Different results with Dwass-steel-critchlow-fligner tests

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)
Post Reply
Josu
Posts: 2
Joined: Fri Feb 21, 2020 6:08 pm

Different results with Dwass-steel-critchlow-fligner tests

Post by Josu »

Hello,

Several months ago I performed a Kruskall-wallis test and a dwass steel critchlow-fligner test on some data. A few days ago, I wanted to check those results, and I conducted the exact sames tests on the exact same data (as we can see on the picture the descriptive stats are exactly the same).

The weird thing is that I got different results. The statistics are the same, but not the indices of significance. Hopefully, I had a screenshot of my old results to compare. As you can see on the picture, the diffrence between 1-3 and 2-3 is significant on the old results, but not on the new one.

Image

How can you explain this difference ?

Thank you for your help.
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Different results with Dwass-steel-critchlow-fligner tes

Post by jonathon »

hi,

there were some circumstances where the DSCF wouldn't produce correct results -- i think the fix landed in may 2019.

cheers

jonathon
Josu
Posts: 2
Joined: Fri Feb 21, 2020 6:08 pm

Re: Different results with Dwass-steel-critchlow-fligner tes

Post by Josu »

Okay, so I better should use the new results instead.

Thank you for your help !
Post Reply