ABSTRACT
M.H. Murphy3 and A.M. Nevill4 1Stranmillis University College, Belfast, Northern Ireland; 2Institute for Sport and Health, University College Dublin, Ireland; 3School of Sports Studies, University of Ulster, Newtownabbey, Northern Ireland; 4Research Institute for Health Care Sciences, University of Wolverhampton, Wolverhampton, England
The accurate assessment of physical activity (PA) in young children has been identified as a research priority (Fulton et al., 2001). However, PA is amultifaceted behaviour, which is difficult to assess accurately (Wareham and Rennie, 1998). Further, there appears to be a relationship between the validity of a measure, the invasiveness of the assessment and the cost of the method (Montoye et al., 1996). Traditionally research interest has tended to focus on the concurrent validity of instruments to assess PA. To this end there is a consensus emerging that accelerometers (Finn and Specker, 2000; Pfeiffer et al., 2006), heart rate monitors (Sirrard and Pate, 2001), direct observation (Puhl et al., 1990) and pedometers (McKee et al., 2005; Hands et al., 2006) have acceptable validity for assessing PA in young children and that four to five days of monitoring yields valid information about weekly activity patterns (Trost et al., 2000). However, what is less clear is the validity of an individual four-or five-day assessment of PA as an indication of typical PA and the seasonal variation of PA in young children. Using crosssectional analysis with measurements of PA in different cohorts of children in the various seasons researchers have found children to be more active in Autumn than inWinter (Burdette et al., 2004), slight difference across seasons with Spring lower than Summer, Autumn or Winter (Fisher et al., 2005) or a trend towards a seasonal difference (Goran et al., 1998). This cross-sectional approach to assessing seasonal effects in the PA of children is beset with difficulties as it has been shown that there is high variability in the PA of young children (McKee et al., 2005) which is likely to mask the influence of seasonal variation. Using a more robust prospective designBaranowski et al. (1993) conducted a study of seasonal variation in the PAof youngTexan children and found PA tended to fall during the Summer months, possibly influenced by a reduction in the amount of time spent outside. The purpose of the current study was to employ a prospective design to examine the influence of season on PA in young Irish children.
