[Feature Request]: GRUBBS test for outliers

[niersparker]

Description

Die Analyse soll Ausreißer innerhalb einer Stichprobe erkennen, eliminieren und durch Iteration die verbleibende Stichprobe verbessern.

Purpose

Datenbereinigung

Use-case

Datenanalyse von Immobiliendaten

Is your feature request related to a problem?

No response

Is your feature request related to a JASP module?

Descriptives

Describe the solution you would like

No response

Describe alternatives that you have considered

No response

Additional context

No response

[EJWagenmakers]

You can use the "filter" functionality to define and eliminate outliers. If this does not address your concern please reopen the issue and clarify what functionality is missing.

[tomtomme]

@EJWagenmakers
this is about the Grubbs test: https://en.wikipedia.org/wiki/Grubbs%27s_test
I do not think we have that one.

@niersparker
We have Boxplots and Residual plots to detect outliers. Does this suffice?

[EJWagenmakers]

There are several such tests...we could add them (and other techniques) as a separate module ("Outliers") if there is sufficient interest.

[niersparker]

A good idea!

[TarandeepKang]

Seconded! This sounds very useful!

[tomtomme]

@niersparker
also note that besides boxplots / rainclouds we also have e.g.

to identify outliers in regression.
Why would you want to put this in descriptives? Usually one wants to detect outliers in context of their analysis so for the model-residuals, not for the single variables that combined give the model.