Thank you for the answer, unfortunately I failed to find afex before you pointed me to it. If I ever bump into any inconsistencies I will post it here.

Cheers,

Balázs

Statistics: Posted by bknakker — Wed Jun 20, 2018 7:03 pm

]]>

We use afex to run these models: https://github.com/jamovi/jmv/blob/c3e3 ... rm.b.R#L41. Can you give me a reproducible example where the two don't match?

Cheers,

Ravi

Statistics: Posted by Ravi — Thu May 31, 2018 6:36 pm

]]>

I'm trying to replicate a repeated measures anova with 2 within and 1 between variable. My purpose is to work onwards with a model object of some kind, e.g. experiment with multcomp::glht(). I'm doing kind of fine using only the within vars, but I could not get even close to the stats seen in jamovi with the between var in the model. How are these models internally implemented in jamovi?

Thanks in advance,

Balázs Knakker

Statistics: Posted by bknakker — Thu May 31, 2018 5:30 pm

]]>

thanks for reporting.

jonathon

Statistics: Posted by jonathon — Tue May 15, 2018 9:33 pm

]]>

you're quite right. i'm surprised this escaped our attention for so long.

will be fixed in the next version.

with thanks

Statistics: Posted by jonathon — Fri May 11, 2018 10:23 pm

]]>

Not at the moment. With the way we created the UI it's not very easy to allow for second order factors and still keep things clean and easy-to-use. I think it would be nicer if more complex CFA/SEM models would be implemented in a separate module, which we are eager to support.

Cheers,

Ravi

Statistics: Posted by Ravi — Fri May 11, 2018 7:20 pm

]]>

With Jamovi, the One Sample T-Test (non parametric) is called ‘Mann-Whitney U’. The displayed statistic is a sum of ranks ($T^{+}$).

The Paired Samples T-Test (non-parametric) is called the ‘Wilcoxon rank’. The displayed statistic is a sum of ranks ($T^{+}$).

The Independent Samples T-Test (non-parametric) is called the ‘Mann-Whitney U’. The displayed statistic is a sum of comparisons that is calculated from a sum of ranks:

$$U_{1}= n_{1}\cdot n_{2}+ \frac{n_{1}(n_{1}+1)}{2} - T_{1}$$

For the sake of consistency, I suggest that the One Sample T-Test (non-parametric) would be renamed ‘Wilcoxon rank’.

Hoping that my remark is understandable, I send you my best regards.

Jantonie

Statistics: Posted by jantonie — Fri May 11, 2018 11:19 am

]]>

Statistics: Posted by scroon — Fri May 04, 2018 3:17 am

]]>

]]>

sorry for the delay!

Statistics: Posted by jonathon — Thu May 03, 2018 8:58 pm

]]>

Is there a way, in which I could add a second order factor to my confirmatory factor analysis?

Best regards

Yann

Statistics: Posted by Yann Cook — Tue May 01, 2018 3:36 pm

]]>

great if you could create an issue on our issues page:

https://github.com/jamovi/jamovi/issues

with thanks

Statistics: Posted by jonathon — Fri Apr 27, 2018 1:34 am

]]>

Whether one sample or two samples are available, the Shapiro-Wilk normality test gives the same result with Jamovi.

I propose to you, when you have two samples, to perform the normality test not on the scores x_ {ij} but rather on the intra-group differences (x_ {ij} - \ bar {x} _j).

Thanks again for your wonderful software.

Jantonie

Statistics: Posted by jantonie — Thu Apr 26, 2018 11:44 am

]]>

]]>

]]>