[Bug]: Error in Psych (Analysis terminated unexpectedly)

[seanlenton-oss]

JASP Version

0.95.0

Commit ID

No response

JASP Module

Reliability

What analysis are you seeing the problem on?

Coefficient Alpha and Guttmans Lambda

What OS are you seeing the problem on?

Windows 10

Bug Description

This analysis terminated unexpectedly.

Error in psych::principal(dataset): I am sorry: missing values (NAs) in the correlation matrix do not allow me to continue.
Please drop those variables and try again.

Stack trace
tryCatchOne(expr, names, parentenv, handlers[[1]])

doTryCatch(return(expr), name, parentenv, handler)

withCallingHandlers(expr = analysis(jaspResults = jaspResults, dataset = dataset, options = options), error = .addStackTrace)

analysis(jaspResults = jaspResults, dataset = dataset, options = options)

.frequentistPreCalc(jaspResults, dataset, options, datasetOld)

gettextf('%1$s%2$s', model[['footnote']], .checkLoadings(dataset, options[['variables']]))

sprintf(gettext(fmt, domain = domain, trim = trim), ...)

.checkLoadings(dataset, options[['variables']])

psych::principal(dataset)

stop('I am sorry: missing values (NAs) in the correlation matrix do not allow me to continue.nPlease drop those variables and try again.')

To receive assistance with this problem, please report the message above at: https://jasp-stats.org/bug-reports

Expected Behaviour

When I used the file previously, it did not give the same error.
The error occurs with the customer service, quality and E-commerce Platform data.

Steps to Reproduce

1. Just carry out the reliability test by adding each titled data set and when you add any of the three mentioned in the Log below, you get this error?

Log (if any)

The error occurs with the customer service, quality and E-commerce Platform data.

Feedback Summary JASP Ordered.csv

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.

[tomtomme]

@seanlenton-oss
there are indeed some variables with many missings:

I do not think we can do anything about this, can we @juliuspfadt ?

The error is also present in 0.95.1, but I think the error message is pretty clear on what to do.

-------- Application Info --------
JASP Version: JASP 0.95.1
Build Branch: HEAD
Build Date: Aug 21 2025 15:02:20 (Netherlands)
Last Commit: 180d14c18d13c107ee7c16b0993ff7e1cc02680b

-------- Basic Info --------
Operating System: KDE Flatpak runtime
Product Version: 6.8
Kernel Type: linux
Kernel Version: 6.15.9-2-MANJARO
Architecture: x86_64
Install Path: /app/bin
Platfotm Name: wayland
System Local: de_DE

[seanlenton-oss]

Morning. Apologies if this is due to my misunderstanding as I am learning the program.
The data is not missing per say, it's just that each title has that number of comments, so more feedback information was given against quality versus price, for example. Thanks.

[juliuspfadt]

@seanlenton-oss, thanks for creating the issue. When the data is read, there is a background function that checks if some of the variables are negatively correlated with the scale, which usually indicates that a variable should be reversely scaled. With your data this check fails because when the program tries to disregard all the missing values what is left are variables with zero variance. They are impossible to use in any meaningful analysis. You are saying that this did not give the same error previously, I checked with 0.19.3 version of JASP and get the same error. What do you mean?

[seanlenton-oss]

Thanks @juliuspfadt, I honestly thought that was the case, but I have a large amount of data and multiple sets, so it's very likely it was one of the other sets then. I was also getting used to the format of how the data needs to be arranged. Apologies. I understand why the analysis failed now and will keep this in mind.