CHANGES IN THE INCIDENCE OF DRUNK DRIVING IN THE UNITED STATES, 1973-1986

The two surveys were very similar in methodology. They were conducted on Friday and Saturday nights from 10 p.m. to 3 a.m., using separate roadside sites from 10-12 p.m. and from 1-3 a.m. The survey locations along heavy and medium volume roads (except freeways) numbered 185 in 1973 and 182 in 1986. At all locations drivers from the passing traffic stream were stopped in a random procedure by a law enforcement officer and sent to the civilian interviewer at the road shoulder or in an adjacent parking area. In 1973 most interviews took place in the interviewer's motorhome, van, or car, but in 1986 all interviews took place at the driver's car window. In 1986 four localities required the survey to take place within a larger sobriety checkpoint operation. Participation in these non-official surveys was necessarily voluntary. The interviewer had to persuade the driver to answer the interview questions and to provide a breath sample at the end. In 1973 a satisfactory breath sample was obtained from 3192 (86.3%) of the 3698 stopped motorists. In 1986 2850 (91.9%) of the 3100 stopped drivers provided a satisfactory breath sample. In both surveys the interviewers made an estimate of the drinking condition of all selected drivers, both participants and refusers, in order to try to see if there was any refusal bias. In 1986 this judgment was aided in most cases by