The Instrument Landing System (ILS) is used to guide aircraft on final approach and is by far the most important and frequently used guidance system for landings [EUROCONTROL, 2008]. However, this system is both expensive and inflexible because it is only able to provide a straight-in approach trajectory for one runway end, so that multiple installations for one airport are required. The successor of this system is expected to be the Global Navigation Satellite Systems (GNSS) with augmentation systems such as SBAS (Satellite Based Augmentation System) or GBAS (Ground Based Augmentation System). Such systems are able to provide safe and reliable guidance with greatly improved flexibility in the definition of approach tracks. The use of SBAS has been proposed to complement ILS for Autoland CAT-I and CAT-II. No new airport infrastructure implementation will be needed as long as the airport is within the service area. Standalone single -frequency GPS is mostly affected by the ionospheric delay and the satellite clock and ephemeris errors. SBAS is therefore providing to the user corrections and integrity information for these three error sources. The satellite clock offset and the ephemeris errors, which are the subject of this article are corrected by Fast Corrections and Long-Term Corrections. The Fast Correction message conveys the PseudoRange Correction (PRC) for each satellite of the current satellite mask. The concerned SBAS message types also contain the UDRE (User Differential Range Error) parameter that expresses a bound on the standard deviation of the correction residual error, providing integrity bounding and used to compute a protection level. The PRC and the computed Range Rate Correction (RRC) are correcting the component of the satellite clock delay that is quickly varying. The Fast Correction is then the sum of the current PRC and the computed RRC multiplied by the elapsed time. The Long -Term Correction conveys the satellite position correction and the slowly varying component of the clock error. These 2 corrections are improving the user position accuracy. In order to quantify this improvement, we propose to study the SBAS residual clock and ephemeris errors in the position domain. We propose in this article a protocol to evaluate the nominal residual clock and ephemeris errors remaining after applying the corrections, reflecting SBAS accuracy. Our methodology uses broadcast GPS navigation data and broadcast SBAS correction messages compared against reference IGS or NGA SV ephemeris and clock data data, and does not use actual pseudorange observations. The results using 2015 and 2016 broadcast data collections will also be described in this article.