Currently browsing posts filed under "Legacy Admissions"

# Eph Fecudity

How many children does an average Eph have? My current guess is at least 1.5, and probably more.

Consider a not-so-randomly selected first-year entry from the mid 1980s. Those Ephs are now into their 50s with, presumably, most of their reproduction complete. The entry had 24 students, 12 men and 12 women. It has produced at least 36 children. Three of the women and (I think) three of the men had no offspring. The remaining 18 averaged exactly two children each. Comments:

1) This is a minimum. If I only relied on the Alumni Directory, I would only have found 32 children. One (male) alum, with 4 children, had not recorded any of them in the directory. I may have missed others.

I am especially suspicious of two other male alums with no children listed. I think — opinions welcome! — that male alums are much less likely to be childless than female alums, and that male alums are less likely to have accurate entries in the directory. Or is that an unfair stereotype?

2) Perhaps some (older!) readers could report the data for their own freshmen entries? Although entries are small, they are (very?) random, so just counting all the children from a single entry probably provides a not-unreasonable estimate.

3) Given that I sampled 24 students out of a class of 500, what is the confidence interval for my 1.5 estimate? I probably should have kept track of the 24 individual values and done a bootstrap . . .

4) This is relevant for our discussions about legacy admissions. If 1.5 is accurate then, for the class of 2024, applying this fall, there are 750 or so high school seniors with an Eph parent (and hundreds more with an Eph grandparent). Around 75 of them will become students at Williams. Is is hard to believe that the top 10% of the distribution of Williams children might be academically equivalent to the other 475 members of the class of 2024? Not at all.

5) A rigorous way of exploring this conclusion would be to calculate the expected regression to the mean of children in terms of the academic abilities of their parents. Smart people have smart children, but generally not children as smart as them. So, the average child of an Eph would not be smart enough to get into Williams. But the top 10%? I bet yes. (Readers are welcome to provide their own calculations in the comments.)

This Atlantic article about legacy admissions is a mess, but it does have a couple of interesting data points. Day 2 of 2.

From the article:

Yale is an interesting case study. The school currently gives the children of alumni an admissions bump, but from 1980 to 2010, the proportion of students in its freshman class with a parent who also attended dropped from 24 percent to 13 percent.

1) Just because Yale tells a gullible Atlantic reporter something — like that it does not discriminate against Asian-Americans — does not make that something true. Yale has many reasons, mostly related to fund-raising, to claim that it gives legacies a “bump,” even if — especially if! — the bump is so small as to be invisible.

2) I don’t think the drop at Williams has been so dramatic, but there has been a drop. I think a recent class was only 10% legacy, whereas the usual number was closer to 15%. I don’t know what it was in the 80’s, although that information is available in the library in the annual letters that the Admissions Office used to produce for the trustees. On my list of projects for reunion week-end!

3) The numbers for Yale help to explain the dramatic increase in legacy quality at Yale. (We don’t have Yale data but there is no reason why the trends would not be the same there as at Harvard/Williams.) First, if you only take half as many legacies, you can reject the really stupid ones. Second, the doubling of the Yale undergraduate class size from the 60s to the 80s means that there are, more or less, twice as many legacies to choose from. Third, Yale students in the 80s are much smarter than those from the 50s, so their children will form a much more academically accomplished pool to choose from.

This Atlantic article about legacy admissions is a mess, but it does have a couple of interesting data points. Day 1 of 2.

Applying to college as a legacy is like having a superpower. It has been estimated to double or quadruple one’s chances of getting into a highly selective school, and has been found to be roughly equivalent to a 160-point boost on the SAT.

I need to re-work my annual post on legacy admissions to deal more directly with (excellent!) comments/criticisms from folks like abl. In the meantime, can we agree that the above is incredibly misleading? The average SAT score for Williams legacies is higher than the average for non-legacies. Nor is this only a Williams phenomenon:

And a Harvard spokesperson told me that admitted legacies tend to have higher median test scores and grades than the rest of admitted students. This doesn’t make the admissions advantage that legacies are given defensible, but it’s possibly another reason that the status quo of legacy admissions persists.

Now “admitted students” are not the same as “enrolled students,” which is the real comparison we want. But Harvard enrolls 80%+ of its admitted students, so the statistics for admitted students are very likely to be similar to those of enrolled students. Moreover, it is not obvious which way any differences would go.

The question is that same as before. If, among enrolled students, the average Harvard legacy scores 1500 (or whatever), and the average non-legacy scores 1480, how much of a preference could being a legacy possibly be?

Entire article below the break:

# Legacy Admissions Play No Meaningful Role at Elite Colleges

tl;dr: Legacy status does not provide a meaningful advantage in admissions to elite colleges like Williams. People like Sam Altman and Arjun Narayan ’10 are wrong, either because of genuine ignorance or because of a (unconscious?) refusal to confront the major beneficiaries of admissions preferences: athletes and (non-Asian) racial minorities. (If Sam has complained about extra considerations that Stanford gives football players and African-Americans, I must have missed it.)

Hasn’t Arjun Narayan ’10 ever read EphBlog? We have been documenting these facts for over a decade. From 2008:

Morty [then Williams President Morton Schapiro] noted that a decade or so ago [or perhaps when he arrived?], the average legacy was a 3.3 on the 1-9 scale of academic ranks while the average non-legacy was 2.3. Morty did not seem to be a huge fan of this gap, or of giving legacies such a preference. He then noted that the latest statistics show that legacy and non-legacy are now equivalent (both at 2.3). Morty confirmed, consistent with all the analysis I have done, that being a legacy is not a meaningful advantage in getting into Williams.

Director of Communications Mary Dettloff kindly provided this update for 2017:

I had a conversation with Dick Nesbitt about this, and he says it has long been our policy not to release academic standing information for specific subgroups of students. That said, he also shared that for at least the last 20 years, the legacy students have had equal, if not marginally stronger, SAT scores and Academic Rating when compared to the rest of their classmates.

Case closed.[1]

More importantly, should we be surprised that students whose parents went to elite colleges are much more likely to win admissions to elite colleges themselves? No! Nature and nurture are passed down through the generations now, just as they always have been.

Consider professional baseball. From the New York Times:

A random US man has a 1-in-15,000 chance of playing in the MLB. The son of an MLB player has a 1-in-75 chance. In other words, your odds of playing in the MLB are 200 times higher of your father played. Given that fact, should we be surprised if your odds of coming to Williams are 200 times higher if your parent is an Eph?

The mechanisms in both cases are the same. Genetics play a major role. The specific genes — probably thousands of them — that help you to hit a curve ball are passed from father to son. The genes that aid in doing well in school and on standardized tests are passed on just as easily. Nurture matters. Baseball players probably provide their sons with a better than average environment in which to learn baseball. Ephs who become parents do the same. You should no more be surprised at the high numbers of legacies at elite colleges than at the high numbers of baseball children in the Majors.[2]

However, it is interesting to consider how legacy admissions have evolved in the last 30 years. In the 1980’s, it was tough for Williams to find 75 high quality legacies in drawing from Williams classes of the 1950s. First, the college was much smaller than, with fewer than half the current student population. Second, Williams was much less academically rigorous. (That is, there were plenty of not-very-smart students.)

In the 80’s, there were 500 academically accomplished students per class. Judging/guessing from what we see at reunions, the total number of children of a typical class is at least 500 and probably closer to 1,000. But only 75 or so find spots at Williams! Do the other 425 go to Stanford? Nope. And the same harsh mathematics apply to the children of other elite schools. Since smart people have smart children, the pool of legacies that the College has to choose from is very impressive. Williams does not need to lower standards at all to find 75 good ones.[3]

—————-
[1] To be fair to Altman/Narayan, there are some subtle counter-arguments. First, if it is the case that legacies, as a group, differ from non-legacies on other dimensions besides academic rating, then it might not be fair to compare the two groups directly. Instead, we should compare legacies with non-legacies who “look” like legacies. For example, if legacies are more likely to be white and non-poor, then comparing them with non-legacies is makes no sense. Instead, we should compare them with similarly white/non-poor non-legacies.

What they don’t tell you is that whites and Asians lacking the legacy hook need to be a lot better than “equal, if not marginally stronger” than the school average. Without a legacy, a student applying to a selective LAC should aim for the 75th percentile, which I take to be approximately the bottom end of AR 1. In terms of the old SAT, this would be 770 on the Math (vs 708 Williams average) and 780 on Critical Reading (vs 720 averages). These 75th percentile scores are each about a half-standard deviation higher than the average scores. So, a half-standard deviation in academic ability is what legacy status buys you at Williams.

Hmm. This is not obviously implausible. I should spend more time on this topic and reader pointers are welcome.

But, first, just how “white” are legacies. Williams was just as Black 30 years ago as it is today and Black Ephs have children too. I would assume (contrary evidence welcome) that the white/black ratio in legacy admissions is similar to the white/black ratio in the general student body. Why wouldn’t it be? (The same argument does not apply, obviously, to Hispanic/Asian admissions.) Second, plenty of legacies are also athletes, at least some of whom are recruited. Indeed, so many Williams sports are really rich-northeastern-elite pastimes that it would hardly be surprising if legacies were over-represented in sports like crew and squash.

So, I agree with KSM that comparing legacies to the overall pool is not perfectly fair but nor it is fair to compare them only to non-athlete white/Asian applicants.

The second subtle counter-argument: it could be the case that legacies come in two flavors: over-qualified and under-qualified. The over-qualified ones are exceptional candidates who turn down Harvard/Yale/Princeton/Stanford for Williams. The under-qualified ones receive substantial preferences in admissions. Combining the two groups creates an overall legacy group which is similar to non-legacies but which “masks” the substantial advantages given to under-qualified legacies.

[2] Of course, legacy students are much more likely to attend their parents’ alma mater than legacy baseball players are to play for the same team as their fathers. Exercise for the reader: Explore the industrial organization of elite colleges and major league baseball to explain this difference. Perhaps a better view is to consider all the legacy students as a whole, in the same way that the New York Times considers all the legacy baseball players. But this post is already long enough . . .

[3] sigh, an EphBlog regular, pointed out this study (pdf) on “The impact of legacy status on undergraduate admissions at elite colleges and universities.” The author (an Eph!) argues that legacy status matters a great (or at least did matter in the fall of 2007). I have now read (and taught!) this reasonable article, although I remain unconvinced, for reasons which will need to await for next year’s version of this post.

# Class of 2021 Admissions Data II

Let’s discus admissions data for the class of 2021. Key table:

Today is Day 2.

No one should be surprised that Williams yields whites better than it yields any other group. Is that a problem or an opportunity? I bet that white students from rich families who attended elite high schools are, on average, the happiest students at Williams. If so, should we admit more of them?

A similar analysis applies to legacies, the vast majority of whom are white. (Note that the legacy numbers are much iffier because the College does not (regularly) publish the exact numbers. President Falk usually provides an estimate of 1/7th but that certainly varies year-to-year. Indeed, the exact definition of “legacy” matters. We always include the children of alumni, never (?) the nieces/nephews and sometimes (?) the grandchildren. In any event, the 79 here is my estimate, equal to 1/7th of the 553 students in the class.)

I am amazed that we yield so well among legacies. Will 79 or so of the 86 legacies we admitted choose to enroll. That seems much too high to me. I know, just in my personal circle of friends, two legacy children admitted to Williams who went Ivy instead. Then again, perhaps the vast majority of those 86 were admitted early decision? Informed commentary welcome.

# Legacy Admissions Play No Meaningful Role at Elite Colleges

tl;dr: Legacy status does not provide a meaningful advantage in admissions to elite colleges like Williams. People like Sam Altman and Arjun Narayan ’10 are wrong, either because of genuine ignorance or because of a (unconscious?) refusal to confront the major beneficiaries of admissions preferences: athletes and (non-Asian) racial minorities. (If Sam has complained about extra considerations that Stanford gives football players and African-Americans, I must have missed it.)

Hasn’t Arjun Narayan ’10 ever read EphBlog? We have been documenting these facts for over a decade. From 2008:

Morty [then Williams President Morton Schapiro] noted that a decade or so ago [or perhaps when he arrived?], the average legacy was a 3.3 on the 1-9 scale of academic ranks while the average non-legacy was 2.3. Morty did not seem to be a huge fan of this gap, or of giving legacies such a preference. He then noted that the latest statistics show that legacy and non-legacy are now equivalent (both at 2.3). Morty confirmed, consistent with all the analysis I have done, that being a legacy is not a meaningful advantage in getting into Williams.

Director of Communications Mary Dettloff kindly provided this update for 2017:

I had a conversation with Dick Nesbitt about this, and he says it has long been our policy not to release academic standing information for specific subgroups of students. That said, he also shared that for at least the last 20 years, the legacy students have had equal, if not marginally stronger, SAT scores and Academic Rating when compared to the rest of their classmates.

Case closed.[1]

More importantly, should we be surprised that students whose parents went to elite colleges are much more likely to win admissions to elite colleges themselves? No! Nature and nurture are passed down through the generations now, just as they always have been.

Consider professional baseball. From the New York Times:

A random US man has a 1-in-15,000 chance of playing in the MLB. The son of an MLB player has a 1-in-75 chance. In other words, your odds of playing in the MLB are 200 times higher of your father played. Given that fact, should we be surprised if your odds of coming to Williams are 200 times higher if your parent is an Eph?

The mechanisms in both cases are the same. Genetics play a major role. The specific genes — probably thousands of them — that help you to hit a curve ball are passed from father to son. The genes that aid in doing well in school and on standardized tests are passed on just as easily. Nurture matters. Baseball players probably provide their sons with a better than average environment in which to learn baseball. Ephs who become parents do the same. You should no more be surprised at the high numbers of legacies at elite colleges than at the high numbers of baseball children in the Majors.[2]

However, it is interesting to consider how legacy admissions have evolved in the last 30 years. In the 1980’s, it was tough for Williams to find 75 high quality legacies in drawing from Williams classes of the 1950s. First, the college was much smaller than, with fewer than half the current student population. Second, Williams was much less academically rigorous. (That is, there were plenty of not-very-smart students.)

In the 80’s, there were 500 academically accomplished students per class. Judging/guessing from what we see at reunions, the total number of children of a typical class is at least 500 and probably closer to 1,000. But only 75 or so find spots at Williams! Do the other 425 go to Stanford? Nope. And the same harsh mathematics apply to the children of other elite schools. Since smart people have smart children, the pool of legacies that the College has to choose from is very impressive. Williams does not need to lower standards at all to find 75 good ones.[3]

—————-
[1] To be fair to Altman/Narayan, there are some subtle counter-arguments. First, if it is the case that legacies, as a group, differ from non-legacies on other dimensions besides academic rating, then it might not be fair to compare the two groups directly. Instead, we should compare legacies with non-legacies who “look” like legacies. For example, if legacies are more likely to be white and non-poor, then comparing them with non-legacies is makes no sense. Instead, we should compare them with similarly white/non-poor non-legacies.

Second, it could be the case that legacies come in two flavors: over-qualified and under-qualified. The over-qualified ones are exceptional candidates who turn down Harvard/Yale/Princeton/Stanford for Williams. The under-qualified ones receive substantial preferences in admissions. Combining the two groups creates an overall legacy group which is similar to non-legacies but which “masks” the substantial advantages given to under-qualified legacies.

[2] Of course, legacy students are much more likely to attend their parents’ alma mater than legacy baseball players are to play for the same team as their fathers. Exercise for the reader: Explore the industrial organization of elite colleges and major league baseball to explain this difference. Perhaps a better view is to consider all the legacy students as a whole, in the same way that the New York Times considers all the legacy baseball players. But this post is already long enough . . .

[3] sigh, an EphBlog regular, points out this study (pdf) on “The impact of legacy status on undergraduate admissions at elite colleges and universities.” The author argues that legacy status matters a great (or at least did matter in the fall of 2007). I have my doubts. Let’s dive into the details in the comments!

# (Don’t) Give Me Your Poor III

Third installment in our two week discussion of the recent New York Times article “Generation Later, Poor Are Still Rare at Elite Colleges” by Richard Perez-Pena. Interested readers should check out our collection of posts about socio-economic issues related to admissions, from which I have plagiarized extensively.

But critics contend that on the whole, elite colleges are too worried about harming their finances and rankings to match their rhetoric about wanting economic diversity with action.

“It’s not clear to me that universities are hungry for that,” said Richard D. Kahlenberg, a senior fellow at the Century Foundation who studies college diversity. “What happens if low-income students start calling the bluff of selective universities, and do start applying in much larger numbers? Will the doors be open?”

Kahlenberg has had a nice career regurgitating the Cathedral’s wisdom and making himself available to New York Times reporters. Unfortunately, he has been very wrong about several important items in his area of (alleged) expertise.

First, his initial claim to fame was to propose affirmative action based on class as a replacement for affirmative action based on race. Alas, he lacked the social science chops to understand that the massive number of poor, but academically successful, Asian Americans meant that, were a place like Williams to use class instead of race, its proportion of African American students would go toward zero very quickly. In fact, for virtually every African American student who enrolls at Williams then are 10 (100? 1000?) or more Asian Americans who come from poorer families but had better high school records.

Second, he edited a whole book decrying legacy admissions without realizing that, among elite schools like Williams, legacy status plays a minimal role in admissions, and that role is diminishing every year. See these posts for details. But the intuition is obvious enough:

In the 80’s, there were 500 academically accomplished students per class. Judging/guessing from what we see at reunions, the total number of children of a typical class is at least 500. But only 75 or so find spots at Williams! Do the other 425 go to Stanford? Nope. And the same harsh mathematics apply to the children of other elite schools. Since smart people have smart children, the pool of legacies that the College has to choose from is very impressive. So, it does not need to meaningfully lower standards to find 75 good ones.

Does Kahlenberg continue his stupidity here? You bet he does! Consider the quote:

What happens if low-income students start calling the bluff of selective universities, and do start applying in much larger numbers?

It all depends on which sort of low-income students we are talking about. If 10,000 poor kids with lousy SAT scores and bad high school grades were to apply to Williams, then Williams would reject them all, just as it rejects thousands of rich kids with poor scores/grades. If lots of poor students with amazing scores/grades started applying, then Williams would accept them, as would other elite colleges.

But the truth, which Kahlenberg either doesn’t know or is too slippery to admit, is that the main issue is poor students with less impressive academic credentials than the non-poor students Williams currently admits/enrolls. If Williams has to choose between poor kids with 1350 SAT scores and rich kids with 1550 (and similar differences in high school grades), then it will (should!) choose the rich kids. If it doesn’t, it won’t be an elite school for long.

No bluff calling is required.

Director of Institution Research Chris Winters ’95 kindly answered my question about legacy admissions in the class of 2014.

Over the past decade or so, the percent of direct plus (unduplicated) skipped legacies in the matriculating class has been in the range of 11-17%. Direct legacies make up the vast majority of those.

Alas, I had hoped that Chris would give us more details about the class of 2014. Recall that last year the data was more detailed:

Director of Institutional Research Chris Winters ’95 reports on the numbers for the class of 2013. There are 69 students (13%) with at least one alumni parent and another 10 (2%) or so with no parent but at least one grandparent. (Some people restrict “legacy” to mean the children of alums, others include grandchildren.)

The most useful thing to know about legacy admissions would be their average academic rating as compared to the class as a whole. Three years ago:

Morty noted that a decade or so ago [or perhaps when he arrived?], the average legacy was a 3.3 on the 1-9 scale of academic ranks while the average non-legacy was 2.3. Morty did not seem to be a huge fan of this gap, or of giving legacies such a preference. He then noted that the latest statistics show that legacy and non-legacy are now equivalent (both at 2.3). Morty confirmed, consistent with all the analysis I have done, that being a legacy is not a meaningful advantage in getting into Williams.

But, since that time, the legacy pool has only gotten stronger and more competitive. Could the average AR of legacies now be higher than that of non-legacies? Perhaps. But a proper comparison would adjust for key confounders like race, athleticism and nationality.

A great topic for a senior thesis!

In Speak Up, JeffZ linked to an interesting op-ed piece in today’s NY Times, which I thought was worth its own post. One of the key sections of the article says:

Among selective research universities, public and private, almost three-quarters employ legacy preferences, as do the vast majority of selective liberal arts colleges. Some admissions departments insist they are used only as tie-breakers among deserving applicants. But studies have shown that being the child of an alumnus adds the equivalent of 160 SAT points to one’s application (using the traditional 400-to-1600-point scale, and not factoring in the new writing section of the test) and increases one’s chances of admission by almost 20 percentage points.

At many selective schools, legacies make up 10 percent to 25 percent of the student population. By contrast, at the California Institute of Technology, which has no legacy preferences, only 1.5 percent of students are the children of alumni.

My impression, based on previous posts/discussions here in EphBlog, is that simply being the child of an alum is of little help in gaining admission to Williams. But I don’t know if my impression is correct, or if there is any way of quantifying the admissions advantage conferred by being a legacy applicant, but its a topic in which I have some interest, as college is at least on the distant horizon (my oldest son is almost 9). This has been a topic has been previously discussed on EphBlog; previous discussions can be found here.

Apparently between 12-15% of each Williams class is made up of students who have either a parent or grandparent who went to Williams. Is that a good thing? And should legacies get any admissions advantage?

1) Director of Institutional Research Chris Winters ’95 reports on the numbers for the class of 2013. There are 69 students (13%) with at least one alumni parent and another 10 (2%) or so with no parent but at least one grandparent. (Some people restrict “legacy” to mean the children of alums, others include grandchildren.)

2) Here are my notes on Morty’s remarks about legacies from reunion last June.

Many schools (not naming names but mentioned Amherst at 9% in this context) seem to want to keep legacies to single digits. That seems stupid to Morty. We are at 13%-15% legacies defined as mom or dad (or both) at Williams. Add another 3% for grand children. Legacies are good kids, more likely to be JA. Williams gives 1/2 the advantage to legacies that it did 15 years ago.

It is not clear to me what it means for Williams to give 1/2 the advantage that it did 15 years ago. Half of what? My guess would be that this refers to the difference in Academic Rating between legacies and non-legacies. But recall what he said in 2008:

Morty noted that a decade or so ago [or perhaps when he arrived?], the average legacy was a 3.3 on the 1-9 scale of academic ranks while the average non-legacy was 2.3. Morty did not seem to be a huge fan of this gap, or of giving legacies such a preference. He then noted that the latest statistics show that legacy and non-legacy are now equivalent (both at 2.3). Morty confirmed, consistent with all the analysis I have done, that being a legacy is not a meaningful advantage in getting into Williams.

3) How can both these claims be true, that legacies get an advantage (if only half as much as they used to) and that the average legacy has the same Academic Rating as the average non-legacy? Easy! The key is whether you are comparing legacies to applicants that are like them (rich, mostly non-URM and non-tip, from good schools, and with college educated parents) or to all applicants. The second group includes many more URMs and athletic tips, both with substantial admissions advantages, than the former. So, legacies are, on average, the same as all students but not (quite) as qualified as the more elite pool which has many fewer URMs/tips.

4) There is still an amazing senior thesis to be written about legacy admissions at Williams. You should write it.

Summary: Legacy status counts for much less at Williams then it did 10 or 30 years ago. The doubling of the number of students in the 70s meant that the (fewer) children of 50s graduates had (proportionately) more open spots. The dramatic increase in student selectivity in the 80s meant that Eph children were becoming smarter and coming from families with more of a focus on elite education. All those trends are continuing. Within a few years, being a legacy will count for, essentially, nothing when you apply to Williams. Till then, the main advantages are: 1) The Admissions Office will give you a secret wink if you really have no chance, thus saving them (and you) the awkwardness of a formal rejection and 2) AR 1 legacies are always (?) admitted.

# Oh, to be a legacy…

I still remember the first time I met a legacy at Williams. It was first days, and I was talking with a few people about our various backgrounds, something I had interest in as one of Williams’s fairly rare southerners from a majority-minority public high school.

My new friend paused after describing her parents’ occupations, leaned forward, and then half-whispered, “I’m a legacy,” as if it was somehow something to be ashamed of, or a secret not revealed lightly. Until that point, I hadn’t really conceived of the notion that legacy was a burden, or something to hide, but we certainly don’t talk about it much here and now. In fact, the only time I’ve heard a legacy identified publicly was when the Octet brought one of its founding members on stage last semester, a man who spoke of the Octet’s founding and the grandfather of a current Williams senior.

Partly, it may have to do with stereotypes about what that means in terms of family wealth, but I think there is a greater factor involved. Following in one’s family footsteps is no small task. If I am walking the same halls and learning in the same classrooms as a parent, I likely have no choice but to compare myself to them and their accomplishments at Williams. Furthermore, legacies know that they have a minor leg-up in admissions – could that give rise to some fear that they don’t belong at Williams, and that they were only admitted because of who came before?

Yet at the same time, a legacy may have been raised with Williams decorating their home. They might have been to the College for a reunion or two, and could be, by the nature of their birth, a better Williams student (and/or a better alum).

As the old Williams, a place of almost exclusively white men, fades away into the yellowed pages of history, what will become of the legacy in this modern era?

Text in italics is edited in @ 8:36 PM

Morty answered a question about legacy admissions with all sorts of juicy details. He mentioned the distinction between direct legacies (one or two parents from Williams) and skip legacies (no parents but at least one grandparent). He noted that the former were more of a focus than the latter. He reported that 12%-15% of a typical Williams class was direct legacy. I think he said that the class of 2012 was 15%, but that went up to 17%-18% once you included Eph grandchildren. Morty mentioned that some people (other LAC presidents?) feel that you want to be wary of having “double digit” legacies, that you want to allow for new blood and not be too inbred. Morty thought this was nuts. He saw no reason to penalize an applicant just because her parents went to Williams. How does that make sense? If anything, he felt that such applicants were particularly desirable. They understood Williams, knew its strengths and weaknesses, and were probably making a very informed decision in coming here.

Morty noted that a decade or so ago [or perhaps when he arrived?], the average legacy was a 3.3 on the 1-9 scale of academic ranks while the average non-legacy was 2.3. Morty did not seem to be a huge fan of this gap, or of giving legacies such a preference. He then noted that the latest statistics show that legacy and non-legacy are now equivalent (both at 2.3). Morty confirmed, consistent with all the analysis I have done, that being a legacy is not a meaningful advantage in getting into Williams. Morty noted that the way that some people measure this — by comparing the general admissions rate (16%) with the legacy admission rate (40%?) — was misleading because legacy applicants are often told ahead of time that they have no chance. So, they don’t apply and/or withdraw their applications, thus artificially increasing the legacy acceptance rate. Non-legacies with no chance are not given this inside scoop. They just apply and get rejected.

Of course, being a legacy is still an advantage. Morty pointed out that the acceptance rate for AR 1’s was only about 30%. [I bet this was for all AR 1’s, not just US citizens, which almost all legacies are.] Legacies with AR 1s are “always” accepted. [I bet that AR 1 legacies are some of the best and happiest members of the Willams community, especially the subset that applies early decision. They could have gone elsewhere but chose Williams because they were more than ready to fall in love with it. In other words, even if you did not want to give legacies an advantage qua legacies, you should still admit the AR 1 legacies over other AR 1s because these sons and daughters of Ephs are much more likely to be happy at Williams and contribute to the happiness of others.]

# No Help for Legacies

From College Confidential:

I spend a lot of time around Williams as a local and an interested applicant. Here’s what I’ve heard from admission information sessions (i’ve attended three this summer…since they all provide somewhat different details)

It doesn’t to help be a legacy anymore. Last year…they only took legacies that were going to be admitted anyway. Part of the problem is that they have significantly more legacies applying than ever before. So unless your parents will be donating a building…

This is consistent with my analysis. Unsophisticated applicants sometimes take heart from the fact that the admission rate for legacies is much higher than that for non-legacies. This (true) statistic results, I think, from the Admissions Office giving a “heads up” to legacy applicants about their chances. Williams will tell a legacy with no hooks and 1300 SATs not to bother to apply. A non-legacy with the same statistics and, therefore, zero chance, will be allowed, even encouraged, to send in an application.

# Fairness

Andrew Triska has a serviceable op-ed in last week’s Record entitled “Fairness Beyond Financial Aid.” But, while his sentiments are reasonable, his actual knowledge about the facts seems lacking. Don’t kids today read EphBlog?
Just kidding! Details below.

# Silver Spoons

Must-read article in Inside Higher Ed on Daniel Golden’s new book, The Price of Admission: How America’s Ruling Class Buys Its Way into Elite Colleges — and Who Gets Left Outside the Gates.

That American higher education is not a pure meritocracy is, of course, hardly news. But Golden’s book has a level of detail about the degree to which he says some colleges favor the privileged that will embarrass many an admissions officer. Golden names names of students — and includes details about their academic records before college and once there that raise questions about the admissions decisions being made. For good measure, he attacks Title IX (saying that the women’s teams colleges create favor wealthy, white applicants), preferences for faculty children (ditto, although substitute middle class for wealthy), and accuses colleges of making Asian applicants the “new Jews” and holding them to much higher standards than other students.

Even before its official release, The Price of Admission is causing considerable fear among the admissions officers of elite colleges. If you want to see an admissions dean really happy, tell her that you can’t find her institution in the index.

Bring it on! I have commented on some of Golden’s Pulitzer Prize-winning articles in the past. There is much here that is worthy of discussion, but I want to focus on Williams-specific references first.

While those whose institutions or causes are attacked in the book are among those taking a critical look, so are those in elite higher education whose colleges come out relatively unscathed. Williams College, for example, doesn’t rate an index mention, but does give preference to alumni children and selected others.

Not in the index! Golden should have checked EphBlog . . .

Richard L. Nesbitt, director of admissions at Williams, said that roughly 11 to 14 percent of each class is made up of alumni children, and that that ratio has been unchanged for about 30 years.

If Williams isn’t in the book, then why is the author talking to Nesbitt? Probably because an editor of Inside Higher Ed, Doug Lederman, was the author of the Alumni Review article on admissions. Still that article does not quote legacy percentages. That’s from EphBlog, perhaps. Note that Jim Kolesar told us it was 11% to 15% (not 14%). That may not seem like a big change, but it decreases the range of results by 25%, and centers it right around 12.5%, what I believe the unofficial “target” to be.

The admit rate of alumni children is significantly higher than that for all applicants, Nesbitt said, although he declined to reveal specifics, saying that Williams does not release that information for any subgroup of applicants. But Nesbitt cautioned against thinking that the higher rate means lower standards for that group. Applicants are generally better prepared if they have well educated parents, who are more likely to have the resources to help their children’s education. Beyond that, he said, one benefit Williams does give to alumni children is to offer more information in interviews, so that students who are unlikely to be admitted get that message early and are less likely to apply.

Correct. And who can complain about that? Although you can bet that non-legacy candidates with no hook and 1250 SATs wish that someone would give them a similar heads-up. We have explored related issues in the past, see here and here. Indeed, I believe that there is more published information about legacy admissions at Williams than for any other similar school. Score one for EphBlog!

Nesbitt said he had difficulty with Golden’s thesis that admissions policies that give any preference to alumni children are limiting the overall socioeconomic diversity of elite colleges, and especially the enrollment of Asian students. Colleges like Williams are more diverse every year, he said, enrolling record numbers of Asian students while changing financial aid policies to attract more low-income students of all races and ethnicities.

Agreed. If you didn’t let in alumni kids (and you wanted to maintain similar credentials in your class in terms of SAT scores and high school grades), you would just let in a bunch of kids who looked just like your alumni kids. The legacy preference is, now, just not that large because there are so many well-qualified legacy applicants. (This was probably less true 20 years ago.)

Now, you could have a different policy in which you went for non-trivially worse academic ranks in order to get, say, more social economic diversity. (See endless discussion here and associated links.) But this is, conceptually, a separate issue from legacy. Once you decide that you want to have 50% (or whatever) of your class be academic rank 1s and 2s, you are going to get a lot of non-poor, non-URM kids. Why not favor the girls whose mothers went to Williams over the girls whose mothers went to Amherst? They are largely interchangeable otherwise.)

The whole “changing financial aid policies” is mostly a crock. (Previous rant here, with lots of links for our new First Year readers.) More commentary later. Try to contain your excitement.

# No Legacy Quota

Thanks to Jim Kolesar ’74 for providing these comments on whether or not Williams has a (minimum) quota of about 12.5% on legacy admissions. Previous discussion here and here. Jim explains:

If a quota is a figure you hit precisely no matter what and a target is a figure you take all reasonable measures to try to hit as a minimum, legacies are neither. We give them extra consideration in the admission process, though their average academic rating is virtually indistinguishable from that of the class as a whole. After each admission cycle, we review the percentage and the academic preparation of the legacy group to see if we’re comfortable with where those measures fall. Virtually always we are. How those considerations play into the admission of individual legacies is another part of what makes the process an art rather than a science.

Also, the number of legacies in the class isn’t as constant as you imply. The percent of them in the class bounces between around 11% and 15%. That means the number can vary by roughly 50%.

1) Thanks to Jim for taking the time to explain this to that subset of the alumni community which is interested.

2) It would still be nice to see the actual data. There is a great senior thesis to be written about the history of legacy admissions at Williams.

3) “Bounce” is not the word that I would use to describe a statistic that moves from 11% to 15% and back. Moreover, if those are the minimum and maximum figures (rounded?) for the last 20 years, then “hold quite steady” is a more accurate desciption. Is there any other aspect of the admissions process at Williams that has been this stable for decades? None comes to mind. Also, I know no one who would describe changes in this range as a movement of “50%.”

4) The main reason to be suspicious — or pleased, depending on your point of view — is that the number/quality of legacy applicants relative to the pool of spots has changed so much in the last two decades. Back in the 80’s, all the legacy applicants had fathers who went to Williams in the era of 250 students per class. Also, those fathers [Sorry, Dad!] were not, on average, in the same academic league as later students. (Williams in the 1950’s was the Bates of its era, if that.) Since test scores and grades are correlated across generations — either because of nature or nurture — the children of 1950s graduates were not, I’ll wager, as nearly accomplished academically as the children of 1970s graduates. Morever, there were twice as many students in the 70s.

So, the number and quality of legacy applicants (assuming more or less stable fertility among Ephs over time, another great thesis topic) is much higher now than it was 20 years ago. And yet the College still enrolls about 60 each year. Why?

5) The obvious conclusion is that there is a “goal” if not a quota. In other words, Jim, as always, is telling us the truth. There is no exact “quota.” But the College has always looked favorably on the children of its alumni. The powers-that-be have always been “comfortable” with a class that is 12% or so legacies. Everything seems to work out. No one complains. 12% is enough to satisfy the Go-Williams faction. 12% does not seem to allow in unqualified students.

6) But, behind the scenes, the competition for those legacy slots has gone up, I bet. One of the most interesting parts of the Admissions Office tell-all (pdf) was this line:

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.

Recall our discussion of academic ratings. My inference is that the only attributes which matter for 3’s and below are URM, tip or wealth. Being a legacy, like being a flutist, is nice but does not matter for admissions for 3’s and below because there are “enough” legacies and musicians among the 1’s and 2’s. This is consistent with our previous analysis of legacy SAT scores. There may be a few AR 3’s who get in just because of their legacy status, but there are not many.

Conclusion: All is for the best in this, the best of all possible worlds. With 12% legacies, there is no grounds for anyone to complain that Williams does not let in enough alumni children. What do they want? 20% legacies? 50%? But, with the so many highly talented legacy applicants — a function of the increasing number and quality in the legacy pool — Williams gets enough legacies from the AR 1’s and 2’s that it does not need to go deeper in the pool.

Prediction: This trend will continue. Recall that “legacy” is defined as being the child or grandchild of an alum. The number of such legacies will be increasing as the larger and higher quality Williams classes of the 1970’s and later enter the grandparent years. Given the continuing test score stratification of American (and global?) society, these Eph families will be richer and more accomplished then ever, able to provide their children and grandchildren with every advantage imaginable. And, even better for Williams, these families will be filled with children who, like me, choose Williams because of those family connections.

There will soon come a day when Williams is able to reach its goal of 12% legacies without actually giving legacy applicants any advantage whatsoever in the process. I wonder how far away that day is.

There is much to blog about in the latest issue of the Record, but consider this line from an article last spring about the class of 2009.

Twelve to thirteen percent of the class are legacy students, a number which Nesbitt said has been constant at the College for 20 years.

This would suggest 65 to 75 legacies. Consistent with what we were later able to confirm. But the interesting part is that this number has been consistent for 20 years.

Hmmm. Thoughts:

1) I believe that Nesbitt is telling the truth, but it would be fun to look at the exact numbers. Are they available?

2) Is there a (maximum) quota here, as there is in International admissions? I doubt it. Why would the College turn down excellent legacy candidates?

3) Is there a (minimum) quota? That would be shocking. My first thought is to dismiss such speculation, but, then again, I was too stupid to realize that there was an International quota, so I might be misinformed again.

4) Why would the number stay constant for 20 years if there were not a quota, or at least a guideline?

5) Given the dramatic rise in the number of International students and in US students of color (as well as continued focus on athletics), the odds of getting into Williams if you are not in one of these categores has plummetted. I am too lazy to do the math, and the are certainly legacies that are athletes and/or of color, but there are only so many spots to go around.

6) It is interesting to note that the International quota is set at 6%, just about exactly one half of the legacy admission percentage. Is that a coincidence? Someone had to pick the 6% number and he didn’t pick it out of thin air . . .

7) Why doesn’t the Record write a story on this topic? I think that it is a scandal. Surely, I am not the only one. I had lunch with a recent graduate and he was shocked to learn about it.

I think that my big 2006 Williams project will be to form a alumni/student/faculty group whose purpose will be to urge the College to increase the percentage of International students by not penalizing them so much in the admissions process. Eps Against International Quotas, perhaps.

# Legacy Math

Time for some more legacy math! Although my own daughters are still several years away from their college applications, it is never too early to start thinking about what their chances might be at Williams. Recall our earlier discussion of the fact that there were 94 legacies admitted into the class of 2009 with an average SAT score of 1446.

Thanks to Jim Kolesar, we now know that 68 of these legacies ended up enrolling in the class. Comments:

1) I wonder about the 26 students who didn’t come to Williams. Are these mostly very strong students who choose places like Harvard/Yale/Princeton/Stanford instead? If so, this would suggest that the average SAT score of the students who came to Williams was lower than the 1446 average of the 94 applicants admitted. It is also possible that they are weaker students who realized that Williams might not be the best place for them, but I doubt that.

2) What is the extreme lower-bound for the enrolled legacies? Assume that all 26 of the students who turned down Williams had perfect 1600s. This implies that the 68 enrolled legacies would average around 1390 since (94 * 1446 – 26 * 1600) / 68 = 1387.

3) Can we reject the null hypothesis that enrolled legacies have similar SAT scores to the overall Williams population? No. Note that the SAT average for the class of 2009 is 1425. Solving the appropriate formula for X:

(94 * 1446 – 26 * X) / 68 = 1425

we can see that X = 1500. In other words, if the average SAT score of the accepted legacies who went elsewhere is 1500, then the average for the enrolled legacies would be the same as the average for the entire class. Is 1500 a plausible estimate for those 26 non-enrollees? Sure.

4) Does this imply that legacy status does not positively impact one’s admissions chances? No. Again, the key issue is that 1425 is the average for the entire class. The average for students from wealthy families with college-educated parents is going to be much higher than 1425 because, for good or for ill, the College gives preferences to applicants, independent of race, from poorer, less educated families. Now, the magnitude of those preferences is not large, but it is not zero.

5) Key unknowns are the characteristics of applicants and admitted students who come from families “like” those of the legacy applicants. Although race, wealth and athletic talent are complicating factors, I doubt that the average SAT scores of similar enrolled students in the class of 2009 is much higher than 1475.

All in all, I will stick with my back-of-the-envelope guess that being a legacy, all else equal, is worth about 50 points in combined average SAT score. This is a minor advantage, much lower than the 100 points that tip-level athleticism counts for and 150 points that accrue to URMs.

In other words, if you’re an Eph and your daughter scores 1300 or below (and she is not an athlete, heiress or deeper shade of purple), you better start looking at Colby.

# 1446

Although College Confidential discussions about Williams are not always reliable, they are often informative. Who knew that the combined average SAT among admitted legacy applicants was 1,446 for the class of 2009? Not me. Comments:

1) The provided link (page 36) provides one other interesting factoid: there were 94 admitted legacies into the class of 2009.

2) An obvious interpretation of the 1,446 figure is that legacy status does not matter much for admissions since 1,446 is significantly higher than the average SAT for the class as a whole. But this is, of course, misleading since we don’t know the average of the admitted non-legacy applicants. We only know the average for the admitted and enrolling students, those who decide to attend. I would guess that the average for all admitted students is at least 1,446 since so many of the admitted applicants are also accepted by Harvard/Yale/Princeton/Stanford and, inexplicably, turn down Williams.

3) Of course, what we would really like to know is how many legacies decided to attend Williams and what the average SAT for this subgroup is. If this is a number like 1,300 then we might conclude that legacy status is similar in the “boost” that it gives an applicant to tips-level athleticism.

4) But, unless the numbers are extremely skewed, there is just no way that this can be the case. After all, if 50 of these students had 1,300 SATs (and those were the 50 that decided to come to Williams), then the other 50 would all have to have 1,600. That seems implausible. The most reasonable conclusion is that legacy status does not give one much of a boost in admissions. Back of the envelope, I would guess 50 points or less. Indeed, these numbers wouldn’t contradict a null hypothesis of zero boost.

5) I am sweeping a lot of complications under the rug. Legacy students are different in all sorts of ways from non-legacy students, so a simple SAT comparison is not easy to make. What we really want is to compare the academic credentials of legacy admits versus matched non-legacy admits, those from similarly wealthly, educated parents.

6) It might also be that the legacy numbers are biased on the high side because so many very highly qualified legacies choose Williams. I would not be surprised if a disproportionate number of students who choose Williams over H/Y/P/S are legacy applicants.

7) Those eager to believe that legacy status matters a lot can take comfort in this quote from a great Malcolm Gladwell article on admissions.

In the 1985-92 period, for instance, Harvard admitted children of alumni at a rate more than twice that of non-athlete, non-legacy applicants, despite the fact that, on virtually every one of the school’s magical ratings scales, legacies significantly lagged behind their peers. Karabel calls the practice “unmeritocratic at best and profoundly corrupt at worst,” but rewarding customer loyalty is what luxury brands do.

A priori I would have thought that similar preferences for legacies exist at all elite schools. Why would Harvard care more than Williams?

Conclusion: I think that this data point provides further evidence for my claim that legacy status, while certainly a boost to ones chances, has nowhere near the power of significant athletic talent (much less URMness) at increasing ones chances at Williams.

Currently browsing posts filed under "Legacy Admissions"