+ COLLEGE-STUDENT SURVEY – PHASE II Stephanie Levonne Fei Yan Yukihiro Tsuya Zhiyong Yang My (Roxy) Dinh
Nov 17, 2014
+
COLLEGE-STUDENT SURVEY – PHASE IIStephanie Levonne
Fei YanYukihiro TsuyaZhiyong Yang
My (Roxy) Dinh
+
AGENDA1. Important Attributes & Survey2. Dummy variables & Regression model 3. Part-worths of the attributes4. Relative importance of an attribute5. Expected utility of each college choice6. Expected market shares7. Conclusion + Q&A
+IMPORTANT ATTRIBUTES:PRICE
10K25K50K
RANKTop 25Others
SIZELargeMediumSmall
LOCATIONCitySuburban
+SURVEY
+REGRESSION MODELUtility = 8.107 – 1.172*Tuition 25K – 2.508*Tuition 50K – 2.065*Rank(others) - .463*Size(medium) - .379*Size(small) -.649*Location(Suburban)
+
Part-Worth Utilities of the Attributes Top 25 Others
-2.5
-2
-1.5
-1
-0.5
0
Part-worths for Rank
Larg
e
Smal
l
-0.5
-0.4
-0.3
-0.2
-0.1
0
Part-worths for Size
Series1
+Part-Worths for Rank
Top 25 Others
-2.5
-2
-1.5
-1
-0.5
0
Part-worths for Rank
+ Part-Worths for Price
10,000 25,000 50,000
-3
-2.5
-2
-1.5
-1
-0.5
0
Part-worths for Price
+Part-Worths for Size
Large Medium Small
-0.5
-0.45
-0.4
-0.35
-0.3
-0.25
-0.2
-0.15
-0.1
-0.05
0
Part-worths for Size
Series1
+ Part-Worths for Location
City Suburb
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0
Part-worths for Location
+
Relative Importance of Each Attribute
RI for Tuition RI for Rank RI for Size RI for Location
44.11% 36.33% 8.14% 11.42%
Lowest
Highest
+ Expected utility of each college choice:
Ten different combinations
+Expected utility of each college choice:
LowestHighest
+Market shares of 10 combinations
Formula:
+Market Shares of 10 Combinations
MS(U1) = Exp(U1)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 23.453/4892.438 = 0.479%
MS(U2) = Exp(U2)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 141.175/4892.438 = 2.885%
MS(U3) = Exp(U3)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 365.602/4892.438 = 7.473%
MS(U4) = Exp(U4)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 646.776/4892.438 = 13.220%
MS(U5) = Exp(U5)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 130.321/4892.438 = 2.664%
MS(U6) = Exp(U6)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 319.862/4892.438 = 6.538%
MS(U7) = Exp(U7)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 68.102/4892.438 = 1.392%
MS(U8) = Exp(U8)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 17.904/4892.438 = 0.366%
MS(U9) = Exp(U9)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 1091.164/4892.438 = 22.303%
MS(U10) = Exp(U10)/[Exp(U1)+Exp(U2)+Exp(U3)+Exp(U4)+Exp(U5)+Exp(U6)+Exp(U7)+Exp(U8)+Exp(U9)+Exp(U10)] = 2088.079/4892.438 = 42.680%
Lowest: Combination 8 (50k, others, large, suburban)
Highest: Combination 10 (10k, Top 25, medium, city)
+Conclusion (from our sample)
Students wants to attend a large-sized college located in the urban area which is ranked in the Top 25 and only costs 10k for tuition => it is difficult to reach this ideal combination, so universities should pay attention to more important attributes.
Tuition is the most important attribute (44.11%), while size is the least important factor that affects a college student’s decision (8.14%)
Hence, universities should give financial incentives e.g. scholarship, financial aids, on-campus jobs to attract the target audience: potential college students and parents
Ranking is the second most important factor, so the admission office should focus a lot on PR and students’ feedbacks in order to improve ranking
+THANKS FOR YOUR ATTENTION
Questions?