or: what to watch out for in after school and summer programs
In writing this post I do not mean to say that there aren't some great programs out there - there are. I also know that most of the programs I'm aware are being developed by people who are really trying to do some good.
Since I wrote a post on my problem with drop in curricula on March 4 (http://cestlaz.github.io/2015/03/04/expedient-vs-good.html) I've been meaning to follow up with some of my thoughts on after school and summer programs.
This itch was further spurred when I read this post on avc.com: http://avc.com/2015/03/numbers-can-ruin-a-good-story/. It talks about startups - you can have a good story and raise some funds but then when numbers, that is, results start to come in, that story might not look so good.
The world of CS Education summer and after school programs are full of "good stories." My favorite is "your kid will learn how to build an iPhone app in 4 weeks!!!" If it were that easy, we wouldn't have a tech talent shortage and my grads wouldn't be in such high demand.
The truth is, it's very easy to get by in CS Education on nothing more than a good story. You never actually have to show real results. Most of the programs out there follow this pattern. I don't think it's intentional. It's just that most people running K12 CS Ed programs aren't educators and frequently they aren't CS people. What they are concerned is scaling so that they can get to as many kids as possible. They aren't dealing with next steps. Besides, to them, it appears they're doing great work - they buy into their own hype.
What's the problem?
Simple - learning takes time.
How do you set up the story?
Start by promising an awesome program - an awesome project after two or four weeks. With today's tools it's easy to make a mock up. Either you use simplified tools or you can use a lot of boilerplate code with pre-built components. These aren't necessarily bad in and of themselves, just don't pretend you're teaching the real deal.
It's easy to pull this off since there's no real follow up. Kids are then sent back to their high schools or off to college and results aren't tracked.
Also, make sure you recruit some high performing kids. I know a number of programs that have taken in my Stuy kids even though they already know more than the program has to offer. This guarantees that your program will look good since these kids will produce great projects.
I recall about a year and a half ago reading an article. It highlighted a youngster who went through one of these summer programs. A bright kid in a well regarded, well hyped, well funded program. I've don't think the article said so, but I heard many of this programs top people repeatedly say "our program better prepares our kids for the AP Computer Science exam than the AP course itself does." Hmmm. Well, in the article, the kid learned all these wonderful things from the program and then went back to high school.
A short time after, according to the piece, the kid was struggling in AP Computer Science and the article quickly threw the teacher under the bus. Now, I have no idea if the teacher was any good but if your program better prepares a kid for the exam than the class, there's not way the kid should have been struggling - unless, of course, it was all a story.
The truth is, that the designers and implementers of many of these programs don't know education and often know little CS. I can't tell you the number of kids that have come into my classes and programs proclaiming to be master programmers as a result of these programs and it turns out all they know is how to build a limited drag and drop prototype or how to tweak a pre-built template.
What's the harm? In my classes and programs, we can educate the kids but if this is in college with less support, self doubt will creep in and more often than not the kid will drop tech. I've seen it happen many a time.
That brings me to the next story – the glue programs. I also see a lot of this. Programs that teach the very rudiments of programming and then use pre-built libraries and packages to make really impressive looking results. They look great, but there's not a lot of substance under the hood. I remember seeing a number of these "finished products" and each time noting that the student contributed maybe 20 to 40 lines of code and they were mostly tweaks to the web site samples.
The final story is less of a story and more legit in terms of teaching but it still largely a story. That's when a program tries to teach too much to quickly. I've seen one and two week programs that promise the word. The kids put together some impressive stuff but since it's so much so fast, facilitated by lots of coaching, there's little retention and questionable understanding.
I was talking to a friend a few months ago. She worked with a number of high achieving kids teaching them to code. All seemed great. They got the kids back together months later for a follow up. I asked "how was the retention?" The reply "that's a real problem."
the truth is learning takes time. There's only so much you can do in an after-school or in a summer and it's not always flashy.
When I designed Google CAPE NYC for the summer of 2010 I knew we couldn't do everything in a month. What we did do is give the kids the fundamentals and a path to follow to learn more. They did neat projects but they weren't the over the top stuff you see other programs produce (as an aside, one of my grads noted - "If it looks too good to be true, it probably isn't true."). While we didn't formally track the kids, 3/4 of the kids we were still in touch with last year were still doing computer science at a second year college level so I know what we did worked but I haven't seen that type follow up from any other program.
What I do know is that we've got the best teachers and the best program and that our kids might not make something quite so flashy after three or four weeks as some of the other programs, our kids are set up for long term success.
So, if you're looking for a great program with great teachers and a solid plan, check out CSTUY. If you want to create a story, just follow these steps:
- Recruit some high achieving kids so you'll have some stories
- Teach a framework or simplified tool
- Coaching, coaching, coaching.
- Make sure they have fun so you get good reviews