Meta analysis error - need finite 'xlim' values

[laurastjohn]

Hi all,

I'm getting an error when I use the MAJOR meta analysis module and because of this it won't produce any graphs. I keep getting the same error saying:

"need finite 'xlim' values"

I have multiple outcomes (as you can see) and some of my papers do not have data for the particular outcome (which is why there are blanks in the data set).

Either way - I can't seem to figure out what to do so that I can get it to produce the graphs. I have attached my dataset in case there is something I need to fix there.

Thanks!

[MAgojam]

Hi @laurastjohn,
I think the problem you raised is due to how you organized the data for the stats to do.

Hoping it will help you:
1) After opening the csv file, make sure that the variables reporting the numerosity have been interpreted as continuous (Measure type = Continous), most likely being few data per row, jamovi will interpret them as factors, attributing (Measure type = Nominal) and (Data type = Integer).
In case it changes from Nominal to Continuous.
2) Make sure you only go through those rows where there is data for the variables of interest and this you can do by simply filtering and excluding the rows that are missing this data.

I am attaching a screenshot where you can see the result of the above, on your data (csv) with the variables (CVD _...) by applying the following filter:
CVD_EXP_N != ""

MAJOR_FilterRows.png (193.3 KiB) Viewed 6162 times

Cheers,
Maurizio

[Beatty01]

Hello,
Your assistance will be extremely appreciated.
Despite confirmation of no missing values in my attached dataset, I am getting 2 error messages when running the Correlation Coefficient option of the MAJOR package to conduct a meta-analysis in Jamovi. No graphs/plots are therefore generated in the results.

Error 1: "need finite xlim value."
Error 2: "argument refers to differing number of rows 66,91"

Please note that I interchanged the study/author labels with practices for which I was interested in comparing. However, maintaining the study labels still produces the same errors.

Thanks, and I will be glad to hear from you on how to resolve this.
Beatrice