Programs

Iridescent’s experiences with teacher and school evaluation

Iridescent recounts its effort to build a data-driven, crowdsourced system for evaluating schools and teachers in Los Angeles, in partnership with Factual and Density Design, and the lessons learned about complexity, focus, and collaboration before shelving the project.

We like to do things at scale, cheaply, fast and for the long term. So when we started working with a lot of schools in Los Angeles, I figured it would make sense to automate the process of finding lifelong partners.

We worked with Factual (a data company) to automatically collect data on school demographics, test scores, crime and parent involvement. This data was already available on the web. We worked with an information visualization group in Milan, called Density Design to develop the user interface that would help parents and organizations like Iridescent, find the right school. You can use the tool and learn all about it here.

Watch: Urban School Needs Map demo video

After this tool was developed, we realized that we needed more data on the classroom level – i.e how committed a teacher would be to our partnership and we entered the labyrinth of teacher evaluation.

I came in very naively thinking that with my love for technology and efficiency and formulae, I would be able to develop a clean system that could provide data on how good schools and each teacher was. I wanted to give a voice to the most important people in this whole equation – children and parents. Using Factual's crowdsourcing tools, our vision was to provide a wiki base of school data (that was aggregated from the web) on top of which we could collect data for each teacher that was contributed by children and parents. Children and parents would fill out surveys that would evaluate various facets of teaching. In addition they would be able to upload videos as well.

Good teachers would be highlighted and awarded “Master Teachers” badges and would be invited to lead best practices conferences. In addition, we would collect videos of great teachers executing their craft and skills.

This was our vision.

We needed partners to make this happen and we dove into the haystack. What we found was many groups working on this problem in different corners of the room, some talking and learning from each other, but mostly not willing to collaborate. Each had spent significant amount of time and money going down a particular direction and didn't want to change. Each direction varied from the other only by a dot or a dash. I didn't come across even one really disruptive approach.

We were working in the Los Angeles Unified School District and heard much open-minded talk about open data, but in reality, the doors to collaboration and sharing of data were shut or going to be shut very soon.

What I learned was this:

the problem is undoubtedly complex. There are many different school districts with different testing systems, different standards as the baseline. On top of that you have different grades and different skill sets to be mastered at every stage. Teachers can only be compared across narrow ranges.
the problem needs undivided attention. It doesn't need a ton of money – as most school districts think. Technology can be intelligently leveraged to go very far, fast and do great things. But it does need undivided attention and focus. And for Iridescent it just didn't make sense as our undivided focus is on making children curious, courageous and persistent.
the problem needs a fresh perspective. Towards the end we were going to use Facebook as a platform and offer badges and $1000 prizes to teachers who would be brave enough to share their own classroom data and solicit feedback from their children and parents. Nobody is the bad person here. Everyone wants to do a good job. That was our starting assumption. Then move forward building systems that help all groups learn and excel.

Although we stopped working on this project, I am very curious to see if some innovative soul takes it on and paints the canvas brilliantly.