Parts of the Body

The Parts of the Body assessment was completed by age-eligible NLSY79 children in 1986 and 1988 only.  Developed by Jerome Kagan of Harvard University, Parts of the Body attempts to measure a one- or two-year-old child's receptive vocabulary knowledge of orally presented words as a means of estimating verbal intellectual development.  The interviewer names each of ten body parts and asks the child to point to that part of his or her body.

Scoring Parts of the Body

The child's score is computed by summing the items that a child correctly identifies (C07972. for 1988 and C05799. for 1986). Thus, a minimum score is 0 and a maximum score is 10.  No proration was attempted since the later items in the sequence are more difficult than the earlier items. A Spanish version of this assessment was available for use with young Hispanic children. A complete protocol for the Body Parts assessment can be found in Section 3 of the 1986 Child Supplement and the 1988 Child Supplement, available on the Questionnaires page.

Because of inconsistency in the way some interviewers interpreted the instructions, the Body Parts assessment was scored in 1988 using three alternate criteria.  First, a child had to answer each of the ten items either (1) correctly or (2) incorrectly on at least one of the two attempts (see page CS-18 in the 1988 Child Supplement). If scoring was completed according to this criteria, then the case was coded a "1" on the Body Parts scoring criteria flag (C07973.00). A second, less restrictive criterion, allowed some of the individual items to be coded "3" (no answer) on some of the attempts. For this subset of children, a code of "3" was treated as an incorrect response and the overall assessment was accordingly scored. These cases can be identified by a value of "2" on the Body Parts criteria flag. Children for whom virtually all the responses were coded "3" (and translated into incorrect responses) received a value of "3" on the Body Parts criteria flag.  Thus, users may restrict analyses to the more constrained sample or opt to include only children who had been scored according to the less conservative definitions. As with all the assessments, users who plan to use a particular assessment are strongly urged to evaluate the scoring schema and data quality according to their own criteria. While we have made every effort to create scores that are faithful to the intentions of the assessment designers, there are instances where researchers could reasonably disagree about what precise scoring procedures should be utilized. The Body Parts assessment was given to very young children for whom there could be considerable ambiguity in differentiating between "incorrect" and "nonresponses."

Age Eligibility for Parts of the Body

In the 1986 and 1988 Child surveys, Parts of the Body was administered to children between the ages of 1 and 2 years.

Norms for Parts of the Body

As no appropriate national norms are available for scoring this assessment, CHRR provided (for 1988) internally normed standard and percentile scores. No normed results are provided for 1986. As the raw score on this assessment is extremely sensitive to the age of the child, users of the raw scores are encouraged to apply appropriate techniques that permit analytical comparisons of children across different ages. When controlling for age, the user should select the appropriate Child Supplement age variable that specifies the child's age (in months) as of the Child Supplement interview date.

Completion, Validity, and Reliability for Parts of the Body

Notwithstanding the availability of a Spanish version of this assessment in the NLSY79, the user should proceed cautiously when interpreting its reliability and validity, particularly with regard to minority and relatively more disadvantaged children. It appears that a child's score may be quite sensitive to the child's English language capabilities as well as rapport with the interviewer. In 1986, the non-completion rate for this assessment was about 17 percent. For about half of the completed assessments, a child is reported to have not responded on at least one question, requiring the alternate assumptions with regard to scoring we describe above to be made. For a more complete discussion of the reliability and internal validity of this assessment and the Memory for Location assessment, please see pages 30-31 in The NLSY Children, 1992: Description and Evaluation, available on the Research/Technical Reports page.