Share this article on:

Seasonal and Annual Variation in Young Children’s Physical Activity


Medicine & Science in Sports & Exercise: July 2012 - Volume 44 - Issue 7 - p 1318–1324
doi: 10.1249/MSS.0b013e3182464db5

Introduction It is well established that regular physical activity (PA) contributes to lower levels of morbidity and mortality. However, little is known about the stability of very young children’s PA habits across seasons and years. The aims of this study were to 1) examine the influence of season and increasing age on objectively assessed PA in preschool children and 2) examine the stability of young children’s PA rankings during 1 yr.

Methods The PA levels of preschool (3- and 4-yr-old) children were measured, using 6-d pedometer step counts, during winter and spring (n = 85, 52 boys). PA levels were measured again 1 yr after the spring data collection when the children had entered primary school (n = 37, 22 boys). Parents completed questionnaires to assess attitudes toward PA, PA habits, and demographic information in the winter of the first year and the spring of the second year.

Results Young children take approximately 2000 (20%) fewer steps per day in winter than in spring with a rank order stability between the two measures of r = 0.04 (P < 0.01). A modest degree of the observed intrachild or seasonal variation was related to the amount of time fathers played with their children (P < 0.05) and the availability of a safe place for children to play (P < 0.05). Children took approximately 2300 (20%) more steps per day at age 5 compared with age 4 (P < 0.01). The rank order stability of young children’s PA during this period was low with correlations ranging from 0.01 to 0.15.

Conclusions Results suggest that a one-off assessment of PA is unlikely to be representative of a young child’s activity during 1 yr and that PA tracks poorly from age 4 to 5.

1Department of Health and Physical Education, Stranmillis University College, Belfast, NORTHERN IRELAND; 2Department of Arts Education and Physical Education, Faculty of Education, Mary Immaculate College, University of Limerick, Limerick, IRELAND; 3Institute for Sport and Health, University College Dublin, Dublin, IRELAND; 4School of Sport, Performing Arts and Leisure, University of Wolverhampton, Wolverhampton, ENGLAND; and 5Sport and Exercise Science Research Institute, University of Ulster, Newtownabbey, NORTHERN IRELAND

Address for correspondence: Marie H. Murphy, Ph.D., Sport and Exercise Science Research Institute, University of Ulster, Newtownabbey, County Antrim BT37 0QB, Northern Ireland; E-mail:

Submitted for publication July 2011.

Accepted for publication December 2011.

There is widespread acceptance that regular physical activity (PA) is essential for good health and that the majority of children in Europe (28) and the United States (7) do not meet the recommended levels of PA (35). Given the importance of early intervention to address public health concerns, it is not surprising that PA in preschool settings has been identified as strategy targets in the US (36) and UK (10) national PA plans. However, inadequate data currently exist on the prevalence of preschool children’s PA levels (17). Understanding of preschool children’s PA patterns and levels is necessary to aid the development of effective strategies to promote PA in young children.

PA levels in young people change during the first decades of life (31); however, evidence on the amount and direction of any change in PA in young children with increasing age is equivocal. Whereas some longitudinal studies have reported increased levels of PA in 3- to 5-yr-old children (18,27), others have found a decrease between the ages of 3 and 4–5 yr (33) and significantly lower PA in first-grade children (6.7 yr) than in preschool children (5.7 yr) (32). This lack of consensus as to the effect of increasing age on observed PA in young children is further complicated by findings that seasonality of PA is not clear in children younger than 6 yr old (5). This may be partly explained by the use of cross-sectional designs (for example, Finn et al. [13] and Fisher et al. [14]), which do not account for the high degree of intersubject variability in the PA of young children (22) likely to mask any influence of seasonal variation. A prospective study of seasonal variation in the PA of young Texan children found that PA tended to fall during the summer months possibly influenced by a reduction in the amount of time spent outside in response to high ambient temperatures (1). Prospective studies examining the seasonality in PA of preschool children in Europe are lacking. If PA varies by season, children may lose health benefits associated with PA when levels fall (5). Attempts to increase PA in this group may be helped by an understanding of this seasonal variation.

An assumption that underlies health promotion programs in childhood and youth is that behavioral risk factors for chronic disease exhibit stability over time (26). Tracking or the tendency for an individual to maintain his/her disease risk factor rank level relative to his/her peers through time (15) is a useful predictor of a child’s later PA behavior (19). However, associations between PA and health outcomes can be weakened by regression dilution when the exposure (PA) contains large measurement error, and this could lead to important associations between PA and health outcomes being missed. Hence, it is important to establish the stability of PA within specific populations. Relatively little is known about the tendency of PA to track during childhood (26). Low to moderate levels of tracking have been reported during a 2-yr period in children age 3.8 yr at baseline (19) and during a 1-yr period in 3- to 4-yr-old children (18). The lack of consensus in findings on seasonal and annual variation of young children’s PA levels and tracking of PA may be due to both a paucity of data, particularly concerning children younger than 5 yr, and methodological limitations of existing research, including a lack of objective measurement tools for PA. The present study will augment the limited body of evidence on these PA constructs in children younger than 5 yr. An understanding of PA behavior will enable interventions to be developed and tailored to meet the specific needs of preschool children.

The aims of the current study were to use a prospective design to (a) examine the influence of season and increasing age on objectively assessed PA in young children and (b) examine young children’s stability of PA rankings during 1 yr. In addition, the study sought to examine correlates of any observed change in PA between seasons.

Back to Top | Article Outline



Ethical approval for the study was received from the local research ethics committee. Four hundred seventy-four children from nine suburban preschools in the greater Belfast area of Northern Ireland were invited to take part in the study. These schools were selected on a convenience basis because they had previously collaborated with university projects. Written informed consent was obtained from the parents of 151 children. Eighty-five (33 girls) of these provided data for both winter and spring. All of these children were asked to undertake another PA assessment 13–14 months after the spring data collection, with 37 children (22 boys) supplying data at this final collection. There are many reasons for the loss to follow-up over the three measurement occurrences including some children not providing adequate days of monitoring or adequate duration of monitoring on each day, some children refusing to wear the pedometer on the second and/or third measurement occasion, some parents indicating that they no longer wished to continue with the study, some children leaving the area, and all the children transferring to new schools between the first and the last assessment periods.

Back to Top | Article Outline



PA was assessed via pedometers (Digiwalker DW-200; Yamax, Tokyo, Japan). The pedometers have previously been shown to have functional utility in assessing PA in children (12), including those of preschool age (22). Mean daily counts were calculated for all participants during four weekdays and two weekend days.

Parents were sent a pack containing a pedometer, a recording sheet for the pedometer scores, cable ties to seal the pedometer, and detailed instructions as to the placement and use of the pedometers. Parents were requested to ensure their child wore the pedometer for six consecutive days (including two weekend days), recording the scores on a daily basis together with the time the pedometer was fitted and removed and any times during the day when the instrument was removed (i.e., for bathing or swimming). All participants were requested to fit the pedometer as soon as practical in the morning and to wear the device until going to bed that evening. Data included in the analysis had a minimum of 9 h of monitoring per day and at least three weekdays and one weekend day. Whereas other researchers have chosen more hours per day (28) as inclusion criteria, we selected a shorter day because these young children spend fewer hours awake.

Back to Top | Article Outline

Anthropometric measures.

Children’s height was recorded to the nearest millimeter on a portable stadiometer (Leicester Height Meter; Child Growth Foundation, London, United Kingdom), and body mass was recorded to the nearest 0.1 kg on Seca 707 digital physician’s scales (Vogel and Halke, Hamburg, Germany) using standard procedures (20). At each time point, these measurements were taken at the start of the school day. Parents recorded their height and body mass on a self-report form. Body mass index (BMI) was derived by dividing mass in kilograms by height in meters squared. With the inclusion of gender, date of birth, and date of assessment, raw BMI scores were converted to SD scores on the basis of the 1990 UK norms (8). Waist and hip circumferences were calculated as the average of two measures with an inelastic flexible tape measure (Rollfix; Hoechstmass Balzer, Sulzbach, Germany) using established procedures (20). Children’s anthropometric measures were assessed in the winter of year 1 and the spring of year 2 in the schools the children attended.

Back to Top | Article Outline

Seasonal assessment, weather, and parental questionnaire.

The seasons for PA assessment were selected after consultation with the teachers in the settings. Briefly, advice indicated that the young children (3 yr old) would be more settled in their school environment by the December of the first term, and hence, December and January were chosen as the first sampling point. In addition, December and January represent the shortest hours of daylight in the year. April was chosen to represent spring because by this stage of the year, historical meteorological data indicated a substantial increase in the duration of daylight. At the project planning stage, it was intended to follow this assessment with another in June (the last month in which the children attended nursery school). However, feedback from parents indicated that compliance with a third assessment period in a relatively short period was likely to be low. The team was advised by teachers against assessing the children in the following fall because all the children had just entered new primary schools. Subsequently, the children were reassessed in the May of the following year to allow an assessment of the stability of PA during 1 yr while controlling for any potential seasonal effect. Data on local weather conditions (rain, hours of sunshine, duration of daylight, wind velocity, and temperature variables) for the specific period of the study were obtained from the meteorological office. A modified version of a previously used lifestyle questionnaire (38) was used to collect data about the children and their parents. The questionnaire was designed by trained and experienced epidemiologists who had involvement with similar populations and included questions about aspects of neighborhood safety, the amount of time mothers and fathers played with children, and the amount of time children spent outdoors. The questionnaire was administered during the winter data collection in the first year and the spring data collection in the second year. The teachers and teaching assistants were asked to complete a five-point rating of each child’s PA (low activity to highly active). This measurement occurred during the winter data collection of the first year.

Back to Top | Article Outline


The effects of season and gender on PA were assessed via repeated-measures ANOVA. Gender differences in PA in year 1 winter and spring measures were evaluated with t-tests. In addition, t-tests were used to examine differences in winter PA between children who had their PA assessed in winter and then dropped out of the study and those who had their PA assessed in both winter and spring. t-tests were also used to examine differences in the daily weather data supplied from the meteorological office for the winter and spring study periods. A Spearman rank order correlation (SRO) was calculated between year 1 winter and spring measures to assess the stability of PA ranking in children. In an attempt to evaluate potential causes of seasonal variability in PA, SRO and Pearson product moment correlations were used as appropriate to compare the observed difference in PA between the seasons with a range of variables assessed from the lifestyle questionnaire completed by parents. Because multiple bivariate correlations between the observed change in PA from winter and spring and the range of variables obtained from the lifestyle questionnaire cannot fully explain observed effects in a multivariate predictor variable space, the initial multiple bivariate analysis was used in an exploratory fashion to identify potential covariates for a repeated-measures ANOVA using winter PA and spring PA as dependent variables and variables from the lifestyle questionnaire as continuous covariates and between-subject categorical factors as appropriate. Variables were entered into the model using a liberal P value of 0.10 and were removed from the model using backward elimination if they were found not to achieve P ≤ 0.05.

The effects of year of measurement and gender on PA were assessed via a between-/within-subjects ANOVA. An SRO was calculated between PA at 4 and 5 yr to assess the stability of PA ranking in children. Statistical analysis was carried out using SPSS v20 for Windows. The α level was set at P ≤ 0.05.

Back to Top | Article Outline



Anthropometric data for the 85 children who provided data during winter and spring and the 37 of these children who repeated measurements 1 yr later are presented in Tables 1 and 2. The t-test results indicated no significant differences between children who completed only the initial winter assessment and children who completed both the winter and spring assessments (mean difference (95% confidence interval (CI)): height = −1.7 (−4.0 to +0.5) cm, mass = −0.2 (−1.5 to +1.0) kg, BMI = 0.4 (−0.4 to +1.2) kg·m−2, all not significant (NS)). There was no evidence to support differences in the initial anthropometric variables between children who completed the initial winter assessment and those who completed the final assessment in the spring of the following year (mean difference (95% CI): height = 0.5 (−1.4 to +2.5) cm, mass = 0.3 (−0.7 to 1.4) kg, BMI = 0.1 (−0.6 to +0.8) kg·m−2, all NS).





Back to Top | Article Outline

Seasonal variation.

The PA results (Table 3) indicate a high degree of variability in the PA levels of young children as evidenced by the large SD. There was no evidence of a difference in the winter measurements between the children who had their PA assessed in winter and spring and those who dropped out of the study after the initial measurement during winter, nor were there any gender differences. Furthermore, there was no evidence of a difference in the seasonal effect on weekdays and weekend days; therefore, subsequent results were analyzed and presented on the basis of average daily counts over all the days monitored, i.e., weekdays and weekend days.



There was no evidence to support a gender difference in the PA levels of the current sample. The ANOVA results showed a main effect for season (P = 0.0001), but this was not differentiated by gender. There was a substantive difference in the children’s PA between winter and spring of 1930 steps per day (3252 steps per day, 95% CI = 1229–2632 steps per day). This difference was not attributable to differences in the time the children wore the pedometers for (mean difference between winter and spring = 0.38 ± 0.74 h, 95% CI = −0.20 to +0.12 h). The SRO correlation between winter and spring PA was r = 0.40 (P < 0.01).

Back to Top | Article Outline

Correlates of seasonal variation.

Correlations between the seasonal difference in PA and a range of variables are presented in Table 4. Repeated-measures ANOVA results indicated that fathers’ daily play with children and child access to a safe place to play were related to PA in winter and spring. Specifically, the duration (min) of fathers’ daily play with children was related to winter PA (β = 13 steps, 95% CI = 0.4–25.3, P = 0.043), whereas having a safe place to play was related to spring PA (β = 1484 steps, 95% CI = 124–2845, P = 0.033). Selected weather variables are presented in Table 5. Mean temperature, maximum temperature, hours of sunshine, and duration of daylight were significantly higher in spring. The differences in the average daily rainfall and wind speed between the two seasons were NS.





Back to Top | Article Outline

PA levels of 4- and 5-yr-olds.

The PA results (Table 6) indicate a high degree of variability in the PA levels of young children as evidenced by the large SD with greater variation at the weekends. There was no evidence to support a gender difference in the PA levels of the current sample, whereas the ANOVA results showed a main effect for year of measurement in weekly (P < 0.01), weekday (P < 0.05), and weekend (P < 0.01) PA. The differences between weekly, weekday, and weekend PA at 4 and 5 yr were 2917 (3854, 95% CI = 1572–4262), 2443 (3923, 95% CI = 1071–3816), and 3673 (6357, 95% CI = 1455–5891) steps per day. This difference was not attributable to differences between the years in the time the children wore the pedometers for (mean difference between the time the children wore the pedometers on a weekly, weekday, and weekend basis was 0.05 ± 0.88, 0.07 ± 0.97, and 0.33 ± 1.28 h, all NS). Correlations between repeated assessment of weekly, weekday, and weekend PA in the 2 yr revealed low rank order stability of PA (r = 0.013 − 0.151).



Back to Top | Article Outline


The current study augments the limited body of evidence internationally concerning PA in children <6 yr old. Our results indicate that young children take just under 2000 (or approximately 20%) fewer steps per day in winter than in spring. This reduction is not attributable to the children wearing the pedometer for shorter periods. Therefore, a central finding of this study is that one 4- to 5-d assessment of PA is unlikely to be representative of a child’s activity during an entire year. Our finding of a seasonal effect is in keeping with data from preschool children in Canada (summer and autumn higher than winter [6]) and 11- to 12-yr-old UK children (summer higher than winter [21]) and consistent with the trend toward a seasonal effect within 4- to 10-yr-old children from Vermont and Alabama (spring higher than autumn [16]) and preschool children from Texas (winter higher than the summer, the latter being characterised by extreme high temperatures [1]).

Fathers’ daily play with children and child access to a safe place to play were related to the observed variation in PA. For every minute that fathers played with their children, there was an increase of 13 steps in winter PA levels. This in turn led to PA being more stable across the seasons in children who had greater play with their father. Parental support and encouragement for PA is positively associated with children’s PA participation (37,40), and children’s overall PA increases if parents are physically active with their child (30). The association between the time that fathers played with their children and child PA found in the present study may be explained by the different ways that mothers and fathers influence their child’s activity (2). For example, whereas mothers provide higher levels of logistic support, fathers are more likely to use their own behavior to encourage activity (9). Further research is needed to examine the distinct influences of father versus mother coparticipation in PA with their children and the subsequent effect on child PA levels. In the present study, access to a safe place to play contributed to children having a higher level of PA in spring and an increased variation in PA between winter and spring. A safe environment is thought to be crucial to increasing opportunities of PA (39). A possible explanation for our finding may be contained in the weather data because average daily temperature and hours of daylight have been shown to be related to PA (5,14). In winter, the days are shorter with a lower average temperature and fewer hours of sunshine than in spring (Table 5). These factors may have influenced time spent outdoors, which has previously been shown to be positively related to PA in preschool children (17). Thus, where safe areas to play were present, the children in the present study may have used them less in the winter and more in the spring. This is of public health concern because children may lose the immediate health benefits of sustained participation in PA if levels fall during the winter period.

Girls and boys reported mean daily step counts of 11,177 and 11,736, respectively, at the age of 4 yr. These PA scores are greater than that reported in Belgian preschoolers, i.e., 9980 steps per day (4). At age 5, step counts for girls in the present study were greater than those recorded for 5-yr-old girls in a Czech study, but similar step counts were found for boys in the present study and their Czech counterparts (32). In contrast with previous studies (3,17,25,34), there was no evidence of gender differences in the present study for PA levels. A longitudinal study found no difference in PA levels between sexes at age 3, although by age 5, PA was higher in boys than in girls (27). Thus, it may be that the gender difference in PA levels of the present cohort may not appear until later, highlighting the need for future research to quantify the onset and magnitude of gender difference in PA levels of young children.

Children took approximately 2300 (20%) more steps per day at the age of 5 yr compared with the age of 4 yr. The magnitude of the increase is similar to the 22% increase in accelerometer-assessed total activity counts during 2 yr reported previously in children who were 3.8-yr-old at the start of the study (19). The direction of the change in PA is also in agreement with results from Scotland in children measured at the ages of 3 and 5 yr (27) and the ages of 3.7 and 4.7 yr (18). Increases in the PA levels of this age group may be due to advancing motor proficiency in young children in the early years. Previous work has demonstrated that the relationship between level of motor skill performance and PA participation was stronger for 4-yr-olds than for 3-yr-olds (41). However, other studies report an age-related decrease in the PA level of young children from the age of 5.7 to 6.7 yr (32) and from the age of 3 to 4 and 5 yr (33) or no change from 5 to 8 yr (24). It is worth noting that Scottish children who were participants in the aforementioned three studies reporting an increase in PA (18,19,27) are subject to a similar climatic and educational milieu as the children reported in this study.

The rank order stability of children’s PA during 1 yr found in the present study was very low with correlations ranging from 0.01 to 0.15. It seems that in the transition from a preschool environment to primary school, the most active children tend to fall in PA ranking compared with their peers. Therefore, the move to the more structured and likely longer formal school day presented a challenge for the more active children to maintain their PA ranking. In keeping with local curricular guidance, children in the current study typically spent 3 h in an informal play-based educational setting at the age of 4, moving to a more structured formal educational setting of 5 h·d−1 at the age of 5. Few studies have examined this specific time frame, and further work is essential if school programs are to promote PA to those at most risk of regression. A dearth of objectively measured data in young children younger than 6 yr makes comparison with other studies limited. Three longitudinal studies using accelerometer data have reported correlations higher than those in the present study of 0.35 to 0.55 (18,19,23). The low tracking in our children is difficult to explain because the aforementioned studies were United Kingdom based and involved children of a similar age. Anecdotally, some parents (14%) reported on the pedometer record sheets that the weather was particularly good or that their children were outdoors more than normal during the PA assessment at 5 yr. Thus, some of the children may have recorded higher than normal PA due in part to the good weather conditions and/or exposure to the outdoors, both of which have previously been linked to PA level (1,29).

Back to Top | Article Outline

Strengths and limitations.

Strengths of the current study include the use of an objective measurement of PA level and its longitudinal design, following the same children during a 1-yr period. The narrow age range of the children at baseline reduced the potential effect of age-related influences on PA behavior. The study also followed children during the transition from preschool to a more structured primary school environment, which may be a time when more sedentary classroom behavior is encouraged. Limitations of the study include the small sample size at 1-yr follow-up, the possibility of self-report bias in the parental questionnaire (11), and the self-selection of participants into the study. Finally, failure to observe stronger evidence of tracking or gender difference could relate to a lack of power, a feature common to many tracking studies (19).

In conclusion, our findings suggest that, because of seasonal variation in young children’s PA, a once-off assessment of PA is unlikely to be representative of a child’s activity during a year. Although children’s mean PA for the sample increased from the age of 4 to 5 yr, the rank order stability was low. Thus, some young children may be at risk for adopting poor levels of PA, highlighting the need for earlier identification and intervention.

No funding was received for this study.

The authors have no conflict of interest to declare.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Baranowski T, Thompson W, DuRant R, Baranowski J, Puhl J. Observations on physical activity in physical locations: age, gender, ethnicity, and month effects. Res Q Exerc Sport. 1993; 64 (2): 127–33.
2. Beets MW, Vogel R, Chapman S, Pitetti KH, Cardinal BJ. Parent’s social support for children’s outdoor physical activity: do weekdays and weekends matter? Sex Roles. 2007; 56 (1): 125–31.
3. Belton S, Brady P, Meegan S, Woods C. Pedometer step count and BMI of Irish primary school children aged 6–9 years. Prev Med. 2010; 50: 189–92.
4. Cardon G, De Bordeaudhuij I. Comparison of pedometer and accelerometer measures of physical activity in preschool children. Pediatr Exerc Sci. 2007; 19: 205–14.
5. Carson V, Spence J. Seasonal variation in physical activity among children and adolescents: a review. Pediatr Exerc Sci. 2010; 22 (1): 81–92.
6. Carson V, Spence J, Cutumisu N, Boule N, Edwards J. Seasonal variation in physical activity among preschool children in a northern Canadian city. Res Q Exerc Sport. 2010; 81 (4): 392–9.
7. Centers for Disease Control and Prevention. State Indicator Report on Physical Activity, 2010. Atlanta (GA): US Department of Health and Human Services; 2010. p. 12.
8. Cole T, Freeman J, Preece M. Body mass index reference curves for the UK, 1990. Arch Dis Child. 1995; 73 (1): 25–9.
9. Davison KK, Cutting TM, Birch LL. Parents’ activity-related parenting practices predict girls’ physical activity. Med Sci Sports Exerc. 2003; 35 (9): 1589–95.
10. Department of Health. Choosing Activity: A Physical Activity Action Plan. London: UK Department of Health; 2005. Chapter 3, Children and young people: starting on an active path; p. 13–7.
11. Dollman J, Okely AD, Hardy L, Timperio A, Salmon J, Hills AP. A hitchhiker’s guide to assessing young people’s physical activity: deciding what method to use. J Sci Med Sport. 2009; 12 (5): 518–25.
12. Eston RG, Rowlands AV, Ingledew DK. Validity of heart rate, pedometry, and accelerometry for predicting the energy cost of children’s activities. J Appl Physiol. 1998; 84 (1): 362–71.
13. Finn K, Johannsen N, Specker B. Factors associated with physical activity in preschool children. J Pediatr. 2002; 140 (1): 81–5.
14. Fisher A, Reilly J, Montgomery C, et al.. Seasonality in physical activity and sedentary behavior in young children. Pediatr Exerc Sci. 2005; 17 (1): 31–40.
15. Fuentes R, Notkola I, Shemeikka S, Tuomilehto J, Nissinen A. Tracking of body mass index during childhood: a 15-year prospective population-based family study in eastern Finland. Int J Obes Relat Metab Disord. 2003; 27 (6): 716–21.
16. Goran M, Nagy T, Gower B, et al.. Influence of sex, seasonality, ethnicity, and geographic location on the components of total energy expenditure in young children: implications for energy requirements. Am J Clin Nutr. 1998; 68 (3): 675–82.
17. Hinkley T, Crawford D, Salmon J, Okely A, Hesketh K. Preschool children and physical activity: a review of correlates. Am J Prev Med. 2008; 34 (5): 435–41.
18. Jackson D, Reilly J, Kelly L, Montgomery C, Grant S, Paton J. Objectively measured physical activity in a representative sample of 3- to 4-year-old children. Obes Res. 2003; 11 (3): 420–5.
19. Kelly L, Reilly J, Jackson D, Montgomery C, Grant S, Paton J. Tracking physical activity and sedentary behavior in young children. Pediatr Exerc Sci. 2007; 19 (1): 51–60.
20. Lohman TG, Roche AF, Martorell R. Anthropometric Standardization Reference Manual. Champaign (IL): Human Kinetics; 1988. p. 122–4.
21. Mattocks C, Leary S, Ness A, et al.. Intraindividual variation of objectively measured physical activity in children. Med Sci Sports Exerc. 2007; 39 (4): 622–9.
22. McKee D, Boreham C, Murphy M, Nevill A. Validation of the Digiwalker pedometer for measuring physical activity in young children. Pediatr Exerc Sci. 2005; 17 (4): 345–52.
23. Metcalf B, Mallam K, Voss L, et al.. The regulation of physical activity in young children. Educ Health. 2004; 22 (4): 61–4.
24. Metcalf B, Voss L, Hosking J, Jeffery A, Wilkin T. Physical activity at the government-recommended level and obesity-related health outcomes: a longitudinal study (Early Bird 37). Arch Dis Child. 2008; 93 (9): 772–7.
25. Murtagh E, Murphy M. Active travel to school and physical activity levels of Irish primary schoolchildren. Pediatr Exerc Sci. 2011; 23 (2): 230–6.
26. Pate R, Baranowski T, Dowda M, Trost S. Tracking of physical activity in young children. Med Sci Sports Exerc. 1996; 28 (1): 92–6.
27. Reilly J, Jackson D, Montgomery C, et al.. Total energy expenditure and physical activity in young Scottish children: mixed longitudinal study. Lancet. 2004; 363 (9404): 211–2.
28. Riddoch C, Mattocks C, Deere K, et al.. Objective measurement of levels and patterns of physical activity. Arch Dis Child. 2007; 92 (11): 963–9.
29. Sääkslahti A, Numminen P, Niinikoski H, et al.. Is physical activity related to body size, fundamental motor skills, and CHD risk factors in early childhood? Pediatr Exerc Sci. 1999; 11: 327–40.
30. Saelens BE, Kerr J. The family. In: Smith A, Biddle S, editors. Youth Physical Activity and Sedentary Behaviour: Challenges and Solutions. Champaign (IL): Human Kinetics; 2008. p. 267–94.
31. Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc. 2000; 32 (5): 963–75.
32. Sigmund E, Sigmundová D, El Ansari W. Changes in physical activity in pre-schoolers and first-grade children: longitudinal study in the Czech Republic. Child Care Health Dev. 2009; 35 (3): 376–82.
33. Taylor R, Murdoch L, Carter P, Gerrard D, Williams S, Taylor B. Longitudinal study of physical activity and inactivity in preschoolers: the FLAME study. Med Sci Sports Exerc. 2009; 41 (1): 96–102.
34. Tucker P. The physical activity levels of preschool-aged children: a systematic review. Early Child Res Q. 2008; 23 (4): 547–58.
35. US Department of Health and Human Services. 2008 Physical Activity Guidelines for Americans Washington, DC: US Department of Health and Human Services; [Internet]. 2008 [cited 2009 Jun 22]. Available from:
36. US National Physical Activity Plan Coordinating Committee. National Physical Activity Plan [Internet]. 2010 [cited 2011 Sep 23]. Available from:
37. Van Der Horst K, Paw MJCA, Twisk JWR, Van Mechelen W. A brief review on correlates of physical activity and sedentariness in youth. Med Sci Sports Exerc. 2007; 39 (8): 1241–50.
38. Watkins D, Murray L, McCarron P, et al.. Ten-year trends for fatness in Northern Irish adolescents: the Young Hearts Projects—repeat cross-sectional study. Int J Obes (Lond). 2005; 29 (6): 579–85.
39. Weir LA, Etelson D, Brand DA. Parents’ perceptions of neighborhood safety and children’s physical activity. Prev Med. 2006; 43 (3): 212–7.
40. Whitt-Glover MC, Taylor WC, Floyd MF, Yore MM, Yancey AK, Matthews CE. Disparities in physical activity and sedentary behaviors among US children and adolescents: prevalence, correlates, and intervention implications. J Public Health Policy. 2009; 30: S309–34.
41. Williams HG, Pfeiffer KA, O’Neill JR, et al.. Motor skill performance and physical activity in preschool children. Obesity (Silver Spring). 2008; 16 (6): 1421–6.


©2012The American College of Sports Medicine