Giving control of NYC's specialized schools to a political appointee

Yesterday I shared my thoughts on Bill de Blasio's plans to "fix" the selection criteria for New York City's specialized high schools. If you haven't read the post, you can find it here.

I was going to get back to CS and CS Ed related blogging today but there's more to the story.

In spite of what BdB stated, it's not enough for him to switch to another measure - the middle school state exam along with some modifiers. This is about ending the exam altogether and not replacing it with anything better. Here's a link to information on the proposed legislation for New York Assembly Bill 10427 including links to the actual text.

There's a lot of beurobabble but the end game is to basically allow the chancellor (not sure if this means city or state) to do whatever they want.

Here are some of the key points:

  • Seats would ultimately go to the top 5 to 7 percent of each middle school's class.
  • These students would have to also be classified in the top 25% city wide.
  • Top is defined by a composite score of multiple measures as determined by the chancellor which means anything goes.
  • Remaining seats will be filled by lottery of all students with a 3.7 GPA or greater.
  • It's not clear how the students would be allocated to each of the specialized schools.

So, what does this all mean?

Multiple measures could mean the state test + average. It could mean average + recommendation. It could be shoe size + favorite color. The chancellor also decides how to weight these multiple measures.

This means that the politicians can do whatever they want to paint any picture they want regardless of how their ideological and political games hurt kids.

Any teacher, in fact, any rational adult will tell you that middle schools are not interchangeable. The top students at one middle school are typically not at the same level as top students at another one. I'm sure you can cluster middle schools in terms of the students they graduate, what the students know and can do and the clusters will likely fall on economic lines. This is the problem that BdB should be addressing not how to rig the specialized school admissions stats.

Even if this is done right, you're assuming an even distribution of top gifted students across all the middle schools. If one school has higher concentration of gifted students, they're out of luck. The end result is that if you attend an SP or other gifted program for eighth grade you'll likely be locked out of the specialized schools. What if a particular middle school is attractive to all of our top math team talent? Only a couple will get in to the specialized schools under the new system since each school is allocated a fixed number of slots.

I guess this does make sense if you don't believe in gifted students or gifted education but then why not just get rid of the specialized schools to begin with rather than playing these games. This "solution" will result in parents having to decide - do I send my kids to the gifted middle school program and hurt their chances to get into a specialized high school or do I try to find the lowest performing school I can find and have my kid stand out. This is pretty messed up.

At the end of the day, though, you have to remember that any change to the process can help one group and hurt another. Stuy, for example has 46% of it's students on free or reduced lunch and is about 70% Asian. If in the new system you're increasing the percent population of an under represented group, you're likely reducing the percentage of Asian students. So, you're creating opportunities for a number of as of yet undetermined students (who might not be properly prepared for the experience) at the expense of economically disadvantaged Asian students who have proven to be prepared for the experience. Wouldn't a better solution be to fix the opportunities presented at the middle schools rather than rig the system of high school acceptance.

Going further, a response on another blog noted that BdB probably refused to do anything with the admissions criteria of the non big three because they all had numbers that he would consider more diverse. The commenter provided the Black / Latino percentages, I added the free and reduced lunch numbers:

School Percent Black / Latino Free / Reduced Lunch
Brooklyn Latin 28% 61%
HSMSE 26% 46%
HSAS 17% 25%
Queens Sci 12% 64%

All the more reason to run some tests or analysis there. Of course, if you look further you'll note that HSAS only has 25% of its students on free or reduced lunch so is it really doing better than Stuy or Sci with ~45% low income students each or Brooklyn Tech which supports 62% or are we to say that there's low income and then there's the right low income?

Certainly the education system needs work but the proposed legislation is not a solution. It's taking a flawed but objective measure and replacing it with essentially chancellor's discretion.

If the legislation passes, by the year 2022 we will have admissions to what used to be some of the nations finest high schools determined by a political appointee rather than an objective test.

It's a shame. It was very important to my wife and me to have our kids go to public schools and I was very happy to have been able to navigate the system with them from first grade through high school graduation with only a few bumps along the way. With this nonsense from the political left and the privatization push from the political right (or actually both sides) pretty soon there will be nothing left.

Changing Admissions to the NYC Specialized High Schools

Yesterday, Bill de Blasio, the current Mayor of New York City outlined how he would "fix" our specialized schools. The schools he was referring to were the "big three" of Stuyvesant, Bronx Science, and Brooklyn Tech and then five additional schools - The High School for Math Science and Engineering at CCNY, The High School for American Studies at Lehman, Brooklyn Latin, The Queens Arts and Science High School at York College, and Staten Island Tech.

All use a single SAT style test, the SHSAT, for admission. The Big Three are locked into that system unless there are changes made by the state legislature. The Mayor can make changes to the admissions criteria of the other five on his own.

The problem, as stated by Chalkbeat writers Alex Zimmerman and Monica Disare is that:

The city’s specialized high schools — considered some of the crown jewels of New York City’s education system — accept students based on a single test score. Over the last decade, they have come under fire for offering admissions to few students of color: While two-thirds of city students are black or Hispanic, only about 10 percent of admissions offers to those schools go to black or Hispanic students.

I normally don't write about things like this but given my history at and with Stuyvesant I thought I should.

Before getting started, my biases - I believe in public schools, not charters, I also believe that a system should have gifted programs and magnet schools and also a mix of large comprehensive schools and smaller niche ones. Additionally, while I do believe in gifted programs I'm not sure at what point it makes sense to start offering them. When we were looking at elementary schools for our own kids we opted for a neighborhood public school rather than applying to the district gifted program as we felt it offered a better overall education. On the other hand, had our kids not made Stuyvesant, I'm not sure what we would have done. Part of that was due to the fact that the Bloomberg administration destroyed just about all the large comprehensive schools so, other than that large test schools and LaGuardia, good luck finding a school that was strong in the humanities, sciences and offered a robust music program.

The overall picture

For reference, here are unofficial cutoff scores for the specialized schools from 2018 along with the NY State data on free and reduced lunch for 2017. Note that there's overlap between the schools. I separated out SI Tech since geography probably affects it's numbers more than the other schools and I separated the big three from the remaining four because I think the three are all substantially larger. The free and reduced lunch percent is important because I believe an entire school qualifies for Title I funding if the number's over 40 so while these schools are not populated entirely by low income students, all but HSAS has a substantial number and there's a chance, some say a good one that at best BdB's plans will merely shuffle which low income kids get in rather than actually solving the root problem.

School Lowest Score admitted High Score Pct free or reduced lunch
Stuyvesant 559 698 46
Bronx Science 518 637 45
Brooklyn Tech 493 668 62
HSMSE 516 616 46
HSAS 516 633 25
Brooklyn Latin 482 555 61
Queens Sci 538 542 64
SI Tech 519 660 40

The plan

Part 1 - test prep

There's idea here is that test prep is rampant for SHSAT and that the test can be gamed. This may or may not be true but if it is then it's safe to assume that a lot of low income kids are doing test prep either by prioritizing test prep programs in family budgets, working off of an SHSAT test prep book or by attending a middle school that has some form of prep.

There has long been a program on the books called "The Discovery Program" which has a free city run test prep component and also a summer component. The summer component is for low income students who score within "a few points" of the school in question's cutoff score. At the end of the summer, I believe the students retest and in general make the school in question.

This program is a good thing and should be reinstated but I don't know if it will result in the changes that BdB is looking for. More likely Stuy will get its discovery kids from Science and Science from Tech. As there are already many low income students at all of the schools there's no indication that this would increase the number of under represented groups which is BdB's goal.

It's also worth noting that there's a sizable chunk of Brooklyn Tech students that don't perform particularly well on regents exams. Everyone knows I'm not a fan of standardized tests and I don't believe that students that score a "few points" lower on the SHSAT are any less qualified to be in a specialized school but at the lowest performing end of these schools, one has to ask how specialized is it really?

Part 2 - use state test scores instead of the SHSAT

This could be a very good thing with one major caveat.

Twenty one middle schools produce over half the students that ultimately are accepted to the specialized schools. For the rest, there are a few reasons

  • Many students don't take the SHSAT.
  • There have been stories of schools steering kids away from the specialized schools.
  • There's no culture in those schools of striving for specialized school acceptance or attendance.
  • The academic programs at those schools don't prepare students for the SHSAT exam for reasons that we're not going to delve into here.

All students already take the NY State middle school Math and English exams. Yes, it would further raise the stakes for those exams but it would ensure that the specialized schools are on the map for everyone,

The big question here is "do the current state exams cover enough material in enough depth to select and differentiate for gifted programs." I don't know the answer to this question. If they don't then using state test scores will be a disaster. If they do, then they could be a suitable alternative.

Part 3 - admissions from every middle school

This is the third part of BdB's proposal - guarantee admissions to the top 10% of students at every middle school.

This could be a real disaster either for the students or for the schools. There's no question that students coming out of different middle schools are at different levels of preparedness for the specialized schools. Arguably the most obvious difference is that some middle schools offer algebra and others don't.

For better or worse, the specialized school's curricula, classes, and overall programs have evolved hand in hand with the exam. If you all of a sudden end up placing a large number of students with a different and arguable weaker academic profile into these schools one of two things will happen. Either the students will struggle immensely or the school will have to lower its standards. There's no two ways about it.

If you set standards for academic preparedness for the specialized schools either by making a hard cutoff on an exam score be it state or SHSAT and then spread those acceptances across all the schools maybe it could work. If not, you're setting up students with high grades at schools with weak programs up for failure.

Another problem is what to do with non-public school students - private schools, charters which take public money but act as private schools, parochial, and home school students. Where do they fall in to this equation?

Is one measure bad?

So there you have my thoughts on BdB's specific proposals but I also want to address his contention that a single measure is bad and makes the comparison to colleges where they by and large use multiple measures. This is a ridiculous comparison. Yes, colleges look at multiple measures but after watching decades of Stuy students go off to college, it's clear to me that the process is by no means fair or consistent. Elite schools can easily game their acceptances and they still can have 100% of their applicants with off the charts SAT scores.

A single test might not be ideal but it can only be gamed by test prep and test prep can be as cheap as buying a test prep book.

As to other measures that come up from time to time - interviews, portfolios, essays - who does that help? The low income kid or the well to do one?

If you're going to add another measure for admission it has to be something that can't be gamed or politically influenced,


Should the city address the fact that there are groups that are under represented at the specialized schools? Certainly, or rather, the city should address deficiencies in opportunities that are like the causes of this under representation.

Should the city be doing things differently? Also yes. Here are some thoughts on what the city can and should be doing.

Do the experiments in the non-big three.

There are five specialized schools that aren't covered by state legislation. Why not run your experiments there. You've got the big three as a control group and what's more if you look at the "acceptance score windows" in the table above, you'll see that there's enough overlap that you can probably get some real data out of these experiments.

Why not change the admissions criteria for one or two of them and see how it goes before jumping in whole hog. Figure out if anything works first.

Check existing data

We've got SHSAT results and State Test score results along with student grades for years. Why not study correlations between them. Rather than making a specious claim about middle school grades in NYC Public Schools vs state tests vs the SHSAT, run the numbers. Take a look at student high school performance based on these predictions. This shouldn't be guess work and city should have the data to do better.

Fix the middle schools

This one is more pie in the sky because no politician really wants to take this one but you've got to look at what's going on in the middle schools and why.

I tweeted this the other day:

The truth is that if you did this, the high performing school would probably continue to be high performing and the low performing one would still struggle. There's only so much a teacher can do. A school with a disadvantaged population needs the resources to succeed - lower class sizes for a start. Add wraparound services, extensive after school and weekend opportunities - make the schools a part of the community and maybe we can get somewhere.

Another question relating to our schools in general is to look at attrition rates to private schools - how many students in our under represented groups are being siphoned off to private schools on scholarship. I don't have any data for this. It could be an insignificant amount or it could go a long way in explaining the downturn in the numbers over the decades.

So where are we

There are a lot of changes I'd love to see in our public schools and I do believe BdB's heart is in the right place. I'm concerned that some of the solutions that he and other politicians come up with will hurt the current low income students at the specialized schools and unintentionally advantage the well to do while not helping those that at least BdB may honestly want to help.

He should enact Discovery and research the State Tests but proceed with caution. The test schools have always been and continue to be a gateway for poor and immigrant students and any changes that are proposed should be weighed carefully and tested before taking steps that could real harm when good is intended.

The Tech or the Teacher

Every morning one of the first things I do is quickly glance at my emails and other notifications. I really should wait until I'm more awake but old habits die hard.

As some of you know over the past couple of years I've been making a series of videos and related post on using my editor of choice, Emacs. I've done 48 videos, have over 2500 subscribers on YouTube and people seem to find some value from the series.

This morning one of my emails was a new comment on one of my videos:

Thanks for not describing a single key combination and how that fits into basic usage of org mode. I was going to subscribe but if all your videos are going to be about showing off how fast you are at typing emacs commands on your annoyingly noisy keyborad, then no thanks!

My first reaction was "you know, you could have said something like 'if you slow down an actually say the keys it would be much easier to follow and would be more valuable' and it would be much better received" but then I started to think about instructional videos in general. I started to make these videos for my students. I thought they might be helpful and then found that the Emacs community by and large liked them. They're not teaching anything deep or complicated just how I use a programmable editor. Even so, I was reminded by this comment about the limitations of videos and ed tech in general.

Videos provide some nice features - you can watch them at any time, you can pause and replay, and in some cases you can watch at super speeds. What you can't do is slow the video down and most importantly you can't ask questions or interact with the instructor. You can leave comments if the video is on a platform like YouTube but that's pretty limited.

On the other hand, if this was a class, we would have been able to interact and I as the instructor would have had a much better chance to make sure this "student" was up to speed.

Of course, this wasn't a class. I don't have the time to teach tools like Emacs in class and since it is a mechanical skill, it can be offloaded to a video. The video has the ancillary benefit of living on and being discoverable and viewable by the public at large. This is a good thing. Videos are tech and tech scales. Unfortunately, good education doesn't.

Videos are tech and tech is the rage. Lots of people are working hard to get more tech in front of our kids - usually at the expense of having a great teacher. This is something we don't need. Technology is nice but I'd rather have my kids in a small class with a knowledgable teacher and a piece of chalk than all the tech in the world.

As a computer science guy and a computer science teacher I think it's good to remind myself that tech is cool but a good teacher is way cooler.

Enter your email address:

Delivered by FeedBurner

Google Analytics Alternative