Okay, so it was interesting to see that the most selective colleges don’t have a monopoly on large numbers of students with high test scores, right? But there are many other factors that influence a student’s college experience, and we have data on some of them, too. Let’s look at how colleges of different levels of selectivity differ on some of the criteria used by US News and Princeton Review to rate them.

I’ve selected ten of the most selective universities–all of which admitted fewer than 10% of applicants–and labelled them Group 1. Ten schools slightly less selective–all admitted between 10 and 29% of applicants–that are often thought of as “safe schools” by the strongest students constitute Group 2. And ten colleges that almost all applicants to the most selective schools would consider sure bets–all admitted between 30 and 49%–are in Group 3.

GROUP 1

Brown

Dartmouth

Duke

Northwestern

Penn

Princeton

Rice

Stanford

Swarthmore

Yale

GROUP 2

Wesleyan

Middlebury

Emory

Carnegie Mellon

Villanova

Tufts

U Richmond

Boston U

Vassar

Colgate

GROUP 3

Fordham

Bucknell

Baylor U

Kenyon College

Lehigh

Santa Clara U

Southern Methodist U

Pepperdine

Lafayette College

Worcester Polytech

Using numbers from the 2019-2020 Common Data Sets (see footnote) and from Princeton Review’s most recent ‘Best Colleges’ book for each school, we’re going to compare these three groups on some of the measures US News and Princeton Review use to create their rankings/ratings. We’ll see that–while there are measurable differences on some factors based on the selectivity of colleges–the differences that are implied by ranked lists are not nearly as great as people often assume based on the distances between the numerical rankings of the schools on those lists. And hopefully, seeing how small the real distance is between colleges, you’ll begin to conclude that those differences certainly aren’t great enough to be worth risking your children’s (and your) physical and emotional well-being.

[This, in my opinion, is the where the real value is in college rankings, not the placement of one school in the number 1 spot, another at number 2, etc. Having all of this data in one place allows you to compare schools on criteria that may matter to you in making a decision without having to visit each separate Common Data Set or school website to get the numbers. It’s all laid out for you in one place, and will likely expand your list to schools you wouldn’t have thought of on your own.]

First up, the percentage of students who return for sophomore year, used by US News as part of their ranking formula. This is a pretty good measure of how much students like a college, and it will also impact your experience on campus. It’s preferable to be in a college where you have a larger percentage of your classmates coming back, right?

Return Sophomore Year

Group 1                                                   98%

Group 2                                                   95%

Group 3                                                   92%

It looks like there is a small difference between the three groups on this measure, but—in my opinion anyway—not enough of one to be worth considering strongly when choosing a college. All but two of these 30 colleges have over 90% coming back for sophomore year.

Smaller class sizes are also desirable, and US News uses this as one of its measures, too. They report the percentage of classes with fewer than 20 students, and it looks like again the higher ranked colleges offer a slightly better experience. Assuming the student takes a total of 40 courses to earn their degree, Group 1 would have 28 classes with 20 or fewer students, Group 2 would have 26, and Group 3 would have 23. That handful of classes with lower numbers is not likely to make much of a difference in a student’s experience.

Classes under 20

Group 1                                                             71%

Group 2                                                             65%

Group 3                                                             58%

Importantly, if you expand it to look at the percentage of classes under 30, the difference almost disappears. Thus, most of the difference in the numbers above would probably be made up for with class sizes just a few students higher.

Classes Under 30

Group 1                                                   83%

Group 2                                                   85%

Group 3                                                   80%

At the other end of the spectrum, US News reports the percentage of classes with 50 students or more. Interestingly, here the advantage goes to the less selective schools, which makes one wonder if some colleges are paying for more classes on the smaller end with more classes on the larger end.

Classes 50+

Group 1                                                   8%

Group 2                                                   5%

Group 3                                                   5%

There are, of course, many arguments about how good a measure of academic ability the SAT and ACT are, and also how well it can predict success in college. Regardless of where the truth may lie on this issue, we know that many folks, including those at US News, measure the quality of an education at any given college at least partly by the achievement of its students on these tests.

US News uses the middle 50% of test scores for colleges as part of its rating system. If we assume a normal distribution within this range, then we can compare median scores at our three groups of colleges.

Median SAT score

Group 1                                                   1504

Group 2                                                   1433

Group 3                                                   1360

As you might guess, the median SAT score shows a large number difference between our different groups, with the most selective colleges 71 points above Group 2, and Group 2 another 73 points above Group 3. These are huge differences, right? More selective schools obviously have significantly smarter students.

Well, not exactly.

Have you ever looked at College Board’s graph of the distribution of SAT scores? It’s important to understand how they’re distributed if you’re going to make conclusions about differences in scores.

SAT score                                                Percentile

       1500                                                           99

       1400                                                           97

       1300                                                           91

       1200                                                           81

       1100                                                           67

As you might guess, a score of 1500 lands you solidly in the top 1% of high schoolers. But did you know that those with a 1400 are in the top 3%? And that those with a 1300 are still firmly in the top 10%?

Knowing this, let’s look at those differences between Groups 1, 2 and 3 again.

Median SAT percentile

Group 1                                                   99

Group 2                                                   98

Group 3                                                   95

Wow, right?!! Sure, the average student at a Group 1 college is in the top 1%, but you can see that the distance between them and average students at schools in Groups 2 and 3 isn’t nearly as big as it appears when you look at the scores, with the average Group 3 student still in the top 5% of nationwide teens. How much of a difference will that small spread really make in a student’s experience?

US News also gives weight to the percentage of professors who hold the highest degree in their field. I’ve never considered this a great measure of how good a teacher someone will be, but I’ve included it here so we can compare it to Princeton Review’s survey of students’ opinions of their professors’ ability to teach.

Professors with Highest Degree

Group 1                                                   90%

Group 2                                                   84%

Group 3                                                   80%

As you can see, the differences between groups are noticeable.

But are they significant? Not in formal research’s sense of significance, but rather in terms of the impact on your child’s education. Yes, those with the highest degrees are almost always going to be among those with the highest intellect and knowledge base in their field. But with under 2 percent of the population holding a PhD and only 12% holding an advanced degree of any type according to the US Census, aren’t we nitpicking to make distinctions between colleges that have 90% and those that have 80% of these highly capable individuals on their faculty? Most college professors are extremely bright and knowledgeable.

Given this, doesn’t the question become which are best at sharing that knowledge with others? Aren’t the students’ opinions of those professors more important than which advanced degree they have? Doesn’t this matter more when considering the quality of education at any given institution? Isn’t the inspiration students feel to delve deeper into an area of study important to an education? Even if you only view college as a means to a career, isn’t a person’s enthusiasm for their work critical to their success in that field?

In addition to providing the basics about each school, Princeton Review’s annual Best Colleges books are an excellent source of students’ opinions of colleges with regard to numerous factors, including the quality of their professors. Over 100,000 students fill out their questionnaires each year, which include both ratings scales and areas for written reviews. The result is an in-depth look at what the consumer thinks of the product, a critical component of ratings that many others don’t take into consideration.

And student opinions, it turns out, support my suspicion that higher academic achievement doesn’t necessarily make one better at imparting what one knows to others.

Professors Good at Teaching?

Group 1                                                   90

Group 2                                                   92

Group 3                                                   89

Data Source: Princeton Review Best 386 Colleges (2021)

On a scale of 60-99 used by Princeton Review, students at Groups 2 and 3 rate their professors just as interesting as students in Group 1 do.

Okay, for the final comparison of our three groups of schools, I looked at ratings of overall quality of life in Princeton Review’s Best Colleges book. This measure, again on a scale of 60-99, is a combination of students’ opinions of their campus, dorm comfort and food quality, among other factors.

Quality of Life

Group 1                                                   90

Group 2                                                   90

Group 3                                                   88

Data Source: Princeton Review Best 386 Colleges (2021)

As you can see, once again the three groups are essentially the same.

Let’s go back and take a look at the composition of these 3 groups again, just so you can have it clearly in your minds at which schools there’s really not that much of a difference on so many factors.

GROUP 1

Brown

Dartmouth

Duke

Northwestern

Penn

Princeton

Rice

Stanford

Swarthmore

Yale

GROUP 2

Wesleyan

Middlebury

Emory

Carnegie Mellon

Villanova

Tufts

U Richmond

Boston U

Vassar

Colgate

GROUP 3

Pepperdine

Bucknell

Baylor

Worcester Polytech

Lehigh

Santa Clara U

Southern Methodist U

Fordham

Lafayette

Kenyon

Given all of the above, is it really worth the risk of all the negatives that can come with chronic stress in order to maybe gain a slight advantage in earning admission to a college that’s a bit higher ranked? Sure, Group 1 is filled with phenomenal places for exceptionally bright, ambitious, hard-working, talented students to get an education. If they meet your criteria for what you’re looking for in a college, then apply. If you’re admitted, that’s wonderful–celebrate!

Groups 2 and 3 are pretty darn awesome, too, though.

Note: Due to the pandemic’s effects on some of the numbers in the 2020-21 Common Data Sets, the 2019-20 data is a more accurate reflection of what students can expect. I will update this page when things get back to ‘normal’.