The specific aims of this project are to investigate empirically the effect that alcohol advertising has on drunk driving. The empirical work in this project is designed to produce estimates of the effect that alcohol advertising has on highway fatality rates for various age and sex categories. Variables measuring television and radio advertising of beer and wine and variables measuring billboard, newspaper and consumer novelty advertising of all alcoholic beverages will be included. The empirical work will also result in new estimates of the effect of alcohol price, and drunk driving deterrence laws on highway fatality rates, and seasonal pattern of highway fatality rates. The focus on highway fatalities is important since highway fatalities are a leading cause of death in the United States. Several public interest groups have asserted that alcohol advertising leads to increased alcohol consumption and increased drunk driving. Separate estimates of the effects of alcohol advertising by age are also important since adolescents may be particularly effected by alcohol advertising. These estimates will be generated by a series of multiple regression models. The data set will be pooled, quarterly time series data, for the years 1985 through 1988, for 75 cities and towns in the United States. The fatality data come from the Fatal Accident Reporting System, the television advertising data come from the Arbitron Ratings Company, the radio advertising data come from the Radio Expenditure Reports the billboard advertising data come from Leading National Advertisers, the newspaper data come from Media Records and the price data come from the American Chamber of Commerce Research Association. These data are both timely and representative of the nation as a whole. The alcohol advertising data are widely used in the advertising industry and represent the most comprehensive measures of alcohol advertising that can be collected.
