Why I Recommend Flatiron School
I’ve been teaching computer science and running schools for 14 years. Since 2012 I’ve created and run three developer training programs (Hungry Academy, gSchool/Galvanize, and Turing), running both the business and academics for each while training hundreds of happy alumni. There is probably no single person with more expertise on the construction and operation of bootcamp-type programs in the world.
With those bonafides aside, let me say this: Flatiron School, both in-person and online, are the only programs I recommend other than my own.
The news of a $375K settlement with the New York Attorney General’s office does nothing to change that. Let’s unpack the two key complaints: state licensure and aggregated outcomes reporting.
State Licensure Is a “Nice-to-Have”
State licensure is damn near irrelevant. It makes for a good pull-quote: “Flatiron has operated for years without a license!” I’ve been through many of these licensure processes, both in K-12 and adult workforce training. They don’t really matter.
Our license in Colorado, for instance, amounts to submitting outlines of our curriculum, basic staff credentials, and setting up a surety bond to provide refunds to students if we unexpectedly shut down. It’s a pile of paperwork. The organization that oversees us, the Colorado Division of Private Occupational Schools, is tasked with consumer protection, not academic quality. Their key job is to be a landing place for student complaints, not to ensure that we know what we’re doing. Along with Turing, they oversee programs that teach yoga instructors, cell tower repair workers, and a school for butlers. They have no interest or capability to determine if we know what we’re doing.
Most developer training programs started without licensure. Many of the California schools even tried to argue with their state legislature that they should be exempt from licensure. If you did an audit of the 400+ programs listed on CourseReport I believe that you’d find over half of them are not licensed by the states they operate in. Even some of the well-known/well-established programs are operating without a state license.
Let’s open the can of worms for online training. Is Bloc.io approved to deliver services to students in all 50 states? Surely not — it would take a department of compliance people to maintain those licenses. They’ll tell you that their online program is as good as or better than an in-person program (in piles of bullshit like this), but surely do not bear the state-level compliance expectations. It’s just not feasible. (BTW, please try to find their outcomes report!)
Should Flatiron get right with the state? Yes. Should they have done it in the past? Probably. Does it make a difference to the students? Unlikely.
Note that the NY AG article doesn’t make mention of student dissatisfaction. Surely there are some grouchy, pissed off people. We all have them. But Flatiron is not ripping off their students by running outside the oversight of state accreditation. The licensure portion of the complaint means nothing.
Outcomes Reporting
To punish Flatiron for the precision of their outcomes reporting is ludicrous. I have many times credited them with launching the first and most important in-depth data report for these types of programs. Efforts like Turing’s data report, NESTA (RIP), and CIRR all draw their inspiration from Flatiron’s Student Outcomes Reports.
When I first heard of these issues with the AG, I wondered “how can that be?” The issue is not with the report, data, definitions or methodology. The issue is that data may have been used in misleading ways in aggregated statistics.
Student tracking and reporting has been an area of focus for me over the last decade and, let me tell you, it’s complicated. Not just because the data is difficult to gather in accurate, authentic ways, but because even the simple definitions are hard to pin down.
Look at graduation rate. Years ago IB Times published “Code Bootcamps Fail Obama” and it was clear that our reported graduation rate was way different than others. How? Why? You just take the number of students who graduate and divide by the number of students who enroll, right? Sure, but how do you account for…
- Students who dropout within the first 10% of the program? My friend Dave Hoover referred to it as “in-and-out the revolving door” — they never actually come inside. Typical four-year colleges will not count these students as having ever enrolled. Should a bootcamp count them in that denominator?
- We’re a job-training program. The goal is not to graduate, it’s to get a job. About 10% of our students get a job before graduation and choose to start it before finishing the program. We love to see them finish, but as their savings accounts dwindle we push them to make the right choice for their life — our stats be damned. Those 10% of successful students are “drop outs” despite the fact that they got exactly what they came for.
- The typical cutoff length in higher-ed is graduation within 150% of expected program length. While our program is expected to last 27 weeks, many students repeat a quarter and/or take time off in the middle. There are some folks who are with us for 40+ weeks, fight tooth and nail for graduation, and finally reach it. Then they’re counted as dropouts because they stayed over 150%.
These cases aren’t every student. But the student body size is so small that just a few of these scenarios playing out can dramatically change the aggregate numbers. It turns out that even the simple stats get complicated. Dealing with inherently difficult ones like “salary” becomes even cloudier.
Should Flatiron have better explained how they calculated aggregate stats? Sure. But there’s no complaint or objection to their detailed outcomes report. It reminds me of food labeling. Their USDA nutrition label was 100% correct. Their fancy cover photo showed a good-case interpretation. That’s typical in any industry.
We’re working with a group of training providers to develop a set of reporting guidelines via the Council for Integrity in Results Reporting. It’s tough to draw clear lines that result in fair, meaningful data for potential students, funders, and regulators. But, as an industry, we’re pushing to do better.
The existing data reports from Flatiron, Turing, and CIRR member schools are the most robust and detailed outcomes reporting anywhere in modern education. You will not find any equivalent in K-12 or higher education.
The Outrage Doesn’t Exist
The Attorney General’s mandate is to protect the people and consumers of New York. Where is their outrage? Check out the reviews of Flatiron on Course Report or SwitchUp. What do you find? Some justified critiques overshadowed by dozens and dozens of people explaining how the program changed their life. You’ll find nearly the same on Turing’s page.
Haters will hear this AG news and say “See! This is all a fraud!” They’re outsiders looking for any chance of a flaw. If places like Flatiron are so wrong, where are the people who have been wronged?
There Is No Substitute for Giving a Fuck
I’ve known Avi Flombaum and Adam Enbar at Flatiron for years. The amazing Peter Bell joined the online program last year. Rebecca Rombom runs the most respected careers team in the industry.
Collectively we’re some five years into figuring out this accelerated developer training industry. The growth and success won’t be won by the best technology choices, the biggest budgets, or the folks who gather the most attention. Sustainable progress happens when people give a fuck.
Our instructional methodologies, techniques, and choices differ. But that team of people cares about their students and does the hard work of learning and improving. That’s how they show up. That’s how they honor the trust of their students. That’s why I have and will continue to recommend students trust in Flatiron.
Jeff Casimir is the Executive Director at the Turing School where he inexplicably writes long blog posts evangelizing the competition.
