I know there are various rules of thumbs such as 5-20 cases per variable in a scale or recommendations like Comrey and Lee's. Also it depends on communalities. But is there any solid way to justify that your sample size is adequate?
Hi,
Most research and statistics books require the minimum sample size in EFA to be 5 cases per each item in the scale.
Hi,
Most research and statistics books require the minimum sample size in EFA to be 5 cases per each item in the scale.
For sensitivity and specificity you can consider Buderer method if you know the prevalence of the disease.
This method can be applied in validation of diagnostic tests. I am not sure if it appropriate to your study.
Please see the attached link for more information.
Good luck
http://www.ncbi.nlm.nih.gov/pubmed/8870764
Ahmed, you could perform a Monte Carlo simulation study if you already have some knowledge about the population values to be expected.
Or, you may decide to read what other researchers think about this problem. Here is a great resource with several suggestions and references gathered by Nathan Zhao which may be helpful for you:
https://www.encorewiki.org/display/~nzhao/The+Minimum+Sample+Size+in+Factor+Analysis
Regards, Karin
Hi
Most research and statistics books require the minimum sample size in EFA to be 5 -10 cases per each item in the scale. 100-200 cases should be "good enough" for 20-50 items.....but for all.....100 as poor, 200 as fair, 300 as good, 500 as very good
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size