ABSTRACT
How to evaluate effect size and conduct power analysis in such designs is also discussed. Worked examples of each of the stages are given in Appendix VIII.
Interactions between IVs An interaction is where a pattern which is found across the levels of one IV differs according to the levels of another IV. Imagine that, in a study of face recognition, researchers have two IVs: familiarity, with two levels-familiar and unfamiliar (that is, familiar and unfamiliar prior to the experiment); and orientation of face, with two levels-correct and upside down. The DV is time taken to recognise a face shown in a photograph. An interaction between the two variables would exist if familiar faces were more quickly recognised than unfamiliar faces when they were presented in the correct orientation, while there was no difference in speed of recognition between the two levels of familiarity when faces were presented upside down. Orientation could be described as a moderator variable because it moderates the relationship between degree of familiarity and reaction time. Similarly, familiarity could be seen as a moderator of the relationship between orientation and reaction time.
