On the Representativeness of Norming Samples for Aptitude Test
Abstract
This paper discusses the extent to which a sample intended for use in norming aptitude tests must be representative of the underlying population. The work is in support of the Defense Manpower Data Center (DMDC) and its planned use of data from the National Longitudinal Survey of Youth (NLSY97) as a norming sample for aptitude tests. We examined data from a previous national administration of aptitude tests to a representative sample of youth known as Profile of American Youth (PAY 80). We regressed aptitude test scores on demographics and concluded that: * Norming sample for aptitude tests must be representative of the target population with respect to age, race"ethnicity, gender, respondent's education, and mother's education. * It is not necessary that the sample also be representative with respect to number of siblings in the household, degree of urbanization, or census region. Although these factors may be correlated with aptitude scores, if the other five variables are representative, then these factors need not be representative.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2003
- Accession Number
- ADA421075
Entities
People
- Catherine M. Hiatt
- William H. Sims
Organizations
- Center for Naval Analyses