NOAA/NESDIS currently generates GOES water vapor winds operationally for the GOES-8 and GOES-10 spacecraft. These data are an important data source for a wide range of applications that include assimilation into regional and global prediction systems, oceanic analyses, and tropical storm analyses. Moisture features observed in the GOES water vapor imagery can be tracked in time in order to estimate wind speed and direction. The moisture features can be clouds or pure water vapor structures which exist in clear sky conditions. The purpose of this paper is to investigate and discuss the quality of the GOES water vapor winds, particularly under clear sky conditions. Two items which will be discussed include feature tracking and height assignment. These two items can contribute heavily to errors in the water vapor product. The errors introduced by the height assignments can be large since the moisture features being tracked are not always at a single level. Results will be presented which: 1) Highlight the impact of using water vapor imagery, with varying temporal resolutions, to derive water vapor winds, and 2) Characterize the height assignment of these winds.