USNWR 2009: Looking at the Top Strata I (Freshmen Retention)

College Search & Selection
1

<p>The publication of the 2009 USNWR College Rankings provides an opportunity to compare schools based on a wide variety of data points. In this and in threads to follow, I urge the reader to think less about the absolute rankings and more about the nature and value-added of the data point being discussed. </p>

<p>Freshmen Retention, National Universities</p>

<p>98% , Princeton
98% , Yale
98% , MIT
98% , Stanford
98% , Caltech
98% , U Penn
98% , Columbia
98% , Dartmouth
98% , Brown
98% , Notre Dame
97% , Harvard
97% , U Chicago
97% , Northwestern
97% , Wash U
97% , Rice
97% , UC Berkeley
97% , U Virginia
97% , Georgetown
97% , UCLA
96% , Duke
96% , Cornell
96% , Johns Hopkins
96% , Vanderbilt
94% , Emory
94% , Carnegie Mellon</p>

<p>Freshmen Retention, LACs</p>

<p>100% , Pomona
98% , Bowdoin
97% , Amherst
97% , Williams
97% , Carleton
96% , Swarthmore
96% , Davidson
96% , Haverford
96% , Claremont McK
96% , Vassar
96% , US Naval Acad
95% , Wellesley
95% , Middlebury
95% , Wesleyan
95% , Harvey Mudd
94% , W&L
94% , Colgate
94% , Bryn Mawr
94% , Bates
93% , Grinnell
93% , Oberlin
93% , Colby
93% , Macalester
92% , US Military Acad
92% , Hamilton
90% , Smith</p>

2

<p>Ya Pomona!!</p>

3

<p>A few comments:</p>

<p>1) These are mostly very small differences, in most cases not large enough to be meaningful. I wouldn't be concerned about the difference between a 98% and a 96% retention rate, for example.</p>

<p>2) On the other hand, US News reports these figures to the tenth of a percent, and the rounding done by the OP may be slightly misleading. The difference between Harvard's 97.2 (which hawkette rounds to 97) and Caltech's 97.8 (which hawkette rounds to 98) is actually less than the difference between Caltech's 97.8 and Yale's 98.5, even though hawkette rounds both Caltech and Yale to 98. Yale is actually tops in the US News chart for national universities, a full 0.5 ahead of its nearest competitors Princeton, Stanford, Penn and Dartmouth, all at 98.0.</p>

<p>3) I wouldn't worry too much about freshman retention rates until you start to get down into the low 90s and upper 80s. At that point roughly one in ten freshmen is not returning---a signal that something may be wrong, whether academically, socially, or financially. Among the top 50 national universities, #23 Carnegie Mellon (93.8), #33 NYU (92.2), #35 Georgia Tech (92%), #35 Wisconsin (93.2), #40 UIUC (92.2), #41 Case Western (91.0), #41 RPI (92.2), #41 U Washington (92.5), and pretty much all the rest from there on down to #50 Yeshiva (88.0) are in this potentially troublesome range. From #50 to #80, retention rates from the mid 80s to very low nineties become the norm. Below #80, retention rates dip into the low 80s.</p>

<p>4) For all the vaunted advantages of LACs---intimate environment, personal attention from faculty and administration, small classes, undergrad focus, etc---the top LACs as a group do a little worse than the (generally much larger) top research universities in freshman retention. Pomona's 99.5% (sorry, it's not really 100%) tops all the charts, and some at the very top (Amherst 97.2, Williams 97.2, Bowdoin 98.2, Carleton 97.0) are very good. But after the top 10 or 15 LACs, retention rates fall off pretty quickly. By my count, more than half of the US News top 50 LACs have retention rates of 93.5 or lower---that is, the range at which I suggest one might begin to get concerned. I don't have a ready explanation for this.</p>

<p>5) Nor does there appear to be a correlation with school size among the research universities. Some big schools like Berkeley (96.8), UCLA (97.0), UVA (96.8), Michigan (96.0), and UNC Chapel Hill (96.2) do just fine; others like Wisconsin (93.2), UIUC (92.2) and U Texas (92.8) struggle a bit. But the same pattern holds for smaller universities vaunted for their "undergraduate focus"---Dartmouth (98.0) is near the top of the charts, while Emory (94.2) and Carnegie Mellon (93.8) are only so-so.</p>

<p>6) I'm not sure a one-year snapshot of data of this type is all that meaningful. A 3-year or 5-year composite might more clearly reveal which schools are having serious retention issues, and which are showing up just as noise in the data.</p>

4

<p>I imagine CMU suffers from the combination of it's lack of strong financial aid and expensive price tag. I also know a number of students come with the intention of switching into the school of computer science from whichever college they were admitted into, and I imagine very few of them are actually successful. I actually didn't know anybody from freshman year that transferred out or dropped out of college, though I do know one of my friends in later years had taken a leave of absence following her freshman year there.</p>

<p>I imagine the largest contributing factor to a lot of these is the amount of financial aid these schools are able to offer their students so they don't have to drop out to make ends meat.</p>

5

<p>IMO Freshmen Retention is one of the more inappropriately weighted factors of the USNWR rankings. I have a hard time believing that this data point has more value than something like Student-Faculty Ratio, which has only 1% value in the overall USNWR score vs 4% for Freshmen Retention. Still, if there are significant differences (7-10%+?) in Freshmen Retention between institutions that compete for the same students, then this can be a useful number in choosing among those schools.</p>

6

<p>I think US News thinks that since class sizes are more heavily weighted, the s:f ratio doesn't need to be weighted much. I agree, though, that freshmen retention isn't very helpful.</p>

7

<p>USNWR actually gives high weight to % of small classes (under 20) with a weighting of 6%, but they only assign a 2% weighting to large classes (over 50). </p>

<p>IMO, the Graduation/Retention data (20%) has less value than much of the Faculty Resources data (also 20%) which really speak to the nature of the undergraduate academic environment that a student will encounter at XYZ College. Furthermore, I think that the Graduation/Retention data (and subsequent ranks) suffers from the lack of inclusion of 4-year graduation data.</p>

8

<p>bclinton is absolutely right--there is very little variation among the top schools. A prospective student who elects one school over another because one has a 98% and the other has a 97% retention rate is, IMO, grasping at a spurious albeit "numeric" reason to distinguish the "better" of the two schools.</p>

<p>hawkette, the inclusion of a four-year rate would be useful for some students, but it could be misleading unless it was adjusted for program mix. Schools with programs that typically run 4.5 or 5 years will look like underperformers when it comes to graduating students on time, even if they do a fantastic job.</p>

9

<p>hoedown,
I completely agree that tiny differences in Freshmen Retention should not be critical elements in one's choice of college (or one's ranking of colleges!). It is exactly because the linear rankings do exactly what you object to that I am creating threads such as this. IMO the proper interpretation of the data presented at the beginning of the thread is that the differences in Freshmen Retention among the top strata of colleges are negligible as nearly all have an excellent record. For me, only Hamilton (92%) and Smith (90%) might prompt further inquiry although others will undoubtedly interpret the data more loosely or tightly. </p>

<p>As for 4-year graduation rates, I plan to present this data in a future thread. It is my understanding, however, that the CDS data (from which USNWR draws) makes allowances for students involved in programs lasting longer than the traditional 4-year period.</p>

10

<p>Differences of 2-3% in retention may or may not be meaningful because of differences among schools in the way retention is calculated. Retention calculations are not as carefully regulated by the US Dept of Education as graduation rate calculations.</p>

<p>The US Department of Education is going to be requiring 8- and 10-year graduation rates in the future because so many students take longer to graduate than the stated program length of 4 or 5 years. This represents an increase from the current standard of 6- and 7.5-year graduation rates. The 4-year graduation rate is not a valid comparison among schools. Students take longer to graduate in programs with a lot of lab courses, internships, and so on. The US Department of Education recognizes this. US News uses 6- and 7.5-year graduation rates. So does the NCAA. </p>

<p>Even then, graduation rates at schools with a lot of science and engineering majors tend to be lower not because the school isn't good but because the programs are so difficult. It is difficult to interpret graduation rates even though they are carefully regulated by the federal government.</p>

11

<p>Looking at 4-year graduation rates is silly. It's not uncommon for students to take a semester or year hiatus for all sorts of reasons, none of them reflecting negatively on the school. I'll bet a bunch of schools with a politically active student base---especially smaller LACs, where even small numbers can move the needly on graduation rates--will see a noticeable dip in their 4-year graduation rates because a fraction of their current students are taking a leave to work on a hotly contested presidential election this year. IMO this is a good thing, not a bad thing, and it has absolutely nothing to do with the quality of the education students receive. </p>

<p>I have my doubts even about 6-year graduation rates. In some states (like Minnesota) a lot of parents still expect their kids to work their way through school on the theory that it teaches self-reliance and the degree will be worth a lot more to the student if they pay for it themselves. Old-fashioned, I know, but is that such a bad thing? Given rising costs, it often takes more than 6 years, pushing down the 6-year graduation rate at schools like the University of Minnesota which takes a pummeling in the US News rankings for it. Does this make Minnesota a worse school than one in a place not imbued this this kind of Midwestern ethos of self-reliance? I don't think so. The whole premise of the US News rankings is that there's a standard model, one-size-fits-all ideal of higher education against which all schools can be measured, reducing all variations to a single metric of "quality." The whole idea is pretty preposterous, when you think about it.</p>

<p>(And hawkette: just askin', but are you sure you aren't discounting freshman retention becuase some of your favorite schools, like Emory (94.2%), Vanderbilt (95.5%), and Wake Forest (94.0%), come out in the bottom half of the US News top 50 on this metric?)</p>

<p>US News top 50 national universities, ranked by retention rate:</p>

<ol>
<li>Yale 98.5</li>
<li>Princeton 98.0</li>
<li>MIT 98.0</li>
<li>Stanford 98.0</li>
<li>Penn 98.0</li>
<li>Dartmouth 98.0</li>
<li>Caltech 97.8</li>
<li>Columbia 97.8</li>
<li>Notre Dame 97.8</li>
<li>Brown 97.5</li>
<li>Harvard 97.2</li>
<li>Chicago 97.2</li>
<li>Northwestern 97.0</li>
<li>WUSTL 97.0</li>
<li>Rice 97.0</li>
<li>Georgetown 97.0</li>
<li>UCLA 97.0</li>
<li>UC Berkeley 96.8</li>
<li>Virginia 96.8</li>
<li>Duke 96.5</li>
<li>Johns Hopkins 96.5</li>
<li>UNC Chapel Hill 96.2</li>
<li>Cornell 96.0</li>
<li>Michigan 96.0</li>
<li>USC 95.8</li>
<li>Vanderbilt 95.5</li>
<li>Tufts 95.5</li>
<li>Boston College 95.5</li>
<li>William & Mary 95.0</li>
<li>Brandeis 94.5</li>
<li>Emory 94.2</li>
<li>UCSD 94.2</li>
<li>U Rochester 94.2</li>
<li>U Florida 94.2</li>
<li>Wake Forest 94.0</li>
<li>Lehigh 94.0</li>
<li>Carnegie Mellon 93.8</li>
<li>UC Irvine 93.5</li>
<li>Penn State 93.5</li>
<li>Wisconsin 93.2</li>
<li>Texas 92.8</li>
<li>U Washington 92.5</li>
<li>NYU 92.2</li>
<li>UIUC 92.2</li>
<li>RPI 92.2</li>
<li>Georgia Tech 92.0</li>
<li>Case Western 91.0</li>
<li>UCSB 90.5</li>
<li>UC Davis 90.2</li>
<li>Yeshiva 88.0</li>
</ol>

12

<p>I think at some large schools, the longer time to graduate can indicate that students have trouble getting the classes they need to fulfill their major requirements.
As a UC graduate, I can tell you that there was a Statistics A class that I needed for my major, only offered in the Fall. I registered for the class as a freshman, sophomore and junior, without getting it. As a senior, and planning to graduate in four years, I still did not get into the class. Talking to the registrar and professor got me nowhere. I had to meet with the Dean in order to get into the class so I could graduate on time.
Some of my classmates in more impacted majors (biology, engineering) faced this issue multiple times and just could not get the courses they needed in sequence to graduate in four years.
I think it's a useful data point for a prospective student.</p>

13

<p>^ I think it's a useful data point, I'm just not sure it has all that much relevance in a one-size-fits-all measurement of "quality" because there are just so many factors that could influence the graduation rate. The situation you describe is an educational quality problem---an inability on the part of the school to deliver the classes students need. The situation I describe is a question of cultural difference---an Upper Midwest/Nordic/Lutheran belief that a little adversity is good for the soul and certainly for character development (see Garrison Keillor), consequently as parents we're going to cut the apron strings at 18 and let the kids figure out on their own how to pay for college. That has nothing to do with educational quality.</p>

14

<p>bc,
Earlier (in # 3) you said,</p>

<p>"I wouldn't be concerned about the difference between a 98% and a 96% retention rate, for example....I wouldn't worry too much about freshman retention rates until you start to get down into the low 90s and upper 80s."</p>

<p>and then you attempt to ridicule me about some of my supposed favorites which all score at 94% or higher? </p>

<p>Sigh.... </p>

<p>Thanks for the constructive contribution...:rolleyes:</p>

15

<p>^ oh, come now, just a little friendly ribbing, nothing to get sore about.</p>

<p>On the other hand, retention rates of 94.2 and 94.0 are getting kind of close to the 93.5 I suggested (post #3, point 4) as a trigger for concern, don't you think? I certainly wouldn't rule out a school at this level but I might want to at least ask some different questions about a school at that level than I would about a school in the 97-98% range. Especially when the 6-year graduation rates at these schools dip below 90%, also below the norm for their peer group (Emory=88.0, second lowest in US News top 25; Wake Forest=89.0).</p>

<p>FWIW, my own D presently has only one major research university at the top of her wish list (97.5% retention rate, no problem there), along with three top LACs (96.2%, 95.2%, and 93.8% respectively). The 6-year graduation rate at all four of these schools is over 90% except for the LAC with the lowest freshman retention rate, where it's 84.0%. So again, not a red light necessarily, but time to ask some additional questions.</p>

<p>I'll repeat what I said before: I'm not sure any of this data belongs in the US News rankings, but it is useful data in its own right, far more useful IMO than the aggregate rankings which lump together so many factors that some potentially troubling factors might get lost.</p>

16

<p>A lot of what USNWR uses to rank schools is misleading - freshman retention is no different. All of the schools that "do a great job" retaining (and, for that matter, graduating) students already have a leg up on everyone else because they are admitting a population of students that is already likely to be retained (and graduate) at high levels: the high-achieving kids with a vested interest in graduating from a "top" school. </p>

<p>Many of the matrixes used by USNWR can be misleading...and manipulated. Some schools, for example, have policies mandating that ALL classes be under 19 students...hmmmm, what's the magic number for class size in USNWR? Some schools have tutorials senior year, where students work 1-on-1 with faculty members - not really a "class," per se - yet they include these in their class size averages. Some schools admit a large percentage of their freshmen for the second semester or next year - often times legacies and athletes that do not come close to the average academic achievements of most admits - which are not included in average SAT score reporting. </p>

<p>STATS DO NOT TELL YOU EVERYTHING!</p>