Many foods and beverages, such as coffee and scotch, are frequently termed "acquired tastes." Pliner (1982) was interested in investigating whether familiarity with a flavor leads to greater approval-the "acquired-taste" phenomenon. She had 24 undergraduates taste each of four unfamiliar tropical fruit juices 0, 5, 10, or 20 times. (For each participant she randomized which juice would be tasted how many times.) Participants were then asked to rate the degree to which they liked the taste of the juice. What statistical procedures are suitable for analyzing these data? What test would she use if she wanted a distribution-free test? (Note: She found the effect that she had expected-greater familiarity led to greater approval.)