The injury reducing effectiveness of child safety seats in frontal crashes was evaluated, based on 36 frontal or oblique sled tests run with two or more GM three-year-old dummies in the simulated passenger compartment of a car. Unrestrained, correctly restrained and incorrectly restrained dummies were tested at the range of speeds where most nonminor injuries occur (15-35 mph). Accident data from NHTSA files were used to calibrate a relationship between the front-seat unrestrained dummies' HIC and unrestrained children's risk of serious head injuries;​ also between torso g's and the risk of serious torso injuries. These relationships were used to predict injury risk for the restrained children as a function of crash speed and to compare it to the risk for unrestrained children. The sled test analysis predicted that the 1984 mix of correctly and incorrectly used safety seats reduced serious injury risk by 40 percent relative to the unrestrained child, in frontal crashes. This prediction is consistent with the effectiveness that has typically been found in statistical analyses of highway accident data. The sled test analysis also predicted that correctly used seats reduce injury by an average of 61 percent, in frontal crashes.

This paper presents the views of the authors and not necessarily those of the National Highway Traffic Safety Administration.