Actually, kids whose parents aren’t paying for private tutoring may be working very hard to prepare, but not as effectively simple because they don’t have the benefit of the paid professional tutoring. Much of the tutoring focuses on test-taking strategy vs. actual substantive content, so a kid who self-studies may be disadvantaged simply because they aren’t given the tricks and short-cuts that the tutored kids know about. Logically the self-studiers would also tend to want to concentrate on their weakest areas, since that makes intuitive sense But as pointed out upthread, it turns out to be strategically better to go for the highest possible scores in areas of strength, because of the high upper tail end score values on the test.
@mom2and says
Again, there are free prep courses. 3 of the 7 Black --they may not be African-American–and 1 of the Hispanic kids who got into Stuy took the FREE Khan Academy course on line. The FREE summer prep courses NYC has offered often had empty seats. The FREE course arranged by Black and Hispanic alumni of these schools has had trouble filling all the seats as well.
And, please note that while a higher percentage of Asian than white test-takers got into Stuy, the difference isn’t that huge; it’s 29% vs. 27.3%. While some of the white kids I know who got into Stuy took a prep course, not a one spent his/her childhood focused on test prep. I seriously doubt that the most of the successful white applicants I don’t know personally did either.
Beginning in 2017, NYC offers every high school junior the chance to take the SAT on a school day at the high school (s)he attends for FREE. This policy dramatically increased the number of Black and Hispanic kids in NYC public schools who take the SAT. However the percentage of Black students who take the test is lower than the percentage of white and Asian kids who do. I’m not talking about how they do on it–just whether they take it. The test is free; it’s given at your high school during the regular school day. And still…a higher percentage of Black and Hispanic kids just don’t take it. (And the stats are based on the total number of kids who take the SAT during the first 3 of all 4 years of high school even if they take it on a different day at a different place.) https://infohub.nyced.org/docs/default-source/default-document-library/2017-sat-results—website—1-11-18.pdf
Trying to engineer a change in the admissions process to try to make these 3 high schools more reflective of the underlying school population is, IMO, smoke and mirrors. It’s treating the symptoms without dealing with the underlying problems.
The previously linked study in post #61 goes into more detail. The specific Stuy first year grade distributions are below, There are indeed quite a few kids at Stuy who are not doing well, more than just slight differences. 1/3 of male students had a FYGPA below 85%, and 17% of male students had a FYGPA in the C range.
<80% FYGPA – 17% of male students, 3% of female students
80-85% FYGPA – 15% of male students, 8% of female students
SHSAT score only explained 5% of the variance in first year GPA among this group, and explained less variance in GPA for later years at Stuy. Being male was a much better predictor of not doing well at Stuy than was SHSAT score, among matriculating students (range restriction issue). The male C students at Stuy averaged the same SHSAT score as did female students with a respectable 85-90% GPA. Being Black also appeared to be correlated with having a lower GPA at Stuy, although the sample size of Black students at Stuy was very small, so it may not be significant. The overall average FYGPA of male Black students at Stuy was in the C range, while the overall average for women at Stuy was in the A range. Black women at Stuy also averaged lower GPAs than women of other races, but well above Black males.
SHSAT appeared to have almost no power in predicting who would do poorly at Stuy (among matriculating students, with limited range), suggesting there would be little impact on expected GPA at Stuy if some kids were admitted with SHSAT scores somewhat below the existing cutoff. However, SHSAT did have a more limited ability in predicting who would be among the top GPA students at Stuy. The few top ~10% of students had an average SHSAT ~20 points higher the overall average of Stuy students. However, there were plenty of exceptions. Several kids directly on the minimum SHSAT score cutoff had a close to a perfect 100% FYGPA at Stuy. While one of the 2 kids at Stuy with the lowest FYGPA in the entire class had an above average SHSAT score, near the average SHSAT score for the 95+% FYGPA group.
I don’t think the SHSAT is intentionally biased against URMs, and instead there are other reasons why URMs tend to score lower. However, basing admissions almost entirely on SHSAT is expected to result in a smaller URM percentage than most other alternatives, and many of those other alternatives would likely do a better job of predicting who will do well at Stuy and similar.
@Data10 Please correct me if I’m wrong, but I think that the Stuy students admitted through the Discovery Program DO have scores below the normal cut off.
It’s my understanding that the Discovery program does support admission for lower income 9th graders who scored just below the threshold to be admitted. Nearly 70% of Discovery program participants are Asian. However, I don’t think the Discovery program was in place during the years of the study, which is consistent with there appearing to be a single score cutoff in the graphs.
I was invited to the Discovery program for Stuy as I scored one point below the cutoff. This was in the late 70s. I don’t remember it being called “Discovery”, but it was presented to me as summer school, which I declined in favor of a better offer.
So it’s been around awhile.
ETA: I found a reference to it being “50 years old”.
Any source for this? When I was in college my ex-girlfriend worked for Kaplan for SAT and ACT prep and they did teach actual content. The only specific test-taking tips were things like “don’t guess” (there was a penalty for wrong answers back then). I highly doubt any standardized test would have so many tricks and strategies that entire courses could be dedicated to them.
I’m pretty sure the scoring information is available for all test takers. In fact, it even appears in the official SHSAT handbook here: https://static1.squarespace.com/static/59b0473fa9db0934a50d87d5/t/5b21753e562fa7589de80759/1528919366035/SHSAT+Handbook+2016-2017.pdf
Are you just saying these things based on your own personal feelings?
Here’s another data point to throw in:
I just looked at data for the percentage of children in single-parent homes in New York:
Asian / Pacific Islander - 14%
Black / African American - 64%
Hispanic / Latino - 50%
Non-Hispanic White - 22%
This is statewide data for 2016 (the latest year available from my source).
Data Source: https://datacenter.kidscount.org/data/tables/107-children-in-single-parent-families-by-race#detailed/2/34/false/870,573,869/10,11,9,12,1,185,13/432,431
That in itself is going to create significant race-based disparities in resources available to children, with “resources” including both financial and the resource of parental time/attention.
@wyzragamer She’s referring to the test’s scoring which favors a very high score on one part rather than high scores on both parts.
That’s not info that is in the guide you posted above.
It does appear on page 21:
^ that doesn’t explain or even mention that the best strategy is to focus on the one of the two tests that you are best in, @wyzragamer
What? It explains exactly the scoring strategy that leads you to conclude it, complete with quantitative estimates.
Whether that’s actually the best strategy or not is a subjective matter that depends on the student.
Seems like that test prep advice to “don’t guess” would have lowered some students SAT scores. The guessing penalty on those old SAT scores made the expected value of random guessing zero, the same as leaving the question blank, so guessing in that situation had an expected neutral effect on the score. But guessing after eliminating at least one answer had a positive expected value, so test takers should guess in that situation.
So it is entirely possible that the existence of test prep of negative value means that the overall average effect of test prep is not that good, even if optimal test prep can result in large gains for some students.
Regarding test taking techniques versus actual content, some test prep books and presumably courses focused on test taking techniques. I remember seeing a Princeton Review SAT prep book (after I started college) and noticing that it mostly focused on test taking techniques (including guessing); the main content knowledge it emphasized was a list of supposed SAT vocabulary words.
@wyzragamer – my son worked for Princeton Review teaching GRE and was trained both for GRE and either SAT or ACT (I don’t remember which) — and he said the training focused almost entirely on test-taking strategies and test short-cuts. He was hired primarily to teach math because of his perfect score on math GRE, and said he taught very little substantive math. Obviously my son’s experience with PR doesn’t tell us what the NY SHSAT schools are teaching, but I figure it’s one up over your ex-girlfriend working for Kaplan 
I’d add that the real key to PR’s success is the repeated practice testing – that is why they tie sitting for the practice tests to their money-back guarantee. The kids getting the professional test prep are probably getting much more practice with a greater volume of material to practice with (test variations, different questions, etc.)
@ucbalumnus I would count memorizing vocabulary as actual content. Studying for any test, even math or science tests in school, generally involves memorizing lots of the relevant information. I don’t remember exactly what the test taking strategies were (so it’s unfair to assume that program was “suboptimal”) but I think they had satisfactory results.
I don’t necessarily think test prepping produces great thinkers, I’m just trying to counter the narrative that other people are pushing. That test prep is just cheating by exploiting structural weaknesses in the test design, and all Asian people force their kids to do it all the time, with the tacit implication that the Asian kids who go to elite schools are systematically less deserving than other races.
@calmom I think both anecdotes are consistent with each other. The person I knew who at Kaplan hated the job (she got a perfect SAT score on the first year that the SAT was changed to 2400 scale) and she felt that all the information was very basic, like a remedial course. They also did a large volume of practice tests, which I think is a legitimate study tactic that reinforces real knowledge. I think there’s a distinction here between teaching low-level, test-specific knowledge and cheating.
Umm…@Data10 …just found a link that explained …
The Discovery Program was begun in the 1960s and was formalized in law in 1971. Initially, it permitted students who had just missed the cut off score to enroll in summer school. If they passed the summer program, they were permitted to enroll. The practical result of this was that Stuy was “stealing” kids who had gotten into Bx Science. BxS was “stealing” from Brooklyn Tech.
When 5 more exam schools were opened, the DOE decided that the Discovery Program would only be open to kids who missed the cut off for ALL 8 exam schools. This means the kids studying to get into Stuy had scores AT LEAST 90 points below the cut off for Stuy. The then principal of Stuy thought that it was impossible to get these kids up to Stuy levels in one summer, so he stopped using the program sometime around 2003. The program seems to have continued at BxS and especially Tech. http://www.theschoolboards.com/showthread.php/1400-Stuyvesant-Principal-explains-why-he-stopped-Discovery-Program-for-promoting-diversity
The Discovery Program was later reinstated at Stuy…I don’t know what the selection criteria are now, though.
Yes, but the point was that Princeton Review test prep books (and classes as @calmom mentions) were mostly test taking techniques, with some content knowledge (SAT vocabulary words), while the Kaplan class you describe had mostly content knowledge with some (apparently incorrect) test taking techniques.
Test preparation is not cheating, of course. But when test preparation becomes widespread, it may reduce the effectiveness of the test measuring things other than the quality of test preparation.
I don’t really want to get into the nitty gritty of test prep, as I’ve never taken it myself and haven’t enrolled my daughter. I also don’t want to defend the single-test approach to admissions, but I haven’t heard policymakers discussing viable alternatives before deferring to race-based solutions (and the problem is described as a diversity one, not a mismatch of low-quality students at elite schools).
I just wanted to share some second-hand anecdotal experience and common sense to explain why I don’t think it’s unfair or unhealthy or means that Asian kids are less deserving or have less real knowledge.
I’m a professor and once taught at a prestigious large public Southern California research university with a relatively high Asian population. I’ve heard (non-Asian) students and teachers talk about how all the Asian kids gather around into cheating rings, and colleagues talk about how Chinese research is all mass-produced and low-quality. Administrators who look out onto the campus and bemoan Proposition 209 for making them have so many Asians.
The local area is segregated, being mostly wealthy and white with some nearby suburbs which are predominantly upper-middle class Asian. Other white parents complain about the test prep and competitiveness of Asian kids pull their kids out of majority Asian suburban public schools (which are some of the best in the country) and put them into one of an extensive network of private schools. The urban public school system, which despite being extremely well-funded is rife with corruption and mismanagement, is almost exclusively black and Latino even in areas with large white populations.
All this while the non-Asian undergraduate students and local rich white kids take advantage of tutoring services which cost $70/hour from PhD students who basically work them step-by-step through their homework.
Maybe it’s a little biased, but I imagine a lot of the parents on CC as white moms and dads who complain about cheating Asian kids and how bad tiger moms are while spending all their free time and money on extracurricular activities for the kids, enrolling them in private school and posting on college message boards to ensure their best outcomes. Why are we so focused on test prep here? Not because it has any substantial effect on de Blasio’s policy (which still uses the SHSAT test, together with an ad hoc proxy for race as its only inputs) but because we rationalize that test prep is what gives Asian kids the advantage, and so they don’t really deserve to be represented in such high numbers at the elite schools. Let’s look inside ourselves and be truthful that’s what we’re doing. And maybe then the conversation can continue in earnest.
Seriously???
“That test prep is just cheating by exploiting structural weaknesses in the test design, and all Asian people force their kids to do it all the time, with the tacit implication that the Asian kids who go to elite schools are systematically less deserving than other races.”
Does this not sound a little racist? Or maybe a lot?
As per my post, my Asian daughter took only one SAT test and no ACT test in high school and got a 1580 with about 4 hours of math review the day before the test. Please do not say all Asians are the same. Not only was she a top student in high school, but she danced 25-35 hrs per week YEAR ROUND on studio dance team and high school dance team. She was a front row dancer of what I believe is a top 10 studio in the US; the studio and was featured on World of Dance and will be on America’s Got Talent in the spring. She won dance scholarships from famous choreographers at multiple conventions.
And BTW- she is Southeast Asian and I really don’t like the stereotyping of Asians and I really also don’t like the grouping of all Asians into one category. And I really don’t like her getting lumped in together with a group of Chinese students that people seem ready to discriminate against with holistic admissions in the same manner that Jewish people were discriminated against by the Ivys in the 1920s.
“all Asian people force their kids to do it all the time”
NO WE DON’T. The only one I know in our modest middle class neighborhood doing test-prep was a white girl in the higher ability class. Her mom told me that she made her get up at 6AM to do Kuman math every day. I told her I worked full-time and I was not about to get my kid up at 6 AM to do extra math… but we did stay out until 10:30 on weeknights for dance.
Please also note that Southeast Asians are considered one of the URM groups by west coast medical schools. Another reason to not to group all Asians together in one group.
Umm… @momofsmartdancer Reread the post. @wyzragamer wasn’t expressing the opinion that annoys you; (s)he is saying that’s what too many white people think and that arguing that the admission process for the sci highs should be changed because it’s too subject to test prep is really about reducing the percentage of Asian students at these schools.