Lecture
An intermediate form of designs between individual and between-group schemes is the cross-individual design.
Cross-individual designs.
- these are experimental designs that involve assigning different sequences, each including all levels of the experimental factor, to different subjects or different groups of subjects.
Cross-individual designs are often used for multilevel experiments.
A multilevel experiment is one in which the experimental factor is represented at more than two levels.
Designs:
1) Reverse counterbalancing.
To two groups of subjects, the conditions are presented in forward and reverse order (ABC to the first group, BCA to the second). The averaging of the dependent-variable measures is then carried out over the total number of trials separately for each level of the independent variable.
2) Complete counterbalancing (control of sequence effects):
- the Latin square
- the balanced Latin square
These are designs for presenting all levels of the experimental factor either to the same subject or, through a single presentation of each micro-sequence, to different subjects. The balanced square differs in that within it every level of the independent variable is immediately preceded exactly once by every other level. However, centration and edge effects are present.
For cross-individual designs, the selection of subjects into groups is the same as in the case of between-group designs. The advantage of cross-individual designs lies in the fact that functional relationships can be constructed on their basis.
Comments