In April 1979 the Albuquerque Police Department began a special enforcement program aimed at reducing DWI accidents. The program consisted of a squad of police officers who manned a van which housed a Breath Alcohol Testing (BAT) device. The data collected to evaluate the program consisted of a quarterly series on night-time mid and week end traffic injuries and fatalities and a proxy for expo- sure to accidents, millions of gallons of fuel consumption in the Albuquerque area. The first 29 quarters beginning in 1972 were used as a control or "before" period, and the last 23 quarters ending with 1984 comprise the experimental or "after" period. A number of analyses could be suggested for evaluating the effects of the program. Some are: 1. Do a "before-after" test of difference in average injuries and fatalities per million gallons of fuel consumed. 2. Regress accidents and fatalities on fuel consumption, seasonal indicators, and a program dummy variable. 3. Sponsors felt that the effect of the Batmobiles may well be cumulative, because knowledge about the program becomes dispersed and acts as a deterrent to driving while intoxicated. This hypothesis can be incorporated by replacing the (0,1) program variable by a variable which takes on "0" for no program and a time sequence or "ramp" function with the values 1,2,3,...,23 for the experi- mental period. A further modification of the regression approach in (2) is to use a trend variable (1,2,3, ...52) to adjust for effects not accounted for by fuel consumption, such as increasing fuel efficiency of automobiles leading to more accident exposure per gallon of fuel consumed.