The TAD vehicle damage rating scale was field tested on a small scale for several months in North Carolina. A test of inter-rater reliability for highway patrolmen using the TAD scale showed there is relatively good agreement among raters as to damage type and severity level. Problems in the use of the scale are noted. A second test involved psychological scaling of the TAD manual pictures and showed that the various TAD scales lacked the desired scale characteristics of equal appearing intervals. A sample of 1329 accident reports was obtained and comparisons made between damage ratings, speed, and estimated damage cost adjusted for vehicle age as predictors of driver injuries. The tests showed that speed is less effective a predictor than either the TAD scale or cost. Cost estimates were as good in predicting driver injury as the TAD scale under some conditions, but the TAD scale proved superior for discriminating serious injuries from minor (or no) injuries.