Roediger and Karpicke (2006) investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to material. They randomly assigned participants to one of two study conditions (study-study or study-test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. How should the analysis of these data be labeled