"Oh, please please do you have the source for that? Not that i doubt you; i definitely don't. Gish always says he is a better driver than most people on the road, and i always say Everyone thinks that. I would love to have actual numbers from a qualified source."
The problem is that this study is just apocryphal, as no such study was ever done so nobody can cite it. I just tell them it comes from the Institute where they figured out (in a different study, of course) that 84.7% of Internet statistics are made up on the spot.
The 97% (or 93% as I've heard it) statistic seems to be apocryphal, but the phenomenon is real. The cognitive bias in which the vast majority of people tend to rate themselves as above average in any number of domains is called illusory superiority, the Dunning-Kruger Effect, or the Lake Woebegone Effect. Some studies have looked specifically at driving and found that anywhere from 60-90% of people consider themselves above average drivers, either in certain ways (e.g., speed, number of violations) or overall. Here are a few references:
Svenson, O. (1981). Are we all less risky and more skillful than our fellow drivers?. Acta Psychologica, 47(2), 143–148.
McCormick, I. A., Walkey, F. H., & Green, D. E. (1986). Comparative perceptions of driver ability— A confirmation and expansion. Accident Analysis & Prevention, 18(3), 205–208.
Delhomme, P. (1991). Comparing one's driving with others': Assessment of ability and frequency of offenses. Evidence for a superior conformity of self-bias? Accident Analysis & Prevention, 23(6), 493-508.
Bathurst, J., & Walton, D. (1998). An Exploration of the Perceptions of the Average Driver’s Speed Compared to Perceived Driver Safety and Driving Skill. Accident Analysis and Prevention, 30, 821-830.