The Wilcoxon signed rank test, also known as the Wilcoxon rank test of symbols, adds the rank of the absolute value of the difference between the observed value and the center position of the null hypothesis according to different symbols as its test statistic.
It is suitable for paired comparisons in T-tests, but it does not require the difference between paired data to follow a normal distribution, only a symmetric distribution is required. Verify whether the difference in paired observation data comes from a population with a mean of 0 (whether the population generating the data has the same mean).
The single sample Wilcoxon test was conducted using A1-A5 variables, and the analysis results are as follows:
From the above table, it can be seen that using a single sample non parametric test to test the differences between the dependent variables A1 and 3.0, the results showed that there was no significant difference between A1 and the mean (p=0.608)
From the above table, it can be seen that using a single sample non parametric test to test the differences between the dependent variables A2 and 3.0, the results showed a significant difference between A2 and the mean (p=0.024)
From the above table, it can be seen that using a single sample non parametric test to test the differences between the dependent variables A3 and 3.0, the results showed that there was no significant difference between A3 and the mean (p=0.289)
From the above table, it can be seen that using a single sample non parametric test to test the difference between the dependent variables A4 and 3.0, the results showed a significant difference between A4 and the mean (p=0.016)
From the above table, it can be seen that using a single sample non parametric test to test the difference between the dependent variables A5 and 3.0, the results showed a significant difference between A5 and the mean (p=0.000)
Reference:
[1]吴喜之 赵博娟.非参数统计:中国统计出版社,2011年
[2]秦山, 汪晓银. 秩和检验在饲料科学研究中的应用[J]. 饲料广角, 2003(21):30-32.
[3]《数学辞海》委员会. 数学辞海(1-6).第6卷[M]. 山西教育出版社, 2002.