This article was downloaded by: [Temple University Libraries]

On: 22 April 2015, At: 07:59

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Sports Sciences

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/rjsp20

Test of gross motor development-2 for Filipino children

with intellectual disability: validity and reliability
ab cd e
Catherine M. Capio , Kathlynne F. Eguia & Johan Simons
a
Institute of Human Performance, The University of Hong Kong, Hong Kong, Hong Kong
b
College of Allied Medical Professions, University of the Philippines, Manila, Philippines
c
Therabilities Inc., Pasig City, Philippines
d
Communicare Life Skills Development, Paranaque City, Philippines
e
Faculty of Kinesiology and Rehabilitation Sciences, University of Leuven (KU Leuven),
Leuven, Belgium
Published online: 17 Apr 2015.
Click for updates

To cite this article: Catherine M. Capio, Kathlynne F. Eguia & Johan Simons (2015): Test of gross motor development-2
for Filipino children with intellectual disability: validity and reliability, Journal of Sports Sciences, DOI:
10.1080/02640414.2015.1033643

To link to this article: http://dx.doi.org/10.1080/02640414.2015.1033643

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Sports Sciences, 2015
http://dx.doi.org/10.1080/02640414.2015.1033643

Test of gross motor development-2 for Filipino children with

intellectual disability: validity and reliability

CATHERINE M. CAPIO1,2, KATHLYNNE F. EGUIA3,4 & JOHAN SIMONS5

1
Institute of Human Performance, The University of Hong Kong, Hong Kong, Hong Kong, 2College of Allied Medical
Professions, University of the Philippines, Manila, Philippines, 3Therabilities Inc., Pasig City, Philippines, 4Communicare Life
Skills Development, Paranaque City, Philippines and 5Faculty of Kinesiology and Rehabilitation Sciences, University of
Leuven (KU Leuven), Leuven, Belgium

(Accepted 21 March 2015)

Downloaded by [Temple University Libraries] at 07:59 22 April 2015

Abstract
This study aimed to examine aspects of validity and reliability of the Test of Gross Motor Development-2 (TGMD-2) in
Filipino children with intellectual disability. Content and construct validity were verified, as well as inter-rater and intra-
rater reliability. Two paediatric physiotherapists tested 81 children with intellectual disability (mean age = 9.29 ± 2.71 years)
on locomotor and object control skills. Analysis of covariance, confirmatory factor analysis and analysis of variance were
used to test validity, while Cronbach’s alpha, intra-class correlation coefficients (ICC) and Bland–Altman plots were used to
examine reliability. Age was a significant predictor of locomotor and object control scores (P = 0.004). The data fit the
hypothesised two-factor model with fit indices as follows: χ2 = 33.525, DF = 34, P = 0.491, χ2/DF = 0.986. As hypothesised,
gender was a significant predictor for object control skills (P = 0.038). Participants’ mean scores were significantly below
mastery (locomotor, P < 0.001; object control, P < 0.001). Cronbach’s alpha was 0.830 for locomotor and 0.792 for object
control components. ICC for locomotor and object control scores ranged from 0.995 to 0.998, suggesting excellent intra-
rater and inter-rater reliability, confirmed by Bland–Altman analysis. This study provides evidence of sufficient content and
construct validity, internal consistency and rater reliability of TGMD-2 for Filipino children with intellectual disability.

Keywords: child, locomotor, object control, intellectual disability, developing country

Background Gross Motor Development – 2 (TGMD-2; Ulrich,

2000). It is a norm and criterion-referenced test that
Fundamental movement skills (FMS) constitute an
consists of six locomotor (run, slide, leap, gallop,
important set of skills that are believed to form the
jump, hop) and six object control skills (catch, over-
essential prerequisite foundations of more complex
hand throw, underhand roll, stationary bat, station-
movements (Gallahue, Ozmun, & Goodway, 2012;
ary dribble, kick). Besides being a measurement tool
Haywood & Getchell, 2005) that are often required
for research, TGMD-2 is used in clinical settings to
in different forms of physical activity. Consisting of
identify and address delay in FMS development
locomotor and object control skills, FMS make up
(Slater et al., 2010).
an important aspect of motor development and life-
Recent studies have shown that children with
long health patterns (Lubans, Morgan, Cliff,
intellectual disability represent one such group who
Barnett, & Okely, 2010). Hence, measurement of
display FMS delay (Simons, Daly, Theodorou,
FMS proficiency continues to be an important area
Caron, & Andoniadou, 2008; Westendorp,
of research and practice. Valid measurement scales
Houwen, Hartman, & Visscher, 2011). Children
for FMS contribute to understanding motor devel-
with intellectual disability are characterised to have
opment and functions, allow identification of possi-
impaired intellectual functioning and adaptive beha-
ble developmental delays (Gallahue et al., 2012) and
viours (Schalock et al., 2010). Intellectual function-
provide the basis of evaluating supportive movement
ing refers to the ability to learn, reason and solve
training programmes (Capio, Sit, & Abernethy,
problems, and is typically measured through an
2011). One of the most frequently used measure-
intelligence quotient (IQ) test (Schalock, 2011).
ment scales for FMS proficiency is the Test of

Correspondence: Catherine M. Capio, Institute of Human Performance, The University of Hong Kong, 111–113 Pokfulam Road, Hong Kong.
E-mail: ccapio@hku.hk

© 2015 Taylor & Francis

 2 C. M. Capio et al.
Adaptive behaviours consist of a range of practical,
conceptual and social skills such as daily personal
care, handling money or communication with peers
(Tasse et al., 2012). It had been suggested that
intellectual functioning of these children have asso-
ciated impairments in information processing
(Planinsec & Pisot, 2006), which may impact their
ability to learn motor skills such as FMS.
While TGMD-2 had been used in studies invol-
ving children with intellectual disability (Valentini,
2012; Westendorp et al., 2011), instrument develop-
ment had been focused on typically developing chil-
dren. Content validation based on the judgement of
three content experts, and conventional item analysis
had shown satisfactory item discrimination and item
difficulty criteria (Ulrich, 2000). Further evidence of
content validity was based on the component scores
Downloaded by [Temple University Libraries] at 07:59 22 April 2015
having a linear relationship with age, as FMS are
known to be developmental in nature (Gallahue
et al., 2012). TGMD-2 had also been shown to
have high internal consistency (α > 0.80). Because
normative scores had been established for North
American children aged 3–10 years old with typical
development, use in other regions (e.g. Brazil
(Valentini, 2012), Korea (Kim, Kim, Valentini, &
Clark, 2014)) and in groups of children with disabil-
ity (e.g. intellectual disability (Simons et al., 2008),
visual impairment (Houwen, Hartman, Jonker, &
Visscher, 2010)) had necessitated further examina-
tion of validity and reliability of TGMD-2.
Among children with intellectual disability,
aspects of validity and reliability of TGMD-2 had
been established in a Flemish sample (Simons et al.,
2008). However, evidence of an instrument’s validity
and reliability cannot be assumed to generalise (Yun
& Ulrich, 2002) across other groups of children with
intellectual disability, particularly those coming from
developing countries. Most aspects of developmental
skills are influenced in varying degrees by the oppor-
tunities that are available to children, as well as by
physical and cultural environments (Fernald,
Kariger, Engle, & Raikes, 2009). Because there are
environmental disparities associated with economic
development (Walker et al., 2007), FMS proficiency
of children from developing countries may occur
through a different trajectory and should be mea-
sured with valid and reliable instruments.
Low- and middle-income countries had been
found to have higher prevalence of intellectual dis-
ability, suggesting the need to establish programmes
that ameliorate the impact of associated impairments
(Maulik, Mascarenhas, Mathers, Dua, & Saxena,
2011). Supportive programmes require valid and
reliable measures of child development, including
the motor skills domain, to provide markers for set-
ting goals and monitoring progress (Engle et al.,
2007). Because child development measurements
in lower-income countries had been generally
focused on the cognitive domain (Grantham-
McGregor et al., 2007), lack of normative motor
development data limits population-appropriate eva-
luation of gains by children who undergo motor
training programmes.
The Philippines is one such developing country,
and recently published studies have used compo-
nents of TGMD-2 among typically developing
Filipino children (Capio, Sit, Eguia, & Abernethy,
2014) and those with intellectual disability (Eguia,
Capio, & Simons, in press). Eguia and colleagues (in
press) found that Filipino children with intellectual
disability had deficits in FMS proficiency that
appeared to be related to insufficient physical activ-
ity. Such finding suggests a need to provide pro-
grammes that would promote FMS development in
children with intellectual disability, the outcome
measures of which should be monitored to verify
the effectiveness of any intervention (Slater et al.,
2010). To evaluate outcomes, valid and reliable
measurement scales are important. In the study by
Eguia and colleagues (in press) the TGMD-2 com-
ponents had acceptable internal consistency.
However, there was no further evidence to support
the validity and reliability of TGMD-2 in Filipino
children with intellectual disability, and this study
aimed to address this research gap.
Content validity refers to the extent that the test
items indeed represent the domain that the test
claims to measure (Thomas & Nelson, 1990).
Consistent with TGMD-2 content validation in typi-
cally developing children, this study also verified
whether a linear relationship exists between age and
locomotor and object control scores, respectively, by
testing correlations between age and scores.
Construct validity, on the other hand, refers to the
extent to which the underlying traits of a test fit the
theoretical model on which the test is based
(Thomas & Nelson, 1990). In this study, the two-
factor model of TGMD-2 (i.e. locomotor, object
control) was tested through factor analysis. As pre-
vious studies had shown children with intellectual
disability to have FMS deficits (Westendorp et al.,
2011), scores of participants in this study were com-
pared to scores that represented mastery to further
verify construct validity. Finally, sex-specific differ-
ences in FMS were also examined because previous
research had shown boys with intellectual disability
generally perform better than girls in object control
skills but not in locomotor skills (Simons et al.,
2008).
One aspect of the reliability of an instrument is
demonstrated by internal consistency, which refers
to the degree to which test components measure the
same construct adequately (Thomas & Nelson,
1990). Internal consistency of TGMD-2 components

had been found to be sufficient in a smaller sample
of Filipino children with intellectual disability
(Cronbach’s alpha = 0.78 for locomotor, 0.81 for
object control), in a study that examined FMS asso-
ciation with physical activity levels (Eguia et al., in
press). However, reliability of TGMD-2 has not yet
been systematically established for this specific popu-
lation. In typically developing children, TGMD-2
aspects of reliability were recently established in an
Australian sample (Barnett, Minto, Lander, & Hardy,
2014). Given the characteristics associated with intel-
lectual disability and being in a developing country, it
is important to ascertain that FMS proficiency may be
measured reliably. This present study aimed to exam-
ine aspects of validity and reliability of TGMD-2
when used among Filipino children with intellectual
disability. Content and construct validity were veri-
Downloaded by [Temple University Libraries] at 07:59 22 April 2015
fied, as well as the consistency of test scores between
two different examiners (inter-rater reliability) or over
multiple tests by a single examiner (intra-rater
reliability).
Methods
Participants
A total of 100 children from three special schools in
two urban regions of the Philippines (Metro Manila
and Cebu) were tested. Participants were enrolled in
a special school, had impaired intellectual function-
ing based on standard cognitive measures and
impaired adaptive functions based on standard func-
tional measures. Those with diagnoses that have
concurrent physical impairments (i.e. Down’s
Syndrome, cerebral palsy) were excluded, bringing
the final study sample to 81 (65 boys, 16 girls).
There were more boys than girls in the sample, but
this was reflective of the gender distribution of the
schools’ enrollees. Participants were aged 5–14 years
(mean = 9.29 s 2.71 years) and their school levels
ranged from preschool to third grade of the national
basic education curriculum. While TGMD-2 was
specifically validated for typically developing chil-
dren aged 3–10 years old (Ulrich, 2000), participants
in this study included those who were aged above
this upper limit. Previous studies had shown FMS
deficits in children with intellectual disability
(Simons et al., 2008; Westendorp et al., 2011), and
in validating the instrument in this particular sample,
older participants may display FMS that are closer to
mastery levels.
The schools from which the participants were
recruited are privately run, and admit students
using similar procedures. Students are typically
referred to schools by other professionals (e.g. occu-
pational therapist, physiotherapist, developmental
paediatrician). Selected components of the
TGMD-2 in children with intellectual disability 3
Vineland Adaptive Behavior Scales-2nd edition
(VABS-2; Sparrow, Cicchetti, & Balla, 2005) are
used to assess cognitive, communication and social-
behavioural functions, which then serve as basis of
admission. IQ tests are neither required nor recom-
mended by the schools, to avoid additional costs
associated with standardised test administration.
Consequently, participants in this study could not
be categorised into severity of intellectual disability
based on IQ cut-points. Nevertheless, the school
administrators confirmed that participants in this
study are those categorised to have mild to moderate
intellectual disability based on selected test compo-
nents from VABS-2. The schools’ decision not to
use IQ test appears to be consistent with how the
construct of intellectual disability had been evolving,
such that an ecological focus has been adopted with
decreasing emphasis on IQ cut-points as a categori-
cal basis of classifying severity of disability (Schalock,
2011).
Procedures
All the study procedures were reviewed and
approved by the institutional review board of the
university, and were consistent with ethical princi-
ples of the Declaration of Helsinki. Parents or care-
givers returned informed consent, and each
participant gave verbal assent prior to testing.
Because TGMD-2 was developed for typically devel-
oping North American children, the cultural rele-
vance of the items was considered. Consistent with
guidelines for measuring child development in low-
income countries (Fernald et al., 2009), discussions
were first conducted with the physical education
teachers and therapists of the participants on the
applicability of the 12 component skills. The con-
sensus was that galloping and stationary striking
should be dropped in testing the current sample
because these did not appear to be frequently
observed in the local physical education context.
Consequently, out of the 12 components of
TGMD-2, 10 skills were tested. In other research,
selected components of TGMD-2 (i.e. not the full
test) had also been used to characterise FMS profi-
ciency of children with and without disabilities (e.g.
Capio, Sit, Eguia, Abernethy & Masters, 2014;
Hardy, King, Farrell, Macniven, & Howlett, 2010).
Two licensed paediatric physiotherapists, each
with >5 years of experience, conducted individual
tests of the participants. All tests were carried out
during school hours, in the school premises and in
the presence of at least one teacher, thereby ensuring
a sense of familiarity for the participants. The tester
provided verbal instructions for each skill, followed
by two rounds of demonstration. While standard
TGMD-2 procedures include one round of
 4 C. M. Capio et al.
demonstration, the current study procedures were
such so as to compensate for participants’ cognitive
impairments that likely limit verbal communication
ability (Schalock et al., 2010). Video recordings were
done at 24 frames per second (fps), to allow post hoc
video analysis. TGMD-2 assessed each skill based
on three to five performance criteria, the presence of
which was scored one and absence was scored zero.
Each participant performed one practice trial, fol-
lowed by two trials that were rated, yielding scores
that ranged from zero to six, zero to eight or zero to
ten per skill. Summation of the skill scores gave full
scores of 40 for locomotor and 38 for object control.
Data analysis
Video recordings of tests were analysed using video
Downloaded by [Temple University Libraries] at 07:59 22 April 2015
playback software that allowed frame-by-frame view-
ing. Each video was first viewed in real-time speed,
followed by a frame-by-frame view, in which
researchers verified the presence or absence of each
performance criterion. Two physiotherapists first
conducted independent analyses of a subsample of
10 participants to verify inter-rater reliability. Each
rater also performed a second analysis of the sub-
sample to assess intra-rater reliability. After complet-
ing the analysis that showed excellent reliability
(intra-class correlation coefficients (ICC) > 0.90;
see Results for details), video analysis of the remain-
ing sample was performed by one of the two raters.
Statistical analyses were performed using IBM SPSS
Statistics and Amos for Windows, version 20.
Normality of data was ascertained based on the
Shapiro–Wilk test, which is considered to be the
best choice for testing data normality (Ghasemi &
Zahediasl, 2012).
Validity. Because FMS acquisition is developmental
(Gallahue et al., 2012), TGMD-2 scores are
expected to be positively correlated with age
(Ulrich, 2000). The association of age with locomo-
tor and object control skills was examined using the
Pearson product–moment correlation coefficient and
multivariate analysis of covariance (ANCOVA) to
verify content validity.
The established two-factor model (i.e. locomotor,
object control) of TGMD-2 (Ulrich, 2000) was
empirically tested via confirmatory factor analysis
(CFA) using maximum likelihood estimation meth-
ods. The two-factor model fit was evaluated based
on the chi-square, chi-square/degrees of freedom
(df) ratio, goodness of fit index (GFI) and the root
mean square error of approximation (RMSEA). A
non-significant chi-square value supports validity of
the two-factor model based on the inference that
there is no reliable difference between the sample
data and the model (Joreskog & Sorbom, 1982).
The chi-square/df ratio should be close to 1.00 for
the model to be correct and any ratio >2.00 repre-
sents an inadequate fit (Byrne, 1989). A GFI value
of 1.00 indicates a perfect fit and ≥0.90 is generally
considered an indicator of a good model fit (Joreskog
& Sorbom, 1982). RMSEA value of ≤0.05 indicates
a good fit of the model and up to 0.08 for tolerable
amount of error (Browne & Cudeck, 1992).
Because gender differences in locomotor and
object control skills were expected, construct validity
was further examined by univariate analysis of var-
iance (ANOVA) with locomotor and object control
skills as dependent variables and gender as the fixed
factor. Finally, one-sample t-test was used to com-
pare participants’ locomotor and object control
aggregate scores to the full scores that represent skills
mastery because children with intellectual disability
were also expected to have deficits in FMS profi-
ciency. The percentage of participants who gained
full scores in both skills was also determined.
Reliability. Internal consistency of the locomotor and
object control components of TGMD-2 was exam-
ined using Cronbach’s alpha, inter-item correlations
and corrected item-total correlations. Cronbach’s
alpha, when each of the individual items (skill com-
ponent) was removed from the scale, was also calcu-
lated. Cronbach’s alpha of ≥0.70 and item-total
correlations of ≥0.30 were considered acceptable
(Nunally, 1978; Nunally & Bernstein, 1994). Inter-
item correlation of ≥0.80 was considered an indica-
tor of redundancy of items. ICC, including 95%
confidence intervals (CI), was calculated to examine
intra-rater and inter-rater reliability. ICC of ≥0.75 is
considered excellent, 0.40–0.74 is moderate and
<0.40 is poor (Fleiss, 1999). Bland–Altman analysis
was performed to verify the intra-rater and inter-
rater agreement (consolidated data of both raters),
by examining the level of agreement between tests
(intra-rater) and between raters (inter-rater). This
method plots the differences between two scores
against their mean, where ±1.96 standard deviation
of the differences provides an interval within which
95% of the differences between the two sets of scores
are predicted to fall (Bland & Altman, 1999).
Agreement is illustrated by a small range of the
95% CI of the differences between the measure-
ments (Myles & Cui, 2007).
Results
Participants’ scores are summarised in Table I.
Normality of data was confirmed using Shapiro–
Wilk tests, which showed non-significant P values
for locomotor (P = 0.054) and object control skills
(P = 0.601).

Table I. Participants’ locomotor, object control and total FMS
scores.
Boys (n = 65) Girls (n = 16)
Locomotora
Mean (s) 24.58 (9.23) 22.50 (9.37)
95% Confidence Interval 22.30–26.87 17.51–27.49
Object controlb
Mean 20.32 (6.65) 18.19 (7.33)
95% Confidence Interval 18.67–21.97 14.28–22.09
Totalc
Mean (s) 44.91 (14.79) 40.69 (15.92)
95% Confidence Interval 41.24–48.57 32.21–49.17
Notes: amastery score = 40; b
mastery score = 38; c
mastery
score = 78.
Validity
Downloaded by [Temple University Libraries] at 07:59 22 April 2015
Age was found to have significant positive correla-
tions with locomotor (r = 0.222, P = 0.047) and
object control skills (r = 0.356, P = 0.001).
Multivariate ANCOVA further showed a significant
effect of age for locomotor skills and object control
skills (F(2, 78) = 5.865, P = 0.004, η2 = 0.131).
It appears that the data fits the hypothesised two-
factor model (locomotor and object control) of
TGMD-2 with fit indices as follows: χ2 = 33.525,
DF = 34, P = 0.491, χ2/DF = 0.986. Absolute fit
indices that directly assess the model fit similarly
indicate sufficient values with GFI = 0.931.
Additionally, RMSEA = 0.000 with 90% CI of
0.000–0.080 suggested good model fit.
Gender was a significant predictor only for object
control skills (F(1, 78) = 4.446, P = 0.038,
η2 = 0.054) and not for locomotor skills (F(1,
78) = 1.831, P = 0.180, η2 = 0.023), such that
boys performed significantly better than girls only
in object control skills. Participants’ mean scores
were significantly lower than the perfect scores that
represent mastery in locomotor (t(80) = −15.421,
P < 0.001) and object control skills (t
(80) = −23.954, P < 0.001). None of the partici-
pants reached the full scores that represent mastery
in both locomotor and object control skills.
Reliability
Cronbach’s alpha for the locomotor components was
0.830 and a range of 0.757–0.814 when each item is
deleted. For object control components, Cronbach’s
alpha was 0.792 and a range of 0.713–0.757 when
each item is deleted. Item-total correlation coeffi-
cients for locomotor components ranged from
0.712 to 0.913, and for object control components
they ranged from 0.673 to 0.816. These findings
suggest acceptable internal consistency (Nunally,
1978; Nunally & Bernstein, 1994). Inter-item
TGMD-2 in children with intellectual disability 5
Table II. Intra-rater and inter-rater reliability indices.
ICC 95% Confidence Interval
Intra-rater
Locomotor 0.995 0.978–0.999
Object control 0.998 0.991–0.999
Total FMS 0.997 0.989–0.999
Inter-rater
Locomotor 0.996 0.984–0.999
Object control 0.998 0.992–1.000
Total FMS 0.998 0.991–0.999
correlations ranged from 0.309 to 0.676 for locomo-
tor components and from 0.253 to 0.577 for object
control components, indicating no redundancy of
items.
ICC for locomotor and object control test scores
ranged from 0.995 to 0.998, suggesting excellent
intra-tester and inter-tester reliability (Fleiss, 1999).
Reliability indices are summarised in Table II.
Bland–Altman analysis confirms intra-rater and
inter-rater agreement (see Figure 1).
Discussion
While FMS are believed to form the basis of more
advanced and complex motor skills (Gallahue et al.,
2012; Haywood & Getchell, 2005), the value of
FMS proficiency has been underscored by evidence
pointing to associated health benefits (Lubans et al.,
2010). Attempts to facilitate FMS development need
to be supported by valid and reliable measurement,
to allow monitoring and programme evaluation.
This study contributes to establishing such FMS
measurement, particularly for Filipino children with
intellectual disability. Currently, there exists paucity
in research to underpin efforts to support FMS
development of this population in developing coun-
tries such as the Philippines.
In this study, two skills – one each for locomotor
and object control components – were dropped as a
consequence of adaptation to the study sample. With
the 10 components, the FMS scores of the Filipino
sample in this study were found to have sufficient fit
with the two-factor model proposed by Ulrich
(2000), as evident in the reported indices. This sup-
ports the notion that these 10 components from
TGMD-2 constitute a valid set of tests to character-
ise FMS proficiency of Filipino children with intel-
lectual disability. The study by Simons and
colleagues (2008) among Flemish children showed
comparable fit indices when using all 12 skills of
TGMD-2. In comparison, the study by Houwen
and colleagues (2010) among children with visual
impairment found lower fit indices, and this was
attributed to constraints associated with impaired
visual input.
 6 C. M. Capio et al.

3.00 3.00

2.00 2.00

Difference (inter-tester)
Difference (intratester)
1.00 1.00

.00 .00

–1.00 –1.00

–2.00 –2.00

–3.00 –3.00

.00 10.00 20.00 30.00 40.00 .00 10.00 20.00 30.00 40.00
Mean (intratester) (a) Mean (inter-tester)

1.00 1.50

.50 1.00

Difference (inter-tester)
Difference (intratester)

.00 .50
Downloaded by [Temple University Libraries] at 07:59 22 April 2015

–.50 .00

–1.00 –.50

–1.50 –1.00

–2.00 –1.50
30.00 40.00
.00 10.00 20.00 .00 10.00 20.00 30.00 40.00
Mean (intratester) (b) Mean (inter-tester)

Figure 1. Bland–Altman plots showing inter-rater and intra-rater agreement in locomotor (a) and object control skills (b) scores.

In interpreting the findings of this study, one has As expected, the participants in this study did not
to consider limitations associated with the rela- reach FMS mastery, and motor proficiency (or lack
tively small sample size. Nevertheless, it is worth of it) is presumably affected by their cognitive
noting that while validation studies of TGMD-2 in impairments (Hartman, Houwen, Scherder, &
typically developing populations had employed Visscher, 2010). Similar results were found among
much bigger sample sizes (e.g. Valentini (2012), Flemish children with intellectual disability who
Kim et al. (2014)), populations of children with scored significantly lower than the normative data
disability are relatively limited. Hence, other stu- from North American typically developing children
dies that had validated the instrument in children (Simons et al., 2008). In typically developing chil-
with disability had also been limited to smaller dren, it has been suggested that lack of FMS mastery
sample sizes (e.g. Simons et al. (2008), Houwen points to the value of providing opportunities for
et al. (2010)). practice during free play as well as teacher-directed
Studies of FMS proficiency in children with and training activities (Hardy et al., 2010). In children
without disability had consistently found differences with intellectual disability, this needs to be empha-
between girls and boys, particularly in object control sised just as well, if not even more so on account of
skills (Simons et al., 2008). While we acknowledge the observed drawbacks that are likely to be asso-
that the subgroup of girls in this study sample was ciated with cognitive impairments.
limited in number, the study nevertheless shows Acceptable internal consistency was found for
similar patterns of gender-related differences, pro- both components in this study and is comparable
viding further evidence of construct validity. In typi- with those found by Simons et al. (2008).
cally developing children, it had been suggested that Moreover, inter-item correlations were sufficient to
gender differences in object control skills are likely support reliability, but not reaching correlation coef-
influenced by social factors that could be related to ficient values to suggest redundancy of items. All
interaction with parents, peers and teachers, as well tests and analyses were conducted by paediatric phy-
as play interactions that lead to sufficient skill prac- siotherapists, who had previous clinical experiences
tice (Hardy et al., 2010). It is likely that in a special in using TGMD-2. The findings of this study show
school environment, similar social factors are present that given trained testers, TGMD-2 can be used in
such that the participants in this study displayed the Philippines with excellent inter-rater and intra-
similar gender differences. rater reliability. This is further supported by the

Bland–Altman analysis that showed small discrepan-
cies in the scores between testers, and by a single
tester performing the test repeatedly.
While this study offers evidence to support aspects
of validity and reliability of TGMD-2 as an instru-
ment to examine the locomotor and object control
skills of Filipino children with intellectual disability,
future work is also recommended to explore other
skills that might be of cultural relevance. For
instance, one of the traditional games in the country
is called “sipa”, which literally translates as “kick”.
The pattern of kicking in this game is different from
that observed in TGMD-2. In “sipa”, the object that
is made up of metal and plastic parts has to be
directed vertically upward by hitting it with the pos-
terior-medial aspect of the foot (Reyes, 2010).
Hence, the movement would require a combination
Downloaded by [Temple University Libraries] at 07:59 22 April 2015
of hip flexion and external rotation, knee flexion, and
foot supination.
The use of pictorial instructions might also be
explored as a means to adapt testing procedures for
children with intellectual disability. Moreover, stu-
dies that would compare children with and without
disability could also lead to further understanding of
FMS development of Filipino children. Finally, it
must also be acknowledged that the sample con-
sisted of students in privately run special schools,
and the findings of this study may not be generalised
to those in the public school system whose access to
resources may be different.
Conclusion
This study aimed to examine aspects of validity and
reliability of 10 components of TGMD-2 in a sample
of Filipino children with intellectual disability, aged
5–14 years old. The findings support sufficient con-
tent and construct validity, internal consistency and
rater reliability (intra-rater, inter-rater). It had been
shown that there is a need to provide supportive
programmes that will develop the FMS proficiency
of Filipino children with intellectual disability. In
developing programmes to address this need,
TGMD-2 is deemed suitable as an instrument for
characterising the extent of FMS deficits, and for
programme monitoring and evaluation.
Disclosure statement
No potential conflict of interest was reported by the
author(s).
Funding
This study was supported in part by the Marden
Seed Funding grant of the Centre for Advancement
of Inclusive and Special Education, the University of
TGMD-2 in children with intellectual disability 7
Hong Kong. The second author was supported by
an Erasmus Mundus scholarship of the European
Commission at the University of Leuven, Belgium.
References
Barnett, L. M., Minto, C., Lander, N., & Hardy, L. L. (2014).
Interrater reliability assessment using the test of gross motor devel-
opment-2. Journal of Science and Medicine in Sport, 17(6), 667–670.
Bland, J. M., & Altman, D. G. (1999). Measuring agreement in
method comparison studies. Statistical Methods in Medical
Research, 8(2), 135–160.
Browne, M. W., & Cudeck, R. (1992). Alternative ways of asses-
sing model fit. Sociological Methods & Research, 21(2), 230–258.
Byrne, B. M. (1989). A primer of LISREL: Basic application and
programming for confirmatory factor analytic models. New York,
NY: Springer-Verlag.
Capio, C. M., Sit, C. H., & Abernethy, B. (2011). Fundamental
movement skills testing in children with cerebral palsy.
Disability & Rehabilitation, 33(25–26), 2519–2528.
Capio, C. M., Sit, C. H., Eguia, K. F., & Abernethy, B. (2014).
Physical activity and movement skills proficiency of young
Filipino children. Pediatrics International, 56(4), 651–653.
Capio, C. M., Sit, C. H. P., Eguia, K. F., Abernethy, B., Masters,
R. S. W. (2014). Movement skills training as a strategy to
promote physical activity in children with cerebral palsy.
Journal of Sport and Health Science. Advance online publication.
doi:10.1016/j.jshs.2014.05.001
Eguia, K. F., Capio, C. M., & Simons, J. (2015). Object control
skills predict the physical activity of children in special schools
in a developing country: Philippines. Journal of Intellectual and
Developmental Disability, 40(3).
Engle, P. L., Black, M. M., Behrman, J. R., De Mello, M. C.,
Gertler, P. J., Kapiriri, L., & Young, M. E. (2007). Child
development in developing countries 3: Strategies to avoid the
loss of developmental potential in more than 200 million chil-
dren in the developing world. The Lancet, 369(9557), 229–242.
Fernald, L. C., Kariger, P., Engle, P., & Raikes, A. (2009).
Examining early child development in low-income countries.
Washington, DC: The World Bank.
Fleiss, J. L. (1999). Reliability of measurement, in the design and analysis
of clinical experiments. Hoboken, NJ: John Wiley & Sons Inc.
Gallahue, D. L., Ozmun, J. C., & Goodway, J. (2012).
Understanding motor development: Infants, children, adolescents,
adults (7th ed.). New York, NY: McGraw-Hill.
Ghasemi, A., & Zahediasl, S. (2012). Normality tests for statistical
analysis: A guide for non-statisticians. International Journal of
Endocrinology and Metabolism, 10(2), 486–489.
Grantham-McGregor, S., Cheung, Y. B., Cueto, S., Glewwe, P.,
Richter, L., & Strupp, B., International Child Development
Steering Group. (2007). Developmental potential in the first
5 years for children in developing countries. The Lancet, 369
(9555), 60–70.
Hardy, L. L., King, L., Farrell, L., Macniven, R., & Howlett, S.
(2010). Fundamental movement skills among Australian preschool
children. Journal of Science and Medicine in Sport, 13(5), 503–508.
Hartman, E., Houwen, S., Scherder, E., & Visscher, C. (2010).
On the relationship between motor performance and executive
functioning in children with intellectual disabilities. Journal of
Intellectual Disability Research, 54, 468–477.
Haywood, K., & Getchell, N. (2005). Life span motor development
(4th ed.). Champaign, IL: Human Kinetics.
Houwen, S., Hartman, E., Jonker, L., & Visscher, C. (2010).
Reliability and validity of the TGMD-2 in primary-school-age
children with visual impairments. Adapted Physical Activity
Quarterly, 27(2), 143–159.
 8 C. M. Capio et al.
Joreskog, K. G., & Sorbom, D. (1982). Recent developments in
structural equation modeling. Journal of Marketing Research, 19
(4), 404–416.
Kim, S., Kim, M. J., Valentini, N. C., & Clark, J. E. (2014).
Validity and reliability of the TGMD-2 for South Korean chil-
dren. Journal of Motor Behavior, 46(5), 351–356.
Lubans, D. R., Morgan, P. J., Cliff, D. P., Barnett, L. M., &
Okely, A. D. (2010). Fundamental movement skills in children
and adolescents: Review of associated health benefits. Sports
Medicine, 40(12), 1019–1035.
Maulik, P. K., Mascarenhas, M. N., Mathers, C. D., Dua, T., &
Saxena, S. (2011). Prevalence of intellectual disability: A meta-
analysis of population-based studies. Research in Developmental
Disabilities, 32(2), 419–436.
Myles, P. S., & Cui, J. (2007). Using the Bland–Altman method
to measure agreement with repeated measures. British Journal
of Anaesthesia, 99(3), 309–311.
Nunally, J. C. (1978). Psychometric theory. New York, NY:
McGraw-Hill.
Nunally, J. C., & Bernstein, I. H. (1994). Psychometric theory. New
Downloaded by [Temple University Libraries] at 07:59 22 April 2015
York, NY: McGraw-Hill.
Planinsec, J., & Pisot, R. (2006). Motor coordination and intelli-
gence level in adolescents. Adolescence, 41(164), 667–676.
Reyes, C. A. D. (2010). From physical recreation to digitisation:
A social history of children’s games in the Philippines. In L.
Hopkins, W. C. Turgeon, & M. Macleod (Eds.), Negotiating
childhoods (pp. 99–109). Oxford: Interdisciplinary Press.
Schalock, R. L. (2011). The evolving understanding of the con-
struct of intellectual disability. Journal of Intellectual and
Developmental Disability, 36(4), 227–237.
Schalock, R. L., Borthwick-Duffy, S. A., Bradley, V. J., Buntinx,
W. H. E., Coulter, D. L., Craig, E. M., & Yeager, M. H.
(2010). Intellectual disability: Definition, classification, and systems
of supports (11th ed.). Washington, DC: American Association
of Intellectual and Developmental Disabilities.
Simons, J., Daly, D., Theodorou, F., Caron, C., & Andoniadou,
E. (2008). Validity and reliability of the TGMD-2 in 7–-
10-year-old Flemish children with intellectual disability.
Adapted Physical Activity Quarterly, 25(1), 71–82.
Slater, A., Bowen, J., Corsini, N., Gardner, C., Golley, R., &
Noakes, M. (2010). Understanding parent concerns about
children’s diet, activity and weight status: An important step
towards effective obesity prevention interventions. Public Health
Nutrition, 13(8), 1221–1228.
Sparrow, S. S., Cicchetti, D. V., & Balla, D. A. (2005). Vineland
adaptive behavior scales – II. Livonia, MN: Pearson
Assessments.
Tassé, M. J., Schalock, R. L., Balboni, G., Bersani Jr, H.,
Borthwick-Duffy, S. A., Spreat, S., & Zhang, D. (2012). The
construct of adaptive behavior: Its conceptualization, measure-
ment, and use in the field of intellectual disability. American
Journal on Intellectual and Developmental Disabilities, 117(4),
291–303.
Thomas, J., & Nelson, J. K. (1990). Research methods in physical
activity (2nd ed.). Champaign, IL: Human Kinetics.
Ulrich, D. (2000). Test of gross motor development (2nd ed.). Austin,
TX: Pro-Ed.
Valentini, N. C. (2012). Validity and reliability of the TGMD-2 for
Brazilian children. Journal of Motor Behavior, 44(4), 275–280.
Walker, S. P., Wachs, T. D., Gardner, J. M., Lozoff, B.,
Wasserman, G. A., Pollitt, E., & Carter, J. A., International
Child Development Steering Group. (2007). Child develop-
ment: Risk factors for adverse outcomes in developing coun-
tries. The Lancet, 369(9556), 145–157.
Westendorp, M., Houwen, S., Hartman, E., & Visscher, C.
(2011). Are gross motor skills and sports participation related
in children with intellectual disabilities? Research in
Developmental Disabilities, 32(3), 1147–1153.
Yun, J., & Ulrich, D. (2002). Estimating measurement validity: A
tutorial. Adapted Physical Activity Quarterly, 19(1), 32–47.