Fatigue test is an important means to evaluate the fatigue performance of materials and parts, under normal circumstances, people pay more attention to the fatigue life value of component fatigue test, and neglect to analyze the reliability and accuracy of fatigue test data itself. In this paper, the fatigue test of the typical fourth-generation aluminum–lithium alloy material 2060-T8 without shot peening and two steel casting shot peening processes is carried out, the normality test of the fatigue test data, the identification of abnormal data, and the hypothesis test of fatigue test data are compared and analyzed, and the reliability life value of each group of materials is calculated under the condition that the fatigue life is significantly different, and the influence of the shot peening process on the fatigue performance of the test specimen, including fatigue life and dispersion, is studied by statistical methods. The reliability of fatigue test results can be improved by using statistical methods for fatigue performance analysis, and effective and reliable support can be provided for the material selection of structures in engineering.