jasp-stats / jasp-issues

This repository is solely meant for reporting of bugs, feature requests and other issues in JASP.
58 stars 29 forks source link

[Bug]: R error on Nonparametrics in 'Repeated Measures ANOVA' #2021

Closed eddie-uu closed 1 year ago

eddie-uu commented 1 year ago

JASP Version

0.17.0

Commit ID

No response

JASP Module

ANOVA

What analysis are you seeing the problem on?

Friedman Test

What OS are you seeing the problem on?

Windows 11

Bug Description

Using the Nonparametrics feature in 'Repeated Measures ANOVA', I am attempting to perform the Friedman test. For this, I am using the data as shown below, with 3 RM factors such that you can fill all of them with for example value(5, 10, 15), block(1, 2, 3), and sign(negative, positive).

<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

T -15 B1 | T -10 B1 | T -5 B1 | T 5 B1 | T 10 B1 | T 15 B1 | T -15 B2 | T -10 B2 | T -5 B2 | T 5 B2 | T 10 B2 | T 15 B2 | T -15 B3 | T -10 B3 | T -5 B3 | T 5 B3 | T 10 B3 | T 15 B3 -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- 794 | 688.25 | 872.625 | 825.25 | 799.875 | 670 | 581.375 | 598.625 | 707.9375 | 676.625 | 649.1875 | 593.5 | 541.375 | 572.9375 | 607.0625 | 655.6875 | 629.375 | 599.75 664 | 661 | 759.3125 | 891.4375 | 760.4375 | 658.125 | 635.625 | 646.625 | 869.125 | 703.25 | 647.9375 | 582.6875 | 588.8125 | 667.375 | 758.375 | 782.375 | 653.1875 | 651.0625

When using these values, I am receiving the 'Please specify appropriate terms for the Friedman/Durbin' error. image

Expected Behaviour

What I am expecting is that the calculation is processed properly, and shows me the results.

Steps to Reproduce

  1. Use the 'Repeated Measures ANOVA'
  2. Use a data file with values such as given in the bug description, with 3 RM factors
  3. Fill RM cells with all the data columns
  4. Perform the nonparametrics with all factors as RM factors

Log (if any)

No response

Final Checklist

JohnnyDoorn commented 1 year ago

Hi @eddie-uu ,

The grouping factor is a between-subjects factor (other than the ID variable) - see also the helpfile. For this data, I do not think the Friedman test is appropriate, since the Friedman test only goes for one-way ANOVA. Also, only having 2 observations per cell seems a bit problematic (but the main analysis still runs, so I guess that is fine?).

Kind regards, Johnny

github-actions[bot] commented 1 year ago

This issue will be automatically closed in 42 days due to inactivity. Feel free to leave a comment if you believe this is still relevant.

github-actions[bot] commented 1 year ago

Automatically closed due to inactivity.