|
所有的外生变量及其系数将被用来测试多个回归模型来考试关系大小和其他因素倾斜。学习要用一个好的方法，不管是从哪个方面来看,结果都是有意义或者没有意义。在回归测试,它将证明每一个因素的影响大小和移除的小费无关的因素。
All of the exogenous variables and their coefficients will be tested by multiple regression models to exam relationships between tipping size and other factors. T-Test will be a good method to see the results are significant or not. During the regression test, it will prove how every factor influence the tipping size and remove the unrelated factors.
为了知道什么因素会导致尺寸的建议,我们应该考试可能的变量之间的关系。我所有的数据收集在一个餐馆在奥格登,犹他州命名富士龙。这个餐厅是中国和日本的食物搭配吸引了来自不同国家的人。服务器有帮助我填写我的调查和记录他们的工作表现和他们的客户反馈。总数量的观察是52。
Data
In order to know what factor will lead the size of tips, we should exam the relationship between possible variables and tipping size. All my data was collected in a restaurant in Ogden, Utah named Fuji Dragon. This restaurant is served with Chinese and Japanese food which attracts people from different countries. Servers there helped me to fill out my survey and recorded their working performances and their customers’ feedback. The total number of observations is 52. I have got right now. There are 12 questions appeared on my survey which contains how many people were at the table, how much the total bill was, do you introduce yourself to the table and so on. After collecting all the data, I processed the data and set variables as following:
Gender: Male=0, Female=1
Time: Lunch=0, Dinner=1
Frequency: 1-2 times=0, several=1, often=2
Method of payment: Cash=0, Credit Card=1
Food Quality: Excellent=3, good=2, ok=1, not good=0
Service quality: 1(poor) to 10(excellent)
Dressing: Very expensive=3, Somewhat expensive=2 , Average=1, Less Expensive=0
Race: White customers=1, Other races=0
Self introduce: yes=0, No=1
The regression result shows that the P-value of server’s gender, bill size, frequency and food quality are very significant. While other variables are not that significant or have no influence on tipping size.
The coefficient of server’ gender is negative 2.331. Since I set male to 0 and female to 1, it means that male servers can earn more tips than female servers when other variables keep constant. The coefficient of women’s percentage in a table is negative as well. It means that higher rate of women percentage in a table will decrease tipping size. It also means male customers usually pay more tips than female, but the influence is not too significant. The coefficient of the product of server’s gender and women percentage is positive. It explains that people pay more tips to the servers who have the same gender as themselves. However, this result doesn’t match my hypothesis. The main reason can be that most of the servers in my survey are female and the P-value of sgwp is 0.342 which is not very significant as well.
The coefficient of bill size is 0.13310 in this regression and the P-value is 0.000 which is very significant. It means that bill size will significantly influence the tipping size and they have a positive relation. The coefficient of frequency is positive too. It demonstrates that the frequently visited customers will pay more tips than in frequent customers. The coefficient of method of payment is 0.1407 which has positive relation between method of payment and tip size. It means that people who pay with credit card pay more tips than people who pay in cash.http://www.ukassignment.org/uklunwen/
The coefficient of food quality is 0.8619 and its P-value is as high as 0.078. It presents that the good food quality will bring a higher tips. While my result of service quality is negative 0.1243 in coefficient. It means that the good service quality won’t help the servers to gain more tips. The result is different from literature review which showed a positive correlation between service quality and tip size. But the P-value and T-test of service quality are not significant. And the service quality is rated by the servers own. It may have the gap between servers themselves feelings and the customers’ feelings.
The customers’ dressing also has positive influence on tipping size. And server’s self introduce can also help increase the tip size. But both of them are not significant.
The coefficient of the race is positive which presents that the while customers do tip more than other races. But the P-value is not significant enough. The coefficient of time is 0.6170 which shows that at dinner time, the server can gain more tips than at lunch time. The coefficient of group size is -0.1051, it means that the bigger the group size the smaller the tips will be. It matches parrett’s result that people will free ride on others when in a big group.
According to this regression, I pull out the 4 most significant and focused variables to do a second regression to see how these four important variables influence the tip size.
Regression Analysis: T versus SG, BS, Freq, FQ
The regression equation is
T = - 0.878 - 1.08 SG + 0.145 BS + 0.443 Freq + 0.784 FQ
Table-2
Predictor Coef SECoef T P
Constant -0.8779 0.8574 -1.02 0.311
SG -1.0827 0.4697 -2.31 0.026
BS 0.14450 0.01510 9.57 0.000
Freq 0.4426 0.2586 1.71 0.093
FQ 0.7842 0.3485 2.25 0.029
S = 1.49249 R-Sq = 71.8% R-Sq(adj) = 69.5%
Analysis of Variance
Source DF SS MS F P
Regression 4 272.804 68.201 30.62 0.000
Residual Error 48 106.922 2.228
Total 52 379.725
From the second regression we can see that all the four variables are statistically significant. The P-values are all less than 0.1. It proved that servers’ gender, bill size, frequency and food quality are all strongly influence the tip size.
The P-value of server’s gender is 0.026 and the absolute value of T-test is 2.31. The coefficient of server’s gender was negative 1.0827. So it means that the server’s gender is negatively related to tipping size. In other word, customers will give more tips to male servers than female servers. And male server can receive about $1 more than female server when other variables no changes.
The P-value of bill size is still 0.000, and T-test is 9.57. The coefficient of bill size in this regression is 0.14450. This means that when bill size increases by $1, the tips will increase by 14.45 cents if keep other variables unchanged. It proves that people do tip as the tipping norm at about 15%.
The P-value of frequency is 0.093 and the T-test is 1.71. The coefficient of frequency is 0.4426. The return customers do know the server well and they want to get good service next time, so they may probably give more tips than others do. At the same time, the server may also pay more attention to those customers and offer better quality. So the frequent customers usually pay more tips at a high rate.
The P-value of food quality is 0.029 and the T-test is 2.25. The coefficient of food quality is 0.7842. It proves that the customers do add their feelings about the food to the tip size. When they like the food, they are more willing to pay the tips. On the contrary, they will reduce the tips if they feel the food quality was not so satisfied. While this factor can not controlled by the servers. #p#分页标题#e#
Conclusion
Tipping behavior is an interesting topic and special economic phenomenon. To find the reasons and the determinants that influence the size of tips is a very good experience and worth to study with. After a series of study with this topic, I get my own result. To answer do people give tips based on a social norm or as a reward for good service, can be concluded as follows.
First of all, tip size is not related to service quality. Although people tip to expect good service quality, as the research found that the service quality will not determine the tipping size.
Secondly, male servers, bill size, and frequency of visit are positively related to tip size. Thus, social norms may dictate tipping behavior. As the result shows, people generally tip around 15% of the bill size which matches the tipping norm in the United States. It means people pay more attention on social norm than any other experience. Additionally, the male servers and the frequency of visit can also help to gain additional tips.
Thirdly, because food quality is an important component of tipping behavior, the total customer experience rather than service quality is important. The result demonstrates that food quality is more important than service quality. When people decide to pay tips, they will not only judge servers’ service quality, but also consider the food quality. It also proves that people will think over the total experience not only consider one or two aspects.
Appendix A: Copy of Survey
How many people were at the table?
How many people were men and how many were women?
Men____ Women___
How much was the total bill?
Were they the frequency customers?
(1-2 time____ several times____ frequently____)
Was it at lunch or dinner? (lunch__ dinner___)
Was everything perfect while you were serving?
What race your customers are?
(White____ Asian___ black or African American___ Hispanic or Latino ethnicity)
8. How they paid? (Credit card or cash)
9. Did you ask if the food was OK? If so how did they respond
(Excellent___ Good____ OK_____ Not Good____)
10. Did you introduce yourself to the table?
11. Please rate how expensive the clothing was of the people at the table
(Very expensive____ Somewhat expensive____ Average___ Less Expensive____)
12. On a scale of 1(poor) to 10(excellent) evaluate the service you gave to the table.
References
Parrett, Matt. (2006, October 1). An analysis of the determinants of tipping behavior: a laboratory experiment and evidence from restaurant tipping The Free Library. (2006). Retrieved January 25, 2012 from http://www.thefreelibrary.com/An analysis of the determinants of tipping behavior: a laboratory...-a0154513788
Ben-Zion, Uri and Edi Karni (1977), "Tip Payments and the Quality of Service," in O.C. Ashenfelter & W.E. Oates,(Eds.), Essays in Labor Market Analysis, New York: John Wiley & Sons, pp. 37- 44.
Azar, O.H. 2004a. “The History of Tipping – from Sixteenth-Century England to United States in the 1910s.” Journal of Socio-Economics 33: 745-764.
Azar, O.H. 2004b. "What Sustains Social Norms and How They Evolve? The Case of Tipping." Journal of Economic Behavior and Organization 54: 49-64.
Azar, O. H. (2007), The Social Norm of Tipping: A Review. Journal of Applied Social Psychology, 37: 380–402. doi: 10.1111/j.0021-9029.2007.00165.x
Lynn, Michael (2006), “Tipping in restaurants and Around the Globe: An Interdisciplinary Review.” Ch. 31, pp. 626-643. In Morris Altman (Ed.) Handbook of Contemporary Behavioral Economics: Foundations and Developments, M.E. Sharpe Publishers.
Feinberg, Richard A. (1986), “Credit Cards as Spending Facilitating Stimuli: A Conditioning Interpretation,” Journal of Consumer Research, 13 (December), 348-356.
Ruffle, Bradley J. (1999), "Gift Giving with Emotions," Journal of Economic Behavior and Organization, 39, 399-420.
Lynn, Michael and Michael McCall. 2000. "Gratitude and Gratuity: A Meta-Analysis of Research on the Service-Tipping Relationship." Journal of Socio-Economics 29: 203-214.
Lynn, Michael (1996), “Seven Ways to Increase Server’s Tips,” Cornell H.R.A. Quarterly, (June), 24-29.
Lynn, Michael (2003), “Restaurant Tips and Service Quality: A Weak Relationship or Just Weak Measurement?,” International Journal of Hospitality Management, 22, 321-325.
Conlin, Michael, Michael Lynn and Ted O'Donahue. 2003. "The Norm of Restaurant Tipping." Journal of Economic Behavior and Organization 52: 297-321.
Lynn, Michael and Andrea Grassman. 1990. “Restaurant Tipping: An Examination of Three 'Rational' Explanations,” Journal of Economics Psychology 11:2, pp. 169-81.
Bodvarsson, Orn B. and William A. Gibson. 1997. “Economics and Restaurant Gratuities: Determining Tip Rates,” American Journal of Economics and Sociology 56:2, pp. 187-204.
|
 |
|||
| 网站地图 |
