What are Learning Analytics?

Learning analytics promises to harness the power of advances in data mining, interpretation, and modeling to improve understandings of teaching and learning, and to tailor education to individual students more effectively. Still in its early stages, learning analytics responds to calls for accountability on campuses across the country, and leverages the vast amount of data produced by students in day-to-day academic activities. While learning analytics has already been used in admissions and fund-raising efforts on several campuses, “academic analytics” is just beginning to take shape.

Learning analytics refers to the interpretation of a wide range of data produced by and gathered on behalf of students in order to assess academic progress, predict future performance, and spot potential issues. Data are collected from explicit student actions, such as completing assignments and taking exams, and from tacit actions, including online social interactions, extracurricular activities, posts on discussion forums, and other activities that are not directly assessed as part of the student’s educational progress. Analysis models that process and display the data assist faculty members and school personnel in interpretation. The goal of learning analytics is to enable teachers and schools to tailor educational opportunities to each student’s level of need and ability.

At its heart, learning analytics is about analyzing a wealth of information about students in a way that would allow schools to take action. This information can include student profiles within an institution’s database, as well as the interactions of students within course management systems. A long absence from a course’s online activities, for example, can trigger faculty intervention. At its best, however, learning analytics goes much further than this, marrying information from disparate sources to create a far more robust and nuanced profile of students, in turn offering faculty members more insight.

Learning analytics need not simply focus on student performance. It might be used as well to assess curricula, programs, and institutions. It could contribute to existing assessment efforts on a campus, helping provide a deeper analysis, or it might be used to transform pedagogy in a more radical manner. It might also be used by students themselves, creating opportunities for holistic synthesis across both formal and informal learning activities.

While EDUCAUSE has announced a major program in partnership with the Gates Foundation, the Hewlett Foundation, and others that identifies learning analytics as one of five key areas for development, it is still very early and most of the work in this area is conceptual. Learning analytics also faces some challenges. It requires combining data from disparate sources, often in different formats. It also carries with it concerns about student privacy and profiling, as well as the sense that students are being reduced to information and numbers. Indeed, learning analytics to date generally falls within the purview of IT departments. For the information and its use to be more productive within curricula and pedagogy, faculty will need both to understand its technical potential, as well its pedagogical usefulness. These challenges will need to be addressed as the work moves forward. The potential for learning is clear, but the technology to deliver that potential is still very young.

INSTRUCTIONS: Enter your responses to the questions below. This is most easily done by moving your cursor to the end of the last item and pressing RETURN to create a new bullet point. Please include URLs whenever you can (full URLs will automatically be turned into hyperlinks; please type them out rather than using the linking tools in the toolbar).

Please "sign" your contributions by marking with the code of 4 tildes (~) in a row so that we can follow up with you if we need additional information or leads to examples- this produces a signature when the page is updated, like this: - Larry Larry Jan 25, 2011

(1) How might this technology be relevant to the educational sector you know best?

  • I think that Learning Analytics has the potential to deliver increased personalised learning. If online systems can measure and benchmark student responses such as incorrect answers vs correct answers then they are able to branch the next response according to the student need, and so providing remedial exercises or more advanced exercises. So the students who need help can get same, and those who need extension can be accomodated. As well, if systems, through data mining, can identify the potential for at risk students to be identified early in the year, then intervention action such as counselling can take to place to help the learner get back on track. So there will be a shift to more personalised and just in time student support, informed by learning analytics. Of course, governments (around the world I believe) are keen to ensure student retention and attainment is improving and so data mining has a huge potential to contribute to accountability and performance. - garry.putland garry.putland Feb 18, 2011 Very much agreed [- roy.pea roy.pea Feb 27, 2011]
  • Helps people find accelerated pathways through curriculum and move to more complex issues. These instruments can be reduced to assist teachers to help students formulate a Personal Learning Plan. Using LA, the teacher can take the time from grading to help guide the student. LA can serve to advance learner self-awareness. (- michael.lambert michael.lambert Feb 20, 2011)
  • This technology and really, this set of analytic methodologies, has the potential to unlock what we mean by personalized learning. As new methods and approaches to educational data sets are tried, some will be more useful than others as far as revealing robust or deeper learning. - chris.brown chris.brown Feb 21, 2011
  • Info on Personalized Learning. Innovate to Educate: System [Re]Design for Personalized Learning. http://www.siia.net/PLI/presentations.asp - jan.morrison jan.morrison Feb 24, 2011
  • However, it is essential that learning analytics don't just become an enhanced form or summative assessment. There is too strong an emphasis on 'grading' rather than investigating the learning process in order to further advance learning opportunities - judy.oconnell judy.oconnell Feb 26, 2011
  • The potential downsides like privacy need to be addressed, for example the danger of creating a data-driven system and turning the joy and complexity of learning into something based on simplistic algorithms, mechanistic and inhuman, students being ciphers. - roger.blamire roger.blamire Feb 27, 2011

(2) What themes are missing from the above description that you think are important?

  • Learning Analytics could be used to assist teacher performance. For example, data mining could show a teacher what parts of the course students are having difficulty and inform them about how well they have taught a specific topic and whether they need to go back over specific topics. More controversially, this technique could compare how two cohorts, taught by different teachers, are faring during a course and administrators could potentially be able to see across a wide spectrum of subjects where teachers may be struggling. If this is used in a positive way, ie administrators identify that a teacher needs some further development (as you say above, to transform pedagogy) in a specific area and supports the teacher to undertake further learning, then I believe that would be good for the student, teacher and school. If it is used in a negative way, that is, a way for the school to remove a teacher, then it simply will not be accepted by the profession, and frankly would be a demonstration of poor leadership. - garry.putland garry.putland Feb 18, 2011
  • How to get faculty buy-in on performance measurement and improvement? How to build a culture of LA, similar to grading. (- michael.lambert michael.lambert Feb 20, 2011)
  • I think this covers most of the area pretty nicely. However, what's missing is a discussion of the legal and policy barriers (privacy for example) to sharing data across entities (states, districtrs, private/public organizations) and across borders. In the US for example, the federal FERPA law is somewhat of a barrier from most folks I have talked with. I also think that while the description talks of the challenges of melding data from a wide array of systems, the problem is multi-faceted. There is the issue of the data formats themselves, but there is also the issue of the underlying meaning of the educational performance data, which is not uniform, i.e., a passing grade in one place might be a failing grade in another. A thorough examination of these issues is required or misleading analysis is likely. Some ways of equating educational data across systems is probably needed. There will be "low hanging fruit" to be harvested and then tougher problems to address.- chris.brown chris.brown Feb 21, 2011

(3) What do you see as the potential impact of this technology on teaching, learning, or creative expression?

  • I think that this has the potential to enhance, amplify and unleash learning(for students and teachers) in ways we haven't seen before. As we move increasingly online, quality learning experiences should be able to be improved through mining appropriate data. Teachers will be better placed with relevant information to know how students are faring and to provide intervention strategies or extension work for students. It will result in a greater focus on student support during the learning process. I suspect that the increase that we are seeing in online tutoring services will benefit from such technologies. - garry.putland garry.putland Feb 18, 2011
  • With LA software, the student can receive immediate feedback. This feedback can be distributed to the teacher and student, allowing both to make adjustments. Imagine an LA system that analyzes a student’s essay using the 6-trait writing. Feedback is given, coupled with the teacher input, and then the student completes another essay focused on one of the 6-traits that was scored low. Student and teacher review the LA, setting a goal for the next essay. This LA software is not intended to replace one-on-one conferences, but it certainly can help reduce grading time. (- michael.lambert michael.lambert Feb 20, 2011)
  • I think as the area progresses it has the potential to help us to more firmly understand how we learn and are impacted from a variety of educational programs and interventions and under which conditions, including teaching/teachers and schools and school systems, virtual learning, etc. I can foresee an educational epidemiology emerging in 5 to 10 years, similar to what we have in medicine. An educational epidemiologist would study large educational data sets to help inform on "better learning". Some folks are doing simliar things already on the narrow question of teacher effectiveness, modeling "teacher value added" with school and the state data sets.- chris.brown chris.brown Feb 21, 2011

(4) Do you have or know of a project working in this area?

  • A software tool named--Social Networks Adapting Pedagogical Practice (SNAPP)
    Aim: to develop a student network visualization tool from the extraction of discussion forum activity. This network diagram of students’ discussions online can: identify disconnected (at risk) students; identify potentially high and low performing students so one can plan interventions before the teacher can mark their work; and more. See http://research.uow.edu.au/learningnetworks/seeing/snapp/index.html (- michael.lambert michael.lambert Feb 20, 2011)
  • We (the Pearson Foundation) are doing an investigation of this area with the National Academy of Sciences, specifically the National Academy of Education, where we are bringing together leading researchers internationally. BTW, any referrals to researchers in this area is appreciated. Contact me - chris.brown chris.brown Feb 21, 2011
  • Higher Ed, but David Wiley gave a great talk on learning analytics at Educause Learning Initiative confernence - http://www.educause.edu/Resources/OpennessLearningAnalyticsandCo/224199 - bwatwood bwatwood Feb 27, 2011
  • See the 1st , 2nd, 3rd and 4th International Conference on Educational Data Mining for the best technical advances in this field [- roy.pea roy.pea Feb 27, 2011]

Please share information about related projects in our Horizon K-12 Project form.