Location,Timing,and Social Structure Patterns Related to Physical Activity Participation in Weight Loss Programs

Abstract

Less than half of the adults in the United States meet national guidelines for physical activity. Physical activity programs can induce short-term improvements in physical activity. To develop effective interventions, researchers and practitioners should consider the timing, location, and social structure patterns of participants. Using a pretest, posttest study design, 329 adults in a weight loss program completed surveys on their patterns and physical activity participation. Men preferred weight lifting, cycling, and jogging, whereas women preferred walking and aerobics. Black participants preferred being active in the home. Additionally, participating in a mix of group and individual activities compared with individual-only activities was predictive of posttest moderate-to-vigorous intensity and total physical activity. Providing a variety of activities for adults in both location and social structure may lead to sustained physical activity participation.

Keywords

adults behavior change environment exercise

Physical activity reduces risk associated with diseases such as cardiovascular disease and diabetes mellitus (Mora, Lee, Buring, & Ridker, 2006). Increasing moderate-intensity or vigorous-intensity physical activity can decrease biomarkers of chronic inflammation and increase anti-inflammatory effects, leading to better heart health and a decreased risk of osteoporosis, diabetes mellitus, and cachexia (McFarlin et al., 2006). However, more than half (50.9%) of American adults do not meet the national guidelines of 150 minutes of moderate-intensity aerobic physical activity or 75 minutes of vigorous-intensity aerobic activity or an equivalent amount of combined moderate- and vigorous-intensity aerobic physical activity (Loustalot et al., 2009; U. S. Department of Health and Human Services, 2008).

Physical activity programs are a popular method to encourage inactive or insufficiently active individuals to increase their physical activity to meet national guidelines. Avenues for physical activity include individual exercises, group activities, in a gym, or at home (Skowron, Stodolska, & Shinew, 2008). The process of finding the appropriate physical activity program to motivate different individuals to participate and adhere to programs can be challenging and complex (Burke, Steenkiste, Music, & Styn, 2008; Daley et al., 2011; Dogra, Meisner, & Ardern, 2010). Program characteristics for design consideration that have been found to be important for successful implementation include instructor quality, location of program services, the type of exercise, style of information communication, and financial expenditure (Cohen-Mansfield, Marx, Biddison, & Guralnik, 2004; Skowron et al., 2008). As a result of the dynamic nature of human beings, tailoring physical activity programs according to individual preferences or targeting toward specific group patterns for physical activity has been suggested as an effective method of engaging and maintaining participation (Dogra et al., 2010).

Demographics and patterns for mode, timing, and with whom an individual exercises (i.e., social structure of the exercise environment; Nies, Vollman, & Cook, 1999) within activities may be important factors for improving the proportion of adults who meet physical activity guidelines. Ethnicity, age, gender, and socioeconomic status should be considered when developing physical activity programs as preferences may vary by these characteristics (Cohen-Mansfield et al., 2004; Skowron et al., 2008). Differences in preferred mode of exercise, time of exercise, social structure of activities, and location of physical activity programs are associated with ethnicity, age, and gender (Dogra et al., 2010). Although few studies have specifically included social structure as a determinant, a qualitative study of Black women defined social structure as the type and number of persons an individual may exercise with, and this emerged as a key facilitator of exercise behavior (Nies et al., 1999).

Targeted and tailored programs can enhance participation and adherence when preferences are taken into consideration. For example, Cohen-Mansfield et al. (2004) found in a community-based sample of 324 older adults that women were more particular about type of exercise, location of activities, music, and peer engagement than men, illustrating programming preferences that differ by gender. Additionally, Dogra et al. (2010) found that ethnic minorities and immigrants in Canada preferred to participate in conventional forms of exercise compared with their White counterparts in the Canadian Community Health Survey. In this sample of more than 230,000 adults, Black participants preferred sports and social activities (activities at church), whereas participants of Latin American origin preferred walking and the Polish population preferred house and yard work (Dogra et al., 2010).

These studies are limited, however, to cross-sectional data and do not consider the time of exercise and social structure of exercise participation. Therefore, the purpose of this study was to describe if patterns for physical activity, including type of activity, timing, and social structure among adults beginning a weight loss program, predicted future physical activity behavior and how these patterns differed by gender and race.

Method

Participants

Adults 18 years and older were recruited at kick-off events for a weight loss program in four cities across the state of South Carolina. The program provided e-mailed newsletters with information on exercise and other physical activity opportunities in the participant’s city, nutrition articles and recipes, and weekly weigh-ins. Participants in the weight loss program were also eligible for discounted membership to a recreation center for up to 3 months. Detailed information on the recruitment and enrollment procedures has been reported elsewhere (Gay & Smith, 2010). The response rate at baseline was 53.8% (N = 560), and all baseline data were collected within 2 weeks of the kickoff event where participants had their first weigh-in and received other health screenings. A 3-month follow-up survey was sent either electronically or by mail based on participant preference. A total of three reminders were sent to participants using Dillman’s (2000) methodology for surveys. All participants were entered into a random drawing for personal MP3 players, valued at $50 each. Of the 560 baseline respondents, 329 returned the follow-up survey (58.8%).

Instrumentation

In addition to measures of the independent and dependent variables, participants were asked to report demographic information, including gender, educational attainment, age, marital status, general health, race/ethnicity, and employment status. Participants also reported height and weight, which were used to calculate self-reported body mass index (BMI) as weight (kg)/height² (m²).

Physical activity patterns

Items were derived from the Canada Fitness Survey (Canadian Fitness and Lifestyle Research Institute, 1997). Specifically, items pertaining to timing and location were included in this study. Participants responded to the following questions: “When do you usually do physical activities?” with response options of weekdays, weekends, or both. “At what time do you usually do your physical activities,” where respondents could select all that apply from “in the morning,” “at lunchtime,” “in the afternoon,” “in the evening,” “at no special time.” Participants were also asked, “Where do you usually do your physical activities” and again could select one or more options from “home,” “outdoor activities in your neighborhood,” “place of worship,” “public school facility such as fields, tracks, gyms, or swimming pools,” “public recreational facility including public parks, pools, community centers, or playgrounds,” and “private recreational facility such as a YMCA, health club, dance, or yoga studio.” These options were adapted from the original Canada Fitness Survey to align with the options open to participants in the weight loss program.

Participants were also asked to report with whom they usually engage in physical activities. The response options were “Exclusively group activities,” “Exclusively individual activities,” or “Both group and individual activities.” This item was taken from a YMCA survey (Wallingford Family YMCA, 2006). Finally, participants were asked to list up to three preferred modes of physical activity (e.g., walking, cycling, aerobics).

Self-reported physical activity

Participants were asked to report physical activity participation using the Godin Leisure-Time Exercise Questionnaire (Godin & Shephard, 1985, 1997). The original questionnaire is worded to assess weekly the number of occasions of physical activity of at least 15 minutes. There are three items, one for each intensity level (mild, moderate, and strenuous). In a study of 306 adults (53.3% male), the physical activity score for these items produced reliability coefficients of 0.83 with Vo₂_max and 0.85 with body fat (Godin & Shephard, 1985). The questionnaire is also able to detect short-term changes in physical activity (Andrykowski, Beacham, & Jacobsen, 2007; Courneya, Friedenreich, Arthur, & Bobick, 1999).

The questionnaire was modified in two ways for this study. First, the minimum duration was changed from 15 to 10 minutes to align with physical activity guidelines that recommend bouts of at least 10 minutes (Haskell et al., 2007; U.S. Department of Health and Human Services, 2008). The second modification asked that participants report the average duration of physical activity participation by intensity level (Rogers et al., 2006). Average minutes per occasion were multiplied by frequency of occasions to determine the number of minutes per week of participation by intensity level. Proportion of participants meeting national physical activity guidelines was calculated as was the combined score of moderate- and vigorous-intensity activity minutes, treating vigorous-intensity activity as twice as many minutes as moderate-intensity activity (Loustalot et al., 2009).

Analysis Plan

Means and standard deviations were calculated for age, BMI, and minutes of physical activity by intensity level and total physical activity participation at baseline and follow-up. Pearson’s bivariate correlation coefficients were calculated for baseline age, BMI, and minutes of moderate-to-vigorous intensity physical activity and total physical activity. Frequencies for baseline data were calculated for gender, proportion meeting physical activity guidelines, and patterns of physical activity, including time of day (morning, lunch, afternoon, evening, no special time), time of week (weekdays, weekend, or both), group size (individual only, groups only, or both groups and individual), and location for activity. As a result of low frequencies for public recreational facilities, public schools, places of worship, and outdoor activities in the neighborhood, they were aggregated into general places open to the public category. Pearson’s chi-square tests of proportions were conducted to examine differences in physical activity patterns by gender and race. Independent samples t tests were used to assess differences in moderate-to-vigorous intensity and total physical activity behavior by gender and race.

To test whether physical activity patterns are predictive of posttest physical activity participation, two separate analysis of covariance models were conducted, one for moderate-to-vigorous intensity physical activity only and the second for total minutes of physical activity. This was important to capture variation in light-intensity activity as participants were enrolled in a weight loss program and not all were meeting physical activity guidelines. The independent variables of interest were social structure, time of day, time of week, location, and mode of activity. As a result of significant relationships between race and gender with physical activity patterns, these interactions were also included in the full model. Light-intensity activity at follow-up and baseline activity by intensity level were controlled for in the model for moderate- and vigorous-intensity activity. Baseline activity by intensity level was controlled for in the total physical activity model. A backward selection process based on Type III sums of squares was used to identify significant predictors of physical activity participation at follow-up. Variables were removed from the model based on the largest p value, starting with interactions. Physical activity participation was transformed to the square root to address skewness and kurtosis. The transformation yielded normally distributed variables. All analyses were conducted with SAS Version 9.2 (SAS Institute Inc., 2008) and inferences are interpretable at the α = .05 level.

Results

There were no differences in minutes of physical activity participation (moderate and vigorous intensity) by gender (t = −0.29, p = .775), race (t = −0.65, p = .519), educational attainment (F = 1.17, p = .313), employment status (F = 1.32, p = .268), or marital status (F = 0.05, p = .949). At baseline, age (r = −0.12, p = .006) and BMI (r = −0.15, p = .001) were both negatively and weakly associated with moderate-to-vigorous intensity physical activity. BMI was also negatively and weakly correlated with total physical activity (r = −0.12, p = .011), but no significant association was found with age.

As shown in Table 1, a greater proportion of follow-up participants reported higher educational attainment, 62.6% with a college degree or higher compared with 54.1% at baseline. A smaller proportion of follow-up participants had a high school education or less (16.1%) compared with baseline (23.4%). No significant differences were found between baseline and follow-up participants in age, BMI, gender, amount of physical activity, race, employment, or marital status.

Table 1.

Participant Characteristics by Time of Measurement

Variable	Baseline, M (SD) or n (%)	Follow-Up, M (SD) or n (%)	t or χ²
Age	44.65 (12.99)	45.76 (13.38)	−1.15
BMI	32.23 (7.63)	31.27 (8.67)	1.59
Gender, % female	417 (86.2)	261 (85.6)	0.05
Baseline physical activity
Light intensity	164.10 (200.10)	164.30 (194.30)	−0.01
Moderate intensity	115.50 (179.40)	188.40 (174.10)	−0.23
Vigorous intensity	83.47 (160.60)	77.63 (149.90)	0.54
Education			8.07*
High school	131 (23.4)	53 (16.1)
Some college	126 (22.5)	70 (21.3)
College degree or higher	303 (54.1)	206 (62.6)
Race			2.30
White	346 (75.7)	187 (70.6)
Black	111 (24.3)	78 (29.4)
Employment status			0.03
Full-time	311 (65.1)	194 (64.5)
Part-time	59 (12.3)	38 (12.6)
Other	108 (22.6)	69 (22.9)
Marital status			1.42
Married	334 (69.3)	222 (73.3)
Not married	62 (12.9)	34 (11.2)
Divorced/separated/widowed	86 (17.8)	47 (15.5)

Note. Column percentages shown. Participants describing themselves as Hispanic (n = 6) or Asian/Pacific Islander (n = 2) were not included in these analyses because of small cell sizes.

p < .05.

Pearson’s chi-square tests were used to examine differences in physical activity patterns by gender (Table 2). The five most frequently reported preferred activities (walking, cycling, aerobics, jogging/running, and weight lifting) were included in the chi-square analyses. A greater proportion of women (51.6%), compared with men (33.3%), preferred both group and individual activities, whereas a greater proportion of men (64.8% compared with 45.6% for women) preferred individual activities only. Almost all women reported that walking was a preferred activity (91.0%) compared with only about two thirds of men. Women also preferred aerobics (28.8%) more than men (14.8%). However, men preferred weight lifting (40.7%), cycling (31.5%), and jogging/running (27.8%) more than women (26.9%, 21.2%, and 13.0%, respectively).

Table 2.

Summary of Pearson’s Chi-Square Analyses for Physical Activity Patterns by Gender and Race

	Gender			Race
Physical Activity Patterns	Males (n = 67)	Females (n = 417)	χ²	White (n = 330)	Black (n = 111)	χ²
Social structure			6.94*			5.74
Groups activities only	1.9%	2.7%		2.7%	1.8%
Individual activities only	64.8%	45.6%		45.5%	58.6%
Both groups and individual activities	33.3%	51.6%		51.8%	39.6%
Time during the week			0.28			0.28
Weekdays	40.7%	43.84%		44.2%	43.6%
Weekends	5.6%	4.4%		4.2%	5.5%
Both weekdays and weekends	53.7%	51.8%		51.5%	50.9%
Time of day
Morning	37.0%	35.1%	0.08	35.8%	31.5%	0.69
Afternoon	22.2%	19.6%	0.21	17.9%	22.5%	1.16
Evening	42.6%	46.7%	0.33	45.4%	41.4%	0.53
No special time	13.0%	18.8%	1.07	16.2%	20.7%	1.21
Location			0.24			6.58*
Private recreational facility	42.6%	40.9%		42.7%	36.0%
Home	29.6%	33.0%		29.1%	42.3%
Places open to the public	27.8%	26.1%		28.2%	21.6%
Activity mode
Walking	68.5%	91.0%	22.85***	83.0%	92.8%	6.51*
Cycling (stationary or outdoor)	31.5%	21.2%	2.86	21.4%	23.4%	0.20
Aerobics	14.8%	28.8%	4.67*	27.5%	21.6%	1.49
Jogging/running	27.8%	13.0%	8.05***	14.7%	14.4%	0.01
Weight lifting	40.7%	26.9%	4.41*	30.1%	20.7%	3.65

Note. Column percentages shown. Participants were instructed to select more than one if they usually were physically active more than once a day and up to three preferred activities. Only 34 participants indicated being active during lunch. Because of the low frequency, this variable was not included in the chi-square analyses.

p < .05. **p < .01. ***p < .001.

A comparison of the demographic, physical activity pattern, and exercise participation variables at baseline and follow-up are presented in Table 1. A greater proportion of participants with a college degree or higher completed the follow-up survey than at baseline (p < .05). No other significant differences were detected by time of survey administration.

Physical activity patterns were also compared by race (Table 2). A greater proportion of Black participants preferred being active in the home (42.3%) compared with White participants (29.1%). Also, more Black participants preferred walking (92.8%) compared with White participants (83.0%). The difference for more White participants preferring weight lifting (30.1% compared with 20.7% for Blacks) approached significance (p = .056), as did the difference for social structure, where more White participants preferred both group and individual activities (51.8%) than Black participants (39.6%; p = .057).

No differences were detected in follow-up physical activity behavior by race or gender for moderate-to-vigorous intensity physical activity (p = .683; p = .996, respectively) or for total physical activity (p = .595; p = .237, respectively).

To identify whether physical activity patterns were predictive of future physical activity behavior, analysis of covariance models were conducted separately for moderate-to-vigorous intensity physical activity and total physical activity. Using listwise deletion, 299 participants were included in the analyses. Physical activity was transformed to the square root to meet the assumption of normality. Interactions between race and gender with physical activity patterns were not significant and were therefore removed from the model. Furthermore, race and gender were not significant predictors and were removed during the backward model selection. As shown in Table 3, social structure and preferring jogging/running were predictive of moderate-to-vigorous intensity physical activity at follow-up, controlling for baseline physical activity participation and follow-up light intensity physical activity. These patterns explained 42% of the variation in follow-up moderate-to-vigorous intensity physical activity. More specifically, participating in both group and individual activities compared with individual activities alone (referent) was significantly associated with greater follow-up moderate-to-vigorous intensity physical activity (p < .01). Participants who reported jogging/running as a preferred activity also reported greater amounts of moderate-to-vigorous intensity physical activity at follow-up (p < .001).

Table 3.

Analysis of Covariance Summary for Pattern Variables Predicting Moderate-to-Vigorous Intensity Physical Activity (Transformed) at Follow-Up

Variable	B	SE B	F	Least Squares Means
Social structure			4.62*
Both group and individual activities	2.27**	0.75		15.91
Group activities only	0.53	2.65		14.17
Individual activities only	—	—		13.64
Jogging/running			18.90^***
Yes	4.61***	1.06		16.88
No	—	—		12.27
Baseline physical activity participation
Light intensity	0.01**	0.00	10.75***
Moderate intensity	0.01*	0.00	5.36*
Vigorous intensity	0.02***	0.00	84.05***
Follow-up light-intensity activity participation	0.01***	0.00	11.50***

Note. $R_{adj}^{2}$ = 0.42 (N = 299, F = 29.91, p < .0001).

p < .05. **p < .01. ***p < .001.

Social structure was also predictive of follow-up total physical activity as was being active in the evening (p < .01; Table 4). Again, participants who preferred both group and individual activities engaged in greater amounts of physical activity at follow-up (p < .01) compared with participants who preferred individual activities only (referent). No interaction terms for race and gender with physical activity patterns were significant. The final model (Table 4), controlling for baseline physical activity participation, accounted for 24% of the variation in total physical activity.

Table 4.

Analysis of Covariance Summary for Pattern Variables Predicting Total Physical Activity (Transformed) at Follow-up

Variable	B	SE B	F	Least Squares Means
Social structure			3.94*
Both group and individual activities	2.15	0.80**		15.06
Group activities only	−1.30	2.80		11.61
Individual activities only	—	—		12.92
Active in the evening			7.75**
Yes	−2.20	0.79		12.10
No	—	—		14.30
Baseline physical activity participation
Light intensity	0.01	0.00	10.87**
Moderate intensity	0.01	0.00	8.28**
Vigorous intensity	0.02	0.00	34.76***

Note. $R_{adj}^{2}$ = 0.42 (N = 300, F = 15.31, p < .0001).

p < .05. **p < .01. ***p < .001.

Discussion

This study explored the differences in physical activity patterns by race and gender. That there were no differences in physical activity behavior by race or gender does not align with the literature, but is not a wholly unique finding. Australian adults were also more likely to report walking behavior regardless of gender and age group (Booth, Bauman, Owen, & Gore, 1997). This finding may be due to the context of data collection, where all participants were enrolled in the same weight loss program. Although there were differences in physical activity patterns, race and gender were not significant predictors of physical activity behavior. However, the relationship between social structure and physical activity behavior at follow-up was strongest for participants who engaged in both group and individual activities. This was true for explaining both moderate-to-vigorous intensity physical activity as well as total physical activity.

The results from this study are similar to other cross-sectional studies that show associations between preferences and physical activity participation. For example, in a 2010 analysis of the Canadian Community Health Survey data (Dogra et al., 2010), non-White participants reported a greater preference for at-home activities, and they were less likely to engage in endurance activities, such as jogging/running and weight training, compared with their White counterparts.

Relating to social structure, in a qualitative study with 33 overweight and obese adults, participants reported they were more likely to engage in physical activity if they had someone to motivate them to increase their activity levels, or to be active with them—participants enjoyed friendly competition offered by group activities (Suggs, McIntyre, & Cowdery, 2010). Furthermore, participants enjoyed contact from coaches, workout buddies, and counselors who would work with their unique situations more so than being physically active alone. Participants in this qualitative study also wanted to have social norms that supported being physically active, which would increase social support for physical activity (Suggs et al., 2010). These findings are also similar to the present study in that adults who engaged in both group activities and individual activities engaged in more moderate-to-vigorous intensity physical activity and total physical activity than those who reported individual-only physical activities. This may be explained by a potential preference for being active with other people. In a separate study of 102 female older adults who preferred to receive support from friends were more active when they engaged in activities with others (Wilson & Spink, 2009). Therefore, providing targeted activities that participants can do on their own or in groups, allowing for alterations in preference, may be instrumental in maintaining physical activity participation over time.

Although the findings from this study found that more participants reported physical activity in the evening (more than 40% for each gender and each race), Cohen-Mansfield et al. (2004) reported that older adults reported a time preference for physical activity, where the majority preferred to exercise between 9:00 a.m. and noon (52.4%). In the present study, the time-of-day pattern could be a result of age differences. Participants in the Cohen-Mansfield et al. (2004) study were all between 74 and 85 years of age, while in this study the mean age at baseline was 44.65 years of age. With a younger sample population, it is more likely that work and child care responsibilities may affect timing of physical activity behavior. All these findings together support the need for providing choices when conducting physical activity programming. Varying time of day, group and individual activities, and providing activities that can be done at home may be instrumental in increasing the proportion of adults who meet national physical activity guidelines to attain health benefits.

Limitations

The results from this study should be taken with consideration of several limitations. The participants in this study were a convenience sample from a weight loss program available across the state of South Carolina. Although the participants were representative of all program participants on demographic characteristics, people who complete surveys may be different from those who did not enroll in the study on motivational or other characteristics. Participants also had a higher level of education than the general statewide population. Furthermore, the results may not be applicable to adults who consistently meet national guidelines for physical activity.

Although patterns for physical activity can be indicative of preferences, this is not always the case. Participants in this study reported actual physical activity patterns rather than preferences, which may reflect the availability and accessibility of physical activity opportunities, and these items have not been assessed for reliability or validity. Self-reported physical activity data, which are prone to overestimation by the participant, were collected rather than objectively measured data. However, the Godin Leisure-Time Exercise Questionnaire has been shown to have acceptable reliability properties with aerobic capacity and body composition (Godin & Shephard, 1985) and can detect change in physical activity participation over a short amount of time (Andrykowski et al., 2007; Courneya et al., 1999). Finally, this study did not address motivation, environmental variables, or level of engagement in the intervention that may also explain physical activity participation. In other physical activity interventions, participants in high implementation groups engaged in a greater amount of vigorous-intensity physical activity, highlighting the importance of understanding the degree of engagement in interventions and the impact on physical activity outcomes (Saunders, Ward, Felton, Dowda, & Pate, 2006).

Conclusion

These novel findings expand on existing descriptive studies by illustrating that emphasizing both group and individual activities, compared with only group or only individual activities, can lead to increased physical activity participation after 3 months. These findings demonstrate that although physical activity patterns for social structure, location, and type of activity differed by gender and race, only social structure consistently explained variation in future physical activity. Physical activity and weight loss programs should consider these differences when targeting interventions. This study adds to the understanding of physical activity participation during interventions and should be used in conjunction with the knowledge base of motivational and environmental determinants that also explain physical activity behavior. As physical activity interventions move more toward the built environment and policy realms, individual-level patterns and preferences may be neglected. But in keeping with ecological models of health behavior, successful interventions must consider the full spectrum of influences on physical activity participation. Developing interventions that provide choices for timing, location, and social structure of activities may improve likelihood of successfully maintaining physical activity engagement in adults Furthermore, in research, community, or clinical settings, promoting autonomy in activity selection, time of day, and whether a participant exercises at home or at a recreation center may lead to increased physical activity participation. Future research may consider how commonalities in individual-level behavioral patterns, such as preferences for exercising alone or in a group, can be influenced through environmental or policy interventions.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interests with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Andrykowski

M. A.

Beacham

A. O.

Jacobsen

P. B.

(2007). Prospective, longitudinal study of leisure-time exercise in women with early-stage breast cancer. Cancer Epidemiology, Biomarkers & Prevention, 16, 430-438.

Booth

M. L.

Bauman

Owen

Gore

C. J.

(1997). Physical activity preferences, preferred sources of assistance, and perceived barriers to increased activity among physically inactive Australians. Preventive Medicine, 26, 131-137.

Burke

L. E.

Steenkiste

Music

Styn

M. A.

(2008). A descriptive study of past experiences with weight-loss treatment. Journal of the American Dietetic Association, 108, 640-647.

Canadian Fitness and Lifestyle Research Institute. (1997). Canada Fitness Survey. Medicine & Science in Sports & Exercise, 29(Suppl. 6), 153-161.

Cohen-Mansfield

Marx

M. S.

Biddison

J. R.

Guralnik

J. M.

(2004). Socio-environmental exercise preferences among older adults. Preventive Medicine, 38, 804-811.

Courneya

K. S.

Friedenreich

C. M.

Arthur

Bobick

T. M.

(1999). Physical exercise and quality of life in postsurgical colorectal cancer patients. Psychology, Health & Medicine, 4, 181-187.

Daley

Stokes-Lampard

Wilson

Rees

Roalfe

Macarthur

(2011). What women want? Exercise preferences of menopausal women. Maturitas, 68, 174-178.

Dillman

D. A.

(2000). Mail and Internet surveys. New York, NY: John Wiley.

Dogra

Meisner

B. A.

Ardern

C. I.

(2010). Variation in mode of physical activity by ethnicity and time since immigration: A cross-sectional analysis. International Journal of Behavioral Nutrition and Physical Activity, 7, 75.

10.

Gay

J. L.

Smith

(2010). Validity of a scale assessing the built environment for physical activity. American Journal of Health Behavior, 34, 420-431.

11.

Godin

Shephard

R. J.

(1985). A simple method to assess exercise behavior in the community. Canadian Journal of Applied Sport Science, 10, 141-146.

12.

Godin

Shephard

R. J.

(1997). Godin Leisure-Time Exercise Questionnaire. Medicine & Science in Sports & Exercise, 29(Suppl. 6), S36-S38.

13.

Haskell

W. L.

Lee

I. M.

Pate

R. R.

Powell

K. E.

Blair

S. N.

Franklin

B. A.

. . . Bauman

(2007). Physical activity and public health: Updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Medicine & Science in Sports & Exercise, 39, 1423-1434.

14.

Loustalot

Carlson

S. A.

Fulton

J. E.

Kruger

Galuska

D. A.

Lobelo

(2009). Prevalence of self-reported aerobic physical activity among U.S. States and territories—Behavioral Risk Factor Surveillance System, 2007. Journal of Physical Activity & Health, 6(Suppl. 1), S9-S17.

15.

McFarlin

B. K.

Flynn

M. G.

Campbell

W. W.

Craig

B. A.

Robinson

J. P.

Stewart

L. K.

. . . Coen

P. M.

(2006). Physical activity status, but not age, influences inflammatory biomarkers and toll-like receptor 4. Journals of Gerontology. Series A Biological Sciences and Medical Sciences, 61, 388-393.

16.

Mora

Lee

I. M.

Buring

J. E.

Ridker

P. M.

(2006). Association of physical activity and body mass index with novel and traditional cardiovascular biomarkers in women. Journal of American Medical Association, 295, 1412-1419.

17.

Nies

M. A.

Vollman

Cook

(1999). African American women’s experiences with physical activity in their daily lives. Public Health Nursing, 16, 23-31.

18.

Rogers

L. Q.

Courneya

K. S.

Robbins

K. T.

Malone

Seiz

Koch

. . . Nagarkar

(2006). Physical activity and quality of life in head and neck cancer survivors. Supportive Care in Cancer, 14, 1012-1019.

19.

SAS Institute Inc. (2008). SAS/STATA (Version 9.2) [Computer software]. Cary, NC: Author.

20.

Saunders

R. P.

Ward

Felton

G. M.

Dowda

Pate

R. R.

(2006). Examining the link between program implementation and behavior outcomes in the lifestyle education for activity program (LEAP). Evaluation and Program Planning, 29, 352-364.

21.

Skowron

M. A.

Stodolska

Shinew

K. J.

(2008). Determinants of leisure time physical activity participation among Latina women. Leisure Sciences, 30, 429-447.

22.

Suggs

McIntyre

Cowdery

J. E.

(2010). Overweight and obese sedentary adults’ physical activity beliefs and preferences. American Journal of Health Studies, 25(2), 69-77.

23.

U.S. Department of Health and Human Services. (2008). Physical Activity Guidelines Advisory Committee Report (ODPHP Pub. No. U0036). Washington, DC: Author.

24.

Wallingford Family YMCA. (2006). Wallingford Family YMCA member satisfaction survey. Wallingford, CT: Author.

25.

Wilson

K. S.

Spink

K. S.

(2009). Social influence and physical activity in older females: Does activity preference matter? Psychology of Sport and Exercise, 10, 481-488.