Three federally funded surveys of almost half a million drivers and right front passengers of cars and minivans in Ohio provided extensive data on shoulder belt use. Shoulder belt use was found to be significantly related to three variables of special interest:​ the type of roadway;​ average daily traffic;​ and posted speed limit of each of the 399 different observation sites in each of the three surveys. Multiple regression analysis demonstrated that, of the three independent variables considered, 90 percent of the multiple r² was attributable to the relationship between speed limit and the rate of belt usage.