The purpose of this study was to examine the prevalence of physical activity and sedentary behaviours in a sample of Greek children and adolescents in Cyprus and present evidence on the association between these two behaviours. On the whole, 52.3% of the participants met current guidelines recommending that young people should engage in moderate to vigorous physical activity for at least 60 minutes per day. These prevalence estimates are slightly higher than self-reported estimates reported in the US and Canada [14, 15]. While, in general, about one third of young people from European countries meet these recommendations, according to Armstrong and Welsman  comparison between countries should be made with caution as wide variations are observed across countries. Further, comparison is even more complicated because of the different measures adapted in each study to measure physical activity. For example, studies using accelerometers in national and international studies, indicate that the proportion of adolescents meeting these recommendations vary between 2.0 to 61.0%  and 62.0% to 97.6% . As this is the first study that presents data on physical activity prevalence based on international guidelines among Cypriot youth from different ages, it may be used for comparison purposes until more data from a more representative sample using objective measures of physical activity is obtained.
Our results indicate that boys are more active than girls across all levels of education with the highest prevalence estimates observed among boys from middle and primary schools (69.4% and 68.2% respectively) and the lowest among girls from technical and high schools (26.4% and 34.4% respectively). Further, their appears to be a marked decline in children's physical activity levels after middle school (i.e. after 14-15 years of age) whereby the overall percentages of physical active adolescents in primary and middle schools were 63.2% and 59.8% respectively while the respective percentages for high and technical schools were 37.8% and 44.3%. These findings are in agreement with studies from European countries [16, 27] and from North America [14, 35] indicating gender and age related differences in physical activity levels. Interestingly, in the study by Whitt-Glover et al.  age-related differences were observed from the age of 12, while in the current study the marked decrease was observed in the age of 15. This may be partly explained by the increased homework obligations among older students. These findings suggest that girls may be especially targeted for physical activity interventions as well as children older than 15 in order to reduce the marked decline of physical activity observed.
Sedentary activities that children devoted most of their time to included television watching (2.7 hours per day), listening to music (2.5 hours per day), in front of the computer (2.1 hours per day) and doing homework (1.7 hours per day). These findings are similar to studies from Scotland  and Hungary  where television watching, doing homework, and playing computer/video games were among the top five most time consuming sedentary activities. The only large effect size difference observed between boys and girls was in hours per day playing video games (means were 2.1 and 0.8 respectively), a finding that confirms findings from other countries [37, 38]. Further, mean hours per day spent watching television in the current study are within the range (1.8 to 2.8 hours per day) reported in a review study by Marshall et al. . Total daily hours spent on screen-based activities (television, video games, DVDs, computer) and non screen-based sedentary activities (talking on the phone, listening to music and motorized transport) were 7.7 and 5.4 respectively. While time spent in both of these types of sedentary behaviours appears to be extensive, it should be noted that screen based activities may be done concurrently with non-screen based activities such as watching television and listening to music or talking on the phone. Interestingly, both of these values observed in the current study are in the range of 5.5 to 8.5 accelerometer derived mean hours per day spent in sedentary activities reported by Whitt-Glover et al.  in a large sample of adolescents from the US.
The finding that about half (52.4%) of the adolescents met the recommendation of watching television for less than two hours per day indicates that there is a need to reduce the time spent in front of the television. These estimates are comparable to data from 11-15 year-old children from mainland Greece  and 15-16 year-old Finnish adolescents  but are more favorable than estimates from children in Italy  and Canada  where 38.0% and 25.0% of children respectively met the recommendation of watching television for less than two hours per day. In our study, boys were more likely to meet the recommendation in comparison to girls (55.3% versus 49.3% respectively), a finding that contradicts findings from previous studies [17, 19, 39]. This finding may be partly explained by the large amount of time that boys spent in other competing sedentary activities such as video game playing, or other pursuits such as sports clubs attendance as observed in this study. Boys from primary and high schools were most likely to meet recommendations (65.7% and 58.6% respectively) and girls from technical and middle schools were the least likely to meet recommendations (37.0% and 44.7% respectively). In general, our findings support previous work that suggests that the percentage of children that meets the recommendations decreases as they grow older [38, 39]. Of interest is the low percentage of girls from technical and middle schools that meet recommendations, a finding that suggests that these groups should be especially targeted for intervention programmes.
In general, our findings provide limited support for the displacement hypothesis, as only two significant associations in the subgroup analyses were observed between physical activity and sedentary behaviours. Interestingly, boys who listened to music for less than one hour per day were less likely to be active in comparison to those who listened to music for more than one hour per day. To our knowledge, this is a novel finding and more research is needed to confirm the present association. A possible explanation may be that during physical activity young people may find music both enjoyable and motivating  and therefore, those who listen to music for more than one hour per day may simultaneously be more likely to engage in physical activity. The only significant association observed between physical activity and screen based activities was in the girls' analyses where those girls who watched television for less than two hours per day were more likely to be physically active. Previous research has produced contrasting results with some studies failing to show any associations [21, 22], other studies showing associations only with boys  and other studies showing small associations with the whole sample .
While our study assessed the association between physical activity and a number of sedentary behaviours as well as between physical activity and composite variables of screen-based and non-screen based sedentary activities, the fact that only two significant associations were observed supports previous research that physical and sedentary behaviours are two separate constructs  and that both need to be targeted in potential intervention programmes to promote physical activity. This is also enhanced by the lack of a significant association between active commuting to school and screen-based sedentary activities, a finding that supports a previous study conducted in Canada .
Another important finding of the present study is the strong association between physical activity and weekly times of sports clubs attendance whereby children who attended sports clubs for two or more times per week were at least three times more likely to meet physical activity recommendations. This finding supports previous results with Greek-Cypriot children using a four-day physical activity recall  and pedometers  as well as findings from the United States using accelerometers, where children accumulated additional 20-minutes of moderate-to-vigorous-activity while attending after school programmes . A higher percentage of boys than girls (61.6% and 44.6% respectively) reported attending sports clubs for two or more times per week. Furthermore, there was a graded decrease in the percentages of adolescents attending sports clubs for two or more times per week from primary (68.8%), middle (58.6%), technical and high schools (45.7% and 38.5% respectively). These differences in sports clubs attendance may partly explain gender and age related differences in the percentages of adolescents meeting physical activity recommendations.
While, to our knowledge, this is the first study to examine physical activity and sedentary behaviours in relation to appropriate guidelines in a large sample of Greek children and adolescents in Cyprus from different levels of education, a number of limitations are also worth addressing. First, the cross-sectional design of the present study precludes the inference of cause and effect relationships between physical activity and sedentary behaviours. Second, physical activity was assessed via self-report and future studies within the Cypriot context that examine the relationship between physical activity and sedentary behaviours should adopt objective measures of physical activity including accelerometers or pedometers. Incorporating an objective measure of physical activity behaviour, at least from a subsample, would strengthen the results of this study. Third, while a number of sedentary activities were assessed, students were asked to indicate the usual time (hours per day) that they spent on each of the above activities. Reporting sedentary activities using one-day recalls or diaries, rather than using 'the usual time' might have improved the validity of these measures. While assessing sedentary behaviours such as television viewing with single items is subject to measurement error, this approach has been used in a large number of studies and is appropriate for surveillance studies . Fourth, socioeconomic status data were not collected in this study and possible differences in SES between levels of education might have biased the results. Furthermore, assessment of physical activity and sedentary behaviours did not differentiate between weekdays and weekends.