The increase in mobile phone use worldwide raises concerns that using mobile phones while driving increases accidents.This result means that previous estimates of the impact of mobile phone use on population risk, based solely on accident samples, may be overestimated by approximately a third.This article differs from previous research in two important ways: first, we use a larger sample of individual-level data; and second, we test for selection effects, such as whether drivers who use cell phones are inherently less safe drivers even if they are not on the phone.