Duke,UNC ,Michigan, or Purdue

system · July 11, 2007, 3:12pm

In terms of Per Capita measures of things such as elite undergrad scholarship winners (like the Rhodes or Fulbright) and top high school awards such as National Merit Scholarships, you’ll see that Duke proportionally has more top students.

Also, Michigan does very well with these indicators as well.

GreenShirt · July 11, 2007, 3:59pm

“However, the communities surrounding those universities aren’t all equally as welcoming. I would say, from my limited exposure, that Ann Arbor and Chapel Hill probably embrace diversity more readily than Durham”

I’m not sure how you’d define the embracing of diversity, but, FYI, Chapel Hill is a “white” suburb while Durham is around 45% African-American.

jack · July 11, 2007, 4:01pm

UNC has the 2nd largest number of Rhodes Scholars among public universities (41 total, I believe), behind UVA. They also a large number of Truman, Goldwater, Mitchell, Churchill, and Mellon Fellows.

Have no information on National Merit, though it doesn’t take a heck of a lot to be a National Merit Finalist. To win a National Merit Scholarship has more to do with some other stuff, other than how great the student is.

GreenShirt: The town of Chapel Hill was built around the University (University was there first), so the town of Chapel Hill and UNC are integrated, and they get along relatively well. Durham and Duke-- well, unless you’ve been living in a cave for the last year or so-- that town/gown relationship isn’t the best. And while Durham has some great things going for it, it also has an extremely high crime rate. Duke is also rather isolated from the town of Durham. UNC, on the other hand, is right on Chapel Hill’s main street–everything within walking distance.

k_amp_s · July 11, 2007, 4:44pm

alex -

Otoh, there is quite a difference in the % of students who scored above 30 on the ACT.

61% - Duke
56% - Cornell
55% - Columbia
55% - Brown
38% - UoM
30% - UNC

hawk -

I don’t know if that would hold true since Duke basically already gets a push as a “Northeastern school”.

hawkette · July 11, 2007, 5:04pm

k&s,
Many people might think that Duke gets a push as a “northeastern” school, but I think that the facts show that to be an incorrect belief.

Recently, collegehelp provided some data on the top 15 states represented at each school. Here is the data for Duke’s incoming class in Fall, 2005:

<ol>
<li> NC-245 students (15%)</li>
<li> FL-161 students (10%)</li>
<li> NY-135 students (8%)</li>
<li> TX-116 students (7%)</li>
<li> CA-96 students (6%)</li>
<li> VA-86 students (5%)</li>
<li> MD-76 students (4.52%)</li>
<li> NJ-74 students (4.44%)</li>
<li> IL-67 students (4%)</li>
<li>PA-54 students (3%)</li>
<li>GA-51 students (3%)</li>
<li>CT-50 students (3%)</li>
<li>OH-46 students (3%)</li>
<li>MA-46 students (3%)</li>
<li>SC-40 students (2%)</li>
</ol>

Also, I think that Duke is seen more as a Southern rather than a Northern college and I believe that the biases towards the South factor into its comparatively low 4.5 PA score. As mentioned previously, ex-PA, Duke would be the 3rd ranked score according to the USNWR methodology.

system · July 11, 2007, 5:07pm

Duke has a NE liberal preppy feeling, but its geographic distribution is a bit different than NE schools.

UNC is less diverse (not a bad thing, a function of it being in-state), and has more students from the South. So I’d assume its correspondingly more conservative, though maybe data from facebook might prove otherwise. Plus John Edwards campaign is head-quartered there.

hawkette · July 11, 2007, 5:12pm

Be careful. John Edwards attended NC State and the Wolfpack would never want one of their own to be associated with Carolina (even if he did get a law degree from Chapel Hill).

loslobos71 · July 11, 2007, 5:23pm

IMO, Duke and UMich are totally comparable.

k_amp_s · July 11, 2007, 5:33pm

hawk -

</li>
</ol>

That’s still about a quarter of the student body coming from the NE.

While geographically located in the South - Duke isn’t seen as the typical Southern school (a la Vandy) - hence, the reputation of having a “NE liberal vibe.”

Besides, there is also the anti-West bias (which is what the Pac-10 teams constantly wail about).

Duke doesn’t have a “low” PA score - it’s grad programs simply aren’t up to snuff with Stanford’s (simply relocating Duke to California wouldn’t make a difference in this regard) - plus, you have to take into account merit aid for the overly high rankings for schools like Duke and WUSTL when foregoing PA scores.

hawkette · July 11, 2007, 5:43pm

loslobos71,
I would respectfully disagree. The objective differences are not small.

O B J E C T I V E D A T A

UNDERGRADUATE ENROLLMENT 
Duke 6330 U Michigan 25,555 U N Carolina 17,124

% AND # OF STUDENTS WHO ARE IN-STATE 
Duke 15% (950) U Michigan 66% (16,866) U N Carolina 85% (14,555)

% OF FEMALE STUDENTS 
Duke 49% U Michigan 52% U N Carolina 60%

% OF WHITE/NON-HISPANIC STUDENTS 
Duke 52% U Michigan 66% U N Carolina 70%

% FROM PUBLIC HIGH SCHOOLS 
Duke 65% U Michigan 80% U N Carolina 84%

% OF STUDENTS IN GREEK LIFE (Male & Female) 
Duke 29%/42% U Michigan 16%/15% U N Carolina 11%/12%

TOP MAJORS AT EACH SCHOOL (acc to collegeboard.com) 
Duke 34% Social Sciences, 15% Engineering, 10% Security & Protective Services, 9% Psychology, 7% Biology, 5% English
U Michigan 17% Engineering, 16% Social Sciences, 8% Psychology, 6% Arts, 6% Biology, 6% Business/Marketing, 6% English,
U N Carolina 17% Communications/Journalism, 14% Social Sciences, 10% Biology, 9% Business & Marketing, 9% Psychology, 7% Health Professions, 5% History, 5% English

IS & OOS COST (Tuition & Fees) 
Duke $33,963 U Michigan $9,723 U N Carolina $5,034
Duke $33,963 U Michigan $29,131 U N Carolina $19,682

TOTAL COLLEGE ENDOWMENT AND PER CAPITA (undergrad and grad)
Duke $4.5bn ($385,079) U Michigan $5.65bn ($141,331) U N Carolina $1.15 bn ($45,284)

AVERAGE HIGH/LOW IN FEBRUARY 
Duke 54-30 U Michigan 34/19 U N Carolina 54-30

GRADUATION RATES 
-% OF STUDENTS EXPECTED TO GRADUATE IN 6 YEARS: 
Duke 93% U Michigan 77% U N Carolina 79%
-% OF STUDENTS WHO DO GRADUATE IN 6 YEARS: 
Duke 93% U Michigan 86% U N Carolina 84%
-% OF STUDENTS WHO GRADUATE IN 4 YEARS: 
Duke 87% U Michigan 70% U N Carolina 65%

FACULTY RESOURCES 
-% OF CLASSES WITH <20 STUDENTS 
Duke 71% U Michigan 43% U N Carolina 50%
-% OF CLASSES WITH 50+ STUDENTS 
Duke 5% U Michigan 16% U N Carolina 11%
-FACULTY/STUDENT RATIO 
Duke 8/1 U Michigan 15/1 U N Carolina 14/1

STUDENT SELECTIVITY 
-% ACCEPTANCE RATE 
Duke 21% U Michigan 47% U N Carolina 34%
-SAT/ACT RANGE (Middle 50%) 
Duke 1380-1550 U Michigan 1210-1420 U N Carolina 1200-1390
-% OF STUDENTS RANKING IN TOP 10% IN HS CLASS 
Duke 87% U Michigan 90% U N Carolina 76%
% OF STUDENTS WITH HS GPA > 3.75 (Unweighted) 
Duke na U Michigan na U N Carolina 93%

<h1>OF NMS FINALISTS IN 2005 (% of student body)</h1>

<pre><code>Duke 142 (8.7%) U Michigan 75 (1.18%) U N Carolina 149 (3.56%)
</code></pre>

<h1>OF 1500 STUDENTS ENROLLED (% of student body)</h1>

<pre><code>Duke 2282 (35%) U Michigan 1645 (6%) U N Carolina 892 (5%)
</code></pre>

ALUMNI GIVING % 
Duke 44% U Michigan 15% U N Carolina 23%

S U B J E C T I V E D A T A

PEER ASSESSMENT 
Duke 4.5 U Michigan 4.5 U N Carolina 4.2

system · July 11, 2007, 6:02pm

k&s - merit scholars make up roughly 2% of Duke’s student body, so what would change if you took that into account?

k_amp_s · July 11, 2007, 6:40pm

Not really, since Duke actually has about 70 merit scholars for each class, making up about 4.5% of its total student body.

Anyway, this edge is enough to make up the small difference in median SAT scores among Duke’s peer schools (these schools wouldn’t be spending the $$ on merit aid if they weren’t seeing some return on their “investment”).

hawkette · July 11, 2007, 7:38pm

k&s,
I really struggle with these objections to merit aid. Why should a school, any school, be marked down for using merit aid to attract students? I just don’t get it. If an institution decides that such a move advances interests of the school, then why not? And if the result is a more qualified class, then that works to the benefit of the class while taking that excellent student away from somebody else. Heck, I applaud those schools using merit aid as it is a key tool to respond to those schools with great historical prestige and presently in control of the status quo. I would expect the Establishment schools to complain and denigrate those schools who are “stealing” great students with merit aid, but that does not make those complaints valid.

By the way, this happens everyday in the business world as you try to hire new employees. Why should the college world be any different?

system · July 11, 2007, 8:22pm

k&s - i’m not sure how you get 70. I’m pretty sure the actual number is about 30 or 40. Each year the ranges are 10-15 Robertsons, 5-10 ABs, 5-10 Uni’s, 10-15 BN/Trinitys, and maybe 5 other ones that I forgot. So between 30-40 - ie 2% or 3% at most. Thats assuming every scholarship offered is accepted by the student, though some turn them down.

Either way, Duke’s SAT 75% range is 1560 or 1550. I don’t think that would shift if you took away 2-3% of top scorers (and thats assuming they are top scorers on SATs in the first place, takes more than a 1600 to get one of these scholarships).

jenny790 · July 11, 2007, 10:00pm

thank you so much you guys, i didn’t expect to get such a huge repsonse in just 1 day!~ You all have helped me in my college search tremendously. Anyone else in the comparison?

mahras2 · July 11, 2007, 11:23pm

Say Duke’s average score is approximately 1465 (midway between 25% and 75%). They have 6330 students. Say 3% of the school is 190. Say every single one of these kids scored a 1600 (I was a finalist…I don’t have that). Through simple math, the average SAT score of the remaining 6140 kids is 1461. 4 pt change in mean. So, the approximate 25%-75% range after removing the top kids still comes to 1376-1546. I still don’t know the reason for removing the top kids but there you go.

To the OP> Can’t go wrong with Duke or Mich. I suggest visiting them as they are pretty different in terms of size which may or may not be important to you. However, generally Duke’s student body will be stronger on average than Michigan’s. Education quality at both are great so you will be in good shape.

jack · July 11, 2007, 11:36pm

Well, I think K&S’s post 24 might be more illuminating if, rather than simply quoting percentages, the actual number of students was put up there instead, or in addition. Percentages can be a bit misleading when you don’t also put up the number of students in the undergraduate population. I mean, 30% of ACT scorers above 30 at UNC (as well as 38% at Michigan) is, in fact, many more students than 61% at Duke, etc. I always think comparing SAT/ACT scores to determine overall quality of education/caliber of student is a bit questionable anyway, but -again- just my opinion. I also think that the percent of white/non-Hispanic for Duke is somewhat misleading as well. While the percentage is, I’m sure, correct, a large percentage of students at Duke is/are Asian American.

Also, and I’ve corrected this so many times on CC that it’s getting silly now-- the percentage of in-state students at UNC is 82% (actually 81.5% in the class of 2006), NOT 85%, as listed on hawkette’s post #30.

Jenny: Again, all the best with your decisions. I encourage you to visit these schools before applying. I also discourage anyone from applying anywhere ED; that only benefits the school-- never the student (in my opinion).

You also “can’t go wrong” with UNC either; as stated before, it will be a more difficult admit than either Duke or Michigan, but you should definitely visit them both.

hoedown · July 12, 2007, 8:23am

I don’t see how, on the one hand, you can trumpet how important O B J E C T I V E data are, but then say the most important things in determining quality are the following:

How do you objectively measure “nature of instruction?” Is there now a measure for “willingness” to use resources for undergraduates?

It’s not that I think those are necessarily bad indicators, but I think you have to acknowledge that measuring those things is either not possible, or hard to do in an “objective” way.

hawkette · July 12, 2007, 8:47am

hoedown,
In response to your question about the “nature of instruction” I would point to factors such as the use of TAs or sub-section classes. If schools want to promote their top professors as indicative of faculty quality, then proximity and access to them would certainly be a legitimate question for students to ponder. If they are doing a lecture class for 100+ students and then a TA is doing all of the interactive teaching in small breakout groups, then this should be understood.

Nature of the teaching is easier to assess than is “quality of instruction” as this is not what PA seems to measure yet this measure is frequently referenced as indicative of undergraduate academic quality. For judging quality instruction, I would think that student surveys would have enormous value as students commonly talk a lot among themselves about which professors are terrific and which teach classes that aren’t even worth attending or where the professor speaks English poorly and communication/understanding is difficult.

As for institutional resources, I am getting at the endowment and other assets and whether the university dedicates its money more to graduate activity, undergraduate activity, or neither. A place like Rice that dedicates part of its endowment monies to keep the tuition down (cost is $27k vs $33-36k at its most common private competitors) is a good example.

On a separate topic, can you clarify how U Michigan reports its SAT and ACT data for the Common Data Set. While U Michigan evaluates applicants on the basis of a single sitting, I am wondering if they report their numbers in the CDS based on a single setting. Would you please officially comment on how this information is being reported by the university?

Alexandre · July 12, 2007, 8:47am

For someone who continuously claims that the Peer Assessment score is subjective and therefore unreliable, Hawkette seems to value subjective criteria a great deal. All of her four criteria are an exercise in subjectivity.

1) Quality of student body cannot be measured statistically, and if it could, you would assume that there was only one standard and one reporting method. As it stands, there are several standards and mutiple reporting methods. SATs are quite possibly the most easily manipulated. For example, Brown, Cornell and Columbia all have SAT averages roughly 100 points higher than Michigan and yet their ACT means are, on average, just 1 point higher. In terms of GPA and class rank, all of those schools have very similar statistics. So, are we to believe that it is class rank, GPA and ACT that are the true measure of student quality (in which case, there is virtually no difference between the student bodies at Brown, Columbia, Cornell and Michigan) or are we to assume that none of those measure ability and that it is in fact the SAT alone, regardless of reporting methodology and test-taking approach, that is the only measure of student quality? I find it very suspicious that the only truly significant gap between the private and public elites is the one element that can be manipulated by the private elites (public may not superscore) and is dependent entirely on separate individual preparation. The ACT, GPA and class rank are all generally related. The SAT requires special preparation and attention.

2) Size and nature of the classroom. This is also impossible to truly measure. Some schools have a huge portion of meaningless fluff classes with fewer than 10 students and restrict classes to 19 students just to look good in the USNWR. Other schools don’t play games and fill classes in a way that benefits the students and the university. 70% of classes at Michigan and UNC have fewer than 30 students, as opposed to 80% of classes at Duke. Duke has many classes with 15-20 students where at Michigan and UNC, those classes probably have 20-25 students. Either way, classes are manageable at any of those universities, but the state schools are not really given the luxury of manipulating class size to look good for the USNWR.

3) Quality and nature of instruction. Of the three, this is the most subjective by far.

4) Institutional resources and willingness to support undergrads. The first can be measured I suppose (endowment, endowment per student, state funding, federal funding, corporate funding, value of alumni donations etc…), but the second part cannot be measured.

At least the Peer Assessment score measures the reputation of undergraduate institutions as it is seen by academe. Whether we agree with their collective opinion or not is irrelevent.

Duke,UNC ,Michigan, or Purdue

CONNECT WITH US