PCS: the Design

In order to allow accurate estimation of the preference associated with each item, the PCS statistical design must be properly prepared and must satisfy some balance criteria.

  • one-way balance: items are included in the design a similar amount of times;
  • two-way balance: each paired combination of items appear together (on the same paired task) in the design a similar amount of times;
  • positional balance: each item appears in each position (left and right) a similar amount of times;
  • within-version balance: items appear within each version (block) a similar amount of times.

The function generate.design.PCS available in R-sw Tradeoff generates a PCS design. It is possible to specify prohibitions, i.e., paired combination of statements that should not appear in the design. In addition, the function automatically provides a report with the design properties, i.e., how it is performing on the various balance criteria.

The function check.design.PCS produces summary statistics for an existing PCS design.