Transforming variables and other improvements

Discuss the jamovi platform, possible improvements, etc.

by Manuel » Sun Dec 15, 2019 11:46 am

We have done the last computing class with jamovi after using SPSS for 9 sessions. The students liked the simplicity and instant results updates. Some issues:

1. They found very confusing the way to transform from continuous to categorical or from categorical to categorical, compared to SPSS with the two-sided boxes (old - new).

2. In regression analysis, it would be interesting to save the predicted and residuals.

3. The non-parametric 2 means comparison (Mann-Whitney) is in the T-Test menu, it would be cleared is separated like in the ANOVA menu below "Non-parametric", or move all non-parametric tests to a specific menu (like in SPSS).

4. The Exploration menu would be better for students distinguishing from the beginning Continuous and Categorical (Nominal and ordinal).

5. In regression analysis, would it be possible to include any heterocedasticity test (BPG, White, etc.)?

Anyway, great software!
Regards
Manuel
Manuel
 
Posts: 6
Joined: Fri Nov 02, 2018 12:46 pm

by coledavis » Sun Dec 15, 2019 10:22 pm

With regards to your 3rd point. It depends. If you want to do it merely so that it is familiar to SPSS users, then it is only a short-term advantage. Personally, I think that having the parametric test and its non-parametric equivalent together makes sense: if when you run a parametric test, you find that there is an assumption error, the more robust test is ready to hand. If the two tests serve the same purpose, functionally, then it doesn't seem to make sense to me to separate them on mathematical grounds.
coledavis
 
Posts: 27
Joined: Sun Jul 02, 2017 1:35 pm


Return to General