Error given in MAJOR continuous moderator

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
Floor Emilie
Posts: 3
Joined: Thu Sep 16, 2021 8:25 am

Error given in MAJOR continuous moderator

Post by Floor Emilie »

Hi everybody,

I am using Jamovi for my master's thesis in Medicine (dep. Sexology) and trying to perform a meta-analysis of sexual outcomes based on studies using testosterone therapy in women by comparing their effect with measured serum testosterone at baseline.

I've conducted the effect sizes (yi) on these sexual outcomes and their variances (vi) and want to use the moderator to study the prediction of the cumulative area under the curve of low baseline testosterone serum samples.

So I went to MAJOR -> Effect Sizes and Sampling Variances -> Effect size (yi), Variance (vi), Study label (author name). So far everything goes well.
But when I put the testosterone values (percin) in Moderator (type "continuous moderator"), I get an error that says "'x' and 'y' lengths differ".
I already made sure that the row with the moderator is as long as the other rows.
Also when I put these in R, I don't get this (or any other) error.

Does anybody know why I get this error and how I can fix it?
(See attachment file)

Thank you very much already for your time!
Attachments
dataFull_desire.csv
(9.33 KiB) Downloaded 571 times
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Error given in MAJOR continuous moderator

Post by jonathon »

try dropping kyle hamilton a line,

he's the author of MAJOR, and i don't think he hangs out much in these forums.

cheers
Floor Emilie
Posts: 3
Joined: Thu Sep 16, 2021 8:25 am

Re: Error given in MAJOR continuous moderator

Post by Floor Emilie »

Thank you for the reply Jonathon!
I am going to do that!
Floor Emilie
Posts: 3
Joined: Thu Sep 16, 2021 8:25 am

Re: Error given in MAJOR continuous moderator

Post by Floor Emilie »

Hi Jonathon and others,

I have contacted Kyle, but did not get a response so far.
Does anybody else recognize the error that has been given by the continuous moderator?

Kind regards,
Floor
User avatar
MAgojam
Posts: 421
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: Error given in MAJOR continuous moderator

Post by MAgojam »

Hi, Jonathon.

I would have a suggestion to fix the bug reported in this and other posts, at least waiting for Kyle to come around and find some time to fix his great MAJOR module.

A colleague's request for help led me to take a look at the module sources and how Kyle commented in the metadv.b.R file in these lines:
https://github.com/kylehamilton/MAJOR/b ... 2C%20alpha)%20%7B

used the TOSTmeta() code (Lakens' TOSTER package) for its plotTOSTfunction().
The error is generated by passing 4 arguments to the plotTOSTfunction() to these 4 lines:
https://github.com/kylehamilton/MAJOR/b ... bound_d%2C

I fixed it like this:

Code: Select all

ES = plotDataTOST$ES[1],
se = plotDataTOST$se[1],
low_eqbound_d = plotDataTOST$low_eqbound_d[1],
high_eqbound_d = plotDataTOST$high_eqbound_d[1],

I cloned the module, fixed the code as above, recompiled and now it does its job both in jamovi and in R on my colleague's PC (also on mine).

Thanks for your attention,
Maurizio
Post Reply