Skip to main content

Table 4 Design of GeDI forms

From: Testing validity inferences for Genetic Drift Inventory scores using Rasch modeling and item order analyses

  Position 1 Position 2 Position 3 Position 4
GeDI form 1 sequence Scenario 1 Scenario 2 Scenario 3 Scenario 4
GeDI form 2 sequence Scenario 2 Scenario 3 Scenario 4 Scenario 1
GeDI form 3 sequence Scenario 3 Scenario 4 Scenario 1 Scenario 2
GeDI form 4 sequence Scenario 4 Scenario 1 Scenario 2 Scenario 3
  1. The presentation sequence of the four scenarios (and accompanying item suites that comprise the GeDI) was rotated among four equivalent test forms using to a Latin square design. A Latin square is an array of n rows and n columns, with each row and each column containing units 1 through n exactly once. By rotating experimental treatments in this manner, the researcher can generate data to determine whether an adjacent treatment (or, in this case, item suite placement) influences overall performance. For a description of scenarios and associated items see Table 2