Author: Renee Ho
Last month, the New York Times published “How Not to Drown in Numbers” an op-ed by two data scientists at Facebook and Google.
The argument is this: Big Data is amazing but it’s simply not enough.
They write, “The things we measure are never exactly what we care about. Just trying to get a single, easy-to-measure number higher and higher (or lower and lower) doesn’t actually help us make the right choice. For this reason, the question isn’t ‘What did I measure?’ but ‘What did I miss?’ ”
In a world that is extremely complicated, small data– surveys and human judgment– can find the holes in the Big Data and provide meaningful insight. For Facebook, Big Data (likes, clicks, comments) is supplemented by small data (“Do you want to see this post in your News Feed?”) and contextualized (“Why?”).
Feedback of this kind, in essence, isn’t that different from the Net Promoter Score. (More on NPS later.)
Supplemented is a key word here. Big data and small data are not mutually exclusive in our quest to have greater social impact in the world. The article highlights a study (the Measures of Effective Teaching study) by the Bill and Melinda Gates Foundation that finds a composite score, using both Big and small data measures, ultimately yields the best results for predicting student outcomes. The composite score uses classroom observation, student perception surveys, and student achievement gains.
Specifically, the MET project report found that “Each measure adds value. Classroom observations provide rich feedback on practice. Student perception surveys provide a reliable indicator of the learning environment and give a voice to the intended beneficiaries of instruction. Student learning gains (adjusted to account for difference among students) can help identify groups of teachers who, by virtue of their instruction, are helping students learn more.”
