Abstract
This chapter presents a comprehensive discussion of educational data mining and its potential for educational research. The origins of data mining and the emergence of educational data mining are discussed. The variety of data generated in education (e.g., text, speech, performance, etc.) are described and the challenges of mining these data for useful information are identified. Techniques for mining these data are discussed. Software used to mine these data are noted and issues of theory and ethics are considered. Examples from published literature are cited throughout the chapter and recommendations for educational researchers are offered.
TopBackground
Data mining emerged in the 1990s in response to the proliferation of mega datasets as corporations, government agencies and other entities increasingly moved operational records and other data to electronic databases (Gandomi & Haider, 2015). These so-called “Big Data” were often linked allowing, for example, businesses to access a prospective customer’s entire history of purchases, employment, and the like (Chopra, Gautreau, Khan, Mirsafian, & Golab, 2018). Similarly, the emergence of social media platforms such as Facebook, normalization of email use, and other developments gave potential employers a history of interactions of prospective employers beyond anything possible in the past (Injadat, Salo, & Nassif, 2016). Finally, the ubiquitous mobile devices that are now part of everyday life provide an uninterrupted stream of data about social interactions and online activity for billions of people (Cheng, Fang, Hong, & Yang, 2017). Data mining grew out of the recognition that these databases provides invaluable information which can be meaningfully extracted, condensed and presented. As a discipline, data mining is concerned with the complexities of this process (Koedinger, D’Mello, McLaughlin, Pardos, & Rose, 2015; Wang et al., 2018).
Key Terms in this Chapter
Educational Data Mining: The application of data mining techniques and methods to educational data for management and research purposes.
Data Mining: The techniques used to acquire, process, analyze and report meaningful results from large datasets.
Big Data: A term used to refer to the massive datasets generated in the digital age. Both the volume and speed at which data are generated is far greater than in the past and requires powerful computing technologies.
Machine Learning: A discipline focused on the development and evaluation of algorithms that permit computers to use patterns, trends, and associations in data to perform tasks without being programmed by a human.
Data Science: A relatively new term applied to an interdisciplinary field of study focused on methods for collecting, maintaining, processing, analyzing and presenting results from large datasets.