Cabrita, R. op den Akkerop den Akker, V. M. Jones, and H. J. Hermens.
2015. “Tailored Motivational Message Generation: A Model and Practical Framework for
Real-time Physical Activity Coaching.” Journal of Biomedical Informatics 55: 104–115.
doi:10.1016/j.jbi.2015.03.005.
Pope, J. P., L. Pelletier, and C. Guertin. 2017. “Starting off on the Best Foot: A Review of Message
Framing and Message Tailoring, and Recommendations for the Comprehensive Messaging
Strategy for Sustained Behavior Change.” Health Communication 33 (9): 1068–1077.
doi:10.1080/10410236.2017.1331305.
Ryan, R. M., and J. P. Connell. 1989. “Perceived Locus of Causality and Internalization: Examining
Reasons for Acting in Two Domains.” Journal of Personality and Social Psychology 57 (5):
749–761. doi:10.1037/0022-3514.57.5.749.
Segar, M. L., E. Guerin, E. Phillips, and M. Fortier. 2016. “From a Vital Sign to Vitality: Selling
Exercise so Patients Want to Buy It.” Current Sports Medicine Reports 15 (4): 276–281.
doi:10.1249/JSR.0000000000000284.
Segar, M. L., J. A. Updegraff, B. J. Zikmund-Fisher, and C. R. Richardson. 2012. “Physical Activity
Advertisements that Feature Daily Well-being Improve Autonomy and Body Image in
Overweight Women but Not Men.” Journal of Obesity Article ID: 354721.
Segar, M. L., J. S. Eccles, and C. R. Richardson. 2011. “Rebranding Exercise: Closing the Gap
between Values and Behavior.” International Journal of Behavioral Nutrition and Physical
Activity 8 (1): 94. doi:10.1186/1479-5868-8-94.
Smith, B. J., and C. M. F. Bonfiglioli. 2015. “Physical Activity in the Mass Media: An Audience
Perspective.” Health Education Research 30 (2): 359–369. doi:10.1093/her/cyv008.
Thai, C. L., J. M. Taber, A. Oh, M. Segar, K. Blake, and H. Patrick. 2019. “’Keep It Realistic’: Reactions
to and Recommendations for Physical Activity Promotion Message from Focus Groups of
Women.” American Journal of Health Promotion 33 (6): 903–911. doi:10.1177/0890117119826870.
U.S. Department of Health and Human Services. 2018. “Physical Activity Guidelines for Americans.
2nd edition”. ed. Washington, DC: U.S. Department of Health and Human Services.
Welch, N., S. A. McNaughton, W. Hunter, C. Hume, and D. Crawford. 2009. “Is the Perception of
Time Pressure a Barrier to Healthy Eating and Physical Activity among Women?” Public Health
Nutrition 12 (7): 888–895. doi:10.1017/S1368980008003066.
Willis, L. E., and S. Knobloch-Westerwick. 2014. “Weighing Women Down: Messages on Weight
Loss and Body Shaping in Editorial Content in Popular Women’s Health and Fitness
Magazines.” Health Communication 29 (4): 323–331. doi:10.1080/10410236.2012.755602.