3

Good day! I'm totally new to SPSS (and the test as well, but after some research I've finally demystified a bit about Mann-Whitney and how to do the test in SPSS. The only problem I'm currently facing is the analysis) And so in this snippet, I want to discover how the researcher came up with this analysis:

Valid XHTML.

How was he able to come up with a conclusion that there is no significant difference between the examination marks? What specific data in the table made him conclude that? What variables did he look at? If I were to use a Mann-Whitney test and have to interpret the results, where will I look at? What are the factors do I have to consider? Any help would be very much appreciated. :)

Also, what's the purpose of the Z and Wilcoxon W?

Ferdi
  • 4,882
  • 7
  • 42
  • 62
troubledwater
  • 77
  • 7

1 Answers1

3

"Asymp. Sig. (2-tailed)" refers to a two-tailed p value for the Mann–Whitney test, which indicates the significance of its comparison of ranks for the two groups of gamers. Since p > .1, his conclusion that these results are insignificant is probably not going to attract much debate. BTW, an exact p value is also calculable for these tests, but may take longer with somewhat large samples such as this.

As for the meaning of Z and the Wilcoxon W, I think another answer of mine has the info you want:

The Wilcoxon signed-rank test is appropriate for paired samples, whereas the Mann–Whitney test assumes independent samples. However, according to Field (2000), the Wilcoxon W in your SPSS output is "a different version of this statistic, which can be converted into a Z score and can, therefore, be compared against critical values of the normal distribution." That explains your z score too then! FYI, Wikipedia adds that, for large samples, U is approximately normally distributed.

Reference
Field, A. (2000). 3.1. Mann-Whitney test. Research Methods 1: SPSS for Windows part 3: Nonparametric tests. Retrieved from http://www.statisticshell.com/docs/nonparametric.pdf.

His sample is fairly large, and the p value should be the same for all test statistics if I'm not mistaken.

Community
  • 1
Nick Stauner
  • 11,558
  • 5
  • 47
  • 105
  • I thought the analysis has something to do with the Z. (I got carried away with the negative sign because my correlation test a while ago yielded a negative value). **Thank you** :D – troubledwater May 14 '14 at 20:33
  • 2
    +1 Field - as so often - is sort of right, but misleading. All of the usual versions of the Wilcoxon rank sum statistics are asymptotically normal and "can be converted to a Z score". – Glen_b May 14 '14 at 23:25