University of Southern California Dissertations and Theses

An exploratory study of physical activity across different adult groups

(USC Thesis Other)

An exploratory study of physical activity across different adult groups

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

AN EXPLORATORY STUDY OF PHYSICAL ACTIVITY ACROSS DIFFERENT
ADULT GROUPS

by
Jian Li

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

May 2010

Copyright 2010 Jian Li

iii
List of Tables
Table 1: Participant Characteristics 16
Table 2: Item Analysis of Knowledge Measure 19

iv
List of Figures
Figure 1. Result of previous structural equation modeling analysis: influence of
knowledge, outcome expectancy on exercise behavior. 4
Figure 2. The base model proposed in the exploratory structural equation
modeling analysis. 14
Figure 3. BMI distributions of different age groups, with Y axis showing number
of participants and embedded numbers showing percentage of participants
in each of three BMI groups. 17
Figure 4. The average daily time spent on different levels of physical activity
across the four groups, with error bars showing the associated standard
deviation. 27
Figure 5. The means of major factors investigated in the current study, with a
common scale of maximum score of 1.0 and standard deviation shown by
the associated error bars. 29
Figure 6. The scatter plot of power vs. sample size for old participant group, with
.05 significant level and sample size of 263. 31
Figure 7. The result of structural model for the young participant group. 32
Figure 8. The result of the structural model analysis for the early-middle-age
group. 34
Figure 9. The result of structural model for the late-middle-aged participant group
(upper) and old participant group (lower). 37

v
Abstract
Research in the area of physical activity has typically focused separately on
internal variables within individuals and environmental variables where the choice of
physical activity happens. The current research examined these influential factors
simultaneously in an integrated framework. A sample of 1552 participants mainly
consisting of students, staff and faculty at a private university in California was recruited.
A wide range of physical activity level was observed and young participants showed
higher level of physical activity than early-middle-age, late-middle-age and old
participants. Exploratory structural equation modeling (SEM) analysis indicated self-
efficacy exerted the strongest impact on physical activity across the four groups of
participants. Social support consistently influenced physical activity through its effect on
self-efficacy. The influence of physical environment got diluted as participants got older.
Other variables including expected benefits from regular physical activity, knowledge
and discounting process did not show a reliable influence on physical activity across the
four groups. The models provided good model fits and explained 31% to 60% of the
variance in physical activity across different groups of participants.

1
Chapter 1: Introduction
A large body of evidence indicates that participation in moderate to vigorous
physical activity contributes to a tremendous amount of health improvements. Major
benefits from regular exercise include but are not limited to decreased risk of a heart
attack and osteoporosis, good management of body weight, increased self-esteem and
mental focus. Despite these well-documented health benefits, the inactive life style is still
a major social problem in the United States. One recent research reports that one-third of
U.S. adults – over 72 million people – were obese in 2005-2006, including 33.3 percent
of men and 35.3 percent of women (Ogden, Carroll, McDowell, & Flegal, 2007). Clearly,
the sedentary lifestyle is likely a major factor contributing to this obesity problem. A
better understanding of what motivates people to keep active is necessary and
tremendously important in practice. The current study aims to investigate what personal,
social and environmental factors may influence one’s physical activity level and how
these factors interact to influence it.
Expected Benefits, Knowledge and Discounting Process
Different theoretical frameworks have been offered to investigate the role of personal,
social and environmental antecedents to physical activity. To the extent that these
antecedents can be successfully identified and manipulated, behavior change can be
brought about more effectively. One important and modifiable personal variable is the
expected benefits from regular physical activities. Similar with other rational behaviors,
the choice of physical activity is a goal-attainment decision. The achievement of expected
physiological and psychological benefits is the basic motivation for one to begin and
maintain an active life style.

2
The theoretical importance of expected benefits has received increasing attention by
current researchers. Two major theories in the physical activity research domain, the
Theory of Planned Behavior (TPB) and the Transtheoretical Model (TTM), have
incorporated outcome expectancy as a major theoretical construct (Ajzen, 1991;
Prochaska & DiClemente, 1983, 1984; Prochaska & Velicer, 1997). In the TPB, attitude
is defined to reflect the positive or negative evaluation of performing a specific behavior
(e.g., favorable/unfavorable, good/bad). It is proposed that attitude will influence
behavior through its effect on an individual’s intention to perform that behavior.
Similarly, decisional balance in TTM represents an individual’s assessment of the
perceived importance of the advantages and disadvantages of performing a behavior, and
it has been has been shown to explain why health behavior change occurs in conjunction
with self-efficacy (Janis & Mann, 1977; Velicer, DiClemente, Prochaska, & Brandenburg,
1985).
One research interest in the current study is what psychological variables can
contribute to the formation of outcome expectancies from physical activity, especially the
expected benefits. Without investigating the precursors of expectations, it becomes
difficult to explore ways to modify “defective” expectations. From a practical perspective,
understanding how changes in one construct might be related to changes in other
constructs can facilitate the selection and development of appropriate interventions
(Courneya & Bobick, 2000). Despite the theoretical importance, the determinant of
outcome expectancy is still ambiguous at best. In the TPB, attitude is hypothesized to be
closely related to behavioral beliefs, which refer to both the perceived advantages and
disadvantages of performing the behavior in question. In the TTM, decision balance was

3
borrowed from decisional balance theory and the theoretical precursor is not clear (Janis
& Mann, 1977). Recently, Courneya and Bobick integrated the TPB with the processes
and stages of change in the TTM (Courneya & Bobick, 2000). They found that 46% of
the variance in attitude could be explained by 4 out of 10 processes of change (i.e.,
counter conditioning, consciousness raising, social liberation, and contingency
management). In spite of the positive relations between attitude and some processes of
change, however, the convincing proximal determinant of attitude still remains unclear.
One potential precursor of the outcome expectancy is physical-activity-related
knowledge. Knowledge allows people to develop realistic expectations of the personal
value of their physical activities and is regarded as one of the main factors stimulating
physical activity. In a study about attitudes and behavior relevant to exercise adoption
among older Australian women, Lee found that stages of change identified by the TTM
model were significantly associated with an individual’s exercise knowledge (Lee, 1993).
Both knowledge and attitudes could be important targets of intervention with middle-
aged women. Similarly, Jeffery and Wing demonstrated that successful weight loss was
related to an increase in nutrition knowledge and a decrease in perceived barriers to
adherence (Jeffery & Wing, 1995). Consistent with these finding, our previous study
found that knowledge could significant predict outcome expectancy which mainly
measured the perceived benefits associated with regular physical activity (Li, Trujillo,
Brougham, Brown, & Walsh, 2009). It was further found that knowledge was a moderate
predictor of physical activity (see Figure 1, standardized coefficient β = .22), compared
with the stronger predictive power of outcome expectancy (standardized coefficient β
= .25). Results supported the hypothesis that knowledge might be an important precursor

4
of expected benefits and suggested that its direct influence on physical activity was
limited and its indirect influence might heavily depend on other mediation factors.

Figure 1. Result of previous structural equation modeling analysis: influence of
knowledge, outcome expectancy on exercise behavior.
*p < .05

Although a number of studies have demonstrated the positive relation between
knowledge and physical activity, opposite results have also been reported by some other
studies. In a survey to determine the effectiveness of a university fitness for living course
on its students’ wellness content knowledge, Downing, Masterson and Gray found no
association between wellness knowledge and the exercise behaviors measured (Downing,
Masterson, & Gray, 2005). Rimal found that the strongest predictors of subsequent
exercise behavior was its prior value, rather than knowledge, and the results suggested
that improving self-efficacy may be a more promising strategy for health promotion
(Rimal, 2001).
Thus, the direct influence of knowledge on physical activity is still unclear.
Knowledge alone has not been consistently found to directly change behavior (Morrow,
Krzewinski-Malone, Jackson, Bungum, & Fitzgerald, 2004). Motivation factors, such as
attitudes and outcome expectancies, could be important intervening variables between
knowledge and physical activity. One goal of the present research was to examine
knowledge as a precursor to expected benefits from physical activity. We hypothesized

5
that knowledge played a key role in the formation of outcome expectancy towards
volitional behavior; and in turn, outcome expectancy mediated the association between
knowledge and physical activity.
In addition to knowledge, the effectiveness of outcome expectancy as a motivating
factor in physical activity adherence can also be influenced by other factors. Outcome
expectancy, mostly the perceived benefits from regular physical activity, must be
sufficiently important to the individual to motivate them to exercise regularly. However,
a number of benefits can only be attained if physical activity is maintained over a
prolonged period of time. Because of this temporal delay, the subjective values of these
benefits are usually devalued as described by the phenomenon known as delay
discounting. This perception of devalued benefits may result in an unmotivated attitude
towards physical activity, and finally a lower likelihood of engaging in the health
behaviors. Additionally, a certain amount of uncertainty is often involved in the same
process. That is, delayed outcomes are usually perceived as uncertain or as probabilistic
ones. Probabilistic discounting, which refers to the fact that the present subjective value
of an outcome (e.g., money, benefits and rewards) decreases as the odds against its
receipt increases, may also contribute to the devaluation of the expected benefits of
physical activity (Myerson, Green, Hanson, Holt, & Estle, 2003). In reality, it is often
difficult to distinguish the impacts of delay and probabilistic discounting (Gretchen B.
Chapman, 2005). For example, a person may devalue the health benefits of regular
physical activity not only because most of the benefits are achieved in the future but also
because one is not sure that he or she will be able to maintain regular physical activity

6
long enough to attain these benefits. Thus, the discounting process might be an
influential factor regarding the expected benefits from physical activity.
Despite the close theoretical relationship between the two types of discounting and
exercise adherence, little research has addressed this connection in detail and no
integrated theoretical framework has been fully established (Levy, Micco, Putt, &
Armstrong, 2006). A significant amount of effort has been spent on addictive behaviors
while little research has examined the degree to which discounting can influence physical
activity (G. B. Chapman, 1998). The lack of relevant studies in the discounting literature
motivated me to conduct a meta-analysis in 2007: 26 independent studies met the
inclusion criteria for this meta-analysis and 31 effect sizes were calculated (Li, 2008). For
studies of non-addictive behaviors such as medical compliance and physical activity, the
estimated mean effect size was .04 (ns). In contrast, addiction studies (e.g., alcohol abuse
and smoking) showed an average effect size of .24 (p < .001). Results demonstrate that
discounting has a moderate impact on addictive behaviors but does not have a significant
influence on non-addictive behaviors. It should be noted that compared with addiction
studies, fewer studies have addressed the impact of discounting on non-addictive
behaviors. It is still not clear whether the scarcity of relevant studies contributes to this
insignificant finding in non-addictive behaviors.
Thus, the present research also aimed to explore the impacts of discounting on one of
the most typical non-addictive health behavior— physical activity —in a general
theoretical framework. We hypothesized that discounting might influence physical
activity directly, and also indirectly through expected benefits. It was hoped that by
investigating outcome expectancy and its closely related antecedent of discounting, it

7
would allow for the identification of significant variables and mechanisms in the
individual’s physical activity decision-making process, as well as fundamental
understanding of the impact of discounting processes on physical activity.
Self-efficacy, Physical, and Social Environment
In addition to outcome expectancy, another important and modifiable personal
variable regarding physical activity adherence is self-efficacy. Self-efficacy refers to the
perceived confidence about one’s ability to execute the behavior (Bandura, 1997;
Strecher, DeVellis, Becker, & Rosenstock, 1986). This perceived confidence about
performing a behavior highly associates with the actual ability to perform that behavior.
Empirical studies suggest that self-efficacy consistently relates to the initiation and
maintenance of a wide range of health behaviors, including physical activity (e.g.,
Bandura, 1997; Edward McAuley, Elavsky, Jerome, Konopack, & Marquez, 2005; Sacco,
et al., 2005; Shields & Brawley, 2006). Some researchers even point out that self-efficacy
sometimes is superior to past performance in predicting future behavior (Marcus & Owen,
1992).
Self-efficacy has close theoretical links with outcome expectancy. In Bandura’s
social-cognitive theory, behavior change and maintenance are a function of two factors,
namely the outcome expectancy of what will result from one’s engaging in a behavior,
and self-efficacy which reflects the expectations about one’s ability to execute the
behavior. Despite the close theoretical relationship between the two variables, self-
efficacy has not been studied extensively in conjunction with outcome expectancy.
Various demographic, environmental and intrapersonal variables intertwine to contribute
to self-efficacy. These variables may also influence outcome expectancy and its

8
determinants. For example, repeated encouragement to physical activity from family
members and friends may lead one to realize the potential health benefits from regular
physical activity and also help one to build up confidence to maintain this behavior. The
relative predictive power of self-efficacy versus outcome expectancy is likely to depend
on the characteristic of the situation in which the behavior occurs. However, a large
amount of physical activity research has only placed emphasis on identifying individual
determinants and failed to consider the influential physical and social context within
which the behavior takes place (Giles-Corti & Donovan, 2002).
In addition, one strength of the self-efficacy framework is that self-efficacy can be
enhanced by interventions and this enhancement is related to subsequent health behavior
change. Specifically, health behavior such as physical activity involves a rational
decision-making process, and also an execution stage which is under the individual’s
volitional control. The execution process requires one has all the resources to provide the
necessary physical facilities, environmental convenience and social support to overcome
different physical activity barriers. Environments rich in resources relevant for physical
activity, such as sidewalks, parks, exercise classes, and encouragement from family
members and friends may make it easier for people to be physically active. The support
from physical and social environments may provide more opportunities for incidental
physical activity, and also encourage the adherence to regular physical activity. A focus
on broader determinants of health behavior with an emphasis on this interaction between
the individual and the social and physical environment will greatly improve our
understanding of the potential intervention methods to manipulate psychological
variables such as self-efficacy, and ultimately to improve people’s health behavior.

9
In recent years, there has been a growing interest in the interaction between
individual and environmental factors such as social, political, cultural and physical
environments. Developments in public physical activity promotion have shifted from
changing individual knowledge, attitudes and skills, to changing social and physical
environmental factors (Phongsavan, McLean, & Bauman, 2007). However, the influence
of environmental attributes is among the least understood of the known influences on
physical activity. Compared with other well-studied psychological variables, the
conceptualization and measurement of physical and social environment comprise a
relatively new area of research. Despite the difficulties in measurement, pilot studies have
consistently shown that physical and social environment factors have stable associations
with physical activity. In general, physical environment correlates of physical activity
include the accessibility, convenience and aesthetic attributes of sidewalks, trails and
recreation facilities. Social environment correlates include community and family support
such as regular participation of friend and family and frequent observation of others
exercising.
Environmental disadvantages such as lack of hills in one’s neighborhood, absence of
enjoyable scenery are often found to be associated with sedentary behavior (Brownson,
Baker, Housemann, Brennan, & Bacak, 2001; King, et al., 2000). Interestingly, the
presence of sidewalks has been found to be a reliable predictor of active lifestyle. For
example, in a telephone interview of 917 African-American women living in two
counties in South Carolina, Ainsworth and colleagues found the presence of sidewalks or
lighter traffic in the neighborhood was significantly related to engaging in sufficient
physical activity (Ainsworth, Wilcox, Thompson, Richter, & Henderson, 2003). Similarly,

10
Jago and colleagues found that sidewalk characteristics, such as sidewalk location,
sidewalk material, presence of streetlights, and number and height of trees, was positively
associated with light-intensity physical activity among male adolescents (Jago,
Baranowski, Zakeri, & Harris, 2005). The presence of trails is also found to be associated
with moderate to strenuous levels of physical activities. One recent study examined the
environmental supports for physical activity in active and inactive adults based on
national recommendations for physical activity and walking (Wilson, Ainsworth, &
Bowles, 2007). Trusting neighborhoods having recreational facilities present, and having
trails available for use were each associated with a 50% reduction in the odds of being
overweight. Using trails was also associated with a large reduction in the odds of being
obese among participants who were not regular walkers.
In addition to sidewalks and trails, the presence of convenient recreational facilities
has also been found to be associated with moderate to strenuous levels of physical
activities. In a randomly selected sample of 449 Australian adults age 60 and older, Booth
and colleagues found self-reported physical activity was significantly associated with
access to local facilities (Booth, Owen, Bauman, Clavisi, & Leslie, 2000). Similarly,
Mowen and colleagues found that perceived proximity of a park was related to physical
activity and health through park use frequency (Mowen, Orsega-Smith, Payne, Ainsworth,
& Godbey, 2007). Additionally, Tester and Baker examined visitation and physical
activity levels in two San Francisco parks in low-income neighborhoods that underwent
field renovations (Tester & Baker, 2009). Both park improvement projects saw significant
increases in male and female visitors. For both genders, there was a significant increase
in sedentary, moderately active, and vigorously active visitors to the improved parks.

11
Physical environments that have undesirable characteristics, such as bad weather or
heavy traffic, may act to reduce the probability that residents will be physically active.
Individuals still have the opportunity to directly change these environmental
disadvantages by purchasing necessary exercise equipment or moving to an area with
more resources. However, social environmental disadvantages such as sedentary friends
and lack of support from family may be difficult to modify or change. These social
environment factors may have greater capacity to hider physical activity. The influence
of others is so powerful that empirical studies found even the infrequent observation of
others exercising in one’s neighborhood is significantly related to sedentary lifestyle
(King, et al., 2000). Similarly, in a longitudinal study, Sallis and colleagues used baseline
perceived environmental variables to predict physical activity at 6 months (Sallis, King,
Sirard, & Albright, 2007). Results showed that men reporting frequently seeing people
being active in their neighborhoods did 50–75 more minutes of physical activity per week
than did those with different environmental characteristics.
Perhaps the most important social environmental characteristic is the support from
family members and friends. The positive association between social support and
exercise behavior has been observed across different age groups. One extensively-studied
group is adolescents. Empirical studies indicate parental and peer support has a strong
influence on the adolescent’s physical activity. For example, in a large survey study with
a sample of 12 –18 year old high school students (n = 3,471) from low SES schools
within South Auckland and New Zealand, Hohepa and colleagues found that low parental
support and low peer support were associated with reduced odds of being regularly active
after school (Hohepa, Scragg, Schofield, Kolt, & Schaaf, 2007). In a secondary analysis

12
of 718 sixth-grade girls between the ages of 10 to 14, Springer and colleagues found that
friend physical activity participation and encouragement from friend and family were
positively related to moderate-to-strenuous physical activity, and friend encouragement
was the only variable positively related to strenuous physical activity (Springer, Kelder,
& Hoelscher, 2006). It was suggested that effective support should involve direct
participation from parents in physical activities with their adolescents.
Social support has also been found to be an influential factor on adults and older
people’s physical activity. Regular participation of friend and family and support from
friends were found to increase leisure and recreational physical activity participation
among older adults (Booth, et al., 2000; Sasidharan, Payne, Orsega-Smith, & Godbey,
2006). In a cross-sectional study on US adults, several social factors were identified to be
associated with physical activity, including surroundings in which many people were
exercising, friends who encouraged exercise, and having at least one friend with whom to
exercise (Brownson, et al., 2001).
Compared with physical environment’s direct impact on an individual’s physical
activity, social support seems to largely influence physical activity through the
enhancement in psychological variables such as self-efficacy. Increasing evidence
showed that self-efficacy was an important psychological construct that mediated the
relation between changes in physical activity and perceived social support. For example,
for older people, a group exercise program was found to increase self-efficacy which
resulted in more positive and less negative feeling states when contrasted with a
condition in which the participants exercised alone (E McAuley, Blissmer, Katula, &
Duncan, 2000) . In a study of the formation of group cohesion and social support in

13
exercise classes among former sedentary adults, Christensen and colleagues found the
mutual support in a cohesive group facilitated development of self-efficacy beliefs among
the participants, and improved their mastery expectation regarding exercise activity
(Christensen, Schmidt, Budtz-Jorgensen, & Avlund, 2006). These authors suggested that
using group dynamics to manipulate mutual support may be a promising intervention tool
in the promotion of leisure-time physical activity. Similarly, in a study of cardiac
rehabilitation maintenance exercise programs, participants higher in social support
reported significantly greater self-efficacy than did their moderate social support
counterparts (Woodgate, Brawley, & Shields, 2007).
Statement of Research Purpose
Although empirical studies have identified a series of important psychological and
environmental variables as predictors of physical activity, many are limited by somewhat
restricted models. Researchers have seldom examined these influential variables in an
integrated framework. It is not clear how these variables interact to influence physical
activity. The purpose of the present study is to examine these determinants of physical
activity in a more complete cognitive-environmental model across different age groups. It
is hoped that this comparison of different aged participants can provide us a better
understanding of the complex roles of the potential determinants of physical activity
adherence.
Data was collected from 1552 participants in which diverse socio-economic
background and a wide range of physical activity was observed. It was hypothesized that
both expected benefits of physical activity and self-efficacy were strong predictors of
physical activity (see Figure 2). It was further hypothesized that greater knowledge and

14
less discounting rate would be separate predictors of more expected benefits. It was also
predicted that higher levels of social support and physical environmental support would
lead to a higher levels of self-efficacy. This general hypothesis has been partially
supported by our pilot study (see Figure 1). It is also in accordance with previous
research on self-efficacy (e.g., Bandura, 1997; Marcus & Owen, 1992; Edward McAuley,
et al., 2005; Sacco, et al., 2005; Shields & Brawley, 2006). In addition, health status was
expected to directly influence physical activity in this framework.

Figure 2. The base model proposed in the exploratory structural equation modeling
analysis.
It should be noted that this study is largely exploratory. The relationship among these
variables within the model can be easily influenced by other demographic variables such
as socio-economic status which was not included in the model. To fully acknowledge the
dynamic nature, the analyses presented here evaluated this hypothetical model across four
different groups of participants, namely young, early-middle-age, late-middle-age and old

15
groups. It is hoped that this exploratory analysis can highlight the complex environment
in which exercise choices happen, and detect important mechanisms that determine
exercise behavior in different groups.

16
Chapter 2: Method
Participants
The majority of participants were undergraduate students, faculty and staff of
USC. Online survey system Qualtrics ® was used as the survey platform. Web records
showed that there were 2,411 responses to our survey invitation and 1,552 participants
finished the survey. Participants with completed data consisted of 966 women and 586
men. The mean age for the participants was 42.9 years (SD = 16.1 years). The mean BMI
for the participants was 30.2 (SD = 6.3). Fifty-four percent are Caucasians, 86 percent
had at least a 2-year college degree and 68 percent had an annual family income above
$60,000. Complete participant characteristics are included in Table 1 and detailed
descriptions of BMI distribution for different age groups are presented in Figure 3.
Table 1: Participant Characteristics
Patients (n=1552)
Race/Ethnicity
Caucasian 53.7%
Asian 18.8%
Hispanic 16.0%
African American 6.1%
Other 5.4%
Education
Less than High School 0.2%
High School Diploma 11.3%
2-year College Degree 8.6%
4-year College Degree 31.6%
Master’s Degree 22.4%
Doctoral Degree 14.4%
Professional Degree (JD,MD) 8.8%
Marital Status
Single 37.2%
Married/Living as Married 48.7%
Divorced/Separated 10.6%
Widowed 1.6%
Employment Status
Full-time/Part-time Student 16.0%
Full-time/Part-time Employee 76.3%
Retired 5.5%

17

Figure 3. BMI distributions of different age groups, with Y axis showing number of
participants and embedded numbers showing percentage of participants in each of
three BMI groups.
Measures
Measures used in the current study were refined from our previous study or adapted
from published studies in the literature. Latent variables were developed based on these
measures and incorporated into the subsequent exploratory SEM analysis.
Physical Activity Measure
Three items were refined to assess participants’ physical activities from a previous
published study (Audrain-McGovern, et al., 2004). Participants were asked to report the
number of days over the previous 2 weeks that they engaged in strenuous, moderate, and
light physical activities. Strenuous physical activities were defined as “hard physical

18
activities that gets your heart rate to 85% or more of your maximum and results in
extremely heavy breathing, sweating and requires extreme exertion that prevents you
from talking.” Examples were given such as “playing full-court basketball at full effort,
running, swimming, or cycling at a very fast and intense pace.” Moderate physical
activities were defined as “Physical activity that gets your heart rate between 65% to 85%
of your maximum and results in heavy breathing and sweating but does not keep you
from talking” and listed examples included “playing full-court basketball at a slow pace,
running, swimming, or cycling at a moderately intense pace.” Light physical activities
were “physical activity that gets you moving but does little to increase your breathing or
heart rate” which was listed as “walking, golf, bowling, gardening, house cleaning, etc.”
The duration of physical activities was assessed by asking participants to indicate the
average number of minutes per day they spent on those days for which they listed
activities. Based on the intensity and duration of the activities, separate estimated calorie
expenditures were calculated (Gellish, et al., 2007; Keytel, et al., 2005) .
Knowledge Measures
The knowledge measure has been adapted from our previous study and updated with
modified knowledge testing questions from current exercise textbooks (e.g., Nieman,
2007; Trujillo, Brougham, & Walsh, 2004). This final version of the knowledge measure
consisted of 15 multiple-choice items and 20 yes-no items. The multiple-choice items
were designed to assess participants’ knowledge of how to exercise appropriately. The
yes-no items were developed to test participants’ knowledge of health benefits from
regular physical activity. In this measure, participants were requested to decide whether
regular exercise could prevent/cure different diseases or health problems.

19
Table 2: Item Analysis of Knowledge Measure
Item Difficulty Point Biserial Correlation
Item 1 .58 .44**
Item 2 .65 .34**
Item 3 .18 .00
Item 4 .67 .25**
Item 5 .40 .28**
Item 6 .32 .40**
Item 7 .52 .27**
Item 8 .16 .33**
Item 9 .59 .46**
Item 10 .86 .37**
Item 11 .26 .42**
Item 12 .50 .22**
Item 13 .62 .39**
Item 14 .16 .29**
Item 15 .33 .32**
Item 16 .50 .28**
Item 17 .81 .25**
Item 18 .22 .09**
Item 19 .92 .31**
Item 20 .23 .09**
Item 21 .83 .28**
Item 22 .82 .31**
Item 23 .90 .32**
Item 24 .80 .25**
Item 25 .83 .27**
Item 26 .74 .39**
Item 27 .74 .29**
Item 28 .88 .31**
Item 29 .48 .19**
Item 30 .58 .36**
Item 31 .84 .34**
Item 32 .26 .06**
Item 33 .95 .28**
Item 34 .55 .32 **
Item 35 .09 -.08**
Note. Multiple-choice items included items 1-15 and Yes-No items included items 16-35. Detailed
descriptions of each item were shown in Appendix E.
** p < .001.

Item difficulty and item discrimination indexes were examined and presented in
Table 2, and detailed descriptions of items are shown in Appendix E. The mean score of
each item was used as the item difficulty index. Possible difficulty ranged from zero to

20
one, with zero indicating no participants answer it correctly and one indicating all
participants answered the item correctly. The actual item difficulties of the knowledge
items ranged from .09 to .94, which indicated a wide range of difficulty had been covered
in the knowledge measure. The point biserial correlation, the correlation coefficient
between each item and the total score, was calculated as the item discrimination index.
Point biserial correlations of items ranged from -.08 to .46. Researchers suggest that
items with point biserial correlation values below .20 should be considered poor and
should be revised or eliminated, and items with values above .30 are good and no
revisions are necessary (Crocker & Algina, 1986). However, in consideration of the
complicated nature of an individual’s knowledge itself and the unknown mechanism of
knowledge’s influence on physical activities, the current data analysis reserved all the
items from our knowledge measure. This decision was consistent with Nunnally and
Bernstein’s recommendation that the final decision to retain or delete an item should not
be based on statistical factors alone and must rest on the judgment of the test developers
(Nunnally & Bernstein, 1994). Two total scores were calculated for multiple-choice items
and yes-no items separately and used in the further analysis.
Expected Benefits Measures
Similar to the knowledge measure, the expected benefits measure has been adapted
from our previous study (Trujillo, et al., 2004). Participants were asked to evaluate 11
consequences associated with exercise. These consequences were categorized as four
groups including improvements in physical capacity, physical health, psychological
capacity and social/interpersonal relations (e.g., “My heart would be stronger and more
efficient, providing better circulation for my whole body”, “I would have more

21
opportunities to make and be with friends”). For each consequence, participants indicated
the importance of this consequence, how pleasant the consequence was, how likely this
consequence would be achieved on 7-point Likert scale. Expected utility scores were
computed by multiplying these ratings together. The perceived benefit measure exhibited
good inter-item correlation (Cronbach’s α = .92). Four separate composite scores were
computed by adding up the expected utility scores across each group of items and used in
the subsequent exploratory SEM analysis.
Discounting Measures
The delay and probabilistic discounting measure was adapted and refined from a
published study (Rosado, Sigmon, Jones, & Stitzer, 2005). Participants were asked to
imagine possessing a winning lottery ticket. In the delay discounting measure, the prize
could be collected after some delay or the ticket could be sold immediately to a lottery
agent for a smaller sum of money. Similarly, in the probabilistic discounting measure, the
prize could be collected with some probability or the ticket could be sold immediately for
a smaller sum of cash. Different levels of delays, winning probabilities and selling prices
of the lottery tickets were presented. Participants’ preferences were monitored in these
different situations. Most importantly, we recorded the crossover points or indifferent
points at which a participant’s choices switched from keeping the lottery ticket to taking
the immediate cash. A series of these indifferent points were used to estimate the
discounting rate for each participant based on an area-under-the-curve method (Myerson,
Green, & Warusawitharana, 2001).

22
Self-Efficacy Measure
Self-efficacy was assessed with six items modified from a measure developed to
assess people’s confidence in their ability to refrain from smoking in different situations
(Baldwin, et al., 2006). Example items included “How confident are you that you could
keep regularly exercising when your family members or friends have a sedentary
lifestyle?” and “How confident are you that you would be able to exercises regularly even
if exercising takes you a lot of time?” Responses were measured on a scale ranging from
0 (not at all confident) to 6 (extremely confident). In addition to the six items,
participants also finally indicated their overall confidence level that they would able to
exercise regularly. Confirmatory factor analysis identified a one-factor structure in the
self-efficacy instrument and good inter-item correlation was achieved (Cronbach’s α
= .92).
Social-Support Measure
Social support for exercise was assessed by four items. These items inquired about
people close to you (e.g., parents, siblings and friends) and asked: “They listen to your
concerns about regular exercise,”“They agree with your decisions about exercising
regularly,” “They encourage choices favorable to your exercise,” and “They assist with
your regular exercise.” Responses were given on a 7-point scale ranging from 0 (disagree
completely) to 6 (agree completely). Participants also indicated their overall belief that
they receive significant social support for their exercise activities. Confirmatory factor
analysis identified a one-factor structure in the self-efficacy instrument and good inter-
item correlation was achieved (Cronbach’s α = .85).

23
Environment Friendliness Measure
Similar to the social support measure, the environment friendliness measure assessed
different types of physical support provided by the environment. Example items included
“My family has exercise equipment which I can use very easily,” “The college/university
where I study now provides convenient exercise facilities,” “My college/university
provides a series of useful courses and guidance on exercise,” and “The neighborhood
where I now live has convenient exercise facilities.” Responses were on a 7-point scale
ranging from 0 (disagree completely) to 6 (agree completely). Participants also indicated
in general whether their living environments provide enough support for exercise
activities. One-factor structure was confirmed and moderate internal consistency was
achieved (Cronbach’s α = .63).
Perceived Health Measure
Perceived health was assessed by six items, including “I am physically able to run,
cycle, swim, etc.”, “I am physically able to lift weights”, “My heart is strong and healthy”,
“My skeletal system is strong and healthy (e.g., back, knees and joints).”, “My kidney
and liver function is strong and healthy.” and “My lungs are strong and healthy.”
Responses were given on a 7-point scale ranging from 0 (completely untrue) to 6
(completely true). Participants also indicated their overall belief that they receive
significant social support for their exercise activities. One-factor structure was confirmed
and the internal consistency was high (Cronbach’s α = .90).
Procedure
Undergraduate students who took an introductory statistics class were recruited
for class credits, and they were also requested to invite one or two of their friends to

24
participate this survey. Faculty and staff of USC, and volunteers from a Gerontology
research subject pool were invited by emails to participate in this survey for a chance at
winning a $50 lottery. A link in the email advertising this study took participants to an
information sheet that explained the research, as well as their rights as research
participants. Then participants were taken to the demographic questionnaire, followed by
exercise behavior, knowledge, self-efficacy, environmental friendliness, social support
and discounting measures. Finally, participants were taken to a separate page to input
their emails addresses for the lottery drawing. Online survey system Qualtrics ® was
used to present all the measures and collect participants’ responses. All materials took
approximately 35 to 55 minutes to complete. Although Qualtrics ® system allowed
participants to complete the survey in several sittings, web records showed that almost all
participants complete the survey in one sitting.
Data Analysis
Participants were divided into four groups: the young group included participants 25
years old or younger, the early-middle-age group included participants between 26 and
45 years old, the late-middle-age group included participants between 46 and 59 years old
and the old group consisted of people 60 years old or older. An exploratory structural
equation modeling (SEM) analysis was used to investigate the psychological and
environmental variables’ influence on physical activity separately for the four groups.
Maximum likelihood estimation and the covariance matrix were used for all analyses.
Before the exploratory structural equation modeling analysis, data were examined for
normality. With participants aged from 17 to 92 years old, a wide range of physical
activity levels was observed. Not surprisingly, the estimated calorie expenditure was

25
heavily positively skewed. A square root transformation was conducted and resulted in a
more normally distributed measure. Three model fit indexes were computed to assess the
overall fit of the models: the χ
2
statistic, Bentler’s Comparative Fit Index (CFI) and the
Root Mean-square Error of Approximation (RMSEA). As sample sizes varied from 237
to 681, the χ
2
statistics were significant in any model. A cut-off value of .95 or more for
CFI and .05 or less for RMSEA were regarded as indicators for good model fit.

26
Chapter 3: Results
Descriptive Statistics
The average time spent on different levels of physical activity across the four age
groups of participants is shown in Figure 4. All four groups reported low level of
strenuous, moderate and light physical activities. As participants got older, they reported
less physical activities. However, the differences across different age groups were small.
In general, nearly 60% participants reported virtually no strenuous physical activities (< 5
min/day) and more than 50% of participants reported lack of moderate physical activities
(<10 min/day). More than 70% of participants did not meet the recommended 150
minutes a week of moderate-intensity or 75 minutes a week of vigorous-intensity aerobic
physical activity (USDHHS, 2008) . On the other hand, 10% of the sample reported 30-
plus minutes of strenuous physical activity per day, and 20% reported 30-plus minutes of
moderate physical activity per day. In addition, 50% of participants reported at least 30
min of light physical activity per day.
Univariate analyses including t-test and ANOVA with significant level of .05 were
used to examine the relations among demographic and physical activity variables. Male
participants spent 31 more minutes per week on strenuous physical activity than females,
and 31.5 more minutes per week on moderate physical activities. No gender difference
was observed for the light physical activities. Post hoc test indicated that Caucasian
participants spent 31 more minutes per week on moderate physical activities than Asian
American participants. No significant relationship among any other ethnicity groups and
any levels of physical activity was found. Annual family income was not significantly
related to level of physical activity. Age of participants was found to be an influential

27
factor regarding all levels of physical activities. Young participants engaged in
significantly more minutes of all levels of physical activities than the other three age
groups while the differences among the other three groups were not significant.

Figure 4. The average daily time spent on different levels of physical activity across
the four groups, with error bars showing the associated standard deviation.
Figure 5 presents the four groups of participants’ average responses to the following
measures: exercise knowledge, perceived benefits, delay and probabilistic discounting
and other social/environmental measures used in this study. For each variable, the
response scale was transformed to a common full scale maximum of 1.0 to provide easy
comparisons and interpretation. T-test and ANOVA with significant level of .05 were
applied to investigate the potential group differences among these cognitive and
environmental variables. Female and male participants showed similar responses to all of
these measures. Given the large sample size, statistically significant gender differences
were observed in several measures. However, the absolute differences were small and
provided little meaningful interpretation (e.g., a 1.5 point difference between the male
0
10
20
30
40
50
60
70
80
90
100
Young Participants Early‐middle‐age
Participants
Late‐middle‐age
Participants
Old Participants
Average Daily Physical Activities Duration
min/day
Strenuous Physical Activities Moderate Physical Activities Light Physical Activities

28
and female participants in the self-efficacy measure with a full score of 42 points).
Similarly, participants in different age groups responded similarly to most of the
measures. Small but statistically significant age group differences were usually found but
provided little meaningful information. For example, on the self-efficacy scale,
significant group differences were identified among the young, early-middle-age, late-
middle-age and old participants (Ms =27.4, 26.7, 29.2 and 29.1, SDs = 7.8, 8.2, 8.6 and
9.2), F(3, 1547) =9.52. p < .001. Thus, the following discussion focuses on the average
results across groups.
In the exercise knowledge measure, the average reported level of exercise knowledge
across the four groups was .57, which suggested that participants understood some basic
knowledge about how to exercise appropriately and what health improvement could be
gained by regular exercise. However, it also indicated that participants’ relevant
knowledge was limited and they might lack more sophisticated exercise-related
knowledge.
In the delay discounting measure, a maximum score of 1.0 indicates participants
strongly prefer the delayed but big reward compared with the immediate but small reward.
Similarly, a maximum score of 1.0 in the probabilistic discounting measure shows that
participants prefer uncertain but big reward compared with small but certain reward.
Participants’ mean response to delay discounting was .77, which suggested that the
majority of participants could prefer to wait for a larger future reward. However, the
mean score of .58 for probabilistic discounting indicated that participants had limited
interested in the uncertain but large rewards, and they preferred to the safe but small
rewards to some extent.

29

Figure 5. The means of major factors investigated in the current study, with a
common scale of maximum score of 1.0 and standard deviation shown by the
associated error bars.

Additionally, the majority of participants reported themselves to be in good to
excellent health. They also perceived that they had received a moderate to high level of
support and encouragement from their family members and friends to exercise (i.e., a
mean score of .71). However, participants’ average rating of the perceived benefits of
exercise was low to moderate, indicated that they had a positive but not strong
expectation about the different psychological and physical health improvements that
might result from regular exercise (i.e., a mean score of .60). In addition, the current
0
0.2
0.4
0.6
0.8
1
Exercise Knowledge Delay Discounting
Rate
Probabilistic
Discounting Rate
Perceived Health
Status
Young Participants Early‐middle‐age Participants
Late‐middle‐age Participants Old Participants
0
0.2
0.4
0.6
0.8
1
Self‐efficacy Social Support Environmental
Friendliness
Perceived Benefits

30
sample had a low to moderate rating of the support from the physical environment for
regular exercise (i.e., a mean score of .55). Participants also showed a low to moderate
rating of their self-efficacy in maintaining a regular exercise program (i.e., a mean score
of .66), which was in accordance with the overall low level of physical activity observed
in the current sample.
Exploratory Structural Equation Modeling Analysis
Before analyzing the structural model, measurement models of each latent variable
were examined separately. The fit of the measurement models were satisfactory for all
latent variables (i.e., RESEA smaller than or close to .05). Although the Lagrange
Multiplier test indicated additional correlations between errors of indicators could
significantly improve the model fit, no adjustments were conducted until the subsequent
evaluation of the structural model.
Once acceptable measurement models were established, all latent variables and their
indicators were evaluated in a single base-line model across the four age groups (see
Figure 2). Post hoc model modifications were performed in order to develop a better and
more parsimonious model. On the basis of the Lagrange Multiplier test, the Wald test and
theoretical relevance, causal relations were updated and necessary correlations between
errors were added. The final structural models with standardized indirect and direct effect
coefficients, squared multiple correlations (R
2
) and model fit indexes are presented in
Figure 7-9 for the four groups separately. The direct effect refers to a variable’s exclusive
influence on a specific dependent variable. In contrast, the indirect effect is its influence
through other variables in the model. The direct effect and indirect effect together
consisted of a variable’s total effect. The squared multiple correlations (R
2
) indicate the

31
variance of each dependent variable explained by other variables in the model. More
detailed full model results are shown in Appendix B, C, D and E.
In addition, separate power analysis were conducted for the four age groups
(MacCallum, Browne, & Sugawara, 1996). Results showed all four models achieved
excellent power level (e.g., power >=.99). As the power of a structural equation model
increases with both model degree freedom (df) and sample size (N), the result of old
participant group, which had the smallest sample size and degree freedom, was shown in
the following Figure 6.

Figure 6. The scatter plot of power vs. sample size for old participant group, with
.05 significant level and sample size of 263.
Figure 7 presents the result for participants who were 25 years old or younger.
Within the model, exercise self-efficacy exerted the strongest total effect on physical
90 100 110 120 130 140 150 160 170 180 190 200 210
Group Sample Size (N)
.88
.90
.92
.94
.96
.98
1.00
Power

32
activity ( β
total
= .65); greater self-efficacy was associated with higher levels of physical
activity. This strong influence was independent of any other environmental or
psychological variable.

Figure 7. The result of structural model for the young participant group.

Environmental support also influenced physical activity ( β
total
= .31); more support
from the physical environment were significantly associated with a higher level of
physical activity. Specifically, the effect of the physical environment on the physical
activity levels of participants was largely direct ( β
direct
= .21), and also indirect through its
effects on self-efficacy ( β
indirect
= .10). That is, as shown in Figure 7, the latent variable
“Environment Friendliness” which measured the physical environment’s influence had a
direct effect of .21 on “Physical Activities”. The indirect effect for “Environment
Friendliness” was .10, which was the product of the coefficients of its direct effect on
“Self-Efficacy” (.15) and the direct effect of “Self-Efficacy” on “Physical Activities”
(.65). Thus, the latent variable “Environment Friendliness” had a total effect of .31 on
“Physical Activities”, which was the sum of its direct and indirect effects.

33
Social support also positively influenced physical activity ( β
total
= .39); more support
from a participant’s family member and friends were significantly associated with a
higher level of physical activity. The effect of social support from family members and
friends on physical activity was indirect through its influence on self-efficacy and no
direct effect was observed. That is, its total effect .39 was the product of the coefficients
of its direct effect on “Self-Efficacy” (.60) and the direct effect of “Self-Efficacy” on
“Physical Activities” (.65). Thus, receiving more social support was associated with
higher levels of exercise self-efficacy, and this enhancement finally contributed to a
higher level of physical activities.
Knowledge was found to be negatively associate with a participant’s physical
activity ( β
total
= -.30). A participant who had more elaborate knowledge about the
recommended scientific way to exercise and who knew more about what kind of diseases
or health problems could be prevented or cured by regular physical activity, was found to
report lower levels of physical activity. Perceived health status was found to be
associated with social support and expected benefits, but not physical activity. A
participant who received more encouragement and support from family members and
friends to exercise, was more likely to have a high level of expectation about the health
improvements associated with regular physical activity, and also reported a better health
status. However, they were no more likely to exercise than other participants.
Figure 8 displays the result for early-middle-aged participants between 26 years old and
45 years old. Similarly with the finding in young participants, exercise self-efficacy had
the strongest total effect on physical activity ( β
total
= .51); greater self-efficacy again was
associated with higher levels of physical activity in early-middle-age participants. The

34
indirect effects of environmental friendliness and social support on physical activity,
through their influence on self-efficacy, were also observed ( β
total
= .09 and β
total
= .15
accordingly). That is, environmental friendliness’s total effect of .09 was the product of
the coefficients of its direct effect on “Self-Efficacy” (.17) and the direct effect of “Self-
Efficacy” on “Physical Activities” (.51). Similarly, social support’s total effect of .15 was
the product of the coefficients of its direct effect on “Self-Efficacy” (.30) and the direct
effect of “Self-Efficacy” on “Physical Activities” (.51). No direct effects of these two
factors were observed. Compared with the finding in young participants, the social
support and especially physical environment’s influence on early-middle-age participants’
physical activity remained significant but were weaker, indicating middle-age
participants’ physical activity was less influenced by the characteristics of the
environment where they lived or work, and their family members’ and friends’ impacts
became less influential.

Figure 8. The result of the structural model analysis for the early-middle-age group.
Different from the finding in young participants, expected benefits from regular
exercise was significantly associated with physical activity level in the early-middle-age

35
group ( β
total
= .22). Both direct and indirect effects contributed to its total effect: its
indirect effect of .12 was the product of the coefficients of its direct effect on “Self-
Efficacy” (.23) and the direct effect of “Self-Efficacy” on “Physical Activities” (.51), and
its direct effect of .10 on “Physical Activities”. That is, realizing the importance and
pleasantness of health improvements from regular physical activities, a participant
showed more confidence in maintaining regular physical activity in adverse circumstance
and this cognitive change contributed to a higher level of physical activity.
Participants’ discounting rate was found to be moderately associated with
physical activity, independent of any other variables in the model ( β
total
= .14). A
participant, who preferred small present rewards compared with big but delayed rewards,
or who preferred small assured rewards rather than big but uncertain rewards, was found
to report lower levels of physical activity. Finally, perceived health status was found to be
indirectly associated with reported physical activity level through self-efficacy ( β
total

= .10), and no direct effect was observed. Its indirect effect of .10 was the product of the
coefficients of its direct effect on “Self-Efficacy” (.19) and the direct effect of “Self-
Efficacy” on “Physical Activities” (.51). A participant with better-perceived health status
was expected to show higher levels of self-efficacy and higher levels of physical
activities.
The results of late-middle-age and old participants were shown in Figure 9. Both
groups showed surprising similarities with each other. Consistent with our findings for
young and middle-age participants, self-efficacy remained the strongest determinant of
physical activity in the two groups ( β
total
= .58 for late-middle-age group and β
total
= .53
for old group). In contrast to the previous young and early-middle-age participant group,

36
the characteristics of the physical environment no longer showed a significant effect on
physical activity in either group. Social support, however, retained its positive impact on
physical activity. For late-middle-age participants, the estimated influence of social
support all came from its indirect effect through expected benefits and self-efficacy
( β
total
= .16). Two paths of indirect effects were observed: social support could influence
physical activity level through self-efficacy only ( β
indirect
= .12), which was the product of
the coefficients of its direct effect on “Self-Efficacy” (.21) and the direct effect of “Self-
Efficacy” on “Physical Activities” (.58); and it could also influence physical activity
level through expected benefits and self-efficacy both ( β
indirect
= .04), which was the
product of the coefficients of its direct effect on “Expected Benefits” (.29), the direct
effect of “Expected Benefits” on “Self-Efficacy” (.24), and the direct effect of “Self-
Efficacy” on “Physical Activities” (.58). For old participants, the pattern of social
support’s impact on physical activity level was identical with our finding in late-middle-
age group: the estimated strength of its influence came from its indirect effects through
expected benefits and self-efficacy ( β
total
= .25). This estimated total influence consisted
of two indirect effects. That is, social support could influence physical activity level
through self-efficacy only ( β
indirect
= .17), and it could also influence physical activity
level through expected benefits and self-efficacy both ( β
indirect
= .07).
Expected benefits indirectly influenced physical activity through its impact on self
efficacy ( β
total
= .14 for late-middle-age group and β
total
= .17 for old group). For late-
middle-age participants, this total effect was the product of the coefficients of its direct
effect on “Self-Efficacy” (.24) and the direct effect of “Self-Efficacy” on “Physical
Activities” (.58). Similarly, for old participants, this total effect was the product of the

37
coefficients of its direct effect on “Self-Efficacy” (.32) and the direct effect of “Self-
Efficacy” on “Physical Activities” (.53). Thus, in both groups, participants who perceived
a higher level of importance and pleasantness of the health improvements from regular
activities, showed more confidence in their ability to maintain regular physical activity,
and this cognitive change predicted a higher level of physical activity.

Figure 9. The result of structural model for the late-middle-aged participant group
(upper) and old participant group (lower).

38
Interestingly, while discounting was not significantly associated with physical
activity in either group, the perceived health status got more influential as participants got
older ( β
total
= .16 for late-middle-age group and β
total
= .25 for old group). For late-
middle-age participants, two paths of indirect effects were observed: perceived health
status could influence physical activity level through self-efficacy only ( β
indirect
= .14),
and it could also influence physical activity level through expected benefits and self-
efficacy both ( β
indirect
= .02). For old participants, one direct effect and two paths of
indirect effects were identified: perceived health status could influence physical activity
level directly ( β
direct
= .13), or through self-efficacy only ( β
indirect
= .12), or it could also
influence physical activity level through expected benefits and self-efficacy both ( β
indirect

= .06). Thus, for both groups, a participant who perceived himself or herself in better
health status, reported more confidence in his or her ability to maintain regular physical
activity, and also showed a higher level of physical activity in reality. As participants got
older, this perceived health status got more important and became one major factor in
determining one’s exercise behavior.

39
Chapter 4: Discussion
Previous research has established that an individual’s self-efficacy in maintaining a
regular physical activity program in adverse circumstances, reliably predicts physical
activity level (e.g., Marcus & Owen, 1992; Edward McAuley, et al., 2005; Motl,
McAuley, Snook, & Gliottoni, 2009). Results of the current study showed that self-
efficacy was the strongest predictor of physical activity across all groups of participants.
In Bandura’s social cognitive theory, there are four sources that can influence one’s self-
efficacy, namely one’s own experience, other’s modeling effect, social persuasions and
physical factors (Bandura, 1997). Consistently, our results also indicated that other
psychological and environmental factors did indirectly influence levels of physical
activity through self-efficacy.
One goal of the current research is to investigate how these factors and self-efficacy
together contribute to physical activity across different age groups. If the ultimate goal of
physical activity research is to promote an active and healthy life style, it is important to
understand how changes in psychological and environmental factors are related to
changes in self-efficacy. It is this knowledge that can be used to facilitate the selection
and development of appropriate interventions. Thus, instead of emphasizing the
importance of self-efficacy itself, the current study put a focus on the manipulable
precedents of exercise self-efficacy in the context of physical activity adherence. For
example, one’s own active lifestyle or previous success in exercise adherence can be a
significant contributor to his or her high level of exercise self-efficacy, and this high self-
efficacy subsequently help one to maintain an active lifestyle in future. However, this
mutual relationship between one’s previous behaviors and self-efficacy is difficult to be

40
manipulated in practice and has few empirical applications. Therefore, in the following
discussions, we mainly focus on some personal variables and environmental variables
that can be easily manipulated to enhance one’s self-efficacy and finally contribute to an
active lifestyle.
Expected Benefits, Knowledge’s Impacts on Physical Activity
The current work investigated the links between physical-activity-related knowledge,
expected benefits of regular physical activity and physical activity. The expected benefits
measure was developed from a previous study and it had already gone through one cycle
of development, data collection and data analysis (Trujillo, et al., 2004). Eleven
consequences associated with physical activities were categorized into four groups
including improvements in physical capacity, physical health, psychological capacity and
social/interpersonal relations. The knowledge measure was designed to assess
participants’ knowledge of how to exercise appropriately and knowledge of the health
benefits from regular physical activity. Compared with the knowledge measures used in
previous published studies, our measure put a special emphasis on more advanced and
elaborated knowledge about the recommended scientific ways to exercise and physical
activity’s impact on disease prevention and treatment (Lee, 1993; Morrow, et al., 2004;
Rimal, 2001). We hypothesized that there was a close relationship among knowledge,
expected benefits and reported physical activity. When one decides to maintain an active
lifestyle, it becomes increasing important to understand how to exercise appropriately
(e.g., intensity and duration) and how to avoid exercise injuries. It is also beneficial to
understand what health problems can be prevented or treated by regular physical activity.
With enough knowledge of these health issues, the process of exercising might become

41
more efficient. One might also perceive physical activity as more cognitively important
and emotionally pleasant, which contributes to the enhancement of expected benefits
from regular physical activity. All of these changes were expected to finally facilitate
one’s maintenance of physical activity.
Consistent with our hypothesis, expected benefits from regular physical activity were
found to be a moderate predictor of reported physical activity in the current study. For
middle-age and old participants, the perception of more expected benefits was associated
with a higher level of one’s confidence in maintaining physical activity in different
adverse circumstances. This enhancement in self-efficacy finally contributed to a higher
level of reported physical activity. However, no significant relationship between expected
benefits and physical activity level was found in the young participant group. For the
young participants, the perception of more outcome expectancy was found to be
associated with a higher level of perceived health status, but had no direct or indirect
effect on physical activity.
Previous studies found individuals who perceived more benefits from exercise
reported more exercise behavior and proposed expected benefits as one important factor
that was associated with higher levels of physical activity (King, et al., 1992; Sallis,
Hovell, & Hofstetter, 1992; Sallis, et al., 1989; Schneider, 1997). However, the present
study found that the expected benefits’ influence might be weaker than proposed, and
might be secondary to other psychological factors such as self-efficacy. For young
participants, given their reported excellent health status, the potential health
improvements from regular physical activity had little effect on the initiation or
maintenance of exercise behavior. For early-middle-age, late-middle-age and old

42
participants, outcome expectancy’s influence was limited and indirectly. This finding is
consistent with Banduar’s proposition that self-efficacy is a stronger predictor of
behavior than outcome expectancy (Bandura, 1997). It suggests that health promotion
materials and programs might emphasize ways to build self-efficacy in addition to
education about potential benefits of exercising. Interventions that target individuals’
self-efficacy belief and barrier-management ability might result in substantial public
health gain.
Different from our hypothesis, the present study indicated that exercise knowledge
was a less important predictor of behavior than previously supposed. Exercise knowledge
failed to predict either expected benefits nor physical activity across the four groups of
participants. These negative findings suggest that sophisticated knowledge about the
recommended scientific ways to exercise and potential health improvement from regular
physical activity, had little practical influence on one’s exercise behavior. Interestingly, a
significant but negative relationship was found in young participants, which was
independent of any other factors. It indicated that young participants with more
knowledge reported lower levels of physical activity. Although it was different from our
hypothesis, this result was somewhat consistent with the mixed results of studies
examining knowledge’s role on health behaviors (e.g., Downing, et al., 2005; Morrow, et
al., 2004; Rimal, 2001).
One possible explanation for the weak link between knowledge and physical activity
is that physical-activity-related knowledge is too general and it involves many different
subjects. It is likely that only a small portion of exercise knowledge can contribute to the
initiation and maintenance of physical activity. Future studies should aim at identifying

43
the most relevant knowledge and its contribution to physical activity. In addition,
knowledge’s weak influence on expected benefits suggest that detailed knowledge about
physical activity’s role in disease prevention had limited effect on promoting a positive
perception about the health improvements from regular physical activity. Although
knowledge plays an important role in the formation of an individual’s understanding of
the potential health benefits from regular physical activity, advanced and detailed
knowledge seems not to strength an individual’s perceived importance and pleasantness
of these outcome expectancies. Thus, volitional behaviors such as physical activity
adherence may be more easily influenced by other psychological factors (e.g., self-
efficacy) or environmental variables (e.g., support from friends) but not by the
knowledge within the individual.
Environmental Friendliness, Social Support, Heath Status and Physical Activity
On the basis of prior research documenting the importance of environmental and
social support for physical activity, the current study also aimed to specifically examine
the physical and social environment’s impacts. The physical environment’s influence was
measured by the latent variable “Environmental Friendliness” in the exploratory SEM
analysis. This construct was designed to assess different types of physical support
provided by the environment (e.g., “My daytime environment (school or work) now
provides convenient exercise facilities”). Similarly, the latent variable “Social support” in
the SEM model was developed to reflect the encouragement and support from parents,
siblings and friends regarding physical activity (e.g., “They listen to your concerns about
regular exercise”).

44
In general, our findings indicated that both environmental friendliness and social
support were positively related to physical activity. Consistently across different age
groups, self-efficacy was found to be an important factor mediating the physical and
social environment’s influence on physical activity. That is, the received support from the
physical environment, encouragements from parents, siblings and friends improve a
participant’s confidence in maintaining a regular physical activity program. This
cognitive enhancement in self-efficacy in turn, reliably contributes to a higher level of
physical activity. This indirect impact of the physical and social environment on physical
activity has also been identified by previous published studies. For example, Dishman
and colleagues tested whether self-efficacy for overcoming barriers to physical activity
had direct, or indirect (i.e., mediated) relations with naturally occurring change in
perceived social support and declines in physical activity during high school (Dishman,
Saunders, Motl, Dowda, & Pate, 2008). Results showed girls who maintained a
perception of strong social support had less of a decline in physical activity if they also
had high self-efficacy. However, girls having high self-efficacy had a greater decline in
physical activity if they perceived declines in social support. Dishman and his colleagues
suggested physical activity interventions should take into account the possibility that the
influence of perceived social support on physical activity might differ according to
efficacy beliefs about barriers to physical activity.
The current study also indicated several important differences between the physical
and social environment’s influence on physical activity. Compared with the social
environment, the strength of physical environment’s impact seemed to be limited and
became weaker as participants got older. As shown by the SEM analysis, the physical

45
environment was influential for young participants; its influence remained significant but
weaker for early-middle-age participants, and finally, had no significant impact for late-
middle-age and old participants. This changing pattern suggests that mature adults may
have enough financial means to cover the costs of different physical activities such as
gym member fee and exercise equipment purchases. Thus, the barriers from the physical
environment may become secondary and the social environmental factors may get more
influential in determining late-middle-age and old participants’ physical activity level.
There is some evidence to support this argument. For example, Giles-Corti &
Donovan examined the relative influence of individual, social environmental and
physical environmental determinants of recreational physical activity (Giles-Corti &
Donovan, 2002). Results suggested that access to a supportive physical environment was
necessary, but might be insufficient to increase recommended levels of physical activity.
Complementary strategies were required that aimed to influence individual and social
environmental factors such as self-efficacy. Although the physical environment’s
influence may be limited for late-middle-age and old individuals, it should be noted that
health behaviors usually develop early in childhood. Thus, policy interventions targeting
the creation of a physical activity friendly environment may prove especially beneficial in
the adoption of an active lifestyle for adolescents and young individuals in practice.
Compared with the physical environment, the social environment’s impact on
physical activity was more stable and robust. For the four groups of participants, one’s
perceived social support from parents, siblings and friends reliably contributed to the
enhancement of self-efficacy, and finally led to a higher level of physical activity. In
addition, social support showed a positive relation with expected benefits from regular

46
physical activities in young, late-middle-age and old participant groups. These findings
suggest that the positive support from family and friends was a strong contributor to the
initiation and maintenance of one’s physical activity. It not only helped build perceived
confidence in one’s ability to maintain physical activity, but also helped one to realize the
potential health improvements from regular physical activity. Given the consistent
influence on individuals at different ages, social support holds considerable potential as a
candidate for health promoting interventions. For adolescents and young individuals,
these interventions could focus on parents’ direct participation in physical activity and
strengthening supportive ties with active friends. For middle-age and old individuals,
such programs might wisely focus on developing spouse’s skills regarding supportive
communication and behaviors patterns for physical activity.
Discounting and Physical Activity
Many health-related behaviors involve a choice between a small but immediate
reward and a delayed but large reward. A key construct that helps us to understand those
choices is the concept of discounting. When choosing between delayed or uncertain
outcomes, individuals discount the value of such outcomes on the basis of the expected
time to receipt of the outcome, or based on the likelihood of their occurrence. As shown
by our meta-analysis, there is ample evidence of a reliable relationship between delayed
discounting and mild addictive behaviors such as smoking, gambling and drinking (Li,
2008). However, this positive finding is limited to mild addictive behavior, and no
consistent relation has been observed between discounting and non-addictive behavior
such as physical activity.

47
In the current research, the early-middle-age group showed that physical activity had
a moderate association with discounting, independent of any other psychological or
environmental variables ( β
total
= .14). A participant, who preferred small present rewards
compared with larger but delayed rewards, or who prefer small assured rewards over
larger but uncertain rewards, was found to report lower levels of physical activity.
However, no significant relationships between discounting and physical activity were
observed in the other three age groups. This mixed pattern of results suggests that
discounting processes are a less important predictor of physical activity than has
previously been supposed, and its influence is at most weak to moderate. One possible
explanation is that individuals at different ages are living in different social/physical
environment, and their psychological characteristics such as personal preferences,
knowledge and perceptions might be different from each other. The differences in
psychological factors and the characteristics of the environment within which the choice
of physical activity happens, might contribute to the mixed results.
Consistent with this argument, recent findings from a behavioral economic study
provide another feasible explanation (Audrain-McGovern, et al., 2004). Focusing on the
influence of alternative reinforcers, Audrian-McGovern et al. propose that the reinforcing
value of substances can be enhanced by complementary reinforcers, and reduced by
substitute reinforcers. And the discounting process can help explain the wide individual
differences in the value of immediate or delayed reinforcers. In their study, substitute
reinforces for smoking (e.g., school involvement, academic performance, and physical
activity), complementary activities to smoking (e.g., peer smoking and substance use)
and delay discounting were measured among 983 adolescents. Latent growth modeling

48
indicated that substitute reinforces reduced the odds of smoking progression almost two-
fold while the complementary reinforces increased the odds by 1.14. More interestingly,
delay discounting was found to indirectly influence the odds of smoking progression
through complementary reinforces (i.e., an increase of one unit in delay discounting
resulted in a 0.08 increase in complementary reinforce use). The authors concluded that
delay discounting might indirectly influence the development of smoking by its impacts
on what types of activities an adolescent finds rewarding, and suggested that discounting
should be considered in a broader environmental context (e.g., available alternative
reinforcers) in which a behavior such as the choice to smoke occurs. Failure to consider
the environmental variables may partially explain why mixed associations between
discounting and health-related behaviors have been reported.
In addition, poor impulse control might also play a critical role. In this argument,
discounting is the tendency to deliberatively devalue the future and poor impulse control
refers to the failure to consider the future. Thus, the weak association between
discounting and physical activity in young and older participants might indicate their
reluctancy to consider the future. For example, young participants are usually in excellent
health and they may lack motivation to consider the long-term benefits of regular
physical activity. The future might be a distant and vague perception and the present may
be much more salient to them. Thus discounting may not be influential in their behaviors.
For late-middle-age and old participants, as more than half of them were obese, they
might want to have more physical activities in order to improve personal health. However,
as they are in the later part of their life cycle, the delayed health benefits from regular
physical activity, in some extent, might be too far away for them to consider. Reflections

49
about the past may seem much more meaningful than looking forward to the future. Thus,
discounting process might also have a less influential role than proposed in these two
groups. Interestingly, for early-middle-age people, the obesity increased from young
people’s 21% rate to 45%, which might contribute to a high level of motivation to
exercise in the early-middle-age group (See Figure 3). In addition, as this group of
participants was still comparatively young, the possible health improvements from
regular physical activity could be an important factor in their decision-making process
and this may be why discounting processess were found to be a moderate factor
influencing the early-middle-age group’sexercise behavior.
Consistent with this argument, Nagin and Pogarsky analyzed data from the National
Longitudinal Survey of Adolescent and found that high discounting was a better predictor
of deliberative or future-oriented problem outcomes, whereas poor impulse control was a
better predictor of urge driven behaviors or conduct involving little forethought (Nagin &
Pogarsky, 2004). Thus, it is possible that the observed weak relations between
discounting and physical activity in young, late-middle-age and old participants indicated
that they just did not reflect on their future seriously.
Limitations and Conclusions
The major strengths of this study are the large sample of participants selected from
the full adult life span, examination of psychological and environmental variables in an
integrated framework and the exploration of potential moderators. However, a number of
potential limitations should be noted.
In the current research, attempts were made to have participants include all physical
activities that would produce an exercise effect by giving them a broad definition of

50
exercise. However, responses were still based on the participants’ general concept of
physical activity and this might result in participants’ reporting activities that may not be
sufficient to produce any significant exercise effects. In addition, the potential distorted
exercising time estimate in which participants unconsciously exaggerate their exercising
time remain unknown. Future research should seek to overcome these shortcomings by
using current daily diaries of exercise activity rather than retrospective reports.
The second potential limitation is that the sample was composed of mostly middle-
aged staff and faculty in a private university. Among our participants, 77% of them
reported to have a four-year college degree or higher and 45% of the participants reported
having an annual family income with $90,000 or more. Thus, it is highly likely that the
participants had a higher level of education and socioeconomic status than would a
random sample of the general population. Therefore, it is necessary to be cautious when
generalizing findings from the current study to other samples. In addition, this study is an
exploratory study, and thus the models presented here are ones that have shown
satisfactory model fits and theoretically reasonable interpretations. However, it is
possible that other potential models, which have not been examined in the current study,
can also show acceptable model fits and provide alternative interpretations. Thus, further
confirmatory research is needed to validate the findings of the current study in a broader
sample of the population.
Third, all variables were measured with self-reported items in the current study and
the reliance on self-reports of physical activity, personal, social and environmental
variables might be a potential limitation. Additional objective measures might be needed
to validate the perception measures used in the current study. However, it should be noted

51
that compared with objective measures, self-reports are less expensive and can be utilized
in more studies in a wider variety of populations and locations. It is also a commonly
used technique in survey studies with a relatively large number of participants, as was the
case here. In addition, results of this study indicated that self-efficacy beliefs were
consistently influenced by the perceptions of physical environment and social support
across different participant groups, which underlined the need for perception
measurements in the current study.
Despite the limitations to the current study, it adds to the increasing evidence for
the influence of personal, social and environmental factors on physical activity levels.
Results showed self-efficacy beliefs consistently exerted the strongest impact on physical
activity across the three groups of participants. Social support and physical environment
influences were of differential importance across different age groups: While social
support retained its influence on physical activity through its effect on self-efficacy, the
influence of physical environment got weaker as participants got older. Other variables
including expected benefits, knowledge and discounting process only influenced physical
activity in certain age group (and to a relatively weak extent) and no reliable influence
across the three groups was observed. The differential influences of these personal, social
and environmental factors on physical activity in different age groups in this study further
support the importance of undertaking segmentation analyses to better understand
barriers and motivations for physical activity. The challenge remains to establish the
causal relationships between these variables with further confirmatory and intervention
studies and to explore the potential mechanisms underlying the observed relationships.

52
References
Ainsworth, B., Wilcox, S., Thompson, W., Richter, D., & Henderson, K. (2003).
Personal, social, and physical environmental correlates of physical activity in
African-American women in South Carolina. American journal of preventive
medicine, 25, 23-29.

Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human
decision processes, 50(2), 179-211.

Audrain-McGovern, J., Rodriguez, D., Tercyak, K. P., Epstein, L. H., Goldman, P., &
Wileyto, E. P. (2004). Applying a behavioral economic framework to
understanding adolescent smoking. Psychology of Addictive Behaviors, 18(1), 64-
73.

Baldwin, A. S., Rothman, A. J., Hertel, A. W., Linde, J. A., Jeffery, R. W., Finch, E. A.,
et al. (2006). Specifying the Determinants of the Initiation and Maintenance of
Behavior Change: An Examination of Self-Efficacy, Satisfaction, and Smoking
Cessation. Health Psychology, 25(5), 626-634.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.

Booth, M., Owen, N., Bauman, A., Clavisi, O., & Leslie, E. (2000). Social-ognitive and
perceived environment influences associated with physical activity in older
Australians. Preventive Medicine, 31(1), 15-22.

Brownson, R., Baker, E., Housemann, R., Brennan, L., & Bacak, S. (2001).
Environmental and policy determinants of physical activity in the United States.
American Journal of Public Health, 91(12), 1995-2003.

Chapman, G. B. (1998). Sooner or later: The psychology of intertemporal choice. In D.
L. Medin (Ed.), The psychology of learning and motivation (Vol. 38, pp. 83-113).
San Diego, CA: Academic Press.

Chapman, G. B. (2005). Short-Term Cost for Long-Term Benefit: Time Preference and
Cancer Control. Health Psychology. Special Issue: Basic and Applied Decision
Making in Cancer Control, 24(4), 41-48.

Christensen, U., Schmidt, L., Budtz-Jorgensen, E., & Avlund, K. (2006). Group cohesion
and social support in exercise classes: Results from a Danish intervention study.
Health Education and Behavior, 33(5), 677-689.

Courneya, K., & Bobick, T. (2000). Integrating the theory of planned behavior with the
processes and stages of change in the exercise domain. Psychology of Sport and
Exercise, 1(1), 41-56.

53
Crocker, L., & Algina, J. (1986). Introduction to classical and modern test theory. New
York: Holt, Rinehart and Winston.

Dishman, R., Saunders, R., Motl, R., Dowda, M., & Pate, R. (2008). Self-Efficacy
Moderates the Relation Between Declines in Physical Activity and Perceived
Social Support in High School Girls. Journal of Pediatric Psychology, 34(4), 441-
451.

Downing, J., Masterson, G., & Gray, L. (2005). Tracking the Effects of a University Core
Fitness for Life Course on the Wellness Knowledge and Exercise Behavior of
Selected Students. Paper presented at the 2005 AAHPERD National Convention
and Exposition.

Gellish, R., Goslin, B., Olson, R., McDonald, A., Russi, G., & Moudgil, V. (2007).
Longitudinal Modeling of the Relationship between Age and Maximal Heart
Rate. Medicine & Science in Sports & Exercise, 39(5), 822-829.

Giles-Corti, B., & Donovan, R. (2002). The relative influence of individual, social and
physical environment determinants of physical activity. Social science &
medicine, 54(12), 1793-1812.

Hohepa, M., Scragg, R., Schofield, G., Kolt, G., & Schaaf, D. (2007). Social support for
youth physical activity: Importance of siblings, parents, friends and school
support across a segmented school day. International Journal of Behavioral
Nutrition and Physical Activity, 4(1), 54-62.

Jago, R., Baranowski, T., Zakeri, I., & Harris, M. (2005). Observed environmental
features and the physical activity of adolescent males. American journal of
preventive medicine, 29(2), 98-104.

Janis, I., & Mann, L. (1977). Decision making: A psychological analysis of conflict,
choice, and commitment. New York: Free Press.

Jeffery, R., & Wing, R. (1995). Long-term effects of interventions for weight loss using
food provision and monetary incentives. Journal of Consulting and Clinical
Psychology, 63(5), 793-796.

Keytel, L., Goedecke, J., Noakes, T., Hiiloskorpi, H., Laukkanen, R., Van Der Merwe,
L., et al. (2005). Prediction of energy expenditure from heart rate monitoring
during submaximal exercise. Journal of Sports Sciences, 23(3), 289-297.

King, A., Blair, S., Bild, D., Dishman, R., Dubbert, P., Marcus, B., et al. (1992).
Determinants of physical activity and interventions in adults. Medicine & Science
in Sports & Exercise, 24(6), 221-236.

54
King, A., Castro, C., Wilcox, S., Eyler, A., Sallis, J., & Brownson, R. (2000). Personal
and environmental factors associated with physical inactivity among different
racial-ethnic groups of US middle-aged and older-aged women. Health
Psychology, 19(4), 354-364.

Lee, C. (1993). Attitudes, knowledge, and stages of change: a survey of exercise patterns
in older Australian women. Health psychology: official journal of the Division of
Health Psychology, American Psychological Association, 12(6), 476-480.

Levy, A. G., Micco, E., Putt, M., & Armstrong, K. (2006). Value for the future and breast
cancer-preventive health behavior. Cancer Epidemiology, Biomarkers &
Prevention, 15(5), 955-960.

Li, J. (2008). Discounting Process in the Health Domain: A Review of the Literature.
Unpublished manuscript.

Li, J., Trujillo, K., Brougham, R., Brown, C., & Walsh, D. (2009). Adherence to Physical
Activity: An Empirical Test of the Knowledge–Perception-Behavior Model
Unpublished manuscript.

MacCallum, R., Browne, M., & Sugawara, H. (1996). Power analysis and determination
of sample size for covariance structure modeling. Psychological methods, 1(2),
130-149.

Marcus, B., & Owen, N. (1992). Motivational Readiness, Self-Efficacy and Decision-
Making for Exercise. Journal of Applied Social Psychology, 22(1), 3-16.

McAuley, E., Blissmer, B., Katula, J., & Duncan, T. (2000). Exercise environment, self-
efficacy, and affective responses to acute exercise in older adults. Psychology &
Health, 15(3), 341-355.

McAuley, E., Elavsky, S., Jerome, G. J., Konopack, J. F., & Marquez, D. X. (2005).
Physical Activity-Related Well-Being in Older Adults: Social Cognitive
Influences. Psychology and Aging, 20(2), 295-302.

Morrow, J., Krzewinski-Malone, J., Jackson, A., Bungum, T., & Fitzgerald, S. (2004).
American adults' knowledge of exercise recommendations. Research Quarterly
for Exercise and Sport, 75(3), 231-237.

Motl, R., McAuley, E., Snook, E., & Gliottoni, R. (2009). Physical activity and quality of
life in multiple sclerosis: Intermediary roles of disability, fatigue, mood, pain,
self-efficacy, and social support. Psychology, Health & Medicine, 14(1), 111-124.

55
Mowen, A., Orsega-Smith, E., Payne, L., Ainsworth, B., & Godbey, G. (2007). The role
of park proximity and social support in shaping park visitation, physical activity,
and perceived health among older adults. Journal of physical activity & health,
4(2), 167-179.

Myerson, J., Green, L., Hanson, J. S., Holt, D. D., & Estle, S. J. (2003). Discounting
delayed and probabilistic rewards: Processes and traits. Journal of Economic
Psychology, 24(5), 619-635.

Myerson, J., Green, L., & Warusawitharana, M. (2001). Area under the curve as a
measure of discounting. Journal of the Experimental Analysis of Behavior, 76(2),
235-243.

Nagin, D. S., & Pogarsky, G. (2004). Time and Punishment: Delayed Consequences and
Criminal Behavior. Journal of Quantitative Criminology. Special Issue: Offender
Decision Making, 20(4), 295-317.

Nieman, D. C. (2007). exercise testing and prescription: a health-related approach. New
York: McGraw-Hill.

Nunnally, J., & Bernstein, I. (1994). Psychometric theory: McGraw-Hill, New York.

Ogden, C., Carroll, M., McDowell, M., & Flegal, K. (2007). Obesity among adults in the
United States--no statistically significant chance since 2003-2004. NCHS data
brief(1), 1-8.

Phongsavan, P., McLean, G., & Bauman, A. (2007). Gender differences in influences of
perceived environmental and psychosocial correlates on recommended level of
physical activity among New Zealanders. Psychology of Sport & Exercise, 8(6),
939-950.

Prochaska, J., & DiClemente, C. (1983). Stages and processes of self-change of smoking:
towards an integrative model of change. . Journal of Consulting and Clinical
Psychology, 51, 390-395.

Prochaska, J., & DiClemente, C. (1984). The transtheoretical approach: Crossing
traditional boundaries of therapy. Pacific Grove, CA: Brooks/Cole.

Prochaska, J., & Velicer, W. (1997). Behavior change: The transtheoretical model of
health behavior change. American Journal of Health Promotion, 12(1), 38-48.

Rimal, R. (2001). Longitudinal influences of knowledge and self-efficacy on exercise
behavior: Tests of a mutual reinforcement model. Journal of Health Psychology,
6(1), 31-46.

56
Rosado, J., Sigmon, S. C., Jones, H. E., & Stitzer, M. L. (2005). Cash Value of Voucher
Reinforcers in Pregnant Drug-Dependent Women. Experimental and Clinical
Psychopharmacology, 13(1), 41-47.

Sacco, W. P., Wells, K. J., Vaughan, C. A., Friedman, A., Perez, S., & Matthew, R.
(2005). Depression in Adults With Type 2 Diabetes: The Role of Adherence,
Body Mass Index, and Self-Efficacy. Health Psychology, 24(6), 630-634.

Sallis, J., Hovell, M., & Hofstetter, C. (1992). Predictors of adoption and maintenance of
vigorous physical activity in men and women. Preventive Medicine, 21(2), 237-
251.

Sallis, J., Hovell, M., Hofstetter, C., Faucher, P., Elder, J., Blanchard, J., et al. (1989). A
multivariate study of determinants of vigorous exercise in a community sample.
Preventive Medicine, 18(1), 20-34.

Sallis, J., King, A., Sirard, J., & Albright, C. (2007). Perceived environmental predictors
of physical activity over 6 months in adults: activity counseling trial. Health
psychology: official journal of the Division of Health Psychology, American
Psychological Association, 26(6), 701-709.

Sasidharan, V., Payne, L., Orsega-Smith, E., & Godbey, G. (2006). Older adults' physical
activity participation and perceptions of wellbeing: Examining the role of social
support for leisure. Managing Leisure, 11(3), 164-185.

Schneider, J. (1997). Self-Regulation andExercise Behavior in Older Women. Journal of
Gerontology, 52(5), 235-241.

Shields, C. A., & Brawley, L. R. (2006). Preferring Proxy-agency Impact on Self-efficacy
for Exercise. Journal of Health Psychology, 11(6), 904-914.

Springer, A., Kelder, S., & Hoelscher, D. (2006). Social support, physical activity and
sedentary behavior among 6 th-grade girls: a cross-sectional study. International
Journal of Behavioral Nutrition and Physical Activity, 3(1), 8-17.

Strecher, V. J., DeVellis, B. M., Becker, M. H., & Rosenstock, I. M. (1986). The Role of
Self-Efficacy in Achieving Health Behavior Change. Health Education &
Behavior, 13(1), 73-91.

Tester, J., & Baker, R. (2009). Making the playfields even: Evaluating the impact of an
environmental intervention on park use and physical activity. [doi: DOI:
10.1016/j.ypmed.2009.01.010]. Preventive Medicine, 48(4), 316-320.

Trujillo, K., Brougham, R., & Walsh, D. (2004). Age differences in reasons for
exercising. Current Psychology, 22(4), 348-367.

57
USDHHS. (2008). 2008 physical activity guidelines for Americans. Retrieved from
http://www.health.gov/PAGuidelines/.

Velicer, W., DiClemente, C., Prochaska, J., & Brandenburg, N. (1985). A decisional
balance measure for predicting smoking cessation. Journal of Personality and
Social Psychology, 48, 1279-1289.

Wilson, D., Ainsworth, B., & Bowles, H. (2007). Body mass index and environmental
supports for physical activity among active and inactive residents of a US
southeastern county. Health Psychol, 26(6), 710.

Woodgate, J., Brawley, L., & Shields, C. (2007). Social Support in Cardiac
Rehabilitation Exercise Maintenance: Associations with Self-Efficacy and Health-
Related Quality of Life 1. Journal of Applied Social Psychology, 37(5), 1041-
1059.

Appendices

58
Appendix A. Results of SEM analysis for young participants.

59
Appendix B. Results of SEM analysis for early-middle-age participants.

60
Appendix C. Results of SEM analysis for late-middle-age participants.

61
Appendix D. Results of SEM analysis for old participants.

62
Appendix E. Original exercise preference questionnaires used in the current study.
Study of Exercise Preference
David A. Walsh, Ph. D., Director
Jian Li, Principal Investigator
Department of Psychology
University of Southern California
Los Angeles, CA 90089-1061

This questionnaire looks at factors we believe contribute to people’s decisions about
exercise. Why do we care about these factors? We want to understand, in a general way,
how and why people make decisions about important life matters. Studying people’s
decisions about exercise offers an opportunity to look at some of the psychological
factors that underlie human decision-making.

Your answers to the questionnaire will be completely anonymous. We do not need you to
put your name on the questionnaire, and we will never attempt to trace these answers
back to you. The questionnaire takes about 15 to 30 minutes to complete.

Since we won’t be able to use your responses unless you complete every section, we hope
you will take the time to complete all parts of the questionnaire—however, do not forget
that your participation is completely voluntary: You may discontinue your participation
at anytime without penalty. We greatly appreciate your willingness to contribute your
time and energy to our research efforts. Based on your participation, you will enter a
lottery for a $50 gift card. In total, 24 gift cards will be given and winners will be notified
via email. Be sure to provide your email address when it is requested at the end of the
survey. Your email address will not be stored with your survey data so that your
anonymity will be preserved.

The University Park Institutional Review Board (IRB) designee determined
that this project meets the requirements outlined in 45 CFR 46.101 (b) (2),
and qualifies for exemption from IRB review. IRB Exemption of this study
was granted on 5/1/2008.

USC UPIRB # UP-08-00116

Email _____________________________

63
Appendix E, continued
Demographic Information

Gender

Age

1 Male
2 Female

______ years old

Marital
Status
(check one)
1 Single, never married
2 Married without children
3 Married with children
4 Divorced
5 Separated
6 Widowed 7 Other

Ethnic
(check one)

1 White/Caucasian
2 African American
3 Hispanic
4 Asian
5 Native American
6 Pacific Islander
7 Other

Education
Level
(check one)
1 Less than High School
2 High School / GED
3 2-year College Degree
4 4-year College Degree
5 Master’s Degree
6 Doctoral Degree
7 Professional Degree (JD, MD)
8 Other

Height

Weight

______ feet and

______inches

_____ pounds

Family
Annual
Income
(check one)
1 Below $20,000
2 $20,000 - $29,999
3 $30,000 - $39,999
4 $40,000 - $49,999
5 $50,000 - $59,999
6 $60,000 - $69,999
7 $70,000 - $79,000
8 $80,000 - $89,999
9 $90,000 or more

Employment
Status
(check one)
1 Full-time student
2 Part-time student
3 Full-time employee
4 Part-time employee
5 Retired 6 Other

Health
Status
(check one)
1 You can perform all physical activities of daily living without
assistance. (Excellent capacity)
2 You can perform all physical activities without assistance but may
need some help with the heavy work such as laundry and housekeeping.
(Good capacity)
3 You regularly require help with certain physical activities and/or heavy
work but can get through any single day without help.
4 You need help each day but not necessarily throughout the day or
night. (Severely impaired capacity)
5 You need help throughout the day and/or night to carry out the
activities of daily living. (Completely impaired capacity)

64
Appendix E, continued
Exercise Knowledge Measure Part 1
There are a number of professional groups in the United States that study exercise and
make recommendations about the types and amount of exercise that is healthy. The
following questions measure what you know about the widely accepted recommendations
many professional groups offer about exercise.

1. It is widely recommended that people engage in which of the following types of
exercise?
a. Aerobic
b. Strength training
c. Low intensity recreational
d. Aerobic and low intensity recreational
e. Aerobic and strength training

2. Three women, each 35 years of age, train at a heart rate of 150 bpm, 30
minutes/session, 5 days a week. One cycles, one runs, and the other uses a rowing
machine. Which one will develop the best aerobic fitness?
a. Woman who cycles
b. Woman who runs
c. Woman who rows
d. All will develop the same level of heart and lung fitness
e. None will improve their fitness because their training program is too easy

3. How many days a week is it widely recommended that you engage in aerobic exercise?
a. at least 1-2 days
b. at least 2-3 days
c. at least 3-4 days
d. at least 4-5 days
e. at least 5-6 days

4. How many minutes is it recommended that you spend on each aerobic exercise period?
a. at least 10 minutes
b. at least 20 minutes
c. at least 30 minutes
d. at least 40 minutes
e. at least 50 minutes

5 What is the minimum level of heart rate intensity recommended to produce an aerobic
exercise benefit?
a. 30% of maximum heart rate
b. 45% of maximum heart rate
c. 60% of maximum heart rate

65
Appendix E, continued

d. 75% of maximum heart rate
e. 90% of maximum heart rate

6. Which of the following activities can best build BOTH cardiovascular health and
muscle strength significantly?
a. Running (brisk pace, 8 mph)
b. Weight training
c. Yoga
d. Cross-country skiing (brisk speed)
e. Tennis (competitive)

7. How often is it recommended that you engage in strength training exercise?
a. at least 1-2 times a week
b. at least 2-3 times a week
c. at least 3-4 times a week
d. at least 4-5 times a week
e. at least 5-6 times a week

8. In order to increase muscular size (hypertrophy training), which of the following load
is mostly recommended in weight training exercise?
a. Use a weight equal to 20% of the maximum you can lift in a single repetition
b. Use a weight equal to 40% of the maximum you can lift in a single repetition
c. Use a weight equal to 60% of the maximum you can lift in a single repetition
d. Use a weight equal to 80% of the maximum you can lift in a single repetition
e. Use a weight equal to 100% of the maximum you can lift in a single repetition

9. Which of the following statements about abdominal exercise is true?
a. Abdominal exercise gets rid of the blubber around your middle
b. For best results, you should do at least 100 repetitions of abdominal exercise in any
single exercise session
c. You need to work your abs every day
d. If you have a bad back, training the abs will worsen it
e. Crunches work better than sit-ups to efficiently develop the abdominal muscles

10. Which of the following strength training guidelines is NOT appropriate?
a. Perform a minimum of 8-10 separate exercise that train the major muscle groups
b. Perform every exercise through a full range of motion
c. Hold your breath while performing each repetition
d. Perform both the lifting and the lowering portion of the resistance exercise in a
controlled manner
e. If possible, exercise with a training partner who can provide assistance and motivation

11. The first step of a good warm-up is
a. Stretching

66
Appendix E, continued

b. Low-intensity aerobic activity
c. Resistance training
d. Isometrics
e. Low-intensity strength training activity

12. It is recommended that static stretching exercises be held for how long to develop
flexibility?
a. 10-30 seconds
b. 30-50 seconds
c. 50-70 seconds
d. 70-90 seconds
e. 90-110 seconds

13. Which athlete is likely to have the lowest bone density?
a. Swimmer
b. Runner
c. Weight lifter
d. Soccer player
e. Basketball player

14. Which of the following nutrition suggestions is NOT appropriate in a typical
endurance training or competition (such as long-distance running and triathlon event)?
a. Increase total energy intake
b. Keep the carbohydrate intake high
c. Drinking large amount of water during training and the event
d. Keep a close watch on possible iron deficiency, especially for women
e. Vitamin and mineral supplements are important

15. Over-exercising, either by doing aerobic exercise too hard, for too long or too often,
can lead to injury, and abandonment of your fitness program. Which one of the following
is NOT a sign of over-exercising?
a. Increased appetite
b. Lack of desire to train
c. Persistent muscle soreness
d. Persistent fatigue
e. Elevated resting heart rate

67
Appendix E, continued
Exercise Knowledge Measure Part 2
The merits of exercise — from preventing chronic health diseases to boosting confidence
and self-esteem — are hard to ignore. However, in some health and disease areas, very
little evidence exists to support a prevention or treatment role for regular exercise. We
have listed a series of health issues in the following table. Regular exercise has consistent
effects on some of them but not on others. Please indicate your own thinking about
exercise’s potential effects on these health issues by placing a check in the corresponding
column.

Potential Health Benefits Ineffective Effective

Prevention of coronary heart disease
Prevention of high blood pressure
Treatment of high blood pressure
Raises HDL cholesterol
Prevention of Type 2 diabetes
Alleviate/prevent depression
Prevention of overweight
Increases your range of motion
Prevention of colon cancer
Helps build up bone density

Prevention of arthritis
Treatment of arthritis
Prevention of asthma
Treatment of asthma
Treatment of cancer
Lowers total blood cholesterol
Lowers LDL cholesterol
Treatment of type 1 diabetes
Slow progression of HIV to AIDS
Treatment of obesity

68
Appendix E, continued
Exercise Habit
Please write down your answers and circle the corresponding choice.

Over the past 14
days, on how many
days did you do the
following listed
activities?
On any of those days
you did the following
listed activities, how
many minutes did you
spend on those activities
on average?
1) Take part in hard physical activities
that gets your heart rate to 85% or
more of your maximum and results in
extremely heavy breathing, sweating
and requires extreme exertion that
prevents you from talking. This might
include playing full-court basketball
at full effort, running, swimming, or
cycling at a very fast, and intense
pace.

_____________ days

_____________ minutes

2) Physical activity that gets your
heart rate between 65% to 85% of
your maximum and results in heavy
breathing and sweating but does not
keep you from talking. This might
include playing full-court basketball
at a slow pace, running, swimming, or
cycling at a moderately intense pace.

_____________ days

_____________ minutes

3) Physical activity that gets you
moving but does little to increase your
breathing or heart rate. This might
include walking, golf, bowling,
gardening, house cleaning, etc.

_____________ days

_____________ minutes

4) Overall, how active do you think you have been is in the past two weeks?
0 – 1 – 2 – 3 – 4 – 5 – 6
extremely sedentary neither sedentary nor active extremely active

69
Appendix E, continued
Self-efficacy
Please indicate your answers by circling the corresponding choice.
0 1 2 3 4 5 6
extremely unsure neither confident nor unsure extremely confident

1) How confident are you that you could keep exercising
regularly when your family members or friends have a
sedentary lifestyle?
0 – 1 – 2 – 3 – 4 – 5 – 6

2) How confident are you that you would be able to exercise
regularly at home?
0 – 1 – 2 – 3 – 4 – 5 – 6

3) How confident are you that you would be able to follow
your exercise plan when you get very busy?
0 – 1 – 2 – 3 – 4 – 5 – 6

4) How confident are you that you would be able to follow
your exercise plan when you are in a bad mood (e.g., anxious,
sad, depressed and irritable)?
0 – 1 – 2 – 3 – 4 – 5 – 6

5) How confident are you that you would be able to follow
your exercise plan when you are tired?
0 – 1 – 2 – 3 – 4 – 5 – 6

6) How confident are you that you would be able to exercise
regularly even if exercising takes you a lot of time?
0 – 1 – 2 – 3 – 4 – 5 – 6

7) How confident are you that you would be able to exercise
regularly even if you do not see any positive effects of
exercises?
0 – 1 – 2 – 3 – 4 – 5 – 6

8) How confident are you that you would be able to resume
exercising even if you stopped for a short time (a couple of
days)?
0 – 1 – 2 – 3 – 4 – 5 – 6

9) Overall, how confident are you that you would be able
to exercises regularly?
0 – 1 – 2 – 3 – 4 – 5 – 6

70
Appendix E, continued
Social Support
Please circle the item that describes your perception of how people close
to you (e.g., parents, siblings and friends) support your exercise
activities.
0 1 2 3 4 5 6
disagree completely neither agree nor disagree agree completely

1) They listen to your concerns about regular exercise. 0 – 1 – 2 – 3 – 4 – 5 – 6

2) They agree with your decisions about exercising regularly. 0 – 1 – 2 – 3 – 4 – 5 – 6

3) They encourage choices favorable to your exercise. 0 – 1 – 2 – 3 – 4 – 5 – 6

4) They assist with your regular exercise. 0 – 1 – 2 – 3 – 4 – 5 – 6

5) Overall, how do you agree with the statement that you
have received significant social support for your exercise
activities?
0 – 1 – 2 – 3 – 4 – 5 – 6

71
Appendix E, continued
Environmental Friendliness
Please indicate the extent to which the environment provides different
types of support for your exercise. Specify your answers by circling the
corresponding choice.
0 1 2 3 4 5 6
disagree completely neither agree nor disagree agree completely

1) My family has exercise equipments which I can use very
easily.
0 – 1 – 2 – 3 – 4 – 5 – 6

2) My daytime environment (school or work) now provides
convenient exercise facilities
0 – 1 – 2 – 3 – 4 – 5 – 6

3) My daytime environment (school or work) provides a series
of useful courses and guidance on exercise.
0 – 1 – 2 – 3 – 4 – 5 – 6

4) The neighborhood where I now live in has convenient
exercise facilities.
0 – 1 – 2 – 3 – 4 – 5 – 6

5) Overall, how do you agree with the statement that your
living environment provides enough support for exercise
activities?
0 – 1 – 2 – 3 – 4 – 5 – 6

72
Appendix E, continued
Consequences of Exercise
First, we want you to rate how important it is to you to achieve each of the
following consequences. This scale ranges from “not important at all” to “very
important.” Please note that we have abbreviated important as “impt.” because of space
limitations. In the second column, we want you to rate how likely it is that you can
achieve the listed consequence. This scale ranges from a low of “0” or I won’t achieve it
(won’t occur) to “6” I’m certain to achieve it (certain to occur). In the third column, we
want you to rate how pleasant or unpleasant each consequence would be for you if it did
occur. This scale ranges from a low of “0” indicating the consequence would be very
unpleasant (horrible) if it occurred, to “6” it would be very pleasant (terrific). The fourth
and final column asks you to rate how long you think it would take you to achieve each
consequence if you exercise regularly. A score of “0” indicates the consequence would
occur quickly, in 1 to 2 months, whereas a “6” indicates it would take at least 11 to 12
months of exercise to achieve the consequence.

Consequence of
exercise
How important
is this to
consider?
How likely that
you can
achieve it?
How pleasant
or unpleasant
would this be?
How long will
it take you to
achieve it?

Left 0: Not
Important at all
Right 6: Very
Important
Left 0: Won’t
occur
Right 6: Certain to
occur
Left 0: horrible
Right 6: terrific
Left 0: occur
quickly
Right 6: at least
11 to 12 months
I would have a
trim body shape.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would have firm
muscle tone.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would be
sexually
attractive.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

My heart would
be stronger and
more efficient,
providing better
circulation for my
whole body.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would be
stronger, faster
and have better
coordination.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

73
Appendix E, continued

Consequence of
exercise
How important
is this to
consider?
How likely that
you can
achieve it?
How pleasant
or unpleasant
would this be?
How long will
it take you to
achieve it?

Left 0: Not
Important at all
Right 6: Very
Important
Left 0: Won’t
occur
Right 6: Certain to
occur
Left 0: horrible
Right 6: terrific
Left 0: occur
quickly
Right 6: at least
11 to 12 months
My friends and/or
family would be
pleased with me.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would have
more
opportunities to
make and be with
friends.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would be able to
work and play
with fewer
injuries and less
muscle soreness.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would have
greater resistance
to common
illness, such as
colds and
influenzas; I
would catch fewer
and the ones I
caught would be
less severe.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would be less
stressed and more
relaxed.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

I would have
better control of
myself, and enjoy
higher self-esteem
and self-
confidence.
0 –1–2–3–4–5– 6

0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6 0 –1–2–3–4–5– 6

74
Appendix E, continued
Discounting Preference 1

This portion of our study measures how you value future events. We think this is
relevant to understanding differences in people’s exercise behavior because the benefits
of exercise often come in the future. For example, improvements in body composition
and cardiovascular health may take many months to achieve, while improvement in mood
and feelings of wellbeing may occur in a few weeks.

Our discounting measure asks you to imagine that you have a winning lottery ticket. You
could collect your prize at some time in the future or the winning ticket could be sold
immediately to a lottery agent for cash now. In the following example, an individual
shows that he would accept $85 today rather than waiting 3 months to get the full $100
prize.

EXAMPLE: Which one do you prefer?

Prize after 3 Months Cash
now
$100 $55
$100 $60
$100 $65
$100 $70
$100 $75
$100 $80
$100 $85
$100 $90
$100 $95
$100 $100

75
Appendix E, continued

76
Appendix E, continued
Discounting Preference 2

This portion of our study measures how you value uncertain events. We think this is
relevant to understanding differences in people’s exercise behavior because the benefits
of exercise are often uncertain. For example, improvements in body composition and
cardiovascular health are only going to occur if you do enough exercise for a long enough
period of time. Whether or not you will be able to achieve these gains is not completely
certain, but depends on many things that may be beyond your control.

Similar with our last discounting measure, we ask you to imagine that you have a lottery
ticket. However, the ticket could be sold to a lottery agent for cash before it is determined
to be a winning or losing ticket. As shown in the following example, a participant circles
his preferred choices and indicates that he would like to keep the ticket with 50%
winning chance unless he can sell it for $45 or more now.

EXAMPLE: Which one do you prefer?

Prize with 50%
Winning Probability
Cash
now
$100 $15
$100 $20
$100 $25
$100 $30
$100 $35
$100 $40
$100 $45
$100 $50
$100 $55
$100 $60

77
Appendix E, continued

Abstract (if available)

Abstract Research in the area of physical activity has typically focused separately on internal variables within individuals and environmental variables where the choice of physical activity happens. The current research examined these influential factors simultaneously in an integrated framework. A sample of 1552 participants mainly consisting of students, staff and faculty at a private university in California was recruited. A wide range of physical activity level was observed and young participants showed higher level of physical activity than early-middle-age, late-middle-age and old participants. Exploratory structural equation modeling (SEM) analysis indicated self-efficacy exerted the strongest impact on physical activity across the four groups of participants. Social support consistently influenced physical activity through its effect on self-efficacy. The influence of physical environment got diluted as participants got older. Other variables including expected benefits from regular physical activity, knowledge and discounting process did not show a reliable influence on physical activity across the four groups. The models provided good model fits and explained 31% to 60% of the variance in physical activity across different groups of participants.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Evaluating social-cognitive measures of motivation in a longitudinal study of people completing New Year's resolutions to exercise

PDF

Sources of stability and change in the trajectory of openness to experience across the lifespan

PDF

Multivariate biometric modeling among multiple traits across different raters in child externalizing behavior studies

PDF

Influence of sad mood and old age schema on older adults' physical symptoms processing

PDF

Physical activity interventions to reduce rates of sedentary behavior among university employees: a promising practice study

PDF

International sex and age differences in physical function and disability

PDF

A functional use of response time data in cognitive assessment

PDF

Objectively measured physical activity and related factors in minority youth

PDF

Neural correlates of learning and reversal of approach versus withdraw responses to in- and out-group individuals

PDF

Life transitions, leisure activity engagement, and cognition among older adults

PDF

Investigating daily effects of activity space-based built environment exposures on physical activity behaviors in Hispanic women during pregnancy and early postpartum

PDF

An examination of in-school and online protective factors for adolescent trajectories of online victimization and harassment over time

PDF

The effects of physical functioning and public stigma on psychological distress as mediated by cognitive and social factors among Korean survivors of childhood cancer

PDF

Aligning digital technology to support motivation, physical activity, and sports

PDF

Effects of the perceived and objectively assessed environment on physical activity in adults and children

PDF

Social interaction moderates enjoyment and perception of physical activity during exergame play in young adults with autism spectrum disorders

PDF

The effects of bilingual acceptance and commitment training (ACT) on exercise in bilingual international university students

PDF

An examination of factors that potentially impact anxiety levels in high school students in an international school

PDF

Insula activity during safe-sex decision-making in sexually risky men suggests negative urgency and fear of rejection drives risky sexual behavior

PDF

Engaging together: exploring the peer effects of civic engagement

Asset Metadata

Creator Li, Jian (author)

Core Title An exploratory study of physical activity across different adult groups

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 03/30/2010

Defense Date 11/23/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,physical activity, adult

Place Name California (states)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Walsh, David A. (committee chair), Callaghan, John L. (committee member), Manis, Franklin R. (committee member), McArdle, John J. (committee member), Read, Stephen J. (committee member)

Creator Email bigcarcar@hotmail.com,lijian@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2887

Unique identifier UC1425501

Identifier etd-Li-3411 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-291751 (legacy record id),usctheses-m2887 (legacy record id)

Legacy Identifier etd-Li-3411.pdf

Dmrecord 291751

Document Type Dissertation

Rights Li, Jian

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu