APPLICATION OF THE SOCIAL COGNITIVE THEORY TO A WEB-BASED
PHYSICAL ACTIVITY INTERVENTION FOR OVERWEIGHT AFRICAN
AMERICAN FEMALE COLLEGE STUDENTS
By
RODNEY P. JOSEPH
DOROTHY PEKMEZI, CHAIR
NEFERTITI DURANT
GARETH DUTTON
CONNIE KOHLER
TERRI LEWIS
LORI TURNER
A DISSERTATION
Submitted to the graduate faculty of The University of Alabama at Birmingham,
in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
BIRMINGHAM, ALABAMA
2012
Copyright by
RODNEY P. JOSEPH
2012
ii
APPLICATION OF THE SOCIAL COGNTIVE THEORY TO A WEB-BASED
PHYSICAL ACTIVITY INTERVENTION FOR OVERWEIGHT AFRICAN
AMERICAN FEMALE COLLEGE STUDENTS
RODNEY P. JOSEPH
HEALTH EDUCATION – HEALTH PROMOTION
ABSTRACT
African American women report low levels of physical activity and share a
disproportionate burden of many health conditions associated with being insufficiently
active, indicating the need for innovative approaches to promote physical activity in this
population. The purpose of the current study was to evaluate changes in physical activity
and associated Social Cognitive Theory constructs of outcome expectations, enjoyment,
self-regulation, and social support following the completion of a six month, culturallyadapted, Social Cognitive Theory-based website-delivered weight loss and physical
activity intervention for young African-American women.
A secondary data analysis was performed on African-American female college
students (N=34) enrolled in a web-based weight loss and physical activity pilot study.
Bivariate regression analyses were conducted to examine associations between the Social
Cognitive Theory constructs and baseline physical activity levels. Paired t-tests were used
to assess pre-post changes in the Social Cognitive Theory variables and physical activity
from baseline to six months. Bivariate regression analyses were used to assess whether
pre-post changes in the Social Cognitive Theory variables were associated with pre-post
changes in physical activity. Due to the preliminary nature of the pilot study, statistical
significance was set at p<.10.
iii
Participants had a mean age of 21.21 (SD=2.31) years and mean BMI of 35.4
(SD=6.82). No significant bivariate relationships emerged between the Social Cognitive
Theory variables and physical activity at baseline. Intent-to-treat analyses showed
participants increased their moderate-to-vigorous intensity physical activity from 75.5
minutes/week (SD=72.9) at baseline to 92.41 minutes/week (SD=85.3) at six months
(p=.15) and reported significant increases in self-regulation for physical activity (p=.03)
and social support for physical activity from friends (p=.03). Analyses among study
completers showed an increase of 33.23 minutes/week of physical activity (SD=97.7) and
similar increases in both self-regulation and social support. Pre-post changes in the Social
Cognitive Theory variables were not significantly associated with pre-post changes in
physical activity.
Participants reported significant improvements in social support from friends and
self-regulation for physical activity. The promising physical activity findings call for
future studies with larger samples to further explore the applicability of culturally adapted
web-based approaches to promote physical activity in this understudied population.
Keywords: Physical Activity, Internet, Website, Social Cognitive Theory, African
American Women
iv
ACKNOWLEDGMENTS
I would like thank Dr. Nefertiti Durant for allowing me to collect and use data
from her Commit2Fit study in order create this dissertation. Without Dr. Durant’s
support, mentorship, and generosity, this dissertation would not have been possible. I
would also like to thank my primary mentor and advisor, Dr. Dori Pekmezi for her
mentorship, patience, and guidance over the past few years and throughout the
dissertation. Dr. Pekmezi’s passion for minority-based physical activity research is
inspiring. I would also like to extend my sincere appreciation to Dr. Terri Lewis for
serving on my dissertation committee and for her help with the statistical analyses and to
Drs. Lori Turner and Gareth Dutton for graciously serving my dissertation committee.
Lastly, I would like to acknowledge the Robert Wood Johnson Foundation for funding
Dr. Durant’s Commit2Fit study.
v
TABLE OF CONTENTS
Page
ABSTRACT ....................................................................................................................... iii
ACKNOWLEDGMENTS ...................................................................................................v
LIST OF TABLES ............................................................................................................. ix
LIST OF FIGURES ........................................................................................................... xi
CHAPTER
I. THE PROBLEM.........................................................................................................1
Introduction ..............................................................................................................1
Significance..............................................................................................................4
The Problem .............................................................................................................5
Hypotheses ...............................................................................................................6
Delimitations ............................................................................................................6
Assumptions.............................................................................................................6
Operational Definitions ............................................................................................7
Summary ..................................................................................................................8
2. BACKGROUND ....................................................................................................10
Health Benefits of Physical Activity......................................................................10
National Recommendations for Physical Activity.................................................10
Prevalence of Americans Meeting the Physical Activity Recommendations ........13
Physical Activity Assessment ................................................................................13
Promotion of Physical Activity..............................................................................16
Review of Internet-based Physical Activity Interventions ....................................17
Behavioral Outcomes and Duration of Internet-based Physical Activity
Interventions ..............................................................................................42
Theoretical Basis of Internet-based Physical Activity Interventions .........42
Behaviors Targeted by Internet-based Physical Activity Interventions ....43
Internet-based Strategies Used to Promote Physical Activity ...................44
Populations Targeted by Internet-based Physical Activity Interventions ..46
Internet-based Physical Activity Interventions Among College Students .47
Summary of Past Research on Internet-based Physical Activity
Interventions ..............................................................................................52
vi
Application of the Social Cognitive Theory to Physical Activity Promotion .......53
Cultural, Environmental, and Individual Factors Associated with Physical
Activity in African American Women ..................................................................57
Selection of a Web-based Approach for the Current Study...................................60
Addressing the Concept of Digital Inequality .......................................................61
Description of the Web-based Approach in the Current Study .............................63
Commit2Fit Website ..................................................................................64
Structured Exercise Sessions .....................................................................67
Summary ................................................................................................................67
3. METHODS .............................................................................................................69
Introduction ............................................................................................................69
Study Design ..........................................................................................................69
Participants .............................................................................................................70
Procedure ...............................................................................................................71
Outcome Measures.................................................................................................72
Physical Activity ........................................................................................72
Outcome Expectations ...............................................................................75
Enjoyment ..................................................................................................76
Social Support ............................................................................................77
Self-Regulation ..........................................................................................77
Covariate Measures ....................................................................................78
Statistical Analyses ................................................................................................79
Specific Aim 1 ...........................................................................................79
Specific Aim 2 ...........................................................................................80
Specific Aim 3 ...........................................................................................81
Missing Data ..............................................................................................81
Power and Sample Size..............................................................................82
Summary ................................................................................................................84
4. RESULTS ...............................................................................................................85
Introduction ............................................................................................................85
Participants .............................................................................................................85
Sample Characteristics ...............................................................................85
Completers versus Non-completers ...........................................................89
Website Usage and Exercise Session Attendance .................................................91
Corroboration of Seven Day Physical Activity Recall with Accelerometry .........92
Reliability Estimates of the Social Cognitive Theory Variables ...........................96
Specific Aim 1 .......................................................................................................97
Preliminary Data Analyses ........................................................................97
Regression Analyses ..................................................................................99
Specific Aim 2 .....................................................................................................102
vii
Preliminary Data Analyses ......................................................................102
Paired T-test Analyses .............................................................................102
Paired Proportion Analyses......................................................................105
Post-hoc Power Analyses .........................................................................107
Specific Aim 3 .....................................................................................................107
Preliminary Analyses ...............................................................................107
Paired T-test Analyses .............................................................................107
Outcome Expectations ......................................................................110
Enjoyment for Physical Activity ......................................................110
Self-Regulation .................................................................................111
Social Support from Family .............................................................112
Social Support from Friends .............................................................113
Post-hoc Power Analyses .........................................................................118
Regression Analyses between Social Cognitive Theory Variables and
Physical Activity ......................................................................................120
Summary ..............................................................................................................121
5. DISCUSSION, CONCLUSIONS, AND PUBLIC HEALTH IMPLICATIONS .123
Introduction ..........................................................................................................123
Summary of Findings ...........................................................................................124
Discussion and Conclusions ................................................................................125
Comparison of Physical Activity and Social Cognitive Theory Findings
with other Studies ....................................................................................126
Sensitivity of the Social Cognitive Theory Psychosocial Measures .......130
Corroboration of Seven Day Physical Activity Recall with
Accelerometers .......................................................................................131
Website Usage and Structured Exercise Session Attendance ..................132
Power and Sample Size............................................................................133
Strengths, Limitations, and Public Health Implications ......................................134
Study Strengths ........................................................................................134
Study Limitations .....................................................................................136
Public Health Implications .......................................................................138
LIST OF REFERENCES .................................................................................................140
APPENDICES
A
INSTITUTIONAL REVIEW BOARD APPROVAL ..................................156
B
DATA COLLECTION INSTRUMENTS ....................................................158
C
COMMIT2FIT WEBSITE SCREENSHOTS...............................................166
viii
LIST OF TABLES
Tables
Page
1
Internet-based Physical Activity Studies ...............................................................19
2
Internet-based Physical Activity Studies among College Students .......................49
3
Overview of the Application the of Social Cognitive Theory to Intervention
Components ...........................................................................................................66
4
Demographic Characteristics of Participants at Baseline ......................................88
5
Baseline Comparison of Study Completers versus Non-Completers ....................91
6
Mean Minutes per Week of Moderate-intensity Physical Activity Data Collected
by the Seven Day Physical Activity Recall and Accelerometers ..........................94
7
Correlations between Self-Reported and Accelerometer Measured Physical
Activity at Baseline ................................................................................................95
8
Correlations between Self-Reported and Accelerometer Measured Physical
Activity at Six Month Follow-up ...........................................................................96
9
Inter-item Reliability Estimates for Social Cognitive Theory Variables for all
Assessment Periods ................................................................................................97
10
Pearson Bivariate Correlations among Social Cognitive Theory Variables at
Baseline ..................................................................................................................99
11
Bivariate Regression Outcomes between the Social Cognitive Theory Variables
and Physical Activity at Baseline ........................................................................100
ix
12
Odds Ratios for the Social Cognitive Theory Constructs Predicting Achievement
of 150 Minutes of Physical Activity per Week ....................................................101
13
Mean Self-reported Physical Activity Levels at Baseline, Midpoint, and Six
Months .................................................................................................................103
14
T-tests for Changes in Self-reported Physical Activity .......................................105
15
Paired Proportion Analyses of Participants who Achieved 150 Minutes of
Moderate-intensity Physical Activity at Baseline and at Six Months..................106
16
Mean Values for Social Cognitive Theory Variables for all Assessment
Periods..................................................................................................................109
17
Changes in Social Cognitive Theory Variables from Baseline to Six Months ....115
18
Changes in Social Cognitive Theory Variables from Baseline to Midpoint .......116
19
Changes in Social Cognitive Theory Variables from Midpoint to Six Months ...117
20
Post-hoc Power Analyses for Changes in the Social Cognitive Theory Variables
from baseline to Six Months ................................................................................119
21
Bivariate Associations between select Social Cognitive Theory Change Scores
and Changes in Physical Activity .......................................................................121
x
LIST OF FIGURES
Figure
Page
1
Participant Recruitment and Retention Flow Diagram ..........................................87
2
Mean Physical Activity Scores across Assessment Periods ................................104
3
Mean Outcome Expectations for Physical Activity Scores across Assessment
Periods..................................................................................................................110
4
Mean Enjoyment for Physical Activity Scores across Assessment Periods ........111
5
Mean Self-Regulation Scores across Assessment Periods...................................112
6
Mean Social Support from Family Scores across Assessment Periods ...............113
7
Mean Social Support from Friends Scores across Assessment Periods ..............114
xi
CHAPTER 1
THE PROBLEM
Introduction
Physical and mental health benefits of regular physical activity have been well
documented during the past 20 years. Physical activity has established benefits of
preventing and treating many adverse health conditions such as heart disease, type II
diabetes, osteoarthritis, osteoporosis, depression, and anxiety (American College of
Sports Medicine [ACSM], 2007; U.S. Department of Health and Human Services
[USDHHS], 2008). Regular physical activity has also been shown to be a strong predictor
in the promotion and maintenance of weight loss (Jakicic & Otto, 2005). The Centers for
Disease Control and Prevention (2008) and American College of Sports Medicine (2007)
currently recommend that adults engage in at least 150 minutes of moderate-intensity
physical activity each week and suggest that health benefits can be achieved in bouts as
short as ten minutes. However, despite the overwhelming positive evidence for
performing regular physical activity, many Americans do not engage in physical activity
at recommended levels.
Recent data indicate that only 48% of Americans achieve the recommended
physical activity levels (Centers for Disease Control and Prevention [CDC], 2010).
Moreover, 38% of Americans perform insufficient amounts of physical activity and
13.5% are inactive. Demographic disparities also exist among those who achieve
adequate levels of physical activity. Forty-one percent of African Americans report
1
performing 150 minutes of moderate-intensity physical activity each week, which is an
approximate 10% lower prevalence than their White counterparts (CDC, 2010).
Furthermore, only 36% of African American females achieve the recommended levels of
physical activity, representing the lowest prevalence for any race and sex demographic
group. Given the low physical activity levels among African American females, it is not
surprising that this population demonstrates a disproportionate disease burden of many
health conditions associated with being insufficiently active. For example, African
American women have higher incidence and mortality rates for colon cancer, type II
diabetes, and cardiovascular disease than White women and have a higher mortality rate
than White women following a breast cancer diagnosis (U.S. Cancer Statistics Working
Group, 2010).
Disparate levels of physical activity among African American females appear to
emerge at an early age. Findings from an eight-year prospective cohort following 2379
adolescent girls beginning at age 10 showed that habitual physical activity levels declined
100% among African Americans over the course of the study in comparison to 64% in
Whites (Kimm et al., 2002). Additional research has shown that physical activity levels
across all populations peak when people are in their late teenage years or early twenties
and continually decline with age (Caspersen, Pereira, & Curran, 2000; CDC, 2010);
indicating a key period in life in which to intervene and promote physical activity.
Internet-based physical activity interventions represent a potential method in
which to intervene and promote physical activity due to the Internet’s widely available
accessibility and ability to reach a large number of people at relatively low-cost (Marcus
et al., 2006). Internet-based physical activity interventions have not only dramatically
2
increased over the past decade, but have shown positive outcomes for increasing physical
activity (Vandelanotte, Spathonis, Eakin, & Owen, 2007). However, little is known about
the efficacy of these approaches among African American women.
The systematic review of Internet-based physical activity interventions conducted
for the current study identified over 60 studies evaluating an Internet and/or web-based
approach to promoting physical activity (see Table 1 on page 19-41). Of these, only one
study (Pekmezi et al., 2010) evaluated the outcomes of a web-based approach to
promoting physical activity exclusively among African American women. Results of this
study supported the use of the web-based approach to promoting physical activity among
African American women; however, the participants of the study were a sub-set of
middle-aged African American women enrolled in a much larger physical activity
intervention. No published studies have examined the efficacy a culturally-adapted webbased intervention specifically designed to promote physical activity among African
American women.
Thus, the lack of research on Internet-based approaches to promoting physical
activity among African American women indicates the need for further research to
explore the acceptability and feasibility these approaches in this underserved population.
The present study adds to the paucity research on web-based approaches to promoting
physical activity among African American women by being the first study (to the authors
knowledge) to evaluate a prospectively designed, culturally-adapted web-based
intervention promoting physical activity among African American women. The purpose
current study was to evaluate the physical activity outcomes of a culturally-adapted webbased weight loss and physical activity pilot intervention for young African American
3
women between ages 19 to 30 years. Specifically, the study assessed the intervention’s
impact on self-reported physical activity and associated Social Cognitive Theory
constructs of outcome expectations, enjoyment, social support, and self-regulation.
Significance
Physical inactivity is an independent risk factor for various health conditions such
as cardiovascular disease, type II diabetes, obesity, and select cancers (ACSM, 2007;
USDHHS, 2008). African Americans report performing the lowest levels of physical
activity when compared to any other race or sex demographic groups (CDC, 2010) and
share a disproportionate burden of the aforementioned diseases associated with physical
inactivity (U.S. Cancer Statistics Working Group, 2010). Intervening to promote physical
activity among African American women during late adolescence and young adulthood
provides the opportunity to establish lifelong physical activity patterns, reduce peak
weight gains, and ultimately reduce physical activity related health disparities in this
population.
Internet-based physical activity interventions represent a potential method in
which to help resolve these physical activity related disparities. Internet-based
interventions have shown positive outcomes for promoting physical activity across
diverse populations (Hamel, Robbins, & Wilbur, 2011; Lau, Lau, Wong del, & Ransdell,
2011; van den Berg, Schoones, & Vliet Vlieland, 2007; Vandelanotte et al., 2007);
however, no published web-based interventions have been designed to address the
specific cultural, environmental, and social factors associated with physical activity
among African American women. The current study evaluated a culturally-adapted web-
4
based pilot study promoting physical activity among young African American women
aged 19 to 30. Results of this study expand the limited knowledge on Internet-based
approaches to promoting physical activity among African American women.
The Problem
The purpose of the current study was to evaluate pre-post intervention changes in
physical activity and associated Social Cognitive Theory constructs of outcome
expectations, enjoyment, self-regulation, and social support following the completion of a
six month web-based physical activity promotion pilot study. Evaluation of these specific
Social Cognitive Theory variables was selected because these constructs were explicitly
targeted by the intervention components. The specific aims of the study were as follows:
1. Examine the relationship between Social Cognitive Theory constructs related to
physical activity (outcome expectations, self-regulation, social support, and
enjoyment) and self reported physical activity levels as measured by the Seven
Day Physical Activity Recall at baseline.
2. Assess self-reported changes in physical activity levels, as measured by the Seven
Day Physical Activity Recall, from baseline to the six month follow-up.
3. Assess changes in Social Cognitive Theory variables from baseline to six months
and evaluate how these changes are associated with changes in physical activity
levels from baseline to six months.
5
Hypotheses
The corresponding hypotheses were postulated for each of the aforementioned
specific aims:
1. Participants with greater outcome expectations, self-regulation, social support,
and enjoyment for physical activity will demonstrate higher physical activity
levels at baseline.
2. Participants will report significant increases in physical activity from baseline to
six months.
3. Improvements in Social Cognitive Theory variables will be associated with
improvements in self-reported physical activity levels at six months.
Delimitations
In this study, only moderate-to-vigorous intensity physical activity performed in
bouts of ten minutes or greater were included in the self-reported physical activity
outcomes (the primary physical activity outcome of the study). This activity threshold
was selected from recommendations from the Centers for Disease Control (2008) and
American College of Sports Medicine (2007) that moderate-to-vigorous intensity activity
must be performed for in bouts of at least ten minutes in order to achieve health benefits.
Assumptions
It was assumed that the web-based intervention was delivered to each of the
participants in the exact same manner; that is, the algorithms of the web-based
intervention performed identically for all participants. It was also assumed that the
6
primary outcome of self-reported physical activity, assessed by Seven Day Physical
Activity recall, was obtained using the exact same interview technique for all participants
and that the interviewer did not provide any leading questions or cues that introduced bias
into physical activity outcomes. Similarly, it was assumed that participants truthfully
reported (to the best of their ability) their physical activity levels and thoroughly read,
comprehended, and answered all psychosocial surveys assessing the Social Cognitive
Theory constructs in a truthful and honest manner. Lastly, it was assumed that
participants did not enroll in any other physical activity promotion program while
participating in the study (which was an exclusion criterion at baseline).
Operational Definitions
Accelerometer: a light-weight activity monitor that is worn on a person’s hip to
assess distance, intensity, and duration of physical activity performed.
African American: an American who has Black African ancestors or who selfidentifies and an African American.
Enjoyment of Physical Activity: the affective response (positive or negative)
associated with physical activity.Moderate-to-vigorous intensity physical activity:
physical activity performed at an intensity that is at least three times greater than the
energy expended at rest.
Obese/Obesity: an adult with a body mass index over 30.
Outcome Expectations for Physical Activity: anticipated outcomes from
performing physical activity.
Overweight: an adult with a body mass index between 25 and 29.9.
7
Physical Activity: any bodily movement produced by the contraction of skeletal
muscle that increases energy expenditure above a basal level.
Self-Regulation for Physical Activity: an individual’s ability to manage his/her
own social, cognitive, and motivational processes in order to perform physical activity.
Social Cognitive Theory: behavioral health theory developed by Albert Bandura.
The theory explains behavior in a triadic and reciprocal model, known as reciprocal
determinism, where the environment, an individual, and behavior itself continually
interact to produce behavior. Constructs of the Social Cognitive Theory include (but not
limited to): social support, self-regulation, outcome expectations, outcome expectancies,
self-efficacy, observational learning, and reinforcement.
Social Support for Physical Activity: the extent to which an individual’s
significant referents (family, friends, peers) approve, encourage, and/or influence
performance of physical activity.
Study completers: participants who provided data at all three assessment points
(baseline, midpoint, and follow-up), logged onto the study website a minimum of once,
and attended at least one structured exercise session during the course of the study.
Summary
African American women report performing low levels of physical activity when
compared to other race or sex demographic groups and also share a disproportionate
burden of various diseases associated with insufficient physical activity levels, indicating
the need for focused physical activity promotion efforts for this population. Internetbased physical activity interventions represent a potential method in which to intervene
8
and promote physical activity due to their accessibility and ability to reach a large
number of people at a relatively low cost. However, few studies have investigated the
efficacy of web-based physical activity interventions exclusively among the African
American population; therefore, the feasibility of web-based approaches to promoting
physical activity in this population are unknown. The current study evaluated the physical
activity and Social Cognitive Theory outcomes of a six month, culturally-adapted webbased pilot study promoting physical activity among young African American women
aged 19 to 30. Promoting physical activity in this defined age group provides the
opportunity for the establishment of lifelong physical activity patterns; which can
positively alter weight and chronic disease trajectories commonly observed in this
population and may ultimately reduce physical activity related health disparities among
African American women.
9
CHAPTER 2
BACKGROUND
Health Benefits of Physical Activity
Performance of regular physical activity provides both physical and psychological
health benefits. Perhaps, the most significant benefit of physical activity is that it is an
independent risk factor for reducing all cause mortality (USDHHS, 2008). Thus, the more
physically active people are, the less likely they are to die prematurely. Additional
physical benefits of physical activity include prevention and treatment of many health
conditions, including: cardiovascular disease, type II diabetes, osteoarthritis, metabolic
syndrome, stroke, high blood pressure, and select cancers such as breast and colon (CDC,
2010). Physical activity is also a strong predictor in weight maintenance and can assist
with weight loss when combined with diet (USDHHS, 2008; Jakicic & Otto, 2005; 2008).
Psychological benefits of regular physical activity include the reduction of depression
symptoms, increased cognitive functioning, and improved quality of life (CDC, 2010).
National Recommendations for Physical Activity
The U.S. Department of Health and Human Services established the first national
recommendations for physical activity with the 2008 publication entitled 2008 Physical
Activity Guidelines for Americans. This publication provides a comprehensive framework
for the intensity, frequency, and duration of physical activity needed to provide important
health benefits. Specifically, this publication recommends that Americans perform
10
strength training activities at least two days per week and achieve one of the following
guidelines for aerobic physical activity each week:
1. 150 minutes of moderate-intensity physical activity; or
2. 75 minutes of vigorous-intensity activity; or
3. A combination of both moderate- and vigorous-intensity physical activity
that is equivalent to meeting the aforementioned moderate and vigorous
intensity recommendations.
The guidelines further suggest that in order to meet these aerobic physical activity
recommendations, physical activity should be performed on most days of the week and in
bouts of ten minutes or greater.
The 2008 Physical Activity Guidelines for Americans (USDHHS, 2008) provides
two methods in which to measure intensity of physical activity: absolute-intensity and
relative intensity. Absolute-intensity refers the amount of energy expended per minute of
activity. Accordingly, in order to achieve moderate-intensity activity, individuals should
perform activity at approximately three to six times greater than they do at rest; for
vigorous activity, physical activity should be performed at an intensity that is six times
higher than at rest. Relative-intensity refers to an activity measure where perceived effort
is assessed in order to estimate physical activity levels. The 2008 Physical Activity
Guidelines for Americans recommend that people use a scale from one to ten when
assessing relative-intensity. On this scale, a rating of a one refers to being at rest and a
rating of ten is considered very intense. Thus, moderate-intensity should be performed at
a perceived effort of five or six and vigorous-intensity at a level of seven or eight.
11
Physical activity intensity can also be measured as a percentage of one’s maximal
heart rate. Using this method, moderate-intensity activity is defined as working at a rate
that is between 55% and 69% of a person’s maximal heart rate (ACSM, 2007); vigorousintensity physical activity is performed at a heart rate of 70% or greater than a person’s
maximal heart rate.
The types of aerobic physical activity that can be performed in order to meet the
national recommendations are endless. People should not solely focus on structured
exercise as the only mode of physical activity to meet the national guidelines as there are
a variety of physical activities that can be performed. Examples of moderate-intensity
activities include ballroom or line dancing, walking at a rate of approximately three miles
per hour, gardening, cycling at a rate of approximately 10 miles per hour, and playing
sports such as tennis or volleyball (USDHHS, 2008). Examples of vigorous-intensity
include running, hiking up-hill, biking greater than 10 miles per hour, aerobic dancing,
and jumping rope (USDHHS, 2008).
Regarding the strength training recommendations, the 2008 Physical Activity
Guidelines for all Americans suggest that adults engage in strength building exercises a
minimum of two days per week. Health benefits of participating in muscle strengthening
activities include improvement and maintenance of muscular fitness and bone mass
(USDHHS, 2008). In order to achieve health benefits from muscle strengthening
activities, exercises should be performed at a moderate-intensity and target all major
muscle groups (i.e. chest, abdominals, shoulders, legs, and back). The guidelines do not
specify how much time should be spent on muscle strengthening activities; rather, the
publication suggest that strengthening exercises for each major muscle group be
12
performed to a point to where it would be difficult to do another repetition without
assistance. Examples of muscle strengthening activities include weight training and body
weight calisthenics (i.e. push-ups, sit-ups).
Prevalence of Americans Meeting the Physical Activity Recommendations
Approximately 48% of Americans report performing the recommended physical
activity levels. However, disparities exist among those achieving the recommended
activity levels. On average, males report a higher prevalence of achieving the
recommended physical activity levels than females. Among races, African Americans
report lower physical activity levels than Whites or Hispanics, with only 41% of this
population achieving 150 minutes of moderate-to-vigorous intensity activity per week
(compared to 53% Whites and 42% Hispanics) (U.S. Cancer Statistics Working Group,
2010). Furthermore, African American women demonstrate the lowest physical activity
levels out of any race or sex group with 36% of this population achieving the
recommended levels (compared to 50% of White females and 42% of Hispanic females);
indicating the need for programs targeted at increasing physical activity in this
population.
Physical Activity Assessment
Methods of assessing physical activity can be classified into two main categories:
subjective and objective. Subjective measures are the most frequently used methods to
assess physical activity and typically involve self-report or interviewer-administered
recall questionnaires or activity logs (Sallis & Saelens, 2000). Objective measures do not
13
require participant recall; instead, these measures assess physical activity in real-time
when people are actually performing physical activity. Examples of objective physical
activity measures include pedometers, accelerometers, and assessment by doubly-labeled
water.
Both subjective and objective measures of physical activity have their own sets of
strengths and weaknesses. Advantages of subjective physical activity assessment include
not requiring participants to wear an obtrusive device in order to assess physical activity,
not relying on participants to actually wear the devices to obtain an assessment, and the
ability to assess the types and modes of physical activity performed (i.e. swimming,
running, cycling, etc) (Shephard, 2003). Additionally, subjective measures of physical
activity are generally a less expensive way to collect physical activity data when
compared to the cost of purchasing accelerometers and pedometers, especially when
assessing physical activity among large sample sizes. Weaknesses of subjective physical
activity assessments include social desirability bias, which often leads to participants
over-reporting their physical activity levels, difficulty of participants accurately
remembering how much activity they performed, and participant error in understanding
and/or assessing the intensity level of physical activity performed (light, moderate,
vigorous) (Janz, 2006; Sallis & Saelens, 2000; Shephard, 2003).
Strengths of objectively measured physical activity assessments include the lack
of human reporting error, avoidance of social desirability bias, and the standardization of
assessing and reporting physical activity across studies (i.e., minutes of daily activity
performed) (Janz, 2006). The standardization of assessing and reporting physical activity
allows for the comparison of physical activity outcomes across studies. Weaknesses of
14
objective physical activity measures include the intrusiveness of wearing an activity
monitor or pedometer, lack of monitor sensitivity to assess upper limb physical activity,
and the inability of these devices to collect the mode/type of activity performed as most
objective measures are limited to assessing distance and/or intensity of physical activity
(Janz, 2006). Another major weakness of objective physical activity assessment is that
participants must wear the monitor in order to collect physical activity data. Therefore, if
participants do not to wear the monitor as prescribed, the validity of the data is
compromised.
Due to the strengths and weaknesses of both subjective and objective physical
activity assessment, researchers have advised the use of both types of assessment to
collect physical activity data (Sallis & Saelens, 2000; Schmidt, Cleland, Thomson,
Dwyer, & Venn, 2008). Assessing physical activity using both methods allows for the
comparison of the subjective and objective findings, which aids in the interpretation of
physical activity performed by the sample population. For example, subjective
assessment measures can provide meaningful information on the context and types of
physical activity performed while objective measures can provide information on the
quantity and intensity level of activity performed (Sallis & Saelens, 2000). Additionally,
comparing both methods allows researchers to identify if the sample is over- or underestimating their physical activity levels via subjective assessment by comparison to the
objective measure.
15
Promotion of Physical Activity
Over the past two decades, researchers have used various strategies to intervene
and promote physical activity. Early strategies employed face-to-face and small group
counseling, mailed print-materials (i.e. newsletters, flyers, pamphlets), and telephone
counseling. Reviews of these strategies have shown moderate-to-strong support for these
methods in promoting physical activity (Dunn, Andersen, & Jakicic, 1998; Goode,
Reeves, & Eakin, 2012; Marcus, Owen, Forsyth, Cavill, & Fridinger, 1998; Marshall,
Owen, & Bauman, 2004). In recent years, advancements in computer technology have
lead researchers to explore technology-based delivery channels such as email and the
Internet to promote physical activity.
Internet and web-based technology holds promise for the delivery of physical
activity promotion programs because many Americans already use this type of
technology on a daily basis. Statistics show that approximately 71% of Americans have
household Internet access (U.S. Department of Commerce, 2011). Furthermore, this
percentage increases to 77% for those who have attended some college, and to 89% for
those who have a college degree (U.S. Department of Commerce, 2011). Advantages of
web-based technology to promote physical activity include: a) the potential to reach a
large number of individuals at a relatively low cost; b) the ability to provide 24-hour
access to intervention materials which may increase convenience, access, and exposure of
intervention messages; and c) the capacity to instantaneously deliver tailored intervention
messages to participants without the delays commonly found in print or telephone
delivered interventions (Marcus, Nigg, Riebe, & Forsyth, 2000).
16
Previous systematic reviews of Internet-based physical activity interventions have
shown positive outcomes (Hamel et al., 2011; Lau et al., 2011; van den Berg et al., 2007;
Vandelanotte et al., 2007). However, five years has passed since a systematic review has
examined the state of the literature on Internet-based approaches to promoting physical
activity among adult populations (reviews by Hamel, Robbins, & Wilbur, 2011 and Lau,
Lau, Wong del, & Ransdel, 2011 evaluated child/adolescent populations). Therefore, an
updated review of Internet-based physical activity studies was conducted in order to
provide insight on the current status of the literature for Internet and web-based
approaches to promoting physical activity.
Review of Internet-based Physical Activity Interventions
The literature review of Internet-based physical activity studies performed for the
current study focused exclusively on Internet and web-based interventions targeting adult
populations. Therefore, studies focusing on child and adolescent populations as well as
those focusing on elderly populations were excluded. Additionally, Internet-based
physical activity interventions for individuals with psychological disorders (i.e.
Schizophrenia or any other mental illness) were excluded. Articles were located using the
PubMed search engine and the Boolean AND search strategy. Keywords used to identify
articles included: “physical activity”, “exercise” “Internet”, “web-based”, and “email.”
Reference lists of the articles retrieved during search procedures were also reviewed in
order to identify additional Internet-based physical activity studies not retrieved during
the online search process. In total, the search process identified 60 unique Internet-based
17
physical activity interventions meeting the search criteria. A brief summary for each
study reviewed is located in Table 1.
18
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Bennett,
Broome,
Schwab-Pilley,
& Gilmore,
2011
Booth, Nowson,
& Matters, 2008
N=145 worksite
managers; mean
age 41.5; 64%
female; 82% White
Randomized
controlled trial;
Groups: a) access
to website, b)
control
Randomized
controlled trial of
two web-based
groups; Groups: a)
web-based weight
reduction program
plus exercise, b)
web-based exercise
only
Exercise;
nutrition; weight
loss
6 months
Not reported
Physical activity;
weight loss
12 weeks
Goal-Setting
Theory
Adapted version
of the Active
Australian
Survey;
pedometers
Bosak, Yates, &
Pozehl, 2009,
2010
N=22 adults with
metabolic
syndrome; mean
age: intervention
group 41.3 years,
usual care group
50.4 years; 27%
female; 91% White
Randomized
controlled trial;
Groups: a) webbased intervention
with email
feedback on goals,
b) usual care
Physical activity
6 weeks
Self-efficacy
Theory
Seven Day
Physical
Activity Recall;
accelerometers
N=53 overweight
Australian adults;
mean age 46.3
years; 79% female;
race not reported
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Godin Leisure
Time Exercise
Questionnaire
Major Findings
Intervention
group
significantly
increased exercise
Both groups
significantly
increased steps
per day via
pedometers (no
between group
differences); no
significant
increases in selfreported physical
activity for either
group
No improvement
in physical
activity for either
group
19
Table 1.
Internet-based physical activity studies.
20
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Accelerometers
Carlson, Sallis,
Ramirez,
Patrick, &
Norman, 2012
N=352
overweight/ obese
adults; mean age:
intervention group
44.5 years; control
group 42.2 years;
52% female; 70%
White
Physical activity;
dietary behaviors
12 months
Social Cognitive
Theory;
Transtheoretical
Model
Carr et al.,
2008;
Carr et al., 2009
N=32 adults; mean
age: intervention
group 41.4 years;
49.4 for control;
Percent
male/female not
described; race not
reported
Randomized
controlled trial;
Groups: a) access
to website and
received monthly
tailored feedback
via telephone on
progress, b)
waitlist control for
males and general
information for
females
Randomized
controlled trial;
Groups: a) ALED-I
web-based
intervention paired
with a virtual
partner, b) delayed
to treat control
Physical activity;
reduction of
cardiovascular
risk
16 weeks
Social Cognitive
Theory,
Transtheoretical
Model
Pedometers
Carr et al., 2012
N= 66 sedentary
health adults; mean
age: enhanced
internet 38.5 years;
standard internet
36.8 years; 87%
Randomized
controlled trial of
two internet-based
interventions: a)
enhanced internet,
b) standard internet
Physical activity
6 months
Social Cognitive
Theory
Seven Day
Physical
Activity Recall
Major Findings
Intervention
group
significantly
increased physical
activity
Web-based group
significantly
increased
steps/day at the
end of 16 weeks;
at 8 month
follow-up
participants
regressed back to
baseline step
levels.
At 3 months, the
enhanced internet
group had a
significantly
greater increase in
physical activity
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
White
Dinger, Heesch,
Cipriani, &
Qualls, 2007
N=56
insufficiently
active females;
mean age 41.5
years; 86% White
21
Randomized
controlled trial of
two emaildelivered
pedometer
interventions;
Groups: a) weekly
emails reminders to
wear pedometers
and return activity
logs, b) same
weekly emails as
group a plus
Transtheoretical
Model-based
messages to
increase steps
Physical activity
6 weeks
Transtheoretical
Model
Physical
Activity
Questionnaire
International
Physical
Activity
Questionnaire
(IPAQ) Shortform
Major Findings
compared to the
standard internet
group; at 6
months, both
groups had
significant
increases in
physical activity
compared to
baseline (no
between group
differences)
Both groups
significantly
increased physical
activity at followup; no difference
between groups.
Table 1.
Internet-based physical activity studies.
22
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Accelerometry;
International
Physical
Activity
Questionnaire
(IPAQ); Godin
Leisure-Time
Exercise
Questionnaire
Dlugonski,
Motl, &
McAuley, 2011
N=21adults with
multiple sclerosis;
mean age 46.4
years; 90% female;
90% White
One-group pre-post
design. Participants
received website
access, participated
in online chat
groups with other
participants, and
email support.
Physical activity
12 weeks
Social Cognitive
Theory
Dunton &
Robertson, 2008
N=156 adult
females; mean age
42.8 years; 65%
White
Randomized
controlled trial;
Groups: a) access
to website and
individually
tailored weekly
emails, b) wait-list
control
Physical activity
3 months
Health Belief
Model;
Transtheoretical
Model
Standardized
activity
inventory format
(Hopkins 1991)
Faghri et al.,
2008
N=206 state
worksite
employees; mean
age not reported;
81.8% female;
59% White
One group pre-post
design assessing
the effects of a
website and email
based pedometer
walking
intervention.
Physical activity
(steps per day)
10 weeks
Transtheoretical
Model
Pedometers,
self-report
physical activity
with a nonspecified
measure
Major Findings
Significant
improvement in
physical activity
at 12 weeks
measured by
accelerometers
and the IPAQ. No
improvement in
activity assessed
by the Godin
Leisure-Time
Exercise
Questionnaire
Both groups
significantly
increased activity;
however, Internet
group increased at
a greater rate.
No improvement
steps per day at 10
weeks.
Table 1.
Internet-based physical activity studies.
23
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Active Australia
Questionnaire
Ferney,
Marshall, Eakin,
& Owen, 2009
N=106 inactive
Australian adults;
mean age 52 years;
72% female; race
not reported
Physical activity
26 weeks
Social Cognitive
Theory
Franko et al.,
2008
N=476 college
students from 6
universities; mean
age 20.1 years;
56.3% female;
58% White
Physical activity;
nutrition
6 months
Not Reported
International
Physical
Activity
Questionnaire
(IPAQ)
Glasgow, Boles,
McKay, Feil, &
Barrera, 2003
N=320 type II
diabetes patients;
mean age 59 years;
Percent
male/female not
described; race not
Randomized
controlled trial;
Groups: a) access
to a neighborhood
environmentfocused website, b)
motivationalinformation
website; both
groups received
emails
Randomized
controlled trial;
Groups: a) My
Student Body webbased intervention
(2 website visits
required), b) My
Student Body webbased intervention
plus booster session
(3 website visits),
c) attention control
Randomized
controlled trial of
D-Net Diabetes
Program: 1) info
only website
(control), 2)
Nutrition; blood
profiles; physical
activity
10 months
Self-Efficacy
Theory; Social
Support Theory
Physical
Activity Scale
for the Elderly
Major Findings
Both groups
significantly
increased physical
activity from
baseline;
however, the
neighborhood
group had a
significantly
greater increase in
physical activity
No improvement
in physical
activity for any
group.
No improvement
in physical
activity for any
group
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
24
Physical
Activity
Measure
Major Findings
4 months: Both
intervention
groups
significantly
increased physical
activity compared
to baseline;
12 months: both
intervention
groups
significantly
increased physical
activity (no
difference
between
intervention
groups)
No significant
increase in
reported
Tailored selfmanagement
training with
website and twice
per week online
coaching, 3) Peer
support internet
groups with basic
website
Glasgow et al.,
2010; Glasgow
et al., 2012
N=463 type II
diabetes patients;
mean age 58.4
years; 49.8%
female; 72% White
Randomized
controlled trial;
Groups: a) website
access only, b)
website access plus
in-person group
support visit, c)
usual care website
access (information
only)
Physical activity;
eating behaviors;
biological
outcomes
4 months
initially
reported; 12
months
reported in
subsequent
publication
Social
Ecological
Theory; SelfManagement
Model
Community
Health Activities
Model Program
for Seniors
Questionnaire
(CHAMPS)
Gow, Trace, &
Mazzeo, 2010
N=170 first year
college students;
Randomized
controlled trial;
Physical activity;
prevention of
6 weeks
Social Cognitive
Theory
International
Physical
Table 1.
Internet-based physical activity studies.
Study
25
Sample
Characteristics
Study Design
mean age not
reported; 73.7%
female; 54% White
(22% Black)
Groups: a) no
treatment, b) access
to website only, c)
6-week weight and
caloric feedback
emails, d)
combined access to
website and
feedback
Randomized
controlled trial;
Groups: a) access
to website, b) inperson class to
increase physical
activity, c) inperson health class
Randomized
controlled trial after
6 month face-toface obesity
treatment; Groups:
of: a) Internet
support with
weekly emails and
internet chat
sessions with
therapist/other
participants, b)
Grim, Hortz, &
Petosa, 2011
N=233 college
students; mean age
21.2 years; 82%
female; 82% White
Harvey-Berino,
Pintauro,
Buzzell, et al.,
2002
N=122 overweight
healthy adults;
mean age 48.4
years; 85% female;
97% White
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
weight gain
Physical
Activity
Measure
Activity
Questionnaire
(IPAQ) – Short
Form
Major Findings
physical activity
for any study
group
Physical activity
College
semester
Social Cognitive
Theory
Seven Day
Physical
Activity Recall
Both the webbased and
physical activity
class significantly
increased physical
activity
Weight loss;
physical activity
6 month inperson obesity
treatment
followed by 12month
maintenance
Not reported
Paffenbarger
Physical
Activity
Questionnaire
All groups
increased physical
activity during
treatment; only inperson support
maintained an
increase at end of
maintenance
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Harvey-Berino,
Pintauro, &
Gold, 2002
N=46 overweight
health adults; mean
age 46.3 years;
80% female;
predominately
White
Harvey-Berino,
Pintauro,
Buzzell, &
Gold, 2004
N=255 healthy
overweight or
obese adults, mean
age 45.8 years;
82% female; race
not reported
Study Design
frequent in-person
support, c) minimal
in-person support
Randomized
controlled trial after
15 week weight
loss study; Groups:
a) internet support
with email and
online chat sessions
with therapist, b)
in-person support,
3) no support.
Randomized
controlled trial after
6-month behavioral
weight loss
program; Groups:
a) internet support,
b) frequent inperson support plus
telephone calls, c)
minimal in-person
contact
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Major Findings
Weight loss
maintenance;
Physical activity
15 week weight
loss treatment
followed by 22
week
maintenance
program.
Not reported
Paffenbarger
Physical
Activity
Questionnaire
Internet group had
an increase of 419
in calories burned
from physical
activity, no
mention if this
increase was
significant
Physical activity;
weight loss
12 months post
6 month
behavioral
weight loss
program
Not reported
Paffenbarger
Physical
Activity
Questionnaire
All groups
significantly
increased physical
activity
26
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Irvine et al.,
2011
N=221
manufacturing
plant employees;
mean age 45 years;
42.2% female;
79% White
Randomized
controlled trial;
Groups: a) access
to the Get Moving
website, b) no
contact control
Physical activity
1 month
Not reported
Huang, Hung,
Chang, &
Chang, 2009
N=130 Taiwanese
female freshman;
mean age not
reported; race not
reported
Randomized
controlled trial;
Groups: a) access
to stage-matched
website messages,
b) access to generic
website, c) in-class
lecture group
Physical activity
2 months
Transtheoretical
Model
Hurling et al.,
2007
N=77 healthy
adults; mean age
40.4 years; 66%
female; 99% White
Randomized
control trial;
Groups: a) access
to website, mobile
phone message,
and email, b) no
contact control
Physical activity
9 weeks
Not reported
Physical
Activity
Measure
Single item: “On
a typical day,
how many
minutes do you
spend in
physical
activity?”
Not reported
International
Physical
Activity
Questionnaire
(IPAQ);
Actiwaches
(accelerometerbased watches
developed for
the study)
Major Findings
Intervention
group
significantly
increased physical
activity
Both website
groups
significantly
increased physical
activity; no
improvement in
physical activity
for the in-class
group
Intervention
group
significantly
increased
moderate intensity
physical activity;
no change for
control group
27
Table 1.
Internet-based physical activity studies.
Study
Kelders, Van
Gemert-Pijnen,
Werkman,
Nijland, &
Seydel, 2011;
Kelders, van
Gemert-Pijnen,
Werkman, &
Seydel, 2010
Kim, Draska,
Hess, Wilson, &
Richardson,
2012
Kosma,
Cardinal, &
McCubbin,
2005
Lachausse,
2012
Sample
Characteristics
28
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Dutch Standard
for Healthy
Physical
Activity
Major Findings
N=269 Dutch
adults; mean age
41.5 years; 66%
female; race not
reported
Randomized
controlled trial;
Groups: a) access
to the Healthy
Weight Assist
website, b) wait-list
control
Physical activity;
dietary habits
12 weeks
Not reported
N=49 post-partum
females diagnosed
with gestational
diabetes; mean
age: control 35.5
years, intervention
35.9 years; 71%
White
N=75 individuals
with disabilities;
mean age not
reported; 72%
female; 89% White
Randomized
controlled trial;
Groups: a) access
to website, b) no
contact control
Physical activity
13 weeks
Not reported
Self-reported
pedometer steps
per day
No improvement
in physical
activity for either
group
Randomized
controlled trial of
(pilot); Groups: a)
website access, b)
control
Physical activity
1 month
Transtheoretical
Model
13 item Physical
Activity Scale
for Individuals
with disabilities
Treatment group
significantly
increased physical
activity
N=320 college
students; mean age
(by group): a) 26.7
years, b) 25.0
years, c) 22.8
Randomized
control trial;
Groups: a) access
to My Student Body
website, b) on-
Physical activity;
weight loss
12 weeks
Not reported
Youth Risk
Behavior Survey
No improvement
in physical
activity for any
group
No improvement
in physical
activity for either
group
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Major Findings
29
years; 73% female;
21% White (44%
Hispanic)
campus course, c)
no contact control
Leslie,
Marshall,
Owen, &
Bauman, 2005;
Marshall,
Leslie, Bauman,
Marcus, &
Owen, 2003
N=655 staff of an
Australian
University; mean
age 43 years; 51%
female; race not
reported
Randomized
controlled trial;
Groups: a) access
to Active Living
website and
biweekly
personalized stagematched emails, b)
Active living print
material
Physical activity
8 weeks
Transtheoretical
model
International
Physical
Activity
Questionnaire
(IPAQ) – short
form
No improvement
in physical
activity for either
group
Lieber et al.,
2012
N=892 Females;
mean age not
reported, range
from 18 to >75
years; 85% White
Physical activity
12 weeks
Transtheoretical
Model
N= 49 individuals
diagnosed with
type II diabetes;
mean age mean
age approximately
54 years; percent
female not
Physical activity
12 weeks
Social Cognitive
Theory
Adapted
questions from
Nurses’ Health
Study and
Women’s Health
Study
Godin LeisureTime Exercise
Questionnaire
Significant
increase in
physical activity.
Liebreich,
Plotnikoff,
Courneya, &
Boulé, 2009
One group pre-post
design assessing
the American Heart
Association’s
Choose to Move
Program
Randomized
control trial;
Groups: a) access
to the Diabetes
NetPLAY website
and weekly and
weekly email
Intervention
groups
significantly
increased physical
activity compared
to the control
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
reported; race not
reported
counseling, b)
control group
received links to
the Canadian
Diabetes
Association’s
Clinical Practice
Guidelines
Randomized
control trial;
Groups: a) access
to website, b)
contact control
Magoc,
Tomaka, &
Bridges-Arzaga,
2011
N=104 college
students; mean age
25.5 years; percent
female not
reported; 86%
Hispanic
Marcus et al.,
2007; Pekmezi
et al. 2010
N= 249 Health
sedentary adults;
mean age 44.5
years;
approximately
82% female; 81%
White
Randomized
controlled trial;
Groups: a) access
to tailored website,
b) tailored print, c)
standard internet
with links to 6
public physical
activity websites
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Physical activity
6 weeks
Social Cognitive
Theory
International
Physical
Activity
Questionnaire
(IPAQ)
Physical activity
12 months
Transtheoretical
Model; Social
Cognitive
Theory
Seven Day
Physical
Activity Recall
Major Findings
Intervention
group
significantly
increased physical
activity; no
change in physical
activity for
control group
All groups
increased physical
activity; no
difference in
physical activity
across groups
30
Table 1.
Internet-based physical activity studies.
Study
31
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Accelerometers
Major Findings
Mailey et al.,
2010
N=47 college
students with
mental disorders;
mean age 25 years;
68.1% females;
68% White
Randomized
controlled trial
(pilot); Groups: a)
access to website
plus two in-person
meetings with an
activity counselor,
b) no contact
control
Physical activity
10 weeks
Social Cognitive
Theory
McKay, King,
Eakin, Seeley,
& Glasgow,
2001
N=78 Type II
diabetes patients;
mean age 52.3
years; 53% female;
82% White
Physical activity
8 weeks
Social
Ecological
Model
BRFSS Survey
McConnon et
al., 2007
N=221 obese
adults; mean age
45.8 years; 77%
female; 95% White
Physical activity;
weight loss
12 months
Not reported
Baecke Physical
Activity
Questionnaire
No improvement
in physical
activity for either
group
Morgan,
Lubans, Collins,
Warren, &
Callister, 2009
N=65
overweight/obese
male staff and
students at an
Australian
Randomized
controlled trial;
Groups: a) online
personal tailored
coaching and study
website, b)
information only
control
Randomized
controlled trial;
Groups: a) access
to website, b) usual
weight loss care
materials
Randomized
control trial;
Groups: a) access
to Shed It website,
b) control
Physical activity;
weight loss
6 months
Social Cognitive
Theory
Pedometers
Both groups had a
significant time
increase of
physical activity;
no difference
Both groups
significantly
increased physical
activity at followup; however,
intervention group
had a greater
physical activity
increase than the
control group.
Significant
increase in
physical activity
for both groups
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
university; mean
age 35.9 years;
race not reported
Morgan et al.,
2011
Motl,
Dlugonski,
Wójcicki,
McAuley, &
Mohr, 2011
Napolitano et
al., 2003
N=110
overweight/obese
Australian male
factory employees;
mean age 44.4
years; race not
reported
N=54 adults
diagnosed with
multiple sclerosis;
mean age (by
group): control
45.6 years,
intervention 46.1
years; race not
reported
N=65 sedentary
hospital
employees; mean
age not reported;
84% female; 91%
White
Major Findings
between groups
Randomized
controlled trial;
Groups: a) access
to POWER website,
b) control
Physical activity;
weight loss
3 months
Social Cognitive
Theory
Godin LeisureTime Exercise
Questionnaire
Intervention
group had a
significant
increase in
physical activity
Randomized
controlled trial
(pilot); Groups: a)
access to study
website, b) control
Physical activity
12 weeks
Social Cognitive
Theory
Godin LeisureTime Exercise
Questionnaire
Intervention
group
significantly
increased physical
activity
Randomized
controlled trial;
Groups: a) website
access plus 12weekly email sheet,
b) control
Physical activity
3 months
Social Cognitive
Theory;
Transtheoretical
Model
BRFSS Physical
Activity Scale
Intervention
group increased
physical activity
at 1 month, but
not at 3 months.
32
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Oenema, Brug,
Dijkstra, de
Weerdt, & de
Vries, 2008
N=2159 Dutch
adults over the age
of 30; mean age
43.6 years; 54%
female; race not
reported
N=441
overweight/obese
male adults; mean
age 43.9 years;
71% White
Randomized
controlled trial;
Groups: a) access
to study website, b)
control
Physical activity;
fat intake;
smoking
cessation
1 month
Precaution
Adoption
Process Model
Randomized
controlled trial;
Groups: a) access
to study website,
weekly feedback on
progress, optional
email access
psychologist for
questions, b)
control
Randomized
controlled trial;
Groups: a) access
to website only, b)
access to website
plus in-person
meetings, c) inperson meetings
only
Physical activity;
weight loss
12 months
Social Cognitive
Theory
International
Physical
Activity
Questionnaire
(IPAQ)
Intervention
group
significantly
increased physical
activity
Physical activity;
weight loss
6 months
Not reported
Paffenbarger
Physical
Activity
Questionnaire
All groups
significantly
increased physical
activity; no
difference
between groups
Patrick et al.,
2011
Pellegrini et al.,
2012
N=51
overweight/obese
adults; mean age
44.2 years; 86.3%
female; 88% White
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
International
Physical
Activity
Questionnaire
(IPAQ)
Major Findings
No improvement
in physical
activity for either
group
33
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
34
Plotnikoff,
McCargar,
Wilson, &
Loucaides, 2005
N=2122 worksite
employees; mean
age 45.0 years;
73% female; race
not reported
Randomized
controlled trial:
Groups: a) weekly
physical activity
and nutrition
emails, b) control
Physical activity;
nutrition
12 weeks; 6
month followup
Social Cognitive
Theory;,
Transtheoretical
Model;
Protection
Motivation,
Theory of
Planned
Behavior
Pressler et al.,
2010
N=140 German
employees of an
automobile
company; mean
age 48 years; 11%
female; race not
reported
Randomized
controlled trial of
two website based
interventions;
Groups: a) website
access with
structured exercise
prescriptions, b)
website access
without structured
exercise
prescriptions
Physical activity;
cardiovascular
risk
12 week
Not reported
Physical
Activity
Measure
Godin Leisure
Time Exercise
Questionnaire
Pedometers
Major Findings
Intervention
group increased
physical activity,
control group
decreased
between baseline
and month 3;
from month 3 to
month 6 both
groups
significantly
increased physical
activity; at month
6, intervention
had significantly
higher activity
than the control.
No improvement
in physical
activity for either
group
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Richardson et
al., 2007
N=35 individuals
diagnosed with
type II diabetes;
age range 38-71
years; 57% female;
77% White
Physical activity
(step counts)
6 weeks
Health Belief
Model
Reid et al., 2011
N=223 adults with
heart disease;
mean age 56.4
years; 15.7%
female; race not
reported
Randomized
controlled trial of
two web-based
pedometer
interventions;
Groups: a) lifestyle
goals targeting step
counts, b)
structured goals
targeting 10 minute
step bouts
Randomized
controlled trial;
Groups: a) access
to CardioFit
website and email
contact with
counselors, b) usual
care (control)
Physical activity
6 months
Not reported
Physical
Activity
Measure
Pedometers
Pedometers;
Godin Leisure
Time Exercise
Questionnaire
Major Findings
Both groups
significantly
increased step
counts; no
difference
between groups
Intervention
group
significantly
increased
pedometer
assessed and selfreport physical
activity;
compared to
control; however,
the time effect
was not
significant for
either group
35
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Rovniak,
Hovell, Wojcik,
Winett, &
MartinezDonate, 2005
N=61 sedentary
adult women;
mean age 20.2
years; 82% White
Slootmaker,
Chinapaw,
Schuit, Seidell,
& Van
Mechelen, 2009
N=102 Dutch
office workers;
mean age 31.8
years; 60% female;
race not reported
Spittaels, De
Bourdeaudhuij,
& Vandelanotte,
2007
N=434 parents and
staff of secondary
schools in
Belgium; mean age
41.4 years; 61.6%
female; race not
reported
36
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Self-reported
walking;
physical activity
logs
Randomized
controlled trial;
Groups: a) high
fidelity Social
Cognitive Theory
emails, b) low
fidelity Social
Cognitive Theory
emails
Randomized
controlled trial;
Groups: a) access
to website with
tailored advice, b)
provided with a
single generic print
materials regarding
physical activity
(control)
Randomized
controlled trial;
Groups: a)
interactive
computer tailored
website, b)
interactive
computer tailored
website plus email
reminders, c) wait
Physical activity
(Walking)
12 weeks
Social Cognitive
Theory
Physical activity
3 months
Physical activity
6 months
Major Findings
No improvement
in physical
activity for either
group
Not reported
The Activity
Questionnaire
for Adolescents
and Adults
No improvement
in physical
activity for either
group
Theory of
Planned
Behavior;
Transtheoretical
Model
International
Physical
Activity
Questionnaire
(IPAQ)
Significant
increase in
physical activity
for both Internet
groups
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Major Findings
list control
R. Steele,
Mummery, &
Dwyer, 2007;
R. M. Steele,
Mummery, &
Dwyer, 2009
Sternfeld et al.,
2009
Tate, Wing, &
Winett, 2001
37
N=192 Australian
adults; mean age
38.7 years; 83%
female; race
reported as
predominately
White
N=787
administrative
office employees;
mean age 44.0
years; 79%
female; 38% White
(55% reported as
“other”)
Randomized
controlled trial;
Groups: a) face-toface delivery, b)
Internet and faceto-face, c) face-toface only
Randomized
controlled trial;
Groups: a) tailored
email messages, b)
no contact control
Physical activity
12 weeks
Social Cognitive
Theory
Active Australia
Questionnaire
All groups
significantly
increased physical
activity no
difference
between groups
Physical activity;
diet
16 weeks
Not reported
Adapted
questions from
the CrossCultural Active
Patterns
Questionnaire
N=91 overweight
healthy hospital
employees; mean
age 40.9 years;
89% female; 84%
White
Randomized
controlled trial
testing two webbased programs;
Groups: a) Internet
education, 2)
Internet behavioral
therapy
Weight loss;
physical activity
6 Months
Not reported, but
included the
construct of selfregulation
Paffenbarger
Physical
Activity
Questionnaire
Intervention
group
significantly
increased physical
activity at 16
weeks and at a 4
month postintervention
follow-up
Significant
increase in
physical activity
for both groups at
3 and 6 months;
no difference
between groups.
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Tate, Jackvony,
& Wing, 2003
N=92 overweight
adults at risk for
type II diabetes;
mean age 48.5
years; 90% female;
89% White
N=137 workplace
employees; mean
age 46.5 years;
6.6% female; 46%
White (54%
reported as “nonWhite”
Randomized
controlled trial;
Groups: a) website
access only, b)
website access plus
e- counseling.
Randomized
controlled trial
testing two
intervention
delivery methods
(delivered the exact
same content): a)
In-person, b)
Internet
Weight loss;
physical activity
12 months
Not reported
Physical activity;
weight loss;
cardiovascular
risk
12 weeks
Not reported
International
Physical
Activity
Questionnaire
(IPAQ)
N=91 college
females; mean age
not reported; race
not reported
Randomized
controlled trial;
Groups: a) access
to website, e-mail
counseling, and
email reminders to
access site, b)
control
Physical activity
6 weeks
Social Cognitive
Theory
International
Physical
Activity
Questionnaire
(IPAQ)
Touger-Decker,
Denmark,
Bruno,
O'SullivanMaillet, &
Lasser, 2010
Wadsworth &
Hallam, 2010
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Paffenbarger
Physical
Activity
Questionnaire
Major Findings
No change in
energy
expenditure
(physical activity)
Both groups
significantly
increased physical
activity at 12
weeks; at a 26
week follow-up,
participants
maintained
physical activity
levels than at
baseline
Significant
increase in
physical activity
at 6 weeks; at 6
month follow-up
the physical
activity increase
was not
maintained
38
Table 1.
Internet-based physical activity studies.
Study
39
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
Short 4 item
Questionnaire
(non-specified);
Accelerometers
Wanner,
Martin-Diener,
BraunFahrländer,
Bauer, &
Martin, 2009
N=1531 German
adults; mean age
43.7 years; 74.9%
female; race not
reported
Randomized
control trial;
Groups: a) access
to Active Online
website, b) nontailored website
Physical activity
13 months
Transtheoretical
Model
Watson,
Bickmore,
Cange,
Kulshreshtha, &
Kvedar, 2012
N=70 overweight
adults; mean age
42.0 years; 84%
female; 76% White
Physical activity
12 weeks
Not reported
Pedometers;
Seven Day
Physical
Activity Recall
Winett,
Anderson,
Wojcik, Winett,
& Bowden,
2007
N=1071 adults;
mean age 53.0
years; 67% female;
23% Black (other
races not reported)
Randomized
controlled trial;
Groups: a) access
to website, virtual
coach, upload
pedometer data, b)
access to
pedometer
company website to
upload pedometer
steps and view
graphs.
Randomized
controlled trail;
Groups: a) access
to Guide to Health
website only, b)
access to Guide to
Health website plus
church support, c)
Physical activity;
weight loss
7 months
Social Cognitive
Theory
Pedometers
Major Findings
Both groups
significantly
increased physical
activity at 6
weeks and 6
months; no
changes at 13
months from
baseline
No change in step
counts or physical
activity levels for
either group
The Guide to
Health plus
church support
group increased
steps compared to
the control group.
No differences
were found
Table 1.
Internet-based physical activity studies.
Study
Sample
Characteristics
Study Design
Behaviors
Targeted
Intervention
Duration
Theoretical
Framework
Physical
Activity
Measure
wait-list control
Major Findings
between the
control and Guide
to Health-only
group.
van Genugten et
al., 2012
N=539 overweight
Dutch adults; mean
age 47.8 years;
69% female; race
not reported
Randomized
controlled trial;
Groups: a) access
to an online
tailored
intervention, b)
access to generic
website
Physical activity,
dietary intake;
weight loss
6 months
Not reported
The Short
Questionnaire to
Access HealthEnhancing
Physical
Activity
van Wier et al.,
2009
N=1386 Dutch
worksite
employees; mean
age 43 years; 33%
female; race not
reported
Randomized
controlled trial;
Groups: a) access
to website plus
telephone
counseling, b)
access to website
plus email
counseling, c) usual
care/control
(lifestyle
brochures)
Physical activity;
weight loss
6 months
Not Reported
The Short
Questionnaire to
Access HealthEnhancing
Physical
Activity
Both groups
significantly
decreased
physical activity
over the duration
of the
intervention; no
difference
between groups
Internet and
phone
significantly
increase physical
activity; no other
differences were
found
40
Table 1.
Internet-based physical activity studies.
Study
Vandelanotte,
Duncan,
Plotnikoff, &
Mummery,
2012
Sample
Characteristics
Study Design
Behaviors
Targeted
N=863 Australian
adults; mean age
52.4 years; 60.7%
female; race not
reported
Randomized
controlled trial
assessing whether
individual delivery
preference
influence physical
activity outcomes
of a website
delivered
intervention;
Groups: a) text
based, b) video
based, c) text and
video based
Physical activity
Intervention
Duration
1 month
Theoretical
Framework
Theory of
Planned
Behavior;
Transtheoretical
Model
Physical
Activity
Measure
Active Australia
Survey
Major Findings
All groups
significantly
increased physical
activity; no
association among
individual
preferences for
delivery methods
and physical
activity outcomes
41
Behavioral Outcomes and Duration of Internet-based Physical Activity Interventions
Out the 60 unique interventions reviewed, 40 (66.6%) reported positive increases
in physical activity at the end of the intervention period. The duration of these
interventions varied from one month to 13 months. Half of the studies reviewed (n=30)
had an intervention period of three months or shorter, nine studies had an intervention
period between three and six months, and 21 studies had an intervention period of six
months or longer. An examination of intervention duration and physical activity
outcomes indicated that 63.3% (n=19/30) of the studies with intervention periods of three
months or shorter reported positive outcomes, 66.6% (n=6/9) of studies with intervention
periods between three and six months reported positive outcomes, and 71.5% (n=15/21)
of studies with intervention periods of six months or longer reported positive outcomes.
The lack of variability in study outcomes according to intervention duration possibly
suggest that that the physical activity outcomes of web-based studies may be not be
related to the duration of the intervention, but rather the intervention activities
themselves.
Theoretical Basis of Internet-based Physical Activity Interventions
The Social Cognitive Theory and Transtheoretical Model were the two most
commonly identified theoretical backgrounds used in the Internet-based physical activity
interventions. Twenty-one studies identified the Social Cognitive Theory as the
theoretical basis for intervention activities and 15 identified the Transtheoretical Model
(five studies reported using both the Social Cognitive Theory and Transtheoretical Model
as the theoretical background for the intervention). Other theoretical backgrounds
42
identified were the Theory of Planned Behavior, Health Belief Model, and the Social
Ecological Model. Approximately one-third (n=21) of the articles reviewed did not report
a theoretical background driving the intervention activities.
To evaluate whether interventions based in behavioral theory were more effective
than non-theoretical interventions, the physical activity outcomes of studies were
compared. Of the 21 studies identifying the Social Cognitive Theory as the primary or
one of the primary theoretical basis for intervention activities, 18 (86%) reported
significant increases in physical activity at the conclusion of the intervention period. In
reference to the 15 studies using the Transtheoretical Model (including those identifying
multiple theories), 11 (73%) demonstrated significant improvements in physical activity.
Among the 10 studies identifying multiple theoretical backgrounds, eight (80%) had
positive outcomes. In contrast, only nine (43%) of the 21 non-theoretical based
interventions reported improvements in physical activity. These findings suggest that
physical activity studies based in behavioral theory may be more effective than nontheory based interventions; which parallels the findings of several other systematic
reviews on interventions promoting physical activity (Pekmezi & Jennings, 2009;
Vandelanotte et al., 2007).
Behaviors Targeted by Internet-based Physical Activity Interventions
Many web-based studies have focused exclusively on physical activity as the only
outcome of the intervention while others have included physical activity as one of the
multiple health behaviors targeted by the intervention activities. Out of the 60 web-based
interventions reviewed for this paper, 30 targeted physical activity only, 16 focused on
43
both weight loss and physical activity, nine targeted physical activity and diet/nutritional
intake, and five focused on physical activity and reduction of various cardiovascular
risks. The physical activity outcomes of these studies showed that 21 of the 30 studies
(70%) exclusively targeting physical activity had significant increases in physical activity
at the end of the intervention period. Among the studies targeting multiple health
behaviors, 18 (60%) of the 30 studies had positive outcomes for physical activity. The
results of these studies suggest that the interventions focusing only on physical activity
and those targeting multiple health behavior are similarly efficacious in promoting
physical activity.
Internet-based Strategies Used to Promote Physical Activity
A variety of web-based strategies have been used to promote physical activity.
Some researchers have relied on email or electronic newsletters to deliver intervention
messages (Anderson, Winett, Wojcik, & Williams, 2010; Hageman, Walker, & Pullen,
2005; Plotnikoff, McCargar, Wilson, & Loucaides; Rovniak et al., 2005) while many
have utilized interactive websites (Huang et al., 2009; McKay, King, Eakin, Seeley, &
Glasgow, 2001; Napolitano et al., 2003; Tate, Wing, & Winett, 2001; Woolf et al., 2006).
A study conducted by Huang et al. (2009) provides a good example of how an interactive
website was used to promote physical activity. In this study, undergraduate females were
assigned to one of three treatment groups: a) a control group that received a one-time
lecture on methods to promote physical activity, b) an intervention group that received
staged-matched web-based physical activity promotion messages according to their
respective Transtheoretical Model stage of motivational readiness for change, or c) an
44
intervention group that received generic non-stage-matched web-based physical activity
promotion messages. Both intervention groups navigated an interactive website
simulating a virtual college dorm room that contained graphics, pictures, and games
illustrating methods to promote physical activity; the only difference between the two
intervention groups was the type of message (stage-matched or generic) participants
received. For example, each time participants logged onto the website, they entered a
virtual college dorm room and were greeted with a welcome message. Participants in the
stage-matched group received a personal, stage-matched message coinciding with their
respective stage of change for adopting regular physical activity. These messages were
designed to target specific processes of change (conscious raising, self re-evaluation,
helping relationships, etc) that would help participants advance through the stages of
change in order to adopt physical activity. Conversely, participants in the generic
Internet group received general information on the benefits of physical activity and
instructions on how to perform various forms of physical activity. Results of the study
indicated that both intervention groups significantly improved physical activity levels at
follow-up; the control group demonstrated no improvements in physical activity.
Other studies have combined multiple web-based strategies to promote physical
activity. Examples of these multiple strategies include individualized email support
(Liebreich et al., 2009; Napolitano et al., 2003; Tate, Jackvony, & Wing, 2003) or online
coaching (Glasgow et al., 2003; McKay et al., 2001) to complement study website usage.
For instance, Leibreich et al. (2009) incorporated email counseling with an interactive
website to promote physical activity among individuals diagnosed with type II diabetes.
Findings from this two-arm randomized controlled trial demonstrated that over the 12
45
week study participants in the intervention group significantly increased physical activity
levels by 35 minutes per week while the control group slightly decreased their physical
activity levels.
Other studies have incorporated intervention websites with varying other forms of
support to facilitate and promote physical activity. Examples include in-person support
from study staff (Glasgow et al., 2003; Glasgow et al., 2010; Mailey et al., 2010;
Pellegrini et al., 2012), telephone support from study staff (Carlson et al., 2012; van Wier
et al., 2009), and web-based or in-person peer-support from other study participants (Carr
et al., 2008; Glasgow et al., 2003; Winett et al., 2007).
Populations Targeted by Internet-based Physical Activity Interventions
Internet-based technology has been used across various adult populations to
promote physical activity. Such populations include sedentary or insufficiently active
individuals (Dunton & Robertson, 2008; Marcus, et al., 2007; Marshall et al., 2003;
Napolitano et al., 2003), overweight and obese adults (Harvey-Berino et al., 2010; Tate et
al., 2003; Tate et al., 2001), worksite populations (Napolitano et al., 2003; Plotnikoff et
al., 2005), university students (Grim et al., 2011; Huang et al., 2009; Magoc et al., 2011),
and individuals diagnosed with type II diabetes (Glasgow, Boles, McKay, Feil, &
Barrera, 2003; Liebreich et al., 2009; McKay et al., 2001; Tate et al., 2003). Findings
from these studies have shown mixed results in reference to improving physical activity
levels. Furthermore, the methodologies across these studies have varied, making it
difficult to conclude whether web-based physical activity interventions are more effective
in some populations than others.
46
Despite the diverse populations targeted by web-based physical activity
interventions, the African American community has been routinely understudied. The
literature review for the current study did not identify a single web-based intervention
promoting physical activity exclusively among African American women. However,
there was one study, conducted by Pekmezi et al. (2010), that examined feasibility of a
web-based approach to promoting physical activity among a subset of African American
women enrolled in a larger randomized controlled trial (Marcus et al., 2007). Findings
from this study showed that African American participants increased their physical
activity levels from 17 minutes per week at baseline to 139 minutes at six months
(p<.001), and to 104 minutes at the 12 month follow-up. Additionally, all participants
who completed the satisfaction survey (70% of the sample) reported reading most or all
of the physical activity information received by Internet or mail, and 80% of these
participants found the information to be helpful (Pekmezi et al., 2010).
Internet-based Physical Activity Interventions Among College Students
The current literature review identified eight Internet-based studies specifically
targeting college/university populations. Table 2 illustrates the characteristics of these
studies. Five of these studies reported positive physical activity outcomes at the end of
the intervention period (Gow et al., 2010; Grim et al., 2011; Huang et al., 2009; Magoc et
al., 2011; Mailey et al., 2010; Wadsworth & Hallam, 2010); however, the Mailey et al.
(2010) study also reported a significant time effect for the control group, so the results of
this study should be interpreted with caution.
47
The Social Cognitive Theory was the most commonly reported behavioral theory
used in these interventions (n=5); one study identified the Transtheoretical Model as the
theoretical basis, and two studies failed to identify a theoretical background. The duration
of these studies varied; three studies were six weeks in duration, three studies lasted
between six weeks and three months, one study lasted a college semester, and one study
was six months long. An attempt to determine if intervention duration was associated
with physical activity outcomes did not reveal any obvious patterns.
Two studies focused exclusively among female populations (Huang et al., 2009;
Wadsworth & Hallam, 2010) and both demonstrated positive physical activity outcomes.
However, none of these studies among college populations exclusively focused on the
African American community. Therefore, the current study is one of the first studies, if
not the first study, to evaluate a web-based approach to promoting physical activity
exclusively among African American college females.
48
Table 2.
Internet-based physical activity studies among college students.
Study
Franko et
al., 2008
Gow et al.,
2010
Sample
Characteristics
N=476 college
students from 6
universities; mean
age 20.1 years;
56.3% female; 58%
White
N=170 first year
college students;
mean age not
reported; 73.7%
female; 72% White
Study Design
Randomized
controlled trial;
Groups: a) My
Student Body webbased intervention
(2 website visits
required), b) My
Student Body webbased intervention
plus booster session
(3 website visits),
c) attention control
Randomized
controlled trial;
Groups: a) no
treatment, b) access
to website only, c)
6-week weight and
caloric feedback
emails, d)
combined access to
website and
feedback
Behaviors
Targeted
Physical activity;
nutrition
Intervention
Duration
6 months
Theoretical
Background
Not reported
Physical Activity
Measure
International
Physical Activity
Questionnaire
(IPAQ)
Physical activity;
prevention of
weight gain
6 weeks
Social Cognitive
Theory
International
Physical Activity
Questionnaire
(IPAQ) – Short
Form
Major Findings
No improvement in
physical activity for
any group
No improvement in
physical activity for
any group
49
Grim et al.,
2011
N=233 college
students; mean age
21.2 years; 82%
female; 82% White
Huang et
al., 2009
N=130 Taiwanese
female freshman;
mean age not
reported; race not
reported
Lachausse,
2012
N=320 college
students; mean age
(by group): a) 26.7,
b) 25.0, c) 22.8; 73%
female; 215 White
(44% Hispanic)
Magoc et
al., 2011
N=104 college
students; mean age
25.5 years; percent
female not reported;
86% Hispanic
Randomized
controlled trial;
Groups: a) access
to website, b) inperson class to
increase physical
activity, c) inperson health class
Randomized
controlled trial;
Groups: a) access
to stage-matched
website messages,
b) access to generic
website; c) in-class
lecture group
Randomized
control trial;
Groups: a) access
to My Student Body
website, b) oncampus course, c)
no contact control
Randomized
control trial;
Groups: a) access
to website, b)
contact control
Physical activity
Long college
semester
Social Cognitive
Theory
Seven Day
Physical Activity
Recall
Both web-based and
physical activity
class significantly
increased physical
activity; health class
did not
Physical activity
2 months
Transtheoretical
Model
Not reported
Both website groups
significantly
increased physical
activity; lecture
group did not
Physical activity;
weight loss
12 weeks
Not reported
Youth Risk
Behavior Survey
No improvement in
physical activity for
any group
Physical activity
6 weeks
Social Cognitive
Theory
International
Physical Activity
Questionnaire
(IPAQ)
Intervention group
significantly
increased physical
activity; no changes
in physical activity
for control group
50
Mailey et
al., 2010
N=47 college
students with
depression; mean
age 25 years; 68.1%
females; 68% White
Randomized
controlled trial
(pilot); Groups: a)
access to website
plus two in-person
meetings with an
activity counselor,
b) no contact
control
Physical activity
10 weeks
Social Cognitive
Theory
Accelerometers
Wadsworth
& Hallam,
2010
N=91 college
females; mean age
not reported; race
not reported
Randomized
controlled trial;
Groups: a) access
to website, e-mail
counseling, and
email reminders to
access site, b)
control
Physical activity
6 weeks
Social Cognitive
Theory
International
Physical Activity
Questionnaire
(IPAQ)
Both groups
significantly
increased physical
activity at followup; however,
intervention group
had a greater
physical activity
increase than the
control group.
Significant increase
in intervention
group physical
activity at 6 weeks;
at 6 month followup the physical
activity increase was
not maintained
51
Summary of Past Research on Internet-Based Physical Activity Interventions
There have been many Internet-based approaches used to promote physical
activity over the past decade. The methodologies and duration of these studies have
varied greatly; however, on the whole, the literature suggests that web-based technology
is advantageous in promoting physical activity. Nevertheless, there are several
weaknesses that exist in the present literature. First, the majority of studies have relied
on self-reported physical activity measures as the primary outcome measures of physical
activity. Incorporating objective measurements of physical activity (such as
accelerometers) to corroborate self-reported physical activity findings will help
strengthen the validity of reported physical activity outcomes (van den Berg et al., 2007;
Vandelanotte et al., 2007). A second weakness of the current web-based physical activity
literature is that various assessment measures have been used to assess physical activity.
Establishment of a common reporting method for physical activity (such as minute of
physical activity per week) is needed in order to compare physical activity effects across
studies; this standard would also allow for meta-analysis of web-based intervention
effects. A final weakness of the current literature is that many studies have not examined
the effects of longer-term (six months or greater) web-based physical activity
interventions (Belanger-Gravel et al., 2011; van den Berg et al., 2007; Vandelanotte et
al., 2007). Studies are needed to examine the longer-term intervention and longer-term
follow-up effects of web-based physical activity interventions.
52
Application of the Social Cognitive Theory to Physical Activity Promotion
In recent years, there has been an increased focus on utilizing behavioral theory to
guide physical activity promotion efforts (Marcus et al. 2006). Behavioral theory
provides a framework for researchers to design, implement, and evaluate the effects of
health promotion programs (National Cancer Institute [NCI], 2005). The Social Cognitive
Theory is the most commonly used behavioral theory in physical activity promotion
(Bélanger-Gravel et al., 2011; Marcus et al., 2006). The theory was developed by Albert
Bandura in the 1970s and expands on the tenets of his own Social Learning Theory
(Bandura, 1977; Bandura, 1986). Social Cognitive Theory explains behavior in a
reciprocal model where personal factors (beliefs, attitudes) and the environment (social
and physical) interact to influence behavior. This triadic relationship between the
individual, environment, and behavior is known as reciprocal determinism. The notion of
reciprocal determinism suggests that changes in any one of its factors (behavior,
individual, or environment) will influence and elicit a change in its other factors, which
ultimately influences behavior (Bandura, 1986).
The literature review for the current study revealed that the most commonly used
Social Cognitive Theory constructs in physical activity research include self-efficacy,
self-regulation (also known as self-control), social support, and outcome expectations. Of
these constructs, self-efficacy, defined as one’s perceived confidence in his/her ability to
perform a series of tasks in order to achieve a desired outcome (Bandura, 1997), is
perhaps most the frequently assessed and the construct that demonstrates the most robust
association with performance of physical activity (Anderson et al., 2010; Anderson-Bill,
53
Winett, & Wojcik, 2011; Anderson-Bill, Winett, Wojcik, & Winett, 2011; Magoc et al.,
2011; Rovniak, Anderson, Winett, & Stephens, 2002).
Self-regulation refers to an individual’s ability to manage his/her own social,
cognitive, and motivational processes in order to perform a behavior to achieve a desired
goal (Cameron & Leventhal, 2003). According to this construct, in order to effectively
change behavior, individuals must be able take control of their actions and self-monitor
their own behavior. Studies assessing self-regulation in relation to physical activity have
found overwhelmingly positive results for the construct being associated with
performance of physical activity (Anderson et al., 2010; Anderson, Wojcik, Winett, &
Williams, 2006; Anderson-Bill, Winett, & Wojcik, 2011; Anderson-Bill, Winett, Wojcik,
et al., 2011; Grim et al., 2011; Rovniak et al., 2002; Wadsworth & Hallam, 2010).
Furthermore, several studies have found this construct to be as strong of a predictor as
self-efficacy (Anderson-Bill, Winett, & Wojcik, 2011) or a better predictor than selfefficacy for influencing performance physical activity (Rovniak et al., 2002).
Studies examining the construct of social support for physical activity have
produced mixed findings regarding their associations with physical activity. Social
support refers to how an individual’s significant referents (family, friends, peers)
influence performance of a behavior (Uchino, 2004). Social support for physical activity
encompasses both tangible behaviors, such as providing rewards for performance of
physical activity, and intangible behaviors, such as encouragement (Beets, Cardinal, &
Alderman, 2010). Several studies have identified positive associations between social
support and physical activity (Anderson et al., 2010; Anderson et al., 2006; AndersonBill, Winett, & Wojcik, 2011; Anderson-Bill, Winett, Wojcik, et al., 2011; Martin,
54
McCaughtry, Flory, Murphy, & Wisdom, 2011; Rovniak et al., 2002) and other have not
(Grim et al., 2011; Magoc et al., 2011).
Outcome expectations are the anticipated outcomes one perceives to receive from
performing a specific behavior (Williams, Anderson, & Winett, 2005). This construct
appears to have a modest but meaningful association with performance of physical
activity. A previous review of this construct showed that outcome expectations for
physical activity have a small, albeit statistically significant relationship with
performance of physical activity among young adult populations (Williams et al., 2005).
Since this review, several studies have shown non-significant relationships between
outcome expectations and performance of physical activity (Anderson et al., 2010;
Anderson-Bill, Winett, & Wojcik, 2011; Anderson-Bill, Winett, Wojcik, et al., 2011;
Grim et al., 2011; Hallam & Petosa, 2004; Rovniak et al., 2002). However, none of these
studies were performed exclusively with an African-American population, warranting the
need for further research on this construct specifically among this community.
The Social Cognitive Theory has been used to promote physical activity across
various populations. Examples include college and university students (Anderson et al.,
2010; Grim et al., 2011; Magoc et al., 2011; Rovniak et al., 2002), overweight and/or
obese adults (Anderson et al., 2010; Carr et al., 2008; Decker, 2007; Tudor-Locke et al.,
2004; Turner-McGrievy et al., 2009; Winett et al., 2007), children and adolescents
(Araujo-Soares, McIntyre, MacLennan, & Sniehotta, 2009; Goran & Reynolds, 2005;
Martin et al., 2011), women who have recently given birth (< 12 months postpartum)
(Fjeldsoe, Miller, & Marshall, 2010), and individuals with type II diabetes (Liebreich et
al., 2009; Tudor-Locke et al., 2004). Among these studies, many have indicated positive
55
outcomes for promoting physical activity (Araujo-Soares et al., 2009; Fjeldsoe et al.,
2010; Grim et al., 2011; Liebreich et al., 2009; Magoc et al., 2011; Turner-McGrievy et
al., 2009; Winett et al., 2007), while only a few have not (Decker, 2007; Goran &
Reynolds, 2005).
Web-based physical activity promotion studies based on Social Cognitive Theory
have shown particular promise for promotion of physical activity. Results from many of
these studies have provided favorable outcomes for the Social Cognitive Theory
constructs predicting and explaining physical activity outcomes (Anderson et al., 2010;
Anderson-Bill, Winett, & Wojcik, 2011; Anderson-Bill, Winett, Wojcik, et al., 2011;
Grim et al., 2011; Liebreich et al., 2009; Magoc et al., 2011; Pekmezi, et al., 2010;
Winett et al., 2007). For example, findings from the Guide to Health Trial, a large webbased physical activity promotion study (n =661), showed that the Social Cognitive
Theory variables of self-efficacy, self-regulation and social support predicted physical
activity levels at baseline enrollment (Anderson-Bill, Winett, & Wojcik, 2011) and that
positive changes in these variables over the 12 month intervention were significantly
associated with improvements in physical activity at the end of the intervention
(Anderson et al. 2010) and at a 4 months post-intervention follow-up (Anderson et al.,
2010; Anderson-Bill, Winett, & Wojcik, 2011).
Finally, it is important to note that some researchers have identified the Social
Cognitive Theory as the theoretical basis for physical activity interventions but have
failed to assess any of the theoretical constructs in relation to physical activity outcomes
(Liebreich et al. 2009; Turner-McGrievy et al. 2009). Lack of testing such relationships
inhibits conclusions as to whether the observed changes in physical activity can be
56
attributed to the constructs of the Social Cognitive Theory. Future researchers designing
physical activity promotion interventions based on the Social Cognitive Theory should
test the mediation effects of the constructs in which intervention activities are based.
Performing such research is necessary to advance the field of physical activity promotion
(Marcus et al., 2006) and the applicability of the Social Cognitive Theory to physical
activity promotion efforts.
Cultural, Environment, and Individual Factors Associated with Physical Activity in
African American Women
African American women possess specific cultural, environment, and individual
perceptions related to physical activity that play an important role in determining whether
or not they perform physical activity (Bopp et al., 2007; Nies, Vollman, & Cook, 1999;
Siddiqi, Tiro, & Shuval, 2011; Young, He, Harris, & Mabry, 2002). Therefore,
researchers suggest that in order to effectively promote physical activity in this
population, intervention efforts should specifically target these perceptions (Nies,
Vollman, & Cook, 1999; Siddiqi, Tiro, & Shuval, 2011; Young, He, Harris, & Mabry,
2002). The following paragraphs describe various perceptions and barriers African
American women commonly associate with physical activity.
African American women are often considered the core of their family, and as a
result, have increased household and caretaker responsibilities that can limit the amount
of time available for them to perform physical activity (Bopp et al., 2007; Richter,
Wilcox, Greaney, Henderson, & Ainsworth, 2002; Sanderson, Littleton, & Pulley, 2002;
Siddiqi et al., 2011; Wilcox, Richter, Henderson, Greaney, & Ainsworth, 2002; Young et
57
al., 2002). Examples of these responsibilities include family meal preparation, providing
care for their own children as well as the children of extended family members, caring for
elderly relatives, and attending to various other chores that contribute to the balance and
maintenance of their household. Furthermore, African American women indicate that
even if there is time available for them to be physically activity, they often feel a sense of
selfishness or guilt for not spending that time with their family (Im et al., 2012;
Sanderson et al., 2002).
Lack of knowledge is also a commonly identified barrier to physical activity for
African American women. Many African American women report never learning about
the positive health benefits associated with physical activity (Nies et al., 1999; Siddiqi et
al., 2011; Wilcox et al., 2002). Further, studies have shown that even if African
American women are can identify the importance of physical activity, they do not know
what activities to perform in order to be physically active (Im et al., 2012; Wilcox et al.,
2002). Some researchers have linked this lack of physical activity knowledge to lack of
participation in physical activity during childhood due to cultural norms of African
American women not participating in physical activity (Young et al., 2002) and/or
negative childhood experiences associated physical activity due to teasing from peers,
which resulted in the withdrawal of participation in physical activity (Im et al., 2012;
Young et al., 2002).
African American women also cite issues associated with their physical
appearance as barriers to performing physical activity. For example, literature suggests
that African American women value a more curvaceous, full-figured body shape, and as
result, are hesitant to engage in physical activity for fear that they will reduce their weight
58
and ultimately alter the shape of their figure (Mabry et al., 2003; Sanderson et al., 2002;
Young et al., 2002). Qualitative studies have also identified hair concerns as a limiting
factor for participation of physical activity (Boyington et al., 2008; Im et al., 2012;
Siddiqi et al., 2011). African American women often spend significant time, effort, and
money to maintain their hairstyles. The process of sweating while performing physical
activity can have a negative effect their meticulous hairstyles, which is undesirable and a
major barrier to performing physical activity. A final physical appearance barrier to
physical activity for African American women is sun exposure. African American
women report unwanted sun exposure as a limiting factor for the performance of physical
activity (Richter et al., 2002; Sanderson et al., 2002; Wilcox et al., 2002).
The social and physical environments of African American women also play an
important role in the performance of physical activity. Lack of social support or someone
to be physically active with is a frequently identified barrier to physical activity for
African American women (Nies et al., 1999; Richter et al., 2002). African American
women report that if they had someone to exercise with or had encouragement from
family members to be physically active, they would be more inclined to engage in
physical activity (Nies et al., 1999). Furthermore, the lack of physically active African
American female role models has also been identified as an influential factor in the
physical activity levels of African American women (Sanderson et al., 2002; Young et
al., 2002).
A final barrier limiting physical activity among African American women is the
physical environment. Studies have shown that African American women residing in
historically African American neighborhoods often lack the facilities (sidewalks, parks,
59
recreation centers, etc) in which to perform physical activity (Bopp et al., 2007; Richter
et al., 2002; Siddiqi et al., 2011). Additionally, some African American women who
report having outdoor spaces or facilities in which to be physically active do not do so for
fear of safety, crime, and/or verbal harassment (Bopp et al., 2007; Boyington et al., 2008;
Richter et al., 2002; Siddiqi et al., 2011; Young et al., 2002).
It should be noted that the majority of the studies discussed in the previous
paragraphs have been conducted with samples of middle-aged African American women;
few studies have evaluated the factors that influence participation in younger adult
African American female populations. Therefore, the generalizability of some of these
findings to a population of younger African American females should be cautiously
interpreted. Additional research is needed to examine the factors that influence
participation in physical activity among younger adult African American women.
Selection of a Web-Based Approach for the Current Study
A web-based approach was selected for the current study for several reasons.
Most importantly, focus group research with the study’s target population, young African
American women aged 19-30 enrolled in college, identified the web-based approach as
the preferred method for delivery of a weight-loss intervention (N. Durant personal
communication, 2011). Additional rationale came from nationally representative data
indicating that adults between the ages of 19-34 spend over twice as much time on the
Internet when compared to any other age group (Kontos, Emmons, Puleo, & Viswanath,
2010) and evidence suggesting that young African Americans use web-based social
networking applications (which is a primary component in the current study website)
60
more frequently than their White or Hispanic counterparts (Smith, 2010). Therefore,
given these data supporting the use of a web-based approach and the previously
established effectiveness of web-based physical activity interventions, an Internet-based
approach emerged as an ideal method to deliver the current intervention.
The decision to develop a culturally-adapted website as opposed to using a
commercially available or previously developed website was selected due to research
suggesting that African American women have specific cultural and environmental
determinants that influence their participation in physical activity (Sanderson et al., 2002;
Wilcox et al., 2002; Young et al., 2002) and the favorable findings of other previously
implemented culturally-adapted (non-web-based) physical activity interventions
(Newton & Perri, 2004; Resnicow et al., 2005). Thus, it was hypothesized that utilizing a
culturally-adapted web-based approach would be advantageous in promoting physical
activity among college-aged African American women.
Addressing the Concept of Digital Inequality
A potential criticism for using a web-based approach to promote physical activity
among young African American women is based upon the notion of digital inequality.
Digital inequality, also referred to as the “digital divide”, refers to the differential access
of computer and/or informational communication technology among particular racial and
socioeconomic groups (Dimaggio, Hargittai, Celeste, & Shafer, 2004; Graham & Smith,
2001). These terms were developed in the in the mid-to-late 1990s when research showed
that racial minority and economically disadvantaged groups had less access to emerging
computer technology. Early research on the topic indicated that African Americans had
61
significantly lower access levels to computer technology in comparison to their nonAfrican American counterparts (Dimaggio et al., 2004). However, current research
suggests racial digital inequality is decreasing and may not be present in all population
segments, particularly the college-age population.
Population statistics show that internet access has increased among all Americans
since the late 1990s regardless of race or ethnicity. In 2001, nationally representative data
indicated 26% of African Americans and 46% of non-Hispanic Whites reported having
internet in their home (Dimaggio et al., 2004). Currently, these numbers have reached
58% and 74% in African American and White households respectively (U.S. Department
of Commerce, 2011). These statistics represent absolute percentages across all age, sex,
and socioeconomic groups, which do not appear to be representative of younger African
American women, the focus of the current study.
Studies focusing exclusively on college and young adult populations in the United
States provide limited evidence of a digital divide. A study on 873 college students
collected from nine colleges located throughout the United States found no significant
differences in Internet use between African American and White college students
(Koregan, Odell, & Schumacher, 2001). More recently, Chou et al. (2009) found that
younger age, specifically between the ages of 19 and 34, was the single best predictor of
increased internet utilization, not race. Studies have also shown that African Americans
demonstrate higher utilization of certain Internet applications in comparison to other
racial groups. For example, African Americans use social networking websites at a higher
prevalence than other racial groups (Chou, Hunt, Beckjord, Moser, & Hesse, 2009;
62
Graham & Smith, 2001) and are more likely to participate in internet blogging (Chou et
al., 2009).
In summary, the literature suggests that while digital inequality exists, it is
declining and may not have a salient presence among young adults. Thus, the utilization
of Internet-based interventions among African Americans, particularly young adult
African American women, may present a practical and innovative method to promote
physical activity.
Description of the Web-based Approach in the Current Study
The current study evaluated a theory driven, culturally-adapted technology-based
program designed to promote physical activity among young African American women.
Two main intervention components were utilized to promote physical activity. The first
component was the study website, entitled Commit2Fit, which served as the primary
physical activity promotion tool. The second study component consisted of structured
exercise sessions designed for participants to achieve a minimum of 150 minutes of
physical activity each week as recommended by the Centers for Disease Control and
Prevention (2008). Both of these study components were products of two previous
formative research phases that were conducted to inform development of the current
study.
In the first phase of the study’s development, focus groups and cognitive
interviews were conducted with the target population to identify web-based applications
young African American women desire in an Internet-based weight loss and physical
activity program. Data collected through these formative assessments were analyzed and
63
findings were used to inform the development of the Commit2Fit website. At the end of
this phase, a website prototype was developed to promote weight loss and physical
activity among young African American women.
In the second phase of study development, a six week feasibility assessment of the
Commit2Fit website (developed at the end of phase one) and intervention walking
protocol were performed. This phase required participants to utilize the study website as a
physical activity promotion tool for six weeks while concurrently attending alternating
weeks of structured walking and focus group sessions. Specifically, participants attended
walking sessions on weeks one, three, and five and focus groups on weeks two, four and
six. Data collected through the focus group sessions identified any technical problems
associated with the website and provided further feedback for website refinement. Data
collected from the walking sessions were used to evaluate the appropriateness of the
walking protocol and provide researchers the opportunity to test data analysis methods to
be used in the final intervention.
Thus, the two previously described formative research phases resulted in the final
version of the Commit2Fit website and structured exercise session protocol that were
implemented in the current study. The specific elements of both of these components are
described below.
Commit2Fit Website
Resources and applications available on the Commit2Fit website were grounded
in the constructs of the Social Cognitive Theory and provided culturally relevant
information to facilitate physical activity among young African American females. The
64
website platform allowed participants to access information regarding physical activity
while interacting with other study participants similar to many popular social network
sites (i.e. Facebook, MySpace, etc.). Participants had the ability create their own profile
page where they can share personal information, photos, and their physical activityrelated goals with other study participants. The purpose of these social networking
applications was to create a social support network that facilitates and encourages
performance of regular physical activity. Additionally, the Commit2Fit website was
accessible via cellular phone using the website’s iPhone application.
The primary applications available on the study website specifically related to
physical activity included social networking tools such as message boards and blogs,
physical activity self-monitoring/tracking tools, exercise videos, workout plans, and
educational blogs promoting methods to incorporate physical activity into daily activities.
Table 3 illustrates how each of these factors target specific constructs of the Social
Cognitive Theory and screenshots of these applications are available in Appendix C.
65
Table 3.
Overview of the Application of the Social Cognitive Theory to Intervention Components
Social Cognitive Theory
Component associated with Commit2Fit Program
Construct
Self-Efficacy
•
Social support from other Commit2Fit peers
participating in the program
•
Observing other study participants participate in
the program (via structure exercise sessions and
website profiles)
Observational Learning
•
Exercise Videos
Outcome Expectations
•
Blog posts of young African American women
providing personal testimonials on the benefits of
being physical activity
•
Videos of African American women performing
physical activity.
Self-Regulation
•
Exercise Tracker
Social Support
•
Message boards and blogs discussing and
encouraging performance of physical activity
•
Wall posts where participants post messages
promoting physical activity
Behavioral Capability
•
Exercise Plans
•
Exercise Videos
•
Blogs
66
Structured Exercise Sessions
Study protocol required that participants attend four structured exercise sessions
each week for the duration of the six month study. For three of these sessions,
participants walked three miles per session on the indoor track at the university recreation
center. On the fourth weekly exercise session, participants had the option to attend a
cardiovascular-based group exercise class such as indoor cycling, Zumba, or kickboxing.
Trained research assistants monitored these exercise sessions and physical activity data
were collected at each session by participants wearing heart rate monitors,
accelerometers, and pedometers. The heart rate monitors were individually programmed
to alert participants (by beeping) when their heart rate fell below a moderate-intensity
zone. Pedometers and accelerometers were used to collect information on the distance
and intensity of physical activity performed. Data collected by these activity monitoring
devices were recorded by study staff and provided to participants so they could upload
the data on the Commit2Fit website and self-monitor their progress throughout the study.
Summary
African American women report low levels of physical activity compared to other
race or sex groups; indicating the need for innovative physical activity promotion efforts
for this population. Internet-based physical activity interventions have shown great
promise for promoting physical activity; however, no previous Internet-based physical
activity interventions have been developed to specifically target the cultural and/or
personal needs of the African American female population. The purpose current study
was to evaluate a culturally-adapted web-based approach to promoting physical activity
67
among young African American college females. The findings of the current study add
to the limited research on web-based approaches to promoting physical activity among
this underserved population.
68
CHAPTER 3
METHODS
Introduction
The purpose of this study was to evaluate the physical activity and associated
Social Cognitive Theory outcomes of outcome expectations, enjoyment, self-regulation,
and social support following a six month web-based intervention promoting physical
activity among young African American women. Data collection and analysis methods
used in the current study are outlined in the subsequent sections in this chapter.
Study Design
The current study employed a secondary data analysis of a six month, single
group pre-post test design study assessing the acceptability and feasibility of a culturally
adapted web-based weight loss promotion tool. The current study focused exclusively on
the outcomes of physical activity and related psychosocial constructs of self-regulation,
outcome expectations, enjoyment, and social support. All study variables were assessed
at baseline, midpoint (three months), and one week following the conclusion of the six
month intervention. The primary outcome of physical activity was assessed by the Seven
Day Physical Activity Recall (Sallis, 1985). To corroborate findings from the Seven Day
Physical Activity Recall, participants were asked to wear an Actigraph accelerometer for
a seven day period that overlapped with the assessment days of the Seven Day Physical
Activity Recall. The secondary outcomes of outcome expectations, enjoyment of physical
69
activity, self-regulation, and social support were assessed by various other measures with
previously established reliability and validity
Participants
Participants of the study were young, overweight or obese African American
women (n = 34) enrolled at the University of Alabama at Birmingham. Inclusion criteria
for the study included: a) being between the ages of 19 and 30 years at time of study
enrollment, b) having a body mass index (BMI) greater than 25, c) self-identifying as
African American, d) being currently enrolled as a student at The University of Alabama
at Birmingham, and e) absence of any self-reported medical conditions that would inhibit
or limit performance of physical activity. Exclusion criteria for the study included a)
participation in another physical activity, nutrition, or weight loss program (commercial
or research), b) having uncontrolled high blood pressure (defined as greater than 140/90),
c) previously having a medical surgery to facilitate weight loss, d) currently on weight
loss medications, or e) losing more than 10 pounds in the previous three months before
enrollment.
It should be noted that since the primary focus of the parent study was weight
loss, participants were not required to be insufficiently active in order to enroll in the
study. Therefore, participants were enrolled in the study even if they were already
meeting the national physical activity recommendations. Additionally, the eligibility
screening process did not assess whether participants had computer or Internet access at
their place of residence. Since participants were required to be enrolled at the university,
70
it was assumed that they had regular access to these forms of technology as most
university coursework requires both computer and Internet access.
Procedure
Participants were recruited via convenience sampling methods at the University of
Alabama at Birmingham during the Spring 2011 semester. Sampling strategies included
face-to-face recruitment by study staff at the university’s student center, flyers placed
around campus, and word-of-mouth referrals from potential participants exposed to
recruitment efforts. Individuals expressing interest in study participation during
recruitment efforts were asked to provide their contact information to research staff.
Research staff then followed-up with potential participants via telephone and explained in
detail the purpose of the study and expectations for participation. Individuals interested in
participating in the study following the in-depth telephone description were immediately
screened for eligibility. At the conclusion of the eligibility screening, eligible participants
were scheduled for their baseline study assessment and emailed an informed consent
document to review prior to attending their initial baseline assessment.
The baseline study assessment was comprised of two clinic visits scheduled
exactly one week (seven days) apart. At the first visit, informed consent was obtained and
demographic, survey, and anthropometric data were collected. Participants were also
given an accelerometer at this initial visit and instructed to wear the device during all
waking hours for the next seven days. Detailed oral and written instructions on how to
wear the accelerometer were provided. At the second baseline clinic visit (seven days
71
after the initial baseline visit), participants returned their accelerometer and completed the
Seven Day Physical Activity Recall.
Midpoint data collection consisted of a single clinic visit during which
anthropometric, survey, and physical activity data were collected. Six month data
collection procedures were similar to baseline with participants attending two clinic visits
one week apart. At the first visit, psychosocial data were collected and an accelerometer
was distributed to participants for seven day wear. At the follow-up visit, participants
returned the accelerometer and completed the Seven Day Physical Activity Recall.
Participants were eligible to receive a total of $150 for study participation.
Compensation was distributed incrementally throughout the duration of the study.
Individuals not completing the study received prorated compensation based on their
participation. The physical activity and weight loss intervention from which data were
collected for the current study was funded by Robert Wood Johnson Foundation and all
study procedures were approved by the University of Alabama at Birmingham
Institutional Review Board.
Outcome Measures
Physical Activity
The Seven Day Physical Activity Recall (Sallis et al., 1985) was the primary
measure used to assess physical activity. The Seven Day Physical Activity Recall is an
interviewer-administered questionnaire that utilizes a standardized, semi-structured
technique to assess duration, intensity, and frequency of physical activity (Pereira et al.,
1997). This instrument assesses moderate-to-vigorous intensity physical activity
72
performed in bouts of ten minutes or greater over a previous seven day period. The
standardized interview process used by the Seven Day Physical Activity Recall has
demonstrated significant test-retest reliability estimates among adolescent (r = .81)
(Sallis, Buono, Roby, Micale, & Nelson, 1993) and young adult populations (r =.99)
(Gross, Sallis, Buono, Roby, & Nelson, 1990), and has strong inter-rater reliability
(r=.78) for assessments performed by multiple interviewers with the same subject (Sallis,
Patterson, Buono, & Nader, 1988). The recall instrument has been validated against more
objective measures of physical activity such as doubly labeled water (Washburn,
Jacobsen, Sonko, Hill, & Donnelly, 2003), physical activity logs (Dishman & Steinhardt,
1988), and accelerometers (Marcus et al., 2007).
To corroborate self-report physical activity findings, participants were asked to
wear an Actigraph activity monitor (GT3X and GT3X-plus models) during all waking
hours for a seven-day period at baseline and at the six-month follow-up. This seven-day
period precisely overlapped with the days assessed by the Seven Day Physical Activity
Recall. The Actigraph is a lightweight accelerometer worn around the waist that assesses
both movement and intensity of physical activity (Trost, McIver, & Pate, 2005). The
Actigraph has been validated to provide an accurate estimate of physical activity when
compared to doubly labeled water (Plasqui & Westerterp, 2007) and room calorimeter
(Rothney, Schaefer, Neumann, Choi, & Chen, 2008).
Physical activity data collected via accelerometry were prepared for analysis
according to the protocol described by Troiano et al. (2008). Accordingly, to be
considered as valid data for analyses, participants were required to wear the
accelerometer for a minimum of ten hours per day for at least four days during the seven
73
day assessment period. The minimal activity count threshold for moderate-to-vigorous
intensity physical activity was set at 2020 counts per minute. Time spent in at least
moderate-intensity physical activity is presented in two ways: 1) time spent in at least
moderate-intensity activity when performed in bouts of ten minutes or longer, and 2) total
minutes spent in at least moderate-intensity physical activity according to the 2020 count
per minute threshold. Ten-minute activity bouts were defined as achieving the
aforementioned cut-points for moderate-intensity activity for at least ten consecutive
minutes, with allowance of one to two minutes below these thresholds during the ten
minute period. Presenting and analyzing accelerometer data in bouts of ten minutes or
greater allows for direct comparison of the accelerometer data with data assessed by the
Seven Day Physical Activity Recall. The total time spent in moderate-intensity physical
activity (assessed by accelerometers) is presented for reference only, as this outcome
provides an elevated estimate of time spent in moderate-to-vigorous physical activity
because it includes all moderate-to-vigorous activity performed; not just physical activity
performed in bouts of ten minutes or greater.
The Seven Day Physical Activity Recall was selected as the primary physical
activity outcome measure (as opposed to data collected by the Actigraph) for several
reasons. First, the Seven Day Physical Activity recall has established reliability and
validity and employs a consistent assessment methodology across all studies that use the
measure. The standardized interview process employed by the Seven Day Physical
Activity Recall allows for comparison of observed physical activity levels in the current
study with the physical activity levels reported by other studies that have also used this
measure; which provides meaningful information on the effects of the current
74
intervention to other similarly designed studies. In comparison, physical activity assessed
by accelerometers lacks a standardized methodology used across studies (Troiano et al.,
2008; Trost et al., 2005). Therefore, researchers have not consistently used the same
validity criteria, activity cut-point thresholds, or epoch lengths to calculate time spent in
moderate-intensity physical activity; which can limit the comparison of physical activity
findings across studies.
Another reason why the Seven Day Physical Activity Recall was selected as the
primary outcome measure is because the Actigraph requires participants to wear the
device for a minimum of ten hours per day for at least four out of seven consecutive days
in order to collect a valid activity estimate. Therefore, if participants fail to wear the
Actigraph for ample time to achieve these validity cut points, an accurate measure of
activity level would not be available for analysis; thus, reducing the usable sample size
for analyses. This was of particular concern in the current study due to the already small
sample size. Furthermore, even if participants met the aforementioned validity cut points,
the possibility exists that they did not wear the accelerometer for all activity performed;
therefore, providing an underestimate of the participant’s true physical activity levels.
Outcome Expectations
The Social Cognitive Theory construct of outcome expectations was assessed by
The Outcome Expectation Scale for Exercise (Resnick, Zimmerman, Orwig, Furstenberg,
& Magaziner, 2000). The Outcome Expectation Scale for Exercise is a nine-item
questionnaire where respondents rate their level of agreement with statements regarding
exercise on a five-point Likert-like scale. This scale has been validated in African
75
American populations (Resnick, Luisi, Vogel, & Junaleepa, 2004) and demonstrates good
reliability estimates (Cronbach alphas ranging between .73-.89) (Resnick et al., 2004;
Resnick et al., 2000). Sample items from this scale include, “Exercise makes me feel
better” and “Exercise helps me feel less tired.”
Enjoyment
The Physical Activity Enjoyment Scale (Kendzierski & DeCarlo, 1991) was used
to assess the construct of enjoyment. While enjoyment is not an official construct
identified by the Social Cognitive Theory, many researchers conceptualize variable as
aligning with the construct of outcome expectations (Marcus & Forsyth, 2009). The
decision to warrant special attention to the concept of enjoyment of physical activity
came from the lack of research on this topic among African Americans and evidence
supporting that enjoyment can positively predict performance of physical activity
(Hagberg, Lindahl, Nyberg, & Hellénius, 2009; Schneider & Cooper, 2011). The
Physical Activity Enjoyment Scale is an 18-item Likert-like questionnaire where
respondents rate their agreement with contrasting statements on the enjoyment of
exercise. For example, one item on the scale reads “I enjoy it” (in regards to exercise) and
“I hate it.” Between these two contrasting statements, respondents rate agreement on a 1
to 7 scale (1 indicating “I enjoy exercise” and 7 indicating “I hate exercise”). This
questionnaire has previously established test-retest validity (Kendzierski & DeCarlo,
1991) and has demonstrated adequate internal consistency estimates ranging from .86 to
.89 when in when used exclusively with African American populations (Bopp et al.,
2006; Bopp et al., 2009).
76
Social Support
Social support for exercise was evaluated using the Social Support for Exercise
Survey (Sallis, Grossman, Pinski, Patterson, & Nader, 1987). This scale provides two
separate outcomes for social support, one in reference to family support (13-items) and
the other in reference to support from friends (10-items). The scale asks participants to
rate their level of agreement on a five-point Likert-like scale with statements regarding
how often their family and friends provide support for exercise (1= never, 5= very often).
A sample item from this scale includes, “During the past three months my family/friends
exercised with me.” The Social Support for Exercise Survey has demonstrated adequate
test-retest reliability (.79 and .77 for the family and friends scales respectively, p<.0001)
(Sallis et al., 1987) and has internal consistency (Cronbach alphas) estimates ranging
from .88 and .91 when used exclusively in African American populations (Bopp et al.,
2009).
Self-Regulation
Self-regulation for physical activity was assessed by the Self-Regulation Scale
from the Health Beliefs Survey (Anderson et al., 2010). This 10-item survey asks
respondents how often in the past month they have performed certain tasks or strategies
to increase physical activity levels. The Self-Regulation Scale has established reliability
estimates ranging from .83 to .91 (Cronbach’s alpha) and has been validated for use
across various adult populations (Anderson et al., 2010; Anderson et al., 2006). A sample
item from this questionnaire includes, “In the past month how often did you take the
stairs instead of the elevator?”
77
Covariate Measures
Body Mass Index (BMI). Body Mass Index was calculated by the formula weight
(kg)/ height (cm)2. Weight and height measures were collected by trained study staff in
order to calculate BMI. Weights were measured to the nearest tenth of a pound using a
Scaletronix (Wheaton, IL) digital scale. Height was measured to the nearest tenth of an
inch using the Digi-kit stadiometer by Measurement Concepts and Quick Medical (North
Bend, WA). To ensure consistency of height measurements, the same research staff
member assessed height for all participants.
Website Usage. Website utilization was assessed by an algorithm employed by
the study website. Each time a participant logged onto the website and engaged in using
an application, the algorithm calculated a number of points based on the amount of time
spent on the website and how many applications were used. Therefore, the more
frequently participants logged onto the website and used the available applications, the
higher the number of points they accumulated over the duration of the study.
Structured Exercise Session Attendance. All participants were expected to
participate in at least four exercise sessions per week for the duration of the study. The
majority of these sessions were held at the university’s student recreation center and were
monitored by study staff. In the event that school was not in session (i.e. Thanksgiving,
Christmas, and spring break holidays) or a special circumstance arose in which
participants were not able to attend a structured exercise session(s), they were given an
activity log to record their physical activity and encouraged to exercise on their own.
78
Over the course of the 26 week study, there were a total of 104 exercise sessions in which
participants were expected to participate. Attendance of these sessions was recorded and
a percentage of sessions attended was calculated for each participant.
Statistical Analyses
General linear models were used to examine the research questions. Physical
activity outcomes (assessed by the Seven Day Physical Activity Recall) were analyzed
and are presented in two ways. The first is a continuous response measure of minutes of
moderate-to-vigorous intensity physical activity performed per week; the second is a
binary or categorical response measure where participants are classified as either
achieving or not achieving the CDC’s recommendations of 150 minutes of moderateintensity physical activity per week. All analyses were performed using the Statistical
Package for Social Sciences (SPSS) version 20. Due to the exploratory nature of the
study statistical significance was set at p < .10. Specific analyses for each research
question are described below.
Specific Aim 1
Aim. Examine the relationship between Social Cognitive Theory constructs
related to physical activity (outcome expectations, self-regulation, social support, and
enjoyment) and physical activity levels at baseline.
Data Analyses. Bivariate linear regression analyses were used to examine the
relationship between Social Cognitive Theory variables and the continuous measure of
79
physical activity. Logistic regression analyses were used to examine the association
between the Social Cognitive Theory variables and the binary outcome of achieving at
least 150 minutes of moderate-intensity physical activity. Independent variables in the
analyses were the Social Cognitive Theory constructs of outcome expectations, selfregulation, social support, and enjoyment. The dependent variable was either the
continuous or binary measures of physical activity. The potential covariate of BMI was
also considered during analysis.
Specific Aim 2
Aim. Assess self-reported changes in physical activity levels, as measured by the
Seven Day Physical Activity Recall, from baseline to the six month follow-up.
Data Analyses. For the continuous outcome of physical activity expressed in
minutes per week, paired t-tests were used to determine if significant changes in physical
activity occurred over the duration of the study. For the binary outcome of achieving at
least 150 minutes per week of moderate-intensity physical activity, paired-proportion
analyses using McNemar’s test was used to determine if the number of participants
achieving the CDC’s recommendations of physical activity changed from baseline to the
end of the study. Paired t-tests and paired proportion analyses were selected because
these statistical techniques correct for multiple assessments on the same subject; thus
reducing the amount of variance in error term and ultimately increasing the power of the
statistical tests.
80
Specific Aim 3
Aim. Assess changes in Social Cognitive Theory variables from baseline to six
months and evaluate how these changes are associated with changes in physical activity
levels at six months.
Data Analyses. Changes in the Social Cognitive Theory variables were assessed
by a series of paired sample t-tests. Specifically, paired t-tests were used to assess
changes in the Social Cognitive Theory variables across the following time points:
baseline to midpoint, baseline to six months, and midpoint to six months. To evaluate if
changes in the Social Cognitive Theory variables were associated with changes in
physical activity, bivariate regression analyses were performed. In these bivariate models,
only the Social Cognitive Theory variables that demonstrated significant changes from
baseline to midpoint and from baseline to six months (determined by the aforementioned
paired sample t-tests) were tested. Bivariate regression analyses between web-site
utilization and walking session attendance with physical activity were also examined
during analyses.
Missing Data
Missing data in research questions two and three were accounted for by
performing two separate types of analyses. The first analysis technique used only data
from participants who provided complete physical activity (assessed by the Seven Day
Physical Activity Recall) and psychosocial data at all three assessment periods (baseline,
midpoint, and six months). This method was selected to gain feedback regarding the
81
effectiveness of the program among those who completed the study. The second type of
analysis utilized the intent to treat principle with participants’ previous assessment data
carried forward to account for missing data at follow-up assessments. For missing
physical activity data, baseline values were carried forward to the midpoint and six month
assessments. Midpoint physical activity data were not carried forward to the six month
assessment due to the fact that participants enrolled in the study were attending structured
exercise sessions as a part of the study protocol; therefore, carrying midpoint physical
activity values forward would have inflated physical activity scores at six months. In
reference to the Social Cognitive Theory variables, either baseline or midpoint values
were carried forward. Intent to treat analyses provide a more conservative estimate of the
intervention effects and is a suggested statistical approach in this research area
(Vandelanotte et al., 2007).
Power and Sample Size
The current study employed a relatively small sample (n = 34 at baseline)
comprised of participants enrolled in a weight loss and physical activity promotion study.
The parent study was originally designed and powered to detect a five kilogram change in
body weight over the course of the study; therefore, the possibility existed that some of
the research questions in the current study were inadequately powered to detect
significant changes, when, in fact, true differences may have actually existed (known as a
type II error) (Pallant, 2007). This concern was taken into account when selecting the
statistical tests for the current study and resulted in selection of analysis techniques that
82
are less sensitive to type II errors (bivariate regression models, sample t-tests, and paired
proportions).
Power assessments for changes in physical activity and the Social Cognitive
Theory variables were performed using post-hoc power analyses. As opposed to a priori
power calculations which use hypothesized effect sizes for power calculations, post-hoc
power tests compute statistical power using observed values collected in a study
(Onwuegbuzie & Leech, 2004). Post-hoc power analyses were particularly useful in the
current study because a secondary data analysis was performed. Therefore, the sample
size was predetermined and additional participants could not be enrolled if a priori tests
indicated that the study was underpowered to detect significant changes in physical
activity or the Social Cognitive Theory variables. Furthermore, even if a priori tests were
conducted, the selection hypothesized effect sizes and variance estimates would be
ambiguous due to the varied physical activity levels reported in the physical activity
literature. This issue was even further compounded due to the lack of studies promoting
physical activity among young African American women, the target population of the
proposed study.
However, despite the aforementioned power concerns, evaluation of the proposed
research questions was warranted even if only trends could be determined. Assessing the
intervention’s impact on physical activity and associated Social Cognitive Theory
variables in which the study was designed to target provided meaningful information on
the feasibility of a web-based approach to promote physical activity among young
African American women.
83
Summary
A secondary data analysis of six month, one group pre-post test design was
performed. The sample consisted of 34 overweight African American female college
students enrolled in a web-based weight loss and physical activity promotion
intervention. Paired t-tests, paired proportion analyses, and bivariate regression models
were used to evaluate the web-based intervention’s impact on self-reported physical
activity levels and associated Social Cognitive Theory constructs of outcome
expectations, enjoyment, self-regulation, and social support.
84
CHAPTER 4
RESULTS
Introduction
The purpose of this study was to evaluate the physical activity and associated
Social Cognitive Theory outcomes of outcome expectations, enjoyment, self-regulation,
and social support following a six month web-based intervention promoting physical
activity among young African American women. Results of the study are presented in the
following sections of this chapter.
Participants
Sample Characteristics
Seventy-two African American females were screened for study eligibility. Of
these, 54 were considered eligible and 38 provided informed consent and were enrolled in
the study. However, only 34 of the consented participants provided full baseline data, and
of these, 27 began the study, which was defined as accessing the study website at least
once and attending a minimal of one structured exercise session. At the conclusion of the
six month study, 17 of the 34 participants who provided full-baseline data provided six
month follow-up data, indicating a retention rate of 50%. Data analyses for the current
were limited to the 34 individuals who provided full baseline data. Figure 1 illustrates the
recruitment and retention flow diagram of participants.
85
At baseline, participants were mostly obese (mean BMI of 35.37) and the mean
age was 21.21 years. The majority of the sample indicated they were never married and
were seeking an undergraduate degree. Complete baseline demographic characteristics of
the study participants are shown in Table 4.
86
Screened
n = 72
Eligible
n = 54
Ineligible:
Age < 19 or > 30: n = 3
BMI < 25: n = 5
Lost 10 lbs in previous 3 months: n = 3
Diagnosis of Hypertension: n = 4
Non-student n = 1
Current participation in another
weight loss study n = 2
Refused n = 14
Loss to contact n = 2
Enrolled
n = 38
Lack of time n = 2
Loss of contact n = 2
Completed all Baseline
Assessments
n = 34
Lack of time n = 2
Loss of contact n = 3
Lack of enjoyment n = 1
Did not provide medical clearance
n=1
Started Study
n = 27
Lack of time n = 8
Loss of contact n = 1
Moved out of state n = 1
Provided Midpoint Data
n = 17
Lack of time n = 1
Loss of contact n = 1
Provided 6 Month Data
n = 17*
Figure 1. Participant recruitment and retention flow diagram.
Note: *Includes 2 participants who did not provide midpoint data.
87
Table 4.
Demographic characteristics of participants at baseline (N=34).
Variable
Mean
SD
Age (years)
21.21
2.30
BMI
35.37
6.82
N
Percent
Married
1
2.9
Divorced
1
2.9
Never Married
31
91.2
No Answer
1
2.9
Undergraduate
31
91.2
Masters
1
2.9
PhD
1
2.9
RN
1
2.9
First Year
2
5.9
Second Year
9
26.5
Third Year
8
23.5
Fourth Year
9
26.5
Fifth Year
6
17.6
Marital Status
Degree Currently Obtaining
Current Year in School
88
Table 1 continued.
N
Percent
High School or GED
17
50.0
Associates Degree
3
8.8
Bachelors Degree
4
11.8
Doctorate
1
2.9
Other
2
5.9
Less than High School
2
5.9
No Answer
5
14.1
High School or GED
15
44.1
Associates Degree
7
20.6
Bachelors Degree
5
14.7
Doctorate
4
11.8
Other
1
2.9
Less than High School
1
2.9
No Answer
1
2.9
Highest Degree Father Obtained
Highest Degree Mother Obtained
Completers versus Non-completers
Study completers were defined as providing data at all three assessment periods
(baseline, midpoint, six months), attending at least one structured exercise session, and
logging on to the study website a minimum of one time. Accordingly, 15 of the 34
89
participants whom provided baseline data were considered completers. Paired t-tests
comparing baseline characteristics of study completers versus non-completers indicated
that completers were older (p=.09), performed more physical activity at baseline (p=.05),
had higher outcome expectations for exercise (p=.07), and greater family social support
for exercise (p=.04). Table 5 illustrates baseline characteristics of study completers
versus non-completers.
90
Table 5.
Baseline comparison of study completers versus non-completers.
Variable
Completers
Non-Completers
Difference
t
p-value
N=15
N=19
BMI
35.60
35.20
.40
.17
.97
Age
21.87
20.53
1.34
1.74
.09
111.00
58.69
52.32
2.08
.05
Expectationsb
4.40
4.00
.40
1.91
.07
Enjoymentb
5.49
4.95
.54
1.65
.11
Self-Regulationb
2.31
2.15
.16
1.23
.22
2.74
2.74
.00
.01
.99
2.53
1.92
.61
2.17
.04
Self-Reported
Physical Activity
(minutes/week)a
Outcome
Social Support from
Friendsb
Social Support from
Familyb
Notes: aPhysical activity assessed by the Seven Day Physical Activity Recall; bMean
survey scores.
Website Usage and Exercise Session Attendance
Data collected via analytic tracking software indicated the study website was
accessed 1,570 times over the six-month study and that the average time spent on the
91
website per visit was five minutes and twenty-seven seconds. It should be noted that the
tracking software did not differentiate between participant and research staff website
visits (study staff regularly accessed the website); therefore, the actual number of
participant website visits and time spent on the website cannot be determined.
Out of the 27 participants who began the study, overall median exercise session
attendance over the six month study was 42.05% (range = 2.88% – 84.61%) and the
median number of website usage points was 56.00 (range = 0 - 1726). Completers (n=15)
attended a median of 54.92% of exercise sessions (range = 13.46% to 84.62%) and had a
median of 130.00 website points (range = 10.00 – 1726.00). Non-completers attended a
median of 9.61% (range = 2.88% - 59.62%) of exercise sessions and had a median of
11.00 (range = 0.00 to 1531) website points. Paired t-tests showed that study completers
attended significantly more exercise sessions than non-completers (t=5.06, p<.001);
however, there was not a significant difference for website utilization between completers
and non-completers (t=1.20, p=.21).
Corroboration of Seven Day Physical Activity Recall with Accelerometry
Mean physical activity values for both subjective and objective assessment
measures are presented in Table 6. Results of the correlation analyses between the Seven
Day Physical Activity Recall and physical activity assessed by accelerometers are
presented in Tables 7 and 8. At baseline, 27 participants provided valid accelerometer
data to corroborate with self-reported physical activity levels. Findings indicated nonsignificant correlations between the Seven Day Physical Activity Recall and both of the
physical activity outcomes assessed by accelerometers. A secondary analysis of study
92
completers at baseline showed a significant positive correlation between the Seven Day
Physical Activity Recall and accelerometer measured physical activity performed in ten
minute activity bouts (r=.53, p=.08); the correlation between the Seven Day Physical
Activity Recall and total accelerometer measured physical activity was not significant
(r=.44; p=1.29)
At the six month follow-up, eight participants provided valid accelerometer data
to corroborate with the Seven Day Physical Activity Recall. Correlation results indicated
a significant positive correlation between the Seven Day Physical Activity Recall and
accelerometer measured physical activity performed in ten minute activity bouts (r=.78,
p=.002). The correlation between the Seven Day Physical Activity Recall and total
accelerometer measured physical activity was not significant (r=.13, p=.76).
93
Table 6.
Mean minutes per week of moderate-intensity physical activity data collected by the
Seven Day Physical Activity Recall and accelerometers.
N
Seven Day
Accelerometer
Accelerometer
Recalla
10-minute Bouts
Total Activity
Mean (SD)
Mean (SD)
Mean (SD)
Baseline
All participants
27
82.59 (74.02)
49.52 (46.76)
176.33 (82.95)
Completers
12
109.19 (86.12)
28.92 (37.40)
127.00 (99.03)
8
169.00 (105.42)
56.50 (47.68)
161.75(92.78)
Six month
Completers
Note: Only participants with valid accelerometry data are presented. Mean values are
minutes of physical activity per week. aSeven Day Physical Activity Recall.
94
Table 7.
Correlations between self-reported and accelerometer measured physical activity at
baseline.
All participants (N=27)
1
2
3
1.00
.00
-.17
2. Accelerometer 10 minute Activity Bouts
-
1.00
.63**
3. Accelerometer Total Minutes of Activity
-
-
1.00
1
2
3
1.00
.58*
.36
2. Accelerometer 10 minute Activity Bouts
-
1.00
.55
3. Accelerometer Total Minutes of Activity
-
-
1.00
1. Seven Physical Activity Recall
Study Completers (N=12)
1. Seven Physical Activity Recall
Notes: Correlations Spearman’s rho. *Correlation significant at the .05 level (two-tailed);
**Correlation significant at the .01 level (two-tailed).
95
Table 8.
Correlations between self-reported and accelerometer measured physical activity at six
month follow-up.
Completers (N=8)
1
2
3
1.00
.78**
.13
2. Accelerometer 10 minute Activity Bouts
-
1.00
.08
3. Accelerometer Total Minutes of Activity
-
-
1.00
1. Seven Physical Activity Recall
Note: Correlations are Spearman’s rho. **Correlation significant at the .01 level (twotailed).
Reliability Estimates of the Social Cognitive Theory Variables
Reliability estimates for the Social Cognitive Theory variables across all
assessment periods are presented in Table 9. All variables demonstrated adequate
reliability estimates with the exception of the Self-Regulation Scale for Exercise at
baseline (α=.56).
96
Table 9.
Inter-item reliability estimates for Social Cognitive Theory variables for all assessment
periods.
Number
Baseline
Midpoint
Six Month
of
α
α
α
Range
Items
N=34
N=17
N=17
Outcome Expectations
1-5
9
.88
.85
.85
Self-Regulation
1-5
10
.56
.77
.77
Enjoyment
1-7
18
.91
.94
.87
1-5
13
.91
.89
.82
1-5
10
.91
.95
.91
Variable
Social Support from
Family
Social Support from
Friends
Note: Alphas presented are Cronbach’s alpha coefficients.
Specific Aim 1
Aim 1. Examine the relationship between Social Cognitive Theory constructs
related to physical activity (outcome expectations, self-regulation, social support, and
enjoyment) and physical activity levels at baseline.
Preliminary Data Analyses
Preliminary analyses indicated that two participants reported outlying physical
activity values at baseline. An examination of the self-reported physical activity data
97
provided by these participants revealed that they reported unusually high levels of
vigorous intensity physical activity (115 minutes/week and 210 minutes/week of vigorous
intensity physical activity respectively), which inflated their total self-reported physical
activity levels. Accordingly, they were dropped from analyses due to the study’s small
sample size and their influence on skewing the mean and standard deviation of total
physical activity levels. Tests for normality showed that the baseline physical activity
levels were mildly positively skewed even after eliminating these two outliers for
physical activity (Kolmogorov-Smirnov test statistic =.163, p=.03). Both log and square
root transformations were explored to correct for the non-normality of the data; however,
neither of these transformations improved normality (log transformation KolmogorovSmirnov test statistic =.22, p<.001; square root transformation Kolmogorov-Smirnov test
statistic = .21, p=.001). Therefore, regression analyses were performed without any
correction for skewedness of the physical activity data.
Bivariate correlation analyses among the Social Cognitive Theory variables
indicated a significant positive correlation between Outcome Expectations for Exercise
Scale and the Physical Activity Enjoyment Scale (p=.002); thus, providing support for
the hypothesis that enjoyment of physical activity is closely aligned with outcome
expectations. Positive correlations were found between self-regulation and social support
from friends (p=.07), and between social support from family and social support from
friends (p=.05). Table 10 illustrates the bivariate correlations among all the Social
Cognitive Theory study variables.
98
Table 10.
Pearson bivariate correlations among Social Cognitive Theory variables at baseline
(N=34).
Variable
1
2
3
4
5
6
1.00
.52***
.18
.04
-.10
-.06
2. Enjoyment
-
1.00
.24
.11
-.09
.11
3. Self-Regulation
-
-
1.00
-.03
.32*
.11
4. Social Support from
-
-
-
1.00
.34**
-.17
-
-
-
-
1.00
-.17
-
-
-
-
-
1.00
1. Outcome Expectations
Family
5. Social Support from
Friends
6. Physical Activity
Notes: *p<.10, **p<.05, ***p<.01
Regression Analyses
Participants reported performing a mean of 75.78 (SD = 72.89) minutes per week
of moderate-intensity physical activity (median = 70.00, range = 0 – 235). Bivariate
linear regression analyses using the continuous outcome of minutes of physical activity
per week indicated that none of the Social Cognitive Theory variables were associated
with physical activity levels at baseline. However, the covariate of BMI demonstrated a
significant inverse relationship with physical activity (ß=.32, p=.08). Table 11 shows the
results from all bivariate regression analyses at baseline.
99
Table 11.
Bivariate regression outcomes between the Social Cognitive Theory variables and
physical activity at baseline (N=32).
Effect Size (f2) p-value
Variable
df
F
Beta
Body Mass Index
1
3.30
-.31
.02
.08
1
.01
.02
.00
.91
1
.08
-.05
.00
.78
Friends
1
1.96
-.25
.06
.51
Family
1
.61
-.14
.02
.17
1
.64
.14
.02
.43
Self-Regulation Scale for
Physical Activity
Outcome Expectation Scale
for Exercise
Social Support for Exercise
Survey
Physical Activity
Enjoyment Scale
Note: Effect size estimated calculated by Cohen’s fI2statistic
Bivariate logistic regression analyses using the binary outcome of achieving the
CDC’s recommendations of achieving at least 150 minutes of moderate-intensity physical
activity are presented in Table 12. Seven participants reported achieving at least 150
minutes of physical activity per week at baseline. Enjoyment for physical activity
demonstrated a marginally significant relationship with achieving 150 minutes per week
100
of physical activity (OR= .43, CI=.17– .99). None of the other Social Cognitive Theory
variables or covariates were associated with this physical activity outcome.
Table 12.
Odds ratios for the Social Cognitive Theory constructs predicting achievement of 150
minutes of physical activity per week (N=32).
Variable
Odds Ratio
90% CI
p-value
.92
.82 – 1.04
.38
1.55
.25 – 9.58
.98
.91
.37 – 2.69
.91
Friends
.59
.25 – 1.39
.13
Family
.48
.21 – 1.07
.50
1.96
.92 – 4.09
.10
Body Mass Index
Self-Regulation Scale for
Physical Activity
Outcome Expectation Scale for
Exercise
Social Support for Exercise
Survey
Physical Activity Enjoyment
Scale
101
Specific Aim 2
Aim 2. Assess changes in self-reported physical activity levels, as measured by
the Seven Day Physical Activity Recall, from baseline to the six month follow-up.
Preliminary Analyses
In alignment with Aim 1, the two participants with outlying physical activity
values at baseline were excluded from analyses assessing changes in physical activity;
thus, reducing the sample size of study completers to 13 participants. Sensitivity analyses
including these participants were conducted for both completer and baseline observation
carried forward analyses. Results from the sensitivity analyses did not significantly differ
from what are reported with the outliers excluded.
Paired T-test Analyses
Study completers (n=13) reported performing a mean of 100.77 minutes per week
of physical activity at baseline and a mean of 134.00 minutes per week at six months. Ttest analysis showed an increase of 33.23 minutes per week activity of physical activity
from baseline to six months (t=1.23, p=.24). Additionally, although not a formal an aim,
physical activity levels at midpoint were assessed. Results indicated a significant increase
of 76.69 minutes per week of physical activity from baseline to three months (t=2.09,
p=.06).
Analysis using baseline observations carried forward for participants with missing
values at the six months showed a mean increase of 16.63 minutes per week of physical
activity from baseline to six months (t=1.47, p=.15). Table 13 shows mean physical
102
activity levels across all assessment periods, Table 14 shows findings from paired t-test
analyses, and Figure 2 illustrates changes in physical activity.
Table 13.
Mean self-reported physical activity levels at baseline, midpoint, and six months.
Baseline
Midpoint
Six Month
Mean (SD)
(Mean SD)
Mean (SD)
Completers (N=13)
100.77 (87.84)
177.46 (102.09)
134.00 (104.50)
Intent-to-treat (N=32)
75.78 (72.89)
115.50 (96.76)
92.41 (85.33)
Note: Mean and standard deviations are presented in minutes/week. Intent-to-treat
analysis utilized baseline physical activity values carried forward for participants with
missing data at midpoint or six months.
103
Minutes/Week of Physical Activity
190
†
170
150
Completers
130
BOCF
Completer Trend*
110
90
70
Baseline
Midpoint
Six Month
Assessment Period
Figure 2. Mean physical activity scores across assessment periods.
Note: †Indicates a significant increase in physical activity. *Completer trend refers to the
linear trend observed by study completers.
104
Table 14.
T-tests for changes in self-reported physical activity.
Assessment Period
Mean Difference (SD)
t
p-value
Completers
33.23 (97.17)
1.23
.24
Intent-to-treat
16.63 (63.82)
1.47
.15
Completers
76.69 (132.28)
2.09
.06
Intent-to-treat
39.72 (95.33)
2.36
.03
Completers
-43.46 (110.57)
-1.42
.18
Intent-to-treat
-23.10 (81.74)
-1.60
.12
Baseline - Six Months
Baseline - Midpoint
Midpoint – Six Months
Note: Positive values denote an increase in physical activity. Mean and standard
deviations are presented in minutes/week. Intent-to-treat analyses utilized baseline
physical activity values carried forward for participants with missing data at midpoint or
six months. Completers N=13; Intent-to-treat N=32.
Paired Proportion Analyses
At baseline, seven of the 32 participants included in analyses reported performing
at least 150 minutes per of moderate intensity physical activity. Among study completers,
five of the 13 participants achieved a minimum of 150 minutes per week of physical
activity at baseline. Results from the six month follow-up showed that the number of
participants achieving 150 minutes of physical activity per week did not differ from
105
baseline for either study completers or analysis with baseline observations carried
forward for missing data. Table 15 shows findings from the paired proportion analyses.
Table 15.
Paired proportion analyses of participants who achieved 150 minutes of moderateintensity physical activity at baseline and at six months.
Achieved 150
Change in Proportion of Participants
min./week
Achieving 150 min./week from
p-valuea
Baseline to Six Months
Yes
No
Completers
Baseline
5
8
Six Months
5
8
Intent-to-treat
Baseline
7
25
Six Months
7
25
0
1.00
0
1.00
Notes: a Exact significance using McNemar’s test to assess change in achieving 150
minutes of moderate-intensity physical activity from baseline to six months. Intent-totreat analyses utilized baseline physical activity values carried forward for participants
with missing data at six months. Completers N=13; BOCF=32.
106
Post-hoc Power Analyses
The post-hoc power analysis using data from study completers (n=13, difference
in physical activity=33.23 minutes per week, SD=97.17) indicated the paired t-test that
assessed changes in physical activity had an observed power of 30.2% (α=.10). The posthoc power analysis for baseline observation carried forward analysis demonstrated an
observed power of 41.4% (α=.10).
Specific Aim 3
Aim 3. Assess changes in Social Cognitive Theory variables from baseline to six
months and evaluate how these changes are associated with changes in physical activity
levels at six months.
Preliminary Analyses
In accordance with analyses performed in previous aims, the same two
participants with outlying values for physical activity at baseline were excluded from
analyses in this aim. Sensitivity analyses including these participants were conducted for
all the Social Cognitive Theory variables and exclusion of these participants did not alter
the findings presented below.
Paired T-test Analyses
Paired t-test analyses were used to assess changes in the Social Cognitive Theory
variables. Mean values for the Social Cognitive Theory variables at all assessment
107
periods are shown in Table 16. Results from the paired t-tests are presented in Tables 17,
18 and 19.
108
Table 16.
Mean values for Social Cognitive Theory variables for all assessment periods.
Variable
Baseline
Midpoint
Six Months
Mean (SD)
Mean (SD)
Mean (SD)
Completers
4.40 (.54)
4.30 (.55)
4.32 (.56)
Intent-to-treat
4.17 (.63)
4.15 (.59)
4.13 (.59)
Completers
2.27 (.35)
2.88 (.67)
2.52 (.60)
Intent-to-treat
2.20 (.61)
2.55 (.68)
2.42 (.61)
Completers
5.55 (.97)
5.30 (.97)
5.43 (.72)
Intent-to-treat
5.19 (1.00)
5.13 (1.00)
5.24 (.88)
Completers
2.58 (.83)
2.34 (.88)
2.45 (.55)
Intent-to-treat
2.19 (.86)
2.14 (.88)
2.19 (.76)
Completers
2.57 (.78)
2.99 (1.04)
2.95 (.88)
Intent-to-treat
2.67 (.95)
2.88 (1.07)
2.88 (1.03)
Outcome Expectations
Self-Regulation
Enjoyment
Social Support Family
Social Support Friends
Note: Intent-to-treat analyses utilized the last observation provided by each participant
carried forward for missing data at midpoint of six months. Intent-to-treat N=32;
Completers N=13.
109
Outcome Expectations. Analyses for both study completers and most recent
observation carried forward did not reveal any significant changes in outcome
expectations over the course of the study. Figure 3 illustrates the outcome expectation
values for all assessment periods.
5
4.5
Mean Score
4
3.5
Completers
3
LOCF
Completer Trend*
2.5
2
1.5
1
Baseline
Midpoint
Six Months
Figure 3. Mean outcome expectations for physical activity scores across assessment
periods. Note: *Completer trend refers to the linear trend observed by study completers.
Enjoyment for Physical Activity. No significant changes were observed for
enjoyment of physical activity over the duration of the study. Results did not differ
among completer or most recent observation carried forward analyses. A graphic
depiction of the mean values of enjoyment for physical activity is presented in figure 4.
110
7
6.5
6
5.5
Mean Score
5
4.5
Completers
4
LOCF
3.5
Completer Trend*
3
2.5
2
1.5
1
Baseline
Midpoint
Six Month
Figure 4. Mean enjoyment for physical activity scores across all assessment periods.
Note: *Completer trend refers to the linear trend observed by study completers.
Self-Regulation. Self-regulation for physical activity among study completers
significantly increased from baseline to midpoint (p=.005). However, subsequently
decreased from midpoint to six months (p= .06), resulting in a non-significant change in
self-regulation from baseline to six months (p=.16). Despite this non-significant finding
from baseline to six months, a positive trend for self-regulation among study completers
was observed.
Results from the analysis using the most recent observation carried forward for
missing data indicated a significant increase in self-regulation from baseline to midpoint
(p=.002) and from baseline to six months (p=.03). Figure 5 illustrates changes in selfregulation scores across assessment periods.
111
5
4.5
Mean Score
4
3.5
Completers
3
LOCF
Completer Trend*
2.5
2
1.5
1
Baseline
Midpoint
Six Months
Figure 5. Mean self-regulation for physical activity scores across assessment periods.
Note: *Completer trend refers to the linear trend observed by study completers.
Social Support from Family. No significant changes were observed for social
support for exercise from family. Mean values for social support from family are
illustrated in Figure 6.
112
5
4.5
Mean Score
4
3.5
Completers
3
LOCF
Completer Trend*
2.5
2
1.5
1
Baseline
Midpoint
Six Months
Figure 6. Mean social support from family scores across assessment periods.
Note: *Completer trend refers to the linear trend observed by study completers.
Social Support from Friends. A significant increase in social support for exercise
from friends was observed from baseline to six months among study completers (p=.05).
Analysis performed with most recent observations carried forward for missing values
also demonstrated a significant increase from baseline to six months (p=.03). Changes in
social support for exercise from friends are graphically presented in Figure 7.
113
5
4.5
Mean Score
4
3.5
Completers
3
LOCF
Completer Trend*
2.5
2
1.5
1
Baseline
Midpoint
Six Months
Figure 7. Mean social support from friends scores across assessment periods.
Note: *Completer trend refers to the linear trend observed by study completers.
114
Table 17.
Changes in Social Cognitive Theory variables from baseline to six months.
Variable
Mean Difference (SD)
Effect Size (d)
t
p-value
Completers
-.85 (.28)
.28
-1.10
.29
Intent-to-treat
-.04 (.24)
.17
-.89
.38
Completers
.25 (.61)
.41
1.48
.164
Intent-to-treat
.22 (.56)
.39
2.27
.03
Completers
-.12 (.56)
.21
-.78
.45
Intent-to-treat
.05 (.55)
.09
.53
.60
Completers
-.13 (.58)
.22
-.81
.43
Intent-to-treat
.00 (.44)
.09
.00
1.00
Completers
.38 (.62)
.61
2.16
.05
Intent-to-treat
.20 (.51)
.39
2.24
.03
Outcome Expectations
Self-Regulation
Enjoyment
Social Support Family
Social Support Friends
Note: Positive values denote increases in Social Cognitive Theory variables. Intent-totreat analyses utilized the last observation provided by each participant carried forward
for missing values at midpoint or six months. Intent-to-treat N=32; Completers N=13.
115
Table 18.
Changes in Social Cognitive Theory variables from baseline to midpoint.
Variable
Mean Difference (SD)
Effect Size (d)
t
p-value
Completers
-.10 (.26)
.38
-1.10
.23
Intent-to-treat
-.02 (.23)
.08
-.42
.68
Completers
.61 (.64)
.95
3.46
.005
Intent-to-treat
.35 (.60)
.58
2.30
.002
Completers
-.24 (.75)
.32
-1.18
.26
Intent-to-treat
-.06 (.51)
.12
-.69
.50
Completers
-.24 (.41)
.59
-2.10
.057
Intent-to-treat
-.05 (.44)
.11
-.86
.40
Completers
.42 (.95)
.44
.99
.137
Intent-to-treat
.21 (.68)
.31
1.79
.083
Outcome Expectations
Self-Regulation
Enjoyment
Social Support Family
Social Support Friends
Note: Positive values denote increases in Social Cognitive Theory variables. Intent-totreat analyses utilized the last observation provided by each participant carried forward
for missing values at midpoint or six months. Intent-to-treat N=32; Completers N=13.
116
Table 19.
Changes in Social Cognitive Theory variables from midpoint to six months.
Variable
Mean Difference (SD)
Effect Size (d)
t
p-value
Completers
.02 (.28)
.07
.21
.84
Intent-to-treat
-.02 (.20)
.10
-.58
.57
Completers
-.36 (.62)
.58
-.210
.06
Intent-to-treat
-.13 (.45)
.29
-1.61
.12
Completers
.12 (.82)
.14
.55
.59
Intent-to-treat
.11 (.64)
.17
1.01
.32
Completers
.11 (.65)
.17
.62
.55
Intent-to-treat
.05 (.44)
.11
.69
.50
Completers
-.05 (.68)
.07
-.23
.82
Intent-to-treat
-.01 (.44)
.02
-.15
.88
Outcome Expectations
Self-Regulation
Enjoyment
Social Support Family
Social Support Friends
Note: Positive values denote increases in Social Cognitive Theory variables. Intent-totreat analyses utilized the last observation provided by each participant carried forward
for missing values at midpoint or six months. Intent-to-treat N=32; Completers N=13.
117
Post Hoc Power Analyses
Results from the post hoc power analyses for the Social Cognitive Theory
variables are presented in Table 20. Findings from these analyses indicated that the
majority of the t-tests were powered at less than 80 percent.
118
Table 20.
Post-hoc power analyses for changes in the Social Cognitive Theory variables from
baseline to six months.
Variable
Mean Difference (SD)
Observed Power
Completers
-.85 (.28)
1.00
Intent-to-treat
-.04 (.24)
.24
Completers
.25 (.61)
.39
Intent-to-treat
.22 (.56)
.70
Completers
-.12 (.56)
.18
Intent-to-treat
.05 (.55)
.14
Completers
-.13 (.58)
.19
Intent-to-treat
.00 (.44)
-
Completers
.38 (.62)
.66
Intent-to-treat
.20 (.51)
.70
Outcome Expectations
Self-Regulation
Enjoyment
Social Support Family
Social Support Friends
Note: Effect size calculated by Cohen’s d statistic. Positive values denote increases in
Social Cognitive Theory variables. Intent-to-treat analyses utilized the last observation
provided by each participant carried forward for missing values at midpoint or six
months. Intent-to-treat N=32; Completers N=13.
119
Regression Analyses between Social Cognitive Theory Change Variables and Physical
Activity
Bivariate regression analyses were conducted to assess whether the significant
change scores observed by the Social Cognitive Theory variables of self-regulation and
social support from friends were associated with changes in physical activity. Findings
from the bivariate regression models assessing the associations between the change
scores for self-regulation and physical activity did not reveal significant associations.
Regression analysis using the change scores for the most recent observation carried
forward for the variable of social support also demonstrated a non-significant relationship
with changes in physical activity. Lastly, the significant baseline to six months change
score for social support from friends among completers was not significantly associated
with changes in physical activity. Table 21 shows the results from these regression
models.
120
Table 21.
Bivariate associations between select Social Cognitive Theory change scores and
changes in physical activity.
df
F
Beta
Effect Size (f2)
p-value
Completers*
12
.70
-.25
.06
.42
Intent-to-treat
31
1.82
.24
.06
.19
Completers
12
.19
.13
.02
.68
Intent-to-treat
31
1.05
.18
.04
.31
Variable
Self-Regulation Scale
for Physical Activity
Social Support for
Exercise from Friends
Note. *Baseline to midpoint change score was used to predict physical activity, the
change score from baseline to six months was not significant. Effect size calculated by
Cohen’s f 2 statistic. Intent-to-treat analyses utilized the last observation provided by each
participant carried forward for missing values at midpoint or six months. Intent-to-treat
N=32; Completers N=13.
Summary
The purpose of the current study was to evaluate the physical activity and
associated Social Cognitive Theory outcomes of outcome expectations, enjoyment, selfregulation, and social support following a six month web-based intervention promoting
physical activity among young African American women. Results indicated that at
121
baseline participants had a mean age of 21.21 (SD=2.31) years, mean BMI of 35.4
(SD=6.82), and reported performing a 75.78 (SD=72.89) minutes per week of physical
activity. No significant bivariate relationships emerged between the Social Cognitive
Theory variables and physical activity at baseline. Completer analyses indicated a
significant pre-post intervention increase in social support from friends (p=.05) and an
increase of 33.23 (SD=97.17) minute per week of physical activity (p=.24). Intent-to-treat
analyses showed an increasing trend of 16.63 (SD=63.82) minutes per week of physical
activity (p=.15) and significant pre-post intervention increases in both self-regulation
(p=.03) and social support from friends (p=.03). Pre-post changes in the Social Cognitive
Theory variables were not associated with pre-post changes in physical activity.
122
CHAPTER 5
DISCUSSION, CONCLUSIONS, AND PUBLIC HEALTH IMPLICATIONS
Introduction
African American women demonstrate low levels of physical activity and share a
disproportionate burden of health conditions associated with being insufficiently active;
indicating the need for innovative and effective approaches to promote physical activity
in this population. Internet-based physical activity interventions have shown great
promise for promoting physical activity; however, no published web-based interventions
have been specifically developed for and tested among the African American female
population. The purpose of the current study was to evaluate the physical activity and
associated Social Cognitive Theory outcomes of outcome expectations, enjoyment, selfregulation, and social support of a culturally-adapted website-based approach to
promoting physical activity and weight loss among young African American females
between the ages of 19 and 30.
The age group of 19 to 30 years was selected as the target population for the study
because this is the period in life in which lifelong physical activity patterns can
potentially be established. Thus, if interventions can successfully promote physical
activity during this age period, the possibility exists that theses increased physical activity
levels can be maintained throughout the lifespan (Britton et al., 2002; Nogueira et al.,
2009), which may ultimately lead to a decrease in physical activity related health
disparities among African American females.
123
Summary of Findings
The current study involved a secondary data analysis of a six month one-group
pre-post test design study assessing the acceptability and feasibility of a culturallyadapted web-based approach to promoting weight loss and physical activity among
African American women aged 19 to 30. The findings presented in this study focus
exclusively on physical activity and associated Social Cognitive Theory outcomes of
outcome expectations, enjoyment, self-regulation, and social support.
At baseline, participants (N=34) had a mean age of 21.21 (SD=2.31) years, mean
BMI of 35.4 (SD=6.82), and reported performing 75.78 (SD=72.89) minutes per week of
physical activity. The purpose of aim one was to cross-sectionally examine the baseline
bivariate relationships between the Social Cognitive Theory variables of outcome
expectations, enjoyment, self-regulation, and social support with the dependent variable
of physical activity. No significant bivariate relationships emerged between the Social
Cognitive Theory variables and physical activity. However, the covariate of BMI had an
inverse relationship with physical activity levels at baseline (ß=.32, p=.08).
For aim two, pre-post intervention changes in physical activity were assessed.
Findings indicated that study completers (n=13) significantly increased physical activity
from baseline to midpoint (three months) by 76.69 (SD=132.28) minutes per week
(p=.06). At the six-month follow-up, completers reported an increase of 33.23
(SD=97.17) minutes of physical activity per week from baseline (p=.24). Intent-to-treat
analyses with participants’ baseline physical activity observations carried forward for
missing values showed a significant increase of 39.71 (SD=95.33) minutes per week
from baseline to midpoint (p=.03) and an increase of 16.63 (SD=63.82) minutes per week
124
at six months (p=.15). The more conservative physical activity findings from the intentto-treat analyses were not surprising as baseline observations were carried forward for
approximately half of the study participants.
The purpose of aim three was to: a) assess pre-post changes in the physical
activity related Social Cognitive Theory constructs of outcome expectations, enjoyment,
self-regulation, and social support, and b) examine whether pre-post changes in the Social
Cognitive Theory constructs were associated with pre-post changes in physical activity.
Both the intent-to-treat and completers analyses indicated a significant increase in social
support from friends over the course of the study (p=.03 and p=.05 respectively). Intentto-treat analysis indicated a significant increase in self-regulation for physical activity
from baseline to six months (p=.03); completers analysis showed a positive trend for
increased self-regulation from baseline to six months, however, this positive trend did not
reach significance (p=.16). No changes in outcome expectations, enjoyment, or social
support from family were observed over the six month study. Bivariate regression
models examining the relationships between the change scores of the Social Cognitive
Theory variables and the change scores in physical activity did not reveal any significant
associations.
Discussion and Conclusions
To the author’s knowledge, the present study is the first to evaluate a culturallyadapted web-based approach to promoting physical activity among African American
women. Findings of the study provide important insight and implications for Internet-
125
based approaches promoting physical activity among young African American women.
The following sections discuss specific findings from the study.
Comparison of Physical Activity and Social Cognitive Theory Findings with other
Studies
Participants who completed the study increased physical activity by 33 minutes
per week at the end of the six month study. While this increase in physical activity was
not statistically significant (most likely due to inadequate power), the findings
demonstrated a positive trend for increased physical activity over the duration of the
study. Due to the lack of web-based studies promoting physical activity exclusively
among African Americans females, comparing the findings of the current study with
other studies focused at the same population was not possible; however, the results can be
placed in the context of web-based physical activity promotion studies among college
students.
The literature review performed for the current study identified eight web-based
studies promoting physical activity among college populations. Of these, three identified
non-significant changes in physical activity similar to the current study (Franko et al.,
2008; Gow et al., 2010; Lachausse, 2012). However, only one of the web-based studies
evaluated the effects of six month intervention (Franko et al., 2008); the durations of the
other studies ranged from six weeks to three months. Therefore, the non-significant
physical activity outcomes of the current study paralleled the findings of the only other
six month intervention identified. The lack of longer-term interventions promoting
126
physical activity in this population indicates a clear need for longer-term interventions in
the college-aged population.
The significant increase in physical activity from baseline to midpoint, and
subsequent decline in physical activity from midpoint to six months found in the current
study is similar to the results of other longer-term physical activity interventions (Carr et
al., 2012; Glasgow et al., 2012; Marcus et al., 2007). These findings suggest that young
African American women, like other populations, may lose interest and/or motivation in
performing physical activity over time. Future studies should investigate potential
strategies to preserve initial intervention improvements in physical activity.
Social support from friends for physical activity significantly increased over the
duration of the study. This finding suggests that the web-based social networking
components and structured exercise sessions were effective in fostering social support for
physical activity. Regression analyses did not show an association between the increase
in social support and changes in physical activity over the duration of the study; however,
this may be due to lack of power due to the small sample size of the study as many
studies have shown social support as a salient variable influencing physical activity levels
(Anderson et al., 2010; Anderson-Bill, Winett, & Wojcik, 2011; Anderson-Bill, Winett,
Wojcik, et al., 2011; Martin et al., 2011; Rovniak et al., 2002)
The non-significant change in social support from family over the duration of the
study was not surprising since the study did not include any family-level or family
focused components. However, the finding that study completers (compared to noncompleters) demonstrated higher levels of social support from family at baseline provides
an interesting point to consider. Perhaps, if the current web-based study included family-
127
focused components in the physical activity promotion efforts, the attrition would have
been lower. Furthermore, given the breadth of research suggesting that African American
women have a prominent caregiver role in the family that influences their available time
to participate in physical activity (Bopp et al., 2007; Richter et al., 2002; Wilcox et al.,
2002), incorporation of family-based components in web-based efforts may be effective
in promoting physical activity in this population.
The Social Cognitive Theory construct of outcome expectations for physical
activity was not associated with physical activity levels at baseline or six months, and the
non-significant change in this construct over the duration of the study was similar to
findings from other studies (Goran & Reynolds, 2005; Grim et al., 2011; Rovniak et al.,
2002). The fact that study completers (compared to non-completers) demonstrated higher
outcome expectations for physical activity and higher physical activity levels at baseline
provides a point for speculation. Perhaps, if the intervention incorporated a lead-in
component promoting positive outcome expectations for physical activity prior to
intervening to promote physical activity, attrition rates would be lower; which would
have provided more intervention exposure to participants and may have ultimately
increased physical activity levels.
Enjoyment of physical activity was not associated with physical activity levels at
baseline or six months and did not change over course of the study. A similar nonsignificant relationship between physical activity levels and enjoyment of physical
activity was noted in a study conducted by (Bopp et al., 2006). Other studies evaluating
the construct in African American populations have shown significant changes in
enjoyment of physical activity over the duration of their respective interventions (Bopp et
128
al., 2009; Papandonatos et al., 2012); however, the association between enjoyment and
physical activity levels were not provided making the findings of the current study
difficult to compare to others. Participants in the current study reported relatively high
levels of enjoyment for physical activity at both time points, potentially implicating that
enjoyment may not influence whether or not young African American females participate
in physical activity. Due to the limited amount of research on enjoyment of physical
activity among African American women, more studies are needed on this topic.
Intent-to-treat analyses showed that self-regulation significantly increased over
the six month study. Completers analysis showed a similar mean improvement in selfregulation; however, the improvement using this analysis technique was not statistically
significant. This discrepancy in significance is likely due to issues related to the wider
standard deviation demonstrated by study completers at six months. Despite the nonsignificant finding for completers, a positive trend for increased self-regulation for
physical activity was observed over the duration of the study. Regression analyses did not
show an association between self-regulation and physical activity levels; which
contradicts the findings of several studies (Anderson et al., 2010; Anderson-Bill, Winett,
Wojcik, et al., 2011; Rovniak et al., 2002). This lack of association between selfregulation and physical activity found in the current study may be due to inadequate
power. Furthermore, the self-regulation scale used in the current study is a relatively new
measure and has not been previously used exclusively among African American
populations. The use of this scale in the current study provides some support for the
applicability of the measure to the African American populations.
129
Sensitivity of the Social Cognitive Theory Psychosocial Measures
African American women possess cultural, environmental, and/or individual
perceptions associated with performance of physical activity that can differ from those of
White populations (Nies et al., 1999; Siddiqi et al., 2011; Young et al., 2002). While all
of the psychosocial measures used in the current study but one, the Self-Regulation Scale
for Physical Activity, have been previously used and/or validated in African American
populations, the possibility exists that some of the measures (specifically, the measures
for outcome expectations and enjoyment) were not sensitive enough to detect the cultural
and/or personal differences associated with performance of physical activity in the
African American female population. For example, researchers have shown that some
African American women avoid physical activity due to the fear that it will lead to an
unattractive body shape (Mabry et al., 2003; Sanderson et al., 2002; Young et al., 2002)
and/or negatively impact their hair styles (Boyington et al., 2008; Im et al., 2012; Siddiqi
et al., 2011). However, these items (or similar) were not addressed by the outcome
expectations scale for exercise; providing a possible explanation of why there no changes
in the variable were observed. Furthermore, participants reported high values for these
constructs at baseline which provided little room for improvement over the duration of
the study; potentially indicating a ceiling effect for these variables.
Measurement issues associated the sensitivity of the psychosocial measures
among African American women is not only a limitation of the current study, but as the
literature as a whole. The psychosocial measures used to assess the Social Cognitive
Theory constructs in the current study (with the exception of the Self-Regulation Scale)
were selected because they are commonly used to assess their respective constructs
130
among African American women. However, their frequent usage does not necessarily
indicate that they are culturally sensitive or relevant for use African American women.
Future studies should explore the development and/or adaptation of existing psychosocial
measures in order to meet the specific personal and/or cultural needs of the African
American community. Development of such measures is warranted in order to advance
the field of physical activity research among African American women.
Corroboration of the Seven Day Physical Activity Recall with Accelerometers
To corroborate self-reported physical activity levels participants were asked to
wear an Actigraph activity monitor for a seven day period at both baseline and six
months. A comparison of these two physical activity assessment methods revealed a
discrepancy in the amount of physical activity reported. At baseline, participants with
valid accelerometer data self-reported 82.59 minutes per week of physical activity while
accelerometer reported physical activity was 49.52 minutes per week. At six months,
mean self-reported physical activity was 169.00 minutes per week compared to 56.50
minutes per week as measured by the Actigraph. These findings suggest that participants
over self-reported physical activity levels at both time points.
Correlation analyses between the two physical activity assessment methods
revealed a non-significant correlation between the Seven Day Physical Activity Recall
and Actigraph for participants at baseline. A secondary analysis of study completers at
baseline demonstrated a significant positive correlation between the two measures
(rho=.58). At six months, completers again demonstrated a significant positive
correlation (r=.78). The significant correlations between self-reported and objectively
131
measured physical activity among completers was intriguing due to the large difference
in reported physical activity levels between the two measures. This discrepancy suggests
that completers reported physical activity when indeed they were actually performing it;
however, they over-estimated the time spent in physical activity. The over-reporting of
physical activity observed in the current study is not uncommon as people often overestimate self-reported physical activity levels (LeBlanc & Janssen, 2010; Marcus et al.,
2006; Sallis & Saelens, 2000).
Website Usage and Structured Exercise Session Attendance
The algorithm used by the web-based study to collect data on participant website
usage did not provide direct information on the frequency of participant login participants
or duration of time spent on the website. Instead, the website awarded participants points
according to the number of applications used and duration of time spent on the website;
making the interpretation of website usage difficult to assess and compare to other
studies. However, anecdotal evidence from participant feedback and the researcher’s
own experience with using the study website suggests that website usage among
participants was relatively low throughout the study. Moreover, most participants
demonstrated only small increases in weekly accumulated website point totals as the
study progressed (if any increase in points were incurred at all); further indicating less
than ideal website usage. The low web-site usage reported in the current study parallels
the findings of several systematic reviews evaluating website-delivered weight loss and
physical activity interventions (Arem & Irwin, 2011; Neve, Morgan, Jones, & Collins,
2010; Vandelanotte et al., 2007).
132
Attendance of the study’s structured exercise sessions was also relatively low.
Study completers attended a median of 54.92% of structured exercise sessions (range
13.46% to 84.62%). Qualitative data collected from post-intervention focus groups
indicated that participants lost interest in walking on the university’s indoor track over
the course of the study and that the scheduled walking sessions were not convenient with
their work/school schedules. A review of the literature showed the majority of web-based
studies have not incorporated in-person exercise sessions as a part of their study
protocols; therefore, making it difficult to compare exercise session attendance with other
web-based studies. Nonetheless, less than 55% adherence to a major study component
such as the structured exercise sessions paired with the low website usage indicates the
need for innovative strategies to enhance and maintain participant engagement with study
activities.
Power and Sample Size
Post hoc power analyses were performed to determine the observed power of the
paired sample t-tests used to determine pre-post changes in physical activity and the
associated Social Cognitive Theory variables. Findings of these analyses indicated that
the majority of the t-tests performed were underpowered to determine significant changes
at 80% power. For example, the post hoc power analysis for changes in physical activity
among study completers indicated the test had a 30.2% power to detect a significant
change at the p=.10 level. Accordingly, for the 33 minute per week increase in physical
observed in the study to be significant (p=.10 level), the study would have needed 54
participants to complete the study (as opposed to the 13 who actually completed the
133
study). However, despite the lack of power to detect a significant increase in physical
activity, completers demonstrated a 33% increase in physical activity at the six month
follow-up; which can be considered a meaningful impact on physical activity levels.
Post hoc power analyses for the Social Cognitive Theory variables were also
performed. However, the results of these post hoc calculations should be interpreted with
caution. The variables with non-significant pre-post intervention findings (outcome
expectations, enjoyment, and social support from family) showed small pre-post test
decreases in completers analyses and small pre-post test increases with intent-to-treat
analyses; suggesting that these psychosocial variables did not change over the course of
the study. Therefore, the information provided by the post-hoc analysis is somewhat
arbitrary due to the lack of change in these variables across assessment periods.
Strengths, Limitations, and Public Health Implications
Study Strengths
The current study has several strengths. First, this study was one of few
examining the effects of a web-based approach to promote of physical activity
exclusively among African American women. The literature review performed for this
study identified only one study assessing the use of web-based approach specifically
among the African American population (a study by Pekmezi et al., 2010). Findings from
the current study add to the limited research on web-based approaches promoting
physical activity among African Americans. A second strength was that the study is
grounded in behavioral theory. Using behavioral theory in health promotion provides a
framework for researchers to move beyond intuition and apply sound, testable approaches
134
to facilitate behavior change (NCI, 2005). Furthermore, physical activity interventions
based on behavioral theory have the potential to be more effective in promoting physical
activity than non-theoretical interventions (Hamel et al., 2011; Lau et al., 2011).
A third strength of the study was that the same research staff member performed
all of the Seven Day Physical Activity Recall assessments and obtained all of the
participants’ heights and weights to calculate BMI outcomes. Utilizing the same staff
member to collect such outcomes at all of the study assessment periods provided a
consistency of these measures throughout the study and reduced the potential for
measurement variation and error for these study outcomes.
An additional strength was that the study evaluated the outcomes of a relatively
long (six months) web-based intervention promoting physical activity. Many studies
utilizing a web-based approach to promote physical activity have employed much shorter
interventions ranging from four weeks to three months (Vandelanotte et al., 2007). A
final strength was that the study used both subjective and objective methods to assess
physical activity. Self-report physical activity levels assessed by the Seven Day Physical
Activity Recall and data collected from accelerometers were significantly correlated
among study completers (baseline r = 53, p = .08; six month r =. 78, p = .002) and
showed approximately the same increase in physical activity form baseline to six months
(Seven Day Physical Activity Recall=33.23 minutes/week; accelerometer=27.6
minutes/week); providing promising findings for the intervention’ impact on physical
activity levels.
135
Study Limitations
As with all studies, the current one is not without limitations. Lack of a control or
comparison group hinders causal inferences and does not control for various threats to
internal validity such as maturation or secular trends. Therefore, the observed
improvements in physical activity, social support, and self-regulation may have been
influenced by external forces and not by the intervention itself. Other limitations include
the study’s small sample size and high attrition rate. The sample size for completers
analysis was 13; which limited statistical analyses and did not allow for the use of
covariates in modeling procedures. Additionally, post-hoc power analyses revealed that
the study was underpowered to detect significant changes in physical activity.
The high attrition rate in the current study indicates the need for future studies to
focus efforts on innovative strategies to recruit and maintain participant retention.
Previous research has shown that increasing monetary incentives and providing other
tangible items (prizes/merchandise) to reward achievement and acknowledge progress
throughout an intervention can increase both study retention rates and behavioral
outcomes targeted by study activities (Marteau, Ashcroft, & Oliver, 2009; Sutherland,
Christianson, & Leatherman, 2008; Volpp et al., 2008). Such approaches should be
considered for use in future studies to promote website usage, exercise session
attendance, and achievement of study goals. Furthermore, increased monetary incentives
among the study’s target population, young African American female college students,
may show particular promise for participant engagement and retention due to financial
strains experienced by many college students (Nelson, Lust, Story, & Ehlinger, 2008;
Vázquez, Otero, & Díaz, 2012).
136
Low structured exercise session attendance and low website usage were also
limitations. Focus group sessions conducted post-intervention indicated that participants
became bored with walking the indoor track and would have liked the website content to
be updated more frequently. Future studies should incorporate these suggestions by
exploring/allowing other types of physical activity to be performed during exercise
sessions and updating/modifying intervention website frequently in order to maintain
participant engagement with study activities.
Another limitation was the use of a convenience sample of university student
volunteers. Participants may not have been representative of the general African
American female population. Furthermore, college students often experience high levels
of stress and have competing interests for their time such as class, studying, social time
with friends (Elliot, Kennedy, Morgan, Anderson, & Morris, 2012); which may have also
influenced the high attrition rate observed in the study. An additional limitation is that
while the study used a targeted approach to promote physical activity among young
African American women, the intervention messages were not tailored at the individual
level. Some researchers have shown that tailored messages (based on stages of change in
the Transtheoretical model) can be beneficial in promoting physical activity (Carr et al.,
2008; Fjeldsoe et al., 2010; Marcus et al., 2007). A final limitation is that the design did
not allow for assessment of longer-term effects on physical activity following the end of
the study. The current study assessed physical activity at one week following conclusion
of the intervention; therefore, there is no way to determine if the positive trend in
physical activity was maintained after the one week follow-up period.
137
Public Health Implications
The current study showed that a web-based approach promoting physical activity
offers promise for increasing physical activity, social support and self-regulation for
physical activity. To the author’s knowledge, no other web-based approaches have been
developed and/or culturally-adapted to specifically meet the needs of young African
American women. The findings in from the current study provide some preliminary
support for the use of a web-based approach to promoting physical activity among young
African American women.
The current study faced challenges of high attrition and less than ideal website
usage, which are commonly reported problems among physical activity interventions for
African American women (Banks-Wallace & Conn, 2002; Bopp et al., 2009; Pekmezi &
Jennings, 2009; Pekmezi, Barera, Bodenlos, Jones, & Brantly, 2009; Wilbur et al., 2006)
and web-based health behavior change interventions (Neve et al., 2010; Vandelanotte et
al., 2007). However, despite these challenges, the culturally-adapted approach promoted
improvements in physical activity and several psychosocial variables. If issues of
participant retention and website usage can be effectively addressed in future studies,
Internet-based physical activity promotion programs may provide an important strategy
to effectively promote physical activity among college-aged African American females.
Development of innovative physical activity promotion programs for African
American women are needed due to the low physical activity levels and high prevalence
of health conditions associated with lack of physical activity in by this population (U.S.
Cancer Statistics Working Group, 2010). Thus, discovery of effective methods to
successfully promote physical activity in this population will not only provide
138
cardiovascular health benefits, but also help decrease the risk for all cause mortality and
the development of obesity, type II diabetes, select cancers, and many other adverse
health conditions.
139
List of References
American College of Sports Medicine. (2007). ACSM’s guidelines for exercise testing
and prescription. (8th ed.). Philadelphia, PA: Lippincott Williams & Wilkins.
Anderson, E. S., Winett, R. A., Wojcik, J. R., & Williams, D. M. (2010). Social cognitive
mediators of change in a group randomized nutrition and physical activity
intervention: social support, self-efficacy, outcome expectations and selfregulation in the Guide-to-Health Trial. Journal of Health Psychology, 15(1), 2132.
Anderson, E. S., Wojcik, J. R., Winett, R. A., & Williams, D. M. (2006). Social-cognitive
determinants of physical activity: the influence of social support, self-efficacy,
outcome expectations, and self-regulation among participants in a church-based
health promotion study. Health Psychology, 25(4), 510-520.
Anderson-Bill, E. S., Winett, R. A., & Wojcik, J. R. (2011). Social cognitive
determinants of nutrition and physical activity among web-health users enrolling
in an online intervention: the influence of social support, self-efficacy, outcome
expectations, and self-regulation. Journal of Medical Internet Research, 13(1),
e28. doi: 10.2196/jmir.1551
Anderson-Bill, E. S., Winett, R. A., Wojcik, J. R., & Winett, S. G. (2011). Web-based
guide to health: relationship of theoretical variables to change in physical
activity, nutrition and weight at 16-months. Journal of Medical Internet
Research, 13(1), e27. doi: 10.2196/jmir.1614
Araujo-Soares, V., McIntyre, T., MacLennan, G., & Sniehotta, F. F. (2009).
Development and exploratory cluster-randomised opportunistic trial of a theorybased intervention to enhance physical activity among adolescents. Psychology
and Health, 24, 805-822.
Arem, H., & Irwin, M. (2011). A review of web-based weight loss interventions in adults.
Obesity Reviews, 12(5), e236-243.
Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive
framework. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York, NY: W.H.
Freeman and Company.
Banks-Wallace, J., & Conn, V. (2002). Interventions to promote physical activity among
African American women. Public Health Nursing, 19(5), 321-335.
140
Beets, M. W., Cardinal, B. J., & Alderman, B. L. (2010). Parental social support and the
physical activity-related behaviors of youth: a review. Health Education and
Behavior, 37(5), 621-644.
Bélanger-Gravel, A., Godin, G., Vézina-Im, L. A., Amireault, S., & Poirier, P. (2011).
The effect of theory-based interventions on physical activity participation among
overweight/obese individuals: A systematic review. Obesity Reviews, 12(6), 430439.
Bennett, J. B., Broome, K. M., Schwab-Pilley, A., & Gilmore, P. (2011). A web-based
approach to address cardiovascular risks in managers: Results of a randomized
trial. Journal of Occupational and Environmental Medicine, 53(8), 911-918.
Booth, A. O., Nowson, C. A., & Matters, H. (2008). Evaluation of an interactive,
Internet-based weight loss program: A pilot study. Health Education Research,
23(3), 371-381.
Bopp, M., Lattimore, D., Wilcox, S., Laken, M., McClorin, L., Swinton, R., . . . Bryant,
D. (2007). Understanding physical activity participation in members of an
African American church: A qualitative study. Health Education Research,
22(6), 815-826.
Bopp, M., Wilcox, S., Laken, M., Butler, K., Carter, R. E., McClorin, L., & Yancey, A.
(2006). Factors associated with physical activity among African-American men
and women. American Journal of Preventative Medicine, 30(4), 340-346.
Bopp, M., Wilcox, S., Laken, M., Hooker, S. P., Parra-Medina, D., Saunders, R., . . .
McClorin, L. (2009). 8 Steps to Fitness: A faith-based, behavior change physical
activity intervention for African Americans. Journal of Physical Activity and
Health, 6(5), 568-577.
Bosak, K. A., Yates, B., & Pozehl, B. (2009). Feasibility of an internet physical activity
intervention. Western Journal of Nursing Research, 31(5), 648-661.
Bosak, K. A., Yates, B., & Pozehl, B. (2010). Effects of an Internet physical activity
intervention in adults with metabolic syndrome. Western Journal of Nursing
Research, 32(1), 5-22.
Boyington, J. E., Carter-Edwards, L., Piehl, M., Hutson, J., Langdon, D., & McManus, S.
(2008). Cultural attitudes toward weight, diet, and physical activity among
overweight African American girls. Preventing Chronic Disease, 5(2), A36.
Britton, J.A., Gammon, M.D., Kelsey, J.K., Brogan, D.J., Coates, R.J., Schoenberg,
J.B…Brinton, L.A. (2002). Characterstics assocaited with recent recreational
exercise among women 20 to 40 years of age. Women's Health, 3(2-3), 81-96.
Cameron, L. D., & Leventhal, H. (2003). The self-regulation of health and illness
behavior. London, New York: Routledge.
141
Carlson, J. A., Sallis, J. F., Ramirez, E. R., Patrick, K., & Norman, G. J. (2012). Physical
activity and dietary behavior change in Internet-based weight loss interventions:
comparing two multiple-behavior change indices. Preventative Medicine, 54(1),
50-54.
Carr, L. J., Bartee, R. T., Dorozynski, C., Broomfield, J. F., Smith, M. L., & Smith, D. T.
(2008). Internet-delivered behavior change program increases physical activity
and improves cardiometabolic disease risk factors in sedentary adults: Results of
a randomized controlled trial. Preventative Medicine, 46, 431-438.
Carr, L. J., Bartee, R. T., Dorozynski, C. M., Broomfield, J. F., Smith, M. L., & Smith, D.
T. (2009). Eight-month follow-up of physical activity and central adiposity:
Results from an Internet-delivered randomized control trial intervention. Journal
of Physical Activity and Health, 6(4), 444-455.
Carr, L. J., Dunsiger, S. I., Lewis, B., Ciccolo, J. T., Hartman, S., Bock, B., . . . Marcus,
B. H. (2012). Randomized controlled trial testing an Internet physical activity
intervention for sedentary adults. Health Psychology. doi: 10.1037/a0028962
Caspersen, C. J., Pereira, M. A., & Curran, K. M. (2000). Changes in physical activity
patterns in the United States, by sex and cross-sectional age. Medicine and
Science in Sports and Exercise, 32(9), 1601-1609.
Centers for Disease Control and Prevention. (2010). U.S. Physical activity statistics.
Atlanta, GA: Centers for Disease Control and Prevention.
Chou, W. Y., Hunt, Y. M., Beckjord, E. B., Moser, R. P., & Hesse, B. W. (2009). Social
media use in the United States: Implications for health communication. Journal
of Medical Internet Research, 11(4), e48. doi: 10.2196/jmir.1249
Decker, E. B. (2007). Physician interventions combined with a pedometer-driven walking
program to increase physical activity. Retrieved from Dissertations & Theses:
Full Text. (AAT 1441146)
Dimaggio, P., Hargittai, E., Celeste, C., & Shafer, S. (2004). Digial Inequality: From
unequal access to differentiated use. In K. Nekerman (Ed.), Social Inequality,
355-400. New York, NY: Russell Sage Foundation.
Dinger, M. K., Heesch, K. C., Cipriani, G., & Qualls, M. (2007). Comparison of two
email-delivered, pedometer-based interventions to promote walking among
insufficiently active women. Journal of Science and Medicine in Sport, 10(5),
297-302.
Dishman, R. K., & Steinhardt, M. (1988). Reliability and concurrent validity for a 7-d recall of physical activity in college students. Medicine and Science in Sports and
Exercise, 20(1), 14-25.
142
Dlugonski, D., Motl, R. W., & McAuley, E. (2011). Increasing physical activity in
multiple sclerosis: replicating Internet intervention effects using objective and
self-report outcomes. Journal of Rehabilitation Research and Development,
48(9), 1129-1136.
Dunn, A. L., Andersen, R. E., & Jakicic, J. M. (1998). Lifestyle physical activity
interventions. History, short- and long-term effects, and recommendations.
American Journal of Preventative Medicine, 15(4), 398-412.
Dunton, G. F., & Robertson, T. P. (2008). A tailored Internet-plus-email intervention for
increasing physical activity among ethnically-diverse women. Preventative
Medicine, 47(6), 605-611.
Elliot, C. A., Kennedy, C., Morgan, G., Anderson, S. K., & Morris, D. (2012).
Undergraduate physical activity and depressive symptoms: A national study.
American Journal of Health Behavior, 36(2), 230-241.
Faghri, P. D., Omokaro, C., Parker, C., Nichols, E., Gustavesen, S., & Blozie, E. (2008).
E-technology and pedometer walking program to increase physical activity at
work. Journal of Primary Prevention, 29(1), 73-91.
Ferney, S. L., Marshall, A. L., Eakin, E. G., & Owen, N. (2009). Randomized trial of a
neighborhood environment-focused physical activity website intervention.
Preventative Medicine, 48(2), 144-150.
Fjeldsoe, B. S., Miller, Y. D., & Marshall, A. L. (2010). MobileMums: A randomized
controlled trial of an SMS-based physical activity intervention. Annals of
Behavioral Medicine, 39(2), 101-111.
Franko, D. L., Cousineau, T. M., Trant, M., Green, T. C., Rancourt, D., Thompson, D., . .
. Ciccazzo, M. (2008). Motivation, self-efficacy, physical activity and nutrition in
college students: randomized controlled trial of an internet-based education
program. Preventative Medicine, 47(4), 369-377.
Glasgow, R. E., Boles, S. M., McKay, H. G., Feil, E. G., & Barrera, M. (2003). The DNet diabetes self-management program: Long-term implementation, outcomes,
and generalization results. Preventative Medicine, 36(4), 410-419.
Glasgow, R. E., Kurz, D., King, D., Dickman, J. M., Faber, A. J., Halterman, E., . . .
Ritzwoller, D. (2010). Outcomes of minimal and moderate support versions of an
Internet-based diabetes self-management support program. Journal of General
Internal Medicine, 25(12), 1315-1322.
Glasgow, R. E., Kurz, D., King, D., Dickman, J. M., Faber, A. J., Halterman, E., . . .
Ritzwoller, D. (2012). Twelve-month outcomes of an Internet-based diabetes
self-management support program. Patient Education and Counseling, 87(1), 8192.
143
Goode, A. D., Reeves, M. M., & Eakin, E. G. (2012). Telephone-delivered interventions
for physical activity and dietary behavior change: an updated systematic review.
American Journal of Preventative Medicine, 42(1), 81-88.
Goran, M. I., & Reynolds, K. (2005). Interactive multimedia for promoting physical
activity (IMPACT) in children. Obesity Research, 13(4), 762-771.
Gow, R. W., Trace, S. E., & Mazzeo, S. E. (2010). Preventing weight gain in first year
college students: An online intervention to prevent the "freshman fifteen". Eating
Behaviors, 11(1), 33-39.
Graham, R., & Smith, D. T. (2001). Internet as digital practice: Examining differences in
African American internet usage. Future Internet, 3, 185-203.
Grim, M., Hortz, B., & Petosa, R. (2011). Impact evaluation of a pilot web-based
intervention to increase physical activity. American Journal of Health
Promotion, 25(4), 227-230.
Gross, L. D., Sallis, J. F., Buono, M. J., Roby, J. J., & Nelson, J. A. (1990). Reliability of
interviewers using the Seven-Day Physical Activity Recall. Research Quarterly
in Exercise and Sports, 61(4), 321-325.
Hagberg, L. A., Lindahl, B., Nyberg, L., & Hellénius, M. L. (2009). Importance of
enjoyment when promoting physical exercise. Scandinavian Journal of Medicine
and Science in Sports, 19(5), 740-747.
Hageman, P. A., Walker, S. N., & Pullen, C. H. (2005). Tailored versus standard internetdelivered interventions to promote physical activity in older women. Journal of
Geriatric Physical Therapy, 28(1), 28-33.
Hallam, J. S., & Petosa, R. (2004). The long-term impact of a four-session work-site
intervention on selected social cognitive theory variables linked to adult exercise
adherence. Health Education and Behavior, 31(1), 88-100.
Hamel, L. M., Robbins, L. B., & Wilbur, J. (2011). Computer- and web-based
interventions to increase preadolescent and adolescent physical activity: A
systematic review. Journal of Advanced Nursing, 67(2), 251-268.
Harvey-Berino, J., Pintauro, S., Buzzell, P., DiGiulio, M., Casey Gold, B., Moldovan, C.,
& Ramirez, E. (2002). Does using the Internet facilitate the maintenance of
weight loss? International Journal of Obesity Related Metabolic Disorders,
26(9), 1254-1260.
Harvey-Berino, J., Pintauro, S., Buzzell, P., & Gold, E. C. (2004). Effect of Internet
support on the long-term maintenance of weight loss. Obesity Research, 12(2),
320-329.
144
Harvey-Berino, J., Pintauro, S. J., & Gold, E. C. (2002). The feasibility of using Internet
support for the maintenance of weight loss. Behavior Modification, 26(1), 103116.
Harvey-Berino, J., West, D., Krukowski, R., Prewitt, E., VanBiervliet, A., Ashikaga, T.,
& Skelly, J. (2010). Internet delivered behavioral obesity treatment. Preventative
Medicine, 51(2), 123-128.
Huang, S. J., Hung, W. C., Chang, M., & Chang, J. (2009). The effect of an internetbased, stage-matched message intervention on young Taiwanese women's
physical activity. Journal of Health Communication, 14, 210-227.
Hurling, R., Catt, M., Boni, M. D., Fairley, B. W., Hurst, T., Murray, P., . . . Sodhi, J. S.
(2007). Using Internet and mobile phone technology to deliver an automated
physical activity program: Randomized controlled trial. Journal of Medical
Internet Research, 9(2), e7. doi: 10.2196/jmir.9.2.e7
Im, E. O., Ko, Y., Hwang, H., Yoo, K. H., Chee, W., Stuifbergen, A., . . . Chee, E.
(2012). "Physical activity as a luxury": African American women's attitudes
toward physical activity. Western Journal of Nursing Research, 34(3), 317-339.
Irvine, A. B., Philips, L., Seeley, J., Wyant, S., Duncan, S., & Moore, R. W. (2011). Get
moving: A web site that increases physical activity of sedentary employees.
American Journal of Health Promotion, 25(3), 199-206.
Jakicic, J. M., & Otto, A. D. (2005). Physical activity considerations for the treatment and
prevention of obesity. American Journal of Clinical Nutrition, 82, 226S-229S.
Janz, K. F. (2006). Physical activity in epidemiology: moving from questionnaire to
objective measurement. British Journal of Sports Medicine, 40(3), 191-192.
Kelders, S. M., Van Gemert-Pijnen, J. E., Werkman, A., Nijland, N., & Seydel, E. R.
(2011). Effectiveness of a Web-based intervention aimed at healthy dietary and
physical activity behavior: A randomized controlled trial about users and usage.
Journal of Medical Internet Research, 13(2), e32. doi: 10.2196/jmir.1624
Kelders, S. M., van Gemert-Pijnen, J. E., Werkman, A., & Seydel, E. R. (2010). Usage
and effect of a web-based intervention for the prevention of overweight; A RCT.
Studies in Health Technologies and Informatics, 160(1), 28-32.
Kendzierski, D., & DeCarlo, K. J. (1991). Physical Activity Enjoyment Scale: Two
validation studies. Journal of Sport and Exercise Psychology, 13(1), 50-64.
Kim, C., Draska, M., Hess, M. L., Wilson, E. J., & Richardson, C. R. (2012). A webbased pedometer programme in women with a recent history of gestational
diabetes. Diabetic Medicine, 29(2), 278-283.
145
Kimm, S. Y., Glynn, N. W., Kriska, A. M., Barton, B. A., Kronsberg, S. S., Daniels, S.
R., . . . Liu, K. (2002). Decline in physical activity in Black girls and White girls
during adolescence. New England Journal of Medicine, 347(10), 709-715.
Kontos, E. Z., Emmons, K. M., Puleo, E., & Viswanath, K. (2010). Communication
inequalities and public health implications of adult social networking site use in
the United States. Journal of Health Communication, 15 (Suppl 3), 216-235.
Koregan, K., Odell, P., & Schumacher, P. (2001). Internet use among college students:
Are there differences by race/ethnicity? Electronic Journal of Sociology, 5.
Retrieved from http://www.sociology.org/content/vol005.003/korgen.html
Kosma, M., Cardinal, B. J., & McCubbin, J. A. (2005). A pilot study of a web-based
physical activity motivational program for adults with physical disabilities.
Disability Rehabilitation, 27(23), 1435-1442.
Lachausse, R. G. (2012). My student body: Effects of an Internet-based prevention
program to decrease obesity among college students. Journal of American
College Health, 60(4), 324-330.
Lau, P. W., Lau, E. Y., Wong del, P., & Ransdell, L. (2011). A systematic review of
information and communication technology-based interventions for promoting
physical activity behavior change in children and adolescents. Journal of Medical
Internet Research, 13, e4.
LeBlanc, A. G., & Janssen, I. (2010). Difference between self-reported and accelerometer
measured moderate-to-vigorous physical activity in youth. Pediatric Exercise
Science, 22(4), 523-534.
Leslie, E., Marshall, A. L., Owen, N., & Bauman, A. (2005). Engagement and retention
of participants in a physical activity website. Preventative Medicine, 40 (1), 5459.
Lewis, C. E., Jacobs, D. R., McCreath, H., Kiefe, C. I., Schreiner, P. J., Smith, D. E., &
Williams, O. D. (2000). Weight gain continues in the 1990s: 10-year trends in
weight and overweight from the CARDIA study. Coronary Artery Risk
Development in Young Adults. American Journal of Epidemiology, 151(12),
1172-1181.
Lieber, S. B., Redberg, R. F., Blumenthal, R. S., Gandhi, A., Robb, K. J., & Mora, S.
(2012). A national interactive web-based physical activity intervention in women,
evaluation of the American Heart Association Choose to Move Program 20062007. American Journal of Cardiology, 109(12), 1754-1760.
Liebreich, T., Plotnikoff, R. C., Courneya, K. S., & Boulé, N. (2009). Diabetes
NetPLAY: A physical activity website and linked email counselling randomized
intervention for individuals with type 2 diabetes. International Journal of
Behavioral Nutrition and Physical Activity, 6, 18.doi: 10.1186/1479-5868-6-18
146
Mabry, I. R., Young, D. R., Cooper, L. A., Meyers, T., Joffe, A., & Duggan, A. K.
(2003). Physical activity attitudes of African American and White adolescent
girls. Ambulatory Pediatrics, 3(6), 312-316.
Magoc, D., Tomaka, J., & Bridges-Arzaga, A. (2011). Using the web to increase physical
activity in college students. American Journal of Health Behavior, 35(2), 142154.
Mailey, E. L., Wójcicki, T. R., Motl, R. W., Hu, L., Strauser, D. R., Collins, K. D., &
McAuley, E. (2010). Internet-delivered physical activity intervention for college
students with mental health disorders: A randomized pilot trial. Psychology,
Health, and Medicine, 15(6), 646-659.
Marcus, B. H., & Forsyth, L. H. (2009). Motivating people to be physically Active (2nd
ed.). Champaign, IL: Human Kinetics.
Marcus, B. H., Lewis, B. A., Williams, D. M., Dunsiger, S., Jakicic, J. M., Whiteley, J.
A., . . . Parisi, A. F. (2007). A comparison of Internet and print-based physical
activity interventions. Archives of Internal Medicine, 167(9), 944-949.
Marcus, B. H., Napolitano, M. A., King, A. C., Lewis, B. A., Whiteley, J. A., Albrecht,
A., . . . Papandonatos, G. D. (2007). Telephone versus print delivery of an
individualized motivationally tailored physical activity intervention: Project
STRIDE. Health Psychology, 26(4), 401-409.
Marcus, B. H., Nigg, C. R., Riebe, D., & Forsyth, L. H. (2000). Interactive
communication strategies: Implications for population-based physical-activity
promotion. American Journal of Preventative Medicine, 19, 121-126.
Marcus, B. H., Owen, N., Forsyth, L. H., Cavill, N. A., & Fridinger, F. (1998). Physical
activity interventions using mass media, print media, and information technology.
American Journal of Preventative Medicine, 15(4), 362-378.
Marcus, B. H., Williams, D. M., Dubbert, P. M., Sallis, J. F., King, A. C., Yancey, A. K.,
. . . Claytor, R. P. (2006). Physical activity intervention studies: what we know
and what we need to know: A scientific statement from the American Heart
Association Council on Nutrition, Physical Activity, and Metabolism
(Subcommittee on Physical Activity); Council on Cardiovascular Disease in the
Young; and the Interdisciplinary Working Group on Quality of Care and
Outcomes Research. Circulation, 114, 2739-2752.
Marshall, A. L., Leslie, E. R., Bauman, A. E., Marcus, B. H., & Owen, N. (2003). Print
versus website physical activity programs: A randomized trial. American Journal
of Preventative Medicine, 25(2), 88-94.
Marshall, A. L., Owen, N., & Bauman, A. E. (2004). Mediated approaches for
influencing physical activity: Update of the evidence on mass media, print,
147
telephone and website delivery of interventions. Journal of Science and Medicine
in Sport, 7 (1 Suppl), 74-80.
Marteau, T. M., Ashcroft, R. E., & Oliver, A. (2009). Using financial incentives to
achieve healthy behaviour. BMJ, 338, b1415.
Martin, J. J., McCaughtry, N., Flory, S., Murphy, A., & Wisdom, K. (2011). Using social
cognitive theory to predict physical activity and fitness in underserved middle
school children. Research Quarterly in Exercise and Sport, 82(2), 247-255.
McConnon, A., Kirk, S. F., Cockroft, J. E., Harvey, E. L., Greenwood, D. C., Thomas, J.
D., . . . Bojke, L. (2007). The Internet for weight control in an obese sample:
results of a randomised controlled trial. BMC Health Service Research, 7, 206.
doi:10.1186/1427-6963-7-206
McKay, H. G., King, D., Eakin, E. G., Seeley, J. R., & Glasgow, R. E. (2001). The
diabetes network internet-based physical activity intervention: A randomized
pilot study. Diabetes Care, 24(8), 1328-1334.
Morgan, P. J., Collins, C. E., Plotnikoff, R. C., Cook, A. T., Berthon, B., Mitchell, S., &
Callister, R. (2011). Efficacy of a workplace-based weight loss program for
overweight male shift workers: the Workplace POWER (Preventing Obesity
Without Eating like a Rabbit) randomized controlled trial. Preventative Medicine,
52(5), 317-325.
Morgan, P. J., Lubans, D. R., Collins, C. E., Warren, J. M., & Callister, R. (2009). The
SHED-IT randomized controlled trial: Evaluation of an Internet-based weightloss program for men. Obesity (Silver Spring), 17(11), 2025-2032.
Motl, R. W., Dlugonski, D., Wójcicki, T. R., McAuley, E., & Mohr, D. C. (2011).
Internet intervention for increasing physical activity in persons with multiple
sclerosis. Multiple Sclerosis, 17(1), 116-128.
Napolitano, M. A., Fotheringham, M., Tate, D., Sciamanna, C., Leslie, E., Owen, N., . . .
Marcus, B. (2003). Evaluation of an Internet-based physical activity intervention:
A preliminary investigation. Annals of Behavioral Medicine, 25(2), 92-99.
National Cancer Institute. (2005). Theory at a glance: A guide for health promotion and
practice (2nd ed.). Bethesda, MD: U.S. Department of Health and Human
Services.
Nelson, M. C., Lust, K., Story, M., & Ehlinger, E. (2008). Credit card debt, stress and key
health risk behaviors among college students. American Journal of Health
Promotion, 22(6), 400-407. doi: 10.4278/ajhp.22.6.400
Neve, M., Morgan, P. J., Jones, P. R., & Collins, C. E. (2010). Effectiveness of webbased interventions in achieving weight loss and weight loss maintenance in
148
overweight and obese adults: A systematic review with meta-analysis. Obesity
Reviews, 11(4), 306-321.
Newton, R. L., & Perri, M. G. (2004). A randomized pilot trial of exercise promotion in
sedentary African-American adults. Ethnicity and Disease, 14(4), 548-557.
Nies, M. A., Vollman, M., & Cook, T. (1999). African American women's experiences
with physical activity in their daily lives. Public Health Nursing, 16(1), 23-31.
Nogueira, D., Faerstein, E., Rugani, I., Chor, D., Lopes, C.S., & Wernech, G.L. (2009).
Does leisure-time physical activity in early adulthood predict later physical
activity? Reviews of Brazilian Epidemiology, 12(1), 3-9.
Oenema, A., Brug, J., Dijkstra, A., de Weerdt, I., & de Vries, H. (2008). Efficacy and use
of an Internet-delivered computer-tailored lifestyle intervention, targeting
saturated fat intake, physical activity and smoking cessation: A randomized
controlled trial. Annals of Behavioral Medicine, 35(2), 125-135.
Onwuegbuzie, A., J., & Leech, N., L. (2004). Post hoc Power: A concept whose time has
come. Understanding Statistics, 3 (4), 29.
Pallant, J. (2007). SPSS survival manual (3rd ed.), (pp. 205-207). New York, NY: Open
University Press.
Papandonatos, G. D., Williams, D. M., Jennings, E. G., Napolitano, M. A., Bock, B. C.,
Dunsiger, S., & Marcus, B. H. (2012). Mediators of physical activity behavior
change: Findings from a 12-month randomized controlled trial. Health
Psychology, 31(4), 512-520.
Patrick, K., Calfas, K. J., Norman, G. J., Rosenberg, D., Zabinski, M. F., Sallis, J. F., . . .
Dillon, L. W. (2011). Outcomes of a 12-month web-based intervention for
overweight and obese men. Annals of Behavioral Medicine, 42(3), 391-401.
Pekmezi, D., & Jennings, E. (2009). Interventions to promote physical activity among
African Americans. American Journal of Lifestyle Medicine, 3, 173-184.
Pekmezi, D. W., Barera, B. L., Bodenlos, J. S., Jones, G. N., & Brantly, P. J. (2009).
Promoting physical activity in low income African Americans: Project LAPS.
Journal of Health Disparities Research and Practice, 3(2), 9. Retrieved from
http://digitalscholarship.unlv.edu/jhdrp/vol3/iss2/7
Pekmezi, D. W., Williams, D. M., Dunsiger, S., Jennings, E. G., Lewis, B. A., Jakicic, J.
M., & Marcus, B. H. (2010). Feasibility of using computer-tailored and internetbased interventions to promote physical activity in underserved populations.
Telemedicine Journal and E-Health, 16(4), 498-503.
149
Pellegrini, C. A., Verba, S. D., Otto, A. D., Helsel, D. L., Davis, K. K., & Jakicic, J. M.
(2012). The comparison of a technology-based system and an in-person
behavioral weight loss intervention. Obesity (Silver Spring), 20(2), 356-363.
Pereira, M. A., FitzerGerald, S. J., Gregg, E. W., Joswiak, M. L., Ryan, W. J., Suminski,
R. R., . . . Zmuda, J. M. (1997). A collection of physical activity questionnaires
for health-related research. Medicine and Science in Sports and Exercise, 29(6
Suppl), S1-205.
Plasqui, G., & Westerterp, K. R. (2007). Physical activity assessment with
accelerometers: An evaluation against doubly labeled water. Obesity (Silver
Spring), 15(10), 2371-2379.
Plotnikoff, R. C., McCargar, L. J., Wilson, P. M., & Loucaides, C. A. (2005). Efficacy of
an E-mail intervention for the promotion of physical activity and nutrition
behavior in the workplace context. American Journal of Health Promotion,
19(6), 422-429.
Pressler, A., Knebel, U., Esch, S., Kölbl, D., Esefeld, K., Scherr, J., . . . Leimeister, J. M.
(2010). An Internet-delivered exercise intervention for workplace health
promotion in overweight sedentary employees: A randomized trial. Preventative
Medicine, 51(3-4), 234-239.
Reid, R. D., Morrin, L. I., Beaton, L. J., Papadakis, S., Kocourek, J., McDonnell, L., . . .
Pipe, A. L. (2011). Randomized trial of an Internet-based computer-tailored
expert system for physical activity in patients with heart disease. European
Journal of Cardiovascular Prevention and Rehabilitation. doi:
10.1177/1741826711422988
Resnick, B., Luisi, D., Vogel, A., & Junaleepa, P. (2004). Reliability and validity of the
self-efficacy for exercise and outcome expectations for exercise scales with
minority older adults. Journal of Nursing Measurement, 12(3), 235-247.
Resnick, B., Zimmerman, S. I., Orwig, D., Furstenberg, A. L., & Magaziner, J. (2000).
Outcome Expectations for Exercise Scale: utility and psychometrics. Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 55(6), S352356.
Resnicow, K., Jackson, A., Blissett, D., Wang, T., McCarty, F., Rahotep, S., &
Periasamy, S. (2005). Results of the Healthy Body Healthy Spirit trial. Health
Psychology, 24(4), 339-348.
Richardson, C. R., Mehari, K. S., McIntyre, L. G., Janney, A. W., Fortlage, L. A., Sen,
A., . . . Piette, J. D. (2007). A randomized trial comparing structured and lifestyle
goals in an internet-mediated walking program for people with type 2 diabetes.
International Journal of Behavioral Nutrition and Physical Activity, 4, 59. doi:
10.1186/1479-58868-4-59
150
Richter, D. L., Wilcox, S., Greaney, M. L., Henderson, K. A., & Ainsworth, B. E. (2002).
Environmental, policy, and cultural factors related to physical activity in African
American women. Women and Health, 36(2), 91-109.
Rothney, M. P., Schaefer, E. V., Neumann, M. M., Choi, L., & Chen, K. Y. (2008).
Validity of physical activity intensity predictions by ActiGraph, Actical, and RT3
accelerometers. Obesity (Silver Spring), 16(8), 1946-1952.
Rovniak, L. S., Anderson, E. S., Winett, R. A., & Stephens, R. S. (2002). Social cognitive
determinants of physical activity in young adults: a prospective structural
equation analysis. Annals of Behavioral Medicine, 24(2), 149-156.
Rovniak, L. S., Hovell, M. F., Wojcik, J. R., Winett, R. A., & Martinez-Donate, A. P.
(2005). Enhancing theoretical fidelity: An e-mail-based walking program
demonstration. American Journal of Health Promotion, 20(2), 85-95.
Sallis, J. F., Buono, M. J., Roby, J. J., Micale, F. G., & Nelson, J. A. (1993). Seven-day
recall and other physical activity self-reports in children and adolescents.
Medicine and Science in Sports and Exercise, 25(1), 99-108.
Sallis, J. F., Grossman, R. M., Pinski, R. B., Patterson, T. L., & Nader, P. R. (1987). The
development of scales to measure social support for diet and exercise behaviors.
Preventative Medicine, 16(6), 825-836.
Sallis, J. F., Haskell, W. L., Wood, P. D., Fortmann, S. P., Rogers, T., Blair, S. N., &
Paffenbarger, R. S., Jr. (1985). Physical activity assessment methodology in the
Five-City Project. American Journal of Epidemiology, 121(1), 91-106.
Sallis, J. F., Patterson, T. L., Buono, M. J., & Nader, P. R. (1988). Relation of
cardiovascular fitness and physical activity to cardiovascular disease risk factors
in children and adults. American Journal of Epidemiology, 127(5), 933-941.
Sallis, J. F., & Saelens, B. E. (2000). Assessment of physical activity by self-report:
Status, limitations, and future directions. Research Quarterly in Exercise and
Sport, 71 (2 Suppl), S1-14.
Sanderson, B., Littleton, M., & Pulley, L. (2002). Environmental, policy, and cultural
factors related to physical activity among rural, African American women.
Women and Health, 36(2), 75-90.
Schmidt, M. D., Cleland, V. J., Thomson, R. J., Dwyer, T., & Venn, A. J. (2008). A
comparison of subjective and objective measures of physical activity and fitness
in identifying associations with cardiometabolic risk factors. Annals of
Epidemiology, 18(5), 378-386.
Schneider, M., & Cooper, D. M. (2011). Enjoyment of exercise moderates the impact of a
school-based physical activity intervention. International Journal of Behavioral
151
Nutrition and Physical Activity, 8, 64. Retrieved from
http://www.ijbnpa.org/content/8/1/64
Shephard, R. J. (2003). Limits to the measurement of habitual physical activity by
questionnaires. British Journal of Sports Medicine, 37(3), 197-206.
Siddiqi, Z., Tiro, J. A., & Shuval, K. (2011). Understanding impediments and enablers to
physical activity among African American adults: A systematic review of
qualitative studies. Health Education Research, 26(6), 1010-1024.
Slootmaker, S. M., Chinapaw, M. J., Schuit, A. J., Seidell, J. C., & Van Mechelen, W.
(2009). Feasibility and effectiveness of online physical activity advice based on a
personal activity monitor: Randomized controlled trial. Journal of Medical
Internet Research, 11 (3), e27. doi: 10.2196/jmir1139
Smith, A. (2010). Technology trends among people of color. Retrieved from
www.pewinternet.org.
Spittaels, H., De Bourdeaudhuij, I., & Vandelanotte, C. (2007). Evaluation of a websitedelivered computer-tailored intervention for increasing physical activity in the
general population. Preventative Medicine, 44(3), 209-217.
Steele, R., Mummery, W. K., & Dwyer, T. (2007). Using the Internet to promote physical
activity: A randomized trial of intervention delivery modes. Journal of Physical
Activity and Health, 4(3), 245-260.
Steele, R. M., Mummery, W. K., & Dwyer, T. (2009). A comparison of face-to-face or
Internet-delivered physical activity intervention on targeted determinants. Health
Education and Behavior, 36(6), 1051-1064.
Sternfeld, B., Block, C., Quesenberry, C. P., Block, T. J., Husson, G., Norris, J. C., . . .
Block, G. (2009). Improving diet and physical activity with ALIVE: A worksite
randomized trial. American Journal of Preventative Medicine, 36(6), 475-483.
Sutherland, K., Christianson, J. B., & Leatherman, S. (2008). Impact of targeted financial
incentives on personal health behavior: a review of the literature. Medical Care
Research Reviews, 65(6 Suppl), 36S-78S.
Tate, D. F., Jackvony, E. H., & Wing, R. R. (2003). Effects of Internet behavioral
counseling on weight loss in adults at risk for type 2 diabetes: A randomized trial.
JAMA, 289 (14), 1833-1836.
Tate, D. F., Wing, R. R., & Winett, R. A. (2001). Using Internet technology to deliver a
behavioral weight loss program. JAMA, 285(9), 1172-1177.
Touger-Decker, R., Denmark, R., Bruno, M., O'Sullivan-Maillet, J., & Lasser, N. (2010).
Workplace weight loss program; comparing live and Internet methods. Journal of
Occupational and Environmental Medicine, 52 (11), 1112-1118.
152
Troiano, R. P., Berrigan, D., Dodd, K. W., Mâsse, L. C., Tilert, T., & McDowell, M.
(2008). Physical activity in the United States measured by accelerometer.
Medicine and Science in Sports and Exercise, 40(1), 181-188.
Trost, S. G., McIver, K. L., & Pate, R. R. (2005). Conducting accelerometer-based
activity assessments in field-based research. Medicine and Science in Sports and
Exercise, 37, 531-543.
Tudor-Locke, C., Bell, R. C., Myers, A. M., Harris, S. B., Ecclestone, N. A., Lauzon, N.,
& Rodger, N. W. (2004). Controlled outcome evaluation of the First Step
Program: A daily physical activity intervention for individuals with type II
diabetes. International Journal of Obesity Related Metabolic Disorders, 28(1),
113-119.
Turner-McGrievy, G. M., Campbell, M. K., Tate, D. F., Truesdale, K. P., Bowling, J. M.,
& Crosby, L. (2009). Pounds Off Digitally study: A randomized podcasting
weight-loss intervention. American Journal of Preventative Medicine, 37(4), 263269.
United States Cancer Statistics Working Group. (2010). United States cancer statistics:
1999–2007 incidence and mortality web-based report. Bethesda, MD: U.S.
Department of Health and Human Services, Centers for Disease Control and
Prevention and National Cancer Institute.
United States Department of Health and Human Services. (2008). 2008 Physical Activity
Guidelines for Americans. Hyattsville, MD: US Department of Health and
Human Services.
U.S. Department of Commerce, National Telecommunications and Information
Administration. (2011). Digital nation: Expanding internet use. Retrived from
http://www.ntia.doc.gov.
Uchino, B. N. (2004). The meaning and measurement of social support social support
and physical health: Understanding the Health Consequences of Relationships
(pp. 9-13). United States: Yale University Press.
van den Berg, M. H., Schoones, J. W., & Vliet Vlieland, T. P. (2007). Internet-based
physical activity interventions: A systematic review of the literature. Journal of
Medical Internet Research, 9 (3), e26. doi: 10.2196/jmir.9.3.e26
van Wier, M. F., Ariëns, G. A., Dekkers, J. C., Hendriksen, I. J., Smid, T., & van
Mechelen, W. (2009). Phone and e-mail counselling are effective for weight
management in an overweight working population: a randomized controlled trial.
BMC Public Health, 9, 6. doi: 10.1186/1471-2458-9-6
Vandelanotte, C., Duncan, M. J., Plotnikoff, R. C., & Mummery, W. K. (2012). Do
participants' preferences for mode of delivery (text, video, or both) influence the
153
effectiveness of a Web-based physical activity intervention? Journal of Medical
Internet Research, 14(1), e37. doi: 10.2196/jmir.1998
Vandelanotte, C., Spathonis, K. M., Eakin, E. G., & Owen, N. (2007). Website-delivered
physical activity interventions a review of the literature. American Journal of
Preventative Medicine, 33, 54-64.
Vázquez, F. L., Otero, P., & Díaz, O. (2012). Psychological distress and related factors in
female college students. Journall of American College Health, 60(3), 219-225.
Volpp, K. G., John, L. K., Troxel, A. B., Norton, L., Fassbender, J., & Loewenstein, G.
(2008). Financial incentive-based approaches for weight loss: A randomized trial.
JAMA, 300(22), 2631-2637.
Wadsworth, D. D., & Hallam, J. S. (2010). Effect of a web site intervention on physical
activity of college females. American Journal of Health Behavior, 34(1), 60-69.
Wanner, M., Martin-Diener, E., Braun-Fahrländer, C., Bauer, G., & Martin, B. W.
(2009). Effectiveness of active-online, an individually tailored physical activity
intervention, in a real-life setting: randomized controlled trial. Journal of Medical
Internet Research, 11(3), e23. doi: 10.2196/jmir.1179
Washburn, R. A., Jacobsen, D. J., Sonko, B. J., Hill, J. O., & Donnelly, J. E. (2003). The
validity of the Stanford Seven-Day Physical Activity Recall in young adults.
Medicine and Science in Sports and Exercise, 35 (8), 1374-1380.
Watson, A., Bickmore, T., Cange, A., Kulshreshtha, A., & Kvedar, J. (2012). An internetbased virtual coach to promote physical activity adherence in overweight adults:
Randomized controlled trial. Journal of Medical Internet Research, 14(1), e1.
doi: 10.2196/jmir.1629
Wilbur, J., McDevitt, J., Wang, E., Dancy, B., Briller, J., Ingram, D., . . . Zenk, S. N.
(2006). Recruitment of African American women to a walking program:
Eligibility, ineligibility, and attrition during screening. Research in Nursing and
Health, 29(3), 176-189.
Wilcox, S., Richter, D. L., Henderson, K. A., Greaney, M. L., & Ainsworth, B. E. (2002).
Perceptions of physical activity and personal barriers and enablers in AfricanAmerican women. Ethnicity and Disease, 12(3), 353-362.
Williams, D. M., Anderson, E. S., & Winett, R. A. (2005). A review of the outcome
expectancy construct in physical activity research. Annals of Behavioral
Medicine, 29(1), 70-79.
Winett, R. A., Anderson, E. S., Wojcik, J. R., Winett, S. G., & Bowden, T. (2007). Guide
to Health: Nutrition and physical activity outcomes of a group-randomized trial
of an Internet-based intervention in churches. Annals of Behavioral Medicine,
33(3), 251-261.
154
Woolf, S. H., Krist, A. H., Johnson, R. E., Wilson, D. B., Rothemich, S. F., Norman, G.
J., & Devers, K. J. (2006). A practice-sponsored web site to help patients pursue
healthy behaviors: An ACORN study. Annals of Family Medicine, 4(2), 148152.
Young, D. R., He, X., Harris, J., & Mabry, I. (2002). Environmental, policy, and cultural
factors related to physical activity in well-educated urban African American
women. Women and Health, 36 (2), 29-41.
155
APPENDIX A
INSTITUTIONAL REVIEW BOARD APPROVAL
156
157
APPENDIX B
DATA COLLECTION INSTRUMENTS
158
Seven Day Physical Activity Recall
159
160
Physical Activity Enjoyment Scale
Please rate how you feel at the moment about physical activity. Below is a list of feelings
with respect to physical activity. For each feeling, please mark the number that best
describes you.
I enjoy it
1
2
3
4
5
6
7
I hate it
I feel bored
1
2
3
4
5
6
7
I feel interested
I dislike it
1
2
3
4
5
6
7
I like it
I find it pleasurable
1
2
3
4
5
6
7
I am very absorbed
in physical activity
1
2
3
4
5
6
7
I find it
unpleasurable
I am not at all
absorbed in physical
activity
It’s no fun at all
1
2
3
4
5
6
7
It’s a lot of fun
I find it energizing
1
2
3
4
5
6
7
I find it tiring
I makes me
depressed
1
2
3
4
5
6
7
It makes me happy
It’s very pleasant
1
2
3
4
5
6
7
It’s very unpleasant
1
2
3
4
5
6
7
I feel bad physically
while doing it
It’s very
invigorating
1
2
3
4
5
6
7
It’s not at all
invigorating
I am very frustrated
by it
1
2
3
4
5
6
7
I am not at all
frustrated by it
It’s very gratifying
1
2
3
4
5
6
7
It’s not at all
gratifying
It’s very exhilarating
1
2
3
4
5
6
7
It’s not at all
exhilarating
It’s not at all
stimulating
It gives me a strong
sense of
accomplishment
1
2
3
4
5
6
7
It’s very stimulating
1
2
3
4
5
6
7
I feel good
physically while
doing it
It does not give me
any sense of
accomplishment
161
It’s very refreshing
I felt as though I
would rather be
doing something
else
1
2
3
4
5
6
7
1
2
3
4
5
6
7
It’s not at all
refreshing
I felt as though there
was nothing else I
would rather be
doing
162
Self-Regulation for Physical Activity
These questions ask about what strategies you have used in the past 3 months to
increase your daily step-count or physical activity.
Use this scale to tell us how often in the past month you did the following:
1
Never
2
Seldom
3
Occasionally
4
Often
In the past month how often did you:
1.
2.
5.
6.
Walk instead of drive when going out for lunch or doing errands?
7.
Find or hire a babysitter so you can increase your daily step-count
or physical activity?
Take short breaks to increase your daily step-count or physical
activity during the day?
Park farther away from school or work to increase your daily stepcount or physical activity?
Get together with someone else to increase your step-count or
physical activity?
4.
8.
9.
10.
How Often
(1-5)
Set aside time each day to increase your daily step-count or
physical activity?
Take the stairs instead of an elevator?
Write down in your calendar each week your plans to increase your
daily step-count or physical activity?
Plan other places to increase your daily step-count or physical
activity if the weather is bad?
Keep track of how many steps you are taking?
3.
5
Repeatedly
163
Outcome Expectations for Exercise
The following are statements about the benefits of exercising (walking, jogging,
swimming, bicycling or any activity where the exertion is as least as hard as these
activities).
State the degree to which you agree or disagree with these statements.
Strongly
Neither Agree
nor Disagree Agree Agree
Strongly
Disagree
Disagree
1. Makes me feel better
physically.
1
2
3
4
5
2. Makes my mood better in
general.
1
2
3
4
5
3. Helps me feel less tired.
1
2
3
4
5
4. Makes my muscles
stronger.
1
2
3
4
5
5. Is an activity I enjoy
doing.
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Exercise…
6. Gives me a sense of
personal accomplishment.
7. Makes me more alert
mentally.
8. Improves my endurance in
performing my daily
activities (personal care,
cooking, shopping, light
cleaning, taking out
garbage).
9. Helps to strengthen my
bones.
164
165
APPENDIX C
COMMIT2FIT WEBSITE SCREENSHOTS
166
Personal Profile Page
167
Exercise Tracker
168
Workout Plans
169
Exercise Database
170
Blogs
171
Message Boards
172