Statements of drivers are often used to gain insights into accident causations, whose understanding is an important factor for the development of effective prevention strategies. The coming into effect of the General Safety Regulation 2019 led to an increasing availability of objective vehicle data in the form of Event Data Recorders (EDR) in Germany as well as the whole European Union. This creates the possibility to examine the accuracy of the driver statements in more detail. Among others, maladjusted vehicle speed is an important crash contributing factor. Therefore, a good understanding about the quality and applicability of the drivers' reported speed can be important. The goal of the present study is to examine the accuracy of speed reports from German drivers in the context of accident research. To this end, the reported speeds and speed violations were evaluated with respect to their consistency with the EDR data. Additionally, it was investigated whether there is a relationship between the accuracy of the statements and the role in causing the accident on the one hand and the time elapsed between the accident and the report of the driven speed on the other hand. Based on data from the Audi Accident Research Unit (AARU), this study compares drivers' self-reported speeds prior to an accident obtained by standardized telephone interviews with the respective recorded EDR speeds of the crash vehicles. It was shown that driver reported speed violations significantly less often and to a smaller extent than they were committed based on the EDR data. The reported speeds were significantly lower than the recorded speeds from the EDR data. Furthermore, this effect was significantly stronger for the other accident participants, which tended to underestimate their speed more than the accident causers. This result might be explained by a certain group of other accident participants which were driving very fast (> 200 km/h) on the motorway, as is elaborated in detail in the discussion. There was no significant correlation between the accuracy of the reported speed and the timing of the respective interviews. In conclusion, the drivers’ statements regarding the driven speed were found to be relatively inaccurate, which is why they can only be used with caution and in conjunction with more reliable data. Still, the reported speed can be valuable to evaluate the general plausibility of the drivers' statements concerning the accident.