R E S E A R C H R E P O R T

Concurrent Validity of the Bayley

Scales of Infant Development II
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A

(BSID-II) Motor Scale and the Peabody

WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023

Developmental Motor Scale II

(PDMS-2) in 12-Month-Old Infants
Barbara H. Connolly, FAPTA, Lauri Dalton, MPT, Jennifer Bengston Smith, MPT, Nichole Grice Lamberth, MPT,
Brent McCay, MPT, and Will Murphy, MPT
University of Tennessee Health Services Center, Memphis, Tennessee

Purpose: This study was designed to examine concurrent validity of the Peabody Developmental Motor Scale II
(PDMS-2) and the Bayley Scales of Infant Development II (BSID-II) Motor Scale. Methods: The PDMS-2 and the BSID-II
Motor Scale were administered to fifteen 12-month-old infants who were developing typically. PDMS-2 raw scores
were converted to the Gross Motor Quotient (GMQ), the Fine Motor Quotient (FMQ), and the Total Motor
Quotient (TMQ). BSID-II raw scores were converted to the Psychomotor Development Index (PDI). Age equivalent
scores were obtained for all PDMS-2 Gross and Fine Motor Subscales and for the BSID-II Motor Scale and compared
using the Pearson Product-Moment Correlation Coefficient. Results: Low correlations were found between the
PDMS-2 FMQ, GMQ, TMQ, and the PDI of the BSID-II. Low correlations were found between age equivalent scores
of the PDMS-2 subtests for grasp, stationary, and Visual Motor Integration and the BSID-II Motor Scale. A low
negative correlation was found between age equivalent scores of the PMDS-2 subtest for Object Manipulation and
the BSID-II Motor Scale. A high correlation (r ⫽ 0.71, P ⬍ 0.05) was found between age equivalent scores of the
PDMS-2 subtest for locomotion and the BSID-II Motor Scale. Conclusions: There is a lack of concurrent validity
between the PDMS-2 standard scores and standard scores of the BSID-II Motor Scale and a lack of agreement
between age equivalent scores of the BSID-II Motor Scale and the PDMS-2 subtests except for Locomotion. The
investigators caution about using only one standard score or age equivalent score for decisions about the need for
early intervention for children at 12 months of age when using the BSID-II or the PDMS-2. (Pediatr Phys Ther
2006;18:190 –196) Key words: child development, comparative study, developmental disabilities/diagnosis, infant,
motor skills, neuropsychological tests/standards, psychometrics

INTRODUCTION tice. Using the patient/client management process as

The use of standardized norm-referenced and criterion-
referenced tests as a part of the examination process has be-
come an integral part of the developmental therapist’s prac-
0898-5669/106/1803-0190
Pediatric Physical Therapy
Copyright © 2006 Lippincott Williams & Wilkins and Section on Pedi-
atrics of the American Physical Therapy Association.
Address correspondence to: Barbara H. Connolly, FAPTA, Department
of Physical Therapy, University of Tennessee Health Sciences Center, 930
Madison Avenue, Memphis, TN 38163. E-mail: bconnolly@utmem.edu
DOI: 10.1097/01.pep.0000226746.57895.57
described in the Guide to Physical Therapist Practice,1 the
physical therapist selects specific tests and measures as a
means of gathering data about the patient. These tests and
measures are used to identify impairments and functional
limitations of the child; then, to help establish a diagnosis,
prognosis, and plan of care; and, finally, to select appropriate
interventions. Tests and measures that are used as a part of the
initial examination allow the therapist to confirm or reject
hypotheses about the factors that may contribute to the child’s
current level of functioning. In addition, tests and measures
may be used to support the therapist’s clinical judgments
about necessary interventions, appropriate goals, and ex-
pected outcomes for the child.

190 Connolly et al Pediatric Physical Therapy

 Early identification of a developmental delay using a
discriminative measure is useful for determining a child’s
eligibility for early intervention services under Individuals
with Disabilities Education Act (IDEA). Sec. 303.16 of Part
C of IDEA2 states that “Infants and toddlers with disabili-
ties. . . are experiencing developmental delays as measured
by appropriate diagnostic instruments and procedures.” In
selecting a specific measurement tool to document a child’s
motor age, the physical therapist must choose a test that
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A
yields both valid and reliable measurements of the child’s
performance. Validity is defined as the extent to which a
test measures what it purports to measure. Three types of
WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023
validity, construct, content, and criterion, may be used to
assess the viability of a test. Construct validity examines
the theory or hypothetical constructs underlying the test.
For example, the Peabody Developmental Motor Scales-2
(PDMS-2) is valid for measuring gross and fine motor pro-
ficiency but not for developmental reflexes or muscle tone.
Content validity refers to test appropriateness, or how well
the content of the test samples the subject matter or behav-
iors about which conclusions are to be drawn. Criterion
related validity is measured by examining concurrent va-
lidity and predictive validity. Concurrent validity repre-
sents the relationship of the performance on the test with
performance on another well reputed test. Concurrent va-
lidity is of specific interest to physical therapists when de-
ciding which testing instrument to use.
Two tests widely used by physical therapists in the
identification of children with developmental delays are
the PDMS-23 and the Bayley Scales of Motor Development,
2nd Edition (BSID-II).4 The BSID-II, like its predecessor,
the Bayley Scales of Infant Development (BSID),5 is a test
that is individually administered to infants and children to
assess their current level of developmental functioning.
The BSID-II consists of three scales: the Mental Scale, Mo-
tor Scale, and Behavior Rating Scale. The BSID-II was de-
veloped for use with children between one and 42 months
of age, and normal values were determined from a sample
of 1700 children. The normative group included 100 chil-
dren in each of 17 specified age groups between one and 42
months of age. Because development occurs more rapidly
in younger children, more age groups were sampled in the
one- to 12-month age range than in the 13- to 42-month
range. The scoring system for the BSID-II consists of Credit
(C), No Credit (NC), Refused (RF), Omit (O), or Caregiver
Report (RPT). However, a point is given only if the child
receives a Credit on an item. Therefore, the scoring can be
considered binary (one point for Credit and no point for
any other score). Scoring of the Mental Scale yields a Men-
tal Development Index (MDI) for the child. Information
from the motor scale is expressed as a Psychomotor Devel-
opment Index (PDI). The mean standard score for each of
the indexes for all age ranges is 100 with a standard devia-
tion of 15 points. A score of 115 and greater is considered
accelerated performance, 85 to 114 within normal limits,
70 to 84 mildly delayed performance, and 69 and below
Pediatric Physical Therapy
significantly delayed performance.4 Concurrent validity of
the BSID-II has been demonstrated with the McCarthy
Scales of Children’s Abilities6 with r values for the MDI
ranging from 0.57 to 0.77 and r values for the PDI ranging
from 0.18 to 0.59. Correlations with the Wechsler Pre-
school and Primary Scale of Intelligence-Revised (WPPSI-R)7
revealed r values ranging from 0.21 to 0.73 for the MDI and
from 0.14 to 0.41 for the PDI.4 The mental scale appears to
relate more highly than the motor scale to tests that assess
general cognitive abilities.
The PDMS-2 was developed as a new and improved re-
vision of the original Peabody Developmental Motor Scales
(PDMS). The original PDMS was created to improve motor
development assessment and programming for young chil-
dren with disabilities and was commonly used by health care
professionals as a means of assessing gross and fine motor
skills.8 The new PDMS-2 was designed to (1) estimate a child’s
motor competence; (2) compare gross and fine motor dispar-
ity; (3) provide qualitative and quantitative aspects of individ-
ual skills; (4) evaluate a child’s progress; and (5) provide a
research tool.3 On the basis of clinical comments about weak-
nesses in the original PDMS,9 changes made in the PDMS-2
include new normative data; new validity studies; new scor-
ing criteria with clearly specified performance levels for scores
of two, one, or zero; revisions in instructions for administer-
ing and scoring the test; and elimination of certain items that
appeared in the earlier edition.
Construct validity of the PDMS-2 as well as construct-
identification validity and concurrent validity have been
assessed and are reported in the test manual. Correlations
between performance on the PDMS-2 and age were found
to be high with r values ranging from 0.80 to 0.93. Con-
current validity of the PDMS-2 has been assessed through
comparisons with the original PDMS (Gross Motor Quo-
tient [GMQ] r ⫽ 0.84 and Fine Motor Quotient [FMQ] r ⫽
0.91) and with the Mullen Scales of Early Learning: AGS
Edition (GMQ r ⫽ 0.86 and FMQ r ⫽ 0.80).3,10 Other
psychometric properties of the PDMS-2 include high test-
retest reliability scores for two groups of children.3 Test–
retest reliability values for the total motor quotient for a
group of two- to 11-month-old infants were r ⫽ 0.89 and
r ⫽ 0.96 for a group of 12- to17-month-old infants. Inter-
scorer reliability for two separate groups of children (one
group of three- to 11-month-old infants and one group of
15- to 36-month-old children) was found to be r ⫽ 0.96
indicating very high test scorer reliability.
The BSID-II test manual includes information on reli-
ability between the Mental, Motor, and Behavioral Rating
Scales for groups of children from one to 42 months.5 Us-
ing the coefficient alpha, the coefficients for the Mental,
Motor, and Behavioral Rating Scales were 0.88, 0.84, and
0.88, respectively, indicating high levels of reliability. The
authors also used the standard error of measurement
(SEM) to assess the Mental and Motor scales and found low
SEMs, indicative of a high level of reliability.11 Interscorer
Concurrent Validity of the Bayley II and the Peabody 2 191
 reliability was assessed using a sample of 51 children, rang-
ing in age from two to 30 months and correlation coeffi-
cients for the Mental and Motor scales were 0.96 and 0.75,
respectively. Interscorer agreement levels of 90.9% were
achieved for a group of one- to five-month-old infants
and 87.5% for a group of 13- to 42-month-old children.4
Provost et al,12 in a study of 38 two-year-old chil-
dren who were developing typically, found a high cor-
relation between the BSID-II Motor Scale and the age
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A
METHODS
Subjects
The participants in the study were fifteen 12-month-
old children who were developing typically and who scored
greater than the 25th percentile on the Alberta Infant Motor
Scales (AIMS). All but one child scored at or greater than
the 50th percentile on the AIMS. The children ranged in
age from 11 months, 18 days to 12 months, 15 days (mean,

equivalent scores on the PDMS. The correlation between
age-equivalent scores of the PDMS Fine Motor Scale and
the BSID-II Motor Scale was r ⫽ 0.87. Correlation between
WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023
12 months, two days, SD ⫽ 9.11 days). Six girls and nine
boys were included in the sample. Table 1 presents demo-
graphic data on the children included in the study.
Although we included 15 subjects in the study, a

the PDMS Gross Motor Scale and the BSID-II Motor Scale
was only slightly lower at r ⫽ 0.83.12 Correlations between power analysis performed before the study had revealed
the standard scores of the BSID-II Motor Scale and the that only seven subjects were needed in order to achieve 80%
PDMS were r ⫽ 0.64 for the Fine Motor Subtest and r ⫽ 0.49 power using a one-tailed test at ␣ ⫽ 0.05 for the Pearson
for the Gross Motor Subtest. Thus, concurrent validity of the Product Moment correlation coefficient.14 We used a one-
BSID-II Motor Scale and the PDMS was supported for age tailed test because we anticipated a positive correlation
equivalents but not for standard scores. Only 58% of the between the scores. If we had used a two-tailed test with the
children were classified the same on the BSID-II Motor anticipation of either a positive or a negative correlation
Scale and the Fine Motor subscale for the PDMS and only between the scores, a power analysis would have revealed
66% were classified the same on the BSID-II Motor Scale that nine subjects would be necessary to achieve 80%
and the Gross Motor Subscale. power. Therefore, our sample size of 15 is large enough to
In a more recent study Provost et al13 studied the con- achieve adequate power for the data analyses using either a
current validity between the BSID-II and the PDMS-2 in a one-tailed or a two tailed test.14 Inclusion criteria were as
sample of children with developmental delays (mean age ⫽ follows: (1) full-term infants born within two weeks of the
25.3 months, SD ⫽ 9.7). Children’s scores in this study typical 40-week gestation period; (2) APGAR scores of
demonstrated a moderate correlation (r ⫽ 0.67) between eight or greater at five minutes; (3) age between 11 months,
the FMQ from the PDMS-2 and the PDI of the BSID-II. 16 days and 12 months, 14 days; (4) described by the par-
However, high correlations were found between the GMQ ent(s) and child care provider as developing typically; and
(r ⫽ 0.75) and the Total Motor Quotient (TMQ; r ⫽ 0.76) (5) scores on the Alberta Infant Motor Scales indicating
when compared to the PDI. However, more than 75% of the typical development. Exclusion criteria included the fol-
children who were deemed significantly delayed on the lowing: (1) any hospitalization since birth of more than 24
BSID-II Motor Scale (PDI ⱕ 69) did not score very poor hours; (2) any visual or auditory impairment; (3) any ge-
(TMQ ⱕ 69) on the PDMS-2. Thus, the findings suggested netic or neuromotor disorder that affected development;
that the standard scores of the BSID-II Motor Scale and the and (4) current illness that would interfere with testing.
PDMS-2 show poor agreement and have low concurrent The sample attended the Boling Child Care Center at the
validity in children with developmental disabilities. Addi- University of Tennessee Health Sciences Center or the
tionally, the marked differences in the standard scores of Methodist LeBonheur Children’s Hospital Child Care Cen-
the two tests might affect a child’s eligibility for services in ter. Informed consent was obtained from parents prior to
certain situations.
The purpose of this study was to compare concur- TABLE 1
rent validity of the BSID-II Motor Scale and the PDMS-2 Demographic Information on the Children in the Study
Gross Motor Subscale and Fine Motor Subscale by ad-
Gender Age Ethnicity
ministering both tests to typically developing 12-month-
old infants. We were concerned that the same low con- Child 1 M 12 months, 15 days Asian
current validity noted in the Provost et al.13 study in Child 2 M 11 months, 18 days White
Child 3 M 12 months, 10 days White
older children with developmental disabilities would be Child 4 M 11 months, 18 days Asian
present in younger child The research questions were as Child 5 M 12 months, 6 days White
follows: (1) Do the age-equivalent scores on the BSID-II Child 6 M 12 months Black
Motor scale correlate with the age equivalent scores on Child 7 M 12 months, 14 days White
the PDMS-2 Gross Motor and the Fine Motor Subscales? Child 8 M 11 months, 19 days White
Child 9 M 12 months, 9 days White
(2) At what level of reliability do the standard scores Child 10 F 12 months, 4 days White
(PDI) on the BSID-II Motor Scale correlate with the stan- Child 11 F 12 months, 4 days White
dard scores on the PDMS-2 Gross Motor and Fine Motor Child 12 F 11 months, 27 days White
subtest or the Total Motor Quotient? (3) Do the classi- Child 13 F 11 months, 27 days White
fications of standard scores on the BSID-II Motor Scale Child 14 F 12 months, 8 days Asian
Child 15 F 12 months, 11 days White
and the PDMS-2 agree?

192 Connolly et al Pediatric Physical Therapy

 testing each infant and the study was approved by the Uni-
versity of Tennessee Health Science Center Institutional
Review Board.
Procedures
To establish interrater reliability for this study, a pilot
study was conducted before the formal testing phase.
Three 12-month-old infants who were developing typically
were selected from the child care centers used for the pri-
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A
were less than 12 months of age, the reflex, stationary, and
locomotion gross motor subtests were used for determin-
ing the Gross Motor Quotient and the Total Motor Quo-
tient for the PDMS-2. For those children who were 12
months of age or older the stationary, locomotion, and
object manipulation gross motor subtests were used for
determining the Gross Motor Quotient and the Total Mo-
tor Quotient for the PDMS-2. The children were tested in
their classrooms at the child care facility. The rooms were

mary study and were tested for the pilot study. The same well lit and large enough for the child to perform test ac-
inclusion and exclusion criteria applied to children used in tivities including walking and throwing. The testing of
the pilot study. Five physical therapist students were each child took approximately one hour.
WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023

trained using the PDMS-2 or the BSID-II by a physical

therapist who had 12 years of experience in using standard-
ized testing instruments and who was a pediatric clinical
specialist. The pediatric physical therapist tested and scored
the three children using both the BSID-II Motor Scale and
the PDMS-2. Although the therapist conducted each test, the
students observed and scored each child at the same time.
Percent agreement between the scores of the students on all
portions of BSID-II Motor Scale and the PDMS-2 and the
scores of the pediatric physical therapist on these tests were
examined to determine whether acceptable levels of inter-
rater reliability were established. The percent agreement of
the PDMS-2 scores between the pediatric physical thera-
pists and the physical therapist students was 95% or higher
for all three children tested in the pilot study. The percent
agreement of the BSID-II scores between the pediatric
physical therapist and the physical therapist students
ranged between 79% and 100%. A minimal criterion of 70%
agreement had been previously agreed upon as evidence of
high correlation needed for the study.11(p. 354)
Five physical therapist students who had completed
the pilot study and who were under the supervision of the
pediatric physical therapist were responsible for data col-
lection. The PDMS-2 and the BSID-II Motor Scale were
administered to all 15 children. For those children who
Data Analysis
Each child’s age equivalent score, gross motor quo-
tient, fine motor quotient, and total motor quotient was
determined for the PDMS-2. For the BSID-II Motor Scale,
the age-equivalent score, and the psychomotor develop-
ment index was determined. Age-equivalent scores for
each child are shown in Table 2. However, age equivalent
scores for object manipulation are not reported for those
children who were less than 12 months of age.
The standard scores for each child were calculated for
both the BSID-II Motor Scale and the PDMS-2 because both
scales are normalized with 100 serving as the average and
15 points as the standard deviation. Standard scores for
each child on both the PMDS-2 and the BSID-II are shown
in Table 3. Concurrent validity was examined using corre-
lational analysis with the Pearson-Product Moment corre-
lation coefficient. The magnitude of each correlation anal-
ysis was interpreted using descriptive terms for the strength of
correlation coefficients11 with correlations of 0.00 to 0.25
indicating little, if any correlation, 0.26 to 0.49 indicating
low correlation, 0.50 to 0.69 indicating moderate correla-
tion, 0.70 to 0.89 indicating high correlation and 0.90 to
1.00 indicating very high correlation. P values were also

TABLE 2
Age Equivalents on BSID-II and PDMS-2

Stationary Locomotor Object Manipulation Grasping Visual Motor Integration

PDMS-2 PDMS-2 PDMS-2 PDMS-2 PDMS-2 BSID-II
Child 1 18 13 13 12 13 11
Child 2 11 11 – 9 12 10
Child 3 18 15 12 14 12 12
Child 4 11 13 – 13 12 11
Child 5 11 14 12 10 11 12
Child 6 11 13 13 12 11 11
Child 7 18 15 12 14 14 12
Child 8 11 10 – 12 12 11
Child 9 14 12 12 9 13 11
Child 10 14 13 12 10 13 11
Child 11 11 10 12 10 10 11
Child 12 11 11 – 10 13 11
Child 13 11 14 – 9 14 12
Child 14 11 14 12 9 13 12
Child 15 11 14 12 15 12 11
MEAN 12.8 12.8 12.2 11.2 12.33 11.27
(SD) (2.88) (1.66) (0.42) (2.08) (1.11) (0.59)

Pediatric Physical Therapy Concurrent Validity of the Bayley II and the Peabody 2 193
 TABLE 3
Standard Scores on BSID-II and PDMS-2

Gross Motor Quotient Fine Motor Quotient Total Motor Quotient Psychomotor Development Index
PDMS-2 PDMS-2 PDMS-2 BSID-II
Child 1 106 103 105 89
Child 2 98 100 98 77
Child 3 106 106 107 105
Child 4 104 109 107 89
Child 5 98 94 96 101
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A

Child 6 102 97 100 89

Child 7 106 112 109 101
Child 8 98 102 101 95
Child 9 100 97 98 85
WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023

Child 10 102 100 101 89

Child 11 94 91 92 89
Child 12 100 103 101 85
Child 13 106 106 107 97
Child 14 98 97 97 101
Child 15 100 109 104 93
Mean 101.2 101.7 101.53 92.33
(SD) (3.76) (5.97) (4.88) (7.62)

calculated to determine if the calculated correlation coeffi-
cients could have occurred by chance even if in fact there
was no relationship between the scores.
Frequencies of agreement between classifications also
were calculated. On the BSID-II, classification labels are
derived from the standard deviations from the mean of the
test. Children scoring between one and two standard devi-
ations below the mean are identified as mildly delayed
while scores more than two standard deviations below the
mean are identified as significantly delayed.3 The PDMS-2
uses standard scores to classify a child’s performance in
relationship to other children. In this study of 12-month-
old infants who were developing typically, we used the
standard scores of 90 to 110 to denote average behavior and
80 to 89 to denote below average as defined in the PDMS-2
test manual to categorize the children for our comparison
with the BSID-II Motor Scale scores.
RESULTS
Mean and standard deviations are presented for the
standard scores and for age equivalent scores for the
children on the BSID-II and the PDMS-2 in Table 2 and
3. Correlations between standard scores for the children
on the BSID-II and PDMS-2 are shown in Table 4. The
correlation of the standard scores of the PDMS-2 fine
motor quotient and the PDI of the BSID-II was low and
nonsignificant. (r ⫽ 0.22, p ⫽ ns). Standard scores on
the PDMS-2 gross motor quotient and the PDI of the
TABLE 4
Correlation (r) Between Standard Scores on BSID-II and PDMS-2
Variable GMQ FMQ
GMQ – –
FMQ 0.73* – –
TMQ 0.91* 0.94*
PDI 0.30 0.22
* Denotes significance at p ⬍ 0.05.
BSID-II also showed a low and non-significant correla-
tion (r ⫽ 0.30, p ⫽ ns) as did the standard scores of the
PDMS-2 total motor quotient and the PDI of the BSID-II
(r ⫽ 0.32, p ⫽ ns). Comparison of each child’s scores on
the PDI of the BSID-II and the standard scores of the
PDMS-2 is presented in Figure 1.
Correlations between the age equivalent score on the
PDMS-2 subscales and the BSID-II Motor Scale are shown
in Table 5. Little if any correlation (r ⫽ 0.13, p ⫽ ns) was
found between the age equivalent scores of the PDMS-2
fine motor subtest for grasp and the BSID-II Motor Scale.
Low but non-significant correlations were found between
the age equivalent scores of the PDMS-2 both for stationary
gross motor subtest (r ⫽ 0.28, p ⫽ ns) and for the fine
motor subtest for visual motor integration (r ⫽ 0.29, p ⫽
ns) when compared to the BSID-II Motor Scale. A low but
negative correlation (r ⫽ ⫺0.41, p ⫽ ns) was found be-
tween the age equivalent scores of the PDMS-2 gross motor
subtest for object manipulation and the BSID-II Motor
Scale. Only the age equivalent scores of those children who
were older than 12 months were included in the correla-
tion between object manipulation and the BSID-II Motor
Scale. A high and significant correlation (r ⫽ 0.71, p ⬍
0.05) was found between the age equivalent scores of the
PDMS-2 gross motor subtest for locomotion and the
BSID-II Motor Scale.
All 15 children’s scores (greater than 90) would have
been considered average on the TMQ for the PDMS-2 and
14 of the children’s scores (85 or greater) would have been
considered to be in the average range on the BSID-II. How-
ever, one of the children would have scored in the mildly
delayed range on the BSID-II. Thus agreement of scores on
TMQ
classification for the sample was 93%.
–
– DISCUSSION
0.32 The results of this study raise questions about the
concurrent validity of two assessment tools commonly

194 Connolly et al Pediatric Physical Therapy

Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A
WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023

Fig. 1. Individual Child Scores on the PDMS-2 (GMQ, FMQ, and TMQ) and the BSID-II (PDI).

TABLE 5
Correlation (r) Between Age-Equivalent Scores on BSID-II and PDMS-2

Stationary Age Locomotion Age Object Manipulation Age Grasp Age Visual Motor Integration Age
Variable PDMS-2 PDMS-2 PDMS-2 (n ⫽ 10) PDMS-2 PDMS-2
Stationary age – – – – –
PDMS-2
Locomotor age 0.45 – – – –
PDMS-2
Object manipulation age 0.13 ⫺0.11 – – –
PDMS-2 (n ⫽ 10)
Grasp age 0.41 0.41 0.12 – –
PDMS-2
Visual motor integration age 0.40 0.39 ⫺0.09 ⫺0.06 –
PDMS-2
BSID-II age 0.28 0.71* ⫺0.41 0.13 0.29
* Denotes significance at p ⬍ 0.05.

used in pediatric physical therapy. Our study with 12-
month-old infants who were developing typically had find-
ings that were not consistent with those reported in a pre-
vious study13 of the concurrent validity of the BSID-II
Motor Scale and the PDMS-2 with children who had devel-
opment delays. We found a low correlation between the
PDI of the BSID–II and the GMQ and the TMQ of the
PDMS-2 in contrast to the report of Provost et al.13 who
reported significant correlations of 0.75 and 0.76, respec-
tively. Although Provost et al13 found a lower correlation
between the FMQ and the PDI than between the GMQ and
the TMQ and the PDI, their study revealed a moderate
correlation between the FMQ and the PDI in comparison
with our low correlation. If very high correlational values
of approximately r ⫽ 0.95 are expected between two forms
of the same test to demonstrate alternate form reliability,15
we propose that the use of either the BSID-II or the PDMS-2
will not yield the same results when testing a child at age 12
months. However, agreement between the classifications
of children using the BSID-II and the PDMS-2 was found in
93% of the children studied. No child in the study scored in
the significantly delayed category on either test. However,
one child scored in the mildly delayed category on the
BSID-II Motor Scale, whereas no children scored at this
level on the PDMS-2. Similar to the study by Provost et al,13
we found that children scored higher on the PDMS-2 than
on the BSID-II Motor Scale. Differences between our find-
ings and those reported by Provost et al.13 may be related to
our focus on children who are typically developing rather
than children with developmentally delays.
The results of this study also raise concerns about the
lack of agreement between the age-equivalent scores on the
BSID-II Motor Scale and the PDMS-2. Low correlations
between age equivalent score for the stationary and the
visual motor integration subtests on the PDMS-2 and the
age-equivalent score for the BSID-II Motor Scale, no corre-
lation between the age equivalents for the grasp subtest
when compared with the age-equivalent score for the
BSID-II Motor Scale and moderate but negative correlation
between age equivalent scores on object manipulation and

Pediatric Physical Therapy Concurrent Validity of the Bayley II and the Peabody 2 195
 the age equivalent scores for the BSID-II illustrate the rea-
son for this concern. Our finding that the locomotion age
was most highly correlated with the age equivalent scores
for the BSID-II Motor Scale was consistent with Provost et
al’s13 finding for children with developmental delays. Our
study demonstrates that if a child’s eligibility for services is
based upon age equivalent scores or percent delays in spe-
cific categories, the results of the BSID-II Motor Scale
might suggest the need for intervention services whereas
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1A
use of age equivalence scores on the PDMS-2 would sug-
gest that no services were needed.
One limitation of our study was that the children were
WnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/13/2023
tested at a child care center. The optimal environment for
testing children is in their homes. For this study, children
were tested in their classrooms with their teachers and
other classmates present. This environment presented a
large number of visual, auditory, and tactile distractions.
However, because both tests were administered to each
child simultaneously, no bias for either test would be ex-
pected. Another limitation may have been the fact that the
children’s parents were not present; however, a child care
provider who was familiar with the child was present. The
small number of children included in the sample could
be cited as a limitation of the study; however, the num-
ber included was adequate for statistical analyses of our
findings.14
Future research is needed on children of varying ages
to determine whether concurrent validity is present be-
tween the PDMS-2 and the BSID-II. The Bayley Scales of
Infant and Toddler Development III16 recently has been
published, and we suggest that further studies should be
conducted to determine whether improved concurrent va-
lidity is noted between the PDMS-2 and this new version of
the Bayley Scales. Additionally, further studies need to be
performed with children who are from a variety of back-
grounds and cultures to determine the cultural sensitivity
of the test. Future research also could be done in various
clinical settings and testing environments, in different geo-
graphical regions, and with children of different socioeco-
nomic backgrounds.
CONCLUSIONS
This study does not support concurrent validity be-
tween the PDMS-2 and the BSID-II using the standard
196 Connolly et al
scores or for age equivalent scores except for locomotion
sub test. Therefore, it can be concluded that the PDMS-2
and the BSID-II performed with children at 12 months of
age may yield dissimilar findings in motor development. If
the same differences found using the standard scores or the
age equivalent scores in these two tests in children who are
nontypically developing are similar, the child’s eligibility
for services may be affected in some states. We support the
use of multiple sources of information when deciding upon
eligibility for services rather than the use of the standard
scores on only one test.
REFERENCES
1. APTA Guide to physical therapist practice. 2nd ed. Phys Ther. 2001;
81:42– 48.
2. Individuals with Disabilities Education Act. Amendments of 1997. 20
U.S.C. §1401.
3. Folio MR, Fewell RR. Peabody Developmental Motor Scales. Examiner’s
Manual. 2nd ed. Austin, Tex: Pro-Ed; 2000.
4. Bayley N. Bayley Scales of Infant Development. 2nd ed. San Antonio,
Tex: The Psychological Corporation; 1993.
5. Bayley N. Bayley Scales of Infant Development Manual. New York, NY:
The Psychological Corporation; 1969.
6. McCarthy D. McCarthy Scales of Children’s Abilities. San Antonio,
Tex: The Psychological Corporation; 1972.
7. Weschler D. Manual for Wechsler Pre-school and Primary Scale of
Intelligence-Revised. San Antonio, Tex: The Psychological Corp; 1989.
8. Folio MR, Fewell RF. Peabody Developmental Motor Scales and activity
cards. Chicago, Ill: Riverside; 1983.
9. Palisano R. Concurrent and predictive validities of the Bayley Motor
Scale and the Peabody Developmental Motor Scales. Phys Ther, 1986;
66:1714 –1719.
10. Mullen EM. Mullen Scales of Early Learning: AGS Edition. Circle
Pines, Minn: American Guidance Services; 1995.
11. Domholdt E. Physical Therapy Research: Principles and Applications.
2nd ed. Philadelphia: WB Saunders; 2000.
12. Provost B, Crowe TK, McClain C. Concurrent validity of the Bayley
Scales of Infant Development II and the Peabody Developmental
Motor Scales in two-year-old children. Phys and Occup Ther Pediatr
2000;20:5–18.
13. Provost B, Heimerl S, McLain C, et al. Concurrent validity of the
Bayley Scales of Infant Development II motor scale and the Peabody
Developmental Motor Scales-2 in children with developmental de-
lays. Pediatr Phys Ther. 2004;16:149 –156.
14. Portney LG, Watkins MP. Foundations of Clinical Research: Applica-
tions to Practice. 2nd ed. Upper Saddle River, NJ: Prentice Hall; 2000.
15. Munro BH. Statistical Methods for Health Care Research. Philadelphia,
Pa: Lippincott-Raven; 1997.
16. Bayley N. Bayley Scales of Infant and Toddler Development. 3rd ed. San
Antonio, Tex: PsychCorp; 2006.
Pediatric Physical Therapy