It is very important to analyze and test the data of the TK2GPS sounding rocket for understanding the applicability and the data reliability of the test instrument and its application prospects in weather modification. Based on the analysis methods of average deviation (BIAS), root mean square error (RMSE), and correlation coefficient, usging the TK2GPS sounding rocket and Lband sounding radar data, the differences and variation characteristics of the sounding data such as temperature, relative humidity, wind direction and wind speed are compared and analyzed, and the possible causes of the differences are discussed. The results show that the TK2 GPS sounding rocket and the Lband sounding have significant positive correlations with the same elements, and passed the significance test of 0.05 and above. Analyses of the differences between the two sounding data reveal that the average deviation and dispersion of temperature and wind speed are small; although the average deviation of relative humidity is large, correlation is good; the dispersion of wind direction at each level is large, and the correlation is poor. The vertical profiles of temperature and wind speed of the two soundings are highly consistent. The vertical profile of relative humidity is significantly different at low altitudes and more consistent at higher altitudes; the relative humidity at low altitudes is greater than that at high altitudes. The difference trend in wind profiles is just the opposite of relative humidity profiles. With the increase of horizontal detection distance, the deviation of each element of sounding rocket from Lband sounding increases. The correction equation of relative humidity is established by correlation analysis. It can correct the relative humidity of sounding rockets very well.