jamovi

Depending on whether the "One-way ANOVA" or "ANOVA" is run, the Shapiro-Wilk P.value changes. The Shapiro-Wilk P.value from the the "One-way ANOVA" was 0.807, while the Shapiro-Wilk P.value from the "ANOVA" was 0.817. I cannot work out why the difference is occurring - what is the difference in coding for these two Shapiro-Wilks?

hi, you'd need to provide a .omv file before we could comment (you may need to zip it up first).

cheers

jonathon

Good Morning,

I have the same problem, I have obtained different shapiro-wilk pvalues in anova one-way and with respect anova. Also, with the library moretest I had different Kolmovogov-Smirnov and A_D tests and pvalues. The differences are insignificant but I would like to know about it.

Thank you in advances and for the great program you have done!
Carmen

MOPOLV wrote: ↑Mon Aug 14, 2023 7:11 am I have the same problem, I have obtained different shapiro-wilk pvalues in anova one-way and with respect anova ...

Hey Carmen,
a simple response on the fly.
The small differences you saw are due to the residuals passed through the shapiro-wilk test.
In anova one-way they are standardized residuals!
You can take a look at the source code here (#L56 and #L279):
https://github.com/jamovi/jmv/blob/mast ... vaonew.b.R

Cheers,
Maurizio Agosti
https://www.jamovi.org/about.html