Abstract
Children’s physical activity (PA) is increasingly restricted by the fast pace of modern life. Schools are the opportune setting in which to increase PA during daily recess. The purpose of this study was to record the degree of PA of 6th grade primary school students during recess and determine whether the playground size and available equipment can influence the degree of students’ PA. The study sample comprised of 625 12-year-old students, 324 (51.8%) boys and 301 (48.2%) girls. The PA during recess was calculated by Metabolic Equivalent Time (MET*min) using a self-reference questionnaire. The impact of schoolyard size and equipment on MET*min was evaluated by Kruskal–Wallis analysis of variance (ANOVA). The students’ PA (MET*min) during recess was higher at schools with a large schoolyard (197.55±133.15 MET*min), compared with medium size (152.65±109.23 MET*min) and small ones (117.69±104.22 MET*min), (χ2=50.629, p<0.001). Moreover, students were more physically active during recess in adequately equipped schools (187.11±124.21 MET*min), compared with partially (158.84±128.61 MET*min) or inadequately equipped ones (114.19±89.13 MET*min). This indicates a strong positive relationship between PA and equipment (χ2=21.277, p<0.001). An encouraging school environment can motivate pre-adolescent children to be more physically active and therefore healthier. These data reveal that spacious schoolyards rich in outdoor facilities can lead to more physically active behaviour among students during recess.
Introduction
Modern lifestyles and minimum free-play areas in the urban environment restrain children from physical activities (PAs) and social experiences arising from active games (Rivkin, 1997). However, play is a basic children’s need and consequently it has to be met. Many researchers claim that satisfying this need results in psychological well-being and positively affects a child’s educational and sports behaviour (Goulimaris et al., 2014). The schoolyard is an ideal setting for the promotion of children’s PA during recess (Mota et al., 2005; Sallis et al., 2000). Throughout recess, schoolchildren have the opportunity to be active and spontaneously interact with each other and their surroundings (Thomson, 2007; Verstraete et al., 2006). The significant proportion of time children spend at school makes schoolyards a critical determinant for motivating students into a more active lifestyle (Erwin et al., 2012). The ecological model maintains that the natural environment along with intra-personal, inter-personal and socio-cultural factors influences PA at behavioural level (Bebetsos and Goulimaris, 2014; Haug et al., 2010; Matsouka et al., 2003a, 2003b; Naylor and McKay, 2009; Sallis et al., 2006). Therefore, the outdoor school setting can encourage students’ PA so that daily required PA levels are met.
The fact that children spend a sustained proportion of their day at school suggests that schoolyards should be ample spaces that allow children to move freely during recess (Simmens et al., 2007). Unfortunately, increasing urbanization rates and recent education policies, such as schools merging to reduce costs, are gradually resulting in overcrowded schoolyards (Goulimaris et al., 2008). This may have a profound impact on outdoor activities, since students are crammed into few square metres and vigorous game playing is discouraged to avoid injuries (Bundy et al., 2009; Jarvis, 2007). Recent studies have shown that schoolyard size and play areas are key factors that determine schoolchildren’s PA during recess time (Escalante et al., 2012; Harten et al., 2008). Although relevant research is sparse, it indicates that schoolchildren in ample schoolyards are more physically active than their counterparts in small, crammed ones (Escalante et al., 2012; Ozdemir and Yilmaz, 2008; Zask et al., 2001). In addition, many researchers argue that children attending schools rich in play equipment and sport facilities supplemented with multicoloured ground markings, as well as rich green areas with grass or trees, have higher levels of PA during recess than those attending schools with poor facilities (Blaes et al., 2013; Dyment et al., 2009; Huberty et al., 2011; Nicaise et al., 2012; Nielsen et al., 2012; Taylor et al., 2011).
The aim of this study was to record the degree of PA of 6th grade primary school students during recess and determine whether the playground size and available equipment can influence the degree of students’ PA.
Methods
Participants
The study sample consisted of 625 students, 324 (51.8%) boys and 301 (48.2%) girls aged 12, out of a total of 844 6th grade students of the 24 primary schools in three urban areas in the Evros Prefecture, NE Greece. Descriptive data of the study sample are given in Table 1. The particular age group was selected after taking into account the following: (i) the measuring method, which was based on a self-reporting questionnaire, considered appropriate and reliable for schoolchildren aged 11+ (Sallis and Owen, 1999); (ii) the fact that 6th grade students have experienced many more years at school and may thus deliver a better judgment; (iii) this age group is on the verge of adolescence, during which documented reduction of PA begins and sedentary behaviour increases (Booth et al., 2002; Kristjansdottir and Vilhjalmsson, 2001); and finally, (iv) initially the sample size was considered sufficient for the study purposes. A larger sample may have caused problems to the participant schools’ daily routine as the questionnaire took 2 hours to complete.
Descriptive characteristics of schools.
Instruments
To assess the degree of PA, participants filled out a self-reference Child and Adolescent Trial for Cardiovascular Health, Self-Administer Physical Activity Checklist form (CATCH-SAPAC-235, Version 30/9/1993), recalling the previous day’s activities (Perry et al., 1990). The SAPAC form was translated into Greek by an English teacher and modified for Greek schools by a group of physical education teachers. To test the validity and reliability of the modified questionnaire, 33 out of 75 6th grade students at one of the 24 participant schools were asked to complete it. Both the school and the students were randomly selected. The back-to-back procedure yielded a reliability score of α=0.69. These 33 students were excluded from the actual survey. The SAPAC form, dividing the day into three parts (before, during and after school) and providing a list of 21 physical activities, asked participants to answer three questions: (i) what physical activities they engaged in the previous day; (ii) how long they actively participated in each activity, recording the actual time in minutes; and (iii) how tired they felt after a documented physical event. As regards the third question, the participants had three options to choose from: “not at all”, “a little” or “a lot”, thus giving a subjective indication of each activity’s intensity.
The evaluation of physical activities was based on the Compendium of Physical Activities: Classification of Energy Costs of Human Physical Activities by Ainsworth et al. (1993), according to which the activities were coded into Metabolic Equivalent Time (MET*min). MET is the energy metabolic equivalent of an activity and is defined as the ratio of the metabolic rate (independent of body weight) to a stable metabolic rate characterized as a unit of 4184 KJ*kg-1*h-1, corresponding to a resting metabolic state (Jetté et al., 1990). To calculate each participant’s MET*min score, the intensity of PA was taken into account with reference to the standard activity tables by Ainsworth et al. (1993), and the MET value was multiplied by the number of minutes each activity lasted. For the purpose of this research, only the “during school” PA part of the questionnaire was taken into account.
Procedure
Two preliminary investigations were conducted in order to classify schoolyards by size and available equipment. In the first study, the data concerning schoolyard sizes were obtained from the Urban Development Agency of the Evros Prefecture, a state service and reliable source of information. Its data are updated each time a change occurs. The schoolyard size in square metres (m2) was divided by the actual number of students to estimate the play area per student during recess. The resulting data were compared with the requirements set by the Hellenic Organization of School Buildings (HOSB, 1982), according to which 6-class, 9-class and 12-class primary schools should have 9.4m2, 7.8m2, 7.0m2 play area per student, respectively. Greek primary schools may have from six to 12 teaching posts depending on the number of students enrolled. There may be more than one class per grade as the student-to-teacher ratio is 27:1. Subsequently, schoolyards were classified into small, medium and large ones. As regards 6-class schools <4.7m2 play area per student was considered small, 4.7–7.8m2 medium and >7.8m2 large. Concerning 9-class schools <3.9m2 per student was considered small, 3.9–6.5m2 medium and >6.5m2 large. For 12-class schools <3.6m2 per student was considered small, 3.7–5.9m2 medium and >5.9m2 large.
The second preliminary investigation classified schoolyards according to equipment availability. The HOSB sets the requirements to be met regarding school equipment. However, the principal of each school, taking into consideration the school needs, is responsible for replacing old equipment or purchasing new with funds from the municipality. Taking into account the HOSB Guidelines (HOSB, 2008: 4-10, 15-16, 19, 37), a 15-feature list was used for the classification. The features were grouped into four categories: (i) facilities (demarcated sports facilities, synthetic track, courts with spectator seats, free-play areas and playground equipment); (ii) green areas (flower beds, trees and shrubs); (iii) recreation areas (benches, outdoor amphitheatres, pergolas and kiosks); and (iv) surface material (grass, dirt/gravel and concrete-asphalt). According to HOSB (2008), a schoolyard should comprise all four feature groups in order to be adequate, meet student needs and contribute to PA during recess. One researcher visited the 24 participating schools and recorded the schoolyard outdoor features. The data were processed and schools were classified into three categories “adequately equipped”, “partially equipped” or “inadequately equipped”. Schoolyards with less than 33.3% of the features, that is fewer than five out of the 15 features, were classified as “inadequately equipped”; schoolyards with 6–10 features fell in the “partially equipped” category and the ones with 11–15 were classified as “adequately equipped”.
A month after the preliminary investigations were completed, all 24 schools simultaneously received an email invitation to allow students to take part in the study, which they all accepted. The researchers visited each school on days convenient to the particular school so that the timetable and the scheduled activities were not disrupted. Parents had been informed in writing and asked for their consent 15 days before the research was conducted. Only students who had their parents’ written consent participated. Before completing the self-reference CATCH-SAPAC questionnaire, students were given easy-to-follow instructions with particular emphasis on how to identify day parts (before, during or after school) and certain time calculation aspects during a 10-minute introductory presentation.
The research, which had been approved by the National Institute of Education, took place between January and May 2011 with a particular focus on ambient weather conditions. The researchers always visited schools after days with no rain, with sunshine and temperatures above 10°C, so that children were more likely to have played in the schoolyard the previous day.
Data analysis
The data obtained from the two preliminary investigations were processed using Microsoft Office Excel 2007 and then the results were coded and entered into SPSS 15.0 (IBM) for statistical evaluation. In order to investigate the relationship between the students’ PA during recess and the schoolyard size or equipment, Kruskal–Wallis non-parametric analysis of variance (ANOVA) was used after carrying out the corresponding tests for normality. Between-sample comparisons were conducted using pairwise Mann–Whitney tests with Bonferroni correction. All the results were considered to be statistically significant at a score of p≤0.05.
Results
A frequency analysis has showed that out of the 24 participating schools, eight (33.3%) had a large schoolyard, five (20.8%) a medium sized one and 11 (45.8%) a small one. As regards schoolyard equipment, only five yards (20.8%) were adequately equipped, 14 (58.3%) partially and five (20.8%) inadequately equipped. The descriptive elements of schoolyards size and equipment are given in Table 2. The correlation between schoolyard size and equipment adequacy showed that only four out of 24 schools (16.7%) were both spacious and fully equipped. Spearman rank correlation between schoolyard size and available equipment yielded a value of r=0.544 (p<0.001), indicating that schoolyard size and available equipment were statistically correlated.
Classification of schoolyards according to size and available equipment.
*N=number of students in the 6th grade.
The mean MET*min of students during recess was high in schools with a large schoolyard (197.55±133.15 MET*min), lower in schools with a middle-size schoolyard (152.65±109.23 MET*min) and minimal in schools with a small schoolyard (117.69±104.22 MET*min) (Table 3). A Kruskal–Wallis test was run to determine if there were differences in MET*min score in terms of student PA during recess and schoolyard size among different schools. Pairwise comparisons were performed using Mann–Whitney tests and applying Bonferroni correction for multiple comparisons. The results showed that the MET*min score differed in terms of schoolyard size in a statistically significant manner χ2(2)=50.629, p<0.001, with a score mean rank of students’ PA during recess being: 371.53 for large, 315.06 for medium and 258.84 for small schoolyards. Post hoc analysis showed that the differences in PA were also statistically significant at all possible combinations, that is between large and medium size of schoolyards U=10.492, p=0.004, Bonferroni corrected p=0.012, p<0.05, between large and small ones U=21.741, p<0.001, Bonferroni corrected p=7.11E-12, p<0.001 and between medium and small schoolyards U=11.535, p=0.004, p<0.01, Bonferroni corrected p=0.011, p<0.05.
Physical activity (MET) of students in relation to size of schoolyard.
Moreover, the mean MET*min during recess was higher in schools that were adequately equipped (187.11±124.21 MET*min), as opposed to schools that were partially equipped (158.84±128.61 MET*min) and minimal in schools that were inadequately equipped (114.19±89.13 MET*min) (Table 4). The students’ PA was directly associated with available equipment as the Kruskal–Wallis ANOVA has shown χ2(2)=21.277, p<0.001 with a mean rank of PA score 361.40 for adequate equipment, 315.87 for partial equipment and 257.93 for inadequate equipment. This was further validated with Mann–Whitney multiple comparison tests, yielding statistically significant differences for all possible combinations; between schoolyards with adequate equipment and the ones with partial equipment the analysis showed U=20.077, p=0.022, p<0.05, Bonferroni corrected p=0.04, p<0.05, between schoolyards with adequate equipment and the ones with inadequate equipment the analysis showed U=5261, p<0.001, Bonferroni corrected p=2.23983E-06, p<0.001 and between schoolyards with partial equipment and the ones with inadequate equipment the analysis showed U=19755.5, p=0.003, p<0.01, Bonferroni corrected p=0.008, p<0.01.
Physical activity (MET) of students in relation to available equipment of schoolyards.
Discussion
Schoolyards constitute a place where considerable school time is spent. Hence, they play a critical role in promoting student PA during recess. Many researchers have investigated the factors that affect student PA during recess and implemented interventions to enhance PA levels without reaching definite conclusions, because they failed to take into account certain PA determinants such as the schoolyard features like area type and size (Escalante et al., 2012). The aim of this study was to record the degree of PA of 6th grade primary school students during recess and determine whether the playground size and available equipment can influence the degree of students’ PA.
As regards schoolyard size (play area m2/schoolchild), the study results have shown that it has a positive impact on student PA during recess, as students in ample schoolyards are more physically active than their counterparts in small, crammed ones. These findings support previous relevant research (Escalante et al., 2012; Ozdemir and Yilmaz, 2008; Zask et al., 2001), and show that primary school students need ample spaces to engage in PA during recess. However, Engelen et al. (2013) did not observe any influence of schoolyard size on children’s PA. The Engelen et al. (2013) study findings come from a specific intervention with loose, primarily recycled materials and refer to students younger than the ones in this particular study. The sample of this research comprised of 12-year-olds who mostly engage in team or structured games (Boulton, 1992; Strong et al., 2005). Therefore there should be separate play areas where they can play undisturbed by younger schoolchildren who engage in more vigorous (Lopes et al., 2006), creative (Thomson, 2007), anaerobic and fantasy games (Strong et al., 2005). Since the nature of PA differs from one age group to another it is recommended that separate play areas are used to accommodate the needs of different age groups. Consequently the schoolyard size is among the deciding factors determining PA during recess. Provided the schoolyard size is adequate for the number of students enrolled, appropriate spatial distribution and arrangement can accommodate all age groups’ needs.
This study also investigated permanent play equipment such as sports facilities, recreation areas and surface materials, as well as plant life such as flower beddings, trees and shrubs to determine whether the existing environment stimulated and encouraged PA during recess. The findings indicated that students attending schools with adequately or partially equipped yards were more physically active than their counterparts in inadequately equipped schoolyards. The study supports previous research findings (Escalante et al., 2014b; Farley et al., 2008; Sallis et al., 2003; Stratton and Mullan, 2005; Taylor et al., 2011; Willenberg et al., 2010) and indicates that, more often than not, children will make use of whatever is available so that their need for play is met (Ozdemir and Yilmaz, 2008; Tsitskari et al., 2014; Weinstein and Pinciotti, 1988). However, it is doubtful if this conclusion can be generalized. Zask et al. (2001) found no significant relation between the available schoolyard equipment and the students’ PA during recess time, with the exception of balls. Engelen et al. (2013) support that an increase in PA due to intervention was not maintained for a long time. There is no conclusive evidence of a link between the type of equipment and the duration of its impact on PA. Escalante et al. (2014b) conducted an interesting systematic review in which relevant studies were classified according to the equipment interventions into the following categories “playground markings, game equipment, playground markings plus physical structures, and playground markings plus game equipment”. As regards school-age children, the findings have demonstrated that portable equipment provision by itself does not increase students’ Moderate-to-Vigorous Physical Activity (MVPA) and Vigorous Physical Activity (VPA) during recess time. However, a combination of multicoloured markings and physical structures seems to result in an increase in MVPA and VPA, but only in the short and medium term. Unfortunately, the impact of the interventions gradually declines and so does PA. The fact that the results are not conclusive may be due to the type of equipment each study investigates (Escalante et al., 2014b) or the different types of games (Escalante et al., 2012; Strong et al., 2005). This may also be attributed to age and gender (Escalante et al., 2011), physical and social factors related to family (Escalante et al., 2014a; Maitland et al., 2013) and culture, and each school’s play habits and policy. Taylor et al. (2011) did not observe any impact of school policies on students’ PA during recess time. However, further research would be necessary since some researchers maintain that teachers restrict or discourage vigorous play, fearing possible injury during recess (Bundy et al., 2009; Jarvis, 2007).
This study, which favours an ecological approach to promoting PA, has certain limitations. One is that the assessment of PA was carried out using a self-reference questionnaire. Although previous research has shown only a medium relation between self-reference and PA objective assessment in children (Sallis, 1991; Sirard and Pate, 2001), the fact that the students had to recall only the previous day’s activities increased the accuracy of the recollection. It is also possible that the children may have failed to precisely estimate the duration of each activity. Nevertheless, the recess time is set and the sample size is large enough to limit errors. A third limitation may be that the study did not examine individual school policies during recess, which seem to either encourage or discourage PA. The researchers assumed that 6th grade students had access to all equipment available in the schoolyard.
In conclusion, 12-year-old students in large schoolyards were more physically active than their counterparts in small ones. The study also indicated that adequately or partially equipped schoolyards stimulated higher PA in the same age group than inadequately equipped ones. The fact that research findings concerning the impact of schoolyard equipment on students’ PA during recess are sparse, coupled with the Escalante et al. (2014b) systematic review results, suggests that the focus should be the environmental factors affecting PA and therefore schoolyard environment enhancement. It would be also wise that students themselves suggested interventions and improvements as to what environment and equipment combinations (Naylor and McKay, 2009; Thomson, 2007) would encourage them to be physically active in each recess period during their long school day.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.