I have around 40 groups. They have different sample sizes. For example in first group 10 people responded, in second group 25 people responded, in third group 5 people responded and so on. The standard deviation of these groups is very different. It is a rating data (not continuous), and distribution is not normal. Can you please guide on how to compare mean difference between these groups statistically ?
Used a non-parametric test Mann-whitney test or separate variances 2-sample t-tests in this case
My Dear Pant..
You can use non-parametric test.. or the best use robust variance analysis (R0bust ANOVA) with 40 groups each group with r replica (sample size)
Regards
Hi. Bootstrap hypothesis for the difference of means for each pair.
https://campus.datacamp.com/courses/statistical-thinking-in-python-part-2/introduction-to-hypothesis-testing?ex=13#skiponboarding
http://courses.washington.edu/matlab1/Bootstrap_examples.html
http://www2.hawaii.edu/~taylor/z631/ips7e_ch16.pdf
Regards
Check these links:
https://www.graphpad.com/support/faq/how-to-compare-two-means-when-the-groups-have-different-standard-deviations/
http://www.napavalley.edu/people/cburditt/Documents/Math%20232%20-%20Statistics/Sullivan%20ANOVA.pdf
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size