RELATION OF HIGHWAY ACCIDENTS TO SHOULDER WIDTH ON TWO-LANE RURAL HIGHWAYS IN NEW YORK STATE

THE RELATION BETWEEN SHOULDER WIDTH AND ACCIDENT RATES ON TWO-LANE RURAL STATE HIGHWAYS IN NEW YORK WERE INVESTIGATED. CORRELATION OF DATA INDICATED 9299 ACCIDENTS DURING 1952 ON 8746 MILES OF TWO-LANE RURAL HIGHWAYS (16-24 FEET IN WIDTH). VARIALBLES WERE: (1) PAVEMENT WIDTH, 18-18 FOOT, 20 FOOT, 22-24 FOOT, (2) YEAR OF CONSTRUCTION OR RECONSTRUCTION, PRE-1944, 1946-1952, AND (3) TRAFFIC VOLUME. ACCIDENT DATA WAS GROUPED AS (1) FATAL AND INJURY, AND (2) PROPERTY DAMAGE. STATISTICAL DATA IS PRESENTED IN TABLES AND GRAPHS. PAVEMENT WIDTH AND PERIOD OF CONSTRUCTION OF THE HIGHWAY DID NOT APPEAR TO BEAR ANY DEFINITE RELATION TO ACCIDENT EXPERIENCE. EVEN THOUGH A NUMBER OF STUDIES SHOW THAT TRAFFIC VOLUME IS RELATED TO ACCIDENT RATES, NO SUCH DEFINITE RELATION IS FOUND HERE. CONSIDERING THE ENTIRE SYSTEM OF RURAL TWO-LANE ROADS IN NEW YORK STATE, THERE EXISTED A MEASURABLE RELATION BETWEEN SHOULDER WIDTH AND ACCIDENT RATE. THE WIDER THE SHOULDER, THE LOWER WAS THE ACCIDENT RATE. THIS IS ESPECIALLY TRUE FOR PROPERTY DAMAGE ACCIDENTS. THE RESULTS ARE DEPENDENT ON AN EQUITABLE DISTRIBUTION OF INTERSECTIONS, CURVES, AND GRADES IN EACH ANALYTICAL GROUPING.