.-
help for ^sampclus^                                                (STB-60: sxd4)
.-

Sample Size for Studies with Cluster Sampling
---------------------------------------------

   ^sampclus^ [^, obs^clus^(^#^) num^clus^(^#^) rh^o^(^#^)^]


Description
-----------

^sampclus^ calculates sample size and number of clusters for cluster sampled
studies, correcting for any intraclass correlation. ^sampclus^ uses the
estimates from ^sampsi^, which must precede it.


Options (Note: must choose either ^obsclus^ or ^numclus^, but not both)
-------

^obsclus(^#^)^ -- number of observations in each cluster; will return 
    corrected sample sizes and minimum number of clusters needed

    Example: ^obsclus(10)^ -- 10 patients per physician; sampclus 
                            determines number of physicians needed

^numclus(^#^)^ -- maximum number of clusters; will return corrected sample
    sizes and number of observations in each cluster. Number of obs.
    per cluster are rounded up so that the number of clusters will
    never exceed ^numclus^, but may be fewer. If too few clusters are
    requested for the other given parameters, ^sampclus^ will prompt
    for the minimum number of clusters possible.

    Example:  ^numclus(40)^ -- 40 physicians chosen; sampclus determines
                             number of patients per physician

^rho(^#^)^ -- intraclass correlation coefficient (default=0)
      

Examples
--------

 . ^sampsi 200 185, alpha(.01) power(.8) sd(30)^
 . ^sampclus, obsclus(10) rho(.2)^
 . ^sampclus, obsclus(10) rho(.1)^

   Corrects sample size and computes number of clusters from a t-test; adjusts
   this sample size calculation for 10 observations per cluster and an
   intraclass correlation of 0.2; repeats for an intraclass correlation of 0.1.

 . ^sampsi .20 .30, power(.8) onesided^
 . ^sampclus, numclus(40) rho(.05)^
 
   Calculates sample size and number of observations per cluster using the
   difference of two proportions; assumes the number of clusters is 40, and an
   intraclass correlation of 0.05.


Author
------

Joanne M Garrett, University of North Carolina
joanne_garrett@med.unc.edu