The objectives of the study are to identify the psychological, socio-demographic and environmental factors that may lead to early smoking initiation of adolescents. What formula should I use? I am planning to do a logistic regression analysis for this study.
I am not sure that there is an anlytic answer with a simple 'formula'. If I am not mistaken you plan perform multiple logistic regressions with several confounding variables. If you have any idea of the variance of your factors (e.g. from literature or, better, from initial brief studies) you might want to simulate your beta(s) by simulating your factors with the given variance.
Take a look at this paper for an example: http://www.biomedcentral.com/1471-2288/9/56
This might not be very helpful, I fear.
ops, I meant: ... to simulate your beta(s) by simulating your factors with the given variance to yield an idea of the required sample size.
Just to avoid serious misunderstandings.
If you do not know how to do that, ask a statistician at your university. (This backup is also recommended in case I misunderstood your question. ;-) )
Your reference is really helpful. I used the formula for logistic regression analysis. :D
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size