.crawford.CI {singlecase}R Documentation

Confidence Intervals based on a non-central t distribution

Description

.crawford.CI is an internal function that computes 95% confidence intervals around the rarity of a score (Crawford & Garthwaite, 2002). These confidence intervals are based on non-central $t$-distributions.

Usage

.crawford.CI(c,n)

Arguments

c test statistics
n number of control subjects

Details

c is an observation from a non-central $t$-distribution on n - 1 degrees of freedom. Confidence intervals are based on two values of the non-centrality parameter of non-central $t$-distribution, such that the resulting non-central $t$-distribution has $c sqrt{n}$ as its $100 α / 2 $ percentile (value 1) or as its $100 (1-α/2)$ (value 2) (for details, see Crawford & Garthwaite, 2002). The purpose of the .crawford.CI function is to find those values.

Value

2.5% Lower bound of the confidence interval.
97.5% Upper bound of the confidence interval.

Author(s)

Matthieu Dubois. matthdub@gmail.com, http://www.code.ucl.ac.be/MatthieuDubois/r_code.html

References

Crawford, J. R., & Garthwaite, P. H. (2002). Investigation of the single case in neuropsychology: Confidence limits on the abnormality of test scores and test score differences. Neuropsychologia, 40(8), 1196–208.

See Also

crawford.t.test, crawford.diff.test


[Package singlecase version 0.1 Index]