what citation can i use for the kendalls w thresholds, you mentioned in the video that there are lots of papers, however i cannot find any with specific thresholds that i can refernce. im talking about the 0.1 (small), 0.3 (moderate effect), 0.5 (strong)
Thank you Sir! I would like to ask what is the difference between Connover Post hoc test and Post hoc test? which one should I use if I am comparing three related groups? Thanks!
Hi Paul! Many thanks for your great video. However, I would like to ask you, how could I determine the effect size for every pair comparison (pos hoc). Thanks again. Best, Enzo.
So for example if I'm comparing values that happened over time, let's say over 5 weeks. Should I use the corrected p values as there are 10 comparisons in total?
There's no hard rule over that- you reduce chances of a false positive (due to multiple testing) if you do. However, you may increase your chance of a false negative. If you do the corrected p values you are erring on the side of caution.
@@DrPC_statistics_guides and I have only 5 subjects in each group (2 groups). As an example, friedman shows me a large sig. diff. (p=0.003) and if I use the corrected p values I might have 1 sig diff between all 10 comparisons. Is this right or in this case I shouldn't use the corrected p values? Thank you.
This video made everything really easy to understand. Already, manged to download JASP and apply here learned knowledge. Cheers.
Awesome, clear, and concise. Learning to use JASP and this was 10 minutes well spent. Thank you.
Thanks. I appreciate that!
what citation can i use for the kendalls w thresholds, you mentioned in the video that there are lots of papers, however i cannot find any with specific thresholds that i can refernce. im talking about the 0.1 (small), 0.3 (moderate effect), 0.5 (strong)
Thank you Sir!
I would like to ask what is the difference between Connover Post hoc test and Post hoc test? which one should I use if I am comparing three related groups? Thanks!
A conover test is still a post hoc test but is used with nonparametric data
You're a legend
Hi Paul! Many thanks for your great video. However, I would like to ask you, how could I determine the effect size for every pair comparison (pos hoc). Thanks again. Best, Enzo.
Have the same issue here. Did you find any solutions to determine pairwise effect size in Firedman test in JASP?
So for example if I'm comparing values that happened over time, let's say over 5 weeks. Should I use the corrected p values as there are 10 comparisons in total?
There's no hard rule over that- you reduce chances of a false positive (due to multiple testing) if you do. However, you may increase your chance of a false negative. If you do the corrected p values you are erring on the side of caution.
@@DrPC_statistics_guides and I have only 5 subjects in each group (2 groups). As an example, friedman shows me a large sig. diff. (p=0.003) and if I use the corrected p values I might have 1 sig diff between all 10 comparisons. Is this right or in this case I shouldn't use the corrected p values?
Thank you.
Thank you very much!