(1) How to analyze relation between two variables in which data is obtained on the ordinal scale from two different groups and the data sets are asymmetric, where one group has significantly more responses than the other? (2) How does SMART PLS address this issue for unequal numbers of observations between the predictor and criterion variables? If not, what other tools are appropriate? (3) What are some assumptions of this type of analysis? What things are important to consider before starting the data collection?
I think for the ordinal nature of the data and the unequal group sizes, non-parametric tests like the Mann-Whitney U test or the Kruskal-Wallis test. Also ordinal logistic regression may be considered.
can you share a part of the data , so as we can think together and analyse
Abhishek Pandey I am not at the data collection stage yet. The study is still in literature review and conceptual phase. This is my first time to handle the different groups in a way described above. That is why I wanted to ask the opinions of peers before going through with this research.
- Badges
- Science topic