I have a 14items questionnaire (Likert data) with 3 variables and sample size 60 (non normally distributed). Guess SEM wouldn't work. Any other alternative?
Well you haven't really described your data but I would suggest ordinal logistic regression. See the attached notes for details. Hope this helps. Best wishes,David Booth
Partial Least Square (PLS) based approaches are appropriate for data not meeting (violating) the assumption of normality. You can use WarpPLS, SmartPLS etc softwares to run Path Analysis, SEM etc. When performing SEM, Bootstrapping would be better option to overcome issues related to sample size.
Use the categorical data analysis such as the Chi-Square test to assess the relationship between row variable and column variable. But the result is very naive.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size