- Email: [email protected]
How important is young children's actual and perceived movement skill competence to their physical activity?
Accepted Manuscript Title: How important is young children’s actual and perceived movement skill competence to their physical activity? Author: Sarah Slykerman Nicola D. Ridgers Christopher Stevenson Lisa M. Barnett PII: DOI: Reference:
S1440-2440(15)00137-1 http://dx.doi.org/doi:10.1016/j.jsams.2015.07.002 JSAMS 1204
To appear in:
Journal of Science and Medicine in Sport
Received date: Revised date: Accepted date:
8-9-2014 26-5-2015 2-7-2015
Please cite this article as: Slykerman S, Ridgers ND, Stevenson C, Barnett LM, How important is young children’s actual and perceived movement skill competence to their physical activity?, Journal of Science and Medicine in Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.07.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
How important is young children’s actual and perceived movement skill competence to their
2
physical activity?
4
Sarah Slykerman1§, Nicola D. Ridgers2, Christopher Stevenson1, Lisa M Barnett1
5
1
6
Australia
7
2
8
VIC 3125, Australia
cr
School of Health and Social Development, Deakin University, Burwood Hwy, Burwood, VIC 3125,
us
Centre for Physical Activity and Nutrition Research, Deakin University, Burwood Hwy, Burwood,
an
9 10 §
Corresponding author: Sarah Slykerman [email protected]
M
11 12
Email addresses:
d
13
SS: [email protected]
15
NDR: [email protected]
16
CS: [email protected]
Ac ce p
18
te
14
17
ip t
3
LMB: [email protected]
19
Word count: 3,279
20
Abstract word count: 252
21
Tables: Three
22
Figures: None
23
1 Page 1 of 19
26 Abstract Objectives: To determine the associations between young children's actual and perceived object
ip t
control and locomotor skills and physical activity and whether associations differ by sex.
cr
Design: Cross sectional study.
Methods: A total of 135 children had actual skill (Test of Gross Motor Development-2), perceived
us
skill (Pictorial Scale of Perceived Movement Skill Competence for Young Children), and moderateto vigorous-intensity physical activity (MVPA) (accelerometers) assessed. Independent t-tests
an
assessed sex differences. A regression (with MVPA as the outcome) was performed with all predictor variables (i.e. Actual Object Control, Actual Locomotor, Perceived Object Control, and Perceived
M
Locomotor). Model 2 also adjusted for age, sex, accelerometer wear time and whether the child was from an English speaking background. Interaction terms between the respective actual or perceived
d
skill factor and sex were added to assess sex differences.
te
Results: Analyses were conducted on 109 children (59 boys, 50 girls; mean age = 6.5 years, SD =
Ac ce p
1.0). Boys had higher actual and perceived object control skill and were more active by an average of 19 minutes per day. There were no sex differences in locomotor skills. There were no associations between skill factors and MVPA, except for girls, where locomotor skill was a significant predictor of MVPA (B = 3.66, p = 0.016).
Conclusions: Actual rather than perceived skill competence was more important to MVPA in this sample. Locomotor skill competence may be more important than object control skill competence for girls as they may engage in types of physical activity that do not require object control mastery. Keywords: Fundamental Movement Skill; Physical self-perception; Object Control Skill Locomotor Skill
3 Page 2 of 19
Introduction
28
There is concern children are not participating in the recommended 60 minutes per day of moderate-
29
to vigorous-intensity physical activity (MVPA) 1. Research has shown it is important to understand
30
the correlates that may influence children’s physical activity participation 2 as without movement skill
31
competence (e.g. the ability to throw, kick and jump proficiently) and a positive perception of such,
32
children may be less likely to be physically active 3, 4. Stodden and colleagues 5 describe this process
33
as a positive spiral of engagement in their conceptual model. Children’s physical activity participation
34
influences their motor skill development 5, and in turn their movement competence influences their
35
physical activity engagement. Perceived movement competence is described as a mediator in this
36
model that also influences levels of physical activity 5. In young children (< 8 years) movement
37
competence is developing and perceptions of competence tend to be inflated. As children age, their
38
perceptions of their abilities better reflect their actual motor competence 5. Thus, the model purports
39
that relationships between the constructs will increase in strength as children age.
cr
us
an
M
d te
40
ip t
27
Whilst the relationship between movement skill competence and physical activity is fairly well
42
established 4, studies in young children (< 8 years) have shown varying associations between skill
43
type (i.e. object control and locomotor) and between boys and girls 6. Hardy 7 and LeGear 8 found
44
young girls (preschool and kindergarten age respectively) had better locomotor skills than boys,
45
potentially due to the rhythm and balance required to perform these skills. However, this ability may
46
not translate to physical activity, with a study in similar-aged children finding an inverse association
47
between locomotor ability and physical activity in girls 9. In contrast, young boys have better object
48
control skill than girls 7 and object control competence has been associated with physical activity in
49
boys 9. Interestingly, two recent studies have recently reported that both object control and locomotor
50
skills were associated with physical activity in young children 10,11, with one of these studies noting
51
that sex did not influence the relationship between object control competence and physical activity 11.
Ac ce p
41
4 Page 3 of 19
Given the mixed findings to date, it is clear that further research is needed to understand these
53
relationships in young children.
Ac ce p
te
d
M
an
us
cr
ip t
52
5 Page 4 of 19
Few studies have focused on perceived competence in young children 3,12, 13. Young children tend to
55
inflate their perception of their physical competence and therefore perceptions are not always accurate
56
when matched with teacher’s perceptions of their fundamental movement skills 14,15. Nevertheless it is
57
still important to investigate perceived competence in young children as it is a known determinant of
58
physical activity in older children and adolescents 3. Existing studies tend to investigate more general
59
perceptions of physical competence. One study in young obese children (aged 5-9 years) found
60
perceived competence was not associated with physical activity 13. In contrast, Robinson and
61
colleagues reported a moderate positive association between perceived competence and physical
62
activity in younger children (mean age of 4 years) 12. Interestingly, Robinson also reported that
63
children in this study exhibited low perceived physical competence scores in comparison to other
64
similar populations. This is a concern as children with low self-perceptions may lose motivation and
65
interest in movement related tasks, leading to lower levels of physical activity 12. This also reinforces
66
the need to investigate perceived physical competence in this age group.
M
an
us
cr
ip t
54
67
Studies investigating perceived movement competence in young children have not aligned the
69
assessment of actual skill with the assessment of perceived movement competence 8, 12. In other words,
70
children complete assessments of actual skills that do not match the same skills they are asked to rate
71
how good they are at. For instance, one commonly used instrument does not have an object control
72
item for the preschool age group, and for the school age group, only one of the six items is an object
73
control item 14. It is possible that children may have different perceptions of how good they are
74
depending on the type of skill 16. Only one other study could be located that has explored skill type in
75
terms of competence perceptions and their association with physical activity. Barnett and colleagues
76
used the TGMD-2 to assess object control skills, but unlike other studies, assessed perceived
77
competence in object control skills using the same pictorial instrument as the current study, which
78
matches skill perceptions to the TGMD-2 skills 17. The purpose of the current study was therefore to
79
determine if actual and perceived skills (in terms of object control and locomotor) are associated with
80
young children’s MVPA and whether observed associations differed by sex.
Ac ce p
te
d
68
81 6 Page 5 of 19
Method
83
This cross-sectional study was approved by the University Human Ethics Advisory Group-Health.
84
Convenience sampling was used to recruit from two primary schools in Victoria, Australia, with
85
School Principals providing written consent for the schools to participate in June 2013. Children were
86
eligible to participate if they were in the first three years of school (Prep-Grade 2; aged 5 to 8 years)
87
and returned their parental consent form. Informed written parent/guardian consent was obtained for
88
135 children (44% consent response rate; 135/308). Demographic information was parent-reported
89
through a survey returned at the time of consent. Information reported included child date of birth,
90
parent country of birth, whether English is the language spoken at home, and parents highest level of
91
education. Data were collected in August 2013.
92
The Test of Gross Motor Development-2 (TGMD-2) was used to evaluate children’s actual
93
Fundamental Movement Skill (FMS) competence, in six object control skills (throw, catch, roll, kick,
94
hit and bounce) and six locomotor skills (run, gallop, hop, leap, jump and slide) 18. Each skill has
95
between three and five performance component indicators 18. If children successfully execute a
96
component of the skill, they received a score of one, or a zero if not. Children completed two trials of
97
each skill live in the field and the sum of each skill score was summed to obtain an overall object
98
control skill (max out of 48) and locomotor skill score (max out of 48) 18. Inter-rater reliability was
99
conducted for 21/63 paired rater combinations in this study, which equated to 252 skill tests. The
Ac ce p
te
d
M
an
us
cr
ip t
82
100
overall ICC was 0.78 (95% confidence intervals 0.54 - 0.91) indicating good reliability between raters
101
for the scores given.
102
Children’s perceived FMS were assessed using a modified version of the Pictorial Scale of Perceived
103
Competence and Acceptance for Young Children 14 which assessed the same 12 skills assessed in the
104
TGMD-2 16. Each child was shown an illustration of six object control and six locomotor skills, where
105
a boy or girl completing that skill ‘competently’ or ‘not so competently’ was presented. All children
106
were asked if they were like the child illustrating the task competently or the child illustrating the task
107
performed not as competently. If the child selected the competent illustration, they were then asked if
7 Page 6 of 19
they were ‘really good’ at the skill or ‘pretty good’ at the skill 16. If the child selected the not so
109
competent illustration, they were then asked if they were ‘not too good’ at the skill or ‘sort of good’ at
110
the skill. Each item was scored on a one to four scale, with one indicating low competency to four
111
being highly competent. This process was repeated for all 12 skills to give a maximum total score of
112
24 for each the object control and locomotor skills 16. This instrument has good internal consistency
113
for object control (Test 1:α = 0.63, Test 2:α = 0.72), and locomotor skills (Test 1:α = 0.64, Test 2:α =
114
0.68) 16. This instrument has also been assessed for 7 day test re-test reliability with good results (0.78
115
[95% CI 0.60-0.89]) for object control and locomotor skill perceptions (0.82 [95% CI 0.63-0.92]) 16.
116
MVPA was objectively assessed for eight consecutive days using GT3X+ ActiGraph (Pensacola, FL)
117
accelerometers. Data from the vertical plane were utilised in this study. To be included in the analyses,
118
children needed to have worn the monitor for ≥8 hours a day for a minimum of 3 days 19. A 15s epoch
119
length was used. Children were shown in small groups how to wear accelerometers, and the
120
circumstances in which they could remove them (e.g. if engaged in water-based activities and when
121
sleeping). Children were instructed to wear the monitor on the right hip, attached using an elasticated
122
nylon belt. This monitor has been validated for use in this population and has acceptable reliability 20.
123
Accelerometer data were initially downloaded using ActiLife Software (version 6.0) and checked for
124
compliance. The accelerometer files were then processed using a customised macro in Microsoft Excel
125
2010. Non-wear time was indicated by periods of 20 minutes of consecutive zeros 21. The
126
accelerometer cut-points developed by Evenson and colleagues 22 were used to define moderate and
127
vigorous intensity physical activity (2296-4011 cpm for moderate, and ≥4012 cpm for vigorous
128
intensity physical activity) 22. These cut-points have been validated and have recently been found to
129
provide acceptable classifications of moderate and vigorous physical activity in children 23. Time spent
130
in moderate and vigorous intensity physical activity were summed to obtain time spent in MVPA.
131
The mean and standard deviation was calculated for all variables. Differences between boys’ and girls’
132
actual object control, actual locomotor, perceived object control, perceived locomotor and MVPA
133
were assessed using independent t-tests. TGMD-2 raw scores were standardised according to the
134
manual18 and used in the analyses. To examine associations between actual and perceived object
Ac ce p
te
d
M
an
us
cr
ip t
108
8 Page 7 of 19
control and locomotor skills and physical activity, regressions with MVPA as the outcome were
136
initially performed with all predictor variables (i.e. Actual Object Control, Actual Locomotor,
137
Perceived Object Control, and Perceived Locomotor) (Model 1). Model 2 also adjusted for age
138
(months), sex, accelerometer wear time and English speaking background or not. Interaction terms
139
between the respective actual or perceived skill factor and sex were added to assess sex differences in
140
the predictor-MVPA association. For any significant or close to significant sex interaction terms (p <
141
0.1) we reran the model separately by sex. Stata version 11 was used for all analyses and p<0.05 was
142
the set significance value.
143
Results
144
Only the 109 children with complete data were included in the analysis. These children (59 boys, 50
145
girls) aged between five and eight (M = 6.5 years, SD = 1.0) completed all assessments, or in the case
146
of accelerometry, met inclusion criteria. A total of 28% of children spoke English at home and 44% of
147
parents were not born in Australia. Three quarters of parents had a tertiary qualification (77%), 12%
148
had completed high school, 8% had completed technical/trade school certificate or apprenticeships and
149
3% had not finished high school.
te
d
M
an
us
cr
ip t
135
150
Table 1 reports children’s actual and perceived object control, locomotor competence and MVPA and
152
the observed sex differences for these variables. Boys had significantly higher actual and perceived
153
object control skill and MVPA compared to girls. There were no significant sex differences in
154
children’s actual or perceived locomotor skills.
155
INSERT TABLE 1.
156
Model 1 (the unadjusted model) found that none of the skill variables were significant predictors of
157
MVPA. Similarly, in Model 2 (after adjusting for age, sex, English speaking background and monitor
158
wear time), none of the skill variables were significant predictors of MVPA (Table 2). However the
159
interaction term for locomotor skill (standardised) was significant (p = 0.029). The adjusted model was
160
rerun separately by sex and showed that for girls, locomotor skill was a significant predictor of MVPA
Ac ce p
151
9 Page 8 of 19
(B = 3.66, p = 0.016). For boys, there was no association between actual locomotor skill and MVPA
162
(B = -1.74, p = 0.245). See Table 3.
163
INSERT TABLE 2 and TABLE 3.
164
Discussion
165
The purpose of this study was to explore the relationship between children’s actual skill competence
166
and MVPA, and children’s perceived skill competence and MVPA, and to determine if any observed
167
associations differed by sex. This study only observed one significant association, which was for
168
actual locomotor competence and MVPA in girls. This finding can be interpreted as a 1.21 increase in
169
physical activity minutes with each successful locomotor skill component, indicating that a girl with
170
one more correct component on each of the six skills would engage in seven more minutes of activity
171
per day.
172
The current study contrasts previous research in young children that have found locomotor skill to be
173
associated with physical activity for both boys and girls. Williams and colleagues 25 found children in
174
the highest tertile of locomotor skill performance spent significantly more time in MVPA, whilst
175
Robinson and colleagues 26 found actual locomotor competence was moderately associated with
176
pedometer steps. Similarly, Iiovnen et al. and Foweather et al. also found locomotor skills were
177
associated with physical activity in young children 10,11. It is not clear why we found this relationship
178
only for girls, not boys. Locomotor skill competence may be more important than object control skill
179
competence for girls of this age as they may be engaging in types of physical activity (e.g. dance,
180
gymnastics) that do not require object control mastery. This is reinforced by Hardy 7 and LeGear 8
181
who found young girls had better locomotor skills than boys. Cliff and colleagues 9, using the TGMD-
182
2 and using accelerometers, also found no association between locomotor skill and physical activity
183
for boys (although it was approaching significance). An association was found between locomotor
184
skill and MVPA for girls, but it was an inverse association 9. The study by Cliff and colleagues
185
stratified by sex in a small sample (<50), which might have produced this unusual result for girls. In
186
another paper Cliff suggests that children may have to sustain a higher movement skill competency
Ac ce p
te
d
M
an
us
cr
ip t
161
10 Page 9 of 19
level for some time before this translates into MVPA 28. It is apparent that the relationship between
188
locomotor skill type and physical activity behaviour is not clear in this age group and further research
189
is warranted.
190
This study did not find an association between actual object control skill and MVPA. One study in
191
very similar aged Australian children using the same instruments and the same physical activity cut-
192
points also found object control skill was not associated with MVPA 17. Similarly, one study in
193
younger children (3 and 4 years) found no relationship between object control and MVPA when sex
194
and other potential confounders were adjusted for 25. Another study with very active 8 year old
195
children (mean of 93 minutes MVPA) also found that object control skill was not associated was
196
physical activity 27. In our study, object control skill initially approached significance with MVPA in
197
the unadjusted model (p = 0.056). Similarly, the previous Australian study found object control skill
198
approached significance with physical activity17, so it may mean that this relationship is only just
199
emerging in this age group. Yet, other studies, all in pre-schoolers 9, 10, 11, 24, have reported a positive
200
association between object control skill ability and MVPA, which suggests this relationship can
201
emerge at a young age in certain samples. Whilst one of these studies used the TGMD-2 24, the other
202
studies all used different instruments to assess motor skill 9, 10, 11, which makes comparisons difficult. It
203
is apparent that the relationship between skill type and physical activity behaviour is not clear and is
204
influenced by age, sex and instrumentation differences. As Stodden and colleagues 5 advocate, the role
205
that movement skill plays in MVPA might not be straightforward, relationships may change by age.
206
There was no association found in this study between perceived skill competence (object control or
207
locomotor) and physical activity. Only one other study could be located that has explored skill type in
208
terms of competence perceptions and their association with physical activity. Barnett and colleagues 17
209
also used the TGMD-2, but unlike other studies, assessed perceived competence using the same
210
pictorial instrument as the current study, which matches skill perceptions to the TGMD-2 skills.
211
Similar to the current study, they found perceived object control competence was not associated with
212
MVPA, after adjusting for sex and age. Also similar to the current study, that study found sex
213
differences in perceived object control skill (with boys higher) which indicates children do have some
Ac ce p
te
d
M
an
us
cr
ip t
187
11 Page 10 of 19
ability to rate themselves as these differences were also present in the actual skill data. Nevertheless
215
children generally had high perceptions of their ability and this lack of variability may have reduced
216
the ability to detect an association with MVPA. No other study could be located that has specifically
217
explored perceived locomotor competence as a predictor of MVPA.
218
Other studies in young children which have used more general perception assessments have shown
219
associations between young children’s perceived physical competence and physical activity 8, 12,
220
suggesting that perceptions could potentially have an influence on actual physical activity behaviour.
221
Although, another study in a similar age group did not find perceived physical competence to be
222
associated with pedometer steps 26. The fact that some studies have found associations in young
223
children suggests that developing general perceptions of physical competence which are broader than
224
just FMS competence (but may need to include FMS competence), is important because previous
225
evidence indicates that this predicts PA 3.
226
There is a limit on the generalizability of this study as it is a cross-sectional design with a relatively
227
small convenience sample. The strengths of the study include the matching of perceived movement
228
skills to a commonly used assessment of actual movement skill ability enabling the field to be
229
progressed by understanding the contribution of specific types of skill perceptions to physical activity,
230
and the use of GTX3+ accelerometers to objectively assess the sporadic nature of children’s physical
231
activity. However, hip-mounted accelerometers cannot capture some object control skills (such as
232
striking a ball or catching) since the motion and movement required to complete those skills are in the
233
arms and upper body. Consequently, accelerometers may register a lack of movement or, at the most,
234
light physical activity. This has potential implications on levels of MVPA recorded.
235
Conclusion
236
Actual rather than perceived skill competence was more important to MVPA in this sample. This
237
reinforces the conceptual model posited by Stodden et al. that actual competence precedes perceived
238
competence as a predictor of PA 5. Although, this study found that only locomotor competence was
239
weakly associated with MVPA, and only for girls. Adequately mastering these skills during childhood
Ac ce p
te
d
M
an
us
cr
ip t
214
12 Page 11 of 19
is important as many sporting activities require locomotor based competence. Therefore, childhood is
241
a prime age at which to encourage skill development through community-based and school-based
242
physical activity interventions as a strategy to promote long-term activity. The recent Morgan and
243
colleagues’ systematic review on the effectiveness of skill interventions found a large effect for both
244
gross motor skill and locomotor skill which provides support for such interventions 29. This study
245
found no associations found between perceived skill competence (object control or locomotor) and
246
MVPA. Developing general perceptions of physical competence is important because previous
247
evidence indicates that this predicts MVPA, especially in older children 3 where competence
248
perceptions become more in line with actual ability. Therefore, perceived physical competence should
249
still be considered during the development of physical education programs within this age group in
250
order to promote confidence and enjoyment of activities undertaken. Although some experimental
251
research has been unable to demonstrate causal associations between FMS, perceived physical
252
competence and objectively measured physical activity in children 28, further evidence is needed to
253
understand causal associations between these correlates and physical activity across different age
254
groups.
255
Practical Implications
257 258 259
cr
us
an
M
d te
Encourage school and community based FMS interventions in order to improve FMS in boys
Ac ce p
256
ip t
240
and girls.
Further research evidence is needed before recommending that perceived movement skill competence be targeted to promote MVPA in children.
260
While children's skill perceptions were not associated with physical activity, activities which
261
build skill and promote positive self-perception are warranted as these relationships are
262
purported to strengthen as children age.
263
13 Page 12 of 19
263
Acknowledgements
265
Acknowledgements provided in cover letter as not anonymous.
Ac ce p
te
d
M
an
us
cr
ip t
264
14 Page 13 of 19
266 267
References
268
1
269
pitfalls, and prospects. The Lancet. 2012; 380(9838): 247-257.
270
2
271
physically active and others not? The Lancet. 2012; 380: 258.
272
3
273
Youth: Systematic Review and Meta-Analysis. Sports Med. 2014. In press doi: 10.1007/s40279-014-
274
0229-z.
275
4
276
Adolescents. Sports Med. 2010; 40(12): 1019-1035.
277
5
278
of Motor Skill Competence in Physical Activity: An Emergent Relationship. Quest. 2008; 60(1): 290-
279
306.
280
6
281
children and adolescents: A systematic review. Psychol Sport Exerc. 2014; 15(4): 382-391.
282
7
283
children. J Sci Med Sport. 2009; 13(5): 503-508.
284
8
285
perceptions of competence of children in kindergarten. Int J Behav Nutr Phys Act. 2012; 9: 29.
286
9
287
objectively measured physical activity in pre-school children. Pediatr Exerc Sci. 2009; 21(4): 436-
288
439.
289
10
290
physical activity in 4-year-old preschool children. Percept Mot Skills 2013; 117(2):627-646.
ip t
Hallal PC, Andersen LB, Bull FC, et al. Global physical activity levels: surveillance progress,
cr
Bauman AE, Reis RS, Sallis JF, et al. Correlates of physical activity: Why are some people
us
Babic MJ, Morgan PJ, Plotnikoff RC, et al. Physical Activity and Physical Self-Concept in
an
Lubans DR, Morgan PJ, Cliff DP, et al. Fundamental Movement Skills in Children and
d
M
Stodden DF, Goodway JD, Langendorfer JS, et al. A Developmental Perspective on the Role
te
Holfelder B, Schott N. Relationship of fundamental movement skills and physical activity in
Ac ce p
Hardy LL, King L, Farrell L, et al. Fundamental movement skills among Australian preschool
LeGear M, Greyling L, Sloan E, Bell RI, et al. A window of opportunity? Motor skills and
Cliff DP, Okely AD, Smith L, et al. Relationship between fundamental movement skills and
Iiovnen KS, Sääkslahti K, Mehtälä A et al. Relationship between fundamental motor skills and
15 Page 14 of 19
291
11
292
and weekday and weekend physical activity in preschool children. J Sci Med Sport 2014; DOI:
293
10.1016/j.jsams.2014.09.014
294
12
295
motor skills in preschool children. Child Care Hlth Dev. 2011; 37(4): 589-596.
296
13
297
obese children. Obesity 2008; 16(12): 2634-41.
298
14
299
children. Child Dev. 1984; 55: 1970-1982.
300
15
301
competence, physical fitness and self-perception in children. Child Care Hlth Dev. 2012; 38(3): 394-
302
402.
303
16
304
for assessing fundamental movement skill perceived competence in young children. J Sci Med Sport.
305
2015; 18: 98.
306
17
307
actual ball skill competence and physical activity. J Sci Med Sport. 2015; 18: 167
308
18
309
2000.
310
19
311
physical activity. Arch Dis Child. 2007; 92(11): 963-969.
312
20
313
(CSA) activity monitor in children. Med Sci Sports Exerc. 1998; 30(4): 629-633.
314
21
315
Sports Exerc. 2005; 37: S523-S30.
316
22
317
activity for children. J Sports Sci. 2008; 26(14): 1557-65.
Foweather L, Knowles Z, Ridgers N et al. Relationship between fundamental movement skills
ip t
Robinson LE. The relationship between perceived physical competence and fundamental
cr
Morgan P, Okely A, Cliff D et al. Correlates of objectively measured physical activity in
us
Harter S, Pike R. The pictorial scale of perceived competence and social acceptance for young
M
an
Vedul-Kjelsas V, Sigmundsson H, Stensdotter AK, et al. The relationship between motor
d
Barnett LM, Ridgers ND, Zask A, et al. Face validity and reliability of a pictorial instrument
Ac ce p
te
Barnett LM, Ridgers ND, Salmon J. Associations between young children's perceived and
Ulrich DA. The Test of Gross Motor Development. 2nd edn. edn. Austin, TX, USA.: Pro-ed
Riddoch CJ, Mattocks C, Deere K, et al. Objective measurement of levels and patterns of
Trost SG, Ward DS, Moorhead SM, et al. Validity of the computer science and applications
Freedson P, Pober D, Janz KF. Calibration of Accelerometer Output for Children. Med Sci
Evenson KR, Catellier DJ, Gill K, et al. Calibration of two objective measures of physical
16 Page 15 of 19
318
23
319
activity intensity in youth. Med Sci Sports Exerc. 2011; 43(7): 1360-8.
320
24
321
children's motor skill proficiency. J Sci Med Sport. 2013; 16(4): 332-6.
322
25
323
preschool children. Obesity. 2008; 16(6): 1421-1426.
324
26
325
Preschool Students. Res Q Exerc Sport. 2012; 83: 20-26.
326
27
327
What about physical activity and weight status?. J PA & Health. 2015. In press.
328
28
329
Randomized Controlled Trial. Med Sci Sports Exerc. 2011; 43: 90-100.
330
29
331
movement skill interventions in youth: A systematic review and meta-analysis. Pediatrics. 2013; 132:
332
e1361.
Barnett L, Hinkley T, Okely AD, et al. Child, family and environmental correlates of
ip t
Williams HG, Pfeiffer KA, O'Neill JR, et al. Motor skill performance and physical activity in
us
cr
Robinson LE, Wadsworth DD, Peoples CM. Correlates of School-Day Physical Activity in
an
Barnett LM, Zask A, Rose L et al. Three year follow-up of an early childhood intervention:
M
Cliff D, Okely A, Morgan P et al. Movement Skills and Physical Activity in Obese Children:
te
d
Morgan PJ, Barnett LM, Cliff DP, Okely AD, Scott HA, Cohen KE, et al. Fundamental
Ac ce p
333
Trost SG, Loprinzi PD, Moore R, et al. Comparison of accelerometer cut points for predicting
17 Page 16 of 19
333
Table 1: Children’s actual and perceived object control, locomotor competence and MVPA mins/day,
334
and sex differences
Range
Boys
Girls
Sex differencesb
(n =109)
(n = 59)
(n = 50)
p-value
ip t
Variables
Total
Mean (SD) Mean (SD) Mean (SD) 9 - 45
29.5 (8.1)
32.1 (8.3)
Actual locomotor (raw score)
16 - 44
31.2 (5.6)
30.4 (5.4)
Perceived object control
13 - 24
20.0 (3.1)
20.8 (2.6)
19.0 (3.4)
p = 0.002
Perceived locomotor
10 - 24
20.6 (2.8)
20.4 (2.5)
20.8 (3.1)
p = 0.526
MVPA (mins/day)a
28 - 127
67.4 (21.2)
57.0 (17.7)
p < 0.001
336
(b) Independent t-tests examined sex differences
p = 0.134
d te Ac ce p
337
32.0 (5.8)
us
an
(a) Assessed by accelerometry
76.1 (20.0)
p < 0.001
M
335
26.4 (6.7)
cr
Actual object control (raw score)
18 Page 17 of 19
Table 2: Regression of MVPA (mins/day) including all predictors: non-adjusted and adjusted Models Predictors
Standard
Lower
Upper
error
95%
of B 1.80
Lower
Upper
95%
error
95%
95%
CI
CI
of B
CI
0.93
-0.05
3.64
0.056 1.35
1.31
0.71
-0.11
2.72
0.070
0.01
1.06
-2.12
2.10
p
B
0.85
control
Ac ce p
(Std) Perceived
locomotor
3.03
0.113
0.07
0.68
-1.41
1.27
0.914
0.991 0.49
1.01
-1.51
2.50
0.625
0.53
0.70
-1.92
0.87
0.455
te
locomotor
d
control Actual
-0.33
M
object
an
(Std) Perceived
CI
us
object
p
ip t
Standard
B
Actual
Adjusted model1
Non-adjusted model
cr
337
0.94
0.75
-2.43
0.55
0.213
338
1
339
monitor wear time statistically significant at the 5% level
340
Std = Standardised Score
Age, Sex, English speaking, monitor wear time. Sex was statistically significant at the 1% level and
341 342 343
19 Page 18 of 19
Table 3: Regression of MVPA (mins/day) for Boys and Girls including all predictors and adjusted Predictors
Boys
Actual
Standard
Lower
Upper
Standard
Lower
Upper
error
95%
95%
error
95%
95%
of B
CI
CI
of B
CI
CI
1.82
1.11
-0.41
4.05
0.108
1.35
-2.77
2.68
0.972
0.29
1.09
-1.91
2.48
0.794
1.74
1.48
-4.70
1.23
1.52
1.09
p
B
0.05
us
object control
an
(Std) Perceived
0.71
0.85
-2.43
1.01
0.409
0.245 3.66
1.46
0.72
6.61
0.016*
0.92
-1.21
2.50
0.488
control
(Std)
-3.70
0.66
0.168 0.64
Ac ce p
Perceived
te
d
locomotor
M
object
Actual
p
ip t
B
Girls
cr
343
locomotor 344
Both models adjusted for age, English speaking and monitor wear time
345
Monitor wear time was significant for boys at the 5% level
346
Std = Standardised Score
347 348
20 Page 19 of 19
S1440-2440(15)00137-1 http://dx.doi.org/doi:10.1016/j.jsams.2015.07.002 JSAMS 1204
To appear in:
Journal of Science and Medicine in Sport
Received date: Revised date: Accepted date:
8-9-2014 26-5-2015 2-7-2015
Please cite this article as: Slykerman S, Ridgers ND, Stevenson C, Barnett LM, How important is young children’s actual and perceived movement skill competence to their physical activity?, Journal of Science and Medicine in Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.07.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
How important is young children’s actual and perceived movement skill competence to their
2
physical activity?
4
Sarah Slykerman1§, Nicola D. Ridgers2, Christopher Stevenson1, Lisa M Barnett1
5
1
6
Australia
7
2
8
VIC 3125, Australia
cr
School of Health and Social Development, Deakin University, Burwood Hwy, Burwood, VIC 3125,
us
Centre for Physical Activity and Nutrition Research, Deakin University, Burwood Hwy, Burwood,
an
9 10 §
Corresponding author: Sarah Slykerman [email protected]
M
11 12
Email addresses:
d
13
SS: [email protected]
15
NDR: [email protected]
16
CS: [email protected]
Ac ce p
18
te
14
17
ip t
3
LMB: [email protected]
19
Word count: 3,279
20
Abstract word count: 252
21
Tables: Three
22
Figures: None
23
1 Page 1 of 19
26 Abstract Objectives: To determine the associations between young children's actual and perceived object
ip t
control and locomotor skills and physical activity and whether associations differ by sex.
cr
Design: Cross sectional study.
Methods: A total of 135 children had actual skill (Test of Gross Motor Development-2), perceived
us
skill (Pictorial Scale of Perceived Movement Skill Competence for Young Children), and moderateto vigorous-intensity physical activity (MVPA) (accelerometers) assessed. Independent t-tests
an
assessed sex differences. A regression (with MVPA as the outcome) was performed with all predictor variables (i.e. Actual Object Control, Actual Locomotor, Perceived Object Control, and Perceived
M
Locomotor). Model 2 also adjusted for age, sex, accelerometer wear time and whether the child was from an English speaking background. Interaction terms between the respective actual or perceived
d
skill factor and sex were added to assess sex differences.
te
Results: Analyses were conducted on 109 children (59 boys, 50 girls; mean age = 6.5 years, SD =
Ac ce p
1.0). Boys had higher actual and perceived object control skill and were more active by an average of 19 minutes per day. There were no sex differences in locomotor skills. There were no associations between skill factors and MVPA, except for girls, where locomotor skill was a significant predictor of MVPA (B = 3.66, p = 0.016).
Conclusions: Actual rather than perceived skill competence was more important to MVPA in this sample. Locomotor skill competence may be more important than object control skill competence for girls as they may engage in types of physical activity that do not require object control mastery. Keywords: Fundamental Movement Skill; Physical self-perception; Object Control Skill Locomotor Skill
3 Page 2 of 19
Introduction
28
There is concern children are not participating in the recommended 60 minutes per day of moderate-
29
to vigorous-intensity physical activity (MVPA) 1. Research has shown it is important to understand
30
the correlates that may influence children’s physical activity participation 2 as without movement skill
31
competence (e.g. the ability to throw, kick and jump proficiently) and a positive perception of such,
32
children may be less likely to be physically active 3, 4. Stodden and colleagues 5 describe this process
33
as a positive spiral of engagement in their conceptual model. Children’s physical activity participation
34
influences their motor skill development 5, and in turn their movement competence influences their
35
physical activity engagement. Perceived movement competence is described as a mediator in this
36
model that also influences levels of physical activity 5. In young children (< 8 years) movement
37
competence is developing and perceptions of competence tend to be inflated. As children age, their
38
perceptions of their abilities better reflect their actual motor competence 5. Thus, the model purports
39
that relationships between the constructs will increase in strength as children age.
cr
us
an
M
d te
40
ip t
27
Whilst the relationship between movement skill competence and physical activity is fairly well
42
established 4, studies in young children (< 8 years) have shown varying associations between skill
43
type (i.e. object control and locomotor) and between boys and girls 6. Hardy 7 and LeGear 8 found
44
young girls (preschool and kindergarten age respectively) had better locomotor skills than boys,
45
potentially due to the rhythm and balance required to perform these skills. However, this ability may
46
not translate to physical activity, with a study in similar-aged children finding an inverse association
47
between locomotor ability and physical activity in girls 9. In contrast, young boys have better object
48
control skill than girls 7 and object control competence has been associated with physical activity in
49
boys 9. Interestingly, two recent studies have recently reported that both object control and locomotor
50
skills were associated with physical activity in young children 10,11, with one of these studies noting
51
that sex did not influence the relationship between object control competence and physical activity 11.
Ac ce p
41
4 Page 3 of 19
Given the mixed findings to date, it is clear that further research is needed to understand these
53
relationships in young children.
Ac ce p
te
d
M
an
us
cr
ip t
52
5 Page 4 of 19
Few studies have focused on perceived competence in young children 3,12, 13. Young children tend to
55
inflate their perception of their physical competence and therefore perceptions are not always accurate
56
when matched with teacher’s perceptions of their fundamental movement skills 14,15. Nevertheless it is
57
still important to investigate perceived competence in young children as it is a known determinant of
58
physical activity in older children and adolescents 3. Existing studies tend to investigate more general
59
perceptions of physical competence. One study in young obese children (aged 5-9 years) found
60
perceived competence was not associated with physical activity 13. In contrast, Robinson and
61
colleagues reported a moderate positive association between perceived competence and physical
62
activity in younger children (mean age of 4 years) 12. Interestingly, Robinson also reported that
63
children in this study exhibited low perceived physical competence scores in comparison to other
64
similar populations. This is a concern as children with low self-perceptions may lose motivation and
65
interest in movement related tasks, leading to lower levels of physical activity 12. This also reinforces
66
the need to investigate perceived physical competence in this age group.
M
an
us
cr
ip t
54
67
Studies investigating perceived movement competence in young children have not aligned the
69
assessment of actual skill with the assessment of perceived movement competence 8, 12. In other words,
70
children complete assessments of actual skills that do not match the same skills they are asked to rate
71
how good they are at. For instance, one commonly used instrument does not have an object control
72
item for the preschool age group, and for the school age group, only one of the six items is an object
73
control item 14. It is possible that children may have different perceptions of how good they are
74
depending on the type of skill 16. Only one other study could be located that has explored skill type in
75
terms of competence perceptions and their association with physical activity. Barnett and colleagues
76
used the TGMD-2 to assess object control skills, but unlike other studies, assessed perceived
77
competence in object control skills using the same pictorial instrument as the current study, which
78
matches skill perceptions to the TGMD-2 skills 17. The purpose of the current study was therefore to
79
determine if actual and perceived skills (in terms of object control and locomotor) are associated with
80
young children’s MVPA and whether observed associations differed by sex.
Ac ce p
te
d
68
81 6 Page 5 of 19
Method
83
This cross-sectional study was approved by the University Human Ethics Advisory Group-Health.
84
Convenience sampling was used to recruit from two primary schools in Victoria, Australia, with
85
School Principals providing written consent for the schools to participate in June 2013. Children were
86
eligible to participate if they were in the first three years of school (Prep-Grade 2; aged 5 to 8 years)
87
and returned their parental consent form. Informed written parent/guardian consent was obtained for
88
135 children (44% consent response rate; 135/308). Demographic information was parent-reported
89
through a survey returned at the time of consent. Information reported included child date of birth,
90
parent country of birth, whether English is the language spoken at home, and parents highest level of
91
education. Data were collected in August 2013.
92
The Test of Gross Motor Development-2 (TGMD-2) was used to evaluate children’s actual
93
Fundamental Movement Skill (FMS) competence, in six object control skills (throw, catch, roll, kick,
94
hit and bounce) and six locomotor skills (run, gallop, hop, leap, jump and slide) 18. Each skill has
95
between three and five performance component indicators 18. If children successfully execute a
96
component of the skill, they received a score of one, or a zero if not. Children completed two trials of
97
each skill live in the field and the sum of each skill score was summed to obtain an overall object
98
control skill (max out of 48) and locomotor skill score (max out of 48) 18. Inter-rater reliability was
99
conducted for 21/63 paired rater combinations in this study, which equated to 252 skill tests. The
Ac ce p
te
d
M
an
us
cr
ip t
82
100
overall ICC was 0.78 (95% confidence intervals 0.54 - 0.91) indicating good reliability between raters
101
for the scores given.
102
Children’s perceived FMS were assessed using a modified version of the Pictorial Scale of Perceived
103
Competence and Acceptance for Young Children 14 which assessed the same 12 skills assessed in the
104
TGMD-2 16. Each child was shown an illustration of six object control and six locomotor skills, where
105
a boy or girl completing that skill ‘competently’ or ‘not so competently’ was presented. All children
106
were asked if they were like the child illustrating the task competently or the child illustrating the task
107
performed not as competently. If the child selected the competent illustration, they were then asked if
7 Page 6 of 19
they were ‘really good’ at the skill or ‘pretty good’ at the skill 16. If the child selected the not so
109
competent illustration, they were then asked if they were ‘not too good’ at the skill or ‘sort of good’ at
110
the skill. Each item was scored on a one to four scale, with one indicating low competency to four
111
being highly competent. This process was repeated for all 12 skills to give a maximum total score of
112
24 for each the object control and locomotor skills 16. This instrument has good internal consistency
113
for object control (Test 1:α = 0.63, Test 2:α = 0.72), and locomotor skills (Test 1:α = 0.64, Test 2:α =
114
0.68) 16. This instrument has also been assessed for 7 day test re-test reliability with good results (0.78
115
[95% CI 0.60-0.89]) for object control and locomotor skill perceptions (0.82 [95% CI 0.63-0.92]) 16.
116
MVPA was objectively assessed for eight consecutive days using GT3X+ ActiGraph (Pensacola, FL)
117
accelerometers. Data from the vertical plane were utilised in this study. To be included in the analyses,
118
children needed to have worn the monitor for ≥8 hours a day for a minimum of 3 days 19. A 15s epoch
119
length was used. Children were shown in small groups how to wear accelerometers, and the
120
circumstances in which they could remove them (e.g. if engaged in water-based activities and when
121
sleeping). Children were instructed to wear the monitor on the right hip, attached using an elasticated
122
nylon belt. This monitor has been validated for use in this population and has acceptable reliability 20.
123
Accelerometer data were initially downloaded using ActiLife Software (version 6.0) and checked for
124
compliance. The accelerometer files were then processed using a customised macro in Microsoft Excel
125
2010. Non-wear time was indicated by periods of 20 minutes of consecutive zeros 21. The
126
accelerometer cut-points developed by Evenson and colleagues 22 were used to define moderate and
127
vigorous intensity physical activity (2296-4011 cpm for moderate, and ≥4012 cpm for vigorous
128
intensity physical activity) 22. These cut-points have been validated and have recently been found to
129
provide acceptable classifications of moderate and vigorous physical activity in children 23. Time spent
130
in moderate and vigorous intensity physical activity were summed to obtain time spent in MVPA.
131
The mean and standard deviation was calculated for all variables. Differences between boys’ and girls’
132
actual object control, actual locomotor, perceived object control, perceived locomotor and MVPA
133
were assessed using independent t-tests. TGMD-2 raw scores were standardised according to the
134
manual18 and used in the analyses. To examine associations between actual and perceived object
Ac ce p
te
d
M
an
us
cr
ip t
108
8 Page 7 of 19
control and locomotor skills and physical activity, regressions with MVPA as the outcome were
136
initially performed with all predictor variables (i.e. Actual Object Control, Actual Locomotor,
137
Perceived Object Control, and Perceived Locomotor) (Model 1). Model 2 also adjusted for age
138
(months), sex, accelerometer wear time and English speaking background or not. Interaction terms
139
between the respective actual or perceived skill factor and sex were added to assess sex differences in
140
the predictor-MVPA association. For any significant or close to significant sex interaction terms (p <
141
0.1) we reran the model separately by sex. Stata version 11 was used for all analyses and p<0.05 was
142
the set significance value.
143
Results
144
Only the 109 children with complete data were included in the analysis. These children (59 boys, 50
145
girls) aged between five and eight (M = 6.5 years, SD = 1.0) completed all assessments, or in the case
146
of accelerometry, met inclusion criteria. A total of 28% of children spoke English at home and 44% of
147
parents were not born in Australia. Three quarters of parents had a tertiary qualification (77%), 12%
148
had completed high school, 8% had completed technical/trade school certificate or apprenticeships and
149
3% had not finished high school.
te
d
M
an
us
cr
ip t
135
150
Table 1 reports children’s actual and perceived object control, locomotor competence and MVPA and
152
the observed sex differences for these variables. Boys had significantly higher actual and perceived
153
object control skill and MVPA compared to girls. There were no significant sex differences in
154
children’s actual or perceived locomotor skills.
155
INSERT TABLE 1.
156
Model 1 (the unadjusted model) found that none of the skill variables were significant predictors of
157
MVPA. Similarly, in Model 2 (after adjusting for age, sex, English speaking background and monitor
158
wear time), none of the skill variables were significant predictors of MVPA (Table 2). However the
159
interaction term for locomotor skill (standardised) was significant (p = 0.029). The adjusted model was
160
rerun separately by sex and showed that for girls, locomotor skill was a significant predictor of MVPA
Ac ce p
151
9 Page 8 of 19
(B = 3.66, p = 0.016). For boys, there was no association between actual locomotor skill and MVPA
162
(B = -1.74, p = 0.245). See Table 3.
163
INSERT TABLE 2 and TABLE 3.
164
Discussion
165
The purpose of this study was to explore the relationship between children’s actual skill competence
166
and MVPA, and children’s perceived skill competence and MVPA, and to determine if any observed
167
associations differed by sex. This study only observed one significant association, which was for
168
actual locomotor competence and MVPA in girls. This finding can be interpreted as a 1.21 increase in
169
physical activity minutes with each successful locomotor skill component, indicating that a girl with
170
one more correct component on each of the six skills would engage in seven more minutes of activity
171
per day.
172
The current study contrasts previous research in young children that have found locomotor skill to be
173
associated with physical activity for both boys and girls. Williams and colleagues 25 found children in
174
the highest tertile of locomotor skill performance spent significantly more time in MVPA, whilst
175
Robinson and colleagues 26 found actual locomotor competence was moderately associated with
176
pedometer steps. Similarly, Iiovnen et al. and Foweather et al. also found locomotor skills were
177
associated with physical activity in young children 10,11. It is not clear why we found this relationship
178
only for girls, not boys. Locomotor skill competence may be more important than object control skill
179
competence for girls of this age as they may be engaging in types of physical activity (e.g. dance,
180
gymnastics) that do not require object control mastery. This is reinforced by Hardy 7 and LeGear 8
181
who found young girls had better locomotor skills than boys. Cliff and colleagues 9, using the TGMD-
182
2 and using accelerometers, also found no association between locomotor skill and physical activity
183
for boys (although it was approaching significance). An association was found between locomotor
184
skill and MVPA for girls, but it was an inverse association 9. The study by Cliff and colleagues
185
stratified by sex in a small sample (<50), which might have produced this unusual result for girls. In
186
another paper Cliff suggests that children may have to sustain a higher movement skill competency
Ac ce p
te
d
M
an
us
cr
ip t
161
10 Page 9 of 19
level for some time before this translates into MVPA 28. It is apparent that the relationship between
188
locomotor skill type and physical activity behaviour is not clear in this age group and further research
189
is warranted.
190
This study did not find an association between actual object control skill and MVPA. One study in
191
very similar aged Australian children using the same instruments and the same physical activity cut-
192
points also found object control skill was not associated with MVPA 17. Similarly, one study in
193
younger children (3 and 4 years) found no relationship between object control and MVPA when sex
194
and other potential confounders were adjusted for 25. Another study with very active 8 year old
195
children (mean of 93 minutes MVPA) also found that object control skill was not associated was
196
physical activity 27. In our study, object control skill initially approached significance with MVPA in
197
the unadjusted model (p = 0.056). Similarly, the previous Australian study found object control skill
198
approached significance with physical activity17, so it may mean that this relationship is only just
199
emerging in this age group. Yet, other studies, all in pre-schoolers 9, 10, 11, 24, have reported a positive
200
association between object control skill ability and MVPA, which suggests this relationship can
201
emerge at a young age in certain samples. Whilst one of these studies used the TGMD-2 24, the other
202
studies all used different instruments to assess motor skill 9, 10, 11, which makes comparisons difficult. It
203
is apparent that the relationship between skill type and physical activity behaviour is not clear and is
204
influenced by age, sex and instrumentation differences. As Stodden and colleagues 5 advocate, the role
205
that movement skill plays in MVPA might not be straightforward, relationships may change by age.
206
There was no association found in this study between perceived skill competence (object control or
207
locomotor) and physical activity. Only one other study could be located that has explored skill type in
208
terms of competence perceptions and their association with physical activity. Barnett and colleagues 17
209
also used the TGMD-2, but unlike other studies, assessed perceived competence using the same
210
pictorial instrument as the current study, which matches skill perceptions to the TGMD-2 skills.
211
Similar to the current study, they found perceived object control competence was not associated with
212
MVPA, after adjusting for sex and age. Also similar to the current study, that study found sex
213
differences in perceived object control skill (with boys higher) which indicates children do have some
Ac ce p
te
d
M
an
us
cr
ip t
187
11 Page 10 of 19
ability to rate themselves as these differences were also present in the actual skill data. Nevertheless
215
children generally had high perceptions of their ability and this lack of variability may have reduced
216
the ability to detect an association with MVPA. No other study could be located that has specifically
217
explored perceived locomotor competence as a predictor of MVPA.
218
Other studies in young children which have used more general perception assessments have shown
219
associations between young children’s perceived physical competence and physical activity 8, 12,
220
suggesting that perceptions could potentially have an influence on actual physical activity behaviour.
221
Although, another study in a similar age group did not find perceived physical competence to be
222
associated with pedometer steps 26. The fact that some studies have found associations in young
223
children suggests that developing general perceptions of physical competence which are broader than
224
just FMS competence (but may need to include FMS competence), is important because previous
225
evidence indicates that this predicts PA 3.
226
There is a limit on the generalizability of this study as it is a cross-sectional design with a relatively
227
small convenience sample. The strengths of the study include the matching of perceived movement
228
skills to a commonly used assessment of actual movement skill ability enabling the field to be
229
progressed by understanding the contribution of specific types of skill perceptions to physical activity,
230
and the use of GTX3+ accelerometers to objectively assess the sporadic nature of children’s physical
231
activity. However, hip-mounted accelerometers cannot capture some object control skills (such as
232
striking a ball or catching) since the motion and movement required to complete those skills are in the
233
arms and upper body. Consequently, accelerometers may register a lack of movement or, at the most,
234
light physical activity. This has potential implications on levels of MVPA recorded.
235
Conclusion
236
Actual rather than perceived skill competence was more important to MVPA in this sample. This
237
reinforces the conceptual model posited by Stodden et al. that actual competence precedes perceived
238
competence as a predictor of PA 5. Although, this study found that only locomotor competence was
239
weakly associated with MVPA, and only for girls. Adequately mastering these skills during childhood
Ac ce p
te
d
M
an
us
cr
ip t
214
12 Page 11 of 19
is important as many sporting activities require locomotor based competence. Therefore, childhood is
241
a prime age at which to encourage skill development through community-based and school-based
242
physical activity interventions as a strategy to promote long-term activity. The recent Morgan and
243
colleagues’ systematic review on the effectiveness of skill interventions found a large effect for both
244
gross motor skill and locomotor skill which provides support for such interventions 29. This study
245
found no associations found between perceived skill competence (object control or locomotor) and
246
MVPA. Developing general perceptions of physical competence is important because previous
247
evidence indicates that this predicts MVPA, especially in older children 3 where competence
248
perceptions become more in line with actual ability. Therefore, perceived physical competence should
249
still be considered during the development of physical education programs within this age group in
250
order to promote confidence and enjoyment of activities undertaken. Although some experimental
251
research has been unable to demonstrate causal associations between FMS, perceived physical
252
competence and objectively measured physical activity in children 28, further evidence is needed to
253
understand causal associations between these correlates and physical activity across different age
254
groups.
255
Practical Implications
257 258 259
cr
us
an
M
d te
Encourage school and community based FMS interventions in order to improve FMS in boys
Ac ce p
256
ip t
240
and girls.
Further research evidence is needed before recommending that perceived movement skill competence be targeted to promote MVPA in children.
260
While children's skill perceptions were not associated with physical activity, activities which
261
build skill and promote positive self-perception are warranted as these relationships are
262
purported to strengthen as children age.
263
13 Page 12 of 19
263
Acknowledgements
265
Acknowledgements provided in cover letter as not anonymous.
Ac ce p
te
d
M
an
us
cr
ip t
264
14 Page 13 of 19
266 267
References
268
1
269
pitfalls, and prospects. The Lancet. 2012; 380(9838): 247-257.
270
2
271
physically active and others not? The Lancet. 2012; 380: 258.
272
3
273
Youth: Systematic Review and Meta-Analysis. Sports Med. 2014. In press doi: 10.1007/s40279-014-
274
0229-z.
275
4
276
Adolescents. Sports Med. 2010; 40(12): 1019-1035.
277
5
278
of Motor Skill Competence in Physical Activity: An Emergent Relationship. Quest. 2008; 60(1): 290-
279
306.
280
6
281
children and adolescents: A systematic review. Psychol Sport Exerc. 2014; 15(4): 382-391.
282
7
283
children. J Sci Med Sport. 2009; 13(5): 503-508.
284
8
285
perceptions of competence of children in kindergarten. Int J Behav Nutr Phys Act. 2012; 9: 29.
286
9
287
objectively measured physical activity in pre-school children. Pediatr Exerc Sci. 2009; 21(4): 436-
288
439.
289
10
290
physical activity in 4-year-old preschool children. Percept Mot Skills 2013; 117(2):627-646.
ip t
Hallal PC, Andersen LB, Bull FC, et al. Global physical activity levels: surveillance progress,
cr
Bauman AE, Reis RS, Sallis JF, et al. Correlates of physical activity: Why are some people
us
Babic MJ, Morgan PJ, Plotnikoff RC, et al. Physical Activity and Physical Self-Concept in
an
Lubans DR, Morgan PJ, Cliff DP, et al. Fundamental Movement Skills in Children and
d
M
Stodden DF, Goodway JD, Langendorfer JS, et al. A Developmental Perspective on the Role
te
Holfelder B, Schott N. Relationship of fundamental movement skills and physical activity in
Ac ce p
Hardy LL, King L, Farrell L, et al. Fundamental movement skills among Australian preschool
LeGear M, Greyling L, Sloan E, Bell RI, et al. A window of opportunity? Motor skills and
Cliff DP, Okely AD, Smith L, et al. Relationship between fundamental movement skills and
Iiovnen KS, Sääkslahti K, Mehtälä A et al. Relationship between fundamental motor skills and
15 Page 14 of 19
291
11
292
and weekday and weekend physical activity in preschool children. J Sci Med Sport 2014; DOI:
293
10.1016/j.jsams.2014.09.014
294
12
295
motor skills in preschool children. Child Care Hlth Dev. 2011; 37(4): 589-596.
296
13
297
obese children. Obesity 2008; 16(12): 2634-41.
298
14
299
children. Child Dev. 1984; 55: 1970-1982.
300
15
301
competence, physical fitness and self-perception in children. Child Care Hlth Dev. 2012; 38(3): 394-
302
402.
303
16
304
for assessing fundamental movement skill perceived competence in young children. J Sci Med Sport.
305
2015; 18: 98.
306
17
307
actual ball skill competence and physical activity. J Sci Med Sport. 2015; 18: 167
308
18
309
2000.
310
19
311
physical activity. Arch Dis Child. 2007; 92(11): 963-969.
312
20
313
(CSA) activity monitor in children. Med Sci Sports Exerc. 1998; 30(4): 629-633.
314
21
315
Sports Exerc. 2005; 37: S523-S30.
316
22
317
activity for children. J Sports Sci. 2008; 26(14): 1557-65.
Foweather L, Knowles Z, Ridgers N et al. Relationship between fundamental movement skills
ip t
Robinson LE. The relationship between perceived physical competence and fundamental
cr
Morgan P, Okely A, Cliff D et al. Correlates of objectively measured physical activity in
us
Harter S, Pike R. The pictorial scale of perceived competence and social acceptance for young
M
an
Vedul-Kjelsas V, Sigmundsson H, Stensdotter AK, et al. The relationship between motor
d
Barnett LM, Ridgers ND, Zask A, et al. Face validity and reliability of a pictorial instrument
Ac ce p
te
Barnett LM, Ridgers ND, Salmon J. Associations between young children's perceived and
Ulrich DA. The Test of Gross Motor Development. 2nd edn. edn. Austin, TX, USA.: Pro-ed
Riddoch CJ, Mattocks C, Deere K, et al. Objective measurement of levels and patterns of
Trost SG, Ward DS, Moorhead SM, et al. Validity of the computer science and applications
Freedson P, Pober D, Janz KF. Calibration of Accelerometer Output for Children. Med Sci
Evenson KR, Catellier DJ, Gill K, et al. Calibration of two objective measures of physical
16 Page 15 of 19
318
23
319
activity intensity in youth. Med Sci Sports Exerc. 2011; 43(7): 1360-8.
320
24
321
children's motor skill proficiency. J Sci Med Sport. 2013; 16(4): 332-6.
322
25
323
preschool children. Obesity. 2008; 16(6): 1421-1426.
324
26
325
Preschool Students. Res Q Exerc Sport. 2012; 83: 20-26.
326
27
327
What about physical activity and weight status?. J PA & Health. 2015. In press.
328
28
329
Randomized Controlled Trial. Med Sci Sports Exerc. 2011; 43: 90-100.
330
29
331
movement skill interventions in youth: A systematic review and meta-analysis. Pediatrics. 2013; 132:
332
e1361.
Barnett L, Hinkley T, Okely AD, et al. Child, family and environmental correlates of
ip t
Williams HG, Pfeiffer KA, O'Neill JR, et al. Motor skill performance and physical activity in
us
cr
Robinson LE, Wadsworth DD, Peoples CM. Correlates of School-Day Physical Activity in
an
Barnett LM, Zask A, Rose L et al. Three year follow-up of an early childhood intervention:
M
Cliff D, Okely A, Morgan P et al. Movement Skills and Physical Activity in Obese Children:
te
d
Morgan PJ, Barnett LM, Cliff DP, Okely AD, Scott HA, Cohen KE, et al. Fundamental
Ac ce p
333
Trost SG, Loprinzi PD, Moore R, et al. Comparison of accelerometer cut points for predicting
17 Page 16 of 19
333
Table 1: Children’s actual and perceived object control, locomotor competence and MVPA mins/day,
334
and sex differences
Range
Boys
Girls
Sex differencesb
(n =109)
(n = 59)
(n = 50)
p-value
ip t
Variables
Total
Mean (SD) Mean (SD) Mean (SD) 9 - 45
29.5 (8.1)
32.1 (8.3)
Actual locomotor (raw score)
16 - 44
31.2 (5.6)
30.4 (5.4)
Perceived object control
13 - 24
20.0 (3.1)
20.8 (2.6)
19.0 (3.4)
p = 0.002
Perceived locomotor
10 - 24
20.6 (2.8)
20.4 (2.5)
20.8 (3.1)
p = 0.526
MVPA (mins/day)a
28 - 127
67.4 (21.2)
57.0 (17.7)
p < 0.001
336
(b) Independent t-tests examined sex differences
p = 0.134
d te Ac ce p
337
32.0 (5.8)
us
an
(a) Assessed by accelerometry
76.1 (20.0)
p < 0.001
M
335
26.4 (6.7)
cr
Actual object control (raw score)
18 Page 17 of 19
Table 2: Regression of MVPA (mins/day) including all predictors: non-adjusted and adjusted Models Predictors
Standard
Lower
Upper
error
95%
of B 1.80
Lower
Upper
95%
error
95%
95%
CI
CI
of B
CI
0.93
-0.05
3.64
0.056 1.35
1.31
0.71
-0.11
2.72
0.070
0.01
1.06
-2.12
2.10
p
B
0.85
control
Ac ce p
(Std) Perceived
locomotor
3.03
0.113
0.07
0.68
-1.41
1.27
0.914
0.991 0.49
1.01
-1.51
2.50
0.625
0.53
0.70
-1.92
0.87
0.455
te
locomotor
d
control Actual
-0.33
M
object
an
(Std) Perceived
CI
us
object
p
ip t
Standard
B
Actual
Adjusted model1
Non-adjusted model
cr
337
0.94
0.75
-2.43
0.55
0.213
338
1
339
monitor wear time statistically significant at the 5% level
340
Std = Standardised Score
Age, Sex, English speaking, monitor wear time. Sex was statistically significant at the 1% level and
341 342 343
19 Page 18 of 19
Table 3: Regression of MVPA (mins/day) for Boys and Girls including all predictors and adjusted Predictors
Boys
Actual
Standard
Lower
Upper
Standard
Lower
Upper
error
95%
95%
error
95%
95%
of B
CI
CI
of B
CI
CI
1.82
1.11
-0.41
4.05
0.108
1.35
-2.77
2.68
0.972
0.29
1.09
-1.91
2.48
0.794
1.74
1.48
-4.70
1.23
1.52
1.09
p
B
0.05
us
object control
an
(Std) Perceived
0.71
0.85
-2.43
1.01
0.409
0.245 3.66
1.46
0.72
6.61
0.016*
0.92
-1.21
2.50
0.488
control
(Std)
-3.70
0.66
0.168 0.64
Ac ce p
Perceived
te
d
locomotor
M
object
Actual
p
ip t
B
Girls
cr
343
locomotor 344
Both models adjusted for age, English speaking and monitor wear time
345
Monitor wear time was significant for boys at the 5% level
346
Std = Standardised Score
347 348
20 Page 19 of 19