jasp-issues
jasp-issues copied to clipboard
jasp-stats
[Feature Request]: Nonparametric factorial ANOVA and pairwise comparisons
Description
Conduct nonparametric factorial ANOVA and pairwise comparisons
Purpose
No response
Use-case
No response
Is your feature request related to a problem?
Given that data often violate the assumptions of normality and homoscedasticity, especially with some outliers, small sample sizes and an unbalanced design, it will be very helpful to allow us to conduct nonparametric factorial ANOVA (two-way, three-way, four-way, …, N-way ANOVA) and nonparametric pairwise comparisons when there is a significant main or interaction effect as well as bayesian testing.
Describe the solution you would like
No response
Describe alternatives that you have considered
In fact, the aligned rank transform (http://depts.washington.edu/acelab/proj/art/index.html) can be used to handle these problems (Elkin et al., 2021; Wobbrock et al., 2011). However, it is not easy for the beginners. If JASP include these functions in the new version, it will become more popular.
Elkin, L.A., Kay, M., Higgins, J. and Wobbrock, J.O. (2021). An aligned rank transform procedure for multifactor contrast tests. Proceedings of the ACM Symposium on User Interface Software and Technology (UIST '21). Virtual Event (October 10-14, 2021). New York: ACM Press, pp. 754-768.
Wobbrock, J.O., Findlater, L., Gergle, D. and Higgins, J.J. (2011). The aligned rank transform for nonparametric factorial analyses using only ANOVA procedures. Proceedings of the ACM Conference on Human Factors in Computing Systems (CHI '11). Vancouver, British Columbia (May 7-12, 2011). New York: ACM Press, pp. 143-146.
Additional context
No response
Yes, but it would not be high on my priority list - it's a great feature enhancement (now JASP only does nonparametric one-way ANOVA, including follow-up tests), but I am very busy at the moment..
While this issue is about classic and bayesian nonparametric testing for ANOVA (two-way, ... N-way)
- [ ] interaction
- [ ] pairwise comparisons for main & interaction effects (done for classical via Dunns post hoc)
When this gets implemented we should also tackle e.g. missing non-parametric effect sizes:
- [ ] for tTests https://github.com/jasp-stats/jasp-issues/issues/450
- [X] for ANOVAs interaction and post hocs https://github.com/jasp-stats/jasp-issues/issues/1905
- [x] move misplaced Dunns post hoc test https://github.com/jasp-stats/jasp-issues/issues/383
