Abstract
THIS PAPER FOCUSES ON AN INDIVIDUAL'S DECISION TO DRIVE OR NOT TO DRIVE AFTER DRINKING. TO EVALUATE THIS DECISION, A UTILITY MAXIMIZING PROBABILISTIC CHOICE MODEL IS SPECIFIED AND ESTIMATED USING A SAMPLE OF COLLEGE STUDENTS. THE ESTIMATION RESULTS PROVIDE INTERESTING INSIGHTS RELATING TO THE POTENTIAL EFFECTIVENESS OF DRINKING-DRIVING COUNTERMEASURES, AND SUGGEST THAT THE MOST EFFECTIVE METHODS OF REDUCING THE PROBABILITY OF DRIVING AFTER DRINKING ARE THOSE ADVERTISING AND AWARENESS CAMPAIGNS THAT FOCUS ON ALTERING INDIVIDUAL PREFERENCES.(Author/publisher).