LOGISTIC REGRESSION IN COMPARISON OF THE PASSIVE SAFETY OF CAR MODELS

In the study logistic regression is applied to a comparison of the passive safety of individual car models. The study examines a total of 4973 drivers injured in the case car. For comparison logistic regression is applied to two sets of data. One consisting of all crash accidents between two passenger cars in private use, and the other representing crash accidents of two passenger cars in private use where at least one driver has been injured. Comparison of these variables shows errors in the comparison of the passive safety of car models' that have been caused as a result of the different basis for inclusion of accidents in the statistics. According to the study it looks as though accidents which have led to the injury of at least one driver have better results in some car models than when all accidents are examined in relation to most car models. However the results given by both methods are similar.