Let's have some fun: Critique my list of what I believe to be the best undergraduate colleges

Public schools are private schools have different mission too. UCSD reaches out for the local area.

US News, Forbes, Kiplinger, and Washington Monthly all use graduation rates (among other measurements) to assess colleges. Look at the 4y graduation rates shown on Kiplinger’s “best value” pages:

http://www.kiplinger.com/tool/college/T014-S001-kiplinger-s-best-values-in-private-colleges/index.php?table=lib_arts&state_code=ALL&id=none&sortby=foury_gradrate_p&sortorder=DESC

http://www.kiplinger.com/tool/college/T014-S001-kiplinger-s-best-values-in-private-colleges/index.php?table=prv_univ&state_code=ALL&id=none&sortby=foury_gradrate_p&sortorder=DESC

http://www.kiplinger.com/tool/college/T014-S001-kiplinger-s-best-values-in-public-colleges/index.php?table=public&state_code=ALL&id=none&sortby=foury_gradrate_p&sortorder=DESC#colleges

75% seems to be at the low end of “elite” universities and LACs.

Other numbers that rankings use to differentiate top schools include average class sizes, average test scores, and admission rates. One also could look at average coverage of demonstrated financial need, endowments per student,the number of faculty awards/distinctions, faculty publication/citation volumes, per capita PhD production, or other post-graduate outcomes (e.g. alumni salary or leadership positions).

If one is going to generate a credible list of “best” colleges, I think one should be prepared to state which of these (or other) metrics distinguish them from other colleges. Although, one might argue that none of these numbers really matter too much (it’s the student’s individual talents and efforts that make the biggest difference.)

“Four year graduation rates mean very little. Elite universities like Michigan, Rice and Stanford have 4 year graduation rates that hover around 75%.”

Once again, schools that have large numbers/high percentages of students enrolled in engineering will typically have lower four year graduation rates. Most LACs and many elite schools have small or non-existent engineering programs.

Some elite universities pass out grade As like candies, from what I’ve remember touring Pomona, the tour guy said no lesser than a B was ever given. While at top public university, you go on reddit, there are lots of posts about people have to repeat a core class one or maybe twice. So 4 year graduation rate has to take into account if that. Sure it’s much harder to get to Pomona than UCSD.

The number of engineering majors (or reputed rigor) doesn’t seem to fully explain the graduation rate patterns (although it may account for part of the difference among otherwise similar schools.)

Swarthmore … 4.9% engineering … 89% 4y graduation rate
JHU … …18.45% engineering … 88% 4y graduation rate
Harvard …4% engineering … 86% 4y graduation rate
Harvey Mudd … 37% engineering … 86% 4y graduation rate
UVa … 13.6% engineering … 86% 4y graduation rate
Wesleyan … 0% engineering … 86% 4y graduation rate
MIT … 39% engineering … 84% 4y graduation rate
UNCCH … 0% engineering … 81% 4y graduation rate
Michigan …16% engineering … 76% 4y graduation rate
Stanford …15.9% engineering … 76% 4y graduation rate
Berkeley …13% engineering … 72% 4y graduation rate
UWashington … 9.16% engineering … 58% 4y graduation rate
New College Fl … 0% engineering … 57% 4y graduation rate
Wisconsin … 9% engineering … 56% 4y graduation rate
Texas … 11.9% engineering … 51% 4y graduation rate
Alabama … 6% engineering … 43% 4y graduation rate
Georgia Tech 60.51% engineering … 41% 4y graduation rate

Sources: Common Data Sets, Kiplinger’s

It appears that “elite” colleges are fairly consistent in having 4y rates above ~70%.
Of the 3 public schools with 4y rates >= 80%, two of them (UVa and UNC) are “public Ivies” that claim to cover 100% of demonstrated need. The 4y rates must be influenced by many factors; I suspect admission selectivity and financial aid are among the more important ones.

Of the schools on Alexandre’s list, Georgia Tech does have one of the highest concentrations of engineering majors. It also has one of the lowest figures for average coverage of financial need (only 51.7% in 2013, compared to the 100% claimed by nearly all the elite private schools on that list.)

In any case, I would think many cash-strapped parents do care about (or should care about) the likelihood of having to cover a 5th or 6th year of college.

tk21769:

During the fall of 2014 there were 6,024 undergraduate students matriculated into the College of Engineering out of a total of 28,395 undergraduates in attendance at The University of Michigan in Ann Arbor. That number expresses a rate of slightly over 21% of the total undergraduate enrollment, which is also 5 percentage points higher than the numbers that you reported. This once again proves that sources such as Kiplinger’s are not to be trusted for accuracy. Based on this one metric, I have to assume that all of your figures for enrollment are inaccurate.

http://obp.umich.edu/wp-content/uploads/almanac/Almanac_Ch2_Jan2015.pdf

Kiplinger as source is also highly inaccurate. Not just for colleges but for most thing I’ve read.
But for UCSD, more than 50% of degrees granted were from STEM major and not just engineering. High concentration of premeds. If these classes are not designed to weed out or fail some students, I don’t know what is then. Also UCSD reaches out for local students, read first generation Hispanics, check out the graduation rate for 4 year for this subgroup and it’s in the low 40%.

I used Kiplinger’s as a source for the 4y graduation rates (Michigan’s 76%, etc.)
http://www.kiplinger.com/tool/college/T014-S001-kiplinger-s-best-values-in-public-colleges/index.php
If these numbers are inaccurate, it would be helpful to suggest a more authoritative source that shows a different number (not only for Michigan but for the other schools we’re comparing).

I did not use Kiplinger’s for the Michigan engineering number (16%). Kiplinger’s does not report those figures.I took that number from Michigan’s 2014-15 Common Data Set, Section J.
http://obp.umich.edu/root/facts-figures/common-data-set/

CDS, section J numbers express the percentage of degrees conferred in various disciplines
(not the percentage of students enrolled in various divisions of the university.)
If we showed the engineering enrollments for all schools (including JHU, Harvey Mudd, MIT, etc.) would that paint a significantly different picture?

“If we showed the engineering enrollments for all schools (including JHU, Harvey Mudd, MIT, etc.) would that paint a significantly different picture?”

Probably not.