Dear All
I am trying to calculate the number of clusters necessary for a
two-stage cluster sampling design for
A) a cross-sectional survey, basing the sample size on a given
prevalence
B) a longitudinal survey, basing the sample size on a given disease
intensity.
I've tried to used Levy and Lemeshow's formula for A) which calculates a
sample size based on proportions, but it seems inappropriate to use with
my available data (I have data from individuals within schools which
consist of binary responses ie does the person have the disease or
not?).
Can anyone help? Replies much appreciated!
Sarah Whawell
Research Statistician
Schistosomiasis Control Initiative
Department of Infectious Diseases and Epidemiology
Imperial College London
St Mary's Campus
Norfolk Place
London
W2 1PG
Tel. 0207 594 3820
|