Self-monitoring and selected measures of aerobic and strength fitness and short-term exercise attendence.
Anshel, Mark H. ; Seipel, Scott J.
While many individuals tend to acknowledge the benefits of exercise
and initiate fitness programs, relatively few people persist at their
exercise regimen. Approximately 60-70% of adults who begin an exercise
program will quit within 6-9 months, despite the widespread belief (82%)
that exercise is beneficial to good health (King, 1994). Nonadherence
with most organized exercise programs ranges from 20% to 90% in the U.S.
(Marcus, King, Bock, Borelli, & Clark, 1998), a figure that is
similar in the United Kingdom (Biddle & Mutrie, 2001). It is
important, therefore, that researchers examine the effectiveness of
various strategies and interventions that enhance exercise participation
and adherence that markedly affect quality of life. Thus, a particular
challenge to researchers and practitioners is developing interventions
that promote exercise adherence. The concept of adherence, however, has
different meanings in the extant literature.
For the purposes of this study, it is important to clarify the
definition of adherence. Rand and Weeks (1998) broadly define adherence
as "the degree to which patient behaviors coincide with the
clinical recommendations of health care providers" (p. 115). King
(1994) defines adherence as "the level of participation achieved in
a behavioral regimen once the individual has agreed to undertake
it" (p. 186). Other definitions of adherence include sticking to or
faithfully conforming to a standard of behavior in order to meet some
goal, and long-term behavior changes associated with preventing
undesirable symptoms or outcomes (Haynes, 2001). Thus, for the present
study, adherence was defined as the person's decision to maintain
his or her participation in an 8-week exercise program after the
participant agreed to undertake it.
There are many reasons for non-adherence to exercise programs.
These include sustaining an injury, self-consciousness about one's
appearance in an exercise facility, engaging in an exercise activity
that one finds overly strenuous, failure to quickly meet (often
unrealistic) goals, the absence of an exercise facility located near
home or work, job-related travel, physical and mental fatigue, lack of
interest, poor weather, family demands, and perceived lack of time
(Biddle, Fox, Boutcher, & Faulkner, 2000; Biddle & Mutrie, 2001;
Sallis & Owen, 1999). Other reasons include lack of instruction,
perceived lack of fitness improvement, and the lack of social support
(Anshel, Reeves, & Roth, 2003; King, 1994; Lox, Martin, &
Petruzzello, 2003). Taken together, these reasons for non-adherence
suggest that many individuals do not have the requisite knowledge and
skills to perform exercise tasks successfully (Anshel, 2006). It is
plausible to surmise, therefore, that novice exercisers require
emotional and instructional support to sustain their initial motivation
in maintaining their exercise program (Buckworth & Dishman, 2002).
Strategies that significantly improve compliance with exercise
programs are referred to as relapse prevention (Marlatt & George,
1998). Relapse is an individual's failure to permanently change an
undesirable behavior, such as returning to a sedentary lifestyle and not
carrying out (or continuing) a prescribed exercise program. Exercise
relapses are an important component of exercise dropout. Lox and
colleagues (2003) concluded "a single lapse may lead an individual
to believe that all hope of behavior change is lost, resulting in full
relapse, termed the abstinence violation effect" (p. 99). Relapse
prevention consists of "a self-control program designed to help
individuals to anticipate and cope with the problem of relapse in the
habit-change process" (Marlatt & George, 1998, p. 33). For
example, in their review of related literature, Buckworth and Dishman
(2002) concluded that more research is needed to examine the
effectiveness of conceptually-based interventions addressing the problem
of exercise adherence and relapse prevention. One way to reduce relapse
is to enhance the patient's feelings of self-control (Marcus et
al., 1998, 2002), which is the primary goal of a process called
self-regulation (SR).
Self-regulation is defined as self-generated thoughts, feelings,
and actions that are planned and cyclically adapted to the attainment of
personal goals" (Zimmerman, 2000, p. 14). To Zimmerman, SR is
cyclical because "the feedback from prior performance is used to
make adjustments during current efforts" (p. 14). According to the
Encyclopedia of Health Psychology (Christensen, Martin, & Smyth,
2004), self-regulation theory posits that "behavior is guided by a
motivational system of setting goals, developing and enacting strategies
to achieve those goals, evaluating progress, and revising goals and
actions accordingly" (p. 263-264). Along these lines, Zimmerman
(2000) categorizes self-regulation as behavioral, environmental, and
covert. Behavioral SR entails self-observing and adjusting performance
(e.g., improving exercise technique). Environmental SR refers to
observing and adjusting environmental conditions or outcomes such as
selecting situational (e.g., time of day) and environmental conditions
(e.g., home versus exercise facility; exercising with friends or alone)
that meet the exerciser's needs. Covert SR involves monitoring and
adjusting cognitive and affective states (e.g., positive self-talk,
imagery, "psyching up"). Each of these forms of SR is relevant
to the current study.
SR is central to exercise adherence. According to Biddle and Mutrie
(2001), "In addition to exercise being reinforcing through positive
psychological outcomes, maintenance of exercise is likely to be
enhanced, at least for some people, through the operation of
self-regulatory strategies and skills" (p. 142). Techniques
designed to improve self-regulatory functioning have been effective in
various performance domains. For example, with respect to sport
performance, SR improves and maintains motivation, and is generally
characteristic of elite athletes (Anshel, 1995; Anshel & Porter,
1996; Anshel, Porter, & Hass, 1996). In promoting desirable
health-related behaviors, adherence to effective self-regulatory
functioning is more likely if the person can effectively carry out
predetermined SR behaviors that, in turn, lead to desirable outcomes
(Clark& Becker, 1998). Exercise adherence is unlikely without
effective self-regulation (Biddle & Mutrie, 2001; Dishman, 1994).
Dishman suggests, "self-regulatory skills and interventions such as
relapse prevention seem necessary for individuals to maintain or resume
a physical activity pattern" (p. 16). According to Biddle and
Mutrie (2001), one popular form of self-regulation for improving health
and performance is called self-monitoring.
Self-monitoring (SM) is defined as "the systematic observation
and recording of target behaviors" (Baker & Kirschenbaum, 1993,
p. 377). To Christensen and colleagues (2004), SM "is a technique
used for health behavior assessment and intervention, which involves
observing and recording information regarding one or more health
behaviors" (p. 263). Types of data that are recorded include target
behaviors (e.g., duration, frequency), contextual factors that
surrounded the target behavior (e.g., time, setting, antecedents,
consequences), and subjective information (e.g., mood, thoughts), each
of which may be recorded quantitatively or qualitatively to understand
the factors that promote desirable outcomes.
Self-monitoring has been shown to be an effective strategy for
achieving behavioral change and improving performance in a variety of
health and psychomotor domains. Kirschenbaum and his colleagues, for
example, have successfully applied SM to improve weight control (e.g.,
Baker & Kirschenbaum, 1993, 1998; Boutelle & Kirschenbaum,
1998). Endler and Kocovsky (2000) concluded, "vigorous
self-monitoring is an important aspect for individuals with some health
problems.., such as treatment for type 2 diabetes and exercise" (p.
580). According to the authors, SM is effective if the individual has
control over some aspect of his or her illness or behavior in reaching a
goal. SM also can bolster self-efficacy.
In other studies examining the efficacy of SM on exercise
compliance, Izawa et al. (2005) found that self-monitoring markedly
improved self-efficacy for physical activity and exercise adherence
among cardiac rehabilitation patients. In a sport study of freestyle figure skating performance, Hume, Martin, Gonzelez, Cracklen, and
Genthon (1985) found that using a self-monitoring checklist, a strategy
used in the present study, "effectively increased the frequency of
jumps and spins performed in a 45-minute free skating session by over
90%, increased the number of times a skater practiced her routine to
music, decreased the time spent in off-task behaviors, led to
improvement in skating, and received positive evaluations from the coach
..." (pp. 343-344).
The effective use of SM techniques on exercise adherence among
healthy participants also has been studied. In an Australian study,
Weber and Wertheim (1989) examined the efficacy of a SM-only treatment
as opposed to SM-plus-attention and a no-SM control group on exercise
adherence over three months (12 weeks). Fitness program attendance at a
community gymnasium served as the criterion for exercise adherence.
Exercise adherence was significantly poorer in the control group, as
compared to both SM groups. However, the SM-alone group demonstrated
statistically superior adherence to the other groups, whereas the
SM-plus-attention group did not markedly differ from the SM-only and
control groups across the 12-week period. The lack of superior adherence
due to additional staff attention may have been due to the absence of
controlling staff behavior. The researchers surmised that prior to the
study it is likely that regular gym staff were "highly attentive
prior to the intervention" to all participants (p. 112), including
those in the control group, thus possibly compromising any potential
benefit of the experimental condition.
Noland (1989) examined the effect of SM and verbal reinforcement on
exercise adherence in an unsupervised program. Participants in the SM
group kept written records of their exercise behavior, while the
reinforcement group verbally reported their exercise behavior to another
person who periodically administered rewards. All participants were
given instruction on proper exercise techniques and then were asked to
exercise on their own for 18 weeks. While the reinforcement group
improved their predicted max [VO.sub.2] by 11% as compared to the SM and
control groups, pre- and post-treatment tests revealed no significant
treatment or interaction effects. Both SM and reinforcement groups
reported a significantly higher frequency of exercise per week than
controls.
While it appears that self-monitoring strategies strongly influence
exercise behavior, the related literature is replete with limitations.
These include the lack of recording baseline attendance, the absence of
scheduled personal coaching, and failure to measure fitness outcomes as
a function of the treatment. One particular shortcoming of the Weber and
Wertheim (1989) study, common in the extant literature, was the
self-monitoring treatment. Based on the researchers' description,
self-monitoring consisted of recording attendance, attitude and emotion
about their exercise participation, and completion of exercise tasks.
Apparently, providing instruction on proper exercise techniques and the
use of proper strategies before, during, and immediately after the
exercise session was not conducted.
Limitations of Noland's (1989) study also are consistent with
the vast majority of previous SM investigations in promoting exercise
adherence. These include the absence of an SM checklist (i.e.,
systematic observation and recording of target behaviors) in exercise
settings, the limited use of quantitative fitness data (e.g., fitness
scores, lipids profile), and not including performance coaching as a way
to provide instruction and performance feedback.
SM has many proven advantages as a vehicle for promoting
self-regulation and adherence to exercise. For example, SM provides an
opportunity for inexpensive home-based fitness programs that would
foster long-term adherence rates (Perri, et al., 2002). Ostensibly, the
result of using SM includes positive reinforcement of properly executed
exercise, increased perceived competence, and consequent improved
long-term adherence. As Kirschenbaum (I 987a) contends, "positivity
of expectancies, positivity of self-monitoring, and task mastery
interact to affect self-regulated performance" (p. S108). Yet,
there is a paucity of research on the potentially beneficial effects of
SM on exercise adherence, with apparently no study of the combined
effects of SM and personal coaching on adherence, among unfit persons.
The current study is an extension of the current exercise adherence
literature in addressing previous shortcomings to the extant related
literature. It is important to examine a more comprehensive SM program
in which participants monitor all aspects of the self-regulatory process
related to their exercise participation in addition to receiving
personal coaching in an effort to promote exercise adherence. In
addition, the SM intervention might be more effective if participants
are provided with instruction and encouragement throughout their
involvement. Whether the sources of instruction and encouragement are
derived from a checklist or from personal coaching was one objective of
this study. Thus, the purpose of this study was to examine the
effectiveness of a comprehensive intervention consisting of
self-monitoring and personal coaching on selected measures of fitness
and exercise adherence among university staff and faculty members. It
was hypothesized that the experimental group, which used a
self-monitoring technique, would experience significantly superior
improved fitness and exercise adherence rates from pre- to
post-intervention as compared to a group that included personal coaching
only, and no self-monitoring.
Method
Participants
Participants were recruited through the e-mail system of a
university campus located in the southeastern U.S., and faculty
newsletter. Information included a description of the program, the
general purpose of the study, and meeting the criteria of being in good
health, confirmed by a note from their personal physician, and no
current involvement in an exercise program. In response to these
recruiting attempts, 103 faculty and staff expressed initial interest in
the program. Of these, 65 individuals engaged in a pre-intervention
seminar and subsequent fitness testing. Thus, the study consisted of 65
staff and faculty, 23 men and 42 women, 62 Caucasian and three African
Americans, who were employed full time at a university in the southeast
U.S. and volunteered to engage in the study. Participants ranged in age
from 24 to 61 yrs. (M= 44.6 yr., SD = 2.98) and were in generally good
health.
Self-Monitoring Exerciser Checklist
The Exerciser Checklist (EC) consisted of a 60-item self-monitoring
strategy to primarily serve as an instructional tool in helping
participants to learn the proper preparation, performance, and mental
skills for engaging in exercise. Sallis, Haskell, Fortmann, Vranizan,
Taylor, and Solomon (1986) studied the factors that best predicted
exercise adherence among 1400 adults. Among the highest predictors of
exercise adherence was the use of behavioral skills, that is, proper use
of exercise techniques prior to, during, and immediately following
exercise sessions. The techniques listed in the current EC consisted of
behavioral exercise strategies endorsed and published by the American
College of Sports Medicine Resource Manual (ACSM, 2006).
The checklist was divided into five segments: (a) lifestyle habits,
(b) day of exercise, (c) pre-exercise activity (at the exercise venue),
(d) during exercise session, and (e) after exercise session. The
checklist (e.g., "I think positive thoughts about my exercise
program immediately before exercising") consisted of indicating the
extent to which the participant followed each guideline, ranging from l
(not at all like me) to 5 (very much like me). Based on the
clients' responses to these items, coaches provided instruction,
remediation, and feedback. The purpose of this checklist was to
encourage clients to adhere to pre-determined targeted exercise
behaviors. More specifically, the checklist measured participants'
levels of agreement concerning whether they effectively engaged in
positive self-regulatory behaviors related to exercise. We followed the
recommendations of Loehr and Schwartz (2003) and Rollnick, Mason, and
Butler (1999), who suggest that changing health behavior through
self-regulation is more effective if routines are developed that support
the target behaviors, in this case, exercise.
As indicated earlier, the EC was used for clients to promote the
self-management of health-related routines and proper exercise habits,
for optimal self-monitoring effectiveness (Creer & Holroyd, 1997).
The authors suggest five distinct steps, each of which was used in the
present study and conducted each week with the participant's
performance coach. First, participants must be able to detect
"significant" internal changes, including self-evaluation and
self-observation of their actions, and record this information. Second,
standards must be established (e.g., the EC used in this study) to
permit the participants to evaluate the data they collect and process
about themselves. Third, participants must be able to evaluate, and make
judgments about the data they obtain. What areas need change or
improvement? What is currently effective and should continue?
The fourth step was for participants to learn to evaluate any
changes prior to (i.e., antecedent conditions), or during performance of
the target behaviors, and the consequences of theft actions. Finally,
contextual factors and conditions must be considered about assessing the
effectiveness of the participant's own management of targeted
behavior. In this study, this might include selecting a proper and
comfortable exercise venue or program, obtaining social support (e.g.,
exercising with a friend or obtaining peer or spousal support),
detecting desirable changes in feelings and attitudes toward exercise
(e.g., less perceived exertion, more enjoyment), and developing routines
that promote the proper management of thoughts, emotions, and behaviors
that accompany an exercise behavior.
Checklist content has differed among related past studies. For
example, in one early study using SM to promote socially appropriate
expressive behavior (Snyder, 1974), checklist items included
"openly expresses his true inner feelings," "is good at
learning what is social appropriate in new situations," and
"often appears to lack deep emotions" (p. 530). More recently,
Baker and Kirschenbaum (1993) asked participants to recall all food
consumed during the week and to count the calories in these foods. Their
SM booklet consisted of blank pages with columns for time, food, and
calories. Boutelle and Kirschenbaum (1998) required participants in a
weight control study to monitor their daily food
intake. In a relatively rare SM study on exercise behavior, Noland
(1989) asked participants to keep written records of their exercise
behavior (i.e., identifying activities).
In a motor behavior study, Martin and Anshel (1995) used SM to note
performance trials that were successful (positive SM) or unsuccessful
(negative SM) using a rating form. While each of these studies showed
that SM favorably affected predicted behavioral patterns, none of these
studies provided instruction. Of these studies, only Martin and
Anshel's study included a condition that monitored improved
performance. The checklist used in the present study, contrary to
previous checklists, requires participants to monitor their lifestyle
and perform certain tasks and have specific types of thoughts or
emotions before, during, and after their exercise session for the
purpose of promoting proper exercise technique. The primary function of
the current checklist was to provide instruction and monitor
improvement, requiring an extensive list of items.
Fitness Testing Equipment
Participants were given four fitness tests prior to and immediately
following the intervention: (a) Single-Stage Treadmill Test to measure
cardiovascular fitness, (b) push-ups to measure upper body strength, (c)
blood pressure, and (d) skinfold. Exercise adherence formed a fifth
dependent measure. Each of these measures has been reported extensively
in the exercise intervention and adherence literature as indicators of
improved physical fitness (Dishman, 1994; Dominick & Morey, 2006).
Body composition. Body composition was measured using a Lange
skinfold caliper. The participant's performance coach who was
trained to measure skinfolds performed a seven-site assessment to assure
accuracy and consistency (Pollock & Jackson, 1984). Body fat
percentage was calculated from estimates of body density using the Siri
(1961) equation.
Blood pressure. Blood pressure was assessed with a WelchAllyn
automatic blood pressure machine, Model CE0050, 420 series. If the
participant's arm circumference was too large for the automatic
blood pressure machine, a skilled technician assessed blood pressure
with a large cuff manual blood pressure cuff. All participants were
measured in the seated position using the right arm.
Cardiovascular fitness. Estimated V[O.sub.2] max was assessed by
the use of the Single-Stage Treadmill Test. Each test was performed on a
Quinton Treadmill, Model number Q55 using standard protocol (Ebbeling,
Ward, Puleo, Widrick, & Rippe, 199 I). Participants were asked not
to hold onto the handrails during the test unless absolutely necessary.
Heart rate was manually palpated for a 10-see. period during the final
minute of the test for use in the prediction equation.
Muscular strength. Strength was ascertained by the number of
push-ups the participant could perform (ASCM, 2001). If the participant
complained of arm or shoulder pain, they performed sit-ups.
Exercise adherence. Adherence to their prescribed fitness program
was based on the number of aerobic and strength training sessions in
which the client engaged each week at any location (e.g., fitness
facility, home equipment, outdoors). The client's performance coach
recorded these data either online or by phone at the end of each week.
Procedure and Intervention
The study consisted of six stages: (1) attending a 3-hour seminar
(including a workbook, lecture, DVD, and group member interaction) that
provided motivational and inspirational content, followed by being
assigned a performance coach, (2) receiving pre-intervention blood and
fitness tests; (3) receiving an exercise program prescription based on
the individual's test data, (4) engaging in an 8-week exercise
program (during which time adherence data were collected), (5) meeting
weekly with their coach, and (6) receiving post-intervention blood and
fitness tests.
Prior to the seminar, these individuals were randomly assigned to
one of two groups, self-monitoring checklist/performance coaching or
performance coaching only (control), and were then assigned to a
performance coach (n = 9 or 10 clients per coach). There were two
criteria for being assigned a particular coach, the individual's
indication they preferred a certain gender and their availability (days
of the week and times of day) to meet with their coach. For instance,
coaches and participants indicated a preferred time of day they were
available to interact and meet the requirements of the intervention.
Coaches were then randomly assigned to one of the two groups,
experimental or control.
The job of the performance coach for checklist participants was:
(1) to conduct pre- and post-intervention fitness tests, (2) to provide
clients with their exercise prescription (which changed during the
program in response to improved fitness), (3) to instruct each client on
proper exercise techniques, including strength training, (4) to monitor
their client's progress, and for the experimental group only, (5)
to instruct each client on completing the Exerciser Checklist (EC) and
to review the checklist in attempting to improve their exercise skills
and checklist scores. It is important to note that the checklist
included items that went beyond exercise behavior. It also included
selected changes in habits that were to be applied before and after the
exercise session in addition to changes in exercise behavior.
As indicated earlier, it was thought that exercise effectiveness
and adherence were more likely if checklist content incorporated
"support behaviors" consisting of routines that were an
integral part of the exercise experience, a strategy suggested by Loehr
and Schwartz (2003) and Rollnick et al. (1999). In addition, checklists
that provide information and behavioral guidance both prior to and
following the actual target behaviors is consistent with the
preparation, action, and maintenance stages of the transtheoretical
model of behavior change, as explained by Prochaska and Marcus (1994)
and Biddle and Mutrie (2001 ). The authors describe the preparation
stage as having a plan of action, obtaining health education, and
discussions and consultations with health professionals or exercise
consultants. The action stage concerns engaging in regular exercise and
applying the knowledge obtained in the preparation stage. The
maintenance stage concerns the use of support groups and co-participants
in promoting exercise adherence. By including behavioral routines that
were expected to occur, during, and following exercise sessions, the
present checklist addressed each of these stages.
Prior to the study, checklist content was reviewed with each coach,
including ways to review their participant's adherence to each item
and how to instruct participants who recorded relatively "low"
scores. Coaches for the non-checklist group engaged in the same tasks
with the exception of completing and reviewing the checklist. Coaches in
the EC group met jointly each week with one of the study's
investigators for in-service training to review the study's
protocol, ensuring adherence to the collection of EC data, review
information about EC content, and to answer any questions. Coaches not
using the EC also met weekly to discuss similar information as the EC
coaches, but avoided discussing EC content. It is important to note the
coaches of both groups shared an equal degree of commitment to help
participants complete the program and to improve strength and aerobic
fitness.
Participants in both groups were asked to engage in aerobic
exercise a minimum of three times per week at a time of day, days of the
week, and location of their own choosing. Strength training was
prescribed a minimum of twice per week. Both aerobic and strength
exercise was emphasized, as was the development of an exercise routine.
Performance coaches contacted their clients by phone or made
appointments to see them in person once per week to monitor their
progress and to determine if they had questions. To promote exercise
adherence, clients were allowed to exercise at times and in venues they
felt most comfortable (e.g., home equipment, campus fitness center, off
campus fitness club).
Participants completed the EC on his or her own time and submitted
it to their coach at the end of each week, either in person or as an
e-mail attachment. Clients were given a new checklist during the weekly
meeting with his or her coach, to be returned the following week. The
coach reviewed the client's responses and then provided instruction
based on areas that needed additional attention (i.e., poorerseores on
the checklist). The goal of using the checklist was to significantly
improve scores for each of the segments, a result that would indicate
improved exercise skill and performance, receive additional instruction
on fitness techniques and mental skills, and to receive positive
information feedback on successfully completing checklist items. In
addition, at the end of each of the eight weeks, clients from both
groups indicated the extent to which they separately adhered to their
strength and aerobic exercise regimens (e.g., exercising 0, 1, 2, or 3
times the previous week) each week. After the eighth week, illness
posttests were conducted. As indicated earlier, the no-checklist group
experienced the usual program, including performance coaching, but did
not use the checklist.
To enhance adherence, the exercise program was tailored to meet the
individual's needs and preferences. Empirical evidence from
contemporary studies have confirmed that tailoring health and
exercise-related messages to meet individual needs are more effective in
motivating people to engage in self-protective action than nontailored
messages (Dijkstra & De Vries, 1999). For example, although the
exercise programs were formally held at the campus's fitness
center, participants were given the choice to exercise at home, at an
off-campus fitness facility, at their home, or anywhere else, as long as
their exercise regimen included both cardiovascular and strength
components. Exercise content was personalized to reflect the
individual's prescription that followed fitness testing, and
included both aerobic and strength conditioning.
Data Considerations
The data to determine exercise adherence was recorded as the
percentage of the prescribed three exercise sessions completed weekly
during each of the intervention's eight weeks. Thus, the inherent
distribution of the data is discrete, ranging from 0% to 100% in
multiples of 33.3%. Due to the time frame in which the data were
collected and the relatively uneven schedule maintained by many
participants during that time frame, adherence on a week-to-week basis
was highly variable. For instance, a client may have scheduled a
vacation during a given week and, therefore, was unable to train,
although participation was at 100% on other weeks. This issue, while
common in field experimental research (Thomas, Nelson, & Silverman,
2005), introduced substantial variability in the adherence data, and
became confounded with variability based on the participant's
ability to adhere to the regimen. A weakness inherent to this study,
then, was the failure to completely isolate these sources of variability
and their magnitude. However, because much of the variability occurred
on a week-to-week basis, a simple solution was to extend the period of
time under consideration; adherence becomes substantially less variable
if it is recorded bi-weekly. Thus, to control for the high variability
factor in this study, an objective aggregate measure of adherence
consisted of comparing the time periods between the intervention's
initial four weeks and the final four weeks. This conversion also
created a relatively stable measure (providing a semi-Gaussian
distribution and homoscedasticity) without sacrificing or eliminating
the underlying data variation.
Results
Changes in Fitness
Four tests were transacted pre- and post-test to determine the
effect of the experimental and control groups on fitness. Descriptive
statistics for the pre- and post-test results for the Astrand sub-max
V[O.sub.2] test, push-ups/sit-ups (strength), systolic and diastolic
blood pressure, and skinfold are shown for each group in Table 1.
Scoring for the V[O.sub.2] test, systolic and diastolic blood pressure,
and skinfold was altered to reflect percentage improvement over the time
of the study. This was because these scores (in scale) would likely
improve dependent on the starting value, and improvements in the mean
would be applicable only to "average" participants. Scoring
for strength was problematic due to the two different approaches to the
measurement and the highly variable base levels established by the
participants. As such, scoring for changes in strength was reduced in
scale to whether improvement was shown over the duration of the study (a
Bernoulli variable). This transformation allowed pushups and sit-ups to
be combined into one scale, albeit at the loss of some information.
Descriptive statistics on the resulting improvement gains are shown in
Tables 1 and 2.
The analysis of changes in fitness typically would include
multivariate techniques for comparisons between groups and genders.
However, much of the data, both in the raw scale and the percent
improvement are not homoscedastic, and follow very non-normal
distributions. No interpretable transformation of the data is known that
would have allowed all of the fitness response variables to be analyzed
using a single multivariate technique. Therefore, fitness measures were
analyzed with the more robust ANOVA procedure, test of proportions, and
the nonparametric Mann-Whitney (MW) test.
The percentage showing strength gain, as measured pretest to
posttest by whether participants showed an increase in pushups or
sit-ups, was significantly higher in the experimental group as compared
to the control group (p = .011). All (100%) experimental group
participants demonstrated increased strength as compared to 83.3% in the
control group, a sample improvement of 16.7%.
A repeated measures (pretest to posttest) ANOVA revealed no
statistically significant improvement in the mean percentage loss in the
experimental group over the control group (p = .090) for changes in
skinfold measures. Differences between genders and the interaction of
gender and group were also not significant (p values of .093 and .433,
respectively). The primary concern in this analysis was a lack of
homoscedasticity (Levene's test of equal variance; p = .001) given
the smaller number of male observations in the experimental group. The F
test statistic of the ANOVA would be considered liberal in this
situation (Stevens, 1996), giving more evidence to the lack of
statistical significance.
For the average percentage gain (pretest to posttest) in the
sub-max V[O.sub.2] test, the experimental group experienced significant
improvement (p = .04) over the control group. Differences between
genders, and the interaction of gender and exercise group, using ANOVA,
were not statistically significant (ps = .92 and .31, respectively). A
marginal result from the test of equal variances among groups
(Levene's test; p = .06) required a subsequent nonparametric MW
test. Findings from the nonparametric test showed a significant
improvement (p = .01) in the experimental group. Note that one
observation was removed from the control group due to a missing value.
For blood pressure, separate repeated measures ANOVAs were used to
compare the effect of group and gender on the systolic and diastolic measures. No significant improvement in the systolic blood pressure of
the SM group over the control group was noted (p = .09). Differences in
gender and interaction were also not significant (p values of .63 and
.27, respectively). Concerning diastolic blood pressure, no significant
improvement was found in the experimental (SM) group compared to the
control group (p =. 11). Tests of gender and interaction effects also
proved not significant (p values of .99 and. 13, respectively).
To test for possible individual differences in coaching
effectiveness, analyses identical as reported earlier were performed at
the coach level. For skinfold, significant differences were found among
the coaches (p = .000). Analyses of the other fitness measures provided
no significant evidence of coach effect. P-values for sub-max V[O.sub.2]
and blood pressure were .06 and. 86, respectively.
Adherence
The means, standard deviations, skewness, and kurtosis values of
aerobic and strength adherence for the control and experimental groups
for the eight-week intervention and the four-week aggregations are shown
in Table 3. Individual observations exhibited clear nonnormal patterns
consistent with the level of measurement. The data indicated client
persistence in maintaining three aerobic training sessions per week. The
vast majority (74%) of all respondents indicated that they participated
in all three (100%) of the prescribed aerobic training sessions in the
initial week, while only 46% maintained that same level for strength
training. Because of this relatively high adherence to aerobic training,
data on aerobic adherence was negatively skewed and highly kurtotic at
times. The four-week aggregations were not as susceptible to these
problems. As they had less of a discrete distribution and were an
aggregate calculation, their distributions were more Gaussian in form
and had similar variance.
As shown in Figures 1 and 2, the weekly data for the control group
appears to follow a pattern--an initial increase, followed by a
prolonged decrease, in both aerobic and strength adherence. For example,
mean adherence of the control group for aerobic exercise started at 85%,
increased steadily to 94% by the fourth week, and then diminished
rapidly to 62% by the final (eighth) week. Mean adherence for strength
training of the control group followed a similar pattern. An initial 58%
mean adherence rate was followed by a rise to approximately 80% in the
third and fourth weeks, with a subsequent drop to 48% mean adherence by
the eighth week.
These patterns were less clear in the experimental group (see
Figures 1 and 2). Mean adherence for aerobic exercise began at 89%,
increased to 92%, but then fluctuated in a somewhat random pattern
between 78% and 85%. Similarly, mean adherence for strength training of
the experimental group commenced at 53%, increased to 71%, but then
oscillated between 58% and 70%. Although there was generally a slight
rise in mean adherence for both aerobic and strength training, the
subsequent reduction in adherence was not nearly as notable and the data
was generally much more stable.
Changes in Adherence
The correlation between the change in adherence (from the initial
to the final four week period) for strength and aerobic training was
moderate (r = .47, p = .000), indicating that individuals who reduced
their adherence in one area of training tended to reduce their adherence
in the other area. Based on the magnitude and significance of this
correlation, multivariate analysis of variance (MANOVA) was utilized to
test the significance of changes in adherence between groups.
Assumptions of MANOVA to control for Type I error and for optimal
power include independent observations, a multivariate normal
distribution of values within groups, and equal population covariance matrices for the dependent variables. Each of these assumptions was
evaluated in the dataset to determine the level of compliance. It was
possible that group effects in the study could have been due to the
effects of individual coach characteristics or the quality of
coach-client relationships, and a lack of independence among
observations within each group is possible. To evaluate this likelihood,
the performance coach for each group was nested, and univariate analyses
of variance (ANOVAs) were performed on changes in aerobic and strength
adherence to determine if at least partial changes in adherence could be
explained by the coach's training skills or their relationship with
respective clients.
Despite the nesting of the coach with group, no significant
differences were found among coaches for either the change in aerobic
adherence (p = .29) or the change in strength adherence (p =.93). Gender
and the interaction of gander and coach also were not significant (p
values of .75 and .06 for gender and interaction, respectively, for
changes in aerobic adherence, and .60 and .33, respectively, for changes
in strength adherence). These results provide partial evidence
supporting the assumption of independence of observations.
MANOVA is relatively robust with respect to Type I error for
deviations from multivariate normality (e.g., Mardia, 1971; Olson,
1974). Univariate and bivariate analyses of the variables within groups
showed no marked deviation from normality. Tests of homogeneity of
population covariances were performed using Box's M test (1949).
The significance level for the analysis was .231, indicating no
significant diversion from this assumption.
Results of a MANOVA using a full factorial model (i.e., group and
gender) indicated that the group variable was a significant determinant of the mean change in aerobic and strength adherence (Wilk's Lambda
= 4.82, p = .011). Neither gender nor the interaction of gender and
group were significant (p values of .59 and. 19, respectively). Post hoc analysis using marginal means indicated that the experimental group
maintained their adherence to aerobic exercise at an average rate of
23.6 percentage points higher (p = .001) than the control group.
Controlling for gender, members of the experimental group averaged an
8.9% increase in adherence as compared to the 14.8% average drop in
adherence for the control group. The 95% one-sided confidence interval for the mean improvement in aerobic adherence for the experimental group
over the control group is 10.8% and greater. In addition, there was a
statistically significant reduction in mean aerobic adherence for the
control group (p = .001), however, there was no significant change in
mean aerobic adherence for the experimental group (p = .913).
The comparative post hoc analysis for adherence to the
strength-training regimen also indicated significant differences. On
average, individuals in the experimental group maintained their
adherence to the strength program at a rate 15.8 percentage points
higher than those in the control group (p = .033). Controlling for
gender, members of the experimental group averaged a .9% increase in
adherence to strength training, as compared to the 14.9% average drop in
adherence in the control group. The 95% one-sided confidence interval
for the mean adherence improvement in the experimental group was 1.7%
and greater. Furthermore, there was a statistically significant
reduction in mean strength adherence over time for the control group (p
= .000). There was no significant increase in mean strength adherence
for the experimental group (p = .483).
Coach as a Mediating Variable
As indicated earlier, participants in both groups received
performance coaching. Because relationships were established with their
performance coach, we conducted post-intervention analyses to examine
the extent to which this relationship, combined with selected
characteristics of their coaches, might have influenced the exercise
outcomes. It is possible that the performance coach, who administered
the fitness tests to his or her clients, affected the test results
(Weber & Wertheim, 1989). To preempt that concern, additional
analyses of each of the fitness tests were performed using coach as the
factor.
Ideally, a coach-treatment interaction would be explored. However,
because the performance coach factor is nested within the treatment,
exploring a coach-treatment interaction was not possible. Therefore, the
exploration of coach as a single factor was examined to determine
whether some coaches exhibited a bias in scoring. The results indicated
no significant differences between coaches for changes in adherence for
either aerobic or strength programs (ps >.05). The gender main effect
and the gender by coach interaction also were nonsignificant for changes
in both aerobic and strength adherence. Thus, it was apparent that
observations were independent of coaching effects, and more likely
reflected effects of the checklist treatment.
To determine if the performance coach, who administered the fitness
tests, influenced test results we analyzed each of the fitness tests
using coach as the factor. Because the performance coach was nested
within each group, significant f'mdings in these analyses could not
conclusively indicate a coach effect. These factors were confounding,
rendering a cause and effect relationship impossible. Significant
differences between the coaches were found for skinfold measures (p =
.000), however, no evidence of a significant effect of coach for the
other fitness measures was found (ps = .06, .73, and .85 for sub-max
V[O.sub.2], strength, and blood pressure, respectively). These results
confirm that the in-service training to promote high quality coaching
lead to reasonably consistent coach-client relationships, and that
performance outcomes were a function of the treatment, not the quality
of the coach's performance.
[FIGURE OMITTED]
Discussion
The purpose of this study was to examine the effect of a
self-monitoring (exercise checklist) technique, combined with
performance coaching, on selected measures of fitness and exercise
adherence over an eight-week program among university staff and faculty.
It was hypothesized that the experimental group, that is, participants
who used self-monitoring in combination with personal coaching, would
experience significantly superior levels of fitness and exercise
adherence as opposed to the control group who also received coaching,
but without self-monitoring. The results strongly supported this
prediction.
Four fitness measures were obtained immediately prior to and
following the intervention. These were: strength, cardiovascular
fitness, skinfold (percent body fat), and blood pressure. Change scores
for strength, as measured by the percentage that showed improvement in
pushups or sit-ups, indicated significant improvement in the
experimental group, as compared to the control group. The average
increase in fitness for exercisers using the checklist was markedly
higher as compared to the non-checklist group. No significant gender
effect was found.
[FIGURE 2 OMITTED]
With respect to cardiovascular fitness, the experimental group
significantly improved on the sub-max V[O.SUB.2] score from pre- to
post-intervention as compared to the control group for both genders.
However, pretest to posttest comparisons on skinfold and blood pressure
both indicated no statistically significant improvement in the mean
percentage loss of the experimental group over the control group. Gender
differences and the interaction of gender and group also were not
significant for both variables.
Significantly better strength and cardiovascular fitness over the
8-week study, but a lack of similar improvement for skinfold and blood
pressure measures may reflect the minimum timeframe needed for changes
in these latter measures for unfit individuals, irrespective of using a
self-monitoring strategy. Significant changes in percent body fat and
blood pressure can take a minimum of 12 weeks, with an intensity level
commensurate with an aerobic training effect (ACSM, 2006). In addition,
the pre- and post-intervention comparisons indicated normal and
relatively small, nonsignificant, changes readings for both systole and
diastole BPs. Thus, while self-monitoring may have contributed to marked
improvement in strength and cardiovascular fitness, additional exercise
intensity may be needed to significantly alter measures of percent body
fat and blood pressure, particularly among novice, unfit, and overweight
exercisers.
With respect to adherence, most (74%) exercisers participated in
all prescribed aerobic training sessions in the initial week, while less
than half (46%) adhered to their strength training program during this
time. An overview of the adherence data indicated a pattern of initial
increase, followed by a prolonged decrease, in both aerobic and strength
adherence. Exercisers in both groups who reduced their adherence in one
type of exercise tended to reduce their adherence in the other type of
exercise. However, participants in the experimental group tended to
adhere more reliably to their respective training regimens. In
particular, the SM group adhered to their aerobic exercise program
significantly better than the no-SM group. Controlling for gender, the
experimental group improved their adherence, while the control
group's adherence dropped. Reduced adherence in the control group
was statistically significant for aerobic exercise, while aerobic
adherence did not drop measurably for the experimental group.
These findings support Fishbein's (2007) contention that
improving a person's frequency of exercise often requires changing
their attitudes toward exercising. It is plausible to surmise that the
checklist provided instruction and ongoing positive feedback that
reflected the participant's improvement in changing
exercise-related behavior. Positive feedback, which fosters
self-efficacy, is an essential feature of self-monitoring (Biddle &
Mutrie, 2001), an outcome that may explain the superior adherence of the
SM group, at least for aerobic exercise.
The comparative post hoc analysis for adherence to the strength
training regimen provided additional significant differences. On
average, individuals in the experimental group demonstrated superior
adherence to the strength program as compared to the control group.
Furthermore, there was a statistically significant reduction in mean
strength adherence over time for the control group, but no significant
decrease in mean strength adherence for the experimental group.
For both groups, exercise adherence was optimal in the first three
to four weeks and then dropped off substantially. The control group
showed a greater initial increase in adherence for both aerobic and
strength exercise, as compared to the SM group. The SM group, however,
showed a superior pattern of exercise stability over the same time
period. Perhaps one reason for this pattern was the time needed for SM
participants to learn and adjust to the self-monitoring technique.
Changes in behavior require adopting self-regulation routines, a pattern
inherent in the Disconnected Values Model (Anshel & Kang, 2007). The
model posits that a combination of data (e.g., checklists, testing)
combined with persistent instruction and the development of routines
leads to the mastery of fimess techniques and subsequent permanent
health behavior change. Along these lines, Boutelle and Kirschenbaum
(1998) contend that consistent self-monitoring is a skill and is
susceptible to noncompliance. Apparently, the benefits of checklists are
not necessarily experienced in the short term.
A post-study manipulation check was conducted to ascertain the
participant's experiences in using the checklist to determine its
effectiveness, a strategy suggested by Whitley (2002). The results
indicated that more than half of the exercisers in the experimental
group found the checklist time consuming. With time, however, they found
the checklist to be more routine and useful as a source of information
for developing proper exercise-related behavioral patterns. Thus,
requiring exercisers to complete the full 60-item checklist every week
may have proven burdensome to selected participants.
One important implication of this study is that using
self-monitoring checklists may be more effective if they are completed
in segments, at least early in the intervention, that checklist content
could be regularly altered, rather than requiring that the full
checklist be completed each week using identical items (Baker &
Kirschenbaum, 1993). Kirschenbaum (1987b) contends that SM procedures
can become overly taxing for the individual, leading to cognitive or
physical disengagement, either on rational (e.g., fatigue, boredom,
available time, opportunity, or knowledge) or emotional grounds (e.g.,
negative mood state). On the other hand, the proper and consistent use
of the EC should have resulted in advancing the stages of behavioral
change (Prochaska & Marcus, 1994). In addition, according to Biddle
and Mutrie (2001), "self-efficacy underpins different stages,"
(p. 261). Thus, if participants reached the action and maintenance
stages of change, their concomitant improved self-efficacy should have
promoted exercise adherence.
Along these lines, Boekaerts and Niemivirta (2000) contend that
self-regulation processes are multidimensional, complex, and highly
demanding. They require attentional control, physical and emotional
energy, and skill. The individual must alternatively activate, inhibit,
and mediate an array of cognitive, emotional, and physical demands. The
cognitive, emotional, and physical demands of applying SR in exercise
settings may have at least partially accounted for the challenges of
consistently using the EC in this study.
While the data provided coaches with areas for further instruction
and reinforcement, it appears that partial completion of the checklist,
or reducing the checklist content, might serve to improve client
compliance and enhance their attitude toward this task. Certainly,
personal coaching embellishes the effectiveness of self-monitoring
strategies. In their review of related literature, for example, Castro
and King (2002) found that personal coaching conducted by phone,
face-to-face, or in combination resulted in significant improvements on
selected fitness measures. The effect of personal coaching was also
shown in an earlier study (Weber & Wertheim, 1989) in which the
effect of staff attention on exercise adherence in Australia on 51
females was examined. As described earlier, the researchers found that
the combined use of self-monitoring and "staff attention"
resulted in superior exercise adherence, as compared to the control (no
self-monitoring) group. There is little question that the fitness
coaches in the present study markedly contributed the effective use of
the EC. Future research is needed on the effect of combining personal
coaching with exercise checklists.
One challenge in future attempts to promote exercise adherence
using self-regulation techniques is to ensure that the
participant's attitude toward exercise is also enhanced. As
Fishbein (2007) suggests, if we want to improve a person's
frequency of exercise, "then it makes perfectly good sense for me
to try to change their attitudes toward exercising" (p. 287). While
the SM checklist used in this study was not intended to promote nor
measure attitude, it is plausible to surmise that checklists intended to
improve healthy habits and exercise skills will also foster the
person's attitude toward engaging in regular physical activity.
Nevertheless, researchers should include a measure of attitude toward
exercise in future related studies.
One additional area that needs attention in adherence research is
improving our understanding of the criteria that operationally defines
adherence. For example, Rand and Weeks (1998) address a common problem
in the adherence literature--the lack of specific criteria to provide
the "gold standard" for determining adherence. What is
acceptable adherence in one study or for one exerciser might be
non-adherence in another study or for someone else. As indicated
earlier, adherence may be classified as "appropriate,"
"erratic/partial," "ideal," "voluntary,"
or "involuntary." In an expanded definition of adherence,
Abrams, Borrelli, Shadel, et al. (1998) contend that adherence also
should include "the degree to which an interventionist and/or
treatment delivery system adheres to a specific protocol, and/or the
extent to which the treatment was proactively delivered to, and
successfully reached, a specific population" (pp. 140-141).
Researchers in previous studies have usually ignored the different
adherence categories. Partial adherence is an important factor that may
partially explain exercise outcomes, as opposed to dropping out of the
program or study. Future studies should include these variations of
common measures in the context of both short-term and long-term
adherence in order to more fully understand the factors that contribute
to different variations of exercise maintenance as opposed to a more
categorical (i.e., adhere/nonadhere) approach.
References
Abrams, D. B., Borrelli, B., Shadel, W. G., King, T., Bock, B.,
& Niaura, R. (1998). Adherence to treatment of nicotine dependence.
In S. A. Shumaker, E. B. Schron, J. K. Ockene, & W. L. McBee (Eds.),
The handbook of health behavior change (2nd ed.; pp. 137-165). New York:
Springer.
American College of Sports Medicine Resource Manual (2006).
ACSM's resource manual for guidelines for exercise testing and
prescription (7th ed.). Philadelphia: Lippincott, Williams, &
Williams.
Anshel, M. H. (2006). Applied exercise psychology: A
practitioner's guide to improving client health and fitness. New
York: Springer Publishing Company.
Anshel, M. H. (1995). Examination of self-regulatory
characteristics and behavioral tendencies of elite and non-elite male
Australian swimmers. The Australian Psychologist, 25, 78-83.
Anshel, M. H., & Kang, M. (2007). An outcome-based action study
on changes in fitness, blood lipids, and exercise adherence, using the
Disconnected Values (Intervention)
Model. Behavioral Medicine, 32, 85-98. Anshel, M. H., & Martin,
M. B. (1996). Effect of divergent self-monitoring strategies on motor
performance and affect as a function of task complexity. Journal of
Sports Sciences, 14, 233-242.
Anshel, M. H., & Porter, A. (1996). A descriptive analysis of
self-regulatory characteristics of competitive swimmers as a function of
skill level and gender. Journal of Sport Behavior, 19, 91-110.
Anshel, M. H., Porter, A., & Hass, A. (1996). Examining the
self-regulatory cognitive and behavioral strategies of elite and
non-elite male and female swimmers. International Journal of Sport
Psychology, 2 7, 321-336.
Anshel, M. H., Reeves, L., & Roth, R. (2003). Concepts in
fitness: A balanced approach to good health. Boston: Pearson Education.
Baker, R. C., & Kirschenbaum, D. S. (1993). Self-monitoring may
be necessary for successful weight control. Behavior Therapy, 24,
377-394.
Baker, R. C., & Kirschenbaum, D. S. (1998). Weight control
during the holidays: Highly consistent self-monitoring as a potentially
useful coping mechanism. Health Psychology, 17, 367-370.
Biddle, S. J. H., Fox, K. R., Boutcher, S. H., & Faulkner, (2
E. (2000). The way forward for physical activity and the promotion of
psychological well-being. In S. J. H. Biddle, K.
R. Fox, & S. H. Boutcher (Eds.), Physical activity and
psychological well-being (pp. 154-168). London: Routledge.
Biddle, S. J. H., & Mutrie, N. (2001). Psychology of physical
activity: Determinants, well-being, and interventions. London:
Routledge.
Boekaerts, M., & Niemivirta, M. (2000). Self-regulated
learning: Finding a balance between learning goals and ego-protective
goals. In Handbook of self-regulation (pp. 417450). San Diego, CA:
Academic Press.
Boutelle, K. N., & Kirschenbaum, D. S. (1998). Further support
for consistent self-monitoring as a vital component of successful weight
control. Obesity Research, 6, 219-224.
Box, Ct E. P. (1949). A general distribution theory for a class of
likelihood criteria. Biometrika, 36, 317-346.
Buckworth, J., & Dishman, R. K. (2002). Exercise psychology.
Champaign, IL: Human Kinetics.
Castro, C. M., & King, A. C. (2002). Telephone-assisted
counseling for physical activity. Exercise and Sport Sciences Reviews,
30, 64-68.
Christensen, A. J., Martin, R., & Smyth, J. M. (2004).
Encyclopedia of health psychology. New York: Kluwer Academic.
Clark, N. M., & Becker, M. H. (1998). Theoretical models and
strategies for improving adherence and disease management. In S.A.
Shumaker et al. (Eds.), The handbook of health behavior change (2nd ed.,
pp. 5-32). New York: Springer.
Creer, T. L., & Holroyd, K. A. (1997). Self-management. In A.
Baum, C. McManus, S. Newman, J. Weinman, & R. West (Eds.), Cambridge
handbook of psychology, health, and behavior (pp. 255-258). Cambridge,
UK: Cambridge University Press.
Dijkstra, A., & De Vries, H. (1999). The development of
computer-generated tailored interventions. Patient Education and
Counselling, 36, 193-203.
Dishman, R. K. (1994). Consensus, problems, and prospects. In R. K.
Dishman (Ed.), Advances in exercise adherence (pp. 1-28). Champaign, IL:
Human Kinetics.
Dishman, R. K., & Gettman, L. R. (1980). Psychobiologic
influences on exercise adherence. Journal of Sport Psychology, 2,
295-310.
Dominick, K. L., & Morey, M. (2006). Adherence to physical
activity. In H. B. Bosworth, E. Z. Oddone, & M. Weinberger (Eds.),
Patient treatment adherence: Concepts, interventions, and measurement
(pp. 49-94). Mahwah, New Jersey: Erlbaum.
Ebbeling, C. B., Ward, A, Puleo, E. M., Widrick, J., & Rippe,
J. M. (1991). Development of a single-stage treadmill walking test.
Medicine and Science in Sports, 23, 966-973.
Endler, N. S., & Kocovsky, N. L. (2000). Self-regulation and
distress in clinical psychology. In M. Boekaerts, E R. Pintrick, &
M. Zeidner, (Eds., pp. 569-599). Handbook of self-regulation. San Diego,
CA: Academic Press.
Fishbein, M. (2007). A reasoned action approach: Some issues,
questions, and clarifications. In I. Ajzen, D. Albarracin, & R.
Homik (Eds.), Prediction and change of health behavior: Applying the
reasoned action approach (pp. 281-295). Mahwah, N J: Erlbaum.
Haynes, R. B. (2001). Improving patient adherence: State of the
art, with a special focus on medication taking for cardiovascular
disorders. In L.E. Burke & I.S. Ockene (Eds.), Compliance in
healthcare and research (pp. 3-21). Arraonk, NY: Futura.
Hume, K. M., Martin, G. L., Gonzelez, P., Cracklen, C., &
Genthon, S. (1985). A self-monitoring feedback package for improving
freestyle figure skating practice. Journal of Sport Psychology, 7,
333-345.
Izawa, K. P., Watanabe, S., Omiya, K., Hirano, Y., Oka, I., Osada,
N., & Iijima, S. (2005). Effect of the self-monitoring approach on
exercise maintenance during cardiac rehabilitation: A randomized,
controlled trial. American Journal of Physical Medicine Rehabilitation,
84, 313-321.
King, A. C. (1994). Clinical and community interventions to promote
and support physical activity participation. In R. K. Dishman (Ed.),
Advances in exercise adherence (pp. 183-212). Champaign, IL: Human
Kinetics.
Kirschenbaum, D. S. (1987a). Self-regulation of sport performance.
Medicine and Science in Sports and Exercise, 19, S106-S113.
Kirschenbaum, D. S. (1987b). Self-regulatory failure: A review with
clinical implications. Clinical Psychology Review, 7, 77-104.
Loehr, J., & Schwartz, T. (2003). Thepower offull engagement.
New York: Free Press.
Lounsbury, E (2004). Cardiac rehabilitation. In A. J. Christensen,
R. Martin, & J. M. Smyth (Eds.), Encyclopedia of health psychology
(pp. 41-62). New York: Kluwer Academic.
Lox, C. L., Martin, K. A., & Petruzzello, S. J. (2003). The
psychology of exercise: Integrating theory and practice. Scottsdale, AZ:
Holcomb Hathaway.
Marcus, B. H., King, T. K., Sock, B. C., Borrelli, B., & Clark,
M. M. (1998). Adherence to physical activity recommendations and
interventions. In S. A. Schumaker, E. B. Schron, J. K. Ockene, & W.
L. McBee (Eds.), The handbook of health behavior change (2rid ed., pp.
189-212). New York: Springer.
Mardia, K. V. (1971). The effect of non-normality on some
multivariate test and robustness to non-normality in the linear model.
Biometrika, 58, 105-121.
Marlatt, Cx A., & George, W. H. (1998). Relapse prevention and
the maintenance of optimal health. In S.A. Schumaker, E.B. Schron, J.K.
Ockene, & W.L. McBee (Eds.), The handbook of health behavior change
(2nd ed., pp. 33-58). New York: Springer.
Martin, M. B., & Anshel, M. H. 0995). Effect of self-monitoring
strategies on motor performance and affect. Journal of Sport &
Exercise Psychology, 17, 112-129.
Noland, M. E (1989). The effects of self-monitoring and
reinforcement on exercise adherence. Research Quarterly for Exercise and
Sport, 60, 216-224.
Oldridge, N. S. (1997). Outcome assessment in cardiac
rehabilitation: Health-related quality of life and economic evaluation.
Journal of Cardiopulmonary Rehabilitation, 17, 179-194.
Olson, C. L. (1974). Comparative robustness of six tests in
multivariate analysis of variance. Journal of the American Statistical
Association, 69, 898-908.
Perri, M. G., Anton, S. D., Duming, P. E., Ketterson, 1". U.,
Sydeman, S. J., Berlant, N. E., Kanasky, W. F., Newton, R. L., Limacher,
M. C., & Martin, A. D. (2002). Adherence to exercise prescriptions:
Effects of prescribing moderate versus higher levels of intensity and
frequency. Health Psychology, 21,452-458.
Pollock, M. L., & Jackson, A. S. (1984). Research progress in
validation of clinical methods of assessing body composition. Medicine
and Science in Sports and Exercise, 16, 606-615.
Prochaska, J. O., & Marcus, B. H. (1994). The transtheoretical
model: Application to exercise. In R. K. Dishman (Ed.), Advances in
exercise adherence (pp. 161-180). Champaign, IL: Human Kinetics.
Rand, C. S., & Weeks, K. (1998). Measuring adherence with
medication regimens in clinical care and research. In S.A. Schumaker,
E.B. Schron, J.K. Ockene, & W.L. McBee (Eds.), The handbook of
health behavior change (2nd ed., pp. 114-132). New York: Springer.
Rollnick, S., Mason, P., & Butler, C. (1999). Health behavior
change: A guide for practitiohers. London: Churchill Livingstone.
Siri, W. E. (1961). Body composition from fluid spaces and density:
analysis of methods. Nutrition, 9,480-491.
Sallis, J. E, HaskeU, W. L., Fortmann, S. P., Vranizan, K. M.,
Taylor, C. B., & Solomon, D. S. (1986). Predictors of adoption and
maintenance of physical activity in a community sample. Preventive
Medicine, 15, 331-341.
Sallis, J. F., & Owen, N. (1999). Physical activity &
behavioral medicine. Thousand Oaks, CA: Sage.
Snyder, M. (1974). Self-monitoring of expressive behavior. Journal
of Personality and Social Psychology, 30, 526-537.
Stevens, J. (1996). Applied multivariate statistics for the social
sciences, iahwah, N J: Lawrence Erlbaum.
Thomas, J. T., & Nelson, J. K., Silverman, S. J. (2005).
Research methods inphysical activity (5th ed.). Champaign, IL: Human
Kinetics.
Weber, J., & Wertheim, E. H. (1989). Relationships of
self-monitoring, special attention, body fat percent, and
self-motivation to attendance at a community gymnasium. Journal of Sport
and Exercise Psychology, 11, 105-114.
Whitley, B. E. (2002). Principles of research in behavioral science (2nd ed.). New York: McGraw-Hill.
Zimmerman, B. J. (2000). Attaining self-regulation: A social
cognitive perspective. In M. Boekaerts, ER. Pintrick, & M. Zeidner,
(Eds.), (pp. 13-39). Handbook of self-regulation. San Diego, CA:
Academic Press.
Mark H. Anshel and Scott J. Seipel
Middle Tennessee State University
Address Correspondence To: Dr. Mark Anshel, Department of Health
and Human Performance, Middle Tennessee State University, Box 96,
Murfreesboro, TN, 37132, E-mail:
[email protected].
Table 1. Descriptive statistics for fitness test scores (n=6).
Panel A. Experimental Group (n=29)
Pre-Test
Mean SD
Strength test (pushups, n=28) 9.54 7.33
Strength test (sit-ups, n=1) 50.00 N/A
Skinfold test (mm) 0.35 0.06
[VO.sub.2] sub-max (mls/kg/min) 22.48 7.31
Systolic BP 124.41 11.52
Diastolic BP 76.28 5.99
Post-Test Percent
Showing
Mean SD Improvement
Strength test (pushups, n=28) 17.63 11.03 100.0%
Strength test (sit-ups, n=1) 122.00 N/A
Skinfold test (mm) 0.32 0.05
[VO.sub.2] sub-max (mls/kg/min) 26.55 7.38
Systolic BP 124.69 10.50
Diastolic BP 77.21 7.60
Panel B. Control Group (n=36)
Pre-Test
Mean SD
Strength test (pushups, n=25) 15.12 12.34
Strength test (sit-ups, n=11) 27.27 11.41
Skinfold test (mm) 0.30 0.07
[VO.sub.2] sub-max (mls/kg/min) 26.95 10.88
Systolic BP 121.56 16.67
Diastolic BP 74.33 9.47
Post-Test Percent
Showing
Mean SD Improvement
Strength test (pushups, n=25) 20.60 13.65 83.3%
Strength test (sit-ups, n=11) 34.91 14.71
Skinfold test (mm) 0.28 0.06
[VO.sub.2] sub-max (mls/kg/min) 30.13 14.90
Systolic BP 118.44 11.54
Diastolic BP 73.31 9.28
Table 2. Descriptive statistics for changes in fitness test scores
(n = 65)
Panel A. Experimental Group (n=29)
Mean Median SD Skewness Kurtosis
Skinfold % Change 8.4% 7.8% 7.8% 0.77 0.81
Sub-max [VO.sub.2] %
Change 20.4% 20.7% 16.6% 0.43 -0.24
Systolic BP % Change -0.5% 0.0% 6.9% 0.06 -0.51
Diastolic BP % Change -1.4% 0.0% 8.8% -0.32 -0.97
Panel B. Control Group (n=36)
Mean Median SD Skewness Kurtosis
Skinfold % Change 6.9% 6.5% 9.9% 0.25 -0.68
Sub-max [VO.sub.2] %
Change 11.7% 3.5% 30.5% 1.22 2.71
Systolic BP % Change 1.9% 0.0% 7.4% 1.53 4.33
Diastolic BP % Change 0.9% 0.0% 8.7% 1.76 5.76
Table 3. Descriptive statistics for adherence to aerobic and strength
exercise regimens (n = 65)
Panel A. Aerobic Training
Experimental Group
4 wk
Week Mean SD Skewness Kurtosis Mean
1 0.89 0.24 -2.45 6.31 0.853
2 0.92 0.25 -2.98 8.01
3 0.82 0.30 -1.68 2.02
4 0.79 0.30 -1.18 0.18
5 0.85 0.23 -1.29 0.49 0.825
6 0.85 0.28 -2.23 4.88
7 0.82 0.29 -1.41 0.95
8 0.78 0.34 -1.45 0.95
Control Group
4 wk
Week Mean SD Skewness Kurtosis Mean
1 0.85 0.27 -1.74 2.10 0.891
2 0.86 0.30 -2.22 3.85
3 0.92 0.24 -3.28 10.45
4 0.94 0.21 -3.59 13.18
5 0.87 0.27 -2.33 5.04 0.736
6 0.72 0.37 -1.02 -0.42
7 0.73 0.41 -1.11 -0.53
8 0.62 0.44 -0.58 -1.53
Panel B. Strength Training
Experimental Group
4 wk
Week Mean SD Skewness Kurtosis Mean
1 0.53 0.47 -0.19 -1.93 0.630
2 0.71 0.39 -1.04 -0.46
3 0.58 0.39 -0.45 -1.25
4 0.70 0.31 -0.78 -0.26
5 0.66 0.37 -0.59 -1.09 0.635
6 0.63 0.40 -0.58 -1.27
7 0.67 0.35 -0.63 -0.81
8 0.59 0.39 -0.39 -1.29
Control Group
4 wk
Week Mean SD Skewness Kurtosis Mean
1 0.58 0.46 -0.35 -1.81 0.729
2 0.72 0.41 -1.02 -0.75
3 0.82 0.34 -1.76 1.62
4 0.79 0.37 -1.47 0.59
5 0.71 0.40 -1.02 -0.62 0.591
6 0.64 0.42 -0.58 -1.39
7 0.53 0.45 -0.15 _1.82
8 0.48 0.45 0.01 -1.85
