由于外界客观条件的变化和实验者主观因素的影响，在实验过程中收集到的数据往往会混有异常值，因而会明显的影响实验结果，甚至会得出错误的结论。本工作按照统计学原理，在计算机上根据σ准则、ｔ检验准则、Ｃｒｕｂｂｓ准则和Ｄｉｘｏｎ准则编制了对实验数据进行识别和剔除的程序，实现对异常值的自动剔除，使实验数据更加准确可靠，实验结果更符合客观实际。该程序使用方便，实际应用效果令人满意，可极大地提高实验资料的统计分析质量。 Due to changes in external objective conditions and the subjective factors of the experimenter, the data collected during the experiment is often mixed with abnormal values, which will obviously affect the experimental results and even lead to erroneous conclusions. This work according to statistical principles, in the computer according to σ criteria, t test criteria, Crubbs criteria and Dixon guidelines for the experimental data identification and removal procedures, to achieve automatic rejection of abnormal values, so that the experimental data is more accurate and reliable, The experimental results are more in line with objective reality. The program is easy to use, the practical application effect is satisfactory, and the statistical analysis quality of experimental data can be greatly improved.