coupon_score: Coupon Score

In TImA97/randfindR: Analysis of Randomness in Human Generated Sequences

Coupon Score

Description

Compute average digit length required for all responses to occur

Usage

coupon_score(x, options)

Arguments

x	vector of distinct options of numbers or characters
options	number of available options in sequence

Details

This function takes a vector x and computes the mean number of values that is required for all possible options to occur. This means that the algorithm starts to count at the beginning of the sequence until all options have occurred. Having observed all options, the counting process starts again at zero at the next value of the sequence. Finally, the mean over the length of these complete sets of values (containing all options) is computed. Incomplete sets of responses at the end of a sequence with at least one option missing are not used for the computation of this index. Consequently, this index cannot be computed for vectors that do not contain all possible options and therefore, NA is returned. The lowest possible score equals the number of options, indicating that each complete set contains every possible option only once. A higher score indicates that at least one complete set contains values of the same kind at least twice.

Value

Coupon Score of x

References

Ginsburg N, Karpiuk P. Random Generation: Analysis of the Responses. Perceptual and Motor Skills. 1994;79(3):1059-1067. \Sexpr[results=rd]{tools:::Rd_expr_doi("doi:10.2466/pms.1994.79.3.1059")}

Towse, J.N., Neil, D. Analyzing human random generation behavior: A review of methods used and a computer program for describing performance. Behavior Research Methods, Instruments, & Computers 30, 583–591 (1998). \Sexpr[results=rd]{tools:::Rd_expr_doi("doi:10.3758/BF03209475")}

Examples

runs_index(ginsburg1994, 10)
runs_index(evans1978[, 1], 10)
runs_index(evans1978[, 2], 10)

TImA97/randfindR documentation built on July 1, 2024, 7:56 p.m.