摘要：This paper examines the claim of the New Zealand Land Transport Safety Authority, that the reduction in the road toll in New Zealand from October 1995 to mid-1996 was predominantly due to a television advertising campaign conducted over that time. The method of analysis was similar to that used to evaluate a similar, successful campaign conducted earlier in Victoria, Australia, on which the New Zealand campaign was modelled. Data on campaign advertising exposure, and other variables thought to be associated with drink-driving behaviour, were subjected to regression analysis to measure the relationship between the incidence of the campaign advertisements and positive evidential breath tests. At best, only a tenuous relationship was found. This result may be because there is no strong relationship. Or, it may be because of the inherent difficulty of extracting useful information from non-experimental data of this type. Or, it may be due to the weak linking of the campaign to enforcement activities; the Victorian study found that it was important for the success of an advertising campaign, that it be linked to enforcement
关键词：drink-driving campaigns; effectiveness; advertising; television; New Zealand; road safety.
