Regression Case
Rock1027
Alumnigiving.pdf
197 Cose Problem Alumni Giving
case Problem Alumni Giving
BostonCollege. Brandeis UniversitY
Brown Universiry
Catifomia Institute of Technology
Camegie Mellon UniversitY
Case Western Reserve Univ.
College ol William and MarY
Columbia UniversitY
Cornell Universiry Dartmouth College
Duke UniversitY
Emory University Georgetown UniversitY
Harvard UniversitY
Johns Hopkins Uni versitY
Lehigh University
Massachusetts lnstitute of Technology
New York Universiry
Northwestem UniversitY
Pennsylvania State Univ.
Princeton UniversitY
Rice University Stanford UniversitY
Tufts University Tulane UniversitY Universiry of Califomia-
Berkeiey
University of Calilornia-Davi s
University of California-Irvine Universiry of Califomi a-
Los Angeles
U niversity of Californi a- San Diego
25
-1J
40 46
28
31
27
31
35
53
45 3i7
29 46 u 40 44
13
30 2t 6't 40
34 ,29
17
18
7 9
13
t3 '8 8
.3r.
10
8
12 '1
13
10
8
7
10
8
9 11
6
13
I 19
5
I 7
9 12
t7
19
20
l8
39
68
60 65
6l: s2, 45
69 '/2 6i 68 6) 54 73
64 t5 65,
63 .66'
32 68
62
69
6:1
56
58
32
42 4L
48
85
7g' .
93 B5
is 72 89
90
9t 94
92 84
91. 9',7
89
81
92
72 90 80
95
92
92 87 72 83
74
74 78
State
MA MA RI CA
NY L PA NJ TX CA MA LA CA
CA CA CA
Alumni donations are an important source of revenue for colleges and universities. If admin-
isffators could determine the factors that could lead to increases in the percentage of alumni
who make a donation, they might be able to implement policies that could lead to increased
revenues. Research shows that students who are more satisfied with their contact with teachers
are more Likely to graduate. As a result, one might suspect that smaller class sizes and lower stu-
dent/faculty ratios-might lead to a higher percentage of satisfied graduates, which in tum might
lead to incieases in the percentage of alumni who make a donation. The following table shows
data for 4g national universities. The Graduation Rate column is the percentage of students who
initially enrolled at the university and graduated. The % of Classes Under 20 column shows the
p"r""rrug", of classes offered with fewer than 20 students. The Student/Faculty Ratio column
L the number of students enrolled divided by the total number of faculty' Finally, the Alumni
Giving Rate column is the percentage of alumni who made a donation to the universify'
, , i%:of, ''Studenfl Afupqi Grailuation, Classes' ''l Faculty Gjving
. Ratel ' Under20,'' , .Ratio I Rate
PA OH VA NY NY NH NC GA DC. MA MB PA MA
19.8
(continued)
Chopter 4 Lineor Regression
Univsr-r.i1y ol Crli lornia_ Santa Barbara
University of Chicago University of Floricla University of Illinois_
Urbana Charnpaign University of Michigan_
Ann Arbor Universiry ol Nonh Carolina_
Chapel Hill University of Notre Dame University of pennsylvania University o1, Rochester University of Southem
Califomia University of Texas_Austin University of Virginia University of Washington University of Wisconsin_
Madison Vanderbilt University Wake Forest Univers:ity Washington University-St. Louis Yale University
_ Graduation ,,!f?r:[: tJ*.#y
t,Xffi,State Rate U"a". ZO 'Riit,r, RatecA704520|2
FIGURE 4.38
L84 FL 67 L77
MI 83
NC 82
IN ol PA 90 NY 76 cA 70
TX VA WA WI
TN NC MO CT
66
92
10
13
82
82
86
94
65
31
29
51
40
53
65
63
53
39
41
37
37
68
59
t3 77
4
ti
i5
l-5
t6
13
7
l0 13
21
13
12
13
9
ll l 7
Jf)
19
23
l3
)A
49
1t 23
22
13
28
12
13
31
38
33
50
Monogeriol Report 1. Use methods of descriptive statistics to summarize the data.2. Develop fln eslinrared simple lir"u;,;;;;:ion rnor-lel rhar r.an be used ro pretlicr rhealumni givinr rate. giren rhe graduar ion rare. Dircur, lour finding:.3' Derctop an.rri*rria m.urripr-e rir;;,:;;e;rrion moder rhar courjbe ured ro precricrthe alumni giving rare using ihe c.rarutri, Rare, 7o of crasr., u;;; 20,,.ndStudent/Faculty Ratio as independent variables. Discuss your findings.4' Based on the results in parh 2 and 3, do y", .,.ri"* urotrr".-.f."ssion model may bemore appropriate ? Estimate this moder, and <.liscuss yor. ..rurt?--
--'
t Hi;J,il[:]:]':TH,11,:':o,,,.nou,io,JJ",.ro, derive r,-om your analysis? what expected,gi,.n*"r"Eiffi:?'ff [:,il:ir:r,:i:Tr1:ffi
a jJf ,,,lH,H::lrrRatio? whar universfties are achieuing u'r.,ur,";;;,;r;oi,.".;;#lgiving
rare rhanwould be expected, given their Graduafion nr,", z, of Classes untter 20, and Student/Faculty Ratio ? what other independent "".iutr". coul<i be inclLrded in the moder?
Appendix Using XLMiner for Regression
Ir lz l,
.t
I 6 7
8 9
IO
1t
To show how XLMiner c3 !e Lrsed tbr regress-ion anarysis, we again develop a murtipreregression moder fbr the B.utler Trucking 6on,pury example. The dependent vzrriabre isTravel Time (v), ana the independent uu.iiuio'ure Miles (x,) and Deliveries (-r,) are shown
lx I'ffi i :i #:i ifj.y*: iff ; t*jF; i : o r,, ui o". f", D " i ;;;;, ;;J i; til#il
