School Of One

When used appropriately, technology can really make things better.  It can also be a big waste of money.  In my math class, I have incorporated an incredible computer program called ‘The Geometer’s Sketchpad’ which first came out in the early 1990s.  With this program, which only costs about $10 per student, I’ve developed activities that encourage critical thinking and collaborative problem solving.  My students have also benefited from my class website where they can download any handouts they lose or were absent for, in addition to all the homework assignments.  I can even email an entire class with the push of a button to clarify something from class that day.  These things have made me a more effective teacher, and they have been practically free.

The School of One is a technology that was hailed by TIME magazine as one of the 50 best inventions of 2009.  By harnessing the power of computers to analyze the skills of students, the computers could create customized learning plans for each student, and find an efficient way to match the students with teachers to maximize learning.

I visited the School Of One in the spring of 2010, during the initial pilot program.  I am (or at least ‘was’ until I wrote this post) friendly with the Joel Rose, the former ‘CEO’ of School Of One, and I asked if I could go to one of the tours.

Before we got to see the school in action, we got a brief summary of what we were about to see.  The concept was that the old teacher-centered model was ineffective.  One teacher would have 30 or more kids.  Some of those kids had the prerequisite skills for the planned lesson, and others didn’t.  So the teacher would struggle to ‘differentiate’ across all the different starting skill levels.  That was true.

So the School of One would alleviate this by having a large room where five different lessons would be taught simultaneously.  Some students would be taught by teachers in large or small groups, depending on what skills they had not yet mastered.  Others would work on individualized lessons at computers.  This would enable the teachers to get the most out of their direct teaching time as students would not participate in a lesson that they were not ready for.  Sounded pretty good to me.

Though the school had only done a small pilot program in the summer of 2009, they claimed “Students in summer school pilot acquired new math skills at highly accelerated rate: estimated by NYC Department of Education’s Research and Policy Study Group as 7 times faster than peers with similar demographics and pre-test scores. Small sample size (n = 76) was a limitation.”

Seven times faster.  I guess if effective teachers teach three times faster, then effective teachers with computer assistance teach seven times faster.  This stat is a huge red flag.  It is similar to a diet that clams you will lose thirty pounds in one week.

After the explanation of the School Of One, we were taken into the room, which was designed a bit like an airport terminal.  Monitors told the students who should go to what station.  There were about 100 kids in the very large room with about six teachers.  Two teachers were teaching in the corners to groups of about 20.  Two teachers were teaching in the middle to small groups of about 10.  Two teachers were supervising students who were working independently on laptops.

I peered over the shoulder of a student working on a computer and saw that the skill he was practicing was to take sentences like:  “three less than t” and convert them into ‘t-3’.  As this is one of the more useless skills in all of middle school math, I was disappointed that they didn’t take the opportunity to drop a few topics in their reinvention of math teaching.

I also noticed that very few students were doing any scrapwork for any of their problems.  This reinforced my own experience in trying to learn things on the computer, like when I tried to learn music theory that way — it is tough to take a computer seriously when it tells you to do something.

I also noticed a lot of kids playing games.  The games came from a website called (Real slogan:  Come for the games, stay for the lessons).  The three most popular (click on links to play the games and see how much math you actually do while playing them) were 1) a pac-man game called math-man where your pac-man tried to clear the board of dots while avoiding ghosts.  The ghosts had numbers on them, and you could eat the ghost with the answer to a math question at the bottom of the screen — or not 2) a game called Bloxorz where the player moved a rectangular box around a maze, and 3) one from where if you got a multiple choice question right, you could try to make a basketball shot.  Part of the idea of individualized learning is that some students will go at an accelerated pace.  But from what I saw, if kids finished early, they were allowed to play these math games.  Unfortunately the games had very little math.  I watched a kid play math-man for fifteen minutes without answering a single math question.  In this screen shot, you can see that the question 83 divided by 1 is at the bottom so if you really want to do the question, you can eat the ghost with the 83 on it, or you could just avoid the ghosts and eat all the dots.

The best thing I saw was the small group instruction, something that I know is effective, where a teacher was working with about 8 students on a math concept, giving each student individualized attention.

Ten minutes before the math period was over was ‘assessment’ time, where each student went to a computer to take a multiple choice test to inform the computer if they would be ready for the next concept tomorrow.  Just like I saw during the lessons, I watched a student do the entire multiple choice test without lifting pencil.  As there were only 4 choices per question, he may have even passed, but it is doubtful he truly knew the concepts.

Next to the large room of School of One was another large room filled with adults with suits on.  They were analyzing, in real time, the data that was coming in from School of One.  Based on the amount of mindless game playing going on in the main room, I wondered what they were analyzing in the other room.

I left School of One feeling a bit funny.  It seemed like there were some good things about it.  Certainly a computer sorting students into ability groups does make it easier for a teacher.  Also computers grading work is helpful.  But the major flaw that I considered a missed opportunity was that in the design it seemed that their main ‘goal’ was ‘test prep.’  If I were to use millions of dollars to ‘re-invent’ the math classroom, I would want to develop a better vision of what the goal of studying math was.  As they were clearly focused on test prep, the entire enterprise, in my opinion, was limited.  Even if they succeeded at getting test scores up, that superficial goal would be all they really accomplished.  There was not one bit of ‘critical thinking’ going on there.

On the tour with me was the founder of Mathalicious, who was just getting started back then, but has since been getting pretty popular in the world of internet math people.  He seemed as unimpressed as me so I talked to him and asked what he thought.  He said “I’m worried that this is what kids will think math is about,” which I agreed with.  The technology, I admit, was pretty new and maybe ‘critical thinking’ can be infused later with better lesson content.  I don’t know if that was part of the long term plan.

My visit was in the spring of 2010.  They were doing the spring pilot at IS 228.  Then they were going to expand to two more schools for the 2010-2011 school year, and then to four more schools the next year.

In the 2010-2011 school year, the three schools IS 228, MS 131, and IS 339 all started their first full year.  By the end of 2011, two of the three schools, MS 131 and IS 339, had dropped the program.  The CEO and the other employee at School of One also resigned from NYC DOE so only one school continued, the original school IS 228.

I checked the school report progress reports to see how math scores were affected by using this program.  I know that these numbers aren’t as meaningful as the DOE hopes they are, but I have no problem using them here since these are the numbers they value so much and it is always ironic the way the DOE chooses to ignore numbers that don’t fit with their theories about what will help the schools.

At IS 228 in Brooklyn, the ‘flagship’ school:

In 2009-2010 had a 41.8% in their math median growth percentile.

In 2010-2011, after doing SO1, they went down to a 37.6% median growth percentile.

At MS 131 in Manhattan:

In 2009-2010 had a 62.7% in their math median growth percentile, which was enough to help the get an ‘A’ on the large ‘progress’ component of the progress report.

In 2010-2011, after doing SO1, they went up to a 74.4% median growth percentile.

At IS 339 in The Bronx:

In 2009-2010, they had a 36.5% in their math median growth percentile, which helped them get a ‘B’ in ‘progress.’

In 2010-2011, after doing SO1, they went down to a 12.9% in math median growth percentile, which contributed to their ‘D’ in ‘progress.’

So of the three schools, one went up on their math progress while two went down.  Two schools dropped the program, including the one school that showed ‘progress’ with it.

Within a year, Joel Rose and Chris Rush, the two main people working on School of One, re-emerged with a new company called New Classrooms Innovation Partners, which will work with cities, including New York City — where they will expand to 50 schools within the next three years.  Though New York City is not immediately giving them a lot of money, this article  questions whether or not this violates the conflict of interest laws.

Former Chancellor Joel Klein who now works for News Corp has just started an ed tech venture called amplify which I believe is working with New Classrooms Innovation Partners in some way.  I haven’t found a direct link between the two companies, but it would seem strange that Klein would be competing against people who he mentored.

As I’ve said, I do think that computers can be used to help teachers work more effectively.  Whether or not this technology will accomplish much is yet to be seen.  As it does not seem to be based on any vision of making math a more interesting or more meaningful experience, I am skeptical.  If the common core standards truly are about making math no longer ‘a mile wide and an inch deep’ I wonder how this framework can be adapted to enable deeper understanding.

I guess the 50 schools in New York City that will get the ‘opportunity’ to participate in the field tests will soon find out.

This entry was posted in Teach For America. Bookmark the permalink.

22 Responses to School Of One

  1. JDM says:

    Jeebus. Will no one stop this maddening fraud? I don’t think any kids – including poor ones – should be grist for this batshit mill o’ money.

  2. Gary, see here:

    Like you, I can’t confirm SOO still contracts with Wireless Gen under their new name of New Classrooms – but it seems very likely they are. The business plan is to expand to Perth Amboy and Chicago first.

    B4K – which is tied to Students First, which gets its funds from Rupert Murdoch among others – came to the superintendent of Perth Amboy’s rescue recently. Quite a web of connections.

    Leonie Haimson’s impressions of the program mirror yours closely.

  3. AmyW says:

    It just seem like this “personalized learning”/”blended instruction” model has so much potential, if the learning they are doing on the computers is of a high level. I’ve been doing some initial searching, and I know there is a lot out there, but it’s hard to figure out what is truly good.

    Are you familiar with Dreambox? I did part of their demo, and it looked like it was promising. And while it had games, playing them was very math related – I had a giant foot trying to smash cars as they drove by with number sentences that equaled a target number. The games also seemed to be limited by how many coins you had earned doing lessons?

    I was also looking at Education City, but haven’t gotten very far with exploring it. It seems to have built-in white board lessons that can start class, and then kids are sent off to individual work.

    I’d love to know if you thought there were any “good” or promising online learning programs/sites so that a teacher could be freed up to work with small groups while other students worked at their own pace.

  4. Liz Rayment says:

    One of the novelties of School of One was the use of lessons from many different publishers. Instead of buying an entire text, lessons are purchased much like an mp3. Tracking is done on how many sessions it takes for a student to learn a lesson with the method from a particular publisher and compared to a “par value” to determine the more effective treatments of a particular topic. SO1 also uses more learning modalities. For example, some bottom-up learners that learn better through building do hands-on activities.

    I see no problem with playing math computer games or solving problems without paper for this level of math. My own district could and should do much more to strengthen math fluency and mental math in a fun way. As it is, most kids hate math and only 30% of the 10th graders in our state test proficient.

    I haven’t met any high end math student that likes “collaborative” work, citing bullying in the classroom when the teacher isn’t looking and continued outside the classroom to extract homework. It isn’t fun for them to try to teach the others so as to get a better grade on team tests. I think there is something to be said for the SO1 model where the students are ability grouped and matched to their modality.

    I don’t think your 1 visit is a sufficient sample size to warrant throwing out everything. Trailblazers always have a bumpy road that’s improved and smoothed by those that come behind. Is SO1 perfect? Of course not. Are our present methods perfect? Details on where Joel Rose et al went can be found at:

    • Gary Rubinstein says:

      I didn’t say that everything should be thrown out. Just that something that has not yet been proven to be effective by any measures is maybe not something taxpayers need to invest hundreds of millions of dollars into yet.

      There might be some good games, but those ones weren’t.

      Also, if I understand your logic. Since what we already have is not perfect and since SO1 is also not perfect, we therefore need to spend a lot of money on SO1?

      And I’m not just judging on my one visit, but also on how the program was tried and dropped by two out of three schools.

      I definitely think that there is something we can learn from SO1, but if the PR for it, like the PR for everything in the ‘reform’ movement is inflated, there is no way we can make real progress in this country.

  5. E. Rat says:

    Beyond their questionable results, I have a hard time believing that these technology-based programs’ cost efficiency is real. What are the startup costs just in terms of hardware? How often will machines and programs be upgraded? What kind of allowance is made for wear and tear?

    Technology has been the great answer since I was in elementary school and we programmed in BASIC on Apple IIs. (I think that TESSERACT program that Michelle Rhee taught in Baltimore was computer-based, too.) But given that my actual classroom technology consists of one eighteen year old eMac and one (donated) six year old PC, I’m not quite sure how that technology piece is supposed to work out.

    The New York Times has published some stories recently on how Apple wines and dines district executives into large contracts. It’s hard to see any of these technology initiatives as more than another way to move public money into private hands.

  6. wendyl says:

    The Connecticut Association of Public School Superintendents is pushing for this “school of one” model to be the model for ALL schooling in Connecticut. Aside from the questionable effectiveness, this model seems to undermine a major goal of public education- to develop responsible, tolerant citizens who work well together. Where is the collaboration, sense of community in a “school of one?”

  7. Emily says:

    Hi Gary,

    I’m a college student who would like to become a high school English or SS teacher. I’ve poked around your blog a bit and your criticisms of TFA are really illuminating. I will NOT be applying for TFA!

    However, I’m not an Ed major and still need to find some sort of alternate certification program so I can teach. Do you have any suggestions or advice?


    • Gary Rubinstein says:

      Hi Emily,
      It depends on what state you want to teach in. I know that Boston has a 1 year residency program. Mississippi has one also. Sorry I don’t know more about this, but let me know what you find and I can tell you what I think.

    • Steve M says:


      Years ago many (large) school districts had their own credentialing programs where people with bachelor degrees in appropriate subjects could apply to become teachers in (generally) hard-to-place schools. The typical applicant was thirty to forty years old and was looking to start a second career. But, many new graduates also got into teaching through such programs (I was one). In a lot of respects, such programs resembled TFA in their methods (but you can see how districts such as Los Angeles were keen on keeping those teachers who completed the program, unlike TFA). However, such programs were fairly rare as they had to meet the same requirements as traditional college programs. Los Angeles Unified’s program was called the “District Intern Program”, as was probably the most successful in terms of teachers credentialed [Norm Marks, the director, was phenomenal].

      Unfortunately, these programs were needed much more back when we had a large deficit of credentialed teachers in the core subjects. Today, most urban districts that are large enough to maintain such a program have dropped them because of budget constraints, increased state credentialing requirements and a surplus of traditionally trained college graduates. The rise of charter schools has also played its part.

      Look up such programs in the larger school districts, some still exist but place an emphasis on getting candidates credentialed in Special Education.

      • Emily says:

        Hi Steve,

        Interesting. I have run across a few district programs – but you are right, a lot of them cater to Special Ed.

        Would you have any other suggestions besides these district programs?


      • Steve M says:


        Sorry to say, but with NCLB, the tightening of teacher credentialing (e.g. need degree related to subject, California’s 3-year BTSA requirements after getting a credential and landing a job, CLAD requirements for all new credentials) and the enormous costs of maintaing such programs, I don’t think there are many alternatives to going the traditional route (through a college credentialing program) unless you really will do SpecEd.

        Unless, of course, you begin teaching at a private school. They do not have to have credentials. But, a warning there: a very high percentage of private school teachers move on to the public sector after a number of years. For some reason they like to have job security and health benefits.

      • E. Rat says:

        San Francisco runs a teaching residency with USF and Stanford that trains candidates for regular education classrooms. The focus is on training teachers to be successful at high-needs schools.

  8. Pingback: Remainders: State and city kindergarten rules seem to conflict | GothamSchools

  9. Steve M says:


    I would suggest that you do a study of the New Technology Foundation (NTF), an organization that has been supported by Bill Gates’ monies for over a decade. The NTF uses technology in many novel ways, and has had some success, but I would give it poor marks overall.

    Even though I am generally a critic of the NTF (I taught at an NTF affiliated school for two years), I would say that their project-based methods and utilization of technology does actually produce gains in ELA and (possibly) Social Science. How can it not, when every teacher is spending 25% of their time reinforcing what amounts to ELA standards?

    What I have seen from the NTF is a historic disregard for strong math and science instruction and also some of the negative characteristics of a charter network: cult mentality; lack of transparency; refusal to take suggestions; unwillingness to change methodologies that are not working for math and science instruction. Indeed, many of the newer NTF schools are, themselves, charters.

  10. Dominick D'Angelo says:

    As the principal of IS228 in Brooklyn, I can tell you that School of One was one of the best things that’s ever happened to our school community. I find it odd that you chose not to include the most recent results from 2011-12, which showed the % of Level 1 students cut by 1/3 and the # of Level 3s and 4s up by almost 10%. Our teachers will never go back to the traditional model…this is a far better way to differentiate instruction that any school would be lucky to have.

    By the way, I have no idea what this program looked like 2 1/2 years ago when you visited in 2010, but I do know that it’s evolved immeasurably in the 2 years we’ve had it in our school. The content has gotten much better over time — and their new model (Teach to One) actually includes Mathalicious as part of it (along with a host of new components design to build critical thinking skills). Just like you probably improved quite a bit as a teacher in the first 2 1/2 years of your career, so too has School of One.

    I’m not saying School of One is perfect. Neither is Everyday Mathematics or any other math program in schools today. But the last thing our kids need are people trying to tear down new, promising ideas in their infancy just because they are different from the way we’ve always done it and not proven the day they come to light.

  11. MV says:

    While you focused on the technology, I was struck by this stat: six teachers for about 100 students. That works out to less than 17 students per teacher. Differentiation becomes a lot easier at these levels; that’s one of the purposes of smaller class sizes. How exactly would you determine if the technology actually had any effect by itself from the data?

  12. dcchillin says:

    Not sure if you were aware, Gary, but School of One is coming to at least one DCPS middle school in fall 2012.

  13. Pingback: Teach To None | Gary Rubinstein's Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s