Friday 2 November Posters / Journal of Science and Medicine in Sport 15 (2012) S265–S327 S299 was to evaluate the agreement between PA levels estimated by the original IPAQ and IPAQ with the EDW Scale. Methods: The EDW Scale–“Mo Tu We Th Fr Sa Su”–allows partic- ipants to indicate on which particular day/s of the week they were physically active. To evaluate the scale, IPAQ (short) was adminis- tered on two occasions, 2 hours apart, to a convenience sample of 89 university students in Austria (36% female, mean age 20 ± 2 yrs). About half of the participants completed the original IPAQ on the first occasion and IPAQ with the EDW Scale on the second occasion, while the other half did it in reverse order. Results: Intraclass correlation coefficient between total weekly level of PA estimated using the original IPAQ and IPAQ with the EDW Scale was 0.88 (95% CI, 0.83–0.92, n = 89). The difference between median PA levels (d = 2.90 MET-hour/week) was not sig- nificant (Wilcoxon test, p = 0.37). Discussion: The EDW Scale provides useful data on the weekly distribution of PA and for the health-related classification of PA, which are imprecisely estimated if the frequency of PA is assessed using open-ended or ordinal-scale questions. These pilot results indicated that the implementation of the EDW Scale in IPAQ does not significantly change PA estimates, which justifies further assessment of its measurement properties. A team of researchers from Australia, Austria, Croatia, Czech Republic, Finland, Serbia, Switzerland and United Kingdom have commenced further testing of the EDW Scale in the IPAQ, Global Physical Activity Questionnaire and Single-Item Physical Activity Measure. http://dx.doi.org/10.1016/j.jsams.2012.11.726 724 Accelerometer use in children: Differences in weekend compli- ance A. Pizarro ∗ , J. Ribeiro, J. Mota, P. Silva, E. Marques, M. Santos Porto University, Faculty of Sports, CIAFEL Introduction: Physical activity (PA) benefits in children are widely recognized however an accurate assessment of the PA habit- ual patterns are crucial, to avoid misinterpretations. PA patterns during weekdays and weekends is known to differ considerable; nonetheless compliance with using accelerometer is greater on weekdays than in weekends which compromise a reliably estimate of habitual PA, and impose restrictions to data analysis in this field. Therefore the aim of our study was to investigate the effect of the day accelerometer was initially placed, on the weekend compliance with wearing accelerometer. Methods: Participants were 479 students with a mean age of 11.6(±0.8) years old. 81 were assigned to start using accelerome- ter on a Monday, 107 on a Tuesday, 83 on a Wednesday, 118 on a Thursday and 89 on a Friday. PA was measured using Actigraph accelerometers, model GT1 M. Participants were educated to use the accelerometer attached to an elastic belt and placed above the right iliac crest for 7 consecutive days. Instructions were given to wear the monitor all times except when sleeping or doing water activities. Data was analyzed with Actlife software and inclusion criteria consisted of a minimum recording of 8 h; 60 min of consec- utive zeros were considered invalid data. Results: In average, children wore the monitors for 5 days if placed on a Monday or Tuesday and for 6 days when the moni- tor was placed on the remaining weekdays. Significant differences were found in general compliance with accelerometer use. Friday (65.2%) and Wednesday (59%) obtained a higher percentage of com- pliance within 7 days of accelerometer use. Qui-Square showed differences in compliance with wearing the accelerometer at least one weekend day and the weekday of initial monitoring (p < 0.05). Excellent compliance rates were obtained for at least one weekend day for children starting the accelerometer-based monitoring on a Friday (92.1%), good compliances were achieved also on Thurs- days (81%) and Wednesdays (76%). Greater compliance (76.4%) with wearing the accelerometer on both weekend days was achieved when children wore the accelerometer for the first time on a Fri- day, followed by Thursday (65%). The worst compliance for wearing the accelerometer on both weekend days was observed in children wearing the accelerometer for the first time on a Tuesday (29%). Discussion: For a precise assessment of habitual patterns of PA in children, especially during weekends, the accelerometer use should be initiated on a Friday as it revealed greater compliances on both weekend days and weekdays. Supported by FCT grant: PTDC/DES/099018/2008, FCT/FCOMP- 01-0124-FEDER-009573 and SFRH/BD/70513/2010. http://dx.doi.org/10.1016/j.jsams.2012.11.727 725 Translation of children’s cycling into steps: The share of cycling in 10 year-olds physical activity A. Raustorp 1,2,3,∗ , C. Boldemann 4 , F. Mårtensson 5 , C. Sternudd 6 , M. Johansson 6 1 Linnaeus University, School of Sport Sciences, Kalmar, Sweden 2 Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden 3 Department of Food, Nutrition and Sport Sciences, University of Gothenburg, Sweden 4 Department of Public Health Sciences, Karolinska Institutet, Stock- holm, Sweden 5 Department of Work Science, Business Economics and Environmental Psychology, Swedish University of Agricultural Sciences, Alnarp 6 Department of Architecture and the Built Environment, Lund Univer- sity, Lund, Sweden Background: Active travel is a potential way to increase chil- drenˇ ıs physical activity (PA). Pedometers are a cost-effective and valid option to measure and communicate PA, however unable to capture cycling activity. Children’s self-reported distances travelled by cycling can be analysed by Geografic Information System (GIS) Aim: The aim was to combine pedometry and GIS mapping to identify the relative amount of cycling in children’s PA. Methods: Of all fourth-graders (N = 187) in Staffanstorp, Sweden 167 (89%) participated. Of these 94 (55%) (49 girls) reported at least one cycle distance and for 88 of them pedometer data of steps/day were available as well. All individual cycle trips were entered into GIS and calculated into a total cycling dis- tance for the investigated week. Previously reported cycling speed, 13,5 km/h, was used to calculate cycling minutes cycleminutesper- day = (cyclelengthmeanpday/(13.5/3.6))/60. Expenditure of 4 and 5 METs has in children been reported equivalent to 122 and 127 steps/min respectively. We estimated 4,7 MET (13,5 km/h) as 126 steps for every minute of cycling (127–122 = 5 × 0,7 = 125,5). Results: Daily mean cycle distance was 676 metres, on average 379 additional steps/day for cycling children (min 21, max 1385, SD = 299) with no gender difference. Additional “steps” obtained by cycling corresponded to on average 3% of their PA. Recorded mean steps per day did not correlate with either the reported number of cycle distances (r = -.12, ns) nor the average distance cycled per day as measured in GIS (r = -.07, ns). Discussion: Pedometers like most motion sensors are unable to capture cycling activity. This lack of information has concerned