scoreci: Wilson Score Confidence Intervals

In rifttable: Results Tables to Bridge the Rift Between Epidemiologists and Their Data

Wilson Score Confidence Intervals

Description

"This function computes a confidence interval for a proportion. It is based on inverting the large-sample normal score test for the proportion." (Alan Agresti, who wrote the original R code)

Inputs for success, total, and level are vectorized.

Usage

scoreci(success, total, level = 0.95, return_midpoint = FALSE)

Arguments

success	Success count.
total	Total count.
level	Optional. Confidence level. Defaults to 0.95.
return_midpoint	Optional. Return midpoint of confidence interval? Defaults to FALSE.

Value

Data frame:

• success Success count

• total Total count

• estimate Proportion

• conf.low Lower bound of the confidence interval.

• conf.high Upper bound of the confidence interval.

• midpoint Mid-point of the confidence interval (for return_midpoint = TRUE).

• level Confidence level.

See Also

https://users.stat.ufl.edu/~aa/cda/R/one-sample/R1/index.html

Agresti A, Coull BA. Approximate is better than "exact" for interval estimation of binomial proportions. Am Stat 1998;52:119-126. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.2307/2685469")}

Brown LD, Cai TT, DasGupta A. Interval estimation for a binomial proportion (with discussion). Stat Sci 2001;16:101-133. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1214/ss/1009213286")}

Examples

scoreci(success = 5, total = 10)
scoreci(success = c(5:10), total = 10, level = 0.9)

rifttable documentation built on June 8, 2025, 1:52 p.m.