Modern classrooms can and will produce incredible volumes of rich data. Such richness, when paired with Machine Learning and Professional Development, can bring data driven decision making to the classroom and drive student success. We present our work doing so and its impacts throughout the school.
Since teachers systematically decided to record their students’ results and analyze them, data driven decisions making (DDDM) has been in the classroom. The past decades have seen incredibly advances in technology, which has placed precise and varied data collection sensors directly in the hands of not just students, but also educational professionals. As a result the opportunity to collect rich data capturing all interactions from every point in a student’s education is not just a sci-fi dream, but a reality.
Big data’s arrival to the educational space has left many schools data rich with impossible volumes and types of data, yet insight poor. This problem of being data rich and insight poor is especially poignant for fully online schools. Without an effective DDDM, online schools limbo in the ironic position of simultaneously having enough data to overcome the loss of signals and intuitions derived from in-person contact while suffering from said loss of in-person contact.
However, implementing DDDM in any data-rich classrooms is, at its heart, two-dimensional. First, it is deeply technical. The explosion of data types means that at some point, each new data type makes the work of digesting and synthesizing data more impossible.
Advancements in the field of machine learning (ML) combined with data mining (DM) have made the synthesis of highly dimension, high volume data possible. Machine learning and data mining are already being used in education in variety of ways from predicting student dropouts to characterizing student engagement. Incorporating ML and DM into the technology supporting DDDM helps educators consume data and arrive at insights as easily as possible
The other side of bringing DDDM to data-rich classrooms is deeply human. Even with the best ML and DM driven analytics, using such data is not intuitive and can leave educational professionals as equally in the dark as they were without any data analytics. A fully comprehensive DDDM support system needs constant and pervasive professional development (PD).
Rather than viewing DDDM as a two-step problem, this work views bringing DDDM to the classroom as an ecosystem of technology and continuous PD. The project begins with the development of a DDDM system which primarily includes a machine learning driven dashboard and teacher education and support system. All elements of the DDDM system were developed iteratively with constant teacher feedback. Once the individual elements of the system reached certain levels of metric success, the full DDDM system was rolled out fully at one online high school throughout the 2017 to 2018 academic year.
Data will collected over 9 class sessions, encompassing over 50 teachers, encompassing around 450 courses and close to 6,000 students. During the evaluation period, all courses will have student academic data and attendance data will be collected. After the evaluation period, this data will be compared against the historical performance at the school from two years prior and controlled for course performance and teacher effects. Data collection and analysis of findings are expected to be completed over the 2017-2018 academic year