14698749, 2012, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1469-8749.2011.04177.x by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [19/11/2022].

See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE

Reliability and responsiveness of the Movement Assessment Battery

for Children–Second Edition Test in children with developmental
coordination disorder
YEE-PAY WUANG 1 , 2 | JUI-HSING SU 1 ,3 | CHWEN-YNG SU 1 , 2

1 Department of Occupational Therapy, Kaohsiung Medical University, Kaohsiung. 2 Department of Physical and Rehabilitation Medicine, Kaohsiung Medical University Chung-Ho
Memorial Hospital, Kaohsiung. 3 Department of Physical and Rehabilitation Medicine, Kaohsiung Municipal Hsiaokang Hospital, Kaohsiung, Taiwan.
Correspondence to Dr Chwen-Yng Su, Department of Occupational Therapy, Kaohsiung Medical University, 100, Shi-Chuan 1st Rd., Kaohsiung City 807, Taiwan. E-mail: cysu@cc.kmu.edu.tw

This article is commented on by Sugden on pages 101–102 of this issue.

PUBLICATION DATA AIM To examine the internal consistency, test–retest reliability, and responsiveness of the
Accepted for publication 19th August 2011. Movement Assessment Battery for Children–Second Edition (MABC-2) Test for children with
developmental coordination disorder (DCD).
ABBREVIATIONS METHOD One hundred and forty-four Taiwanese children with DCD aged 6 to 12 years (87 males,
DCD Developmental coordination disorder 57 females) were tested on three separate occasions: two baseline measurements with a 20-day
MABC-2 Movement Assessment Battery for interval before the intervention, and a follow-up measurement after 6 months of rehabilitation.
Children–Second Edition The therapists rated the performance of children in school-related physical tasks at baseline and
MDC Minimal detectable change after intervention.
MID Minimal important difference
RESULTS Internal consistency for the MABC-2 Test was a = 0.90. Test–retest reliability for the total
PTPS Physical tasks performance scale
ROC Receiver operating characteristic score was excellent, with an intraclass correlation coefficient of 0.97. A small to medium magni-
SRM Standardized response mean tude of treatment effect was captured by the MABC-2 Test. The minimal detectable change (MDC)
was 0.28 points whereas the minimal important difference (MID) values were from 2.36 to 2.50. All
subscales except balance showed acceptable validity in differentiating groups of children whose
physical performance had improved or remained stable.
INTERPRETATION The MABC-2 Test is a reliable and valid measure to assess motor competence
in children with DCD. The MID and MDC scores provide the reference point for clinical decision-
making in managing the individual child.

Developmental coordination disorder (DCD) is one of the
most commonly reported disorders in school-age children,
with prevalence rates of up to 6%.1 A most recent UK popula-
tion-based study found that 1.8% of 7-year-olds have DCD
according to strict Diagnostic and Statistical Manual of Mental
Disorders, 4th edition (DSM-IV) criteria.2 These children
form a heterogeneous group, differing in the extent of their
motor difficulties and the extent to which they display other
conditions such as attention deficits, speech ⁄ articulation diffi-
culties, and non-verbal learning disabilities.3,4 Several studies
have shown that children with DCD are at greater risk of devel-
oping secondary impairments (e.g. decreased strength and lack
of fitness) owing to their limited participation in physical activ-
ity.5,6 Furthermore, in the absence of intervention, many of the
these abnormalities persist into adulthood and might continue
to interfere with performance in various aspects of community
functioning. As a result, early identification and therapeutic
intervention is particularly important to enhance motor func-
tion and promote success in school and daily living in children
with DCD. To monitor the effectiveness of an intervention, it
is crucial to use reliable and sensitive measures that are able to
yield consistent results across repeated measurements and
detect subtle changes in motor function.
The Movement Assessment Battery for Children (MABC)7
is one of the most widely used measures for identifying impair-
ments in motor performance of children and has been trans-
lated into several major European and Asian languages, such
as Italian, Chinese, and Japanese.8–10 The second edition
(MABC-2)11 is an updated version of the original MABC and
differs from the first in several respects, including age exten-
sion, reduction of age bands from four to three, item revision
and addition of new items, installation of an innovative score
interpretation method, inclusion of a more representative
standardization sample, and rearrangement of subtests. The
MABC-2 Test has shown good to excellent reliability in
healthy children,11,12 but there are as yet no data in this area
for children with DCD. For validity, several investigators have
studied the effectiveness of different treatment strategies for
children with DCD using the MABC.13,14 Nevertheless, in
none of these studies was an attempt made to determine the
ability of the MABC to detect small but important clinical
changes in motor performance over time (responsiveness).

160 DOI: 10.1111/j.1469-8749.2011.04177.x ª The Authors. Developmental Medicine & Child Neurology ª 2012 Mac Keith Press

These are changes that have been induced by treatments of
known efficacy, are associated with changes in another out-
come measure of similar constructs, or that clients and clini-
cians perceive as important.15 Knowledge of an instrument’s
responsiveness aids the interpretation of score changes after
treatment and selection of an appropriate outcome measure.
To address such shortcomings in previous reliability and
validity studies, the purpose of the present work was to evalu-
ate the MABC-2 Test’s internal consistency, test–retest reli-
ability, and responsiveness to change in Taiwanese children
with DCD.
METHOD
Participants
Children with DCD were recruited from 13 hospitals, seven
clinics, and 14 elementary schools in southern Taiwan. Inclu-
sion criteria for participation in the study were (1) age between
6 and 12 years, (2) a diagnosis of DCD according to DSM-IV
criteria,16 (3) absence of serious emotional or behavioural dis-
turbances, and (4) participation in physical or occupational
therapy programmes at the time of research. Children who had
previous history of neurological disorders such as traumatic
brain injury, muscular dystrophy, and epilepsy were excluded.
A total of 216 children met the inclusion and exclusion criteria.
Of these, 72 (33.3%) children dropped out of the study for var-
ious reasons. In the final sample (n=144), 40% (n=57) were
female, and the mean age was 7 years 7 months (SD 2y 1mo,
range 6y–12y 9mo). The number of children in each age group
was as follows: ages 6 years to 6 years 11 months (n=22);
7 years to 7 years 11 months (n=22); 8 years to 8 years
11 months (n=21); 9 years to 9 years 11 months (n=21);
10 years to 10 years 11 months (n=20); 11 years to 11 years
11 months (n=19); and 12 years to 12 years 11 months (n=19).
Instruments
MABC-2 Test
The MABC-2 Test is designed to identify and describe
impairments in motor performance of children and adoles-
cents from 3 to 16 years of age divided into three age bands
(3–6y, 7–10y, 11–16y). The test is composed of two parts: the
Performance Test and the Checklist. Only the Performance
Test was used in the current study; it involves children com-
pleting a series of eight fine and gross motor tasks grouped
into three subscales: Manual Dexterity, Aiming and Catching,
and Balance. The MABC-2 Test yields several scores: raw
score and standard score on the individual test items, subscale
standard score with equivalent centiles, and a total standard
score and centile equivalent. The total impairment score is
translated to age-related centile norms with the fifth centile as
the cut-off point to indicate children with movement difficul-
ties. Standard scores have a mean of 10 and an SD of 3.
School Function Assessment–Chinese version
The School Function Assessment–Chinese version is designed
to measure a student’s performance of functional tasks that
support his or her participation in both academic and social
school-related activities for students in grades kindergarten to
14698749, 2012, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1469-8749.2011.04177.x by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [19/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
What this paper adds
• It assesses the reliability of the Movement Assessment Battery for Children–
Second Edition (MABC-2) Test in developmental coordination disorder (DCD).
• It evaluates the responsiveness of the MABC-2 Test in DCD.
• It performs cross-cultural validation of the MABC-2 Test to determine its clini-
cal use in children with DCD in Taiwan.
sixth grade.17 It is a judgment-based criterion-referenced
assessment that comprises three sections: participation, task
supports, and activity performance. There are two major cate-
gories in the activity performance section: physical tasks and
cognitive ⁄ behavioural tasks. The physical tasks performance
scale (PTPS) was used as an external criterion for clinically rel-
evant change in this study. The PTPS consists of 161 items
(activities) divided into 12 domains and is rated on a scale of 1
to 4, where 1=does not perform, 2=partial performance,
3=inconsistent performance, and 4=consistent performance.
Reliability and discriminant validity of the Chinese version of
the School Function Assessment were similar to those of the
original English version.18,19
Procedure
This study was approved by the Institutional Review Board at
the Kaohsiung Medical University Chung-Ho Memorial Hos-
pital. After obtaining informed consent from all parents, the
principal investigator (Y-PW) of the study invited physical
and occupational therapists (n=15) who treated these children
before the study to participate in the intervention stage. All
children received a conventional paediatric rehabilitation pro-
gramme, at least 1 day a week, for 6 months.
Children were assessed with the MABC-2 Test on three
different occasions. Two baseline measurements (T1 and T2)
were performed with a 20-day interval in between. No inter-
ventions were undertaken between these two time points. A
third follow-up measurement (T3) was conducted immedi-
ately after completing the 6-month rehabilitation programme.
All three assessments were performed by the principal investi-
gator, a certified occupational therapist with 15 years’ clinical
experience in paediatric rehabilitation, using standardized
procedures for administration specified in the test manual.
Assessments were made at the same time of day for each visit.
To decrease possible experimenter bias, the examiner did not
re-acquaint herself with the child’s scores from the first assess-
ment when conducting the retests. Testing lasted approxi-
mately 30 minutes and was performed on an individual basis
in quiet locations identified at each child’s respective school,
or paediatric occupational therapy unit at hospitals. In addi-
tion, seven therapists who did not participate in the interven-
tion stage were recruited to complete the PTPS
independently at T1 and T3. The therapists were provided
with written and oral instructions for filling out the PTPS and
had high levels of agreement (intraclass correlation coefficient
[ICC]=0.95–0.99).
Statistical analysis
All data were analysed using SPSS for Windows (SPSS, Chi-
cago, IL, USA). Standard scores of the MABC-2 Test were
used in the following analyses.
Reliability and Responsiveness of the MABC-2 Test Yee-Pay Wuang et al. 161

Distribution
The score distribution of the MABC-2 Test was examined for
floor and ceiling effects. Floor or ceiling effects exceeding
20% of the participants were considered substantial.20
Internal consistency
Internal consistency indicates the extent to which items are all
measuring the same construct. Cronbach’s alpha (a) coefficient
was calculated for internal consistency of the MABC-2 Test
using data on all children receiving initial assessment. Values
of Cronbach’s a above 0.7 are generally regarded as accept-
able, over 0.8 good, and over 0.9 excellent.21
Test–retest reliability
The stability of the MABC-2 Test between the first and sec-
ond measurements was assessed using the ICC with a two-way
random effects model that allowed for the results to be gener-
alized to testing conditions beyond the one in this study. In
accordance with Cicchetti,22 ICCs are considered excellent if
>0.74, good from 0.60 to 0.74, fair from 0.40 to 0.59, and poor
if less than 0.40. The standard error of measurement (SEM)
was also calculated to give an indication of the precision of
individual scores over repeated administrations. The SEM was
calculated by the SD at baseline multiplied by the square root
of one minus its reliability coefficient, where scale reliability
was estimated using ICC. SEM£SD ⁄ 2 was taken as the crite-
rion of acceptable precision.23 The higher the reliability, the
lower the SEM and the more precise the scale.
Responsiveness
There appears to be little consensus over how responsiveness
should be assessed. To obtain a comprehensive picture of the
responsiveness of the MABC-2 Test, both internal and exter-
nal responsiveness were tested.24 Internal responsiveness
describes the ability of a measure to change during a defined
period of treatment. In the present study, the statistical signifi-
cance of the change was determined using variants on Cohen’s
effect size25,26 (standardizing changes in scores over time with
a standard deviation) as well as estimation of the minimum
detectable change (MDC) score. The MDC can be used to
interpret whether the observed change in the construct of
interest reflects a true change. When a change score surpasses
the MDC threshold, there is a reasonable level of confidence
that this change is real and beyond measurement error.
External responsiveness reflects the extent to which changes
in a measure correspond to changes in a reference measure. It
leads to the determination of the minimal important difference
(MID), which is the smallest change in score that clients or cli-
nicians perceive as important.27 There are multiple methods
for estimating the MID. In the current study, two anchor-
based approaches were used to calculate the MID scores of
the MABC-2 Test measures. The first one compared groups
of children who were different in terms of therapist perception
of progress using a four-point Likert scale (anchor). The dif-
ference in mean values of the MABC-2 Test across groups
was then used to ascertain the MID. The second approach
used the receiver operating characteristic (ROC) to evaluate
162 Developmental Medicine & Child Neurology 2012, 54: 160–165
14698749, 2012, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1469-8749.2011.04177.x by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [19/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
the correspondence between the chosen anchor and the
change scores from the MABC-2 Test. Each of the above-
mentioned methods was further specified as follows.
Effect size
The effect size is a standardized measure of change obtained
by dividing the mean change between initial and the third fol-
low-up measurements by the SD of the initial measurement.
As a guide to interpreting these values, Cohen25 labelled an
effect size ‘small’ if the effect size is between 0.2 and 0.5, ‘mod-
erate’ if the effect size is between 0.5 and 0.8, or ‘large’ if the
effect size is at least 0.8.
Standardized response mean (SRM)
The SRM is calculated as the mean change in scores between
initial and the third follow-up measurements divided by the
SD of that change score. A positive SRM indicates improve-
ment whereas a negative SRM indicates deterioration.
MDC
The MDC was computed as 1.65 · 2 · SEM. A z-score of
1.65 was chosen to reflect an acceptable 90% confidence inter-
val for clinical application to an individual child.
MID
The therapist-assessed PTPS score was used as an external cri-
terion (also known as anchor) to classify children as having
experienced an ‘important improvement’ or ‘no change’. The
‘improved’ group was defined as those with change score on
the PTPS of 1 or more, whereas the ‘no change’ group was
defined as those with a change score of less than 1 and up to 0.
The MID values for the MABC-2 Test were calculated by sub-
tracting the mean change score of all patients classified as hav-
ing no change from the mean change score of all patients who
were classified as experiencing an ‘important improvement’.
To establish the validity of the anchor, a one-way multivariate
analysis of variance was performed to test the difference
between groups across MABC-2 Test change scores.
ROC curves
ROC curves were used to examine various cut-off values for
the MABC-2 Test change scores for their sensitivity and spec-
ificity in correctly distinguishing children who were rated as
improved from those who were not. The point closest to the
upper left corner of the ROC curve is assumed to represent
the optimal trade off between sensitivity and specificity for
detecting clinical improvement. The area under the curve rep-
resents the probability that an instrument correctly discrimi-
nates between ‘important improvement’ and ‘no change’.
MABC-2 Test discrimination capability was considered insuf-
ficient if the area under the curve was less than 0.6, acceptable
when between 0.6 and 0.8, and excellent if above 0.8.28
RESULTS
Descriptive analysis of the data
All 144 children completed two baseline measurements
(T1 and T2), and five participants dropped out at the

6-month follow-up (T3). Descriptive statistics for the
MABC-2 Test scores obtained during the initial visit are
summarized in Table I. Mean scores for items ranged
from 4.75 for drawing trail to 8.97 for static balance.
The mean Manual Dexterity subscale score was lower
than those of other subscales.
Reliability
Internal consistency of the MABC-2 Test total score was
excellent (Cronbach’s a=0.90). The Cronbach’s a values for
Manual Dexterity, Aiming and Catching, and Balance sub-
scales were 0.81, 0.84, and 0.88 respectively. This implies a
sufficient homogeneity of all the individual domains as well as
the total test. In the test–retest analysis (Table II), ICCs varied
between 0.88 and 0.99 for the items and the subscales, whereas
the ICC of the total score was 0.97, indicating very high reli-
ability. SEM values all met the criterion (SEM£SD ⁄ 2), sug-
gesting an acceptable measurement precision of the MABC-2
Test.
14698749, 2012, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1469-8749.2011.04177.x by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [19/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Responsiveness
Table III presents the responsiveness indices calculated for
the MABC-2 Test. For the total score, the effect size and
SRM values reflected small to moderate effects (0.42, 0.63,
respectively) between visit 1 and the 6-month follow-up.
Effect size values ranged from 0.11 to 0.67 for the items and
from 0.37 to 0.63 for the subscales; the corresponding SRM
values ranged from 0.17 to 0.80 for the items and from 0.41
to 0.68 for the subscales. The highest effect size and SRM
values occurred with the Aiming and Catching 2 (throwing)
item and the Manual Dexterity subscale, whereas the lowest
values occurred with the Balance 1 (static) item and the
Balance subscale.
The MDC at the 90% confidence interval for the MABC-2
Test total score was 1.21, implying that in 90% of the cases, a
change of 1 point or more is likely to represent a true change
in overall motor function measured by the MABC-2 Test.
Among the 144 children, 41% and 59% were categorized as
‘improved’ and ‘no change,’ respectively, according to their

Table I: Description of the Movement Assessment Battery for Children–

Second Edition Test (MABC-2) standard scores at the initial visit (n = 144) Table III: Responsiveness statistics for the Movement Assessment Bat-
tery for Children–Second Edition (MABC-2) Test (n = 139)
MABC-2 parameters Mean (SD)

Items MABC-2 Difference scorea Effect ROCe

Manual Dexterity 1 6.1 (2.1) parameters mean (SD) range size SRMb MDCc MIDd cut-off
Manual Dexterity 2 7.7 (2.1)
Manual Dexterity 3 4.8 (1.8) Items
Aiming and Catching 1 6.3 (2.2) Manual 1.1 (1.4) )2 to 4 0.52 0.78 0.96 0.81 1.50
Aiming and Catching 2 8.2 (3.1) Dexterity 1
Balance 1 9.0 (3.5) Manual 0.8 (1.1) )2 to 4 0.37 0.42 0.70 0.89 1.50
Balance 2 6.4 (2.6) Dexterity 2
Balance 3 8.0 (3.1) Manual 1.0 (2.0) )2 to 8 0.55 0.49 1.19 1.51 1.50
Subscales Dexterity 3
Manual Dexterity 6.8 (1.8) Aiming and 1.4 (1.9) )1 to 7 0.65 0.75 1.72 1.90 2.50
Aiming and Catching 8.2 (3.1) Catching 1
Balance 8.0 (3.1) Aiming and 2.1 (2.6) )1 to 11 0.67 0.80 1.42 0.19 0.50
Total 8.0 (3.0) Catching 2
Balance 1 0.4 (1.0) )2 to 5 0.11 0.17 0.82 0.49 0.50
Balance 2 0.9 (1.3) 0 to 5 0.35 0.46 1.03 0.79 1.50
Balance 3 1.2 (1.9) )1 to 7 0.40 0.63 1.44 2.24 2.50

Subscales
Table II: Test–retest reliability of the Movement Assessment Battery for Manual 1.4 (1.7) )1 to 5 0.63 0.68 0.72 0.76 1.50
Dexterity
Children–Second Edition (MABC-2) Test (n = 144) Aiming and 1.5 (2.8) )4 to 7 0.49 0.54 2.14 1.78 2.50
Catching
MABC-2 ICC (95% CI) SEM Balance 1.1 (2.8) )5 to 9 0.37 0.41 1.26 0.52 0.50
Total score 1.3 (2.1) )1 to 7 0.42 0.63 1.21 2.36 2.50
Items
a
Manual Dexterity 1 0.96 (0.94–0.97) 0.41 Difference scores were calculated by subtracting the first baseline
Manual Dexterity 2 0.98 (0.97–0.99) 0.30 score from the third retest score (i.e. T3–T1). A positive score means
Manual Dexterity 3 0.92 (0.89–0.95) 0.51 that performance increased at retest, a negative score means that
Aiming and Catching 1 0.88 (0.83–0.92) 0.74 performance decreased at retest. The numbers under ‘range’ represent
Aiming and Catching 2 0.96 (0.95–0.98) 0.61 the lowest and highest difference score.
b
Balance 1 0.99 (0.98–0.99) 0.35 Standardized response mean (SRM) is calculated as the average
Balance 2 0.97 (0.95–0.98) 0.44 change in scores between initial and the third follow-up
Balance 3 0.96 (0.95–0.97) 0.62 measurements divided by the SD of that change score.
c
Subscales Minimal detectable change (MDC) was calculated as 1.65 · 2 · SEM,
Manual Dexterity 0.97 (0.95–0.98) 0.31 at a 10% level of significance (90% confidence interval).
d
Aiming and Catching 0.91 (0.82–0.95) 0.92 Minimal important difference (MID) was defined as the difference in
Balance 0.97 (0.95–0.98) 0.54 mean change of the MABC-2 Test scores between children classified
Total score 0.97 (0.96–0.98) 0.52 as ‘improved’ and those classified as ‘no change’.
e
The Receiver operating characteristic (ROC) curves were used to
ICC, intraclass correlation coefficient; SEM, standard error of examine various cut-off values for the MABC-2 Test change scores for
measurement. their sensitivity and specificity in detecting clinical improvement.

Reliability and Responsiveness of the MABC-2 Test Yee-Pay Wuang et al. 163

PTPS scores. None of the children had a change in PTPS
score of less than 0. The MID for the MABC-2 Test total
score was identified to be 2.36 points; that is, a change of 2
points or more is likely to represent a therapist-perceived
important change in the MABC-2 Test (Table III). A multi-
variate analysis of variance applied to the mean change scores
for the three subscales revealed significant differences between
the two groups (Wilks’ k = 0.84, F4,96 = 6.123, p < 0.01). Sub-
sequent analyses of variance on the individual composite
scores were performed. Significant univariate differences were
found in the Manual Dexterity (F = 211.235, p < 0.001), Aim-
ing and Catching (F = 322.005, p < 0.001), and Balance
(F = 184.382, p = 0.003) subscales, with the ‘improved’ group
scoring higher than the ‘no change’ group.
The MID calculated from the ROC curve using the cut-off
point nearest the upper left-hand corner of the graph was 2.5
points for the MABC-2 Test total score (sensitivity 71%, spec-
ificity 46%), 1.5 points for the Manual Dexterity subscale (sen-
sitivity 63%, specificity 49%), 2.5 points for the Aiming and
Catching subscale (sensitivity 76%, sensitivity 49%), 0.5
points for the Balance subscale (sensitivity 60%, specificity
44%). The values of the area under the curve for the changes
in total score, Manual Dexterity, Aiming and Catching, and
Balance subscales were 0.80, 0.64, 0.68, and 0.50 respectively.
DISCUSSION
This study examined, for the first time, the reliability and the
responsiveness of the MABC-2 Test in 6 to 12-year-old
Taiwanese children with DCD and showed that this test is a
reliable and valid outcome measure, one that can be used in
clinical practice to monitor treatment progress. The MABC-2
Test demonstrated good to excellent internal consistency
(Cronbach’s a range 0.81–0.90). The total instrument, as well
as its separate items and subscales, exhibited excellent test–
retest reliability. These ICC values were higher than those
reported for healthy adolescents.29 Apart from these, because
the reliability coefficients alone do not provide sufficient infor-
mation about the precision of test scores of individual child,
this study provides additional psychometric data about the var-
iability of the errors of measurement based on the SEM. The
SEM for the total MABC-2 Test was 0.52, for the items ran-
ged from 0.30 to 0.74, and for the subscales ranged from 0.31
to 0.92. These values were smaller than those reported in
healthy children (range 1.20–1.56),11 lending further support
to the stability of the MABC-2 Test scores over repeated test-
ing in children with DCD.
The values of effect size and SRM for the MABC-2 Test
corresponded to small to moderate effect sizes except for the
first item (Balance 1) of the Balance subscale, which yielded
negligible effect. Indeed, the Balance subscale had the lowest
14698749, 2012, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1469-8749.2011.04177.x by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [19/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
effect of all subscales. One possible reason might be that bal-
ance skills measured by this subscale were somewhat simple
for children with DCD because over half of them (58%)
scored between the mean and 1SD above the mean at baseline.
This leaves little room for improvement. Effect size scores
were found to be slightly lower than SRM scores, suggesting
that the SD of the initial assessment scores (used in calculating
the effect size) were more heterogeneous than the SD of
change scores (used in calculating the SRM).
Given that average effects across a group might not be
meaningful to the individual child, reference values are
required to determine if changes in the child’s scores before
and after the intervention reach statistical significance. The
MDC is such a reference, which can guide clinicians in mak-
ing the decision to modify the child’s therapy programme
accordingly. In other words, this information allows the clini-
cians to differentiate between true change and change due to
measurement error in score. The MDC at the 90% confidence
interval ranged from 0.72 to 2.14 points for the total and sub-
scales of the MABC-2 Test, and from 0.70 to 1.72 points for
the items. In contrast, the MID serves as another threshold
that determines the importance of changes in outcome scores
from the clinician’s viewpoint. Our results indicated that the
values of the MID calculated from the difference in mean
change scores between the improved and stable groups were
in the neighbouring range to those from the ROC curves,
ranging from 0.5 to 2.5 points. These MID values are benefi-
cial in terms of aiding the clinical decision-making process
because the anchor (performance on school-related motor
tasks) was measured reliably and was relevant to the motor
skills tapped by the MABC-2 Test. However, care must be
taken in using the MID score of the Balance subscale gener-
ated from the ROC statistic when interpreting a child’s perfor-
mance, as this subscale had poor discriminating ability.
The major limitation of the present study is that generaliza-
tion of the findings is limited to Taiwanese children with
DCD aged 6 to 12 years. Further studies including children
aged 3 to 5 years and 13 to 16 years are warranted to explore
fully the use of the MABC-2 for younger children and adoles-
cents with DCD. In addition, our findings should be
replicated in Western countries as studies have found cross-
cultural differences in some M-ABC items when comparing
performance between preschool children from Asian countries
and those from the USA.12
In conclusion, the MABC-2 Test can produce reliable
results in repeated assessments over a 20-day interval and is
responsive to change in children with DCD enrolled in reha-
bilitation programmes. The MID and MDC scores serve as
guidelines for clinicians in determining intervention effective-
ness from either clinical judgment or statistical significance.

REFERENCES
1. Wann J. Current approaches to intervention in children with
developmental coordination disorder. Dev Med Child Neurol
2007; 49: 405.
2. Lingam R, Hunt L, Golding J, Jongmans M, Emond
A. Prevalence of developmental coordination disor-
der using the DSM-IV at 7 years of age: a UK
population-based study. Pediatrics 2009; 123: e693–
700.
3. Chen YW, Tseng MH, Hu FC, Cermak SA. Psychosocial
adjustment and attention in children with developmental
coordination disorder using different motor tests. Res Dev
Disabil 2009; 30: 1367–77.
4. Tsai CL, Pan CY, Cherng RJ, Hsu YW, Chiu HH. Mecha-
nisms of deficit in visuospatial attention deficit in children
with developmental coordination disorder: a neurophysiolog-

164 Developmental Medicine & Child Neurology 2012, 54: 160–165


ical measure of the endogenous Posner paradigm. Brain Cogn
2009; 71: 247–58.
5. Mandich AD, Polatajko HJ, Rodger S. Rites of passage:
understanding participation of children with developmental
coordination disorder. Hum Mov Sci 2003; 22: 583–95.
6. Missiuna C, Rivard L, Bartlett D. Early identification and
risk management of children with developmental coordina-
tion disorder. Pediatr Phys Ther 2003; 15: 32–8.
7. Henderson SE, Sugden DA. Movement Assessment Battery
for Children. London: Psychological Corporation, 1992.
8. Livesey D, Coleman R, Piek J. Performance on the
Movement Assessment Battery for Children by Australian
3 to 5-year-old children. Child Care Health Dev 2007; 33:
713–19.
9. Chow SMK, Henderson SE. Interrater and test–retest reli-
ability of the Movement Assessment Battery for Chinese pre-
school children. Am J Occup Ther 2003; 57: 574–7.
10. Missiuna C, Rivard L, Bartlett D. Exploring assessment tools
and the target of intervention for children with developmen-
tal coordination disorder. Phys Occup Ther Pediatr 2006; 26:
71–89.
11. Henderson SE, Sugden DA, Barnett AL. Movement Assess-
ment Battery for Children, 2nd edn. London: Pearson, 2007.
12. Chow SMK, Hsu Y, Henderson S, Barnett A, Lo S. The
Movement ABC: a cross-cultural comparison of preschool
children from Hong Kong, Taiwan and the USA. Adapt Phys
Activ Q 2006; 23: 31–48.
13. Watemberg N, Waiserberg N, Zuk L. Developmental coor-
dination disorder in children with attention-deficit–hyperac-
tivity disorder and physical therapy intervention. Dev Med
Child Neurol 2007; 49: 920–5.
14. Sugden DA, Chambers ME. Intervention in children with
developmental coordination disorder: the role of parents and
teachers. Brit J Educ Psychol 2003; 73: 545–61.
15. Deyo RA, Diehr P, Patrick DL. Reproducibility and respon-
siveness of health status measures. Statistics and strategies for
evaluation. Control Clin Trials 1991; 12:(Suppl. 4) 142S–
58S.
16. American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders, 4th edition. Washington DC:
American Psychiatric Association, 1994.
17. Huang JL. School Function Assessment–Chinese Version.
Taipei: The Psychological Corporation, 2008.
18. Coster W, Deeney T, Haltiwanger J, Haley S. School Func-
tion Assessment. San Antonio, TX: Therapy Skill Builders,
1998.
19. Huang JL. The reliability and validity of the School Function
Assessment-Chinese version for cross-cultural use in Taiwan.
Occu Ther Int 2005; 11: 26–39.
20. McCarthy ML, Silberstein CE, Atkins EA, Harryman SE,
Sponseller PD, Hadley-Miller NA. Comparing reliability
and validity of pediatric instruments for measuring health
and well-being of children with spastic cerebral palsy. Dev
Med Child Neurol 2002; 44: 468–76.
21. Vangeneugden T, Laenen A, Geys H, Renard D, Mole-
nberghs G. Applying concepts of generalizability theory on
clinical trial data to investigate sources of variation and their
impact on reliability. Biometrics 2005; 61: 295–304.
Reliability and Responsiveness of the MABC-2 Test Yee-Pay Wuang et al. 165
14698749, 2012, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1469-8749.2011.04177.x by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [19/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
22. Cicchetti DV. Guidelines, criteria, and rules of thumb for
evaluating normed and standardized assessment instruments
in psychology. Psychol Assess 1994; 6: 284–90.
23. Wyrwich KW, Nienaber MA, Tierney WM, Wolinsky FD.
Linking clinical relevance and statistical significance in evalu-
ating intraindividual changes in health-related quality of life.
Med Care 1999; 37: 469–78.
24. Husted JA, Cook RJ, Farewell VT, Gladman DD. Methods
for assessing responsiveness: a critical review and recommen-
dations. J Clin Epidemiol 2000; 53: 459–68.
25. Cohen J. Statistical Power for the Behavioral Sciences. New
York: Academic Press, 1977.
26. Middel B, Van Sonderen E. Statistical significant change ver-
sus relevant or important change in (quasi) experimental
design: some conceptual and methodological problems in
estimating magnitude of intervention-related change in
health services research. Int J Integr Care 2002; 2: e15.
27. Revicki DA, Cella D, Hays RD, Sloan JA, Lenderking WR,
Aaronson NK. Responsiveness and minimal important differ-
ences for patient reported outcomes. Health Qual Life Out-
comes 2006; 4: 70.
28. Afessa B, Tefferi A, Dunn WF, Litzow MR, Peters SG.
Intensive care unit support and Acute Physiology and
Chronic Health Evaluation III performance in hematopoietic
stem cell transplant recipients. Crit Care Med 2003; 31:
1715–21.
29. Chow SMK, Chan LL, Chan CPS, Lau CHY. Reliability of
the experimental version of the Movement ABC. Br J Ther
Rehabil 2002; 9: 404–7.