Thank you very much for this video! It helped me alot!
@alaric49
4 жыл бұрын
Thanks so much for posting this!
@reginakolzenburg2936
6 жыл бұрын
How did you arrange that data? I can't quite figure it our from the first shot. Cheers for the video :)
@pub361USC
6 жыл бұрын
Should be the same as any other dataset if I take your meaning? 1 row per observation and 1 column per variable, you can download the data from their website www.statsci.org/data/general/eysenck.html
@agnesberesinski3429
2 жыл бұрын
Hello PUB361, in about 90% of the cases after the rejection of the null hypothesis one sees thanks to pairwise comparisons (post-hoc with Bonferroni corecture) which groups differ significantly. Now, however, I have a few cases where KH rejects the null hypothesis (somewhat close but already 0.044) but pairwise comparisons show no significant differences. Here, however, I thought I might look at the post-hoc values without Bonferorroni corectures and base the significance of the differences on them, because some of the significant values are already below 0.05. Is it possible to understand such an approach? Are the corrected significances according to Bonferroni possibly too imprecise if KH is just below 0.05?
@pub361USC
2 жыл бұрын
Hi Agnes, I'm afraid you might not like my answer - but what I would say is that even though your overall test is below the mythical 0.05 there is nothing scientific definitive about this - its just a line in the sand. I would say that the evidence is not strong, and even if your posthoc tests don't show a difference you can say there is some evidence of a difference and just describe the distributions.
@aminelerian9836
4 жыл бұрын
Hi I have a question,I run Kruskal wallis test and it show significance differences but when get to the pairwise comparison table there are no significance difference, i.e. no yellow highlighted cell.. can you help me out
@pub361USC
4 жыл бұрын
difficult for me know exactly what is happening without seeing the data but this page might be helpful to you as it explains why this can sometimes happen - www.researchgate.net/post/How_to_interpret_the_result_of_a_Kruskal-Wallis_test_revealing_p005_but_with_a_p005_between_two_groups
@wendyteal4204
5 жыл бұрын
Hi! I have a question: I am running into the problem of not being able to see the data in my pairwise comparison. There are so many tests I cant see the numbers. I've tried coping it to the original SPSS output, and transferring to Word but neither of these work. Suggestions?
@pub361USC
5 жыл бұрын
I'm not sure - in the version I'm using at the moment it's not a problem but I know what you mean, and I can't see how to copy the the table as text or get the data for it although you can for other parts from the model viewer. If you go to Edit -> Options -> Viewer in the main window you can play around with the font settings, perhaps shrinking it so the rows don't overlap might help?
@DrHanjabamBarunSharma
6 жыл бұрын
which test is being employed here for post hoc testing by default? is it Dun's test?
@pub361USC
6 жыл бұрын
Probably ?! but I have never actually gone through the documentation ....
@libertadsagrada
3 жыл бұрын
i have the same question, because i need the name for my thesis :(
@libertadsagrada
3 жыл бұрын
@@pub361USC help :(
@pub361USC
3 жыл бұрын
@@libertadsagrada I just googled it and that seems right, try this page here: www.ibm.com/support/pages/can-spss-perform-dunns-nonparametric-comparison-post-hoc-testing-after-kruskal-wallis-test
@martinorsomething1669
3 жыл бұрын
If I pass this Quantitative Methods module, I'll be asking you to marry me!
Пікірлер: 16