Abstract: Aiming to improve the quality of GPS/MET water vapor retrieval data in Anhui Province, the quality control of GPS/MET water vapor retrieval data was carried out by using the comprehensive discrimination method. Firstly, the authors analyzed GPS/MET water vapor retrieval data from July 2011 to September 2014, formed the boundaries of the data value, as the first step of comprehensive discrimination analysis method. On the basis of the first step, the bias (GPS/MET water vapor actual observation value and global reanalysis GDAS calculated precipitation values deviation) threshold method to control the overall quality of the data was adopted as the second step, and the third step was to further conduct the quality control on large deviations in water vapor data by double weight method, and preliminarily obtained the sequence relationship of GPS/MET water vapor retrieval data and the precipitation, and ultimately formed the data sets of GPS/MET water vapor retrieval data quality control of Anhui Province. The development of data sets and the corresponding quality control method, which were based on the use of the GPS/MET observation system of water vapor data with characteristics of high precision and high spatial and temporal resolution, could lay a foundation for water vapor data assimilation fusion.