Development of a Computer-Assisted Self-Interview

Government and private sector organizations need to collect large amounts of detailedinformation from a variety of individuals, on a variety of topics, in a variety of settings, for a variety ofreasons, in a timely and cost-effective manner. These last two constraints have spurred thedevelopment or adaptation of computer-assisted measurement technologies. Implicit to this massive datacollection and consumption is its accuracy (reliability): inaccurate data have zero, or negative, utility.However, most organizations, including research organizations, underestimate the cost of "good" data,substituting face-valid assumptions about its collection for more rigorous procedures. We willsystematically explore and delineate those factors which assist or hinder acquisition of data of varyingsensitivity from substance abusing individuals, identify potential solutions and strategies to overcomeobstacles, to develop an affordable computer-assisted self-interview (CASI) of substance usage, and toevaluate the quality of the data collected by this prototype (in terms of reliability and criterion validity)with that collected by a more traditional method -- a face-to-face interview -- among relatively lowfunctioning individuals.