How Harvard and Yale cook the books -- Read at your own peril!

Parent Cafe
21

In the above paragraph, you have supported the anti-standardized testing army’s argument. By categorizing high schools in the same way that American society is caste (“elite” versus “low performing”), you argue that the tie-breaker is a high standardized test score (but only if that lifts someone from a “low performing” school). The elites will still have their private prep schools. The public high schools can never be equal because of resources. So let’s create another hurdle for those less fortunate to overcome (the SAT or ACT).

All 4.0s are not created equal - we all know that but isn’t that where rigor of the student’s course and high school profiles come into play?

As opposed to making all high schools better and/or telling students to be happy at lower level colleges, how about this? Let’s outlaw all private prep schools, private SAT preparation, tutoring, etc and reduce all high schools to the lowest performing level. Then we can see who really “has it” - Survival of the fittest.

22

@YZamyatin‌, you hit the nail on the head. Based on the facile generalizations and the treatment of data in this article I have no interest in reading the book. What “true facts” could one learn from it? Nonetheless, it may have value if it exposes a broad readership to some of the flaws of the testing industry and mentality – flaws that many of us who frequent this discussion board are well aware of.

I recall being dumfounded when I first encountered the SAT in my senior year of high school. My Mom, daughter of immigrant parents, was salutatorian of her high school class, but never went to college. My Dad, a son of immigrant parents, earned a college degree during the Depression. But they weren’t clued into the testing system and didn’t encourage me and my sibs to prepare for it. “National Merit” exam? I never heard of it until my senior year of high school – too late. Prepping for standardized tests outside of school? Never heard of it then. But my Mom and I used to enjoy the “It Pays to Increase Your Word Power” section of the Reader’s Digest. And she recommended literature for me to read during summers when I was in high school. My father was mathy, an engineer (his father was an auto mechanic) but really a polymath, who had devoured books when he was a youngster. So my parents were into reading, learning, not test-taking. That was important background for me and my 4 sibs.

Now, I had good grades in high school (5th in my graduating class of 577), and my test scores were high-mediocre but high enough on my SAT and achievement tests to allow me admission to a fine college, which later led me to a doctoral degree and a career of college teaching and research. However, nothing in college prepared me especially for standardized multiple-choice tests. The LSAT and GRE were the only multiple-choice tests that I ever took in college! My first encounter with these exams was when I opened the test booklets on test day. I did well enough on those tests to get into some of the finest graduate programs in the country, but not because of prep for the exams (unless one counts my exposure to the SAT’s 4 years earlier).

I think a college admissions regime that is free of standardized admissions tests would be just fine. I applaud colleges that have the courage to take a holistic approach to admissions without relying on standardized tests. It’s got to be harder for those admissions committees to evaluate applicants, especially those who are not applying to special programs such as art and music. But it should be harder than it is now.

23

I alluded to this somewhat facetiously above, but what would really happen if Harvard got rid of the SAT and concentrated on other factors? Is there some reason to believe that lower income people exhibit more service than richer people? I note also that CC is filled with people complaining that Harvard doesn’t value the SAT enough.

24

http://www.nytimes.com/1993/06/05/opinion/the-lani-guinier-mess.html

25

Impossible, since her two statements follow like night and day.

No, what she is saying, without a shred of proof, is that 'testocracy" results in students not interested so much in service.

quote

[/quote]

ONE, just one supports your pov, jhs? Come on, you have been much better than that. :smile:

btw: I would submit that it is the “testocracy” that results in more numbers-oriented law students. (The LSAT requires critical thinking skills for unhooked to get into top schools, and math folks get critical thinking skills, at least those measured on the LSAT.)

26

Menefrega Don’t get your point at all here whether you are being sarcastic or real. The SAT may be a hurdle, but not clear how getting rid of it will lead to more admissions of kids from poor high schools.

There is a vast different between many top public suburban high schools and inner city schools, not just prep schools. Many top public high schools send lots of kids to Ivys each year. The SAT is not a tie breaker, but allows some leveling of comparison among high schools.

JHS: how would Harvard do that? Don’t many of the less fortunate kids that go to Harvard also choose to pursue Wall St. or other lucrative careers? She doesn’t seem to make that distinction. Nor offer a way for Harvard or other elite schools to weed through thousands of applications in a fair way without using test scores.

I agree that elite schools need to do much more to admit and support candidates that show promise but don’t have the SAT scores for admission. However, not sure that means the tests should be scrapped.

27

The article is an excerpt from a book, so presumably she goes on to say how selective schools should go about selecting students. Her anecdote about the Latino student who was brilliant but couldn’t do well on the LSAT suggests that she thinks it’s still OK for Harvard to favor students who are brilliant, as determined by some evaluation by somebody. I guess she thinks that this real brilliance can be identified, and she also thinks that if it is, the result will be that more of the seats in selective colleges will go to people who aren’t rich. Personally, I have some doubts about this. I’m not convinced that Harvard’s current approach enrolls rich plodders at the expense of brilliant “diamonds in the rough” who don’t score well because they are poor or URMs. I think Harvard’s current system enrolls people who have a lot of achievements at the expense of people who have good grades and scores because they work very hard at getting good grades and scores.

28

^^ And who have the opportunities to obtain those achievements, which often requires parents with the education, knowledge and connections to find those opportunities and the financial and other means to make them happen. That, to me, is the fallacy of the get rid of SAT and all will be fair argument. College admissions are not and never will be fair.

29

Not to mention endless accounts of claimed discrimination against high test scorers and Harvard being sued! Poor H, it cannot win as it enrolls either too many rich people (I supposed Lani) or too few (Carnevale and Rose) and too few high scoring Asians (Blum et al) or too few “service oriented people.” I guess that such is the price one pays to be on the top of the mountain in academic circles.

I do not think those debates will abate anytime soon, and especially when fueled with half-truths, selective use of “facts” and shoddy research that makes it easy to discard the opinions that might be worth hearing.

30

"^^ And who have the opportunities to obtain those achievements, which often requires parents with the education, knowledge and connections to find those opportunities and the financial and other means to make them happen. "

Yes, because college adcoms can’t possibly figure out that the kid from New Trier has the means to afford the expensive tennis coach / lessons and the access to the science lab for an internship that the kid in the South Side of Chicago doesn’t have. Come on now.

31

I think there’s another point that’s kind of getting lost in the sauce here, which is that the SAT and ACT provide data about high schools, not just individual students.

Preliminarily, law school is a whole different kettle of fish than undergrad. Almost every candidate has 4 years of undergraduate education at an accredited college and the LSDAS folks have a process which both standardizes GPAs across institutions AND tells law school where the applicant 's GPA stands vs. the other law school applicants from the same college.The way in which it determines how one applicant’s GPA “stacks up” among other applicants from the same school doesn’t require the cooperation of the colleges.

It’s harder for colleges to get a grasp on high school applicants. There are many more high schools; there are far more students who attended secondary education institutions at which instruction was in a language other than English; there are more home school students. It’s just not feasible to standardize GPAs. Nor can colleges find out where most of the applicants stand vis a vis other college applicants from the same school. Look at the CDS for the most elite schools. At many of them, fewer than half those admitted submitted high school class ranks.

Add to this grade inflation. It’s bad in college, but it’s far worse in high schools.

So, what are colleges to do? They need some way to compare the applicants and the SAT and ACT provide one way to do this. Moreover, the fact that the college bound students from any given high school usually take this test gives the CB and colleges information about the high school. The public high school in my dad’s hometown–the only non-vocational high school in a city of over 100,000 many of whom are immigrants —has a median SAT score in the mid 400s (Out of 1600). My offspring’s public magnet alma mater, which also has many immigrants, has one of about 1460. Even if I know nothing more about these two high schools, I’m going to draw some conclusions about them.

There are all sorts of issues with AP tests too. But again, these provide standardized info for many college applicants. More importantly, they provide info about high schools. Colleges can buy high school profiles from the CB. These profiles tell CB how many students took AP exams, in which subjects, and how the students scored.

My offspring’s alma mater lists this information on its profile. It also lists the number of students who took each SAT II test and the median scores. Thus, I know that about 30% took the SAT II in chemistry and the median score was in the mid-700s. Again, this tells colleges info about this high school.

It might find out, for example, that the kid from an inner city school who scored a 650 on the Math 2c exam is the only kid from his high school to take it in the last 5 years and that the median math score on the “regular” SAT at that high school is a 420. That puts his score “in context.” And if you want to help the poor, you need to give context.

So, scrap the SAT and ACT and you not only do away with a tool that, IMO, is useful in assessing a kid’s readiness for college, you also do away with a means by which an admissions committee can get a quick, thumbnail sketch of the high school he attended.

32

Well said. Specifically, the following elements would have to be established: the statement, which must be about another person, must be false.
The statement must be ‘published’ to a third party, who cannot also be the person who is being defamed. Publishing in this context does not mean that it must be printed, but purely that the statement has to be ‘made available’ to someone other than the person about whom the statement was made.
If the nature of the statement is ‘of public concern’ the person who has published it must be at least liable in negligence. Public figures who seek to prove that they have been defamed must prove an additional element under the First Amendment of the US Constitution, that in publishing the statement the defendant was acting with ‘actual malice’ (by publishing something they know to be a lie) or at least to have a total disregard for whether the statement is true or not.
The person about whom the defamatory statement is made must be ‘damaged’ by the statement.

As much if this is rather arbitrary as to US News and how they establish and promulgate such rankings it would be very hard road to tow…and that is not even addressing the issue of actual malice…in short, no.

33

All this article shows is that law professors don’t understand statistics.

Lanier writes “The LSAT predicted 14 percent of the variance between the first-year grades. And it did a little better the second year: 15 percent. Which means that 85 percent of the time it was wrong.” Its hard to parse this out since it is so baffling, but Lanier seems to believe that to the extent other variables predict variance the 1st variable is “wrong”.

There is a deeper issue, though. In fact Lanier has no understanding of what regression coefficients mean. You need to look at the design of the experiment to understand them. Suppose, for example, we take a batch of identical seeds and grow them in different environments. Some more water, some more sun, some more nutrients. Run a regression and the contribution of genetic factors to plant height is zero! Lanier would look at this study and claim it “proves” genetics does not affect plant height. The same effect is happening here with the LSAT/SAT. By looking at schools that have admitted only highly-qualified students then the variation between them accounted for by test scores, similar to the variation in our experiments with plants accounted for by genetics, is quite low. Randomly admit students to Penn Law and I bet you’ll find that LSAT predicts quite a lot!

34

These schools do not want to scrap the tests precisely because they believe, despite what they say and what the CB says publicly, that they are identifying the smarty-est applicants with them.

Let’s say they are. So what? Is there evidence that the 2300 scorer has different life outcomes than the 2000 scorer, equating for other variables? (see what I did there? where’s my acceptance letter?)

The SAT is really a remarkable test, standardized like no other (ACT, with section scores jumping all over the place between sittings, is a distant second). I think it does an amazing job of highlighting a certain type if reasoning ability at a snapshot in time.

My reason for thinking the test is overused in admissions is simply that I don’t see how useful that snapshot is in predicting school or life outcomes at a granular level. Once above 1sd is there any real difference in life being even smarter, for the vast majority of us?

35

I guess I don’t see why it needs to be so linear. If holistic admissions are used (as so often claimed) and the real purpose is level of academic background and ability to succeed in college, why are the SAT/ACT required if the student has AP scores, subject test scores, DE classes, etc? What does the SAT score reveal about academic ability that wouldn’t be evident from the complete resume? Why mandate their inclusion vs multiple “equal” options of possible portfolios which take access of educational opportunities into consideration? Test scores are already not apples to apples (which is why background information factors into admission and scores, grades, lists of courses are not looked at out of context of the applicant’s whole.)

36

Here’s my suggestion: Harvard should auction off half of its admission slots, and fill the other half randomly. They’d get a lot more money, and the composition of the resulting class would be (according to Guinier’s analysis) about what it is now.

37

Test scores are not perfectly apples to apples, but they are as apples to apples as one can get, since everyone takes the exact same SAT/ACT, etc. - unlike the classroom experience.

38

Mom2 (#34): Because lots of students don’t have AP scores at all (which are pretty expensive, to boot), and their subject test scores may well be more a measure of their schools’ quality than of their ability (especially if few students from their schools ever take the subject tests – which is going to be the case for the majority of applicants from non-privileged backgrounds, unless they have been going to prep school on a scholarship or one of a handful of excellent public schools serving low-income students.

YZamyatin (#33): I don’t think the colleges believe at all that SATs identify the smarty-est applicants. I think they believe that SATs (and ACTs) are the best available tool to help compare students of different backgrounds, with different schools and different curricula, on a single scale. I also think they believe USNWR and others use the SATs, too. The really smarty-est applicants kind of jump out at you, which is why the SATs are probably used less at HYPS than elsewhere. For colleges that have to admit lots of applicants who don’t distinguish themselves from one another easily, test scores are a nice crutch.

badgolfer (#32): You may understand statistics, but your command of analogy isn’t so hot. In your example, the contribution of genetics to plant height appears to be zero because genetics is not a variable that was being measured in the study design. The numbers on the predictive effect of SAT scores on grades come from the College Board itself, are what it uses to justify the SAT, and are precisely what their studies are designed to measure. They take students who are otherwise identical students, taking identical classes, and see how the differences in their SAT scores relate to their grade outcomes. Something like five million kids or so take the SAT or ACT every year, and most of them go to college somewhere, so there are pretty extensive data with which to study the interrelationship of standardized tests, other factors, and college grades. And . . . the predictive effect of standardized tests is not so strong. Not that anything else is much better. Sure, if you look at large differences and big populations, the predictive effects will get better. A bunch of 2100-scorers are probably going to outperform a bunch of 1500-scorers in a meaningful way. But there isn’t much value in comparing a particular kid with 2300 SATs to another kid with 2200.

39

All students that them take the same AP and subject tests as well. If a student has numerous 5s and high subject test scores, what precisely does the SAT validate that they didn’t?

Also, knowing how to take the SAT impacts score. Ironically, Xiggi has an entire document stating that fact! http://www.mediafire.com/view/xbientdegj4/Xiggi+Advice.PDF

That is a quote I can agree with. :wink: It is also accurate since kids demonstrate all the time that prep improves scores. That is not academic excellence. (At least not the type of academic excellence I value.)

40

@JHS Admissions is already subjective. Admissions say they look at the student within the context of their school’s academics. I’m not suggesting that admission turn to AP scores as “the” source. I am saying that the idea that the SATs are the only way to validate education is flawed. There are multiple ways to demonstrate “quality of classroom experience.”

Why if kids have the others require the SAT/ACT? What do they specifically offer that is insightful that the others do not reveal?