The Inequality Machine

From Michael\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\'s Personal Wiki
Jump to navigation Jump to search

CHAPTER II: Getting In

Among economists, though, this question has long been a matter of debate. Proponents of a more laid-back approach to college admissions base their argument largely on a 2002 paper, updated and extended in 2011, by Stacy Dale, a researcher, and the late economist Alan Krueger. Dale and Krueger analyzed records from twenty-seven selective colleges and universities, as well as a national database, and concluded that for well-off white and Asian students, there was no benefit at all, in terms of future earnings, to attending a more selective college. Yes, a graduate of Princeton will earn more than a graduate of Penn State, on average--but, the researchers said, that has everything to do with the talents and ambitions of the students whom Princeton admits.


Hoxby has focused her research for more than a decade on the college-admissions process. But for many years, she avoided the debate over the relative value of attending a selective college. She was always somewhat skeptical of the methods that Dale and Krueger used, but she considered the available data too muddy either to confirm or rebut their conclusion. More recently, though, that muddy picture clarified for Hoxby, when she found a way to leverage many years of tax records--using the same IRS database that Chetty draws on for his research--to calculate "value-added" estimates for thousands of American colleges and universities.

Her method, briefly, involves combining the IRS data showing which college each student attends and their earnings a decade after graduating from high school with information from the College Board on where students apply, plus data from a national clearninghouse that reveals which college each student actually winds up attending. By merging these data points, and comparing them for millions of students who graduated from high school between 1999 and 2003, Hoxby was able to estimate, quite precisely, the causal effect of the college a student attends on his or her average lifetime earnings.

Her conclusion? Dale and Krueger were simply wrong. High-prestige colleges do pay off for the students who attend them, and in fact they pay off in a big way.


CollegePoint counted an advising relationship as a success only if the student ended up enrolling in an Aspen 270 college. But the advisers I spoke to found the Aspen list to be rigid and arbitrary; in the real world of high school students, they believed, sometimes a school that wasn't on the list was a better choice than a school that was. For instance, the only historically black college on the Aspen 270 list was Spelman, a women's college.

CHAPTER III: Fixing the Test

It was true that studying for twenty hours on Official SAT Practice was associated with an average score gain of 115 points. But what the College Board didn't tell the public was that very few students had studied that many hours. The College Board sent invitations to 1,075,000 students in the high school class of 2017, encouraging them to link their College Board account to a free account on Khan Academy. Fewer than a quarter of those students -- 250,000 -- actually linked their accounts. Most of those 250,000 students studied for just a few hours, or not at all. Only 8,000 students across the country did at least twenty hours of practice and achieved that average gain of 115 points...

This is where dealing with the College Board grew especially frustrating--and my faith in the organization's good intentions began to wear thin. If only a tiny percentage of students studied for twenty hours, why lead your public presentation with that group's results? When I spoke later than summer with Aaron Lemon-Strauss, the College Board executive who led the collaboration with Khan Academy, he explained that the decision had been driven by a desire for positive media coverage. There was an original draft of the press release, Lemon-Strauss explained, that featured a chart comparing the effect of zero hours of Khan study (an average gain of sixty points over students' PSAT scores), with the effect of five hours (an average gain of eighty-four points) and ten hours (an average gain of ninety-seven points). That chart would have indicated much smaller gains than the numbers the College Board eventually released, but they would have been more representative of what typical students actually achieved using Official SAT Practice.

Lemon-Strauss told me he liked that version, but he explained frankly that it was rejected by the College Board's communications team.


And because the effect of an hour of study on a student's score gain was consistent among all these groups, the extra hours of study that the more privileged students took advantage of meant that their scores increased more, as well. On the whole, according to the data from the hackathon, Official SAT Practice on Khan Academy had increased the existing gaps between students in groups that tended to score higher on the SAT and students in groups that tended to score lower.

CHAPTER IV: Fitting In

[Tony] Jack spent two years doing intensive fieldwork at an elite American college, conducting in-depth interviews with low-income students, and those interviews became the basis for his PhD dissertation and also for his first book, The Privileged Poor, which was published in the spring of 2019. Following the traditional practice of sociologists, he uses pseudonyms in the dissertation and the book for the students he interviewed, and he also uses a pseudonym -- Renowned University -- for the institution where he did his fieldwork. Since I'm not a sociologist, I'm not obliged to follow the conventions of sociology, so I can tell you that I'm pretty certain the university where he did his fieldwork was Harvard, since that's where he was working and studying during the years he was conducting his interviews.


When KiKi first arrived, she was assigned to a classroom full of students who looked a lot like she did. That didn't seem out of the ordinary to KiKi -- she had always been in classrooms full of kids who looked like her -- until she noticed that there was a fourth-grade classroom right next door where the kids looked different. They were all white. Their classroom was nicer, too -- better equipped and better organized. She asked one of her new classmates what was going on, and she was told that at Brook Park, students were grouped by ability. The classroom next door was a "stretch" classroom. Those kids were the smart kids.

If it hadn't been for Ms. Denise the year before, KiKi might not have had the thought that came into her mind next, or at least she might not have expressed it out loud: "But I'm on the of the smart kids." But because Ms. Denise had helped convince her to be confident in her intelligence, KiKi did say it, and she kept saying it: _I belong next door._

She talked to her teacher, and the she talked to the principal, and then she persuaded her mom to come in and raise a fuss on her behalf. Though they were resistant at first, the school's administrators finally agreed to let her take a placement test. KiKi aced it, and the next day she was moved to the stretch classroom.


At Myers Park, the nameless fury that KiKi had felt as a fourth-grade student was focused and honed into an intensely felt racial awareness. kiKi was still reading the biggest, fattest, white-establishment novels she could carry: _The Satanic Verses, Gravity's Rainbow, Sense and Sensibility._ But she also started reading Angela Davis and Eldridge Cleaver. She studied the Blank Panthers and MOVE, the radical black group whose headquarters were firebombed by the Philadelphia police in 1985. She started a pro-black group at Myers Park called PRIDE, and she organized a race symposium to discuss the racial disparities in the makeup of the school's standard and honors and IB classes.

But despite KiKi's full embrace of her black identity, she still felt stuck int he middle. The high-octane education she was receiving meant she spent her days physically separated from the black kids in the standard classrooms. "I don't have a connection to the black student body," she told me when I visited Myers Park during her senior year. "Which is significant to me because that's who I'm fighting for. But to be frank, I don't have much of a connection with anyone at my school."


Over the summer, KiKi had been invited to take part in Princeton's Freshman Scholars Institute, or FSI, a free seven-week orientation program for first-generation and low-income students. She had fun, and she met some people she liked, but there were only a few black kids in the group, and in general the students she met at FSI seemed more working-class or lower-middle-class than gennuinely poor. Their experiences growing up were nothing like KiKi's. "My definition of low income is a lot different than Princeton's definition of low income," KiKi explained with a laugh when we spoke on the phone that summer. When the rest of the freshman class arrived in September, the income divide on campus only grew sharper. Everyone around KiKi seemed downright wealthy.

So what was the deal? Demographic revolution, or all-pervading affluence? What was the Washington Post seeing on the Princeton campus that KiKi was not -- and vice versa?

The answer to that question beings with an understanding of the way Princeton and other institutions calculate the economic diversity of their campuses. The statistic that the post cited is the statistic that pretty much everyone uses: the percentage of the student body, or of the freshman class, that is eligible for a federal Pell grant. "Pell percentage," as it is sometimes called, is a useful marker in many ways, and it is a figure that the efderal government collects and reports for each college. But it is also something of a moving target... some families making more than $80,000 receive Pell grants, and some earning less than $30,000 do not.


In the same way that admissions officials at those college might find it easier to admit middle-class Pell grant recipients rather than poor ones, or the black children of immigrants instead of the black descendants of American slaves, Jack believes that they favor students from the Privileged Poor over the Doubly Disadvantaged as a way to "hedge their bets on diversity." Graduates of Choate or Andover are a known quantity in Ivy League admissions offices. Graduates of the large, mostly low-performing public high schools where the vast majority of low-income black American students spend their days are not...The second big conclusion Tony Jack reached in his research was that Doubly Disadvantaged students had a much rockier experience once they got to college than Privileged Poor students did. And the most stressful part of the transition wasn't the academic work (though that was often stressful as well). It was their daily interactions with their fellow students.


The other students were prepared and informed as well. But when they contributed their thoughts, they often gave the impression of doing so ironically, of being somehow above it all. They cracked knowing jokes and leaned back in their chairs with a studied nonchalance. The alpha dude took great pleasure, at one point, in comparing the character of Dionysus in Euripides's _Bacchae_ to Walter White, the meth-dealing antihero of _Breaking Bad_. The agreed-upon posture in precept, for everyone but KiKi, was to be ostentatiously laid back -- amused, but not aroused. Pop culture, high culture: same trope, different day.

In the game KiKi was playing, you scored points by being the smartest and the most well-read. In the game everyone else around the precept table was playing, you scored points by being the cleverest and most at ease.


If you are looking for reassurance that the American meritocracy is fair and democratic and open to all, Pedigree is not the book for you. Backed up by her meticulous research, Rivera managed, in 350 pages, to undercut and overturn many of the stories we like to tell ourselves about who succeeds in the United States and why. It turns out it really _does_ matter where you go to college -- at least if you want to work in certain high-paying professions. Lacrosse bros really _do_ run the world. It really _is_ who you know, not what you know. An investment banker really _did_ tell Rivera that it was "difficult" for women to be hired by his bank if they weren't "pretty." ("I mean, you don't have to be _hot," he clarified. "But you do need to be reasonably attractive.")

Rivera's central finding was that top law firms and management consulting companies and investment banks -- known collectively as "elite professional service" or EPS firms -- based their entry-level hiring decisions not on what candidates had achieved in college, but on who they were when they arrived at college years earlier...

If you have the misfortune to attend a college below the second tier, your application is more or less doomed. "I'm just being really honest, it pretty much goes into a black hole," a female recruiter explained to Rivera. "Unfortunately, it's just not a great situation. There's not an easy way to get into the firm if you're not at a target school."

CHAPTER V: Letting In

Over the last decade, two distinct conversations about college admissions and class have been taking place in the United States. The first one has been conducted in public, at College Board summits and White House conferences and meetings of philanthropists and nonprofit leaders. It is the conversation that informs some of the interventions I wrote about in chapters 2 and 3, from Caroline Hoxby's packets to Khan Academy's SAT prep to CollegePoint's video counseling. The premise of this conversation is that the problem of inequity in higher education is mostly a demand-side problem: Poor kids are making regrettable miscalculations as they apply to college. They are "betraying themselves." Selective colleges would love to admit more low-income students--if only they could find enough highly qualified ones who could meet their academic standards.

The second conversation is the one going on in the exhibit hall at the NACAC conference, among the enrollment professionals who labor behind the scenes in the admissions-industrial complex. This conversation, held more often in private, starts from the premise that the biggest barriers to opportunity for low-income students in higher education are on the supply side-- in the universities themselves and specifically in the admissions office. Enrollment managers know there is no great shortage of deserving low-income students applying to good colleges; they know this because they regularly reject them -- not because they don't want to admit these students, but because they can't afford to. Admissions professionals are well aware that they spend much of their time and energy looking not for more high-achieving low-income students but for more low-achieving high-income students -- more CFO Specials. That's how they make their budget. That's how their institutions stay afloat.


Among the millions of students who take the SAT each year, about two-thirds receive scores that are in line with their high school grades. For those students, the SAT doesn't really matter at all -- their test scores send exactly the same signal to college-admissions offices that their high school grades do. The students for whom test scores make a difference in admissions are the two groups who have what testing researchers call "discrepant" scores -- meaning either that their SAT score is much higher than their high school grades would predict, or their high school grades are much higher than their SAT score would predict. Those two categories each make up about a sixth of each national cohort of high school seniors.

So if we're trying to determine the overall effect of standardized admission tests -- and of test-optional policies -- on American students, one big question is this: Who are the students in each of these two groups, the inflated-SAT students and the deflated-SAT students? How do they compare with each other -- apart from the obvious fact that one group does better on standardized test and other does better in the classroom?

Here's what the College Board found: The students with the inflated SAT scores were more likely to be white or Asian that the students in the deflated-SAT group, and they were much more likely to be male. Their families were better off, too. Compared to the students with the deflated SAT scores, the inflated-SAT students were more than twice as likely to have parents who earned more than $100,000 a year and more than twice as likely to have parents with graduate degrees. These are the students -- the only students -- who get a big boost in admissions from the SAT.


In 2017 the College Board launched a public relations campaign to counteract and critique the movement toward test-optional admissions and to call into question the credibility of students' high school grades. The strategy the company's leaders chose was an audacious one. They decided to make the case that high school grades gave an unfair advantage to privilege students, while the SAT benefited less privileged students -- despite the fact that their research had for years demonstrated the opposite.

The cornerstone of the College Board's strategy was a new academic paper by two in-house researchers, Michael Hurwitz and Jason Lee. The paper, titled "Grade Inflation and the Role of Standardized Testing," begins by observing a genuine phenomenon which is that the high school grades that students report to the College Board when they take the SAT have been gradually increasing over the last couple of decades, rising from an average GPA of 3.27 in 1998 to an average GPA of 3.38 in 2016. Hurwitz and Lee analyzed this change in high school grades by the type of school that students attended, and they showed that the biggest increase in GPA during those years took place among students at private schools. They used that fact to argue that grade inflation was benefiting white and Asian students and wealthier students, at the expense of everyone else.

The problem with this argument is that the paper itself includes data that shows that grade inflation has actually been quite consistent across racial and socioeconomic groups. What has not been consistent, the data tin the paper shows, is changes in _SAT scores_ among these different groups.

Consider the data for black students and Asian students -- the racial groups that typically score the lowest and the highest, typically, on the SAT. Between 2001 and 2016 high school GPAs for Asian students did increase - they went up by 0.12 grade points. But GPAs went up for black students too, by almost the same amount -- 0.11 grade points. But while their GPAs were moving in parallel, the SAT scores that black and Asian students were receiving diverged quite sharply over the same period. Asian students' SAT scores went up by fifty-five points, on average, between 2001 and 2016, while black students' scores went down by sixteen points.


It is part of the paradox of the project that Angel Perez has undertaken at Trinity -- and the broader paradox of elite college admissions in the age of enrollment management. With each new freshman class, Perez is demonstrating that test-optional admissions can help bring in highly qualified students from diverse and challenging backgrounds, and that they can succeed at Trinity, academically, despite test scores that would under different rules have kept them far from any selective college. But the financial pressure Perez is working under makes it hard, or maybe impossible, for him to build u pa critical mass of those students on campus. And meanwhile, the dominant culture of the institution Perez is trying to transform is strong and deep-rooted and often resistant to change. Trinity these days sometimes feels like a host trying to reject the transplanted organ it needs to survive.

CHAPTER VI: Staying In

Laude needed to figure out exactly which students needed help. It was one thing to identify the students who might struggle in a five-hundred-person chemistry class. Laude was now responsible for the entire freshman class –more than seven thousand students. He conscripted a team of data scientists in the university’s office of institutional research who specialized in predictive analytics. By deconstructing the records of tens of thousands of recent UT students, they were able to develop a statistical model that combined fourteen separate variables, from family income to SAT score to high school class rank to parents’ educational background into a single algorithm that could reliably predict an incoming student’s likelihood of on-time graduation. They called this new tool the Four-Year Graduation Rate Dashboard.

One of the first students to arrive in Buis’ office on a mild morning in March, just after spring break, was Nicolas, a thin, soft-spoken, sharply dressed freshman with a sweep of sandy blond hair. When Buis opened his file on her screen, a big rec circle popped up that read “Predicted risk level: High!”

Early in 2012, a young researcher named David Yeager was hired by the psychology department at the University of Texas as an assistant professor. Yeager had just earned his PhD from Stanford, where he had worked closely with Walton and Cohen, and hw as steeped in this new body of research and this new approach to intervention. Up until that point, the experiments Yeager and Walton and Cohen had tried were all relatively small in scale. But soon after Yeager arrived at UT, he was approached by administrators who asked him if he might beable to create an intervention that could be given to all seven-thousand-plus members of the next incoming freshman class...

For students whose parents had gone to college, the intervention didn't make any difference -- perhaps because they didn't have as much belonging anxiety in the first place. But for first-generation and other disadvantaged students, there was a small but distinct impact of encountering the "it gets better" messages...


The alternative method they were using at UCA was called the corequisite model, which aims to let students get the help they need catching up while at the same time earning college credit in more challenging and engaging courses. The class I sat in on, Writing 1310, was not remedial; it was the standard first-year writing class that every UCA freshman is required to take. The professor, Becky Bogolavsky, taught a class of twenty students that met three times a week. (On the day I was there, the class was discussing strategies for revising essays.) Though she never mentioned this fact in class, ten of Bogoslavsky's twenty students had scores 18 or below on the writing section of the ACT, meaning they were required by UCA's rules to complete the university's remedial writing course, while the other ten had scores 19 or above. On Mondays, Wednesdays and Fridays, Bogoslavsky taught both the high-scoring and low-scoring students together, in the same class, without distinguishing between them at all.

And then on Tuesday and Thursday mornings, Bogoslavsky would meet in a separate class called Transitional Writing with the ten students from her Writing 1310 course who had lower ACT scores.


One of the central questions that Taslim and Orry and many other high school seniors and college students have been asking in the last few years is this: *is college worth it?* When you pose this question to economists, they usually give you either a quick answer or a slightly more deliberate one. The quick answer is: Yes. What economists call the college wage premium -- the amount by which wages for college-educated adults exceed, on average, wages earned by those who don't have a degree -- is about as high in the United States as it has ever been. Back in the 1970s, that gap was relatively modest; college graduates earned just 40 percent more, on average, than non-grads. Now they earn 84 percent more.

The college wage premium is a basic function of supply and demand: when there are more college graduates than the market can bear, the wage premium goes down...Despite the dominant political rhetoric suggesting that the United States has too many college graduates, the clear signal from the labor marketplace is that in fact we have too few.

The somewhat more nuanced answer economists give to the is-college-worth-it question is: it depends. It depends on who you are and where you go and what you take and how you do when you're there and how much debt you amass along the way...According to research by the economists Tim Bartik and Brad Hershbein, if you grow up in a lower-income family, your BA will add an average of $335,000 to your lifetime earnings. But if you grow up in a higher-income family, your BA will add an average of $901,000 to your lifetime earnings.

CHAPTER VII: Getting an A

One of the great oddities in the history of mathematics is the fact that calculus was invented more or less simultaneously in the late seventeenth century by two men in two countries, working independently: Isaac Newton, the English scientist, and Gottfried Wilhelm Leibniz, the German philosopher. The most pressing scientific questions of that era all had to do with motion and change -- how the planets moved in the sky; how ships navigated across the ocean; how fluids and gases and heat could e manipulated to produce power. In each field, the central issue confounding scientists was the same: When you observe a body in motion at a particular point, what can you determine about where it has been and where it is going? It was a question that the existing tools of mathematics, in geometry and algebra and trigonometry were unable to answer.

Newton and Leibniz both came to understand that the way to solve this problem was to consider the process of continuous change as the accretion of an infinite number of infinitesimally small moments. Their system of calculus was the scalpel that allowed scientists and mathematicians to carve time and space into those minute fractions and then to reassemble the parts into a new and more comprehensive whole. Calculus became the essential mathematical tool of the industrial revolution, its methods enabling engineers to harness steam power and astronomers to chart the orbits of the planets and economists to codify the laws of supply and demand that underlay mass production and global trade.

As pretty much every AP Calculus student or teacher today will tell you, the recent national enthusiasm for AP Calculus is not really about calculus. It's about college. Taking AP Calculus has become a general-purpose symbol of academic achievement, a signal to college-admissions officers of "eliteness," the way taking Latin was seen a generation or two ago. In one recent survey, 80 percent of college students who had completed AP Calculus said they took it because they thought it would look good on their college applications. And they were right: highly selective colleges are much more likely to admit students who have taken AP Calculus, even if those students have no plans to pursue math or science or engineering in college.

One particularly striking example: in 2017, 93 percent of freshman admitted by Harvard University reported that they had taken calculus in high school...Whatever your interests, whatever you plan to study in college, if you want to get into a school like Harvard, you better first take AP Calculus.