Step Up To Analysis — Recipes, Analytics and Data Science at Learnmetrics
Learnmetrics is a platform for data aggregation, analysis, and activation. In previous blog series, we have detailed how the first step of data warehousing, ingestion and organization into a standardized and coherent format, leads to quick wins in terms of getting raw data to the correct stakeholders, in a single location, and in real-time.
However, we believe the true value of Learnmetrics lies in closing the feedback loop with actionable insights backed by data. We want to make sure school administrators, teachers and parents have the information they need to answer relevant but otherwise inscrutable questions about their students in real time. How can we identify at-risk students and be certain an RTI initiative translates to academic improvements? How can we justify the cost of an ed-tech reading app and determine if it actually helps students improve reading comprehension? Unless the data loop is closed, they can’t. This is a missed opportunity.
But such a task can quickly become overwhelming. Even assuming access to a well-implemented data solution, data science in education is generally limited to large, well-resourced districts and networks with dedicated teams of data scientists and engineers.
We have spent over five years collaborating with our partner schools to develop an end-to-end data platform that provides educators insights on the metrics driving student success at their institution, and we want to use this summer blog series to reflect on what has worked and share what’s coming next. Each blog will cover a specific recipe or analytic technique, the real situations we’ve used it for, and what we’ve found to be the best way to visualize and interpret the results for stakeholders. There are some great instructional resources out there, but we’ll include any formal underpinnings in the intro, to be accessible to beginners and anyone aspiring to increase their educational data literacy. We’ll start off slow in the first blog, with techniques for grouping and identifying student tiers given a set of indicators.
What tough questions about your student population would you love to see answered? Is there something you’d like to see us include? Reach out! Tweet us @Learnmetrics and we’ll be sure to fold it into the series.
Brian Dellabetta
Chief Data Scientist
brian@learnmetrics.com
