Hierarchical regression in JASP?
Hi,
One question.
Will it be possible in the future versions of JASP to run hierarchical regressions wit more than 2 blocks? Right now I can only get two models by using the "tick" function under Models.
Per (the JASP over 😍)
Comments
Not a JASP person, but I think this question should be placed here https://jasp-stats.org/feature-requests-bug-reports/
But you can run them sequentially in different tabs, right? So you test for predictor set B, with set A marked as nuisance; then you do "duplicate analysis", add set B as nuisance to set A (so A+B is nuisance now), and test for added set C, etc.
E.J.
Hi E.J.
That is great👍🏻. But then the predictors in each block are not accounted for? Like for example in a 3-block hierarchical regression in SPSS. Or?
Kind regards
Per
I am not sure what you are saying -- what predictor is not accounted for? Could you give a concrete example?
E.J.
Hi EJ,
I'm not exactly sure what Per means either, and I think it is possible to mimic the results of the typical hierarchical regression with blocks, in the way that you suggest. That being said, it does seem to be an often requested feature, and would certainly be more convenient for the user if blocks could be created within one analysis without duplication. It seems to be often requested on Github. Indeed I have found at least two separate issues requesting this, one of which by me!
Request: Block Function for Hierarchical Linear Regression · Issue #176 · jasp-stats/jasp-issues (github.com)
[Feature Request]: multivariate multiple regression and hierarchical linear regression (entering predictors in blocks) · Issue #1565 · jasp-stats/jasp-issues (github.com)
I'm well aware you have a significant number of more interesting features to work on, but I do hope you will give this some further thought and maybe a priority nudge. It will save time for a substantial number of users.
Best,
Tarandeep