Shop By:

The MacArthur-Bates Communicative Development Inventories User's Guide and Technical Manual, Second Edition

The MacArthur-Bates Communicative Development Inventories User's Guide and Technical Manual, Second Edition

Authors: Larry Fenson Ph.D., Virginia A. Marchman Ph.D., Donna J. Thal Ph.D., Philip S. Dale Ph.D., J. Steven Reznick Ph.D., Elizabeth Bates Ph.D.

ISBN: 978-1-55766-884-4
Pages: 208
Copyright: 2006
Available Stock
Request Print Exam Copy
Paperback $59.95 Qty:

Size:  8.5 x 11.0
Stock Number:  68844
Format:  Paperback
Promotion Code: 

SAVE when you order this item as part of a set.

This easy-to-read manual provides professionals with a thorough introduction to the standardized, parent-completed report forms designed by top language researchers to assess language and communication skills in young children ages 8–30 months. The second edition of the English User’s Guide and Technical Manual includes:

  • more demographically balanced normative data
  • norms up to 17-18 months for the CDI: Words & Gestures
  • more directions on administrating and scoring the CDIs
  • an introduction to the automated CDI Scoring Program
  • guidance on how scores for various subpopulations should be interpreted
  • expanded information on machine scanning and a new option using desktop scanners
  • key updates on research, clinical findings, and reliability and validity
  • detailed information and normative values for the CDI-III, an extension for children 30–37 months of age

This manual is part of the MacArthur-Bates Communicative Development Inventories (CDIs). The CDIs and their Spanish adaptation, the Inventarios, are standardized, parent-completed report forms that track young children's language and communication skills. Top language researchers developed the report forms, designing them to focus on current behaviors and salient emergent behaviors that parents can recognize and track.

Learn more about the CDIs and the Inventarios.

Review and Rate this Item
Be the first to submit a customer review on this product!
Review and Rate this Item
List of Tables and Figures
The CDI Advisory Board
About the Authors

  1. Introduction
    • Overview
    • Organization of this Manual
    • Methods of Assessing Early Language Skills
      • Structured Tests
      • Language Samples
      • Parent Report
    • Description of the Inventories
      • CDI: Words and Gestures
      • CDI: Words and Sentences
      • Guidelines for Choosing Between the CDI: Words and Gestures and the CDI: Words and Sentences Forms in the 16- to 18-Month Age Range
    • The MacArthur-Bates Communicative Development Inventory Short Forms
  1. Administration, Scoring, and Interpretation
    • Administering the Inventories
      • General Instructions
      • Other Administration Considerations
    • Scoring and Interpreting the Inventories
      • Scoring Examples
      • Automating the Scoring Process
      • Using the Norms
      • Obtaining and Interpreting Children's Scores
  1. Clinical and Research Applications
    • Clinical Applications of the Inventories
      • Screening for Language Delay
      • Evaluation of Older Children with Language Delay
      • Formulation of Intervention Strategies
      • Evaluation of Intervention Effects
    • Research Applications
      • Screening and Preselection of Children at Different Levels of Language Development
      • Screening and Preselection of Children with Particular Language Characteristics or Those with Unusual Profiles
      • Matching Children on Language Skills
      • Examination of the Influence of Other Variables on Language Development
    • Non-English Versions of the Inventories
  1. Phases of the Development of the Inventories
    • Origins of the Instruments
      • A Preliminary Norming Study
      • Development of the CDI: Words and Gestures Form
      • Development of the CDI: Words and Sentences Form
      • Changes Reflected on Both Forms
    • The Norming Study
      • Sampling Procedure
      • Medical Exclusion Criteria
      • Demographic Distribution of the Sample
    • Developmental Trends in Communicative Skills
      • Statistical Methods
      • Developmental Trends in Communicative Development Specific to the CDI: Words and Gestures Form
      • Developmental Trends in Communicative Development Specific to theCDI: Words and Sentences Form <
    • Impact of Demographic Factors on Language Outcomes
      • Maternal Education
      • Birth Order
    • Relations Among the Inventory Subscales
      • Intercorrelations Among the Components of the CDI: Words and Gestures
      • Intercorrelations Among the Components of the CDI: Words and Sentences
    • Reliability of the Inventories
      • Internal Consistency
      • Test-Retest Reliability
    • Validity of the Inventories
      • Face Validity
      • Content Validity
      • Convergent Validity
      • Concurrent Validity
      • Predictive Validity
    • Looking Ahead
  1. Percentile Tables and Figures
  2. The MacArthur-Bates Communicative Development Inventory-III
    • Components of the CDI-III
      • Vocabulary Checklist
      • Sentences
      • Using Language
    • The Norming Study
    • Developmental Trends, Norms, and Sex Differences
    • Validity of the CDI-III
    • Research Findings with the CDI-III
    • Conclusions



  1. Basic Information Form Used in Original Normative Study
  2. Expanded Basic Information Form
  3. Child Report Forms


Excerpted from Chapter 1 of MacArthur–Bates Communicative Development Inventories User's Guide and Technical Manual SECOND EDITION, by Larry Fenson, Ph.D. Donna J. Thal, Ph.D. Virginia A. Marchman, Ph.D. Philip S. Dale, Ph.D., & Elizabeth Bates, Ph.D.


Copyright © 2007 by The CDI Advisory Board. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.




Parents notice the emergence of language. A child's first words are eagerly awaited, often dutifully recorded in baby books and diaries, and always shared with family and friends. Subsequent progress is noted and enjoyed until the child's language is so far along that it is finally taken for granted. Popular books for parents abound with advice for fostering early communication and guidelines suggesting when language milestones should be reached, but there is no question that early language is salient for parents, and they need no impetus to notice this remarkable aspect of early development. From this perspective, it is surprising that only since the 1980s have parents come to be viewed as a reliable source of information about the young child's communicative skills. The obvious obstacle was to find a technique that allows researchers and clinicians to harvest the parent's rich view of the child's language.


The first systematic attempts to use questionnaires to tap parents' knowledge about their children's language skills were reported by Bates and her colleagues in the 1970s and 1980s (e.g., Bates, Camaioni, &Volterra, 1975). These questionnaires evolved into a set of inventories with detailed questions about vocabulary and grammar. At around the same time, Rescorla (1989) developed a 310–item parent–report checklist of productive vocabulary that was designed as a screening tool for detecting language delay in 2–year–old children. The Mac– Arthur–Bates Communicative Development Inventories (CDIs) build on the foundations laid by each of these earlier initiatives (see section ‪Origins of the Instruments‪ in Chapter 4). The first two CDIs (sometimes called Inventories in this manual) were published in 1992. These were the CDI: Words and Gestures and the CDI: Words and Sentences. Since that time, the CDIs have been expanded to include one–page brief versions of CDI: Words and Gestures and CDI: Words and Sentences (Fenson, Pethick, et al., 2000). The MacArthur–Bates Communicative Development Inventory—III (CDI–III; described in detail in Chapter 6) was also designed to be used with children from 30 to 37 months of age.


Members of the CDI team have also developed Mexican Spanish versions of CDI: Words and Gestures and CDI: Words and Sentences, as well as brief versions of each of these instruments. The Spanish CDIs and an accompanying manual were published in 2003 (Jackson–Maldonado et al.). In addition, the CDI Advisory Board has authorized and encouraged the adaptation of the CDI forms into a number of other languages. These adaptations are not direct translations to other languages but, rather, assessment tools that take linguistic and cultural differences into account (see Dale, Fenson, &Thal, 1993, &http://www A number of these instruments are now available for public use, many with supporting normative data (see the CDI web site,, for a complete list of non–English versions of the CDIs). Finally, the CDIs will soon be available in an adaptive–testing format, which uses a computer–based administration of a dynamically selected subset of CDI questions to converge on relevant language parameters quickly and efficiently.


There are two distinct advantages to the strategy of using parents as informants regarding the child's language. First, because the CDIs tap knowledge that most parents and caregivers have about their young children's communicative development, the CDIs lead to a more ecologically valid assessment (Crais, 1995) than commonly used laboratory and clinical measures. Second, the CDI strategy is responsive to Part H of the Education of the Handicapped Act Amendments of 1986 (PL 99–457), Part C of the Individuals with Disabilities Education Act (IDEA) Amendments of 1997 (PL 105–17), and the Individuals with Disabilities Education Improvement Act of 2004 (PL 108–446)—all of which require parent input in child evaluation procedures. Because of these advantages, the CDIs have found wide application among researchers and clinicians.


The clinical utility of the CDIs is particularly noteworthy. Clinicians have found that the CDIs are useful in screening and developing a prognosis for children with language delays (Crais, 1995; Heilmann, Weismer, Evans, &Hollar, 2005; Miller, Sedey, &Miolo, 1995; Thal, Reilly, Seibert, Jeffries, &Fenson, 2004; Yoder, Warren, &McCathren, 1998). Researchers have used the CDIs extensively with a wide variety of populations, including children with language delay, children with Down syndrome and Williams syndrome, bilingual children, children in child care, children from different socioeconomic status (SES) groups, babies who were born preterm or with focal lesions, and so forth (see Chapter 3 for a review of the diverse applications of the CDIs).


This new edition of the manual expands the information presented in the first edition (Fenson et al., 1992/1993) in a number of significant ways:


1. We present an expanded set of normative data that is more demographically balanced than the original norms. The expanded norms also extend the upper age of the CDI: Words and Gestures form to 18 months (formerly 16 months).

2. We provide an overview of the brief versions of the CDIs that parallel the CDI: Words and Gestures form and the CDI: Words and Sentences form.

3. We present detailed information and normative values for the CDI–III, a brief adaptation that is applicable to children 30—37 months of age.

4. More detail is presented on the administration of the Inventories.

5. More detailed directions are provided in the text and appendices on how to tabulate scores for the various sections of the Inventories. Guidelines for scoring the three longest utterances and deriving a mean score (in terms of the mean length of the three longest sentences [M3L]) are greatly expanded, and directions for avoiding common scoring errors are provided.

6. Specific guidelines are provided for using the normative tables for the CDI: Words and Gestures form and the CDI: Words and Sentences form.

7. The section on machine scanning is greatly expanded, and a section is provided on the new option of using desktop scanners.

8. A new CDI automated scoring program is introduced.

9. A new section presents recent data on SES variations in CDI performance and addresses the issue of how scores for various subpopulations should be interpreted.

10. The section on research and clinical applications is updated to reflect new findings.

11. The sections on reliability and validity are greatly expanded, incorporating new data.

One additional change in the CDIs calls for special notice. In December 2003, our beloved colleague Elizabeth Bates died after a long battle with pancreatic cancer. The role that Liz played in the development of the CDIs was vital. The instrument itself was created within Liz's vision of a productive science of child language development, its early stages of development benefited greatly from Liz's nurturance and encouragement, and the continued evolution and impact of the CDIs reflect Liz's influence in every aspect. The CDI team has renamed the Inventories the MacArthur–Bates Communicative Development Inventories to reflect Liz's many contributions.




The main body of this manual is divided into five chapters. In Chapter 1, methods of assessing early language development are reviewed, the role and value of parent report is considered, and the instruments themselves are described in detail. Chapter 2 provides directions for administering, scoring, and interpreting the CDIs. Chapter 3 outlines possible applications of the CDIs for clinical and research purposes. In Chapter 4, details of the developmental trends for the norming sample are presented, the intercorrelations among the various components of each inventory are described, and measures of the reliability and validity of the instruments are reported. Chapter 5 presents percentile tables and figures for major variables separately for boys and girls and for both sexes combined. Chapter 6 describes the CDI–III.


There are three appendices at the end of the book. Appendix A is the original Basic Information Form that was used in the normative studies to record demographic and other background information about the child and family. Appendix B offers an expanded, photocopiable Basic Information Form that was used in the Spanish normative study and by some of the investigators who contributed data for the expanded CDI norms presented in this book. Appendix C presents the Child Report Forms that provide templates for creating summary sheets for reporting scores. The photocopiable Child Report Forms are useful for research or clinical applications and for distribution to physicians and/or parents when accompanied by an interpretive letter. Versions of these forms are automatically generated by the CDI Scoring Program (see Chapter 2).




The language assessment techniques used in research and clinical assessment of infants and toddlers fall into three categories: structured tests, language samples, and parent report. We consider the advantages and disadvantages of each approach.


Structured Tests


In this method, responses are elicited from the child in a relatively formal manner in order to assess the child's knowledge of vocabulary and grammar. Examples of well–known structured tests include the Sequenced Inventory of Communication Development–Revised (SICD–R; Hedrick, Prather, &Tobin, 1984), the Reynell Development Language Scales: U.S. Edition (Reynell &Gruber, 1990), and the Preschool Language Scale, Fourth Edition (PLS–4; Zimmerman, Steiner, &Pond, 2002). Each of these instruments assesses several aspects of communicative development, has been normed on a representative population, and has demonstrated validity and reliability. However, the limitations of structured tests often outweigh their benefits when used with infants and toddlers. Tests require a considerable amount of time to administer, personnel must be trained, and the cooperation of the young child is essential. The latter constraint is particularly problematic for infants and toddlers, who may fear and distrust strangers (e.g., test administrators) and whose cognitive and emotional states vary significantly from hour to hour. The usefulness of structured tests is also limited by their expense, the modest validity that has been demonstrated for most of these instruments when conducted prior to 21/2 to 3 years of age, and the inherent problems that arise when standardized tests are used with linguistically and culturally diverse populations (Crais, 1995; Gutierrez–Clellen, 1996; Kayser, 1995; Lahey, 1988; Langdon, 1992b; Mattes &Omark, 1991).


Language Samples


Language sampling is a second popular technique used by researchers and clinicians to gather information about a child's language skills. The aim is to create conditions that prompt the child to talk in a typical manner so that key features of the child's speech can be evaluated. These features include vocabulary range and grammatical complexity. Researchers and clinicians are trained to follow the child's lead, to make conversational responses to the child's utterances, and to ask open–ended questions. Transcription and analysis of naturalistic language samples require highly trained personnel and a substantial amount of time. As a result, most studies incorporating language samples use small numbers of children. Special care needs to be taken when generalizing results from case studies or when data are obtained from a small sample. Another constraint associated with language sampling is that the behavior observed during interactions is highly variable across contexts and interlocutors. In all interactions between a child and another person, it is the dyad that is assessed—whether or not that is the intent of the examiner. Clearly, parent—child interactions differ from those between a child and a specifically trained clinician or researcher. They also differ across parents and across clinicians/researchers, regardless of the level of training. In addition, context has a strong effect on the nature of the behaviors that are observed, especially with young children.


Although a language sample may yield a broader reflection of a child's vocabulary and grammar than a structured test and may provide relatively detailed indices of child language, the method still assesses a restricted range of the child's linguistic abilities. Researchers and clinicians are often most interested in emerging language skills. It is precisely those skills that are unlikely to be observed in a spontaneous play interaction, in which the interlocutor tends to follow the child's lead and purposely refrains from asking pointed or difficult questions.


An additional complication of using language samples emerges from the fact that the technique was developed in the United States with middle–class children who are likely to have had extensive experience with the types of interactions used to obtain the language sample (e.g., Bloom &Lahey, 1978; Lund &Duchan, 1988; Miller, 1981; van Kleeck &Richardson, 1990). These types of interactions may be unfamiliar to children from other cultural groups (Delgado– Gaitán &Trueba, 1991; Farver, 1993; Heath, 1983; Langdon, 1992a; Roopnarine, Johnson, &Hooper, 1994), qualifying the validity of the language samples.


Differences in conversational styles during different types of play interactions are also found within the same cultural or socioeconomic group (Heath, 1983; Laosa, 1982). For example, Jackson–Maldonado (1994) compared the free play language behavior of Spanish–speaking children and parents who were from very low socioeconomic backgrounds in two different play contexts. When provided with a set of colorful plastic toys like those often used in child language research, the mothers typically did not allow their children to play with the colorful toys during the language sampling sessions, saying ‪Don't touch it‪ when the children tried to do so. When presented with common household objects found in a local market, such as clay pots, wooden spoons, and fruits or vegetables made from natural materials, however, the mothers eagerly encouraged the children to play with them (Carrillo Aranguren, Jackson–Maldonado, Thal, &Flores, 1997; Jackson–Maldonado, 1997).


Parent Report


The oldest type of parent report is the diary, which has a long and respected history in developmental psycholinguistics (e.g., Dromi, 1987, Hernández–Pina, 1984; Leopold, 1949; Llorach, 1976; López Ornat, Fernandez, Gallo, &Mariscal, 1994; Stern &Stern, 1907; Tomasello, 1992; Truex, 1982). This form of parent report constitutes an invaluable source of information about the child's day–today progress in the early stages of language development. However, most diary studies have been published by parent–scientists who have had extensive training in the observation of language behavior—training that the average parent lacks. Nonetheless, the average parent is quite adept at observing his or her child's behavior. Modern–day parent–report instruments have therefore been designed to capture the advantages of the diary study method in a user–friendly format.


Parent report is a component of many widely used screening tools that have been developed for English–speaking infants and toddlers. Examples include the Denver Developmental Screening Test (Frankenburg, Dodds, Fandal, Kazuk, &Cohrs, 1975), the Brigance Inventory of Early Development—Revised (Brigance, 1991), the Birth to Three Developmental Scale (Bangs &Dodson, 1979), the Receptive–Expressive Emergent Language Test, Second Edition (Bzoch &League, 1991), and the Rossetti Infant–Toddler Scale (Rossetti, 1995). Parent report is also incorporated into many structured assessments, such as the SICD–R (Hedric et al., 1984) and the PLS–4 (Zimmerman et al., 2002). Stand–alone parentreport instruments such as the CDIs and the Language Development Scales (Rescorla, 1989) are now widely used with infants and toddlers, both those who speak English and those who speak a variety of the world's languages.


Parent report has a number of inherent advantages over other means of assessing child language. First, because the parent has the opportunity to observe the child in a wide range of situations, parent report can provide data that are more representative of the child's actual language than would be observed in a laboratory sample or a structured test. Second, a child's performance in a laboratory or clinic setting may be strongly influenced by aspects of the child's personality such as shyness and by transient factors such as mood. Parent report should be less susceptible to such factors (Fenson et al., 1992/1993). As Bates, Bretherton, and Snyder (1988) emphasized, parent report reflects ‪what the child knows,‪ whereas observations at home or in the laboratory are better assessments of ‪what the child prefers to use‪ (p. 96). Third, parent report is a cost–effective means for a rapid general evaluation of language that can be valuable for screening purposes in clinical and educational settings as well as research applications. Fourth, parent report can be obtained in advance of actually seeing a child and, therefore, can help in selecting assessment procedures for more in–depth analyses. Clinicians can also use parent report as an adjunct to check the validity of their own evaluations, which would be based on a more limited interaction with the child. Finally, because parent report is based on behavior in contexts outside the clinic or laboratory, it is especially valuable in monitoring changes in language over time—including, when applicable, monitoring the progress of intervention.


Of course, parent report is not without its limitations as a means to assess language abilities (Feldman et al., 2000; Fenson, Bates, et al., 2000; Tomasello &Mervis, 1994). Parents may be accused of being biased observers, tending to either under– or overestimate their child's developing abilities, but other domains of development appear to be more vulnerable to this bias than is language. For example, Reznick and Schwartz (2001) asked mothers to rate their child's tendency to behave intentionally (using a parent–report questionnaire) and to rate their child's language development (using a CDI). One group of mothers made these ratings longitudinally when infants were 8, 10, and 12 months old, and three other groups of mothers made ratings cross–sectionally when their infants were at comparable ages. Successive ratings of intentionality were higher in the longitudinal cohort, suggesting that mothers altered their view of the infant (or their behavior toward the infant) due to the assessment experience. In contrast, CDI scores were identical for mothers completing them repeatedly or for the first time. This finding suggests that language assessment may be relatively objective. Moreover, the assessment context per se can be altered to enhance the objectivity of assessment. For example, Fenson and colleagues argued that parent report is most likely to be accurate under two conditions: 1) when assessment is limited to current and emergent behaviors and 2) when a recognition format is used. Each of these conditions places fewer demands on the respondent's memory, relative to retrospective and/or free–form reports. The recognition strategy capitalizes on the greater ease of picking out possible examples from a set, as contrasted with recalling those examples anew. The perspective underlying the CDIs is that it is better to ask parents to report on their child's vocabulary by selecting words from a comprehensive or representative list rather than to have them write down all of the words they can recall hearing their child use. The recognition format also reduces the need for parents to make inferences about the nature of the questions being asked and/or their child's level of ability relative to other children (e.g., ‪What do they mean by animal words?‪ ‪What do they mean by 'lots' of animal words?‪).




The CDIs are designed to be completed by parents, to profit from the advantages of good parent diaries, and to minimize the limitations of language samples and structured tests. Specifically, the CDIs target current and emerging behaviors, rely on recognition memory for specific information, and avoid retrospective accounts. The CDIs are easy to administer and interpret and allow researchers to collect comparable data on a small scale or a large scale. The general effectiveness of the CDIs, specifically their reliability and validity in the assessment of key language milestones, is documented in detail in the sections in Chapter 4 called ‪Reliability of the Inventories‪ and ‪Validity of the Inventories.‪


The goal of the CDIs is to yield reliable information on the course of language development from children's early signs of comprehension, to their first nonverbal gestural signals, to the expansion of early vocabulary and the beginnings of grammar. Because language changes so dramatically during the period from 8 to 30 months, it was necessary to develop two separate forms designed for different age ranges.


The CDI: Words and Gestures form, appropriate for use with 8– to 18–monthold typically developing children, generates scores for vocabulary comprehension, vocabulary production, and the use of communicative and symbolic gestures. Users who are familiar with the first edition of the CDIs will note that the upper age limit for CDI: Words and Gestures has been extended by 2 months in this second edition. The upper age limit was extended because few ceiling effects were observed at 16 months on key variables in the normative study and many clinicians and researchers are interested in obtaining information about receptive vocabulary and gesture use in children up to 18 months of age. In some cases, it may be appropriate to use CDI: Words and Gestures with children who are chronologically older than 18 months. (See the section called ‪Evaluation of Older Children with Language Delay‪ in Chapter 3 for information about the use of CDI: Words and Gestures with older children who may have delays in language development.)


The CDI: Words and Sentences form, designed for 16– to 30–month olds, yields scores for vocabulary production and a number of aspects of grammatical development, including sentence complexity and the mean length of the child's longest utterances. Because word comprehension usually grows so rapidly during the second year of life, it is unrealistic to expect parents to accurately appraise their child's vocabulary and distinguish between the words their child says and the words their child knows much beyond the middle of the second year. Hence, the vocabulary section of the CDI: Words and Sentences form is limited to expressive language. The second edition of CDI: Words and Sentences covers an age range identical to that in the first edition. The norms for forms of the CDI permit a child's scores on the major components of each inventory to be converted to percentile scores, reflecting the child's rank relative in comparison with other children of the same age and sex. The following two sections provide more specific information on each of the two forms.


Guidelines for Choosing Between the CDI: Words and Gestures and the CDI: Words and Sentences Forms in the 16– to 18–Month Age Range


In the 16– to 18–month range, CDI users have the option of choosing either the CDI: Words and Gestures form or the CDI: Words and Sentences form. Selection of one form over the other depends on the specific aims of the investigator. Here are some considerations to aid in making the best choice.




Reasons for Choosing CDI: Words and Gestures


  • If interested in gestures and/or vocabulary comprehension

    If interested in comparing children's scores on vocabulary comprehension and production

    If interested in comparing children's scores on vocabulary comprehension and production with gesture scores

    If interested in tracking a child's scores on one or more measures across the 8– to 18–month age range

  • Reasons for Choosing CDI: Words and Sentences


  • If interested in obtaining measures of early grammatical skill

    If there is reason to expect relatively large vocabularies, and therefore a need for a longer word list, or there is specific interest in a subset of vocabulary that has limited representation in CDI: Words and Gestures

  • Note: The grammatical skills measured by the CDIs are often at a beginning or preemergent stage in the 16– to 18–month period. Thus, the CDI: Words and Gestures form may offer a richer picture of communicative development in this range because most children develop comprehension vocabulary and gestures by the 16– to 18–month period.




    Reasons for Choosing CDI: Words and Gestures (at or younger than 18 months)


  • If interested in comparing a child's standing on vocabulary comprehension and/or gestures with grammatical skills beyond 18 months

    If there is reason to expect language development is proceeding more slowly than the typical rate of development

  • Reasons for Choosing CDI: Words and Sentences


  • If interested in cross–sectional or longitudinal comparisons on the same scales when testing begins in the 16– to 18–month age range and when it continues beyond 18 months


    Despite the efficiency and statistical integrity of the CDIs, the time needed to complete the form and the requirement that the parents be literate restrict the applicability of the CDIs in some research, clinical, and educational settings when a rapid assessment of a child's language level is needed. The short forms were developed to capitalize on the demonstrated effectiveness of parent input in a briefer format. These forms can be completed in less than 10 minutes and, when necessary, can be administered by interview to parents with limited or absent literacy skills (although normative data using an interactive approach has not been collected). The CDI short forms provide reliable indices of vocabulary development that are highly correlated with vocabulary scores on the full CDIs.


    Separate forms are available for 8– to 16–month–old children (Level 1) and for 16– to 30–month–olds (Level 2). The Level 1 version contains an 89–word vocabulary checklist with separate columns for comprehension and production. There are two equivalent Level 2 forms. Either can be used for a single admin– istration. In addition, a question appears at the bottom of each of these two forms, asking parents if their child has begun to combine words, with the 3 response options being not yet, sometimes, and often.


    These forms should be of value to researchers and clinicians who are seeking a quick assessment of early language. The forms may be particularly valuable when time and/or parental literacy is limited. The short forms may also prove useful in longitudinal studies, in which repeated administration of the full inventory may be impractical. For children in the 16– to 30–month age range, investigators have the option of alternating between the two equivalent forms to reduce effects accruing from repeated administration of the same items. The forms may also be useful for children with developmental delays who are beyond the specified age ranges for typically developing children.


    Because the short forms include only vocabulary measures, and a much briefer list than in the full forms, they are not likely to be as precise as the full CDI. For example, the word list is too brief to be useful for studies of vocabulary composition (Bates et al., 1994). Also, the brevity of the list exacerbates the ceiling effect after 27—28 months, especially for children with higher abilities. Finally, the Level 2 forms do not include any measure of grammatical development following the emergence of word combinations. Researchers and clinicians should carefully consider their goals in language assessment prior to selecting a particular form. Detailed information about the CDI short forms and normative data can be found in Fenson, Pethick, et al. (2000).