Affected by the global change, climate change in Shanxi presented new features. Based on the latest climate data from 38 meteorological stations in Shanxi, this paper analyzed the climate change characteristics of precipitation in Shanxi during 1958-2013 using linear trend estimation, average distribution and EOF analysis. The results show: (1) Annual mean precipitation in Shanxi was 494.9 mm, it varied from 382.8 mm to 637.2 mm in the past 56 years and had a linear decreasing trend of 12.6 mm/10a (significantly higher than the national level). (2) The average seasonal precipitations were 77.6 mm (spring), 290.5 mm (summer), 114.3 mm (autumn) and 13.0 mm (winter), respectively. Except winter, precipitation had decreasing trend in the other season, which was consistent with the research result on North China. (3) The precipitation in spring, summer and winter had obvious decadal characteristics: precipitation in spring had reversed fluctuating features with precipitation in winter, the more (less) the spring precipitation, the less (more) the winter precipitation; precipitation in summer had significant decreasing trend and inter-decadal variation, the biggest decline was 9.8 mm/10a; the winter precipitation had the most dramatic fluctuation, it varied from 1.1 mm to 28.3 mm in the 56 years, the maximum value was 24.7 times of the minimum value. (4) The EOF analysis indicated that annual & four season's precipitation in Shanxi have good consistence in space and show consistent more or less in most years, but the high load zone had some regional characteristics: high load zone of the annual precipitation was located in the central east region, high load zone of the spring precipitation was located in the south central region, high load zones of the summer, autumn and winter precipitation were located in the south region. (5) The modals decomposed with EOF for annual and summer precipitation had slow convergence speeds, variance contribution rates of the first modals were 33% and 49%, respectively, the cumulative variance contribution rates of the first three modals were 69% and 70%, respectively. Variance contribution rates of the modals decomposed with EOF for spring, autumn and winter precipitation were large, their convergence speeds are fast. Variance contribution rates of the first modals were 65%, 62% and 62% respectively, the cumulative variance contribution rates of the first three modals are 81%, 84% and 86%, respectively.