No abstract

The CAGE alcohol screening tool has been touted as a good choice for clinical settings because of its brevity. We administered the CAGE and three other alcohol screening instruments (the Short-MAST, AUDIT, and JELLINEK) by microcomputer to 296 clients at a drinking driver treatment program and three of the four scales to a second sample of 270 clients from six drinking driver treatment programs. The average response times for the CAGE were 31 and 32 sec, respectively, in the first and second samples. The average response time for the JELLINEK was approximately five times longer than it was for the CAGE; response time for the AUDIT averaged four times longer; and response time for the Short-MAST was two and a halftimes as long. The estimated reliability of the CAGE was the lowest and its standard error of measurement was the highest of the four scales. We recommend the Short-MAST as the tool of choice if the extra minute of administration time it requires in comparison with the CAGE is not critical.

No abstract

The Microcomputer Assessment System (MAS), written in C language, allows a researcher to design and administer computer-based surveys by manipulating ASCII text files and requires no knowledge of a traditional programming language. The system was designed to provide maximum user flexibility: MAS allows both multiple-choice and open-ended (fill-in) questions and allows control over text color, allowable answers, and summary output. MAS also assesses response latency and includes support for random assignment of different instruments to respondents. MAS runs under MS-DOS 3.0 or higher, requires 640K of RAM, a color screen (EGA or better), and an extended keyboard. Preliminary results from an application of MAS to a sample of 121 clients at an impaired-driver treatment program are summarized.Self-reports of alcohol use have widespread application in impaired-driver treatment programs because other satisfactory and practical measures are generally unavailable (cf. Sobell, Maisto, Sobell, & Cooper, 1979; Williams, Aitken, & Malin, 1985). Alcohol-use self-reports may be influenced by the nature of response options offered (Hesselbrock, Babor, Hesselbrock, Meyer, & Workman, 1983;Poikolainen & Karkkainen, 1985), the nature of the setting in which they are given (Gfroerer, 1985), and whether reports are about use of alcohol in specific settings or generic use (Hilton, 1986). Babor, Stephens, and Marlatt (1987, p. 420) concluded that "alcohol research would benefit from a new generation of methodological research directed at procedures that enhance the validity of verbal report data.... The use of computerized assessments may yield more valid data ... than other techniques." Clearly, there is a need to examine altematives to traditional paper-and-pencil surveys that will minimize errors of measurement of alcohol use (Allen & Skinner, 1987;Skinner & Allen, 1983).We have developed a data-collection program that allows a researcher to design and administer computerPreparation of this article was supported in part by Grant 07852 from the National Institute on Alcohol Abuse and Alcoholism. The opinions expressed are those of the authors and do not necessarily reflect the views of RAND or the sponsor. The authors appreciate the comments of five anonymous reviewers on an earlier version of this manuscript. Thanks are due to Karen Boseker for outstanding secretarial support. Correspondence concerning this article should be addressed to R. D. Hays, RAND, 1700 Main Street, Santa Monica, CA 90407-2138. based surveys by manipulating ASCII text files and that requires no knowledge of a traditional programming language. This paper summarizes the development of the program and preliminary results of its use in a pilot sample of 121 clients at a treatment program for persons convicted of impaired driving. FEATURES OF NUCROCOMPUTER ASSESSMENT SYSTEMBecause of the flux in the state of the art of microcomputer programs for survey administration and our desire to tailor a program to our specifications and ensure quality control, we dec...

No abstract