The efficacy of physical activity in the prevention and management of coronary artery disease as well as atherosclerotic risk factors, including hypertension, insulin resistance, glucose intolerance, lipid abnormalities, and obesity, has been consistently demonstrated.1–3 It is currently recommended that all adults participate in moderate-intensity physical activity, such as brisk walking for 30 minutes on most, if not all, days of the week. 2,4
Although rates of physical activity are below recommended levels among all demographic groups, adult women particularly fall short of meeting such levels.2,4,5 Furthermore, certain subgroups of women (eg, older women, African American women, women with low monetary incomes) are among the least physically active.6–8 Some have argued that these women have the most to gain in disease prevention and health promotion through even modest increases in physical activity levels.5 Such a debate is pointless, however, if we are unable to persuade sedentary women to participate in regular exercise. Therefore, identifying factors associated with exercise participation is an important focus of exercise and health behavior research.
One potentially important but relatively understudied area of adult exercise behavior involves the study of exercise self-definitions and how having an exercise identity relates to exercise participation. On the basis of theories of the self, including Markus's9 self-schema model, Bem's10 self-perception theory, and Stryker's11 identity theory, it has been proposed that individuals who define themselves as exercisers are more likely to engage in exercise than individuals without such self-definitions. To the extent that women consider themselves to be exercisers, they should be motivated to participate in exercise to verify their self-view.12
The idea that we define ourselves in multiple ways or have multiple internal “selves” (eg, mother, daughter, spouse, nurse) has been an enduring one in the study of sociology and psychology, and examining the relationships between self-definitions and behavior is of particular importance. Markus9 refers to these multiple senses of self as self-schemas. A self-schema is a cognitive knowledge structure, formulated through experience and focused on an aspect of the self, regarded by the individual as important.9 The study of self-schemata has focused on the relevance of one's self-image and how it influences personal factors, environmental situations, and behavior by guiding the processing of information.9
Stryker's11 identity theory emphasizes the role society plays in the emergence and maintenance of the self. Identity theory views the self-concept as a hierarchical ordering of identities that are tied to roles within the social structure.13 This ordering of identities is determined by salience (the probability of activating a given identity in a situation) and commitment (the number and affective strength of ties to others as a result of having a particular role identity). It is thought that identity hierarchies influence our behavioral choices, behavioral consistency, and resistance to behavioral change. The greater one's commitment to an identity, the greater will be the salience of the identity and in turn, the salience of an identity directly influences the behavioral choices made among available choices.11
Bem's10 theory of self-perception posits that individuals use observations of their own behavior as a basis for inferring attitudes and presumably other cognitions about the self; behavior is observed and attitude formation follows. Self-perception theory10 acknowledges the importance of behavioral self-monitoring on attitude formation. For example, if a woman exercises 3 times a week, she infers that she is an exerciser and the more she participates in exercise, the stronger this attitude becomes. Thus, the formation of self-definitions is influenced by our values and perceptions, our interactions with others, and our behavior. We behave in ways that are consistent with our self-views and that behavior, in turn, reinforces the self-definition.
Recent research has shown that young adults who define themselves as exercisers exercise more frequently,14–17 follow through on their intentions to start exercising,14,17 use more strategies to help them exercise regularly,15 and report higher expectations to exercise in the future16 than those who do not define themselves as exercisers. These studies have demonstrated that exercise self-definitions do exist, that such definitions have been associated with both self-reported and objective measures of exercise, and that defining oneself as an exerciser has been shown to predict future exercise behavior. This body of research shows great promise but has been limited by an overreliance on data collected from white, healthy, college students. The influence of exercise self-definitions on exercise participation among middle-aged and older adults is not clear.
Therefore, the purpose of this study was to explore the relationships between exercise self-definitions and participation in a group-based exercise program among a sample of mostly sedentary, middle-aged to older women at high risk for health problems related to physical inactivity. Specifically, a prospective, observational design was used to: (a) describe the exercise self-definitions of study participants, (b) examine the relationships between exercise self-definition scores and exercise participation over a 6-month period, and (c) and report changes in such self-definitions from baseline to 6 months. On the basis of theoretical and empirical research, we hypothesized that (1) exercise self-definitions would be positively correlated with exercise participation and (2) that exercise self-definition scores would increase over time as exercise participation increased (ie, 6-month scores would be significantly greater than baseline scores). Such information may assist nurses working with individuals who need to increase physical activity levels to understand the psychological impact of seeing oneself as an “exerciser,” assist in the development of interventions to strengthen exercise self-views, and ultimately, increase regular exercise participation.
The study was a secondary analysis of selected data from a larger study (Exercise Referral and Older Women Study [EROWS], R01-AG15098) designed to evaluate the long-term effectiveness of physician-initiated exercise referral among women receiving primary care.18 Potential participants were identified using the electronic Regenstief Medical Records System (RMRS), a computerized medical record system that tracks patient visits to health centers, emergency department, or hospital, and records diagnoses, prescriptions, and lab and vital data. Selected women were then screened by their primary care provider for eligibility to complete a submaximal exercise test and ultimately, participate in a group-based, community exercise program. Eligible women were also asked to complete telephone surveys at baseline and at 6-month follow-up.
The study targeted women aged 50 years and older who received primary care from 1 of 2 health centers associated with an urban county hospital. The study health centers serve predominantly low-income adults, and the majority of patients are African American. Compared to US adults, adults who receive healthcare from the study health centers are 4 times more likely to be poor, are 2 to 3 times more likely to report fair or poor health, are more likely to have a medical diagnosis of hypertension, type 2 diabetes mellitus, or coronary heart disease, and fewer than 10% meet US Public Health Service recommendations for physical activity.7
Eight hundred sixty women met the age criterion and were randomly selected from the RMRS as potential study participants. For budgetary and timeline reasons, enrollment was stopped after 500 of the 860 (58%) women had visited 1 of the 2 study health centers during the 6-month enrollment period. Of these 500 women, 404 (81%) were screened by their primary care provider and were determined eligible to participate in submaximal exercise testing and were encouraged to participate in the study's exercise program. Of these 404 women, 192 completed the baseline telephone survey (100 were never reached by phone, 84 refused to be interviewed, and 28 were eliminated from the study because of incomplete data) and comprised the sample for the current study. Eighty-six of the 192 study participants completed the 6-month telephone survey.
Data for the study were derived from the RMRS (age, ethnicity, and chronic disease diagnoses), responses to baseline and 6-month telephone surveys, and observational accounts of the frequency of participation in the exercise program. The telephone surveys consisted of items designed to assess health status and self-beliefs, including exercise self-definitions. Trained research assistants conducted the telephone surveys. No incentives were offered for completing the telephone interviews or for exercise class participation.
A submaximal exercise diagnostic test was provided free of charge to those women who were determined eligible for exercise testing. The purpose of exercise testing was to prescreen individuals for potential health-related risks to participation in moderate-intensity exercise. Exercise testing took place in the study health centers and was conducted by a trained, Advanced Cardiac Life-Saving certified exercise physiologist according to American College of Sports Medicine (ACSM) guidelines.19 Observational records were kept as to which women did or did not complete the submaximal exercise test. The test included riding a stationary bicycle while an electrocardiograph machine monitored heart rate and rhythm. All women who participated in exercise testing successfully completed it.
The group exercise sessions were offered several times a day (Monday through Friday) in a church or community center near the health centers and were free to study participants. The sessions were designed to provide a low- to moderate-intensity exercise level and lasted approximately 60 minutes. Sessions consisted of 20 minutes of chair-based arm and leg movements and up to 30 minutes of indoor walking. A research assistant checked heart rates at each exercise session to ensure that participants were exercising at appropriate levels of intensity (ie, heart rates between 55% and 70% of age-adjusted maximum heart rate). Participants were encouraged to complete at least 3 exercise sessions per week, totaling the US Public Health Services recommendation of moderate-intensity exercise for 150 minutes per week.5
Health Status and Current Exercise Measures
Data on age, ethnicity, and chronic disease diagnoses were obtained from the RMRS. The diagnoses included coronary artery disease, hypertension, type 2 diabetes mellitus, and chronic obstructive pulmonary disease.
During the baseline telephone survey, participants were asked to rate their overall perceived health from excellent to poor. The response set included excellent, very good, good, fair, or poor. Perceived level of difficulty with physical mobility was assessed by totaling the responses to the following 4 items: (a) “Do you have any difficulty walking 1 block,” (b) “…walking 10 blocks,” (c) “…climbing 1 flight of stairs,” and (d) “…climbing several flights of stairs.” Each item was coded 1 (yes) and 0 (no). Possible scores ranged from 0 to 4, and a higher score indicated greater perceived difficulty with physical mobility. The perceived health item and the physical mobility difficulty items have been used in nationally representative surveys of adults and have been shown to perform adequately.7,20
Items that focused on walking assessed exercise behavior at baseline. Participants were asked whether over the past 4 weeks they ever walked 1 block or more, and if so, how long the walk usually lasted. Summing the reported number of minutes walked per week provides a simple, summary measure of exercise. These items had been used previously in a cross-sectional survey on exercise among a random sample of older adults recruited using the RMRS.7 In this prior study, the stability of the responses was assessed using 2-week test-retest reliability. The reported intraclass correlation coefficient for the total minutes walked per week was 0.59, indicating adequate stability for the responses.
The strength of an individual's exercise self-definition was assessed using a measure developed specifically for the current study. We defined exercise self-definition as the degree to which an individual considered herself an exerciser. Item development was guided by the characteristics of the targeted population (ie, middle-aged to older, mostly sedentary women), theories of the self,9–11 and empirical research related to exercise and the self,9–16 in particular, research reported by Kendierski et al.21 Kendierski and colleagues suggested that the development and maintenance of exercise self-definitions are influenced in 3 ways: (a) from inferences made by observing their own behavior and/or the circumstances in which the behavior occurs, (b) by factors that affect one's motivation to exercise, and (c) the acknowledgment of the self-definition by the social world. The validity of these influences was supported by significant correlations between exercise self-definitions and (a) perceptions about behavior (eg, perceived effort), (b) motivation-related variables (eg, perceived competence, enjoyment of engaging in the activity), and (c) social world variables (eg, perception that others acknowledge the self-definition) among undergraduate exercisers and athletes.
Therefore, drawing from the theoretical and empirical literature cited earlier, we developed 11 items to assess the exercise self-definitions of our targeted population. Principal components factor analysis was used to examine the structure of the measure, and results yielded 3 factors that accounted for 67% of the total variance. Examination of the scree plot supported the decision to include all 3 factors in the analyses. All items within each factor had factor loadings that exceeded 0.30. The factors were labeled Acknowledgment, Value of Exercise, and Competence (Table 1). Acknowledgment refers to an individual's view of herself as an exerciser as well as her perceptions of how others view her as an exerciser. Value of Exercise pertains to the perceived enjoyment and importance of exercising, and Competence measures one's perceived ability to perform specific exercises in comparison to others of same age. Cronbach coefficients for the total scale and the subscales (Acknowledgment, Value of Exercise, and Competence) were .86, .88, .72, and .78, respectively. Correlations among the subscale scores ranged from 0.35 to 0.64; none of the correlations among these scores were interpreted as problematic.
Exercise participation was a main outcome variable, and this study focused on exploring the relationships between exercise self-definitions and the decision to adopt and maintain participation rates in a group-exercise program. We measured participation in terms of the number of exercise sessions completed. A completed session was defined as a minimum of 20 minutes of continuous exercise at 55% to 70% of maximal heart rate (moderate intensity as defined by the ACSM,19 1995) as determined by the research assistant who was present at each exercise session. Participation rates were measured at 2 points in time: weeks 1 through 8 and weeks 1 through 24. Data describing exercise behavior were collected by observational accounts of participation at each exercise session.
Baseline characteristics of the study participants were described using sociodemographic (age and ethnicity), health status (prevalence of chronic diseases, perceived health status, and perceived difficulty with physical mobility), and baseline exercise behavior (self-reported number of minutes walked per week) data. These data were retrieved from the RMRS and the baseline telephone survey.
The means and standard deviations of exercise self-definition scores at baseline (total scale and each of the 3 subscales) are presented. Differences in scores by participants' age (< 65 years vs ≥ 65 years), ethnicity (African American vs Caucasian), perceived health status (fair and poor vs excellent, very good, and good), perceived difficulty with physical mobility (more vs less), and baseline exercise behavior (walked < 30 minutes per week vs walked ≥ 30 minutes per week) were explored using independent t tests. Descriptive statistics (means, standard deviations, standard errors, skewness, kurtosis, and ranges) were calculated and analyzed for the indicators of exercise participation.
The hypotheses were tested using Pearson's product-moment correlation coefficients to determine relationships between exercise self-definition scores and exercise participation and paired t tests to determine differences between baseline and 6-month self-definition scores. For all analyses, we used a P value of .05 or less to indicate statistical significance.
The mean age of the 192 study participants was 64.1 years (range = 50–88). The majority of the women were African American (64.2% [n = 118], remaining were white). Rates of chronic disease were high; 92% (n = 172) of the women had a medical diagnosis of hypertension, 43% (n = 82) had been diagnosed with type 2 diabetes mellitus, 18% (n = 35) had been diagnosed with coronary artery disease, and 12% (n = 23) had a medical record indication of chronic obstructive pulmonary disease.
Forty-one percent of the women (n = 78) described their health as fair or poor. The mean score for the difficulty with physical mobility measure was 2.6 (SD = 1.3; range = 0–4), indicating that the women perceived their overall mobility to be somewhat limited. The women were sedentary; 58% (n = 112) reported 0 minutes of weekly walking over the 4 weeks prior to the baseline telephone interview and the average number of minutes walked per week was 25 (range = 0–180). Twenty-three percent (n = 52) of the participants reported walking 30 minutes or more per week, for an average of about 4 minutes per day as compared with current physical activity recommendations of 30 or more minutes per day and at least 150 minutes of activity per week.2
Exercise Self-Definitions at Baseline
As indicated in Table 2, at baseline, participants had moderate total exercise self-definition scores, indicating that they were a little to somewhat sure of defining themselves as exercisers. However, an analysis of the subscale scores provides us with more specific information about the exercise self-definitions of the study participants. Acknowledgment scores were fairly low. In fact, 25% of the sample (n = 48) scored 3 or less on this subscale, indicating that they were very unsure that they or others would acknowledge them as an exerciser. In contrast, Value of Exercise scores were high, indicating that participants strongly believed that exercising was enjoyable and important to them and their significant others. Competence scores were moderate, implying that participants were somewhat certain that their ability to perform the identified exercises was comparable to others their age. As can be seen in Table 3, there were significant differences in self-definition scores across sociodemographic, health status, and baseline exercise variables.
Despite being determined eligible for exercise testing and being encouraged by their primary care provider to participate in the study exercise program, only 45% (n = 86) of the participants completed at least one exercise session while 55% (n = 106) of the participants completed zero sessions. Exercise participation rates were disappointing: the mean number of exercise sessions completed during weeks 1 through 8 was 5.69 (SD = 8.84, range = 0–39) as compared with a recommended participation level of 24 sessions for the first 8 weeks (3 times per week for 8 weeks) and the mean number of completed sessions for weeks 1 through 24 was 11 (SD = 20, range = 0–99) as compared to the recommended 72 sessions.
Hypothesis 1: Relationships Between Exercise Self-Definition Scores and Exercise Participation
As Table 4 indicates, our first hypothesis was supported with significant relationships between exercise self-definitions and exercise participation. Examining the pattern of these correlations was both interesting and informative. Value of Exercise scores were significantly correlated with the mean number of exercise sessions completed during weeks 1 through 8 (r = 0.20, P < .01) while Acknowledgment and Competence scores were significantly correlated with the number of exercise sessions completed during weeks 1 through 24 (r = 0.16, P = .03; r = 0.16, P = 0.03, respectively). The correlations between total scores and the indicators of exercise participation did not reach statistical significance.
Additional support for our hypothesis was found when we examined the relationships between 6-month exercise self-definition scores and exercise participation. Six-month scores were significantly correlated with the number of exercise sessions completed during weeks 1 through 24; total scores (r = 0.27, P = .04), Acknowledgment (r = 0.29, P = .02), Value of Exercise (r = 0.28, P = .01), and Competence scores (r = 0.35, P = .00).
Hypothesis 2: Differences Between Baseline and 6-Month Self-Definition Scores
As can be seen in Table 5, our second hypothesis was also supported, as there was a significant difference between baseline and 6-month self-definition scores. Six-month values for the total scale and all 3 subscales were significantly higher than baseline reports.
Exercise Self-Definition Scores
The current study explored the relationships between exercise self-definitions and exercise participation in a sample of women that is seldom represented in exercise studies—older, minority women with a high prevalence of chronic disease and low levels of exercise participation. The moderate total self-definition scores reported at baseline reflected the total of the high Value of Exercise scores and the moderate competence scores. Acknowledgment scores however were very low, which was not surprising given the participants' modest engagement in physical activity and their prevalent health issues. According to Markus,9 if an individual has relatively little experience in a given domain and has not attended to the behavior in this domain (eg, regular exercise participation), then it is unlikely that she will have a well-developed self-view. It is possible that participants felt uncomfortable with the term “exerciser” and that exercise participation was seen more as a means of avoiding being considered “old.” Similar to findings reported by Kendzierski et al,21 it is likely that participants believed the term “exerciser” was better suited for individuals who spent considerably more time exercising and expended greater effort than they did. Furthermore, low Acknowledgment scores might have reflected the lack of physically active peers in the participants' social world; if significant others do not exercise, it seems unlikely that they would pressure the participant to start exercising.21
However, despite low rates of reported baseline exercise participation, low Acknowledgment scores, and reports of greater difficulty with physical mobility, baseline Competence scores were moderate in strength, suggesting that perceived ability to perform the exercises was not perceived as a barrier to participation. Value of Exercise scores were high, signifying that even though the majority of the participants did not exercise, they were aware of the benefits of exercise participation and acknowledged the importance of exercise to themselves and their significant others. This finding supports previous research that older adults believe in the value of healthy personal behaviors such as exercise, regardless of the level of participation.22 However, it is possible that Value of Exercise scores were inflated because of self-presentational concerns (eg, social desirability) on behalf of the participants during the telephone survey.
Examination of self-definition scores by demographic, health status, and exercise variables indicated variation of self-views among the participants. For example, older women reported higher self-definition scores than did younger women and African American women reported higher self-definition scores than did Caucasian women. These results were unexpected, since exercise behavior is more typically associated with younger Caucasian women.6 The findings are also counter to age-related myths that older individuals, particularly older women, will receive no benefits from exercising.22 The higher self-definition scores among African American women might possibly reflect previous findings that African American women hold more positive views about their bodies than do white women.23–25 Continued examination of the differences in exercise self-definitions by age and ethnicity is recommended. It seems logical to assume that women who perceived themselves to be healthier and less constrained by physical mobility would be more likely to define themselves as exercisers than those who do not. However, this finding might be a red flag for nurses working with women with poor health perceptions who could benefit from exercise, but do not view themselves as capable of exercising. Consistent with Markus'9 premise that individuals with a well-defined self-view in a particular behavioral dimension are most likely to behave in a manner consistent with their self-view as well as findings reported by related empirical studies,12,14–17 self-definition scores were higher among participants who reported higher levels of baseline exercise than those who did not. Although not the purpose of this study, these findings provide preliminary evidence to support the construct validity of the exercise self-definition measure.
Relationships Between Exercise Self-Definitions and Exercise Participation
It has been acknowledged that behavioral change, such as initiating an exercise program, is a complex and difficult process.26 Several phases in the behavior change process have been identified, including initiating or adopting behavior directed toward the desired outcome and maintaining behavioral change over time and across various situations.26 Different determinants are thought to influence individuals at these different phases of change.27 We examined the level of exercise at 2 points in time and found that baseline Value of Exercise scores were significantly related to the number of exercise sessions completed during weeks 1 through 8 while Acknowledgment and Competence scores were related to the number of exercise sessions completed during weeks 1 through 24. Interestingly, if we refer to weeks 1 through 8 as the phase of exercise adoption or the initiation of an exercise program and weeks 1 through 24 as the maintenance phase, or the performance of regular exercise over an extended period of time28 then our results suggest that exercise self-definitions differentially influence the phases of behavioral change. Thus, emphasizing the health-related Value of Exercise may have a greater effect on the decision to start exercising while acknowledging one's exercise identity and focusing on exercise competence may strengthen the decision to maintain an exercise program. Further exploration of the influence of exercise self-definitions on different phases of behavior change is warranted.
In addition to the relationships between exercise self-definitions and exercise participation, we documented a significant difference between baseline and 6-month self-definition scores. Six-month values were significantly higher than baseline values implying that exercise self-definitions can be strengthened with exposure to an exercise program.9,15
Of course, there are several limitations to this study. Even though the women included in the study were among the most vulnerable to chronic disease and physical disability resulting from a lack of exercise, the sample was not randomly selected and the demographic group targeted might have affected the external validity of the results. It is suggested that the relationships between exercise self-definitions and exercise participation be examined among a larger, randomly selected, more diverse sample of women. Second, the low response rate to the telephone surveys, particularly the 6-month follow-up survey, is viewed as a limitation to this report. Third, the correlations reported by the current study were low to moderate in magnitude (ie, correlations ranged from 0.15 to 0.35). The majority of the study participants did not complete even one exercise session; correlations might have been stronger among women with higher levels of exercise participation.
To increase participation rates in future studies, researchers might consider offering programs that include a home exercise component that encourages and monitors exercise in addition to class participation and using as many techniques as possible to strengthen exercise self-definitions. Assessing the strength of exercise self-definitions might increase our understanding of what differentiates exercising women from women who do not exercise. It is possible that some drop out early in the program because their exercise self-definitions are not yet strong enough to influence them to continue or even start exercising. For example, a woman may be motivated to start exercising to lose weight rather than to become an exerciser. If the behavior (eg, exercise) is not consistent with how she defines herself (someone who does not enjoy exercising), she is at high risk for dropping out of the program before her exercise self-definition has strengthened.
Nurses can offer tools that foster exercise self-definitions by bringing subconscious thoughts that affect our self-views, such as how good we feel after completing an exercise session, to the conscious level by simply talking about them during class and by reminding participants of the exercise behavior that they already do, such as climbing stairs or walking to the bus stop. It may be necessary to rectify misconceptions about exercise, such as being “too old to exercise,” or address fears of injury by describing the purpose of the class, what takes place during class, the age and health status of women who attend the class, and a description of the typical clothing worn. Furthermore, provide participants with health-promotion literature that incorporates pictures of similar women engaging in exercise accompanied by the phrase “picture yourself here” and when addressing participants use phrases that acknowledge an exercise self-definition, such as “Congratulations on becoming an exerciser” or “Great to be working with people so committed to keeping healthy.” If possible, provide T-shirts with printed captions such as “I'm keeping my heart healthy by exercising.”
In summary, the results of the present study document the importance of assessing exercise self-definitions and demonstrate their relationship with exercise participation. These results strengthen previous research by using a prospective design and documenting a change in exercise self-definitions over a 6-month time frame among a unique sample of older, mostly sedentary women. Since the ultimate goal is to design interventions based on salient factors that lead to an increase in exercise participation, the pivotal issues are to strengthen the exercise self-definition of participants to increase regular participation and to determine the extent to which exercise self-definitions predict the adoption and maintenance of exercise behavior.
1. Thompson PD, Buchner D, Pina IL, et al. Exercise
and physical activity
in the prevention and treatment of atherosclerotic cardiovascular disease. Circulation
. 2003; 107:3109–3116.
2. Fletcher GF, Balady GJ, Amsterdam EA, et al. Exercise
standards for testing and training. Circulation
3. US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General
. Atlanta, Ga: Centers for Disease Control & Prevention, National Center for Chronic Disease Prevention & Health Promotion; 1996.
4. Pate RR, Pratt M, Blair SN, et al. Physical activity
and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA
5. US Department of Health and Human Services. Healthy People 2000 Review 1998–1999
. Hyattsville, Md: Public Health Service; 1999. DHSS publication PHS 99-1256.
6. Brownson RC, Eyler AA, King AC, Brown DR, Shu YL, Sallis JF. Patterns and correlates of physical activity
among US women
40 years and older. Am J Public Health
7. Clark DO. Physical activity
and its correlates among urban primary care patients aged 55 years and over. J Gerontol Soc Sci.
8. Clark DO, Patrick DL, Grembowski D, Durham M. Socioeconomic status and exercise
self-efficacy in late life. J Behav Med.
9. Markus H. Self-schemas and processing information about the self. J Pers Soc Psychol.
10. Bem DJ. Self-perception theory. In: Berkowitz L, ed. Advances in Experimental Social Psychology
. Vol. 6. New York: Academic Press; 1972:1–62.
11. Stryker S. Identity theory: developments and extensions. In: Yardley K, Honess T, eds. Self and Identity: Psychosocial Perspectives
. New York: Wiley; 1987:89–103.
12. Kendzierski D. Schema theory: an information processing focus. In: Dishman RK, ed. Advances in Exercise Adherence
. Champagne, Ill: Human Kinetics Publishing; 1994:137–159.
13. Cook KS, Fine GA, House JS, eds. Sociological Perspectives on Social Psychology
. Boston: Allyn & Bacon; 1995.
14. Estabrooks P, Courneya KS. Relationships among self-schema, intention, and exercise
behavior. J Sport Exerc Psychol.
15. Kendzierski D. Self-schemata and exercise
. Basic Appl Soc Psychol.
16. Yin Z, Boyd MP. Behavioral and cognitive correlates of exercise
self-schemata. J Psychol
17. Kendzierski D. Exercise
self-schemata: cognitive and behavioral correlates. Health Psychol.
18. Damush TM, Stump TE, Saporito A, Clark DO. Predictors of older primary care patients' participation in a submaximal exercise
test and a supervised, low-impact exercise
class. Prev Med.
19. American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription
. 6th ed. Philadelphia, Pa: Williams & Wilkins Inc; 2000.
20. Guralnik JM, LaCroix AZ, Abbott RD, et al. Maintaining mobility in late life. Demographic characteristics and chronic conditions. Am J Epidemiol.
21. Kendzierski D, Furr M, Schiavoni J. Physical activity
self-definitions: correlates and perceived criteria. J Sport Exerc Psychol.
22. Henry J. Promoting exercise
behaviors in older adults. Health Educ Monogr Ser
23. Harris SM. Racial differences in predictors of college women
's body image attitudes. Women Health
. 1994;21: 89–104.
24. Stevens J, Kumanyika SK, Keil JE. Attitudes toward body size and dieting: differences between elderly black and white women
. Am J Public Health
25. Whaley DE, Ebbeck V. Self-schemata and exercise
identity in older adults. J Aging Phys Activity
26. Stein KF, Markus HR. The role of the self in behavioral change. J Psychother Integration
27. Dishman RK. Advances in Exercise Adherence
. Champaign, Ill: Human Kinetics; 1994.
28. Marcus BH, Dubbert PM, Forsyth LH, et al. Physical activity
behavior change: issues in adoption and maintenance. Health Psychol.
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
exercise; exercise self-definition; physical activity; women