Prom Date Survey Report _______________ AP Statistics Mr. Shubleka 04/ 10/ 09 Presented by: Annika Lin, Nancy Wu, Johnnie Nie, Mark Shen 1
Introduction Picking a good survey topic is a tough part of our project. Our goal is to make our survey question useful, interesting and creative. We began with whether Yuhao fits in his new skinny jeans and opinions someone’s new hair color. But these questions are too inappropriate, trivial, and hard to randomize at the same time. After a long time period of discussion and comparison, we decided to focus on the prom, which is coming soon in May. As everyone knows, Prom is a ritual to celebrate youth and entering the world of adulthood, so choosing an appropriate date is very important. Because of that, everyone at least has a thought about it, thus it makes easier for us to collect different opinions. More importantly, different people face it with different attitudes. In general, female tends to act reluctantly to express themselves, but males, on the other hand, are more open‐minded and they likely hold the initiative. Here is our survey question: “If you are going to prom, would you actively ask someone to be your date for the night or passively wait for others to come to you?’ The purpose of our survey is obvious; we want to see if there is a gender difference regarding how people find a prom date, either being active or passive (two choices). Our target population is the whole student body at Cushing. 2
Process of Randomization When collecting the data, randomization is a very crucial step because it determines the value of the statistical outcome. Careful randomization makes sure that the result does not turn out to be biased and the outcome reflects the true distribution of the responses. Therefore, our group made 98 questionnaires in total and distributed them to different teachers from different classes. The class categories include Math, Science and English, so the survey will not be biased because people from certain classes tend to think in the same way. Moreover, we make sure that any individual who has done the survey once don’t do it again. The reason for this is that we want to eliminate the outliers so anyone who has done it twice will weaken the outcome by ruining the randomization. Furthermore, the classes include students from each grade: freshman, sophomore, junior, senior and post‐graduate. In this case, the data includes all the age groups on campus. There are not many chances for the survey to favor certain outcomes because all of the age groups are included, so the chance a certain age group will likely choose the same answer would be as small as possible. Similarly, people are not allowed to choose both answers or else the answer will be meaningless. In this way, we can possibly eliminate the outliers in this survey. Finally, during the survey, people are not allowed to discuss the topic. Therefore, the opinions from every individual will not be influenced by any other person. Overall, we are very careful about the randomization and it is carried out well.
3
Data Analysis Male Expected Cell Count Female Expected Cell Count Total
Actively Asking 37 26.755 9 19.245 46
Passively Waiting 20 30.245 32 21.755 52
Total 57 41 98
40 35 30 25 20 15 10 5 0
Male Female
Observed Active
Expected Active
Observed Passive
Expected Passive
Ho: There is NO difference between males and females’ thoughts on actively finding a prom date or passively waiting for others to come. Ha: There IS a difference between males and females’ thought on actively finding a prom date or passively waiting for others to come. 1) Calculate the Expected Cell Count Expected Cell Count = (row total × column total) ÷ over all total ‐Female –Ask : (46 × 57) ÷ 98 = 26.755 –Wait: (52 × 57) ÷ 98 = 30.245 ‐Male –Ask : (46 × 41) ÷ 98 = 19.245 –Wait: (52 × 41) ÷ 98 = 21.755 2) Calculate the Degrees of Freedom Degrees of Freedom = (number of rolls-1) × (number of columns-1)
(2-1) × (2-1) = 1
3) Calculate the Chi‐Square Statistic 4
X 2 = ∑ [ (Observed Count-Expected Count) 2 ÷ Expected Count ]
Observed-Expected ‐Female –Ask : 37-26.755 = 10.245 –Wait: 20-30.245 = ‐10.245 ‐Male –Ask : 9-19.245 = ‐10.245 –Wait: 32-21.755 = 10.245
(Observed Count-Expected Count) 2 ‐Female –Ask : 10.245 2 = 104.960 –Wait: (‐10.245) 2 = 104.960 ‐Male –Ask : (‐10.245) 2 = 104.960 2
–Wait: 10.245 = 104.960
Calculator 1) 2nd –X −1 (MATRIX) –EDIT 1: [A] 2×2 [37 20] [9 32] 2: [B] 2×2 [26.755 30.245] [19.245 21.755] 2) STAT –TEST –C: X 2 –Test Observed: [A] Expected: [B]
÷ Expected ‐Female –104.960 ÷ 26.755 = 3.923 –104.960 ÷ 30.245 = 3.470 ‐Male –104.960 ÷ 19.245 = 5.454 –104.960 ÷ 21.755 = 4.825
∑
X 2 = 17.671 p = 2.625 E −5 df=1
Sum X 2 = 3.923 + 3.470+ 5.454 + 4.825 = 17.042
4) P‐value for the Test ( df = 1,X 2 = 17.042 ) p < 0.0005 ‐Choose the significant level as α= 0.05
p(<0.0005) < 0.05
Therefore we reject Ho and are in favor of Ha, which means that males and females do have significantly different thoughts about being active or passive when they are going to prom. Males tend to be more active in finding their prom date for the night. Females tend passively wait for others to come.
5
Conclusion Well, it seems that the result came out exactly the way we expected: most guys are aggressive predators looking for food and most girls are quiet herbivors waiting to be preyed. Awesome. God is indeed a balance creator, at least at Cushing. Imagine a world with the opposite situation. It would be, actually, it might be fun… But anyway, let me be a little bit more formal. With our survey question targeting on the prom date, all students in our school could have an easy ‘yes’ or ‘no’ answer to it. We distributed our survey (with rules applied) to many different classes consisting of students from different age groups, different thoughts, thus the inaccuracy due to influence among the short random samples decreases to the minimum. Each student has to, first, choose the answer, either “being active” or “being passive”; and second, choose the gender. By practicing the Chi‐Square Test, we have an X 2 value of 17.042. With a degree of freedom of 1, the p‐value of the test is smaller than 0.0005. Given a significant level of 0.05, we reject the null hypothesis and conclude that two genders at Cushing have different opinions regarding how they find their prom date, either being active or passive. Special thanks to: Mr. Wilson, Mr. Wenning, Mr. Sheridan, Ms. Lefcourt, Dr. Shields.
6
Additional material:
A Prom Date Survey ‐
Which is also a very important AP Statistics Project about gender difference
Attention, Prom is coming up pretty soon guys. Did you have your prom date yet? If yes, good job and please tell us how you got your date. If no, well, let’s answer the following question and good luck, seriously. If you are going to prom, would you actively ask someone to be your date for the night or passively wait for others to come to you? I will: Go out and ask someone or
Wait for someone to come to me and your gender please: Male
/ Female
Thanks for your time! ☺ Presented by: Annika Lin, Nancy Wu, Yizhuo Nie, Mark Shen
7