Boxplot outliers

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
RangoMcRoony
Posts: 2
Joined: Thu Mar 25, 2021 6:28 pm

Boxplot outliers

Post by RangoMcRoony »

Under descriptives there's an option to do a box-plot of data, this can be used to see if there are outliers. However i cannot seem to figure out how i get jamovi to indicate the specific participant that is an outlier. I know there's an option for this in Jasp, PSPP, SPSS, sofa-statistics. Pretty much all the popular statistical software. Is this something that is possible with jamovi or is this something that still needs to be developed?




I know it isnt really a sophisticated question, this is my first post so let me know if i did something wrong or posted in the wrong forum

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

Re: Boxplot outliers

Post by jonathon »

hi,

there's a bug on ggplot which is currently preventing us from providing this feature.

we're waiting on a fix there.

kind regards

jonathon
RangoMcRoony
Posts: 2
Joined: Thu Mar 25, 2021 6:28 pm

Re: Boxplot outliers

Post by RangoMcRoony »

any updates on this?
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Boxplot outliers

Post by jonathon »

i think the bug in ggplot persists.

jonathon
peszka
Posts: 1
Joined: Wed Jun 22, 2022 10:04 pm

Re: Boxplot outliers

Post by peszka »

I found that you can first calculate what an outlier would be...say for your data it is anything bigger than 15 or smaller than 2. Then use jamovi's filters feature to find these outliers...say in one filter you put =variablename > 15. If you then change the variable to a nominal or ordinal scale (if it isn't already), then go to the exploration module and create a frequency distribution by clicking "Frequency tables". In the output, you'll see how many outliers you have next to N in the descriptives table and in the frequencies table you'll see the row number where each outlier is located under the column labeled "Levels"
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Boxplot outliers

Post by jonathon »

an update, the latest version of jamovi does give you case numbers on box plots for outliers.

jonathon
Solène M
Posts: 1
Joined: Sat Jul 30, 2022 8:53 am

Re: Boxplot outliers

Post by Solène M »

Hello,I'm sorry if this problem has already been solved earlier but I didn't find a solution...
I'm currently trying to identify outliers with Jamovi to exclude them from my analysis.
Is there a way to make Jamovi identify the outliers that I should remove?
Thank you very much for your hepl!
User avatar
MAgojam
Posts: 421
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: Boxplot outliers

Post by MAgojam »

Solène M wrote:Is there a way to make Jamovi identify the outliers that I should remove?
Hey,
you might be interested in taking a look here, where is my suggestion to a question similar to yours?
viewtopic.php?f=2&t=2234&p=8298

All cases that are outliers for the variable of interest (in the example age) are excluded from the analysis using the "Median Absolute Deviations".

Cheers,
Maurizio
Post Reply