For the last two years, I’ve been a data engineer at Schoolzilla, and currently spend my days helping to define the technical vision for our achievement products. Prior to joining the team here, I worked in Research and Evaluation at the KIPP Foundation across the Bay in San Francisco, supporting KIPP schools around the country to turn data into insight, and insight into action. My experiences at both KIPP and Schoolzilla have led me to think a lot about how school systems can get the most out of their work with data.
Here are my top tips for school systems looking to become more “data-ready”:
- Start with goals.
Data is a means to an end, not an end in itself. Your school or district won’t get much out of stronger data practices if it isn’t already goal-driven. Do your teachers, parents, and students have a growth mindset? Are SMART goals being set on a regular basis organization wide.
- Lead with questions, not metrics.
“Are our students climbing the mountain to and through college?” It’s far easier to hook someone with a question that they care about than by showing them a table or chart that reports a metric. That was a key learning that led to the formulation of KIPP’s Six Essential Questions.If you ask a good question, then you’ll likely have someone asking you for the data to answer that question. If you start with a bag of metrics, you probably see eyes glaze over. Also, a thoughtful, mission-driven question will outlast a well-designed metric. Metrics may change as the specific thing to measure—or the best way to measure it—evolves as you make progress towards that overarching goal.
- Think about the data you wish you had, rather than the data you already have.
Just because you have data on a particular dimension, doesn’t mean it should be integrated or presented. As an example, data that has very little variability across classrooms or over time isn’t going to be very useful. You may not be able to easily collect or measure the data that would be ideal for your organization, but it’s important to have a clear vision of what that would look like so that you can always be moving toward it.
- Invest in systems that play nice.
Some companies build walled gardens; others make it easy to get your data out. If you want to get the most of your data, you’ll need to have a scalable, automated way to get it out of systems.- Are external student IDs a part of the exports or does the vendor only include a unique student identifier that is used in that product?
– Do the data exports include everything you care about or do they only contain a small subset of the data in the system?
– Are they structured and formatted in a way that will be easy to use?
– Does the product have a CSV export? What about an API or an automated file drop?
- Start narrow in terms of audience and scope.
The dream is all your data in one place, presented in a way that is perfectly tailored to each stakeholder. The reality is that it’s close to impossible to accomplish that all at once and the path to the “data-driven dream” is smoother if you start by serving a particular audience especially well. Further, you’ll learn things along the way that will move the goalposts anyway. Time is our most important resource and it’s better to cater to one group of stakeholders really well, than all stakeholders somewhat poorly. It’s better to really nail the collection, organization, presentation and use of one domain of data when you’re starting out, rather than trying to shoot the moon.
- Start with goals.