Perceptual and MotorSkills, 2001, 93,275-280.

O Perceptual and Motor Skills 2001

RELIABILITY AND CONCURRENT VALIDITY OF THE MOVEMENT

ASSESSMENT BATTERY FOR CHILDREN '
RONALD V. CROCE
Department of Kinesiology
Universily of New Hampshire
MICHAEL HORVAT, ERICK McCARTHY

Deportment of Physical Edricatio~zand Sporf Sludies

Universliy of Georgia
Summay.-Reliability and concurrent vahdity of the Movement Assessment Battery for Children were evaluated with a sample of 106 boys and girls, distributed into
four age groups: 5-6 yr. (rz = 20), 7-8 yr. (11 = 20), 9-10 yr. 0 2 =46), and 11-12 yr. (n =
20). Test-retest reliabihty of [he Movement Assessment Battery for Children, estimated using intradass correlation coefficients, was high across all age groups, and concurrent validity yielded moderate Pearson correlation coefficients benveen the Movement battery and long and short forms of the Bruininks-Oseretsky Test of Motor Proficiency. These results support the use of the Movement battery as a measure of motor ability in children, ages 5 to 12 years.

Several tests are available currently to assess the motor a b h y of children. Two such tests are the Bruininks-Oseretsky Test of Motor Proficiency
(Bruininks, 1978) and the Movement Assessment Battery for Children (Movement ABC) (Henderson & Sugden, 1992). The Movement ABC is a relatively new test that has evolved over the past 30 years for identhing children with subtle motor impairments (Stott, Moyes, & Henderson, 1972,
1984; Henderson & Sugden, 1992) and is cited as the most popular test outside of North America (Crawford, Wilson, & Dewey, 2001). The BruininksOseretsky test has remained relatively unchanged since its inception, is often
considered the "Gold Standard" of motor abhty tests and has often been
used as the standard for establishing the concurrent validity of newly developed procedures in motor assessment (Riggen, Ulrich, & Ozmun, 1990; Spiegel, Steffens, Rynders, & Bruininks, 1990). Although reliabhty and vahdity
of the Bruininks-Oseretsky test has been consistent (Beitel & Mead, 1980,
1982; Verderber & Payne, 1987; Wilson, Kaplan, Crawford, & Dewey, 2000;
Hassan, 2001), its norms are quite old and many investigators have questioned its u t h t y (Burton & Miller, 1998). One of the most impressive aspects of the Bruininks-Oseretsky test is its sustained popularity throughout

'Address enquiries to Ronald Croce, Ph.D., De artment of Kinesiology, University of New

Hampshire, Durham, NH 03824 or e-mail (rvc9cfrista.unh.edu).

276

R. V. CROCE, ET AL.

North America (Miles, Nierengarten, & Nearing, 1988; Crawford, et al.,

2001).
Comparatively, there is h i t e d research on the reliability and v&dity of
the Movement ABC as a research and diagnostic tool (Mon-Wfiarns, Pascal,
& Wann, 1994; Miyahara, Tsujii, Hanai, Jongmans, Barnett, Henderson,
Hori, Nakanishi, & Kageyama, 1998; Crawford, et al., 2001). Burton and
Miller (1998) stated that there has not been adequate study of either the reliabhty or validity of the Movement ABC but acknowledge that cltnicians
might find it useful for screening, planning, and intervention. Thus, further
study is warranted to ascertain whether the Movement ABC covers the entire domain of motor abdity as does the Bruininks-Oseretsky test, and whether reliabdq and v&dlty are consistent across age bands. Harnmill, Brown,
and Bryant (1992) argued that test evaluation is a constant, ongoing process
and that it is imperative to conduct evaluations of the psychometric properties of a test when revisions of the original version are conducted. The purpose of this study was, then, (1) to examine test-retest reliabhty of the
Movement ABC across age bands and (2) to estimate its concurrent val~d~ty
using long and short forms of the Bruininks-Oseretsky test.
METHOD

Subjects
One hundred six boys (n = 67) and girls (n =39) between the ages of 5
and 12 years participated subsequent to parental approval. Subjects were recruited from two elementary schools in New Hampshire and two elementary
schools in Georgia and were free from orthopedic and intellectual conditions that might h i t participation. AU methods and procedures were approved by the University of New Hampshire and the University of Georgia
Institutional Review Boards and by the individual school boards.
Procedure
Tests.-Each child was tested individually a total of three times, twice
with the Movement ABC (test-retest) and once with the Bruininks-Oseretsky
test (short-form scores were derived from long-form data). Order of administering the two tests was counterbalanced. Retests with the Movement ABC
were conducted within one week of the initial test. Children were tested in
an open area on a wooden floor while wearing tennis shoes and loose clothing. Tests were conducted and scored in accordance with instructions provided by the respective manuals. Testers were graduate students in adapted
physical education and individually trained by the investigators. Each tester
had two years experience in administering the tests.
Ana1ysir.-Based on the designated age bands in the Movement ABC,
subjects were divided into the following age groups: 5-6 yr. (n=20), 7-8 yr.

RELIABILITY AND VALIDITY OF MOVEMENT BATTERY

277

(n =20), 9-10 yr. (n =46), and 11-12 yr. (n =20). Because the number of
subjects was limited, and a two-factor repeated-measures analysis of variance
indicated no significant sex differences ( p > .Ol) on either set of test data,
boys and girls were combined for statistical analyses. Since placement decisions in motor remedlation programs are most often based on percentile
rankings rather than on raw scores, percentile scores were analyzed. In a d d tion, raw scores were excluded because on the short form of the BruininksOseretsky test, the raw score only represents
scores for 14 of
the 46 long-form items. If raw scores were used, short-form scores would
typically be lower than long-form scores, which would invalidate comparisons (Verderber & Payne, 1987). On the other hand, because percentile
scores are derived from standard scores, which have been equated mathematically, comparisons can be made. Use of percentile scores for correlational
analyses of motor proficiency tests can be found in the literature (Verderber
& Payne, 1987; Henderson & Sugden, 1992).
Test-retest reliabhty of the Movement ABC was estimated usmg an inrraclass correlation coefficient (intraclass r ) from a one-way analysls of variance. This statistic was the estimate of consistency of performance across trials using Baumgartner and Jackson's formula (1995). Concurrent validity of
the Movement ABC was estimated by calculating Pearson correlation coefficients between percentiles on the Movement ABC and the long and short
forms of the Bruininks-Oseretsky test. Data on Trial 1 on the Movement
ABC were used in all Pearson r calculations.
RESULTS
A N D DISCUSSION
Means and standard errors of percentile scores on the Movement ABC
and long and short forms of the Bruininks-Oseretsky test are shown in Table
1. Values of intraclass and Pearson correlation coefficients are shown in Table 2. Analyses indicated that there was a high trial-to-trial reliabhty for the
Movement ABC, and concurrent validity, using long and short forms of the
Bruininks-Oseretsky test as the criterion measures, was moderately h~ghand
statistically significant.
Based on the analysis, the Movement ABC emerges as a promising motor assessment device that can be used confidently by teachers, chicians,
and researchers. Test-retest reliability was high (range across one week from
.92 to .98), indicating that the Movement ABC provides stable values over a
1-wk. period. This is important for three reasons.
First, it suggests that the test will provide consistent measures on separate weekly occasions. Often two separate sessions are used by therapists
and educators to obtain a rehable assessment of a child's motor ability, with
the mean performance score being used for final considerations. This approach helps to account for the so-called "good" and "bad" day phenom-

278

R. V. CROCE, ET AL.

TABLE 1
MEANSA N D S T A N D A D
F ~m n o ~ FOR
s PERCENTILE
RANKINGS
OF MOVEMENT
ABC AND
BRUININKS-OSERETSKY
TESTO F MOTORPROFICIENCY:
LONGAND SHORTFORMSBY AGE GROUPS
Test

All Groups
M
SE

Movement ABC
Test
31.6
Retest
35.4
Bruininks-Oseretsky
Long Form
43.3
ShortForm
46.8

5 - 6 yr.

7-8 yr.

9-10 yr.

11-12 yr.

SE

SE

IM

SE

SE

2.7
2.8

23.1
27.2

5.9
6.3

24.3
31.2

5.2
5.9

37.8
39.1

4.2
4.2

33.1
68.9

6.5
7.2

3.0
3.1

32.8
38.4

6.3
6.2

42.4
49.5

7.4
7.1

51.5
53.1

4.7
4.8

36.0
38.2

7.0
6.7

enon that occurs when assessing many children with disabhties. Secondly,
high test-retest reliability supports the use of the Movement ABC examining
the effectiveness of motor intervention programs. Finally, the high intraclass
r value allows practitioners to be confident of the initial assessment of a
child so that the identification of motor deficiencies and subsequent intervention programming can occur in a timely manner. As evidence of the
reliabhty of their test, Henderson and Sugden (1992) reported a 73% agreement of assignment when test and retest performance scores were used to
assign indlv~dualsto groups above or below the 15th percentile cut-off.
TABLE 2
INTRACLASS AND PEARSONPRODUCT-MOMENT
CORRELATION
COEFFICIENTS
FORMOVEMENT
ABC A N D
BRUININKS-OSERETSKY
TEST O F MOTORPROFICIENCY:
LONGAND SHORTFORMSBY AGEGROUPS
Test

AH

5-6 yr.

7-8 yr.

.95t

.98t

.95+

.92t

.97t

.76t

.77+

.76t

.70t

.90t

.7 1i

,791

.76t

.60"

.90t

9-10 yr. 11-12 yr.

Groups
Intradass r
Movement ABC Test-Retest
Pearson Product-moment r
Movement ABC Test-BruininksOseretsky Long Form
Movement ABC Test-BruininksOseretskv Short Form

The second purpose of this study was to estimate the concurrent validity of the Movement ABC using both the long and short forms of the Bruininks-Oseretsky test as the criterion measures. Before dscussing these results,
we note that a caveat in comparing the Movement ABC against the Bruininks-Oseretsky test was put forth by Henderson and Sugden (1992). They
proposed that, while these two tests are similar in some respects, they are
radically different in others. The authors stated "Bruininks dsslgned his test
to measure abhcy across the whole range [motor ability], whereas the Movement ABC focuses primarily on the identification of impairment. For this

RELIABILITY AND VALIDITY O F MOVEMENT BATTERY

279

reason alone it can be argued that a very high correlation should not be expected'' (p. 206). Talung this into consideration, present results indicated
that concurrent validitv between the Movement ABC and the Bruininks-Oseretsky test was good (range of Pearson r values was from .60 to .90).
Based on the present results, we concluded that the Movement ABC
can be used with confidence by professionals who require a short and simple assessment of a child's motor behavior. We had no problem in administering items, and subjects did not find testing to be tedious, difficult, or
dscouraging. This was not the case with the Bruininks-Oseretsky test. Given
the longer time to administer the test, subjects often seemed preoccupied,
and motivation appeared to wane. Along with its brevity, two distinguishing
aspects of the Movement ABC are hkely to appeal to practitioners.
One is the provision of a movement checklist. The primary focus of the
checkhst is the assessment and management of motor problems in an educational setting. This allows the examiner to not only identlfy children who
may have motor difficulties but also provides information on the nature of
the motor difficulties exhibited by the child during testing (Wright, Sugden,
Ng, & Tan, 1994). The other is that the Movement ABC is comprehensive
for idendying motor impairments in children and provides a manual with
guidelines for linlung test results to the management of a child's unique
physical and motor needs.
REFERENCES
BAUMGARTNER,
T., &JACKSON,
A. (1995) Measurement /or evaluation irr physical edrrcatiort and
exernie science. (5th ed.) Dubuque, IA: Brown & Benchmark.
BEITEL,I? A,, &MEAD,
B. J. (1980) Bruininks-Oseretsky Test of Motor Proficiency: a viable
measure for 3- to 5-year-old children. Perceptual and Motor Skills, 51, 919-923.
BEITEL.
P. A,, &MEAD,B. J. (1982) Bruininks-Oseretsk Test of Motor Proficiency: further verMotor Skills, 54, 268-270.
ification wich 3 - to 5-year-old children. ~erce~lzraland
BRUININKS,
R. H. (1978) The Brzrininks-Oseretsky Test of Motor Proficiency: examiner's marttral.
Circle Pines, MN: American Guidance Service.
A. W., & MILLER,
D. E. (1998) Movement skill assessment. Champaign, IL: Human
BURTON,
Kinetics.
CRAWFORD,
S. G.. WILSON,
B. N., &DEWEY,
D. (2001) Idendying developmental coordination
disorder: consistency between tests. Physical & Occzrpa~iorzalTherapy in Pedialrics, 20, 2950.
HAMMILL,
D., BROWN,
L., & BRYANT.
B. (1992) A consumer's gzride to tests in print. (2nd ed.)
Austin, TX: PRO-ED.
HASSAN.
M. H. (2001) Validity and reliability for the Bruininks-Oseretsky Test of Motor Proficiency-Short Form as applied in the United Arab Emirates culture. Perceplual arzd Motor Skills, 92, 157-166.
HENDERSON,
S. E., & SUGDEN,
D. A. (1992) Movement Assessment Batley for Children. London: Psychological Corp.
MILES,B. H., NIERENGARTEN.
M. E.. &NV\RING,R. J. (1988) A review of the eleven most often-cited assessment instruments used in adapted physical education. Clinical Ki~zesiology,
42, 33-41.
MTYAHARA,
M., TSUJII,M., HANAI,T., JONGMANS.
M.. BARNETT,
A,, HENDERSON,
S., Horn, M.,
NAKANISHI,
K., &KAGEYAMA,
H. (1998) The Movement Assessment Battery for Children:

280

R. V. CROCE. ET AL.

a preliminary investigation of its usefulness in Japan. Hzrman Movement Science, 17, 679697.
MON-WILLLAICIS,
M. A,, PASCAL,
E., &WANN,1. l? (1994) Ophthalmic factors in developmental
coordination disorder. Adapted Physical Activity Quarterly, 11, 170-178.
RIGGEN,
K. J.. ULRICH,D. A,, &OZMUN.
1. C. (1990) Reliability and concurrent validity of the
Test of Motor Impairment-Henderson Revision. Adapted Physical Activity Qt~arterly,7,
249-258.
SPIEGEL,
A. N., STEFFENS,
K. M., RYNDERS,
J. E., & BRUININKS,
R. H. (1990) The Early Motor
Profile: correlation with the Bruininks-Oseretsky Test of Motor Proficiency. Perceptrial
and Motor Skills, 71, 645-646.
S ~ o r r D.
, H., Moues, F. A,, & HENDERSON,
S. E. (1972) Test of Motor Impairme~zt.Guelph,
Ontario: Brook Educational.
STOIT, D. H., MOYES.F. A., & HENDERSON,
S. E. (1984) Test of Motor Impairment-Henderson
Revision. San Antonio, T X : Psychological Corp.
V E ~ E R B E1.RM.
, S.. & PAYNE,
V. G. (1987) A comparison of the long and short forms of the
Bruininks-Oseretsky Test of Motor Proficiency. Adapted Phystcal Activity QzLarterly, 11,
170-178.
WILSON.B. N., KAPLAN.
B. J., CRAWFORD,
S. G., &DEWEY,
D. (2000) Interrater reliabhty of the
Bruininks-Oseretsky Test of Motor Proficiency-Long Form. Adapted Physical Activity
Quarterly, 17, 95-110.
WRIGHT,H. C., SUGDEN,
D A . NG, R., &TAN.J. (1994) Identification 06 children with movement problems in S~ngnpore:usefulness of the Movement ABC checklist. Adapted Physical Activity Qz,arterlv 1 1 . 150-157.

Accepted Azigtist 13, 2001