distractor analysis error

[Elvire Alice]

Hello,

My name is EA and I am working as a language test developer. I have recently taken my first steps in the world of test analysis and I am still learning about the wonderous world of Jamovi. We know that ideally a test should comprise of more than 10 items to do a proper analysis, but for now, this is what we have to work with. We have an issue with the distractor analysis of a particular test batch and even though the parallel test forms are equally comprised of 10 items, there haven't been any issues with the other analyses. We keep getting the 'breaks are not unique' error and although we tried to resolve the issue, we cannot find the problem. We have inserted the test takers' answers again in an excel file, changed it to csv, changed the language settings, etc. I am just a beginner so that is why I am reaching out to you. Maybe I have overlooked something? If you need extra info, please do ask. Thank you so much in advance for your time and effort.

[seol]

Hi EA

The newest version of snowIRT resolved the issue of 'breaks are not unique'.
please download the following version of snowIRT module in jamovi library.

Best Regards
Seol

[Elvire Alice]

Hi Seol,

Thank you for your reply. I have been out for a while, so I have missed your message, but I'm back and I'll try to do as you suggest.

Thank you very much!

[Elvire Alice]

Hi again,

I've installed the update - which is the 5.1.8 version, and I have tried again, but I still get the error 'breaks are not unique'. We only have three possible answers, A, B and C, might it have something to do with that?

Thank you for your help!

[seol]

Hi
Yes, that is very likely the reason.

The error "breaks are not unique" is a common R issue that occurs when the data have only a few distinct response categories (such as A, B, C). In these cases, jamovi internally relies on R functions that automatically compute break points for plotting or categorization, and this process can fail when the number of unique values is very small. This behavior is common in multiple-choice or ordinal data and is not related to the jamovi version itself.

If you can send the .omv file in which the error occurs, I would be happy to take a look and see if there is anything else contributing to the issue.

Best
Seol

[Elvire Alice]

Hi Seol,

Thank you so much for your reply. Unfortunately, - we test English listening and reading skills - all our listening tests only have 3 options, so our analysis of our new pretests will have the same issue over and over again.... I'm still very new and unexperienced and I would love to send you the files, but I'm still figuring out how to send it to you ... I will get back to you and thank you again for your help!

[seol]

Hi
I’ve recently updated this issue in the latest version((VERSION 5.1.9) of snowIRT.
Could you please install the most recent version and see if the problem persists?

Best Regards
Seol

[Elvire Alice]

Oh, I was looking for the way to add attachments here and I see you have already replied!

I will install the version you mention.

Meanwhile here are two files I have already tried to analyze, but with the error in distractor analysis.

Thank you!

[Elvire Alice]

Hi Seol,

I've updated again, tried to run the distractor analysis again, but unfortunately I still get the 'breaks are not unique'....

[Elvire Alice]

Sorry, here I am again ....

I've tried to analyse a new data set of listening and for some or other dark reason, I do get the distractor analysis for this set. So happy!
Should I maybe try the whole analysis again for the previous two files? I have updated the IRT module, but since I had already worked in the document, it might have gone wrong? I'll wait for now to give you the chance to reply! Thanks again for your time!