This paper examines experimental field research on drinking and driving, with emphasis on the evaluation of intervention programmes, and comments on Perrine's conceptual model. Considerable research has been conducted recently on: (1) criminal justice and treatment programmes, designed to reduce impaired driving by individuals convicted of drinking and driving (special deterrence); (2) drunk-driving laws and law enforcement designed to produce general deterrence among drinking drivers. Despite increased interest in and funding of this research area, there have not been very many credible studies of laws about impaired driving. Compared with laboratory research, the evaluation of laws and enforcement programmes presents several difficulties, including assessment of the effect of economic conditions on the extent of both drinking and driving. Because of the complexity of such external factors, field research has emphasised controlling them by using quasi-experimental designs and time series analysis. However, researchers often overlook the need to demonstrate endogenous relationships within a programme, that provide a credible basis for assuming a causal relationship between the programme and outcome measures. The creation of a `chain of action' makes research results much more credible and useful.
Abstract