NCJ Number
104580
Journal
Journal of Quantitative Criminology Volume: 2 Issue: 4 Dated: (December 1986) Pages: 293-327
Date Published
1986
Length
35 pages
Annotation
Several potential problems with the reliability and validity of self-report delinquency measures warrant procedures and additional research to improve these measures.
Abstract
A measuring instrument's reliability is its precision level, which refers to the extent to which the instrument would produce identical scores if it made multiple measures of the same object. Some reliability problems with self-reported delinquency measures are associated with the use of internal consistency as the reliability measure and the precision level required for different types of analyses. A test's validity is its ability to measure what it intends. Potential validity problems with delinquency self-reports are related to their content validity, overreporting and underreporting, the use of official records as a validity check on self-reports, and the lack of a reliable criterion for assessing the empirical validity of self-report measures. Given these potential problems with self-report measures of offending, bounding and other recall aids should be routinely used with self-report delinquency measures, as should indepth followup questions. Particular attention should also be given to item selection, wording, and scale construction. Additional research is required to improve the reliability and validity of self-report measures and to identify the circumstances associated with overreporting and underreporting. 4 tables and 69 references. (Author abstract modified)