Returns the necessary sample size to achieve a given width of a binomial confidence interval, as calculated by BinomCI(). The function uses uniroot() to find a numeric solution.
BinomCIn(p = 0.5, width, interval = c(1, 100000),
conf.level = 0.95, sides = "two.sided", method = "wilson")probability for success, defaults to 0.5.
the width of the confidence interval
a vector containing the end-points of the interval to be searched for the root. The defaults are set to c(1, 100000).
confidence level, defaults to 0.95.
a character string specifying the side of the confidence interval, must be one of "two.sided" (default),
"left" or "right". You can specify just the initial letter. "left" would be analogue to a hypothesis of
"greater" in a t.test.
character string specifing which method to use; this can be one out of:
"wald", "wilson", "wilsoncc", "agresti-coull", "jeffreys",
"modified wilson", "modified jeffreys", "clopper-pearson",
"arcsine", "logit", "witting" or "pratt". Defaults to "wilson".
Abbreviation of method are accepted. See details in BinomCI().
a numeric value
The required sample sizes for a specific width of confidence interval depends on the proportion in the population. This value might be unknown right from the start when a study is planned. In such cases the sample size needed for a given level of accuracy can be estimated using the worst case percentage which is p=50%. When a better estimate is available you can you can use it to get a smaller interval.
BinomCI()
# NOT RUN {
BinomCIn(p=0.1, width=0.05, method="pratt")
# }
Run the code above in your browser using DataLab