[Bug]: Coffecient table in logistic regression display wrong value when reversing reference value (categorical data)

[PerPalmgren]

JASP Version

0.19.0

Commit ID

No response

JASP Module

Regression

What analysis are you seeing the problem on?

Logistic regression

What OS are you seeing the problem on?

Windows 10

Bug Description

In logistic regression. When switching the reference category in the label editor, thus, changing 0=No and 1=Yes the categorical value in the table thus not follow the change. The footnote changes but no the value in the table. (See screen shot)

Expected Behaviour

The table should follow the change of reference category.

Steps to Reproduce

1.See screen shot 2. 3. ...

Log (if any)

No response

More Debug Information

No response

Final Checklist

• [X] 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]

@PerPalmgren which value exactly do you mean? Most of them are changing the sign as expected.

[PerPalmgren]

@tomtomme In the left column “excercise prescription” in the second screen shot, the label with (Yes) ought to change to (No) when changing reference category, thus corresponding with the footnote. Now this does not make great sense.🙄 See highlighted screen shot below.

[PerPalmgren]

[JohnnyDoorn]

Hi @PerPalmgren,

If I understand correctly, the footnote describes the dependent variable (Heart attack) reference class, not that of the independent variable (Exercise). The results will stay the same, but flipped (Exercise will either decrease probability of heart attack, or decrease probability of no heart attack).

Cheers Johnny

[PerPalmgren]

@JohnnyDoorn You can close this issue because it was me who was wrong. JASP works fine.😜 All the best Per