Research ~ll,.l Stati,,tic_., Section. ('alit'ornia Department ol Motor Vehicles. Sacramento. Calit'ornia. U.S.A.
D~P, ER record data on types of violations is a readily available source of information on driving behavior in our present state of knowledge. Direct observation of driving behavior is a preferred-source of information, but there is little available at the present time. Driver record data cannot only provide knov, iedge of driving behavior useful in program development. but can point up areas worthy of more direct, detailed observation.
A few research projects have been done of spot observations of driving behavior related to violations. Women made,turn signals a greater proportion of the time than males (Blomgren, Scheuneman and Wilkins, t963: Batch, 1957). Normann (1944) found that generally men and youths broke the speed limits more frequently than women and older persons, respectively. Edwards and Hahn (1964) are conducting an extensive study of filmed driver behavior and accidents. White male drivers in Washington. D.C., were t)llowed l'c,r 5 rain. and, unknown to the drivers, a film was made ot" their behavior. The films ,,,,ere scored for errors. Speeding was committed by 87 per cent of the drivers, changing lanes ,,vithout signalling by 80 per cent. thilure to stay in lane by 63 per cent, and turning without signalling by 46 per cent, with smaller percentages for other errors.
Other studies used driver record data from motor vehicle department files. The Oregon Department of Motor Vehicles (1963) tabulated the type of most recent violation for approximately 13.000 drivers. Violations were grouped into four categories: (1) speed~ (2) hazardous. (3) equipment, and (4) miscellaneous. As age increased, speed and equipment violations comprised a decreasing percentage of all violations, while hazardous violations comprised :.m increasing percentage of all violations. Speed and equipment violations comprised a higher percentage of all male violations than they did of all female violations, while hazardous violations comprised a greater percentage of all female violations than of all male violations. Kaestner (1964) obtained similar results for a sample of male drivers involved in lhtal accidents. Waller and Goo (1969) obtained similar results for samples of drivers v,,ith and without chronic medical conditions.
The relationship between types of violations and accidents is discussed in Parts 7, 8, and 9 of the California Driver Record Study.
The present study t'orms an extension of the previous research in that it presents means and mileage-adjusted means for many types of violation for a large sample analyzed by age. sex, and marital status. In addition, comparisons are made with violations resulting from an accident investigation. METHOD All 147,984 drivers whose license number ended in 00 or 01, and who were licensed for the three years preceeding the date of coding were selected. The data were collected between September. 1963, and March, 1964. Violations were included which occurred within the Mental Hygiene. for writing the computer program for the analysis of variance. The results and interpretations are those of the authors, and not necessarily those of the California Department of Motor Vehicles.