Matrix Sampling Design

Share:

Description

This function produces the forms, each including a subset of pairs of stimuli, for use in a matrix sampling design.

Usage

1
matrix.sampling(x, paper = FALSE, f = 2)

Arguments

x

a vector of stimuli

paper

logical value. If paper=TRUE, the stimuli will not be randomized so that the results in Davison, Zieffler, Cabrera, and Karl (in press) can be replicated. The default value is paper=FALSE.

f

an integer value specifying the number of forms to create. The default is f=2.

Value

A data frame is returned. The data frame includes the first and second stimulus to be presented in each pair of stimuli. It also includes the block number for matrix sampling and the form to which the pair should be assigned.

Author(s)

Andrew Zieffler zief002@umn.edu, Stacy Karl karlx028@umn.edu, and Julio Cabrera cabre031@umn.edu

References

Davison, M. L. Zieffler, A. Cabrera, J. Karl, S. R. and Cohen, H. S. in press Automated Assembly of Optimally Spaced and Balanced Paired Comparisons: Controlling Order Effects. Behavioral Research Methods. DOI 10.3758/s13428-011-0170-0

Cohen, H.S. and Davison, M.L. 1973 Jiffy-Scale: A FORTRAN IV Program for Generating Ross-Ordered Paired Comparisons. Behavioral Science, 18, 1, 76.

Ross, R.T. 1934 Optimimum Orders for the Presentation of Pairs in the Method of Paired Comparisons. Journal of Educational Psychology, 25, 5, 375–382.

See Also

forms, ross.ordering, incomplete.design, big.nine, big.eight

Examples

1
2
3
4
5
6
7
## To choose the ideal number of forms
## forms(9)

## Matrix sampling design of the Big Nine data set for 2 forms
## Reproduces Table 3 in Davison, Zieffler, Cabrera, and Karl (in press)
data(big.nine)
matrix.sampling(big.nine, f = 2, paper = TRUE)