1. data were collected by Non probability sampling Techniques.
2. Data is non- normal.
with references.
Thanks in Advance.
@Shahul PLS Approach does not require the normality assumption of the data. You can use Smart-PLS to run SEM model on non normally distributed data.
Yes you can but the statistical tests will be meaningless since they require a probability sample. See any statisics book to find out why this is so. David Booth
I don't think it would be a good idea to go for SEM. However, as mentioned by @Imran Anwar, PLS being a robust approach could be used to run the analysis. But @Shahul I am interested in knowing about the sample size and other sample representativeness attributes (other than normality); whether these criteria are fulfilled?
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling