Problem: In a study on effectiveness of therapy in treating depression, participants are randomly assigned to groups so that 10 receive cognitive behaviour therapy (CBT) and 10 are assigned to a no-treatment control condition. The outcome improvement scores for participants in the CBT group are quite different from each other (e.g., some had minimal improvement scores, while others showed large benefits). This discrepancy is referred to as: The third variable problem The directionality problem Within-group variance Between-group variance