Description: The MacArthur Communicative
Development Inventories (CDI) assess early language skills
of children between the ages of 8 and 30 months using parent
reports. The CDI consists of two inventories, each with two
sections. The CDI/Words and Gestures inventory is for infants
between the ages of 8 and 16 months. The inventory’s
words section, which has a 28-item list of phrases and a 396-word
checklist, is used to assess the infant’s production
and understanding of words and phases. The gestures section
covers 63 gestures for communication, play, imitation of parents
and other adults, and activities with objects. The CDI/Words
and Sentences inventory is for toddlers between the ages of
16 and 30 months. The inventory’s word section assesses
vocabulary using a 680-word checklist. The second part assesses
the toddler’s use of possessives, plurals, and tenses,
and development of complex sentences. A short version of the
inventories is also available. The short version of the CDI/Words
and Gestures inventory consists of an 89-word vocabulary checklist
and there are two equivalent short versions of the CDI/Words
and Sentence Inventory, both with a 100-word vocabulary checklist.
Uses of Information: The CDI
can be used to screen for delays in language development,
to identify problematic skills, help formulate intervention
strategies, and evaluate treatment outcomes.
Reliability: For the long form,
(1) internal consistency reliability: Cronbach alpha coefficients
for the CDI/Words and Gestures vocabulary production, vocabulary
comprehension, and gestures scales were .95, .96, and .39,
respectively. The low alpha for the gestures scale resulted
from the low correlation of two gestures subscales with a
third. The CDI/Words and Sentences vocabulary production and
sentence complexity scales had alphas of .96 and .95, respectively.
(2) Test-retest reliability (6-week and 6-month intervals):
correlation coefficients ranged from .60 to .90 on the infant
inventory and .90 and higher on the toddler inventory. (3)
Inter-rater reliability: no information available.
For the short form, internal consistency: the infant form
had a Cronbach alpha of .97 and the toddler form A and B each
had an alpha of .99. Test-retest (with 2-week interval): the
infant form had a correlation of .88 for vocabulary comprehension
and .90 for vocabulary production. The vocabulary production
test-retest correlations were .74 and .93 for the toddler
form A and B, respectively.
Validity: For the long form,
(1) Concurrent validity: several comparisons of parent reports
using the earlier version of the CDI/Word and Sentences inventory
(little changed from the current inventory) with assessments
made through laboratory observations found correlations that
ranged from .40 to .67 when compared to the Preschool Language
Scale and .53 to .85 when compared to the Expressive One Word
Picture Vocabulary Test. Similar tests on an even earlier
version of the inventory with Bayley language subscales yielded
correlations ranging from .33 to .79. Tests of the gestures
scale showed significant differences between high- and low-gesture
children on three of the four laboratory assessment instruments
and “substantial and significant concurrent correlations”
between parent and laboratory gestures reports for 12-month-old
children and one of the gestures subscales at 10 months. Correlations
between the CDI/Word and Sentence syntactic development scale
and laboratory measures ranged from .74 to .88. (2) Predictive
validity: administrations six months apart of the CDI/Words
and Sentences inventory for total vocabulary and grammatical
complexity yielded correlations of .71 and .62, respectively.
The CDI/Words and Gestures inventories administered six months
apart had a total vocabulary correlation of .38. The correlation
between the inventories on vocabulary production was .69.
For the short form, the correlations between the infant short
and long form were .98 on vocab ulary comprehension and .97
on vocabulary production. The overall correlations between
the toddler long form and short form A and B were each .99.
Method of Scoring: Scoring can
be done manually or by computer. The User’s Guide
provides instructions for manual scoring. Scoring inventories
usually involves counting the number of marked items or
affirmative responses by sections. Using tables in the
User’s Guide, raw scores can be converted into gender-
and age-specific percentile rankings. Scoring software
is available at no cost at www.utdallas.edu/~vamarch/cdi/.
The software scores both the long and short and English
and Spanish versions of the inventories.
Interpretability: The manual
provides instructions for interpreting the results. The normed
percentile ranking allows the infant/toddler’s performance
to be compared to other infants/toddlers.
Training Support: None described.
Adaptations/Special Instructions for
Individuals with Disabilities: The manual cautions
against using the CDI with developmentally delayed children
whose chronological age exceeds the upper limits of the inventory.
Report Preparation Support:
The manual contains report forms for each inventory for the
user to complete.
References:
Fenson, Larry, Philip S. Dale, J. Steven Reznick, Donna Thal,
Elizabeth Bates, Jeffery P. Hartung, Steve Pethick, and Judy
S. Reilly. MacArthur Communicative Development Inventories,
User’s Guide and Technical Manual. Singular Publishing
Group, Inc., San Diego, CA, 1993.
Fenson, Larry, Philip S. Dale, J. Steven Reznick, Elizabeth
Bates, Donna J. Thal, and Stephen J. Pethick. Variability
in Early Communicative Development. Monographs of the Society
for Research in Child Development, Vol. 59, No. 5, 1994.
Fenson, Larry, Steve Pethick, Connie Renda, Philip S. Dale,
and J. Steven Reznick. “Short-form Versions of the MacArthur
Communicative Development Inventories.” Applied Psycholinguistics.
Vol. 21, No. 1, 2000, pp. 91-115.
|
