[Bug]: Neighborhood-Based Clustering - Means reporting error

[adjwojcik]

JASP Version

0.19.3

Commit ID

No response

JASP Module

Machine Learning

What analysis are you seeing the problem on?

Neighborhood-Based Clustering

What OS are you seeing the problem on?

Windows 10

Bug Description

I am writing to report a critical issue encountered with the Neighborhood-Based Clustering (NBC) technique in the latest JASP version, 19.3.

We previously conducted an NBC analysis and identified a 4-cluster model based on the Bayesian Information Criterion (BIC) using an earlier version of JASP. Upon re-running the identical analysis with JASP 19.3, we have observed a significant discrepancy in the reported cluster means.

Specifically, while the overall model fit statistics (BIC and AIC) remain consistent with our previous findings, the means presented in the "Cluster Means" table are markedly different. These values not only deviate from our initial model's results but also contradict the means displayed in the "Cluster Information" table within the same JASP 19.3 output. The magnitude of these differences is substantial, including instances where the order of means is reversed across clusters.

Furthermore, this issue appears to propagate to the graphical output. The cluster means chart seems to be generated based on the erroneous data from the "Cluster Means" table, whereas the density chart remains unaffected.

Our subsequent investigation suggests that the values reported in the "Cluster Information" table are, in fact, accurate. We have re-analyzed our dataset using alternative statistical software, and the results align with the "Cluster Information" table from JASP 19.3. This consistency strongly indicates a problem with the data import or processing specifically affecting the "Cluster Means" table.

This inconsistency poses a serious concern for the reliability of the NBC results in the current version of JASP. We request your urgent attention to investigate and resolve this apparent anomaly in how data are being handled for the "Cluster Means" output.

Expected Behaviour

ix this. Now. This isn't some minor inconvenience; it's a fundamental flaw that invalidates the results. Get your act together and sort out this data import catastrophe in the Means Table.

Steps to Reproduce

1. Open Jasp.
2. Use Machine Learning Module and the Neighborhood-Based Clustering
3. Compare the means values between the Cluester Means and Cluster Information tables - cluster centers.
4. Look at the Means graphical output.
5. Compare it with density plot for clusters.

Log (if any)

No response

More Debug Information

No response

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.

[github-actions]

@adjwojcik, thanks for taking the time to create this issue. If possible (and applicable), please upload to the issue website (#3446, attaching to an email does not work) a screenshot showcasing the problem, and/or a compressed (zipped) .jasp file or the data file that causes the issue. If you would prefer not to make your data publicly available, you can send your file(s) directly to us, issues@jasp-stats.org

[koenderks]

This bug has already been noticed and fixed in jasp-stats/jaspMachineLearning#423, and the fix will be included in the next release, scheduled for release in about two weeks. Feel free to reopen this issue if you think this is not the same issue as in #3297.