NCJ Number
76160
Journal
Accident Analysis and Prevention Volume: 12 Issue: 1 Dated: (1980) Pages: 55-65
Date Published
1980
Length
11 pages
Annotation
A performance test that could increase police efficiency in detecting drivers under the influence of alcohol was assessed on the basis of its ability to differentiate sober and impaired drivers.
Abstract
The performance test required drivers to align the middle of their vehicles to a series of lines marked in or near the middle of the lane, while maintaining a prescribed driving speed. Three levels of impairment were used: sober, just over the legal impairment level, and half the legal impairment level. Subjects were asked not to eat or drink after the noon meal and were given an orange drink with or without vodka at 6 p.m. Subjects participated in the test on a test track with no other vehicles present. Results showed that 30 percent of the impaired drivers failed the test, while only 1 percent of the sober drivers failed. Using more stringent fail criteria, under which failure on any one component of the test represented failure on the total test, 66 percent of the impaired drivers failed while 23 percent of the sober drivers failed. Results clearly showed that such a performance test is possible. Use of such a test must balance the incovenience to the sober driving public against the rate of detection of impaired drivers. Findings may also be limited by the nature of the test conditions, which are not similar to field conditions. A field evaluation of the test is recommended. Tables, figures, and a list of 16 references are provided. (Author abstract modified)