In: Statistics and Probability
From: Pat Turner
Sent: Thursday, July 7, 2016 8:57 a.m.
Subject: Mystery Data Shopper Stats and Store Performance?
Good morning! Welcome back from vacation J I hope you had a wonderful Fourth of July.
The last mystery shopper surveys came in and I have the final numbers. I am interested in whether there is a way to predict the final average based on the initial survey score. Also, is there a statistically significant relationship between how stores initially performed and what the overall average is?
The initial survey score and the final average data for all seven store locations is in the table below:
Store |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
Initial Survey Score |
83 |
97 |
84 |
72 |
85 |
64 |
93 |
Final Average |
78 |
98 |
92 |
75 |
88 |
70 |
93 |
Also, how good is the relationship between Initial Survey Score and the Final Average? Could I use an Initial Survey Score to predict a Final Average? In fact, could I predict a Final Average if I have an Initial Survey Score of 90?
If you could have this to me before the weekend, that would be great.
Thanks so much!
Pat Turner, Owner
Chic Sales Consignment, LLC
Ans:
Store | Intial(x) | Final(y) | xy | x^2 | y^2 |
1 | 83 | 78 | 6474 | 6889 | 6084 |
2 | 97 | 98 | 9506 | 9409 | 9604 |
3 | 84 | 92 | 7728 | 7056 | 8464 |
4 | 72 | 75 | 5400 | 5184 | 5625 |
5 | 85 | 88 | 7480 | 7225 | 7744 |
6 | 64 | 70 | 4480 | 4096 | 4900 |
7 | 93 | 93 | 8649 | 8649 | 8649 |
Total= | 578 | 594 | 49717 | 48508 | 51070 |
slope,b=(7*49717-578*594)/(7*48508-578^2)=0.8565
intercept,a=(594-0.85654*578)/7=14.131
Regression eqn:
Perdicted final score=0.8565*intial score+14.131
Correlation cofficient,r=(7*49717-578*594)/SQRT((7*48508-578^2)*(7*51070-594^2))=0.929
There is strong positive correlation between intial and final scores
cofficient of determination,r^2=0.929^2=0.863,which indicates that 86.3% of the linear variation in final score is explained by the intial score.