[Bug]: Correlation Analysis "This analysis terminated unexpectedly"- worked for larger dataset but not smaller subset

[owebhs]

JASP Version

0.19.3

Commit ID

No response

JASP Module

Regression

What analysis are you seeing the problem on?

Classical Correlation

What OS are you seeing the problem on?

Windows 11

Bug Description

I was running a Classical Correlation analysis on a subset of data (note I did not get this bug when working on the whole dataset) and kept getting the following error, for different subsets and different files:

I was able to bypass the error by clicking and dragging a couple items into Variables at a time, rather than all into variables at once, but eventually the results page would also turn blank:

I'm currently downloading the newest version of JASP but wanted to post in case the new version doesn't fix this issue (as more likely it's something I'm doing wrong).

Apologies, I cannot share the file itself as it contains sensitive information that I can't redact without substantially changing the file.

Expected Behaviour

Rather than the error or blank page, I expected the output to appear as normal after a few minutes of analysis. I have been able to successfully bulk load all required variables using the whole dataset, but not subsets.

Steps to Reproduce

1. Open file
2. Select Regression > Classical > Correlation Analysis
3. Drag all the relevant columns into variables (bulk drag)
4. Receive error in red text box after a few minutes
5. OR drag relevant columns into variables, a few at a time
6. Get 90% of outputs, but then get the blank results screen for the last 10%. EDIT: If it helps, I am attempting to do analysis across about 80 columns, and the blank screen appears when I'm adding the last 8 or so.

Log (if any)

JASP 2025-04-03 08_50_20 Engine 0.log

Apologies, getting an error "!-- Failed to upload "JASP 2025-04-03 08_50_20 Desktop.log" --"

More Debug Information

-------- Application Info --------
JASP Version: JASP 0.19.1
Build Branch: lockfiletesting
Build Date: Sep 9 2024 19:00:04 (Netherlands)
Last Commit: 8ac10c9c7aec029fdd695262d7bb4c3742a8ec89

-------- Basic Info --------
Operating System: Windows 11 Version 24H2
Product Version: 11
Kernel Type: winnt
Kernel Version: 10.0.26100
Architecture: x86_64
Install Path: C:/Program Files (x86)/JASP
Platfotm Name: windows
System Local: en_GB

-------- Extra Info --------
Current code page
Active code page: 850
Active code page: 65001

Final Checklist

• I have included a screenshot showcasing the issue, if possible.
• I have included a JASP file (zipped) or data file that causes the crash/bug, if applicable.
• I have accurately described the bug, and steps to reproduce it.

[owebhs]

Edit: I am experiencing the same issue with the newest version also. I will update the bug report accordingly.

[boutinb]

Hi @owebhs Maybe the problem comes from one variable. You wrote that you could use 90% of the variables Can you start another analysis, and use only the last 10% of the variables: if this fails, can you try to find out which variable is causing the problem?

[owebhs]

Thank you @boutinb , sorry I should've mentioned, I have tried with the last 10% of variables only, and they work fine. It seems to be the volume that's the issue.

[boutinb]

Hi @owebhs I have just tried with 100 columns (it is the number of columns I could see in your log) with 100 rows (how many rows do you have?) on a cheap Windows laptop, and it works for me. But it takes already quite long (~3 minutes).
Can you try the analysis on another (more powerful) laptop/computer ?

[owebhs]

Hi @boutinb, good news. We've solved the issue - some of the columns had a lot less data (e.g., only 10% of the column was filled). By removing these we could run the analysis with no issues. Not sure why that fixed it, but glad it did! I'll close the bug off with this.