Currently browsing the archives for October 2017

« Newer Posts

Survey for Williams presidential search: Please participate

To the Williams Community,

I write to follow up on my letter in early September regarding the search for Williams’ 18th President. The Presidential Search Committee has begun its work, with the assistance of a senior team from Spencer Stuart, a national executive search firm with expertise in higher education leadership searches, whom we retained to assist in the process. Members of the Committee along with the Spencer Stuart team have held a series of on-campus forums with faculty, staff, students and alumni in order to answer questions and, especially, solicit ideas from the Williams community regarding the search. As I indicated in my last note, information about the Presidential Search Committee, Spencer Stuart and, as we proceed, other matters related to this important work can be found on our search website.

As another important step in the input gathering phase, the Committee has prepared a survey to allow the broadest possible participation from our community. I encourage you to use it to share your views, which will help guide the drafting of a prospectus to be shared with potential candidates, and also help to guide the Committee in the process of interviewing and of refining the candidate pool. In order to be able to incorporate your input, the Committee asks for a few minutes of your time to complete the survey by October 18th.

In addition, the Committee welcomes nominations for the position. If you would like to suggest a candidate, please send an email with any supporting materials to the confidential address: WilliamsPresident@spencerstuart.com.

On behalf of the Presidential Search Committee, we thank you for taking part in this survey and look forward to updating you on our progress over the coming months.

Sincerely,

Michael Eisenson ’77
Chair, Williams College Board of Trustees

Facebooktwitter

Historically Waspy Campuses

Two interesting comments from Muddy:

And this can be accomplished only if the “others” on these historically Waspy campuses exist is such a critical mass that they feel empowered and heard in a meaningful way. My feeling (I’ve worked my entire career on college campuses) is that the current numbers of kids of color at Williams and elsewhere is pretty much at the minimum it needs to be in order for the entire community to benefit from the immeasurable good their presence adds to the educational quality everyone enjoys.

If you really mean “at a minimum,” then I have a deal for you! Let’s replace 25 (or 50!) of the white students in the bottom admission bands (say, AR 4 and below) with 25 (or 50!) Asian-American (or Asian-Asian) applicants with AR 1 that the College currently rejects. This would increase the “current numbers of kids of color at Williams” while, simultaneously, improving the academic quality of each class. Since many (most?) such white admits are athletic tips, the cost might be a few more losses in certain sports, but, even then, Williams would still have an above average athletic programs.

I am honestly curious what you think of this proposal.

The best, most aspiring, most intrinsically interesting white kids will not generally feel compelled by a campus that minimizes the kinds of values I am talking about or one that is seen to be backtracking on its commitment to diversity. Less kids of color means less high value students of every background.

Perhaps, but I doubt it. Consider Middlebury and Caltech, two very different schools, both of which place much less emphasis on African-American enrollment than Williams currently does. Middlebury is at 4% and Caltech at 2% for the class of 2020. I have never heard of a white (or Asian-American!) student reporting that such low African-American enrollment was a reason why they turned down Middlebury/Caltech. Have you? I find the whole thing absurd because the number of white/Asian students who are even aware that Williams is 8% (twice as much as Middlebury!) and Amherst is at 12% (6 times more than Caltech!) is, essentially, zero.

But, as always, contrary opinions welcome. Do you know a white/Asian-American student who turned down Middlebury or Caltech because there were too few African-Americans?

The most subtle argument involves critical mass. While I have never met a white/Asian-American student who knew/cared about differential percentage of African-American enrollment across Middlebury/Williams/Amherst, I know that many African-American students themselves care a great deal. So, perhaps if we didn’t accept 20 or so African-American students from AR 6 and below, we would not be able to enroll the AR 3 and above African-American students whom we most want. Perhaps. Informed commentary welcome!

Facebooktwitter

Preview of Coming Attractions …

Screen Shot 2017-10-05 at 5.40.37 PM                     Williams_seal

 In the spirit of Wikileaks, may I submit for avid readers a preview of what may be next on your favorite blog. The first column are the titles of ‘scheduled’ posts. The second are ‘drafts’ being worked on. All of the ‘scheduled’ are Dave’s. All of the ‘drafts’ are Dave’s with the exception of ‘no title’ and ‘College’s Sexual Assault Policy’

 I do this as a service to readers waiting for the next topic. I know Dave won’t mind given his own questionable use of materials

My fervent hope is that no one reads this blog except the disillusioned, the dyspeptic, and those given to the dissection of data.

 

Facebooktwitter

Comments on International Admissions

This comment merits further discussion.

There are two issues with that [loosening the international quota]. The first is that international students have considerably lower graduation/retention rates than any other demographic group at the top schools. That’s not a consequence of ability but rather of uncertainty: financial aid for international students often doesn’t increase in later years, there is a geographic barrier, and foreign political/economic situations can complicate their coming back.

False. Here is the latest data on graduation rates:

gradu

International student 6-year graduation rate is about the same as that of white/Asian students, as we would expect. African-American/Hispanic students are about twice as likely to fail to graduate from Williams in 6 years.

Now, this data has evolved over time and you may be right about both earlier periods and about 4-year graduation rates. But, even then, a big driver is “diversity” among the international students. Not all international students are AR 1, after all. Indeed, I would not be surprised if some (many?) international students are AR 3 and below, if they come from the sorts of countries (not China, South Korea, England, et cetera) that Williams likes.

When I recommend increasing international enrollment, I mean for AR 1 students. Speaking roughly, I would start with about 25 more students from China/Korea/Japan.

The second issue is that the international pool is not as strong as it is constantly made out to be. Many of these students are not informed about how competitive US colleges are, so you get a lot of weak applicants applying when they have no chance of getting in. This is backed by the statistics of need-blind for international students schools like MIT and Amherst: the international acceptance rate is a third of the domestic one, even though these colleges have made assurances to not let ability to pay influence the likelihood of getting in. Many colleges (Williams, Wesleyan, Swarthmore) report a similar pattern: an international acceptance rate 1/4-1/2 that of domestic students.

Is the acceptance rate low because the pool is weaker or because these schools, like Williams, have a quota on international students?

Everyone that I have discussed this with — although contrary opinions are welcome — suggests that there are, at least 50 AR 1 international applicants (many not requiring any financial aid) who are currently rejected by Williams but who would enroll if given the chance. Do you disagree?

Even if students stand out academically, it isn’t enough. Prominent international universities like India Institute of Technology and Tsinghua University admit solely by performance on a test. The UK institutions- Cambridge, UCL, LSE, Oxford- don’t care about extracurricular activities at all. On the contrary, The top US colleges don’t just want perfect scorers. Williams doesn’t either. As a residential college, it wants committed students who will engage critically and meaningfully with their peers and their community. As a distinguished and scholarly place, it wants those who are committed to learning and open to having their viewpoints expanded and challenged across a broad spectrum of fields. Those things can only be evaluated by subjective perspectives, not the SAT.

False. First, there is no evidence that AR 1 applicants are, relative to AR 4 applicants, any less willing to “engage critically and meaningfully with their peers and their community.” If anything AR 1 students are more willing, or, at the very least they are much more willing to engage in academic work, and with a talent for doing so.

Second, are you arguing that the current Williams admissions process uses “subjective perspectives” in evaluating candidates? As if! Or are you arguing that it should? Perhaps. I am always happy to entertain a discussion of changes in the admissions process.

Not to say that Williams has done enough or that it should be content with where it is- the simple fact that you have 8400 students applying compared to 40000 at some top universities means that there is a significant cohort of good fit, high stats international students who should apply and largely be admitted. But here’s another question: how will Williams convince them to apply and attend over HYPS + other Ivies + other top 20 universities? The LAC name brand is virtually non-existent outside of the States, even for Williams and Amherst (beyond maybe Oxford/Cambridge/London).

Williams doesn’t need to convince 40,000 (or 40) high schools students (who don’t apply) to apply. We have plenty of applicants already! We just need to change who we admit and who we reject.

Facebooktwitter

Any response to Jordan Hampton ’87 email re input for new prexy search?

Screen Shot 2017-10-04 at 10.01.45 AM

 

Rather nice line drawing of campus landmarks.

I’ve attached my input beneath the fold.  Read more

Facebooktwitter

There are Hundreds of Rejected AR 1s

Some readers doubted whether or not there were enough high quality applicants (currently) rejected by Williams who could be admitted as part of this plan. Allay those fears! There are hundreds of rejected AR 1s (and even more AR 2s) who would love to attend Williams if we were to accept them. Evidence:

Recall the 2005 Recipe (pdf) article:

The admission staff wait-listed or rejected nearly 300 of the 675 applicants to whom they had given their top “Academic 1” rating — a pool of students that, on average, ranked in the top 3 percent of their high school classes and had SAT scores of 1545.

Note Adam Falk’s report that, in the fall of 2013, Williams received more than 1,200 applications from students with academic ratings of 2. Since Williams accepts many fewer than 1,000 students in total from this bucket, there must also be hundreds of AR 2s who are rejected.

Amherst, to its credit, is much more transparent with its admissions data. Consider:

am4

Amherst admissions are not Williams admissions and SAT verbal scores are not the same thing as academic ratings. But, if there are almost 2,000 students with 700 and above verbal SAT scores who are rejected by Amherst, then there must be at least a few hundred AR 1 students rejected by Williams.

Facebooktwitter

Students: Consider the Spanish Major. Our hemisphere needs active participation in ecologic, economic, and social inputs and programs!

Map of Spanish speaking countries

From the Williams College Spanish Course listings:

Our graduates have gone on to secure prestigious Fulbright teaching and research grants in Latin America and Spain, and many have used their linguistic and cultural training as they pursue careers in fields including law, health care, journalism, labor and community organizing, education, and doctoral degrees in various fields.

https://spanish.williams.edu/

This may be even more true today after the President’s speech in Puerto Rico, as fluent Spanish speakers grasped for the exact word:  el pendejo, el huevón, el pelotudo.

Facebooktwitter

How Admissions Works at Williams

Williams admissions work the same as admissions at most other elite colleges. If you understand the process at Swarthmore or Princeton, then you understand 99% of what happens at Williams. There are a variety of books about admissions at elite colleges, e.g., The Gatekeepers and A is for Admission. They capture 90% of the details. (These books are somewhat dated and may guild the lily a bit when it comes to race.) Williams Magazine published (pdf) an excellent 2005 article, “Recipe for Success,” about admissions. Virtually everything in it is true, but it also leaves out many of the more controversial aspects.

The purpose of this post is to explain how the Williams admissions process works in reality, not how it should work.

First, the most important part of the admissions process is the “academic rating,” often abbreviated as “AR.” From the Recipe article:

The full-time admission staffers, plus a handful of helpers like Phil Smith ’55 (Nesbitt’s predecessor as director), pore over the folders. Two readers examine each folder independently, without seeing each other’s comments, and assess them in three major ways. Each applicant gets an academic rating from 1 to 9 that focuses heavily on his or her high school grades, standardized test scores, the rigor of his or her academic program within the context of the school setting and the strength of teacher recommendations.

Nurnberg ’09 et al (pdf) provide a similar description:

After evaluating the applicant’s SAT scores, high school grades, essays, class rank, high school academic program, support from the high school administration, AP test score — or IB test scores — and teacher recommendations, admissions readers assign the applicant an academic rating from the scale 1 — 9, with 1 being the best.

Amherst, and all other elite colleges, use essentially the same system. The College does not like to reveal the details of these ratings, but we know from Peter Nurnberg’s ’09 thesis that:

While the academic reader ratings are somewhat subjective, they are strongly influenced by the following guidelines.

  • Academic 1: at top or close to top of HS class / A record / exceptional academic program / 1520 – 1600 composite SAT I score;
  • Academic 2: top 5% of HS class / mostly A record / extremely demanding academic program / 1450 – 1520 composite SAT I score;
  • Academic 3: top 10% of HS class / many A grades / very demanding academic program / 1390 – 1450 composite SAT I score;
  • Academic 4: top 15% of HS class / A – B record / very demanding academic program / 1310 – 1400 composite SAT I score;
  • Academic 5: top 20% of HS class / B record / demanding academic program / 1260 – 1320 composite SAT I score;
  • Academic 6: top 20% of HS class / B record / average academic program / 1210 – 1280 composite SAT I score;
  • Academic 7: top 25% of HS class / mostly B record / less than demanding program / 1140 – 1220 composite SAT I score;
  • Academic 8: top 33% of HS class / mostly B record or below / concern about academic program / 1000 – 1180 composite SAT I score;
  • Academic 9: everyone else.

These ratings are high-school-quality adjusted. At an elite school like Boston Latin or Exeter, you can be in the top 5% or even lower and still be an AR 1. At a weaker high school, you need to be the valedictorian. At the weakest high schools (bottom 25%?), even the valedictorian is almost never considered smart enough to go to Williams, at least in the absence of top standardized test scores.

Note that the working paper (pdf) from which these details are taken was co-authored by then-Williams president Morty Schapiro, so one hopes that it is accurate! Nurnberg’s senior thesis included a copy of the “Class of 2009 Folder Reading Guide, Academic Ratings,” which provided these details:

      verbal   math   composite SAT II   ACT    AP
AR 1: 770-800 750-800 1520-1600 750-800 35-36 mostly 5s
AR 2: 730-770 720-750 1450-1520 720-770 33-34 4s and 5s
AR 3: 700-730 690-720 1390-1450 690-730 32-33 4s

Williams, and all other elite schools, use this system because academic rating does a wonderful job of predicting academic performance at Williams and elsewhere.

Perhaps the main reason that this post is necessary is that Williams, when politically convenient, likes to deny the fundamental realities about how it decides who to admit and who to reject. Consider President Adam Falk and Director of Admissions and Financial Aid Liz Creighton ’01 writing in the Record:

[T]he very notion that the “quality” of students can be defined on a single linear scale is preposterous

Academic rating is, precisely, a “single linear scale” and it is, by far, the major driver of admissions decisions. This is true both for the process as a whole and within sub-groups. For example, African-American applicants with academic rating 1 to 3 are virtually certain to be admitted while those with academic rating 8 or 9 are almost always rejected. The College may have different standards across sub-categories but, within each subcategory (except athletes and development prospects), the academic rating explains 90% of the variation.

Second, students with an academic rating worse than 2 (i.e., 3 or higher) are summarily rejected unless they have a specific “hook” or attribute.

The Recipe is explicit:

In general, all applicants with a combined academic rating of 3 or higher are rejected at this point, unless the first and second readers have identified one or more “attributes” that warrant additional consideration.

Details:

The readers also assign any of more than 30 “attributes” that admission uses to identify exceptional traits. Some of these are easily quantified, such as being the child or grand-child of an alumnus, a member of a minority group, an “impact” athlete or a local resident. Other more subjective “tags” draw attention (usually but not always favorably) to something special about a candidate, like a powerful passion or aptitude for scientific research or an interest in getting a non-science Ph.D.

From Nurnberg ’09 el al, attributes (in addition to race/ethnicity/gender) include:

alumni grandparent, alumni other, alumni parent, alumni sibling, studio art, development or future fundraising potential, dance, institutional connection,
intellectual vitality, local, music, politically active, religious, research science, economically disadvantaged, social service, theater, top athlete, tier 2 athlete, and tier 3 athlete

At this stage, the naive reader will assume that all these attributes have a similar effect. Being a great musician or a great athlete will help some AR 4s get into Williams, and that is OK. (And the College wants you to think that.) In fact, some attributes matter much more than others. Recall (from 2004!) Admissions Director Dick Nesbitt ’74:

We are able to admit roughly 120 top rated musicians each year from the top of the academic reader rating scale–what we refer to as academic 1′ and 2’s (broadly defined as 1500+ SAT’s and very top of the class).

In other words, for many/most attributes, the College does not need to dip below AR 1s and 2s. Yes, being a top musician may help you in the competition with other outstanding students, but, if you are AR 3 or below, it won’t. You will be rejected. And the same applies to other attributes. Top students are also, often, deeply involved in social service or theater. In high school, they often excel in research science or political activism. If Williams were to admit only AR 1s/2s, it would have plenty of students in all these categories.

Third, for applicants with AR 3 or below, the attributes that matter most are race, income and athletics.

Does this mean that no other attributes ever matter? No! It is certainly the case that the daughter of a prominent alum could get into Williams as an AR 4 or the son of a Williams professor as an AR 3. But the major categories, the ones that account for the vast majority of AR 3 and below admissions are race, income and athletics.

Don’t want to read all the posts from those links? Here is a brief summary:

1) There are 100 or so admissions decisions which are driven by a Williams coach. You are either on her list or you are not. These “tips” and “protects” are, by definition, only used for students with AR 3 and below. Best single post overview of the topic is here.

2) In the class of 2020, Williams has (pdf) 115 African-American/Hispanic students. Many of these are AR 1 or 2 applicants who would have been accepted at Williams regardless of which box they checked. But a majority, probably a vast majority, are AR 3 or below. Recall this discussion of SAT scores:

ccf_20170201_reeves_2

Asian-Americans in the 700+ range are at least 6 times more common than African-Americans/Hispanics. So, how can Williams have more African-Americans/Hispanics than Asian-Americans enrolled? (Hint: It isn’t because there aren’t 100+ Asian-Americans among the AR 1/2 applicants who are currently rejected by Williams.) The reason is that Williams admits scores of African-American/Hispanic applicants with AR 3 and below. Williams does this because it wants a class which “mirrors” or “reflects” the US population, at least when it comes to African-Americans and Hispanics. Note that the average African-American student at Amherst has an SAT score consistent with AR 5. It is highly unlikely that Williams does a better job than Amherst at attracting highly rated African-American students.

3) Unlike athletics (which the college is, sometimes, transparent about) and race (on which there is good data), family income and parental education are trickier. The College reports (and is proud of the fact) that about 20% of students are eligible for Pell Grants and that about 20% of students are first generation college students, meaning that they come from families in which neither parent has a 4 year BA. (Of course, there is a big overlap between these two groups, and, to a lesser extent, between these two groups and African-American/Hispanic students.) The problem is that all standardized test results (and, therefore, academic rating) are skewed against such students. So, in order to get to 20%, Williams must admit scores of such students with AR 3 or below.

About 1/2 of a Williams class is AR 1 or 2. (The median math+verbal SAT score at Williams is 1450, which is the bottom of AR 2.) There are 100 recruited athletes (all of whom, by definition, are AR 3 or below), 100+ African-American/Hispanic students, 100+ first generation and 100+ Pell Grant recipients. That adds up to 400+ in a class of 550! Many students fall into more than one category. Many (outside the athletes) are AR 1 or 2. But, given that we only have 275 spots left beneath AR 1/2, a large majority of the bottom half of the class are members of at least one of these 4 categories. The bottom 100 students in each class (approximately AR 5 and below) is almost completely dominated by these students. And, in the categories outside of athletes, academic rating drives the decisions. Williams is much more likely to accept an African-American and/or a first generation student and/or a future Pell Grant recipient if her academic rating is 1 to 3. Every single AR 9 applicant is rejected, regardless of her other outstanding attributes.

And that is how admissions works at Williams, and almost all other elite colleges.

Facebooktwitter

Admissions Data

An anonymous source sent me this file (csv) of data related to Williams admissions.

[UPDATE: Data removed at the request of the College.]

> library(readr)
> x < - read_csv(file = "http://ephblog.com/wp-content/uploads/2017/09/admissions.csv")
> x
# A tibble: 2,110 x 10
   class enrolled state       country      ethnicity   sex   act reading  math writing
                                    
 1  2017        0    AZ United States Asian American     M    NA     770   790     770
 2  2019        0    AZ United States Asian American     F    35     730   770     760
 3  2019        0    AZ United States Asian American     M    NA     800   720     800
 4  2019        0    BC        Canada Asian American     F    NA     800   750     750
 5  2013        1    CA United States Asian American     F    NA     790   800     800
 6  2013        1    CA United States Asian American     M    NA     760   780     790
 7  2013        1    CA United States Asian American     M    NA     790   800     710
 8  2013        1    CA United States Asian American     F    NA     650   590     670
 9  2014        1    CA United States Asian American     F    NA     790   780     720
10  2014        1    CA United States Asian American     F    35     750   800     700
# ... with 2,100 more rows
> 

Comments:

1) Does this look real to you? It does to me, although it is obviously just a sample. Opinions welcome.

2) Should I spend a week exploring this data?

3) The sample is a strange subset of what the “complete” data must look like. For example:

> table(x$class, x$enrolled)
      
         0   1
  2011  86  99
  2013  96 119
  2014 123 105
  2015 124 116
  2016  77 125
  2017 232 159
  2019 172 143
  2020 164 170

a) Note that there is no data for the class of 2018. Perhaps removing this data is one way that Williams keeps track of who it gave this data to and, therefore, who it can go after for leaking it to me.

b) The numbers of students range for 185 for the class of 2011 to 391 for the class of 2017. Since around 1,250 applicants are admitted to Williams each year, we definitely don’t have the complete data.

c) It is interesting to see data for applicants that we admitted — I assume that everyone in this data was admitted — but who chose not to enroll.

d) Would you believe a 230 point difference between Asian-American and African-American SAT scores among Williams students?

> x %>% filter(enrolled == 1) %>% group_by(ethnicity) %>% 
     summarise(count = n(), act = round(mean(act, na.rm = TRUE)), 
               sat = round(mean(reading + math, na.rm = TRUE))) %>% 
     arrange(desc(sat))
# A tibble: 7 x 4
        ethnicity count   act   sat
               
1  Asian American   186    34  1506
2    Unidentified    18    34  1488
3           White   569    33  1480
4          Non-US    24    31  1374
5 Hispanic/Latino    99    30  1341
6 Native American     7    26  1302
7           Black   133    29  1274

That is what the data suggest . . .

Can’t resist adding an image:

density

Code for generating this below the break.
Read more

Facebooktwitter

« Newer Posts