The school playground environment as a driver of primary school children’s physical activity behaviour: A direct observation case study

Abstract

The school playground can promote PA for large numbers of children. This study identifies areas of the playground that children visited at break-times, the decisions according to gender and the influence of contextual and environmental variables on PA levels. The playground of a culturally diverse primary school was observed during morning break-times and lunchtimes. Counts of sedentary, LPA, and MVPA episodes, and the contexts in which they occurred were recorded using the system for observing play and leisure in youth (SOPLAY). Ball sports areas had higher counts of boys (mean ± SD; 9.9 ± 4.8) compared to girls (2.0 ± 3.5); areas promoting climbing and social interaction had higher counts of girls (7.9 ± 7.2) compared to boys (3.5 ± 2.9). The proportion of MVPA episodes during break-times was 34% ± 26%. Areas of the playground with organised activities had 2.70 (95%CI: 1.87 to 3.91) times higher MVPA counts than areas “not organised”. Areas with “supervision” were associated with higher MVPA counts (1.34; 1.18 to 1.53) compared with “not supervised” areas. Organisation and supervision might influence PA choices and PA levels of children in the primary school playground. Further investigation is required to explore different playgrounds settings, and context and gender preferences.

Full text

Title – The School playground environment as a driver of primary school children’s
physical activity behaviour: A direct observation case study
Authors
Michael Graham1* (Corresponding author) Michael.graham@tees.ac.uk
Matthew Wright1 M.Wright@tees.ac.uk
Liane B. Azevedo2 L.Azevedo@hud.ac.uk
Tom Macpherson3 T.Macpherson@tees.ac.uk
Dan Jones1 D.Jones@tees.ac.uk
Alison Innerd1 A.Innerd@tees.ac.uk
Affiliations
1 = School of Health and Life Sciences, Teesside University, Middlesbrough, United
Kingdom
2 = School of Human and Health Sciences, University of Huddersfield, Huddersfield,
United Kingdom
3 = Division of Sport and Exercise, School of Health and Life Sciences, University of
the West of Scotland, United Kingdom
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

Abstract
Background: The school playground can promote physical activity for large numbers of children.
The aim of this study was to identify areas of the playground that children visited at break-times,
the decisions according to gender and the influence of contextual and environmental variables on
PA levels. Methods: The playground of a culturally diverse primary school was observed during
morning break times and lunchtimes. Counts of sedentary, LPA, and MVPA episodes, and the
contexts in which they occurred were recorded using the system for observing play and leisure in
youth (SOPLAY). Results: Areas promoting ball sports had higher counts of boys (mean ± SD; 9.9
± 4.8) compared to girls (2.0 ± 3.5) and areas promoting climbing and social interaction had higher
counts of girls (7.9 ± 7.2) compared to boys (3.5 ± 2.9). The proportion of MVPA episodes during all
break-time was 34% ± 26%. Areas of the playground with organised activities had 2.70 (95%CI:
1.87 to 3.91) times higher MVPA counts than areas ‘not organised’. Similarly, areas with
‘supervision’ were associated with higher MVPA counts (1.34; 1.18 to 1.53) compared ‘not
supervised’ areas. Conclusion: Organisation and supervision might influence PA choices and PA
levels of children in the primary school playground. Further investigation is required to explore
different playgrounds settings, and context and gender preferences.
Keywords: play, break-time, recess, observation, MVPA
1. Background
Current recommendations suggest that children and young people should engage in moderate to
vigorous intensity physical activity (MVPA) for an average of 60 minutes a day over the course of
the week.1 Self-reported data shows that despite an increase of 3.6% on the previous year, more
than half of all 5 to 16 year olds (53.2%) are still failing to reach the recommended level of physical
activity and nearly a third of all children (29%) are achieving less than an average of 30 minutes a
day of MVPA in the UK.2 This is supported by accelerometry measured physical activity data which
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42

highlighted that 49% of boys and 76% of girls, aged 4 to 10 years old failed to meet the earlier
MVPA recommendations in England of at least 60 minutes per day.3 Furthermore, according to
recent analysis of pooled physical activity data (accelerometry) from 47,497 children and
adolescents from across Europe, only 29% (95%CI: 25% to 33%) of children (2 to 9.9 years of
age) were sufficiently active according to the most recent physical activity guidelines (average ≥ 60
minutes of MVPA per day across the week).4
In order to increase physical activity levels of children, a number of school, community and home-
based initiatives have been designed and delivered in the UK and across the world with varied
successes.5,6 School-based interventions possess an additional advantage over other settings as
children spend a large proportion of their waking hours within the school environment.6 Physical
education (PE) lessons7 and break-times5 provide the majority of opportunities for physical activity
during school hours.8 However, there is some evidence that structured PE lessons account for only
a small proportion of a child’s recommended daily MVPA6 and have been found to elicit insufficient
levels of MVPA.8,9 Although it is important to note that the target of primary PE is not primarily to
increase children’s activity levels, but should develop children’s movement skills, which might lead
to improvement in movement skill competence and increase engagement in physical activity during
other periods of time and long term.6,10
Despite the importance of PE for providing a setting to develop movement skill competence and to
contribute to children’s total physical activity, opportunities to engage in quality PE at primary
school are limited11 with no statutory guidance on the amount of the curriculum timetable UK
primary schools should dedicate to PE. Furthermore, primary school PE is often delivered by
teachers without specialist PE teacher training12 and lack of confidence in delivering physical
activity sessions.13
Therefore, there is a need for children to be provided with other opportunities for physical activity
during school hours,14 including classroom15 and break-times (also commonly referred to as recess,
internationally).16 Understanding the nature and complexity of physical activity during break and
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68

lunchtime could be important in the development of school-based interventions and playground
design. For example, observing the complex interactions that exist in the primary school
playground would offer valuable insight for schools and researchers when considering the
implications of child interactions with the playground environment to promote physical activity
engagement. This would support the development of time efficient and sustainable playground
activities that encourage a higher level of movement competence and a greater amount of MVPA
during break and lunchtimes.
School break-times (morning break and lunchtime) provide an opportunity for primary school
children to engage in freely chosen play and physical activity. Break-times are an ideal context
within the school day to promote physical activity as they are offered universally in UK primary
schools17 and targeting these periods in the school day would not interfere with existing academic-
focussed activities.18 However, despite the evidence supporting the benefits of regular breaks on
physical, social/emotional, psychological and cognitive outcomes,19 there has been a marked
decline in the amount of break-times offered to children at all school ages in the last 20 years. 17
Children in key stage one (KS1) (5 to 7 years old; School years 1 and 2) and key stage two (KS2)
(7 to 11 years old; School years 3 to 6) now receive an average of 45 minutes less break-time per
week.17 Furthermore, the proportion of time provided for breaks during school time reduces for
each stage of school (KS1 to KS4),17 predominantly as result of curricular related pressures.17,20
However, evidence suggests this is counterintuitive, with physical activity associated with
increased break-time duration,19 and increases in physical activity, within lesson time and whole
day, having beneficial effects of academic outcomes.20
The reduction in time available for break-times, paired with a large variation in playground
environments (equipment provision, staffing, physical structures and surfaces) available to children
during break-time can result in varied levels of physical activity in primary school children.5,16,21
The understanding in how children from different age groups, for example interact with their school
playground to provide safe, fun, engaging and developmentally appropriate opportunities for
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94

physical activity throughout the school years, to our knowledge has not been previously reported.
Previous narrative reviews found mixed results of intervention strategies (e.g. playground
markings, games/sports equipment, active video games, fixed/non-fixed equipment) aimed at
promoting playground engagement and improving physical activity.16,22 A recent systematic review
and meta-analysis highlighted the promising effect of interventions focussed on school break-
times, on children’s physical activity levels, but re-iterated that the small number of studies
focussing on the same components restricts any definitive conclusions or future
recommendations.5
Accelerometer and GPS determined physical activity ‘hot’ and ‘cold’ spots have previously been
identified in pre-school (5 to 6 years old), middle school (8 to 11 year old) and senior school (12+
years old) children in America,23 Denmark24 and the Netherlands,25 with contrasts in the
environments which promote higher physical activity engagement. For, example, high bars and
climbing equipment in the Denmark study24 were used as a ‘hang-out area’ promoting largely
sedentary behaviours, whilst ‘high bars’ in the Netherlands study25 promoted higher levels of
MVPA. Furthermore, Clevenger et al.23 highlighted that hot and cold spots in their study were time
dependent, changing over the course of the day. Therefore, it is likely that there are differences
between these environments that result in these observed differences in physical activity behaviour
(demographics, geography, environmental, accessible and usable spaces).
Further, evidence on which areas of the playground (sports pitches, climbing frames, creative play
spaces) are ‘most successful’ in engaging children in MVPA in UK primary school children is
needed. Furthermore, modifiable contextual characteristics of the playground (such as playground
supervision and access to larger playground areas) have previously been associated with physical
activity levels, having both negative19,26 and positive effects.21,27 The negative association between
playground supervision and access to some play spaces and physical activity levels may be due to
a priority of school and break-time supervisors being to keep children safe, which has been
suggested to suppress children’s physical activity levels during break-times.26
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120

Likewise, it has recently been suggested that many health promotions and primary school
interventions fail to consider the influence of class cultures and previous dispositions towards
physical activity on interventions success.28 For example, children may actively choose not to
engage in an activity, play with a particular piece of equipment, or play in an area of the playground
which they perceive to be ‘inappropriate’. Further, the school playground is a visible arena where
gender identities are formed, destroyed and contested29 with hegemonic masculine activities (such
as football) dominating much of the playground space.30 Traditional playground hierarchies such as
these may act to promote and prevent physical activity participation, dependant on the child’s
disposition, gender and motivations.
Therefore, the aim of this study was to identify the areas of the playground that elicit higher
proportions of MVPA episodes during morning break-times and lunchtimes, the choices according
to gender, and to explore the contextual and environmental characteristics present in these areas
(i.e. area size, provision of equipment, organisation and supervision). These findings might help to
inform playground design and influence the development of future playground interventions.
2 Methods
2.1 Participants
The participating school approached the University interested in evaluating their playground
equipment. Consequently, the lead author took the opportunity to design a research study to
understand the effect of playground on physical activity behavior. The primary school for this case
study had 528 children with an age range from 5 to 11 year old (50.8% male). The school was
located in a neighbourhood that was in the lowest 10% on the Index of Multiple Deprivation. The
school had 44% of children eligible for free school meals and 70% of children having English as an
additional language. The study received ethics approval from the School of Social Sciences,
Humanities and Law, research ethics committee at Teesside University (Application
number: SSSBLREC055). Study and participant information sheets were provided to Head
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145

teachers, parents and children and there was an opportunity to ask questions of the research
team. Members of staff (including teachers, administration and support staff) at the school were
made aware of the project and the reason for the presence of the research staff. Following Head
Teacher and parental informed consent (‘opt out’) and child informed assent, primary school
children were observed in the playground environment during morning break and lunchtime on
three separate occasions over an eight-week period during July and September 2017 (separated
by the summer holidays). There was no facility for ‘indoor play’ within the case study school.
Therefore, data collection days were completed when the weather forecast projected dry days. The
temperature on data collection days was 18, 20 and 17 degrees Celsius on day 1, day 2 and day 3,
respectively.
The collective term break-times is used in this study when referring to multiple periods of
break throughout the day (i.e. morning break and lunchtime combined). The term ‘recess’ is
commonly used in other countries to describe this period in the day where children are given ‘free
play’ opportunities. Where reference to a specific break period is needed the associated term will
be used (i.e. morning break).
2.2 Observations
Children participation
Children were informed about general details of the project and the presence of the observers prior
to any data collection. Research staff visited the school on three occasions prior to recording the
school playground and were present from the start of the school day until the end of the lunch
break-time. These initial visits served the joint purpose of research staff becoming familiar with the
playground and the identified target areas, and for children to become familiar with the research
staff and cameras. Cameras were not active during these initial visits and data collection began on
the fourth visit to the school.
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169

Children were asked to take part in the usual playground activities and ignore the presence of the
research staff in order to reduce the reactivity of the children. In situations where children
approached research staff, the staff were asked to inform the children ‘we are busy at the minute
but I can talk to you later’ as recommended by Darst, Zakrajsek and Mancini31 to further reduce
reactive behaviour. To reduce the reactivity to the video camera, any recordings taken in the period
before school started (i.e., before 9am, when children who arrived early used the time to access
the school playground) were considered ‘habituation’ and not used in the scoring of playground
activity levels.31 All children had access to the play space that was observed by researchers at
some point during the school day.
In any instance where the camera affected the behaviour of the children (e.g., children “acting up”
for the camera), the decision was made to delete this observation from the recordings. ‘Acting up’
was determined as behaviour that was perceived by the trained observers to be ‘performed’ for the
camera/observer. These instances of reactivity were minimal and included children gathering
around or repeatedly walking in front of the camera at short range, therefore blocking the observed
target area and the children’s behaviour occurring behind them. These instances were viewed by
all three observers and agreements reached on whether to include or discard. This resulted in the
removal of two clips (2 x 30 seconds) in total. In the scenario where a portion of the area was
blocked, we attempted to score as much of the clip as possible. Finally, a designated ‘no
observation zone’ was made available for any child that did not wish to take part in the study
(Figure 1).
SOPLAY method
The System for Observing Play and Leisure Activity in Youth (SOPLAY)32 was utilised for observing
the children’s physical activity levels. The SOPLAY is based on momentary time sampling
techniques in which systematic and periodic scans of individuals and contextual factors within pre-
determined target areas are made.33
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194

Six observers (MG. AI, MW, LA, DJ, TM) were trained in the mapping protocols for SOPLAY and
were familiarised to the school playground and the target areas (Table 1). Observers used practice
and gold standard assessment videos (using available resources from
www.activelivingresearch.org) and recorded observer agreement and relative reliability.
Researchers (MG and AI) engaged with the school staff (teachers and playground supervisors) to
ensure the target areas did not cross any boundaries or restrictions enforced by the school. Twelve
target areas were originally identified (Table 1) and the boundaries made clear to each observer
(Figure 1). The playground was split according to KS1 (5 to 7 years; School year 1 and 2) and KS2
(7 to 11 years; School year 3 to 6) in line with the restrictions put in place by the school to separate
the KS1 and KS2 children during break-times. This is common in UK primary schools, to separate
the range in behaviours and the physical capabilities between the youngest and oldest children.
Figure 1. Playground map and boundaries
No observation
zone
195
196
197
198
199
200
201
202
203
204
205
206
207
208

Table 1. Defining characteristics of playground areas
Target area Key area characteristics
1 (KS2) Large play/climbing frame - climbing Ropes, climbing and scramble nets, balance beams, tyre
swings, climbing wall built on a cushioned surface
2 (KS2) Large open play area – Tarmac surface, playground markings, basket for netball/basketball,
walled area (used predominantly with tennis and football)
3 (KS2) Stage area – a small stage constructed from wood (3mx2m), viewing benches/beams
(sometimes used to climb/balance), surrounded by paved surface
4 (KS2) Multi-sport court – Tarmac surface ball court with markings, goals available on occasion (x2)
and balls provided on request. This court is timetabled to make it available to all year groups
5 (KS2)
(excluded)
Seating area with small climbing wall – climbing wall on the school building next to a bench
provided for children wanting to speak with staff. This area is used more as access to other
areas of the playground
6 (KS2) Tyre climbing area – Multiple rubber truck/tracker tyres fixed to the ground in a different
positions surrounded by grass surface – designed to encourage balance
7 (KS2) Astro turf –5 a-side court providing all weather court for children to play football. Fenced off
area with permanent goals and markings
8 (shared) Seating area – A small area between more active areas with toad stools and a wooden throne
built on a cushioned surface
9 (KS1) Free play – Tarmac surface with markings, wooden tepee’s, wooden tunnel, sand box, wooden
bench and a sandpit – this area is provided with scooters, tricycles, toy prams etc.
10 (KS1) Small play/climbing frame – similar to area one but built for younger year groups
11 (KS1) Sheltered seating area – a hexagonal seating area with 6 benches covered with an aluminium
roof (in the shape of a flower).
12 (KS1) Large open play area – mix of tarmac and grass surface, playground markings, tunnel under a
grass mound, wooden posts to encourage balancing, small kitchen area (for role play) and
wooden benches.
Three pairs of observers used three separate cameras to record the target areas during each
break and lunchtime. Camera locations were decided during the familiarisation visits. Locations
were chosen for their superior vantage point, clear of obstructions and on the perimeter of the
playground area, so not to restrict freedom of movement for the children. Camera operators took
209
210
211
212
213
214
215
216

video recordings of each target area in sequence (target area 1 through 12) for 30 seconds at a
time throughout morning break (15 minutes) and lunchtimes (45 minutes) on each day of data
collection. Each camera started recording at the same time but at a different scan area (camera
one started at area 1, camera two started at area 5, and camera three at area 10) and worked
sequentially, ending once break-time was over. This resulted in 611 video clips totalling 306
minutes of recordings between the three cameras, resulting in an average of 34 minutes per
camera, per day (over three days). Time was lost moving between playground areas and at the
end of break-times when children were called to line up prior to going back into class. Furthermore,
the lunchtime period was inclusive of children eating their lunch so time on the playground was
less than the 45 minutes of break-time scheduled. The same number of recordings were taken for
each playground area. Some areas had recordings when there were no children present in the
area. In situations where there were no children recorded in a playground area, the clips were
scored in the same way, and a zero recorded for episodes of each activity category.
Three trained observers (MG, MW, AI) then scored the clips retrospectively, and independently.
The use of video recordings has a high degree of agreement with live assessments34 and resulted
in a larger number of observations for each target area than would have been possible by using
live observations. Target area 5 was subsequently excluded as there were no children observed in
this area during any of the video recordings.
When scoring video clips, observers were asked to score the number of children (‘episodes’)
observed in sedentary (SED), light physical activity (LPA) and moderate to vigorous physical
activity (MVPA). Definitions for SED (i.e., lying down, sitting or stationary standing; <1.5 MET’s),
LPA (i.e., walking or activity resulting in similar energy expenditure; >1.5 MET’s and <3 METs) and
MVPA (i.e., jogging, running, gymnastic/strength exercises or activity resulting in similar energy
expenditure; >3 MET’s)35 align with the original SOPLAY categories (‘sedentary’, ‘walking’ and ‘very
active’) and were agreed prior to observations in order to minimise errors.
Contextual factors
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242

The five contextual factors in the SOPLAY were scored as counts (number of times the contextual
variables were present) during each observation and summed for each target area. A score of ‘1’
was given to any instance that a contextual factor was present and ‘0’ in any instance it was
absent. These scores were used to calculate the number of times the contextual variables
‘accessible’, ‘usable’, ‘supervised’, ‘organised’ and/or ‘equipped’ were present in the playground.
‘Accessible’ and ‘usable’ were scored if children had access to the area in the observation and that
area was in a usable condition (i.e., not broken, damaged or flooded etc.). ‘Supervised’ was scored
whenever there was an adult in the observation area whose role was primarily behaviour
management and child safety. ‘Organised’ was scored if there was adult supervision in a role
beyond that of the break-time supervisors which involved instruction and facilitation of an
organised activity. Finally, ‘equipped’ was scored if there was movable equipment provided beyond
what would normally be expected in that area. This equipment included footballs, tennis balls,
skipping ropes, hula hoops, scooters, tricycles, prams (for dolls) and bean bags. Areas with fixed
playground equipment (climbing frames, goals, sandpits) were not scored as ‘equipped’ as these
structures were used in the mapping of the playground zones and would have misleadingly
amplified the effect of the contextual variable ‘equipped’ on physical activity levels. All the
equipment used by the children was provided by the school and was available during every break
period (i.e. morning break and lunchtime). Children were not allowed to bring equipment from
home.
Process of observations
Observations began by first scanning the area and recording the number of girls who were
observed as either sedentary, light or moderate to vigorously active. The clip was then re-started
and the scan repeated immediately for boys and then repeated again to score the contextual
variables. Observations started at the left most boundary and were completed left to right at a rate
of one child per second.32 Each child was observed once during each scan (even if they moved
back into view). Backtracking to count new children entering the scan area was discouraged. This
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268

was made easier by adding a small observation window to the video clips during video editing
which moved across the playground area on the video clip at the required speed and helped
maintain the correct scan tempo.
Any child within a specified target area during a scan was identified as actively participating and
scored accordingly. The video clips for each target area were watched in full and a score noted for
females followed by males for each clip before moving to the next clip. If children’s activity levels
were unclear on first observation, then recordings were watched back to resolve any uncertainties
in activity codes. Once all clips for the target area were completed the observers moved onto the
next set of clips for a different target area. The order that each observer viewed the clips were
counterbalanced so that the order of observations did not have an effect on the scoring. There
were unavoidable instances where children were ‘unobservable’ due to moving behind a structure
or hiding inside structures (e.g., tunnels). However, in line with the momentary sampling method
these children were no longer in the observable space and therefore not counted in the
observations.
2.3 Observer reliability
Inter-observer agreement (IOA) was calculated to establish the agreements between observers
and the gold standard score. An IOA equal to or above 0.8 (or 80%) has previously been used as
an acceptable threshold for acceptable observer agreement.26,32 Furthermore, the Inter-class
correlation coefficient (ICC) for agreement (ICC(A, 1)) was calculated in accordance with McGraw
and Wong36 to establish criterion-referenced reliability.
Inter-rater reliability (IRR) was calculated using Cohen’s kappa (k) with qualitative inference based
on the following; 0.81-1.00, almost perfect; 0.61-0.80, substantial; 0.41-0.60, moderate; 0.21-0.40,
fair; 0.00-0.20, slight; <0.00, poor.37 Practice videos were scored independently then discussed as
a team. Once observers met the criteria for acceptable reliability (IOA>80% and ICC>0.75) for the
training and gold standard assessment videos they moved to the next stage of testing (see
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293

supplementary file for the stages of training and calibration procedures for observers). As a result
of the reliability and agreement scores for the practice videos, one observer was removed from
further assessment due to been unable to reach the acceptable level of IRR and IOA. A further two
out of the six observers were placed on a backup list and asked to continue with training videos
once per month (until the end of live observations) in case they were needed in future
observations.
Table 2. Reliability of observers against a gold standard criterion score for assessment videos.
Gold
assessment
Sedentary LPA MVPA
Observer
ID
ICC; (95%CI) IOA %
IRR
k
ICC; (95%CI) IOA %
IRR
k
ICC; (95%CI)
IOA
%
IRR
k
MG 0.97; (0.93-0.98) 89 0.77 0.99; (0.97-1.00) 96 0.92 1.00; (1.00) 100 1.00
AI 0.95; (0.87-0.97) 86 0.69 0.93; (0.84-0.96) 79 0.60 0.95; (0.87-0.97) 82 0.75
MW 0.96; (0.91-0.98) 86 0.70 0.96; (0.92-0.98) 79 0.51 0.98; (0.96-0.99) 93 0.83
Abbreviations: CI = Confidence Intervals; ICC = Inter-class correlation coefficient; IOA = Inter observer agreement; IRR
= Inter-rater reliability; LPA = light physical activity; k = Cohen’s Kappa; MVPA = moderate to vigorous physical activity
The ICC for the practice videos were classed as ‘high’ to ‘extremely high’ (0.89-1.00) based on the
following thresholds: >0.99, extremely high; 0.99–0.90, very high; 0.75–0.90, high; 0.50–0.75,
moderate; 0.20–0.50, low; <0.20, very low.38 The ICC, IRR and IOA for the assessment videos are
presented in Table 2.
2.4 Data Analysis
2.4.1 Activity levels
Video clips from the three days were pooled and grouped by target area. The percentage of SED,
LPA and MVPA was calculated for each clip by dividing the number of episodes for each activity
code by the total number of observations. Proportions (mean ± standard deviation (SD)) of SED,
LPA and MVPA episodes were then calculated for each target area and for the playground as a
whole. Target area proportions for each activity were calculated for male and female children
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312

separately; and males and females combined, in order to establish hotspots for MVPA. Heat
mapping software (http://heatmapper.ca; GenomeAlberta and Wishart research group, Canadian
Institute of Health Research (CIHR), Alberta, Canada) was used to develop a heat map of
individual target area counts for MVPA. A distribution map was generated using each of the eleven
target areas by using the number of episodes of MVPA observed in each area. Markers with a
larger Gaussian radius and a larger contrast between blue (‘cold’) and red (‘hot’) represent a target
area as a ‘hotspot’ for MVPA.
2.4.3 Statistical analysis
Raw data was processed in Microsoft Excel (Microsoft Office 2016) spreadsheets to produce
counts (number of MVPA episodes) and the proportions of SED, LPA and MVPA episodes (males,
females and combined) for each target area. Data are presented descriptively as mean ± SD.
There was a high probability for the same children to be observed in numerous clips and in a
number of different target areas. Therefore, inferential analysis was not possible, as counts were
not independent for each child and these therefore could lead to over/under estimations of activity
levels. However, the purpose of this study was to determine the activity and context of each
playground area and from the playground as a whole during break-times, and not to track the
activity levels of individual children. Negative binomial regression was used to determine the extent
to which the presence or absence of each contextual factor relates to the MVPA counts and total
activity counts (TA = LPA + MVPA) during break-times. The variables “Accessible” and “Usable”
were present in 100% of the observations so were excluded from the analysis. Data are presented
as incidence rate ratios (RR) with 95% confidence intervals (CI). The RR represents the ratio of
counts of MVPA episodes for the presence versus absence of the contextual variable.
3 Results
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336

To explore a gender preference for area, irrespective of activity level we calculated the average
number of boys and girls present in each target area during the 30 second video clips. Table 3
presents the mean difference per area and for combined areas for team sports and social
interaction according to gender. The number of boys were higher in areas that promoted team
sports, while number of girls were higher in areas that promoted social interaction.
3.1 Activity levels
The proportion of MVPA episodes for the playground as a whole during break-times (boys and girls
combined) was 34% ± 26% (Table 4). The MVPA total counts (male and female combined) for each
target area can be seen in Figure 2. Target area 1 and 12 can be considered MVPA hotspots due
to a higher count of MVPA episodes in these areas. Areas containing climbing and play apparatus
(area 1, 10 & 12) and multi-use courts/pitches (area 4 & 7) had a higher count of MVPA episodes,
compared to areas with creative play equipment (area 3, 6 & 8).
337
338
339
340
341
342
343
344
345
346
347
348

When the total number of episodes (sum of SED, LPA and MVPA for boys and girls combined) for
each target area was the denominator, target areas 4 and 7 (KS2), which were the sports areas,
had a higher proportion of MVPA episodes for boys (area 4: 30% ± 23% and area 7: 35% ± 26%)
compared to girls (area 4: 5% ± 10% and area 7: 4% ± 8%). However, target areas 1 and 6 (KS2),
which were used as climbing and social areas, had a higher proportion of MVPA episodes for girls
(area1: 36% ± 18% and area6: 31% ± 27%) compared to boys (area1: 20% ± 18% and area 6:
16% ± 19%) (Table 4).
Figure 2. Moderate to vigorous physical activity (MVPA) heat map for individual target area
MVPA episodes in KS1 target areas (9 to 12) had a consistently higher contribution from boys
observed in MVPA, compared to girls (Table 4).Table 5 displays the proportions of SED, LPA and
MVPA episodes for boys and girls when using target area totals (sum of SED, LPA and MVPA) for
boys and target area totals for girls separately as the denominator. When looking at the
playground as a whole, 38% ± 30% (mean ± SD) of the total number of episodes recorded for boys
were MVPA and for girls it was 31% ± 32%.
349
350
351
352
353
354
355
356
357
358
359
360
361
362

Table 5, highlights that in most cases there are similarities in the areas that promote a higher
amount of MVPA for the KS2 areas for both boys and girls (target area 1 to 7). However, for KS1
there were differences according to gender with boys presenting a higher proportion of MVPA
episodes in areas 9, 10, 11 and 12 compared to girls (Boys: 9= 31 ± 29%; 10= 50 ± 25%, 11= 14 ±
26% and 12= 41 ± 26% vs. Girls: 9 = 21 ± 27%, 10 = 37 ± 25%, 11 = 7 ± 13% and 12 = 29 ± 28%).
363
364
365
366
367

Table 3. Mean number of boys and girls that were present in each target area (data from the 3 days combined)
Mean ± SD 95% CI
Target
Area Males Females Difference lower upper Target area favours males Target area favours females
1 4.2 ± 2.6 11.7 ± 8.8 7.53 5.22 9.84
-15 -10 -5 0 5 10 15
Mean difference in female and male children that were present in each target area
2 6.9 ± 5.3 4.5 ± 6.3 -2.45 -5.45 0.55
3 2.1 ± 1.9 2.8 ± 2.4 0.80 -1.20 2.80
4 10.7 ± 4.9 2.9 ± 4.8 -7.73 -9.13 -6.33
6 3.2 ± 3.5 5.1 ± 3.9 1.96 0.40 3.53
7 9.9 ± 4.7 1.4 ± 1.5 -8.50 -9.91 -7.09
8 4.6 ± 3.9 6.2 ± 3.0 1.56 -0.54 3.66
9 6.9 ± 5.2 5.7 ± 4.7 -1.27 -2.36 -0.18
10 7.6 ± 3.2 5.5 ± 3.1 -2.17 -3.43 -0.91
11 3.4 ± 2.4 5.3 ± 2.9 1.93 0.30 3.26
12 9.5 ± 6.6 7.1 ± 4.9 -2.34 -3.56 -1.11
4 & 7 9.9 ± 4.8 2.0 ± 3.5
Areas that promoted team
sport
1, 6 & 11 3.5 ± 2.9 7.9 ± 7.2 Areas that promoted
social interaction
Key: CI = Confidence interval; SD = Standard deviation
368

3.2 Contextual factors
Negative binomial regression was applied for MVPA counts and TA counts based on
the contextual variables supervised, organised, and equipped. In areas where the
contextual variable ‘organised’ was present, the count of episodes of physical activity
was 2.70 (95%CI: 1.87 to 3.91) and 1.79 (1.23 to 2.60) times that observed in areas
‘not organised’, for MVPA and TA, respectively. The contextual variable ‘supervised’
was associated with 1.34 (1.18 to 1.53) and 1.40 (1.24 to 1.58) times the counts for
MVPA and TA, respectively, compared to areas that were not ‘supervised’. For areas
where the equipment was provided (‘equipped’), without the contextual variables
‘organised’ or ‘supervised’, the number of episodes was 0.85 (0.75 to 0.96) and 0.99
(0.89 to 1.12) times the counts for MVPA and TA, respectively, compared to areas
without provided equipment. The proportion of observations, separated for each target
area, that were scored for the presence of each contextual variable (‘Organised’,
‘Supervised’, ‘Equipped’) are presented in Table 6.
369
370
371
372
373
374
375
376
377
378
379
380
381
382

Table 4. Target area activity levels – Mean and standard deviation of the proportion (%) of observed episodes for each activity threshold contributing to the
total number of episodes (boys and girls combined for target area totals) (Mean (SD))
Mean (SD)
Key stage 2 Shared Key stage 1
Target area 1 2 3 4 6 7 8 9 10 11 12 Playground
Girls SED 12 (15) 4 (9) 22 (29) 3 (7) 12 (13) 1 (3) 30 (22) 12 (15) 7 (10) 41 (25) 17 (21) 13 (19)
Girls LPA 20 (18) 18 (25) 19 (26) 7 (15) 19 (19) 5 (8) 16 (15) 23 (21) 20 (17) 20 (22) 15 (12) 17 (19)
Girls MVPA 36 (18) 15 (26) 12 (23) 5 (10) 31 (27) 4 (8) 14 (20) 11 (17) 15 (12) 6 (11) 14 (18) 14 (19)
Boys SED 6 (11) 12 (23) 19 (29) 18 (24) 6 (10) 8 (12) 20 (19) 13 (15) 8 (10) 21 (23) 15 (15) 13 (18)
Boys LPA 6 (9) 37 (32) 15 (18) 38 (25) 15 (22) 46 (21) 10 (13) 25 (21) 21 (16) 7 (11) 17 (14) 22 (22)
Boys MVPA 20 (18) 14 (18) 13 (26) 30 (23) 16 (19) 35 (26) 10 (12) 16 (18) 29 (19) 5 (11) 21 (17) 21 (20)
% MVPA episodes
during break-times 56 (23) 29 (28) 24 (30) 35 (27) 47 (30) 39 (26) 24 (23) 27 (23) 45 (21) 11 (14) 35 (22) 34 (26)
% LPA episodes
during break-times 26 (20) 55 (30) 24 (28) 44 (25) 35 (26) 51 (23) 26 (19) 48 (25) 41 (22) 27 (23) 33 (18) 39 (25)
% SED episodes
during break-times 18 (19) 16 (25) 42 (36) 21 (24) 18 (17) 9 (12) 51 (28) 25 (24) 15 (14) 62 (23) 32 (22) 26 (26)
Abbreviations: SD = standard deviation; LPA = light physical activity; MVPA = Moderate to vigorous physical activity; SED = sedentary activity
Table 5. Target area activity levels - Mean and standard deviation of the proportion (%) of observed episodes for each activity thresholds by Gender (totals for
boys and girls separated) (Mean (SD))
Mean (SD)
Key stage 2 Shared Key stage 1
Target area 1 2 3 4 6 7 8 9 10 11 12 Playground
Girls SED 17 (21) 13 (27) 39 (40) 23 (34) 23 (28) 11 (26) 51 (27) 28 (31) 16 (20) 62 (29) 33 (30) 28 (31)
Girls LPA 28 (27) 49 (36) 43 (39) 45 (41) 32 (31) 51 (44) 28 (25) 50 (34) 47 (28) 30 (26) 38 (27) 41 (33)
Girls MVPA 55 (29) 38 (39) 18 (26) 33 (35) 45 (34) 39 (44) 21 (27) 21 (27) 37 (25) 7 (13) 29 (28) 31 (32)
Boys SED 20 (27) 18 (26) 40 (39) 19 (25) 17 (30) 9 (12) 48 (37) 24 (26) 13 (15) 62 (35) 27 (24) 24 (28)
Boys LPA 23 (25) 59 (33) 39 (34) 43 (26) 33 (36) 52 (23) 26 (31) 46 (30) 27 (26) 24 (33) 31 (20) 38 (29)
Boys MVPA 57 (33) 24 (28) 22 (32) 37 (28) 50 (40) 39 (26) 26 (31) 31 (29) 50 (25) 14 (26) 41 (26) 38 (30)
Abbreviations: SD = standard deviation; LPA = light physical activity; MVPA = Moderate to vigorous physical activity; SED = sedentary activity.
383
384

Table 6. The percentage of observations per target area that were scored for the
presence of each contextual variable
Contextual variable
Target area Supervised (%) Organised (%) Equipped (%)
1 47.1% 0.0% 0.0%
2 31.0% 0.0% 58.6%
3 29.6% 0.0% 55.6%
4 36.1% 19.7% 90.2%
6 16.7% 0.0% 3.3%
7 0.0% 0.0% 92.2%
8 16.7% 0.0% 0.0%
9 47.2% 0.0% 65.6%
10 61.2% 2.0% 18.4%
11 13.3% 0.0% 26.7%
12 38.3% 0.0% 32.8%
Definitions: ‘Supervised’ = The presence of adult supervision for safety; ‘Organised’ = The
presence of an adult in a role involving instruction of an organised activity; ‘Equipped’ = movable
equipment provided by the school
4. Discussion
The aim of this study was to identify the areas of the playground that elicit higher
proportions of MVPA episodes during break-times and explore the contextual and
environmental characteristics present in these areas. The secondary aim is to explore
the effect of providing equipment, playground supervision and the delivery of organised
activities on the number of MVPA episodes (counts) during break-times. A key finding
was that the environmental characteristics of playground zones and the activities they
promote (e.g. team sport, socialisation, and adventure play) had an effect on the
activity levels and behaviours of children during break-times. By observing the
children’s physical activity behaviours in these target areas, we were able to map
385
386
387
388
389
390
391
392
393
394
395

MVPA ‘hotspots’ on the primary school playground. Further we were able to identify the
impact of ‘organisation’, ‘supervision’ and ‘equipment’ provision on the behaviours of
children during break-times.
An important question to consider in the relationship between playground areas and
children’s activity levels is how environmental variables act on children’s decision
making when selecting areas of the playground to “play” in. Schools with manufactured
equipment and installed play equipment; such as climbing frames and goals, have
previously been found to promote the highest levels of MVPA.16,25,39 Findings which are
supported by this study, with areas that had some form of fixed exploratory play
equipment (climbing and balancing) (area 1), resulted in counts for MVPA much higher
than areas without (area 7). The footprint for area 7 was much larger than area 1
(Figure 2); therefore, it had greater capacity to take a larger number of children before
becoming saturated and creating more opportunities for children to be active. However,
target area 1, which included adventure play equipment, had a higher number of MVPA
observations despite being much smaller in size. Furthermore, areas which had a
larger footprint with multiple games and activities promoted (KS area 2; KS area 9 and
12) had higher amounts of sedentary and LPA compared to MVPA. Pawloswki et al.24
highlighted that children spend a large amount of time ‘waiting’ in the school
playground. For example, areas with playground markings and basketball hoops had a
large number of children waiting to take their turn. This may explain the sedentary and
LPA observed in areas with comparable characteristics in this study (areas 2, 9 and
12). In contrast, the momentary sampling method used in this study has the potential to
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417

miss the short, sporadic burst of activity children engage in. Nonetheless, this finding
has the potential to lead playground designers to make more effective use of play
space that may have previously been considered too small to promote physical activity.
Previous systematic reviews affirmed that primary school playgrounds which offer a
variety of strategies (i.e., playground markings, physical structures and games
equipment) aimed at increasing physical activity levels are the most effective.16,21 The
primary school playground observed in this study matches the description of “variety”
implied in Escalante et al.16 and had an extensive range of activities and play spaces
which were accessible and usable during all observations. Findings from the current
study support Escalante’s conclusions, with 73% of all children observed during break-
times showing engagement in some form of physical activity at light or moderate to
vigorous intensity. However, only in one third of the activity episodes children were
observed at MVPA in the playground. A recent meta-analysis highlighted significant
heterogeneity in playground interventions, with just a few studies in each playground
category (e.g., loose equipment, multi-component, playground markings) to be able to
confidently draw conclusions on the ‘most’ effective designs.5 The current study, though
limited by the findings from one school playground, contributes important findings
regarding the type of playground areas which are more effective at encouraging higher
levels of MVPA.
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437

Additionally, there was a trend for slightly higher proportion of MVPA episodes for boys
(38% ± 30%) compared to girls (31% ± 32%). These findings are consistent throughout
the break-time/recess literature, with boys regularly reported as engaging in higher
amounts of physical activity during break-times.21 In addition, despite potential issues
with pseudo-replication in their analysis methods a previous observation study by
Anthamatten, et al.,40 who explored the patterns in children’s physical activity
behaviours within different playground spaces, identified that there was a significant
mean percentage difference (6.73%: 95%CI; 3.5 to 9.9; p value: 0.001) between the
percentage of boys (39.6%) and girls (32.9%) engagement in MVPA in the playground.
4.1 Gender playground choices
There was a higher proportion of MVPA episodes for KS2 boys compared to girls in
area’s which promoted team sports (areas 4 and 7). On the other hand, there was a
higher proportion of MVPA episodes for KS2 girls in area’s that climbing and balancing
activities (areas 1 and 6). However, though there were instances of LPA and MVPA in
these areas, there were also large numbers of children that were perceived to use
these areas as a venue to ‘hang out’ with friends, a predominantly sedentary activity.
These findings are similar to previous research which identified that areas that might be
assumed to promote activity (e.g., balance bars/climbing/tyres) elicited behaviours that
were observed to be ‘social’ and scored lower on physical activity promotion compared
to other (e.g., field) areas of the playground.24 These gender dissimilarities may be
partially explained by a tendency for girls to play in areas not dominated by sport but
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458

instead chose areas which promote social interaction24,29,41 whilst boys tend to dominate
areas within school playgrounds that are designed for competitive sports.41 Previous
studies observations have shown similar differences in utilisation of target area’s by
gender with girls more likely to use areas with a wide variety of play features and
manufactured equipment,5,21,24, and less likely to use areas that have sports equipment
provided.40
Similarly, in this study gender differences were evident in the KS1 playground with
higher numbers of KS1 boys than girls observed in active play areas and a higher
number of KS1 girls than boys observed in the more inactive, social areas of the
playground. The proportion of MVPA observations were also larger for KS1 boys
compared to KS1 girls, in all areas of the KS1 playground (Table 5). The differences
observed between genders in the KS1 playground may be partly explained by a lack of
specified areas for the more dominant activities such as ball sports (e.g., football) in the
playground.27 Previous research has highlighted that providing and enforcing rules that
indicate specific areas for engaging in dominant activities, reduced the use of those
areas for sedentary behaviour and created safer and more accessible situations for
other types of play where ball sports were ‘banned’.25 The KS2 playground in this study
had these allocated ball sports areas, which may explain some of the larger differences
in target area utilisation (Table 3) on the KS2 playground. The KS1 children, however
were forced to share areas for a multitude of activities which might explain the higher
levels of MVPA observed in KS1 boys.
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479

Conversely, there were KS2 boys and girls observed in this study in playground
behaviours that refute the traditional gender stereotypes highlighted previously.29,41 For
example, there were a number of girls in this playground that chose to access areas
that were more popular with boys (areas 4 and 7; team sports) and although fewer in
number, they were as active as the boys observed in the same areas (area 4: girls =
33%; boys = 37%; area 7: girls = 39%; boys = 39%). This was a similar picture when
observing the KS2 social areas that were more favourable with KS2 girls, with boys in
these areas as active as the girls (area 1: girls 55%; boys 57%; and area 6: girls 50%;
boys 45%). This suggests that something other than playground preferences is
responsible for the difference in target area utilisation between boys and girls. Although
we are unable to provide evidence of a relationship between physical activity and
fundamental movement skill competence, the research surrounding this association
might help to explain the differences in children’s activity choices.42 Furthermore, it is
likely that the type of activities children take part in are a result of similarities in
movement ability and movement skill competency, with children of low physical
competence reluctant to approach activities requiring a higher level of ability43
irrespective of differences in gender.
The trend for a proportion of boys and girls to occupy and maintain a similar level of
physical activity in areas more popular with the opposite gender was not apparent in
the KS1 playground areas (table 5). There was a higher proportion of KS1 boys
observed in MVPA in every KS1 area compared to KS1 girls. These age group, gender
differences are important and warrant further investigation; such as: 1) if there is a
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501

driver for boys to seek out activities and areas which promote a higher activity level; 2)
if there are existing barriers for girls to seek these activities; 3) if there is an influence of
the context in which children find themselves at break-times?.
4.2 Contextual factors
Organisation
It has been suggested that the activity levels of primary school children benefit from the
structured break-times by getting help in planning and developing games.24 Although
this is likely to be very different at each school (due to resources and funding) it is
important to understand how this organisation of activities by a teacher, coach or
lunchtime supervisor increase or decrease the likelihood that a child will take part in
MVPA. A recent review and meta-analysis,5 highlighted that interventions which had
implemented ‘structured/organised’ playground activities were not effective at engaging
children in higher intensity physical activity during break-times.
In contrast, the findings from this study would suggest that areas of the playground that
have organised activities (definitive structure with direct adult instruction, and
purposeful activity) are likely to result in an almost three fold (RR = 2.7) increase in the
rate of MVPA episodes compared to areas with less structure. In support of these
findings, Dyment et al.44 established that areas without organisation had the lowest
rates of MVPA on the playground. However, Parrish et al.5 suggest that the time taken
to ‘organise’ playground activities will reduce the already limited time available during
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521

break-times. It was clear that the school in this study had invested time and money in
this area by providing a member of staff whose role was to organise and lead these
activities. Nevertheless, it is apparent that there is a limited amount of time and space
available for facilitating organised activities during break-time, as the presence of
‘organisation’ was limited to very specific areas of the playground (KS2: area 4). The
restriction of organised activities to one area of playground in this study might limit the
opportunities that providing structured activities might have to other areas of the
playground. On the contrary, care should be taken when identifying areas of the
playground to deliver these structured activities to avoid removing opportunities for
children who value unstructured play and the benefits this brings (e.g., socialisation).5
Supervision
Baines and Blatchford17 identified from self-report questionnaires that over the past
twenty-two years there has been a marked increase in adults supervising at break-
times. Organised activities are often supervised by an adult providing informal
instruction, encouragement and a safe space, free from aggressive and dominating
behaviours45 where a child may be more inclined to make an attempt to take part in
something they usually avoid for fear of unwanted negative attention. However,
providing adult supervision in the absence of more structured activities may have a
similar desired effect in promoting physical activity.25 In this study, areas that had
supervision but without formal instruction resulted in more episodes of MVPA (RR 1.34)
compared to areas with no supervision. A systematic review highlighted that the current
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542

evidence on supervised break-times is inconclusive.21 However, the six studies in the
review which focussed on supervised break-times had a variety of staff roles across the
studies. It can be argued that the behaviours and actions of supervisors can result in
different physical activity levels of children.46 In the current study, areas that were
observed as ‘organised’ were almost always ‘supervised’. Study differences in MVPA
levels observed may be the result of the activities that were organised and/or the
actions and behaviours of the supervisor46 and not simply the presence of each
contextual variable in isolation. In contrast, McKenzie et al.26 observed that in
supervised areas, the odds of children engaging in MVPA were half as likely as
unsupervised areas. Future research should explore the variety of roles within primary
schools that occupy the role of playground supervisor and methods of supervision
(e.g., active vs. passive).
Equipment provision
Previous studies have found positive associations between the provision of additional
playground equipment (e.g. sports equipment) and higher rates of MVPA.21,26
Conversely, target areas with additional equipment provided in this study resulted in a
lower number of MVPA episodes observed. The school in this study supplied children
with scooters, tricycles and prams/pushchairs. The presence of dolls and buggies
(stereotypically girls toys), which were provided in this playground resulted in lower
levels of physical activity. This study did not track the gender of the children using the
equipment provided and therefore is unable to provide clarity on this area.
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563

Nevertheless, it is possible that the negative association between provision of
equipment and MVPA in this study is the result of the types of equipment provided. This
is an important consideration as the activity levels in the playground may be mediated
by the gender biased equipment offered.47
Likewise, the children from this study were primarily from low-income families. Wiltshire
et al.28 suggested that disengagement in certain activities because of differences in
social class might already exist in childhood. Any interest or desire to engage with a
target area in this study may have been overwhelmed by this predefined set of
activities or behaviours perceived as “acceptable”, by the predominantly low SES
population in this study. This is important, as children from a lower SES have lower
FMS competence,42 lower physical activity levels48 and a higher incidence of obesity49
contributing to an increased health risk compared to children with a higher SES.
Primary schools comprising of a larger number of children from a higher SES may find
their children react differently to a playground environment.48
Furthermore, the use of scooters and tricycles could quite easily have been scored as
a lower intensity of activity and at times sedentary during the window of observation. It
is likely an assumption of the staff providing children with these forms of equipment that
they will result in a higher intensity of physical activity. Indeed the youth compendium of
physical activities35 suggest these activities can lead to activities of at least moderate
intensity. However, this was not the case during the observations in this study. As the
equipment was predominantly provided to the KS1 area, it is possible that this
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584

particular type of equipment required more advanced movement skills (e.g., balancing
on a scooter) and as a consequence resulted in a higher percentage of children
observed in sedentary activities in these areas (target areas 8, 9 & 12; Table 4).
Moreover, the tricycles had a passenger seat and therefore, for each observation with a
child attempting to pedal their friend around the playground (MVPA), there was a child
observed as sedentary, as a passenger. Further research is needed to address the
impact that the ‘management’ of the playground environment (organisation, supervision
and equipment provision) has on physical activity during playtime.
4.3 Strengths, limitations and future directions
This study set out to explore the different contexts in which children play and the
relationship to physical activity levels. Though the results of this study might not be
generalizable to the primary school population due to the case study methodology,
similarities do exist between the current study and previous research.24,40 However, the
diversity between the town centre location in this study and more rural residing schools
(SES, ethnicity, playground size etc.) might present in a completely different range of
activity behaviours.48
A key strength of this study was the use of validated systematic observation methods to
understand the context in which children are active during break-times. Many
alternative methods (accelerometers and pedometers, self-report) fail to consider the
importance of the context in which physical activity occurs.33 Although systematic
observation is not without its limitations, when the correct methods for training and
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605

calibration of observers is maintained and monitored, it is the most effective method of
measuring child and youth physical activity behaviours.50 Furthermore, the use of video
recordings in this study was considered an advantage in ensuring a higher level of
accuracy and for the ability to resolve uncertainties in activity codes.34
Observations and regression analysis used in this study allowed for comparisons
between contextually diverse target areas. The observational method used in this study
allowed us to get a population level measure of children’s activity levels during break-
times. Therefore, children may have moved between playground areas throughout the
break-time periods, contributing to the activity counts in more than one target area.
Although the regression model and outcomes are largely unaffected by this, the
authors wish to advise readers that for this reason the regression analysis in this study
should be considered exploratory and warrants cautious interpretation of the
confidence intervals for the model as the counts (episodes) of MVPA are not from
independent samples. Therefore, the activity levels reported should be interpreted as
the number of activity episodes observed during break-times and not as individual
children’s physical activity levels. We made an informed decision to not run statistical
analysis since each observation was not from a separate child, and therefore there was
a risk of ‘simple’ pseudo-replication (multiple measures per experimental unit; in this
case a child) and ‘temporal’ pseudo-replication (multiple measures taken successively
in time and treated as different experimental units). Future research should consider
combining measurement methods (GPS tracking, accelerometers and observation)23,24
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626

in order to get individual data and a more holistic view of playground physical activity
levels during break-times in UK primary schools.
Future research should also consider recording the number of boys and girls that use
particular types of equipment during observations. According to Cherney and
Dempsey,51 children determine (in)appropriate gender related behaviours by
responding to their environments very early in life and avoid things they like due to their
perceptions of gender appropriateness. Identifying the variation in activity levels in
response to different types of equipment would enhance the development of future
playgrounds and the activities on offer to the children. Lastly, whilst we acknowledge
that break-times offer opportunities beyond promoting physical activity, this study was
not designed to explore the social and cultural dynamics of the playground, but to
identify the key factors in promoting higher levels of MVPA during break-times.
However, future research should consider examining these variables in addition to, and
in relation to physical activity during school break-times.
5 Conclusion
Physical activity levels of children during break-times are influenced by a number of
external and internal variables. The differences in playground utilisation between
genders is unsurprising, however; the reasons behind these differences are less clear.
The findings from this study suggest the levels and methods of appropriate adult
supervision, and equipment provision can influence the levels of MVPA children
engage at break-times. Further, adult organised activities, though restricted to specific
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647

areas of the playground might also lead to increases in MVPA. Exploring some of the
defining characteristics of the playground areas which have resulted in a higher
number of MVPA counts would aid playground designers in manufacturing play
equipment that promotes a higher level of activity over a larger play space,
incorporating some of the more inactive areas. For example, playground areas which
promote climbing, team sports and adventure play currently promote the highest levels
of MVPA during break-times. This should be considered when deciding on any specific
playground areas to target in the intervention as to not reduce MVPA levels in these
already active areas. Furthermore, consideration should be given to the allocation of
playground space when designing activities, to accommodate larger numbers of
children in the areas aimed at promoting MVPA. Finally, understanding the areas of the
playground which have the potential to promote the higher levels of MVPA can aid
playground supervisors in supporting children to be more active during break-times.
Acknowledgements
The authors would like to thank all volunteers who participated in this study; the
primary school staff, pupils and parents. We would also like to thank the research and
technical staff at Teesside University and Professor Alan Batterham for provided
support with data analysis and provision of equipment during the data collection
process.
Authors’ contributions
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667

MG is the lead author and led the design, data collection and manuscript writing. MG
completed the video editing, and led the SOPLAY training and assisted with data
analysis. DJ and TM assisted with data collection. LA assisted with study design and
SOPLAY training. AI assisted with study design, data collection, SOPLAY training and
data analysis. MW assisted with study design, data collection, SOPLAY training and
data analysis. All authors read and approved the final manuscript and agree with the
order of presentation of the authors.
Competing interests and funding sources
The authors declare that they have no competing interests. This research did not
receive any specific grant from funding agencies in the public, commercial, or not-for-
profit sectors
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685

References
1. UK Department of Health and Social Care (2019) UK Chief Medical Officers’
Physical Activity Guidelines.
2. Sport England (2019). Active lives children and young people survey 2017/18.
Accessed January 2020.
3. National Centre for Social Research, Joint Health Surveys Unit (2008). Heath
Survey for England, 2008: volume 1. Physical activity and fitness. The NHS
information centre, Leeds.
https://files.digital.nhs.uk/publicationimport/pub00xxx/pub00430/heal-surv-phys-
acti-fitn-eng-2008-rep-v2.pdf. (accessed 23 January 2020).
4. Steene-Johannessen J., Hansen BH., Dalene KE. et al. Variations in
accelerometry measured physical activity and sedentary time across Europe –
harmonized analyses of 47,497 children and adolescents. International Journal
of Behavioural Nutrition and Physical Activity; 2020: 17(38)
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702

5. Parrish, AM., Chong, K.H., Moriarty, A.L. et al. Interventions to Change School
Recess Activity Levels in Children and Adolescents: A Systematic Review and
Meta-Analysis. Sports Med; 2020: 50, 2145–2173
6. Gråstén A, Watt A, Liukkonen J, Jaakkola T. Effects of School-Based Physical
Activity Program on Students Moderate-to-Vigorous Physical Activity and
Perceptions of Physical Competence. J Phys Act Health; 2017: 14: p.455-464.
7. Kerr C, Smith L, Charman S, et al. Physical education contributes to total
physical activity levels and predominantly in higher intensity physical activity
categories. European Physical Education Review. 2018; 24(2):152-164.
doi:10.1177/1356336X16672127
8. Fairclough S, Stratton G. Physical education makes you fit and healthy.
Physical education’s contribution to young people’s physical activity levels.
Health Educ Res 2005; 20:14–23. DOI:10.1093/her/cyg101
9. Wood, C., Hall, K. Physical education or playtime: which is more effective at
promoting physical activity in primary school children?. BMC Res Notes 8, 12
(2015). https://doi.org/10.1186/s13104-015-0979-1
10. Tudor-Locke C., Lee SM., Morgan CF. et al. Children’s pedometer-determined
physical activity during the segmented school day. Med Sci Sports Exerc 2006;
38: 1732-1738. DOI: 10.1016/S1728-869X(09)60016-2
11. Griggs G. (2010) For sale – primary physical education. £20 per hour or nearest
offer. Education 3-13; 38(1); p.39-46
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723

12. Rudd JR. Crotti M. Fitton-Davies K et al. Skill Acquisition Methods Fostering
Physical Literacy in Early-Physical Education (SAMPLE-PE): Rationale and
Study Protocol for a Cluster Randomized Controlled Trial in 5–6-Year-Old
Children From Deprived Areas of North West England. Front. Psychol; 2020:
11, 1128.
13. Quarmby, T., Daly-Smith, A., & Kime, N. (2019). ‘You get some very archaic
ideas of what teaching is…’: primary school teachers’ perceptions of the
barriers to physically active lessons. Education 3-13, 47(3), 308-321.
14. Daly-Smith, A., Quarmby, T., Archbold, V.S.J. et al. Using a multi-stakeholder
experience-based design process to co-develop the Creating Active Schools
Framework. Int J Behav Nutr Phys Act 17, 13 (2020).
https://doi.org/10.1186/s12966-020-0917-z
15. Innerd AL, Azevedo LB, Batterham AM. The effect of a curriculum-based
physical activity intervention on accelerometer-assessed physical activity in
schoolchildren: A non-randomised mixed methods controlled before-and-after
study. PloS one. 2019 Dec 5;14(12):e0225997.
16. Escalante Y., García-Hermoso A., Backx K. et al. Playground designs to
increase physical activity levels during school recess: a systematic review.
Health Educ Behav 2014; 41(2): pp.138-144. DOI: 10.1177/109019811349072
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742

17. Baines E., and Blatchford (2019) School break times and young people’s social
lives: A follow up national study. Executive summary, Department of Pscyhology
and human development, UCL institute of Education, London
18. Erwin HE., Ickes M., Ahn S., Fedewa, A. Impact of recess interventions on
children's physical activity—a meta-analysis. American journal of health
promotion; 2014: 28(3): p.159-167.
19. Broekhuizen, K., Scholten, AM. & de Vries, S.I. The value of (pre)school
playgrounds for children’s physical activity level: a systematic review. Int J
Behav Nutr Phys Act; 2014: 11, 59. https://doi.org/10.1186/1479-5868-11-59
20. Norris E., van Steen T., Direito A., Stamatakis E. Physically active lessons in
schools and their impact on physical activity, educational, health and cognition
outcomes: a systematic review and meta-analysis. British Journal of Sports
Medicine,2019: pp.bjsports-2018
21. Ridgers ND., Salmon J., Parrish AM., Stanley RM., Okely AD. Physical activity
during school recess: a systematic review. American journal of preventive
medicine; 2012: 43(3), 320-328.
22.Parrish, AM., Okely, A.D., Stanley, R.M. et al. The Effect of School Recess
Interventions on Physical Activity. Sports Med; 2013: 43, 287–299.
https://doi.org/10.1007/s40279-013-0024-
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761

23. Clevenger KA., Grady SC., Erickson K., Pfeiffer KA. Use of a spatiotemporal
approach for understanding preschoolers’ playground activity. Spatial and
Spatio-temporal Epidemiology; 2020: 35, 100376.
24. Pawlowski CS., Andersen HB., Troelsen J., Schipperijn J. Children’s physical
activity behavior during school recess: A pilot study using GPS, accelerometer,
participant observation, and go-along interview. PloS one; 2016: 11(2),
e0148786.
25. Van Kann DH., de Vries SI., Schipperijn J. et al. Schoolyard characteristics,
physical activity, and sedentary behavior: combining GPS and accelerometry.
Journal of school health; 2016: 86(12), 913-921.
26. McKenzie TL, Crespo NC, Baquero B. et al. Leisuretime physical activity in
elementary schools: analysis of contextual conditions. J Sch Healt 2010;
80(10): 470–477 DOI:10.1111/j.1746-1561.2010.00530.x
27. Ridgers, N.D., Fairclough, S.J. & Stratton, G. Variables associated with
children's physical activity levels during recess: the A-CLASS project. Int J
Behav Nutr Phys Act; 2010: 7, 74. https://doi.org/10.1186/1479-5868-7-74
28. Wiltshire G., Lee J., Williams O. Understanding the reproduction of health
inequalities: physical activity, social class and Bourdieu’s habitus. Sport Educ
Soc 2017; 1-15. DOI: 10.1080/13573322.2017.1367657
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780

29. Renold E. ‘All they’ve got in their brains is football.’ Sport, masculinity and the
gendered practices of playground relations. Sport, Educ Soc;1997: 2: p.5–23
30. Pearce G., Bailey RP. Football pitches and Barbie dolls: young children’s
perceptions of their school playground. Early Child Development and Care;
2011: 181(10): p.1361-1379
31. Darst PW., Zakrajsek DB., Mancini VH. Analyzing physical education and sport
instruction 2nd ed. Human Kinetics Inc. Champaign, IL. 1989
32. McKenzie TL. System for Observing Play and Leisure Activity in Youth
(SOPLAY). Measurement Instrument Database for the Social Sciences 2012.
Retrieved from www.midss.ie
33. McKenzie TL. Context matters: systematic observation of place-based physical
activity. Res Q Exerc Sport 2016; 87(4): pp.334-341. DOI:
10.1080/02701367.2016.1234302
34. Boonzaaijer M., van Dam E., van Haastert IC. et al. Concurrent Validity
Between Live and Home Video Observations Using the Alberta Infant Motor
Scale. Pediatr Phys Ther 2017; 29(2), p.146. DOI:
10.1097/PEP.0000000000000363.
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797

35. Butte NF., Watson KB., Ridley K. et al. A youth compendium of physical
activities: activity codes and metabolic intensities. Med Sci Sports Exerc 2018;
50(2): p.246
36. McGraw KO., Wong SP. Forming inferences about some intraclass correlation
coefficients. Psychol methods 1996; 1(1): p.30.
37. Landis JR., Koch GG. An application of hierarchical kappa-type statistics in the
assessment of majority agreement among multiple observers. Biometrics 1977;
p.363-374
38. Malcata RM., Vandenbogaerde TJ., Hopkins WG. Using athletes' World
rankings to assess countries' Performance. Int J Sports Physiol Performance
2014; 9(1): 133–138. DOI:10.1123/ijspp.2013-0014
39. Farley TA, Meriwether RA, Baker ET. et al. Where do the children play? The
influence of playground equipment on physical activity of children in free play. J
Phys Act Health 2008; 5(2):319–331. DOI: 10.1123/jpah.5.2.319
40. Anthamatten P., Brink L., Kingston B. An assessment of schoolyard features
and behavior patterns in children’s utilization and physical activity. J Phys Act
Health 2014; 11(3): pp.564-573. DOI: 10.1123/jpah.2012-0064
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814

41. Lucas AJ., Dyment JE. Where do children choose to play on the school ground?
The influence of green design. Education 2010; 38:2: 177-189.
DOI:10.1080/03004270903130812
42.Barnett, L.M., Lai, S.K., Veldman, S.L.C. et al. Correlates of Gross Motor
Competence in Children and Adolescents: A Systematic Review and Meta-
Analysis. Sports Med 46, 1663–1688 (2016). https://doi.org/10.1007/s40279-
016-0495-z
43. Barbour AC. The impact of playground design on the play behaviors of children
with differing levels of physical competence. Early Child Res Q 1999; 14(1):
pp.75-98. DOI: 10.1016/S0885-2006(99)80007-6
44. Dyment JE, Bell AC, Lucas AJ. The relationship between school ground design
and intensity of physical activity. Child Geogr 2009; 7(3): 261–276.
DOI:10.1080/14733280903024423
45. Ramstetter C., Murray R., Garner AS. The crucial role of recess in schools. J
Sch Health 2010; 80: 517-526.
46. Caro HEE., Altenburg TM., Dedding C. et al. Dutch primary schoolchildren’s
perspectives of activity-friendly school playgrounds: A participatory study. Int J
Environ Res Public Health 2016; 13(6): 526. DOI: 10.3390/ijerph13060526
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832

47. Weisgram, E. S., Fulcher, M., & Dinella, L. M. (2014). Pink gives girls
permission: Exploring the roles of explicit gender labels and gender-typed
colors on preschool children's toy preferences. Journal of Applied
Developmental Psychology, 35(5), 401-409.
48.Baquet, G., Ridgers, N.D., Blaes, A. et al. Objectively assessed recess physical
activity in girls and boys from high and low socioeconomic backgrounds. BMC
Public Health 14, 192 (2014). https://doi.org/10.1186/1471-2458-14-192
49. Lioret S, Maire B, Volatier JL, et al. Child overweight in France and its
relationship with physical activity, sedentary behaviour and socioeconomic
status. Eur J Clin Nutr. 2007; 61: 509-516
50. Saint-Maurice PF., Welk G., Ihmels MA. et al. Validation of the SOPLAY Direct
Observation Tool With an Accelerometry-Based Physical Activity Monitor. J
Phys Act Health 2011; 8: 1108-1116. DOI: 10.1123/jpah.8.8.1108
51.Cherney ID., Dempsey J. Young children’s classification, stereotyping and play
behaviour for gender neutral and ambiguous toys. Educ Psychol 2010; 30(6):
651-669, DOI:10.1080/01443410.2010.498416
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848