MULTIPLE MEASURES AND THE VALIDITY OF RESPONSE IN RESEARCH ON DRINKING DRIVERS

Drivers were stopped on city streets on weekend nights in a large midwestern city. In the course of a 5-minute interview, four measures of alcohol consumption were made. Interviewers indicated a personal initial reaction as to whether the driver had been drinking. Second, the interviewers attempted to observe the presence of positional alcohol nystagmus by asking driver to tilt their heads. Third, driver were asked if they had been drinking alcohol. Finally, a breath sample was taken and later analyzed for blood alcohol concentration (BAC). Analysis of the results, using BAC as a criterion measure, indicates that interviewers were conservative in their estimates, making a high percentage of false negatives, and conservative in their observations, indicating "can't tell" most often in positive cases. A high percentage of drivers gave false negatives although false positives also occurred. Situational factors are suggested as possible reasons for invalid responses. (Author)

  • Availability:
  • Corporate Authors:

    National Safety Council

    444 North Michigan Avenue
    Chicago, IL  USA  60611
  • Authors:
    • Zusman, M E
    • Huber, J D
  • Publication Date: 1979

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00302473
  • Record Type: Publication
  • Source Agency: National Safety Council Safety Research Info Serv
  • Files: TRIS
  • Created Date: Dec 29 1979 12:00AM