Using Multitrait-Multimethod (MTMM) Techniques to Examine the Convergent and Discriminant Validity of Social Disorder

• In Event: Do We See the Same Things? Methodological Issues in the Study of Social and Physical Disorder

Abstract

The classic multitrait-multimethod (MTMM) approach was used to explore the reliability, convergent and discriminant validity of social disorder, prostitution and drug activity across resident interview survey, systematic social observations, and police calls for service data collected in Jersey City, NJ. We found support for the reliability of all measures across data sources, except for observed drug activity. In terms of validity, prostitution and drug activity both showed convergent and discriminant validity, with prostitution performing slightly better. However, there was little evidence that social disorder was a valid measure across the three methods. When evaluating merits of the data sources, calls for service data seem to be the most reliable and correspond to social observations the best, while measures drawn from perceptual data show the highest amount of method-driven association.

Authors

• Sue-Ming Yang, George Mason University

• Josh Hinkle, Georgia State University