MAKEUP AND TIPPING 285 PROCEDURE A waitress (22 years old, three years of restaurant practice), regularly employed, was used to help us. She was not informed of the goals of our experiment and she did not receive any information about previous studies on the effect of cosmetics on people’s perception or behavior. Precaution was taken to select a waitress who usually wore makeup herself, to help us. The experiment was conducted each day for six weeks (excluding Saturdays and Sundays) during the lunch hour, because there was a suffi cient amount of patrons who were alone at this time in the restaurant in which the experiment was carried out. Thirty observation periods were obtained (fi ve days a week for six weeks). The two experimental conditions used were 16 lunches with makeup and 15 lunches without makeup. Another volunteer was a young female beautician, who “made up” the waitress for the duration of the exper- iment. In the makeup-free condition, the beautician cleaned and moisturized the face of the waitress. In the makeup condition, the beautician applied makeup to her eyes, cheeks, and lips in a way that enhanced the attractiveness of the waitress. The confederate was instructed to act as usual. She acted in the same way in both condi- tions when she returned to the patron’s table with a drink and the bill. Once the patron had left, the waitress returned to the table to clear it. She reported in a notebook how the client had behaved, namely, whether he/she had left a tip or not, and how much he/she had given her. A 20-year old male, who was standing in a separate room, observed the restaurant room with the help of a video security system that previously existed in the restaurant. He was instructed to leave the restaurant through a separate exit until he saw that the patron, who was alone, stood up to leave the restaurant. The observer waited until the patron was out- side and approached him/her and said, “Excuse me sir/madam, I am a student in social sciences and I am conducting a study on the judgment of individual appearance, such as physical attractiveness, apparel, and so on. Would you be willing to respond to one question about the physical appearance of somebody you have met very recently?” Once the par- ticipant agreed, which all of them did, the interviewer asked the participant to rate the physical attractiveness of the waitress with whom he had interacted in the restaurant. The evaluation was performed with a scale from 0 = low physical attractiveness to 9 = high physical attractiveness. The interviewer noted the rating of the participant, thanked him/ her, and then reentered the restaurant to wait for the waitress in the kitchen, where it was not possible for the patrons to see him. When the waitress arrived, he described the partic- ipant to the waitress in order to obtain data about his/her tipping behavior. In each case, the waitress was informed about the interaction between the interviewer and the participant. RESULTS In France, giving a waiter or a waitress a tip is an unusual practice as French legislation mandates that a 12% service charge is to be included in the cost of the item on the menu. Thus, in this experiment, the number of patrons who left a tip was the fi rst dependent variable and the amount left by the patron who gave a tip was the second dependent vari- able. The rating of the waitress’s physical attractiveness was the third variable. Data are presented in Table I.

JOURNAL OF COSMETIC SCIENCE 286 With the number of participants who gave a tip, a loglinear analysis of the 2 (patron gender) × 2 (experimental condition) × 2 (compliance) design was performed. A signifi - cant interaction effect between the experimental conditions and the patrons’ behaviors was observed, revealing that, overall, patrons tipped the waitress while she was wearing makeup more often than when she was not wearing makeup (χ2(1, N = 174) = 4.16, p = .04, r = .15). However, this signifi cance was found only with male patrons (χ2(1, N = 112) = 4.51, p = .03, r = .20), not with female patrons (χ2(1, N = 62) = 0.20, ns, r = .06). Finally, a near signifi cant main effect of gender on tipping behavior was found: male pa- trons gave tips more often than female patrons (χ2(1, N = 174) = 3.68, p = .06, r = .14). With the 70 patrons who gave tips, a 2 (patron gender) × 2 (experimental condition) ANOVA analysis was performed with the amount of money given as the dependent var- iable. A slightly main effect of the experimental condition was found (F(1, 66) = 3.45, p = .07, eta2 = .05), whereas neither the effect of participant gender nor the interaction between participant gender and experimental condition was signifi cant (p .20). However, when considering each group according to gender, it was found that male patrons participating in the makeup condition gave more money to the waitress than they did in the non- makeup control condition (t(49) = 2.34, p. = .02, d = .67). No signifi cant difference was found when comparing the female patrons in the two experimental conditions (t(17) = 1.09, ns, d = .53). With the ratings of physical attractiveness of the waitress, a 2 (tipping behavior) × 2 (patron gender) × 2 (experimental condition) ANOVA analysis was performed. A main effect of the experimental condition was found (F(1, 166) = 60.77, p .001, eta2 = .27), revealing that the rating of physical attractiveness was higher in the makeup condition (M = 6.17) than in the non-makeup control condition (M = 6.17). A main effect of tip- ping behavior was found (F(1, 166) = 8.27, p .001, eta2 = .05): participants who had given a tip to the waitress rated her more favorably (M = 6.04) than participants who had not given a tip (M = 5.38). A main effect of participant gender was found (F(1, 166) = 20.55, p .001, eta2 = .11), revealing that men’s ratings of physical attractiveness were higher (M = 5.86) than women’s ratings (M = 5.12). Two-way interaction analysis re- vealed no signifi cant effect (p .10 in each case), whereas the three-way interaction ap- peared to be near signifi cant (F(1, 166) = 3.46, p = .06, eta2 = .02). Table I Percent of Customers Who Gave a Tip and Amount of Tip Given According to Experimental Condition and Sex of Customer Male patrons Female patrons Makeup No makeup Makeup No makeup N = 58 N = 54 N = 30 N = 32 Percentages of patrons who gave a tip 55.2% (32/58) 32.2% (19/54) 33.3% (10/30) 28.1% (9/32) Mean (SD in brackets) of amount of tip given (in euros) 1.61 (0.81) 1.12 (0.74) 1.15 (0.34) 1.00 (0.25) Mean (SD in brackets) of rating of physical attractiveness of waitress Among participants who gave a tip 6.63 (0.75) 5.68 (0.74) 6.00 (0.67) 4.77 (0.67) Among participants who did not give a tip 6.19 (0.80) 5.00 (0.77) 5.5 (0.69) 4.96 (0.71)