Learn R Programming

sensR (version 1.5-4)

discrimSim: Simulates replicated difference tests

Description

Simulates the outcome of sample.size replicated sensory difference tests (for any one of eight protocols: 2-AFC, 3-AFC, duotrio, tetrad, triangle, two-out-of-five, two-out-of-five with forgiveness and hexad tests) for a given d-prime value and a given overdispersion (default 0).

Usage

discrimSim(sample.size, replicates, d.prime, sd.indiv = 0,
           method = c("duotrio", "halfprobit", "probit", "tetrad",
             "triangle", "twoAFC", "threeAFC", "hexad", "twofive", "twofiveF"),
           double = FALSE)

Value

A vector of length sample.size with the number of correct answers for each subject.

Arguments

sample.size

the sample size - number of subjects

replicates

number of replications per subject

d.prime

the value of d-prime

method

the discrimination protocol

sd.indiv

the individual variability in d-prime values. A value of 0 (default) corresponds to complete independence

double

should the 'double' variant of the discrimination protocol be used? Logical scalar. Currently not implemented for "twofive", "twofiveF", and "hexad".

Author

Rune Haubo B Christensen and Per Bruun Brockhoff

Details

The d-prime for each subject is a random draw from a normal distribution with mean d.prime and standard deviation sd.indiv. All negative values are set to zero.

References

Brockhoff, P.B. and Christensen, R.H.B. (2010). Thurstonian models for sensory discrimination tests as generalized linear models. Food Quality and Preference, 21, pp. 330-338.

See Also

triangle, twoAFC, threeAFC, duotrio, tetrad, twofive, twofiveF, hexad, discrimPwr, discrim, AnotA, discrimSS, samediff, findcr

Examples

Run this code

## Running simulations:
discrimSim(sample.size = 10, replicates = 3, d.prime = 2,
           method = "triangle", sd.indiv = 1)
# \dontshow{
set.seed(1)
a <- discrimSim(sample.size = 10, replicates = 3, d.prime = 2,
           method = "triangle", sd.indiv = 1)
expect <- c(3L, 3L, 2L, 3L, 3L, 1L, 3L, 3L, 1L, 2L)
stopifnot(all(a == expect))
# }

Run the code above in your browser using DataLab