Surface-based GPS measurements of zenith path delay (ZPD) can be used to derive vertically integrated water vapor (IWV) of the atmosphere. ZPD data are collected in a global network presently consisting of 160 stations as part of the International GPS Service. In the present study, ZPD data from this network are converted into IWV using observed surface pressure and mean atmospheric water vapor column temperature obtained from the European Centre for Medium-Range Weather Forecasts' (ECMWF) operational analyses (OA). For the 4 months of January/July 2000/2001, the GPS-derived IWV values are compared to the IWV from the ECMWF OA, with a special focus on the monthly averaged difference (bias) and the standard deviation of daily differences. This comparison shows that the GPS-derived IWV values are well suited for the validation of OA of IWV. For most GPS stations, the IWV data agree quite well with the analyzed data indicating that they are both correct at these locations. Larger differences for individual days are interpreted as errors in the analyses. A dry bias in the winter is found over central United States, Canada, and central Siberia, suggesting a systematic analysis error. Larger differences were mainly found in mountain areas. These were related to representation problems and interpolation difficulties between model height and station height. In addition, the IWV comparison can be used to identify errors or problems in the observations of ZPD. This includes errors in the data itself, e.g., erroneous outlier in the measured time series, as well as systematic errors that affect all IWV values at a specific station. Such stations were excluded from the intercomparison. Finally, long-term requirements for a GPS-based water vapor monitoring system are discussed.