findcr {sensR} | R Documentation |
Finds the critical value in a one-tailed binomial test
findcr(sample.size, alpha = .05, p0 = .5, pd0 = 0, type = c("difference", "similarity"))
sample.size |
the sample size of the binomial test (must be a positve integer) |
alpha |
the type I error-level of the test (must be between zero and one) |
p0 |
the guessing probability under the null-hypothesis (must be between zero and one); 1/2 for the duotrio and twoAFC tests and 1/3 for the triangle and threeAFC tests |
pd0 |
the proportion of discriminators in the population of interest |
type |
the type of test |
The critical value of the standard one-tailed difference test of "no
difference" is obtained with pd0 = 0
.
The probability under the null hypothesis is
given by pd0 + p0 * (1 - pd0)
.
The critical value in a one-tailed binomial test, that is, the smallest integer such that the null hypothesis binomial probability of being larger (smaller for similarity hypotheses) than or equal to this number is smaller than or equal to the type I error-level of the test.
Rune Haubo B Christensen and Per Bruun Brockhoff
triangle
, twoAFC
,
threeAFC
, duotrio
,
discrim
, discrimPwr
,
discrimSim
, AnotA
discrimSS
, samediff
## Find the critical value for a triangle test for the level 0.05 test ## with 25 subjects: findcr(sample.size = 25, , p0 = 1/3) ## Similarity example: findcr(sample.size = 25, p0 = 1/3, pd0 = .2, type = "simil")