THE INTERRUPTED TIME-SERIES IS A QUASI-EXPERIMENTAL DESIGN FOR STUDYING THE EFFECT OF A GIVEN TREATMENT ON A VARIABLE MEASURED REPEATEDLY OVER A PERIOD OF TIME BEFORE AND AFTER THE APPLICATION OF THE TREATMENT. LIKE ALL QUASI-EXPERIMENTAL TECHNIQUES, IT SUBSTITUTES FOR THE TRUE EXPERIMENT WHICH CANNOT BE CONDUCTED UNDER CIRCUMSTANCES IN WHICH SUBJECTS CANNOT BE ASSIGNED RANDOMLY TO EXPERIMENTAL AND CONTROL GROUPS. VARIOUS THREATS TO THE VALIDITY OF THE INTERRUPTED TIME-SERIES DESIGN ARE DISCUSSED. THE EFFECT OF THE BRITISH ROAD SAFETY ACT OF 1967 ON CAR ACCIDENTS WAS MEASURED. THE ACT POSSESSED FEATURES PROVIDING FOR A RELATIVELY STRINGENT BLOOD ALCOHOL LEVEL, AN ON-THE-SCENE BREATH TEST, AND MANDATORY SANCTIONS. DATA-GATHERING PROCEDURES, THE APPLICATION OF THE TIME-SERIES DESIGN, AND AN EXAMINATION OF THE THREATS TO THE VALIDITY OF THE STUDY ARE DISCUSSED. THE ONLY SERIOUS CONTENDERS TO THE HYPOTHESIS THAT THE LEGISLATION SAVED LIVES AND PREVENTED INJURIES INVOLVED SIMULTANEOUS EVENTS THAT MIGHT BE EXPECTED TO HAVE A SIMILAR EFFECT. EACH OF THESE RIVAL HYPOTHESES WAS EXAMINED AND DISMISSED AS A PLAUSIBLE EXPLANATION FOR THE REDUCTION IN ACCIDENTS AT THE TIME OF ENFORCEMENT OF THE LEGISLATION. IT IS CONCLUDED THAT THE LEGISLATION DID SAVE LIVES AND PREVENT INJURIES AND THAT IT CONTINUES TO HAVE AN IMPORTANT BENEFICIAL EFFECT ON BRITISH HIGHWAYS. IT IS BELIEVED THAT THE METHOD OF EVALUATION USED CAN BE REPLICATED IN EVALUATING THE EFFECT OF SOCIAL CHANGES GENERALLY AND LEGAL CHANGES IN PARTICULAR. (RCB)