What are Learning Analytics?

Learning analytics is an educational application of web analytics, a science that is commonly used by businesses to analyze commercial activities, identify spending trends, and predict consumer behavior. Education is embarking on a similar pursuit into data science with the aim of learner profiling, a process of gathering and analyzing large amounts of detail about individual student interactions in online learning activities. The goal is to build better pedagogies, empower students to take an active part in their learning, target at-risk student populations, and assess factors affecting completion and student success. For learners, educators, and researchers, learning analytics is already starting to provide crucial insights into student progress and interaction with online texts, courseware, and learning environments used to deliver instruction. Students are beginning to experience the benefits of learning analytics as they engage with mobile and online platforms that track data to create responsive, personalized learning experiences.

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?

  • Global outreach, unbundling of content and assessment into smaller chunks and cross- institutional delivery via educational marketplaces (open and in exclusive subsets) seems to be a natural flow-on from analytics applied across the sector. - david.c.gibson david.c.gibson Sep 24, 2015
  • Learning analytics to better understand learners' behaviors and adapt to personalized learning paths is an essential added value for the educational sector, as well as better assessing the quality of learning objects. - agermain agermain Oct 18, 2015- lkoster lkoster Oct 22, 2015 - kelvin kelvin Oct 25, 2015
  • Learning and assessment are closely related. Thus, it is apparent that learning analytics embrace assessment analytics and all the sub-categories deriving from assessment methodologies and technologies (computer-based assessment, computer-based testing, computerized adaptive testing, mobile assessment, etc.). ICT systems will monitor, track and record educational data (related to students, teachers, activities, infrastructure etc.), interpret and map the real current state of these data, organise these data (e.g., filter, classify, prioritise), use these data (e.g., decide adaptations, recommend, provide feedback, guide the learner) and predict the future state of these data. The target is to inform and empower students, teachers and organization about performance and goal achievement, and facilitate decision making accordingly. Some of the learning analytics goals include the prediction of students' performance, prediction of dropout/retention rates, recommendation of educational resources, increase sel-reflection and self-awareness, improve learning and assessment http://www.ifets.info/journals/17_4/4.pdf http://conta.uom.gr/conta/filter_pub.php?filter=3&lang=en - Anastasios.A.ECONOMIDES Anastasios.A.ECONOMIDES Oct 23, 2015


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

  • The extension of analytics into other parts of the higher education delivery system might be worth mentioning: recruitment, (all phases mentioned above re learners learning), adaptive curriculum, adaptive assessment, instructor effectiveness, curriculum management, assessment management, alumni networking, employer relationships, work integrated learning, internships, graduate employablity strategies, life-long portfolios and learning opportunities are examples of the other parts of the business of higher education that will be transformed via analytics. - david.c.gibson david.c.gibson Sep 24, 2015 Yes.- bryan.alexander bryan.alexander Oct 1, 2015 +1 - kvogt kvogt Oct 6, 2015 + another 1- helga helga Oct 6, 2015 - JoanLippincott JoanLippincott Oct 17, 2015- cevetell cevetell Oct 22, 2015 +1- lkoster lkoster Oct 22, 2015
  • Agreed. Not least learners' learning is of importance to me.- ole ole Sep 29, 2015
  • The description is written with a strong focus on online learning, but can be useful in other learning contexts as well. - kvogt kvogt Oct 6, 2015 The data needs to be collected via online activities but that doesn't mean that the entire course is online. - JoanLippincott JoanLippincott Oct 17, 2015 I agree with Joan. Any digital system on campus produces data that could inform any learning activity be it online or face-to-face, or hybrid. The most widely used example of this is the LMS, which is used extensively to support on campus instruction, as well as the online medium.- cevetell cevetell Oct 22, 2015 - kelvin kelvin Oct 25, 2015
  • I would add to this list that learning analytics helps also to assess and therefore improve the relevance and the quality of learning materials/objects used in formal and informal education. - agermain agermain Oct 18, 2015
  • Integrating assessment analytics into the education practice might help build better e-assessment services and systems. For example, provision of suitable audio-visual feedback during assessment could reduce students' on-task stress and fear. http://conta.uom.gr/conta/filter_pub.php?filter=32&lang=en - Anastasios.A.ECONOMIDES Anastasios.A.ECONOMIDES Oct 23, 2015
  • Also Campus experience the pathways, churn and use of space using analytics not only from LMS but the library systems, geo location, wifi data, activity etc - extensively used by mall/shopping centre Logicalis Spacial Intelligence Solutions http://www.uk.logicalis.com/news/logicalis-uk-launches-spatial-intelligence-solution/- capturing motion and location - DaveP DaveP Oct 25, 2015
  • Obstacles include: data in silos, legacy systems, FERPA--or rather local FERPA policy and interpretation, since often it's how FERPA is implemented that causes issues, rather than FERPA itself. - edward.oneill edward.oneill Oct 25, 2015 - kelvin kelvin Oct 25, 2015
  • Data Governance is a huge issue. LA is something we tend to do to students without looking at the issues surrounding what students really want and need- neil.witt neil.witt Oct 25, 2015- escience escience Oct 25, 2015 - kelvin kelvin Oct 25, 2015
  • Note the links to affective computing and big data -- affective computing is one of the emerging tools that can augment analytics, and the data generated by these create big-data challenges - escience escience Oct 25, 2015
  • There has been more "sizzle" than "steak" in the learning analytics front thus far. As promising as LA is, we must be careful not to overpromise and, thus, find ourselves headed for the "trough of disillusionment." Simpler analytics applied well can yield many helpful insights if diverse institutional groups are involved in interpreting and devising appropriate interventions or other actions. (I'm focusing on student success here as the frame. I understand that there are other BI outcomes desirable for LA as well.) - kelvin kelvin Oct 25, 2015
  • Big Data "In education, digital learning technologies such as games and online learning systems collect vast amounts of data as students progress through the game, test, or activity. This type of incremental information can give a more complete picture of the learning process than traditional measures such as grades and test scores, which only measure outcomes. It can also help educators and researchers gain valuable insight into how to improve and personalize learning for students," says Aubrey Francisco. The Chinese version of the bestseller Learning with Big Data: The Future of Education will be published later this year by Shanghai-based East China Normal University. I wonder if this may fit with Visual Data. -
    michael.lambert
    michael.lambert
    michael.lambert Oct 2, 2015-
    bryan.alexander
    bryan.alexander
    bryan.alexander Oct 18, 2015 -
    ole
    ole
    ole Oct 19, 2015 -
    helga
    helga
    helga Oct 20, 2015

    We have access to an amazing array of data about our students. Trouble is, we just ignore it... (-
    damian.mcdonald
    damian.mcdonald
    damian.mcdonald Oct 20, 2015
    ) That's because, like many innovations, it's interesting to talk about but we seem unwilling to make the organizational/structural changes necessary for big data to have any meaningful impact on what we're doing. -
    tom.haymes
    tom.haymes
    tom.haymes Oct 21, 2015

    I think Big Data as well Educational Data Mining are covered in the Horizon Topic "Learning Analytics"-
    Anastasios.A.ECONOMIDES
    Anastasios.A.ECONOMIDES
    Anastasios.A.ECONOMIDES Oct 21, 2015-
    cevetell
    cevetell
    cevetell Oct 22, 2015

    I agree that data relating to specific student's progress in a course/activity/game is probably more like Learning Analytics. However, I believe Big Data goes beyond data from specific learning activities. The data that a school collects could be used to look at patterns (for example, looking at success rates in first year courses vs. courses taken in high school, success in 2-4 year courses based on success in first year, etc.). -
    lkoster
    lkoster
    lkoster Oct 22, 2015
    Are there off-the-shelf commercial solutions for big data in education?-
    bryan.alexander
    bryan.alexander
    bryan.alexander Yesterday
    The way in which big data is important in this sector is the way that more detailed/more continuous data collected from less-designed/transactional sources enable new forms of analytics -- which in turn pose novel challenges for interpretation and validation. One area of application for these analytics is learning -- the other is scholarly publishing and impact.-
    escience
    escience
    escience Yesterday
    Completely agree about the inclusion of Big Data, but can I put in a plea to make sure we talk to our students about their needs, hopes, fears and expectations about what we are going to do with THEIR data -
    neil.witt
    neil.witt
    neil.witt Yesterday
    [Editor's Note: Added here from RQ2.]

(3) What do you see as the potential impact of this technology on higher education?

  • Significant new business models and new delivery models. - david.c.gibson david.c.gibson Sep 24, 2015. More student focused teaching. Research on learning processes and strategies.- ole ole Sep 29, 2015
  • Advising will be positive impacted because learning analytics will find students at risk at early stages, allowing actions for preventing students withdraw academic programas. - jreinoso jreinoso Oct 3, 2015 - JoanLippincott JoanLippincott Oct 17, 2015 - kelvin kelvin Oct 25, 2015
  • Finding the "right" student for each university will be another impact of this technology because this technology facilitate recruitment. - jreinoso jreinoso Oct 3, 2015
  • Supports more personalized/customized learning pathways - kvogt kvogt Oct 6, 2015 - DaveP DaveP Oct 25, 2015 - kelvin kelvin Oct 25, 2015
  • Improvement of learning software design - gordon gordon Oct 15, 2015
  • Possible crossover impact on security based on identifying outlying patterns in user behavior - gordon gordon Oct 15, 2015
  • Improve successful completion of required courses, especially in STEM, resulting, hopefully, in better retention. - JoanLippincott JoanLippincott Oct 17, 2015
  • Offer more/new education models and blurring even more the distinction between formal and informal learning. This is a plus to open higher education to a more diverse population of learners.- agermain agermain Oct 18, 2015
  • Civitas Learning, the higher education software service provider for student success, has received a commitment of up to $60 million as it looks to expand its suite of products. The Austin-based company started its development with six universities in 2011 and has since grown to facilitate and monitor student progress for 80 public and private universities and colleges. According to the company, the idea is to use data science to encourage retention, student engagement, and after-school success among colleges across the country. Check their partners list http://www.civitaslearning.com/strategic-partners/ - Maya Maya Oct 19, 2015Maya Georgieva
  • Educational research has suffered because of the limits of traditional statistical modeling and sampling techniques, and the limits of control group method. It is obvious that every brain is unique, and so sampling has not done well to draw strong connection between research and outcomes. Now all learning interactions in an LMS and other systems can be captured for all students over long stretches of time. This completely changes the nature of the type of research we can conduct, no longer must we sample, we can look at the totality of the data pool to observe patterns and see correlations of significant magnitude.- cevetell cevetell Oct 22, 2015 (And to add to this, not only does this allow for whole-population inference, but also allows for many small experiments to be run. This is incredibly important -- in most education research the treatment applies to a class -- so for some analysis N=1 or very small; and results are assessed at the end of the semester. In a on-line learning environment, experiments can be run on samples of students over much shorter time scales -- yielding effectively much higher N's and much faster iterations.) - escience escience Oct 25, 2015
  • The issue of understanding the deeper learning processes by deconstructing them into more simple, distinct mechanisms can be potentially enhanced by learning analytics procedures. Due to learning analytics, every type of interaction can be coded into behavioural schemes and decoded into interpretable guidance for decision making. Thus, this technology can be beneficial for students, instructors and organizations by timely revealing their personalized needs and successively increasing their (self-)awareness, self-regulation and the overall re-construction of the learning processes. - Anastasios.A.ECONOMIDES Anastasios.A.ECONOMIDES Oct 23, 2015


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


Please share information about related projects in our Horizon Project Sharing Form.