Using structural equation modeling (LISREL 8.71), drivers' everyday speeding behavior was predicted using the theory of planned behavior as a frame of reference. One hundred and twelve test drivers had a device installed in their vehicles that continuously logged their speeding behavior in a large area under an extended period of time. The test drivers also completed a questionnaire including questions inspired by the theory of planned behavior. It was found that the independent variables stipulated in the theory afforded a level of prediction of drivers' self-reported speeding as well as of their logged speeding. Attitude towards speeding, subjective norm, and perceived behavioral control were significant determinants of self-reported speeding. Self-reported speeding, and subjective norm, but not perceived behavioral control, did then contribute to the prediction of drivers' logged speeding. The fact that perceived behavioral control did not directly contribute to the prediction of drivers' logged speeding may be due to the possibility that drivers with several years of experience already take into account the actual control they have over the target behavior. As the theory of planned behavior can be used as a frame of reference to predict drivers' everyday speeding behavior, it is suggested that the drivers might decide on a target behavior and in living up to this decision they continuously monitor their target speed during everyday driving.