The Department of Education’s (ED) Office of Educational Technology today released a draft issue brief — Enhancing Teaching and Learning Through Educational Data Mining and Learning Analytics — representing the results of a months-long discourse among 8 academic and 15 industrial data mining and learning analytics experts conducted by SRI International. The brief, inspired by ED’s 2010 National Educational Technology Plan (NETP), articulates the challenges and opportunities of Big Data in improving student outcomes and overall productivity of K-2 education systems. It focuses on three key research areas — educational data mining, learning analytics, and visual data analytics — and offers a set of corresponding recommendations, categorized by various stakeholders. ED is now seeking public comment on the draft.
Commerce, entertainment, and social life are amplified more and more across the Web and, as a result, the amount of data generated is skyrocketing. Commercial entities are harvesting this data stream to provide personalized advertisements. Public discourse is trending toward questions such as “What data am I creating, where is it going, and what are we getting from it?”
Big data, it seems, is everywhere — even in education. Researchers and developers of online learning systems, intelligent tutoring systems, virtual labs, simulations, games and learning management systems are exploring ways to better understand and use data from learners’ activities online to improve teaching and learning.
The Office of Educational Technology at the U.S. Department of Education asked SRI to talk to industry experts and convene a panel of researchers to understand the state of the art, the state of the practice, and the emerging field of learning analytics and educational data mining.
We did our best to cover learning at all levels, from early learning to adult; we tried to tease apart educational data mining from learning analytics, and we drew on industry applications to understand what was possible — and promising — in education. Along the way, we also encountered challenges such as, who owns the data? How can data mash-ups support improved understanding? Who does this work and what is needed to succeed? And we present those along with recommendations for work going forward — not only how to collect, analyze, and visualize data, but also how to help people become smarter consumers of data and how to ensure integrity regarding privacy and ethics issues.
We’ve tried to cook the technical into something palatable without watering it down and we think the report will be interesting to many stakeholders. We’re interested in hearing what you think of the result.
We welcome your input as we continue the dialog on what is likely to be a game-changing approach to providing today’s learners with more personalized and effective learning opportunities.
Here’s one defining passage from the draft brief:
Using data for making decisions is not new; companies use complex computations on customer data for business intelligence or analytics. Business intelligence techniques can discern historical patterns and trends from data and can create models that predict future trends and patterns. Analytics, broadly defined, comprises applied techniques from computer science, mathematics, and statistics for extracting usable information from very large datasets.
An early example of using data to explore online behavior is Web analytics using tools that log and report Web page visits, countries or domains where the visit was from, and the links that were clicked through. Web analytics are still used to understand and improve how people use the Web, but companies have now developed more sophisticated techniques to track more complex user interactions with their websites. Examples of such tracking include changes in buying habits in response to disruptive technology (e.g., e-readers), most-highlighted passages in e-books, browsing history for predicting likely Web pages of interest, any changes in game players’ habits over time. Across the Web, social actions such as bookmarking to social sites, posting to Twitter or blogs, and commenting on stories can be tracked and analyzed.
Analyzing these new logged events requires new techniques to work with unstructured text and image data, data from multiple sources, and vast amounts of data (“big data”). Manyika et al. (2011) defined big data as “…datasets whose size is beyond the ability of typical database software tools to capture, store, manage, and analyze.” Big data captured from users’ online behaviors enables algorithms to infer the users’ knowledge, intentions, and interests and to create models for predicting future behavior and interest.
Research on machine learning has yielded techniques for knowledge discovery or data mining that discover novel and potentially useful information in large amounts of unstructured data. These techniques find patterns in data and then build predictive models that probabilistically predict an outcome. Applications of these models can then be used in computing analytics over large datasets.
Two areas now developing that are specific to the use of big data in education are educational data mining and learning analytics. Although there is no hard and fast distinction between these two fields, they have had somewhat different research histories and are developing as distinct research areas. Generally, educational data mining is looking for new patterns in data and developing new algorithms and/or new models, while learning analytics is applying known predictive models in instructional systems. We discuss each in turn…
And on a related note, be sure to review the results of a recent CCC visioning activity on learning technology.
(Contributed by Erwin Gianchandani, CCC Director)