network profile
| bio | website | |
|---|---|---|
| location | ||
| age | ||
| visits | member for | 10 months |
| seen | Mar 9 at 12:38 | |
| stats | profile views | 9 |
|
Mar 7 |
awarded | Popular Question |
|
Aug 10 |
accepted | Which result to choose when Kruskal-Wallis and Mann-Whitney seem to return contradicting results? |
|
Aug 10 |
comment |
Which result to choose when Kruskal-Wallis and Mann-Whitney seem to return contradicting results? @MichaelChernick I have became to learn from you that there is much more on statistics than just looking for "p<0.05". Michael Lew: I've downloaded your paper and will give it a read for sure. I'll follow your suggestion to have a good reasoning about my data in this situation. Thank you all! |
|
Aug 9 |
asked | Which result to choose when Kruskal-Wallis and Mann-Whitney seem to return contradicting results? |
|
Aug 3 |
accepted | Which W to report for the Wilcoxon test? |
|
Aug 3 |
comment |
Which W to report for the Wilcoxon test? Yes, I did not forget to mention the sample sizes in each group in my boxplots. So, W and U are the same thing for Wilcoxon rank sum/Mann-Whitney U, and the lowest value of the groups is chosen. Thank you! |
|
Aug 2 |
comment |
Which W to report for the Wilcoxon test? I have boxplots showing the median, the mean, etc, then I have a Wilcoxon rank sum test, say returning W=0 (the lowest), p=0.0034, so from what I understood I can report this as "The Mann-Whitney U test confirms a statistical significance effect (U=0, p<0.05)". Is this correct ? |
|
Aug 1 |
comment |
Which W to report for the Wilcoxon test? ?! Should I report whichever I prefer, or always the lowest W (even if 0), as this is the value that is compared to the critical one in the Wilcoxon Rank Sum Table? |
|
Aug 1 |
comment |
Which W to report for the Wilcoxon test? @whuber The "note" seems to be talking about the lack of unanimity on how to calculate the test statistic. But my question is this: R has function wilcox.test to compute Mann-Whitney U test, but this function returns "W=...", so is this W (the lowest) the value of U? |
|
Aug 1 |
comment |
Which W to report for the Wilcoxon test? @whuber Thank you! I also found this table useful. Two questions: is W the same as U in case of a Wilcoxon rank sum test (equivalent to Mann-Whitney's U)? And when W=0 is this the real W, even when the "other W" is >0? |
|
Aug 1 |
asked | Which W to report for the Wilcoxon test? |
|
Jul 6 |
accepted | Contradicting p-values for Anova and Kruskal-Wallis on same data: Which is right? |
|
Jul 5 |
comment |
Contradicting p-values for Anova and Kruskal-Wallis on same data: Which is right? If I did the transformation into ratios correctly, in R the Shapiro-Wilk results are: shapiro.test(1/A): W = 0.9064, p-value = 0.04657
shapiro.test(1/B): W = 0.9026, p-value = 0.06388
shapiro.test(1/C): W = 0.6018, p-value = 2.057e-06
That is, at alpha=0.05, Shapiro-Wilk normality test fails for groups A and C, so Anova is not adequate again ... |
|
Jul 5 |
comment |
Contradicting p-values for Anova and Kruskal-Wallis on same data: Which is right? I think your point about "The anova F test wrongly interprets group A to have a significantly larger mean" makes sense in my setting especially after my comment to @GregSnow's answer, where I analyse why Kruskal Wallis seems to be the most appropriate choice here. |
|
Jul 5 |
comment |
Contradicting p-values for Anova and Kruskal-Wallis on same data: Which is right? @GregSnow A is highly skewed because one task took much more time (the values above 21 min) than the others in A. A's effect surely makes individuals take more time in that task compared with the same task in B and C, in which individuals did not do the extra work caused by the effect of group A. However, as Kruskal-Wallis compares medians (centrality), and the durations for all other tasks are close among groups, I tend to accept the result of Kruskall-Wallis, especially as the extra time (bug fix) for the task in group A would be spent later (at bug discovery) by individuals in B and C. |
|
Jul 5 |
comment |
Contradicting p-values for Anova and Kruskal-Wallis on same data: Which is right? @Glen The Shapiro-Wilk test for the residuals of group A is the same as for group A (the residuals where computed as indicated here) |
|
Jul 5 |
asked | Contradicting p-values for Anova and Kruskal-Wallis on same data: Which is right? |
|
Jul 4 |
comment |
Is there a software package designed to automatically check the assumptions of various statistical tests? @MichaelChernick Thanks for those references! While searching in the internet I also found this one and this very useful. |
|
Jul 3 |
accepted | Is there a software package designed to automatically check the assumptions of various statistical tests? |
|
Jul 2 |
comment |
Is there a software package designed to automatically check the assumptions of various statistical tests? @MichaelChernick I do computer science research, and sometimes I need to do user controlled experiments, but generally, I can only get small samples because users are hard to get. Mainly, I take measures of task durations, how people use the tools (these sometimes implement different treatments), and users frequently answer surveys/questionnaires with Likert scales. So, in light of what I do, which statistics do I need, and which books do you recommend? Do these books you listed above are suitable for my work? Or there are others more suitable? Thanks |
