Precipitable water vapor is one of the key factors in study of atmospheric radiation transfer and absorption, global wide heat transfer, and prediction of heavy rainfall. Using precipitable water vapor data in 2015 retrieved by GPS/MET, microwave radiometer and L-band sounding respectively at Zhangqiu Station, the characteristics of the 3 kinds of datasets are studied. The results show that the variation trends of the 3 kinds of datasets are similar, but with significant systematic deviations. The mean values of precipitable water vapor from big to small are retrieved by GPS/MET, microwave radiometer and L-band sounding respectively. The mean deviations among the three kinds of datasets have little changes in different seasons, while the mean deviation of dataset retrieved by GPS/MET from dataset retrieved by microwave radiometer is about 4.5 mm. However, there are significant seasonal changes in the spread of deviation, which reaches its largest value in summer, followed by autumn, spring and winter. Since atmospheric water vapor content at 12:00 UTC is greater than that at 00:00 UTC, the spread of deviation of precipitable water vapor is always larger at 12:00 UTC than that at 00:00 UTC, and the relative error at 12:00 UTC is always less than or equal to that at 00:00 UTC.