Amongst other topics, Johnson, Adams Becker, Estrada and Freeman (2014) discusses key trends accelerating higher education technology adoption as well as important developments in educational technology for higher education. This blog looks at one of the aforementioned key trends (i.e., the rise of data-driven learning and assessment) and one of the aforementioned important developments in educational technology (i.e., learning analytics). The trend of the rise of data-driven learning and assessment is envisaged to drive changes in higher education within three to five years while the time-to-adoption horizon for the learning analytics science and technology is one year or less.
Background
Just as data mining and data analytics techniques and tools have been applied in the business domain for analyzing and predicting customer/consumer behavior (including preferences, spending, consumption rates, etc.), these techniques and tools are now being applied in the education domain. Online learners, especially, produce data; this data is suitable or amenable to the application of data mining and data analytics (including data/statistical analysis) techniques and tools – applying these techniques and tools creates the discipline so-called learning analytics. Johnson et. al. (2014) describes learning analytics as an application of big data analytics in the education domain. The interests in as well as the applications and implementations of big data analytics techniques and tools will increase as even greatly more education data gets generated. Privacy and ethics concerns and issues will need to be better addressed so that the effects of this potential mitigating factor to the progress and adoption of this approach will alleviated.
Figure 1 below facilitates a description of the discipline of learning analytics, namely: It can be construed to be a multidisciplinary field, which brings together expertise and knowledge from a range of different fields, it can be construed to be a method and it incorporates, considers or is constrained by a number of inputs and constraints.
The science, technology and applications
The education data that gets produced comprise new data sources and this data is used to drive a couple of areas in education, namely: Learning and assessment; more specifically, the couple of applications in education, for which this data is used, include personalizing the learning experience and conducting performance measurement or evaluation. Other areas of application of this data and the attendant learning analytics include decision-making and policy-making. This data is in very large amounts – as is typical with traditional big data analytics applications – and the techniques and tools employed are those that are, commensurately, suited for handling such huge volumes of data.
Data analytics is used in experiments and demonstration projects to examine ways by which online education data may be used to modify learning strategies, content, processes and student engagement levels and methods. Online education data can be collected, filtered and, therefore, possibly monitored real-time, in order to determine the progress that students undergo. Johnson et. al. (2014) describes learning analytics as a bourgeoning, emerging discipline and, as this discipline matures, there is the promise of continuous improvement of the entire learning activity, both to students (the learners) and the teachers (the instructors).
Adaptive learning software, whose development is currently underway, shows promise in being employed to support this approach and sophisticated web-tracking tools are used to capture the education data, which is fed into the learning analytics process (Johnson et. al., 2014); novel visualization tools and techniques and analytical reports, which are being developed, will usefully and aptly provide guiding empirical evidence for education administration and governing bodies; these bodies can use these modern, innovative and effective means of communication as well as sources of reliable information to plan the entire education delivered.
The adoption trends of learning analytics show that the number of educational institutions adopting this approach is increasing and the US Department of Education has considerably studied the approach; there are many learning analytics projects currently ongoing in educational institutions. An important and a typical testbed for conducting research on learning analytics is the Massive Open Online Course (MOOC) paradigm. Figure 2 below illustrates how learning analytics is indeed data-driven learning and assessment.
Benefits and impacting forces
Benefits of this approach include analyzing, understanding and predicting which students are likely to fail, when they are likely to fail, so that timely assistance can be provided to them, as well as improving student retention; the key benefit includes personalizing the learning experience – rather than dishing out education content/package that is one-size-fits-all, a responsive and flexible system enables the tailoring/customization/adaptation of the education package/content so that it meets the unique needs and requirements of each individual student/learner.
What forces will either drive or impede the rise of data-driven learning and assessment and learning analytics? There is a range of benefits of data analytics, applied in education (i.e., learning analytics); furthermore, there will be certain sociotechnical issues, such as privacy, security and ethics, which will also accompany the emergence and adoption of learning analytics – these benefits and sociotechnical issues will no doubt be forces that will, respectively, drive or impede the key trend and the important development in educational technology, which have been described and discussed in this blog.
References
Carnegie Mellon University (2012). Open Learning Initiative. Retrieved from http://oli.cmu.edu/get-to-know-oli/learn-more-about-oli/
Greller, W. (2011). Learning Analytics framework. Retrieved from http://www.greller.eu/wordpress/?p=1467
Johnson, L., Adams Becker, S., Estrada, V., Freeman, A. (2014). NMC Horizon Report: 2014 Higher Education Edition. Retrieved from www.nmc.org/pdf/2014-nmc-horizon-report-he-EN.pdf
Background
Just as data mining and data analytics techniques and tools have been applied in the business domain for analyzing and predicting customer/consumer behavior (including preferences, spending, consumption rates, etc.), these techniques and tools are now being applied in the education domain. Online learners, especially, produce data; this data is suitable or amenable to the application of data mining and data analytics (including data/statistical analysis) techniques and tools – applying these techniques and tools creates the discipline so-called learning analytics. Johnson et. al. (2014) describes learning analytics as an application of big data analytics in the education domain. The interests in as well as the applications and implementations of big data analytics techniques and tools will increase as even greatly more education data gets generated. Privacy and ethics concerns and issues will need to be better addressed so that the effects of this potential mitigating factor to the progress and adoption of this approach will alleviated.
Figure 1 below facilitates a description of the discipline of learning analytics, namely: It can be construed to be a multidisciplinary field, which brings together expertise and knowledge from a range of different fields, it can be construed to be a method and it incorporates, considers or is constrained by a number of inputs and constraints.
Source: Greller (2011)
Figure 1: Learning Analytics as a Multidisciplinary Field and a Method
The education data that gets produced comprise new data sources and this data is used to drive a couple of areas in education, namely: Learning and assessment; more specifically, the couple of applications in education, for which this data is used, include personalizing the learning experience and conducting performance measurement or evaluation. Other areas of application of this data and the attendant learning analytics include decision-making and policy-making. This data is in very large amounts – as is typical with traditional big data analytics applications – and the techniques and tools employed are those that are, commensurately, suited for handling such huge volumes of data.
Data analytics is used in experiments and demonstration projects to examine ways by which online education data may be used to modify learning strategies, content, processes and student engagement levels and methods. Online education data can be collected, filtered and, therefore, possibly monitored real-time, in order to determine the progress that students undergo. Johnson et. al. (2014) describes learning analytics as a bourgeoning, emerging discipline and, as this discipline matures, there is the promise of continuous improvement of the entire learning activity, both to students (the learners) and the teachers (the instructors).
Adaptive learning software, whose development is currently underway, shows promise in being employed to support this approach and sophisticated web-tracking tools are used to capture the education data, which is fed into the learning analytics process (Johnson et. al., 2014); novel visualization tools and techniques and analytical reports, which are being developed, will usefully and aptly provide guiding empirical evidence for education administration and governing bodies; these bodies can use these modern, innovative and effective means of communication as well as sources of reliable information to plan the entire education delivered.
The adoption trends of learning analytics show that the number of educational institutions adopting this approach is increasing and the US Department of Education has considerably studied the approach; there are many learning analytics projects currently ongoing in educational institutions. An important and a typical testbed for conducting research on learning analytics is the Massive Open Online Course (MOOC) paradigm. Figure 2 below illustrates how learning analytics is indeed data-driven learning and assessment.
Source: Carnegie Mellon University (2012)
Figure 2: Data-driven Learning and Assessment
Benefits and impacting forces
Benefits of this approach include analyzing, understanding and predicting which students are likely to fail, when they are likely to fail, so that timely assistance can be provided to them, as well as improving student retention; the key benefit includes personalizing the learning experience – rather than dishing out education content/package that is one-size-fits-all, a responsive and flexible system enables the tailoring/customization/adaptation of the education package/content so that it meets the unique needs and requirements of each individual student/learner.
What forces will either drive or impede the rise of data-driven learning and assessment and learning analytics? There is a range of benefits of data analytics, applied in education (i.e., learning analytics); furthermore, there will be certain sociotechnical issues, such as privacy, security and ethics, which will also accompany the emergence and adoption of learning analytics – these benefits and sociotechnical issues will no doubt be forces that will, respectively, drive or impede the key trend and the important development in educational technology, which have been described and discussed in this blog.
References
Carnegie Mellon University (2012). Open Learning Initiative. Retrieved from http://oli.cmu.edu/get-to-know-oli/learn-more-about-oli/
Greller, W. (2011). Learning Analytics framework. Retrieved from http://www.greller.eu/wordpress/?p=1467
Johnson, L., Adams Becker, S., Estrada, V., Freeman, A. (2014). NMC Horizon Report: 2014 Higher Education Edition. Retrieved from www.nmc.org/pdf/2014-nmc-horizon-report-he-EN.pdf
