Measurement of Theoretical Constructs and the Use of Data Drawn From the National Youth Survey

Daniel R. Lee, Indiana University of Pennsylvania
Gaylene S. Armstrong, Arizona State University - West
Todd A. Armstrong, Arizona State University West

The analysis of self-reported data began to emerge as a popular means to investigate the etiology of crime and delinquency during the 1960s. The National Youth Survey (NYS) has become a prominent example of a large-scale collection of self-reported data and has been used to support or refute many criminological theories. The public availability of several waves of annual data from the NYS via the Inter-University Consortium for Political and Social Research has added to the prolific use of these data. Along with measuring specific (e.g., violent, property, drug, status) and general (i.e., scaled) forms of delinquency, these data have been used to construct a wide array of independent variables that seek to measure and test the theoretical constructs related to a variety of individual and integrated criminological theories (e.g., developmental/life course, differential association/social learning, labeling, social control, strain). This analysis reviews nearly 100 empirical assessments of data drawn from the NYS in an attempt to clarify the reliability and validity of theoretical constructs as measured by individual and scaled survey items. Our argument is that the extensive use of these data has led to an imprecision in measurement and, subsequently, empirical conclusions that could be, at times, diluted, less than reliable, and invalid. Implications for further analyses of these and other publicly available data are discussed.

(Return to Program Resources)

Updated 05/20/2006