Using regression coefficient method, high steady coefficient method, variation coefficient method and adaptable qualitative parameter method, high-yield, steady-yield and adaptability characters of spring wheat varieties of northeast early-maturing group planted on nonirrigated farmland in 2002 were tested. The results indicated that all four methods brought similar results with little differences. The regression coefficient method, expressing the stability with r2 and error with b, is comparatively scientific and reliable. But its calculation is rather convoluted. As to other three methods, calculations are relatively easy, results can be directly perceived through the senses but they are relatively rough.