Can Data Analytics Improve Chinuch?
We dedicate enormous resources for our kids' chinuch, but are some schools ignoring a powerful tool for becoming better at what they do?
Through my 20 years of teaching, administrating, and board-of-directoring in Torah schools, one theme became gradually more obvious: while most of the people involved were dedicated and talented, no one really had full conceptual control over the process. We were all reasonably confident about our long-term goals (the terminal objectives), but we weren’t nearly so sure about what we should be doing to get there (the enabling objectives). And all while one crisis after another happily kept us off-track and off-balance. On a good day, it was controlled chaos.
I believe that this lack of clarity about the process was partly the result of our inability to really understand the impact that today’s choices might have on tomorrow’s outcomes.
For instance, business decisions (“Can we afford this program?”, or “Can we afford to split a large class into two?”) are often made without enough solid information. Sure, administrators might know the current budget numbers and even revenue projections for the coming six or 12 months. But they almost certainly won’t be able to tell you what kind of long-term impact similar decisions had had in the past. How many families transferred their kids to other schools when class sizes hit a certain threshold? Has there been a measurable historical correlation between class size, academic performance, and school reputation (and, by extension, recruiting)?
Similarly, remedial programs designed to help struggling students are critical. But are you sure they actually work? Or, to ask the question more intelligently, are you sure they work well enough to justify the financial and social costs? Is there anyone actually thinking about and measuring outcomes?
These are just the kinds of questions that data analytics can answer.
But first, a word of sober caution. Just because cool data tools exist doesn’t mean they’ll necessarily be productive. And just because they’re productive doesn’t mean they’re necessarily appropriate. Data, after all, can be misinterpreted (inspiring wasteful and even dangerous policy decisions) or misused (disclosing sensitive private information). Choose carefully.
Ok. So, in the context of underfunded and, often, technology-weak chinuch institutions, what useful data is likely to exist? Well it can be as simple as spreadsheets containing enrollment and attendance records. You might be surprised how many powerful insights can be mined from even such simple datasets. Adding tuition scholarship information (while carefully protecting privacy and confidentiality) can significantly increase value.
Beyond those demographics, it shouldn’t be too difficult to convert assessment scores to a format that’s analytics-ready. And if you haven’t been keeping regular records yet, then now’s a great time to start.
Of course, there are already Torah schools deeply engaged in just this kind of analysis. But if I were to walk into a less engaged school wearing my data analyst hat, here’s what I might do:
Build an inventory of existing demographic and educational assessment data, including report card, assessment, enrollment, and attendance records
Identify the school’s educational and institutional goals
Identify major historical policy and demographic changes
Propose use-cases and protocols for collecting new data sets
Ingest existing data and integrate data from multiple sources
Interpret data using, where relevant, multivariate analysis
Measure conclusions against official goals (through collaboration with local domain experts)
Report findings
The thing I love most about working with data is the sense of adventure. Sure, you might come in with some preconceptions, but you never really know what the data can show you. Sometimes - especially when you do it right - you discover something truly new and shocking. And that’s when you can hope see the big payoff for both your students and your teachers.