Despite the health risks of physical inactivity and sedentary behavior (26), most adults do not engage in recommended levels of physical activity (22). Promoting leisure time recreational sports and exercise is a promising way to increase activity levels, as it accounts for an estimated 60%-70% of physical activity reported among adults (32). There has been increasing support for the role of social and physical environmental factors in recreational physical activity (14,28,35). Research has shown that the availability and accessibility of environmental resources and social support are positively associated with leisure time physical activity levels in adults (7,8). However, little is known about which types of physical settings and social contexts are more conducive to recreational sports and exercise, especially in terms of their impact on the intensity and duration of individual exercise bouts. Policies and programs to promote physical activity at the population level could benefit from information about how social features and physical settings influence the parameters of activities taking place in those contexts.
A small body of intervention and observational research has examined the impact of the immediate social context and physical setting on exercise behavior. For example, a home-based exercise program for obese women had a greater impact on long-term exercise participation, adherence, and weight loss compared with a group-based program held at a clinic facility (25). Also, older adults participating in an overground walking training program had more favorable attitudes and found exercise more enjoyable compared with a treadmill training group (21). In a study using experience sampling, children reported more intense physical activity when they were with their peers and close friends (30). Research comparing self-selected activity intensities of various tasks found differences between home and laboratory settings (44). Taken together, this evidence suggests that physical activity performance varies across contexts. However, the ability to draw meaningful conclusions from this research is limited by the focus on only a small number of settings and use of nonrepresentative samples.
Time use studies offer a novel approach to overcoming challenges associated with studying the effects of different environments on physical activity. Time use surveys sequentially assess the type and duration of all activities that a person performs across the day (using diary or short-term recall) and are commonly used by sociologists and economists to describe human behavior patterns (5,18,32). An advantage of time use survey methodology is that exposure to and use of environments can be temporally linked to specific behaviors (i.e., the interview assesses where and with whom behavior occurs throughout the day). The American Time Use Survey (ATUS), which uses a nationally representative U.S. sample, has been recently used to study walking behaviors (37) and demographic and temporal patterns in the use of social and physical environments for sports and exercise (12).
The current study used data from the ATUS to examine whether the intensity and duration of sports and exercise bouts performed by U.S. adults differ across social and physical environments. Several hypotheses were tested. First, on the basis of the evidence showing a favorable impact of social interactions on physical activity intensity (30) and adherence (9), it was hypothesized that the intensity and duration of exercise bouts occurring with friends/acquaintances, family members, or coworkers would be greater than exercise occurring alone. Second, prior research suggests that people often prefer to exercise outdoors (42) and that there is a positive association between time spent outdoors and overall physical activity level (21). Therefore, it was hypothesized that outdoor environments would have a favorable impact on exercise performance. We expected that exercise occurring outdoors would be greater in intensity and duration than exercise occurring in any other physical environments. We took an exploratory approach to comparing exercise intensity and duration among bouts performed with family, friends/acquaintances, and coworkers as well as among bouts performed at home, at work, and at a gym/health club, mainly due to a lack of prior research on these contexts.
METHODS
ATUS procedures.
The ATUS is sponsored by the Bureau of Labor and Statistics and conducted by the U.S. Census Bureau (http://www.bls.gov/tus/). It uses a stratified, three-stage survey design to measure time use patterns in a nationally representative sample of individuals (ages 15 yr and older). To obtain reliable estimates and adequate sample size, the survey oversampled weekend days and some demographic groups (e.g., households with Hispanics or non-Hispanic Blacks, households with children). Within each selected household, a telephone interview was conducted with a single randomly chosen individual to assess time use on the previous day. The reporting day was randomly preassigned to the household. To encourage households without a telephone to call to complete the survey, they received a $40 debit card in the mail. The ATUS, which is collected through the Census Bureau, does not release identifiable participant information to the public and is not subject to institutional review board approval. Authorization to collect the ATUS is granted by the U.S. Office of Management and Budget.
Trained ATUS interviewers used a Computer-Assisted Telephone Interviewing system to pose a series of standardized questions designed to lead the respondent through the recall of the previous day. Respondents were asked to sequentially describe each activity and its duration for the 24-h period beginning at 4:00 a.m. Follow-up questions assessed where and with whom each activity occurred. Each interview lasted approximately 15 to 20 min.
Study sample.
To obtain reliable estimates of activity across different times of the day and days of the week, data from the 2003-2006 surveys were combined in the current study. Only respondents aged ≥21 yr and reporting at least one bout of sports or exercise were included in the analyses.
Data treatment.
ATUS interviewers at the telephone center assigned a six-digit classification code to each activity (31). The present study focused on activities coded as participating in sports, exercise, or recreation (General Activity Category) within sports, exercise, and recreation (Major Activity Category). Walking (for exercise) and hiking fell into this categorization scheme. Other walking or biking activities coded under traveling (General Activity Category), household, nonhousehold (Major Activity Category), or volunteer activities (Major Activity Category) were not examined in this study. Likewise, moderate-intensity physical activities performed for nonrecreational purposes (e.g., transportation, occupation, household chores) were not included in these analyses.
Activity intensity expressed in METs was calculated using a scheme that links the ATUS variables to the Compendium of Physical Activities (1,38) (http://riskfactor.cancer.gov/tools/atus-met/). Activity intensity levels were based on classifications used by the CDC/ACSM (25) (moderate, 3.0-6.0 METs; vigorous, >6.0 METs).
Follow-up interview questions assessed the social and physical environments of each activity. After the activity was reported, respondents were asked, "Who was with you/Who accompanied you?" Interviewers coded responses into 22 different categories. Responses were given more than one category code if a heterogeneous group was reported. Before data analyses, this variable was recoded into four general categories representing the overarching type of social companion: Alone, Family (e.g., household children, parents, siblings), Friends/Acquaintances (e.g., friends, neighbors, acquaintances, other nonhousehold children, other nonhousehold adults), and Coworkers. A fifth general subgroup called Multiple categories was created, which encompassed responses reporting social companions from more than one of the other four general categories.
To assess the physical environment of the activity, respondents were asked, "Where were you while you were [activity]?" The responses were coded by the interviewers into 26 different categories including travel modes. Before data analyses, less common locations for exercise, such as someone else's house, school, and place of worship, as well as locations coded as other place and unspecified place, were combined to create an overall Other/unspecified category. Exercise and sports reported in unusual or illogical locations (e.g., grocery store, library, restaurant/bar, bank) (<3%) and sedentary travel modes (e.g., car/truck/motorcycle, bus, subway/train, boat/ferry, other mode of transportation, unspecified mode of transportation) (<2%) were not included in the analyses. Therefore, the resulting physical environment variable consisted of five levels: Outdoors, Home, Work, Gym/health club, and Other/unspecified.
Temporal and demographic variables were also recoded. Before data analyses, a season variable was created by recoding the interview date as follows: fall (September 21 to December 20), winter (December 21 to March 20), spring (March 21 to June 20), summer (June 21 to September 20), and target days were coded as weekday or weekend day. Each activity was coded for the time of day that it began (i.e., morning [4:00 to 11:59 a.m.], afternoon [12:00 to 5:59 p.m.], and evening [6:00 p.m. to 3:59 a.m.]). A four-level age variable was also created: 21-34, 35-44, 45-59, and ≥60 yr. Respondents' highest level of education completed was recoded into the following categories: less than high school, high school graduate, some college, graduate of 4-yr college or greater. Race/ethnicity was recoded as non-Hispanic white only, non-Hispanic black only, Hispanic only/Hispanic mixed, Asian/Pacific Islander only, American Indian/Alaskan Native, and non-Hispanic mixed.
Statistical analyses.
Because of the complex sampling design of the ATUS, statistical analyses were conducted using SUDAAN 9.0 (RTI International, Research Triangle Park, NC). ATUS sample weights were applied to all statistical analyses to reduce bias in producing U.S. population estimates as a result of differential sampling and response rates across subpopulations and days of the week. For all analyses, the Fay replicate weight method was used to estimate SE that accounted for the complex sample design (41). This method for variance estimation used replicate weights provided by survey administrators of the ATUS, with an adjustment factor that was set to Fayadj = 0.3. Sample weights across the years 2003-2005 were standardized to the weighting method used in 2006, allowing combined analysis of results (40). All analyses accounted for nonindependence of observations for individuals reporting more than one exercise bout per day through the replicate weight method of variance estimation (27).
Sample-weighted estimates for the proportion of sports and exercise bouts that were of moderate and vigorous intensity were generated. Logistic regression analyses were subsequently used to test whether the type of social and physical environment predicted the likelihood that a sport or exercise bout was of vigorous (compared with moderate) intensity. Multiple regression analyses were used to estimate whether bout duration (in minutes) differed by type of social and physical environment. For the logistic and multiple regression models testing social environments, alone was designated as the reference group. Planned contrasts comparing family members, coworkers, and friends/acquaintances were tested using the "Effects" statement. Outdoors was designated the reference group for the models testing physical environments. Differences in sports and exercise intensity and duration among the home, work, and gym/health club were tested with planned contrasts. The multiple regression analyses testing differences in the duration of sports and exercise bouts examined moderate- and vigorous-intensity activities separately. All of the models adjusted for sex, age, education level, and race/ethnicity. Temporal variables such as season, day of the week, and time of day were also statistically controlled because they could hypothetically influence both the environment in which a behavior is performed and the intensity and duration of that behavior. Predicted marginal proportions (for dichotomous outcomes [moderate vs vigorous intensity]) and predicted marginal means (for continuous outcomes [minutes]) were calculated from the logistic and multiple regressions, respectively. These predictions are standardized values that adjust for all of the other covariates in the model (19).
RESULTS
Descriptive statistics.
A total of 55,902 adults participated in the ATUS across the 4 yr (2003-2006). Of these participants, 7700 individuals (13.8%) reported at least one bout of moderate- to vigorous-intensity (≥3.0 METs) sports or exercise on the previous day (exercise bouts reported in unusual/illogical locations were excluded). The demographic characteristics with weighted population estimates for the current study sample (i.e., adults reporting one or more exercise bouts) compared with the overall adult ATUS sample are shown in Table 1.
;)
TABLE 1: Demographic characteristics for ATUS participants: adults reporting one or more exercise boutsa versus the overall adult sample
Whereas the majority of the current study sample reported one sports or exercise bout on the previous day, approximately 22% reported more than one bout, which contributed to a combined total of 9819 bouts across all participants. The most frequently mentioned types of activity were walking (for exercise; 32%), using cardiovascular equipment (13%), weightlifting/strength training (11%), participating in water sports (7%), and running (5%). Overall, 70.4% of the sports and exercise bouts were moderate (3.0-6.0 METs), and 29.6% were vigorous (>6.0 METs). The sample-weighted mean duration of moderate and vigorous exercise bouts were 78.43 (SE = 0.37) min and 55.47 (SE = 0.32) min, respectively
Environmental influences on the intensity of sports and exercise.
The likelihood of vigorous activity differed by type of social (Adj. Wald F = 161.60, df = 4, P< 0.001) and physical (Adj. Wald F = 532.68, df = 4, P<0.001) environment (Table 2). A greater percentage of sports and exercise bouts performed alone (36%) were vigorous compared with bouts occurring with family members (18%), friends/acquaintances (30%), coworkers (25%), or multiple categories (19%). Results for the physical environment showed that a greater percentage of outdoor sports and exercise bouts were vigorous (26%) compared with bouts taking place at work (16%). In contrast, a greater percentage of sports and exercise bouts occurring at home (40%) or at a gym/health club (40%) were vigorous compared with outdoor exercise bouts. The percentage of vigorous exercise bouts performed at a gym/club and at home was comparable.
TABLE 2: Results of logistic regression analyses comparing social and physical environments by intensity of sports and exercise bouts (vigorous vs moderate) in the ATUS
Environmental influences on the duration of sports and exercise.
Exercising alone generally resulted in shorter moderate and vigorous bouts. On average, moderate exercise bouts occurring alone (56 min) were shorter in duration than bouts occurring with family members (83 min), friends/acquaintances (120 min), or multiple categories (122 min) (P values < 0.001; Fig. 1A). However, the average duration of moderate exercise bouts occurring alone did not differ from moderate bouts occurring with coworkers (55 min). Vigorous bouts occurring alone (44 min) were shorter than bouts taking place with family members (67 min), friends/acquaintances (79 min), coworkers (68 min), or multiple categories (86min; Pvalues < 0.001; Fig. 1B).
FIGURE 1: Predicted marginal means for duration (in min) of moderate (A) and vigorous (B) exercise bouts by social environment. Multiple categories = social companions from more than one of the other four general categories. Predicted marginal means are adjusted for sex, age, education level, race/ethnicity, season, day of the week, and time of day (N = 9819 bouts). All SE are ≤2.42. Differences between values with common letters (e.g., abc) are statistically significant at P < 0.001.
For moderate and vigorous activity, bouts taking place outdoors and in other/unspecified locations lasted the longest. Specifically, moderate bouts performed outdoors (84 min) were longer in duration than bouts occurring at home (52 min), at work (27 min), or at gym or health club (60 min; P values < 0.001) and were shorter than moderate exercise in other/unspecified locations (94 min; P < 0.001; Fig. 2A). Likewise, vigorous exercise bouts performed outdoors (66 min) lasted longer than vigorous bouts taking place at home (42 min), at work (40 min), at gym or health club (47 min), or in other/unspecified locations (62 min; Pvalues < 0.001; Fig. 2B).
FIGURE 2: Predicted marginal means for duration (in min) of moderate (A) and vigorous (B) exercise bouts by physical environment. Examples of other/unspecified locations are someone else's house, school, and place of worship. Predicted marginal means are adjusted for sex, age, education level, race/ethnicity, season, day of the week, and time of day (N = 9819 bouts). All SE are ≤1.86. Differences between values with common letters (e.g., abc) are statistically significant at P < 0.001.
DISCUSSION
The current research examined the influence of social and physical environments on the intensity and duration of recreational sports and exercise bouts reported by adults in the ATUS (years 2003-2006). This is one of the first studies among adults to investigate characteristics of exercise bouts across the full range of settings that they naturally occur in the U.S. population. Time use data offer a novel research strategy to examine context-specific exercise behaviors. Results showed that social company and physical location were differentially associated with exercise intensity and duration. These findings contribute to a growing body of work examining the role that social contexts and physical settings may play in shaping exercise behaviors.
Results from the current study are consistent with behavior setting theory, which suggests that behavior is impacted by characteristics of the immediate context in which it occurs (4). There is emerging support for the notion that microcontextual factors, such as with whom and where a person is located, influence characteristics of the health behavior taking place in that environment (39). This microcontextual approach is supported by calls for research on context-specific behaviors and behavior-specific aspects of the environment relevant to physical activity (29). Overall, findings suggest that sports and exercise behaviors can take on different characteristics depending on the microcontext in which they take place.
The results suggest that exercising alone has a beneficial impact on bout intensity and an unfavorable effect on bout duration. In contrast to children, who were found to engage in more intense physical activities when in the presence of others compared with when they were alone (30), exercise among adults in the current study was significantly more likely to reach a vigorous intensity when performed alone. A potential explanation for this finding is that high-intensity activities, such as speed and endurance training, are simply less feasible when occurring with family members or other groups with mixed ages and activity levels. The decreased likelihood of intense exercise when accompanied by people from multiple categories (e.g., friends and family members) supports this interpretation. Another possible reason is that the goal of successfully completing the exercise task at a desired intensity level is more salient during solitary bouts compared with bouts performed with others, which could also have process-oriented social or entertainment objectives. The unfavorable effect of exercising alone on bout duration, however, is consistent with a large body of work emphasizing the advantages of social support and peer interactions for physical activity (2,17,30,35,43). There is some evidence to suggest that exercising with other people is more enjoyable, pleasant, or entertaining than exercising alone (10,11), which could result in longer sessions. Also possible is that the cooperative or competitive nature of engaging in sports and exercise with other people has a positive impact on enjoyment and performance (13,34). Exercise occurring with people coming from multiple categories is performed with more than one other person. The added element of this group dynamic may further account for its beneficial impact on bout duration. Taken together, results from the current study show that social environments can have different implications for exercise intensity and duration.
Differential findings also emerged for the effect of the physical environment on exercise intensity and duration. The greater likelihood of vigorous activity occurring at a health club/gym and at home may be due to the increased availability of cardiovascular and strengthening equipment at these locations. These results diverge from studies on children, which suggest that activity intensity is higher outdoors than inside their home (3), and may indicate age differences in preferences for types of vigorous activity (e.g., adults prefer indoor treadmill running, whereas children prefer outdoor sports). Yet, this hypothesis has not been tested. The fact that exercise sessions taking place in outdoor and other/unspecified locations persisted longer, however, may reflect unique features of those settings. For example, a few studies suggest that exercising outdoors and in natural settings is perceived to be more pleasant than exercising in artificial environments (6,44). The ATUS was not specifically designed to study exercise environments in-depth, and therefore, a substantial number of physical locations have been coded as other/unspecified. Some of these activities took place at someone else's house, school, or a place of worship. Unfortunately, we are unable to infer much about the remainder of the locations included in this category, but they may consist of recreational facilities such as ice/roller skating rinks, climbing walls, obstacle courses, or other places not included in the general ATUS coding schema. Distinctive characteristics of these settings, such as the availability of specialized and/or novel equipment, could account for in the extended exercise bouts taking place in those settings.
Overall, these findings provide evidence supporting the existence of distinct typologies of context-specific exercise behaviors. Some environments may be more conducive to higher-intensity activities, whereas others are more conducive to activities that are longer in duration. The prioritization of intensity or duration for an activity bout could vary on the basis of contextual factors. Specifically, when people are exercising in the company of others or outdoors, fun and enjoyment might be important. In this case, people might prioritize activity duration over intensity. In contrast, when people exercise alone, especially in a health club (where there is training equipment and help from professional trainers) or at home (if they own exercise equipment), the primary goal could be to engage in high-intensity exercise to improve fitness. Prioritizing intensity in health clubs and at home might naturally result in shorter exercise bouts. These results shed light on the larger issue of environment-behavior specificity that has been receiving growing attention (23,29). The choice of settings for specific programs may need to take into account individual differences and whether increasing the intensity or duration of activity (or both) is the overall objective for the program.
Using the ATUS to examine differences in the intensity and duration of sports and exercise bouts across environments offers several strengths. First, because the ATUS relied on short-term recall of activities and their environments, the data may be less vulnerable to recall errors than typical self-report assessments of physical activity, which ask respondents to recall activities over longer periods (15). Second, physical activity is only one of the many daily activities assessed in the ATUS. Thus, physical activity reporting in the ATUS may be less influenced by social desirability reporting bias. In addition, reporting time use chronologically across the day can provide temporal and contextual cues that may help to improve the accuracy of activity recall (16). Lastly, after applying the appropriate population weights, information reported on the ATUS is nationally representative and, therefore, describes sports and exercise performed by U.S. adults on any given day.
There are a few limitations to this study, however. First, the ATUS is a cross-sectional study and therefore, we need to exert caution in making causal interpretations of the findings. Second, the "other/unspecified" category within the physical environment variable is quite large and heterogeneous because the ATUS was not specifically designed to measure physical activity environments. Third, despite efforts taken by the ATUS developers to standardize response explanations and coding (31), respondents may have experienced some confusion and ambiguity with regards to how to report certain environments such as exercising outdoors when at work. Also, 24-h recall time use surveys describe the average behavior of a group (e.g., proportion of individuals reporting physical activity) on any given day. They are not intended to estimate an individual's usual behavior (i.e., average daily physical activity level), which can vary from day to day. Furthermore, it is possible that the threshold levels for moderate and vigorous activity (3.0 and 6.0 METs, respectively) do not accurately reflect intensity levels for different individuals because of between-people heterogeneity in fitness levels. Lastly, the current study focused only on recreational sports and exercise. The analyses did not include activities performed for the purpose of transportation, work, or household chores.
In summary, research using a nationally representative time use survey of U.S. adults found that the intensity and duration of sports and exercise bouts differ across social and physical environments. However, the effect of context was not consistent for intensity and duration. Therefore, the choice of settings for specific exercise programs may need to take into account whether increasing the intensity or duration of exercise takes priority. Overall, results support the existence of context-specific exercise behaviors.
The first author was supported by the Cancer Prevention Fellowship Program, Office of Preventive Oncology, National Cancer Institute, National Institutes of Health during the preparation of this article.
We would like to thank the ATUS staff at the U.S. Bureau of Labor and Statistics for their technical assistance with this project. The views and opinions expressed in this article are those of the authors and not necessarily those of the Department of Health and Human Services, the National Institutes of Health, or the National Cancer Institute. The results of the present study do not constitute endorsement by ACSM.
REFERENCES
1. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of physical activities: an update of activity codes and MET intensities.
Med Sci Sports Exerc. 2000;32(9 suppl):S498-516.
2. Anderson ES, Wojcik JR, Winett RA, Williams DM. Social-cognitive determinants of physical activity: the influence of social support, self-efficacy, outcome expectations, and self-regulation among participants in a church-based health promotion study.
Health Psychol. 2006;25(4):510-20.
3. Baranowski T, Thompson WO, DuRant RH, Baranowski J, Puhl J. Observations on physical activity in physical locations: age, gender, ethnicity, and month effects.
Res Q Exerc Sport. 1993;64:127-33.
4. Barker RG.
Ecological Psychology: Concepts and Methods for Studying the Environment of Human Behavior. Palo Alto (CA): Stanford University Press; 1968. p. 219-20.
5. Basner M, Fomberstein KM, Razavi FM, et al. American Time Use Survey: sleep time and its relationship to waking activities.
Sleep. 2007;30(9):1085-95.
6. Bertucci W, Grappe F, Groslambert A. Laboratory versus outdoor cycling conditions: differences in pedaling biomechanics.
J Appl Biomech. 2007;23(2):87-92.
7. Booth ML, Owen N, Bauman A, Clavisi O, Leslie E. Social-cognitive and perceived environment influences associated with physical activity in older Australians.
Prev Med. 2000;31(1):15-22.
8. Brownson RC, Baker EA, Housemann RA, Brennan LK, Bacak SJ. Environmental and policy determinants of physical activity in the United States.
Am J Public Health. 2001;91(12):1995-2003.
9. Chiang KC, Seman L, Belza B, Tsai JH. "It is our exercise family": experiences of ethnic older adults in a group-based exercise program.
Prev Chronic Dis. 2008;5(1):A05.
10. Deci EL, Ryan RM. The "what" and "why" of goal pursuits: human needs and the self-determination of behavior.
Psychol Inq. 2000;11(4):227-68.
11. Duncan SC. The role of cognitive appraisal and friendship provisions in adolescents' affect and motivation toward activity in physical education.
Res Q Exerc Sport. 1993;64(3):314-23.
12. Dunton GF, Berrigan D, Ballard-Barbash R, Graubard B, Atienza AA. Social and physical environments of
sports and exercise reported among adults in the American Time Use Survey.
Prev Med. 2008;47(5):519-24.
13. Durand M, Delaplace J, Fourchart F. Effects of competition and cooperation on intrinsic motivation in a motor task.
Curr Psychol Cogn. 1991;11(5):521-36.
14. Gebel K, Bauman AE, Petticrew M. The physical environment and physical activity: a critical appraisal of review articles.
Am J Prev Med. 2007;32(5):361-9.
15. Jacobs DR, Ainsworth BE, Hartman TJ, Leon AS. A simultaneous evaluation of 10 commonly used physical activity questionnaires.
Med Sci Sports Exerc. 1993;25(1):81-91.
16. Johnson-Kozlow M, Matt GE. What respondents recall about walking and what self-report items elicit about walking.
Prev Med. 2004;38(2):227-36.
17. Kahn EB, Ramsey LT, Brownson RC, et al. The effectiveness of interventions to increase physical activity. A systematic review.
Am J Prev Med. 2002;22(suppl 4):73-107.
18. Kapur A, Bhat CR. Modeling adults' weekend day-time use by activity purpose and accompaniment arrangement.
Transp Res Rec. 2007;2021:18-27.
19. Korn EL, Graubard BI.
Analysis of Health Surveys. New York (NY): Wiley; 1999. p. 126-40.
20. Manini TM, Everhart JE, Patel KV, et al. Daily activity energy expenditure and mortality among older adults.
JAMA. 2006;296(2):171-9.
21. Marsh AP, Katula JA, Pacchia CF, Johnson LC, Koury KL, Rejeski WJ. Effect of treadmill and overground walking on function and attitudes in older adults.
Med Sci Sports Exerc. 2006;38(6):1157-64.
22. National Center for Health Statistics.
Health, United States, 2007 with Chartbook on Trends in the Health of Americans. Hyattsville (MD): National Center for Health Statistics; 2007. p. 286-7.
23. Ommundsen Y, Klasson-Heggebø L, Anderssen SA. Psycho-social and environmental correlates of location-specific physical activity among 9- and 15-year-old Norwegian boys and girls: theEuropean Youth Heart Study.
Int J Behav Nutr Phys Act. 2006;3:32.
24. 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. 1995;273(5):402-7.
25. Perri MG, Martin AD, Leermakers EA, Sears SF, Notelovitz M. Effects of group- versus
home-based exercise in the treatment of obesity.
J Consult Clin Psychol. 1997;65(2):278-85.
26. Physical Activity Guidelines Advisory Committee.
Physical Activity Guidelines Advisory Committee Report, 2008. Washington (DC): U.S. Department of Health and Human Services; 2008. p. E1-E22.
27. Research Triangle Institute.
SUDAAN Release 9.0. Research Triangle Park (NC): RTI International; 2005. p. 47-9.
28. Sallis JF, Cervero RB, Ascher W, Henderson KA, Kraft MK, KerrJ. An ecological approach to creating active living communities.
Annu Rev Public Health. 2006;27:297-322.
29. Salmon J, Timperio A. Prevalence, trends and environmental influences on child and youth physical activity.
Med Sport Sci. 2007;50:183-99.
30. Salvy SJ, Bowker JW, Roemmich JN, et al. Peer influence on children's physical activity: an experience sampling study.
J Pediatr Psychol. 2008;33(1):39-49.
31. Shelley KJ. Developing the American Time Use Survey activity classification system.
Mon Labor Rev. 2005;128(6):3-15.
32. Srinivasan S, Bhat CR. An exploratory analysis of joint-activity participation characteristics using the American Time Use Survey.
Transportation. 2008;35(3):301-27.
33. Steffen LM, Arnett DK, Blackburn H, et al. Population trends in leisure-time physical activity: Minnesota Heart Survey, 1980-2000.
Med Sci Sports Exerc. 2006;38(10):1716-23.
34. Tauer JM, Harackiewicz JM. The effects of cooperation and competition on intrinsic motivation and performance.
J Pers Soc Psychol. 2004;86(6):849-61.
35. Trost SG, Owen N, Bauman AE, Sallis JF, Brown W. Correlates of adults' participation in physical activity: review and update.
Med Sci Sports Exerc. 2002;34(12):1996-2001.
36. Tudor-Locke C, Bittman M, Merom D, Bauman A. Patterns of walking for transport and exercise: a novel application of time use data.
Int J Behav Nutr Phys Act. 2005;2:5.
37. Tudor-Locke C, van der Ploeg HP, Bowles HR, et al. Walking behaviours from the 1965-2003 American Heritage Time Use Study (AHTUS).
Int J Behav Nutr Phys Act. 2007;4:45.
38. Tudor-Locke C, Washington TL, Ainsworth BE, Troiano R. Linking the American Time Use Survey (ATUS) and the Compendium of Physical Activities: Methods and rationale.
JPhy Act Health. (in press). Available from:
http://riskfactor.cancer.gov/tools/atus-met/jpah_2007_tudor-locke_atus-met.pdf.
39. Turner L, Mermelstein R, Flay B. Individual and contextual influences on adolescent smoking.
Ann N Y Acad Sci. 2004;1021:175-97.
40. U.S. Bureau of Labor and Statistics Web site [Internet]. American Time Use Survey Users Guide: Understanding ATUS 2003-2006. June 2007; [cited 2008 May 8]. Available from:
http://www.bls.gov/tus/atususersguide.pdf. p. 24-25.
41. U.S. Census Bureau.
Current Population Survey Design and Methodology. Technical Paper 66; 2006. p. 14-1-3.
42. Wanko NS, Brazier CW, Young-Rogers D, et al. Exercise preferences and barriers in urban African Americans with type 2 diabetes.
Diabetes Educ. 2004;30(3):502-13.
43. Wendel-Vos W, Droomers M, Brug J, van Lenthe F. Potential environmental determinants of physical activity in adults: a systematic review.
Obes Rev. 2007;8:425-40.
44. Withers RT, Brooks AG, Gunn SM, Plummer JL, Gore CJ, Cormack J. Self-selected exercise intensity during household/garden activities and walking in 55 to 65-year-old females.
Eur J Appl Physiol. 2006;97(4):494-504.
