This Week in Learning Analytics (February 7 – 13, 2015)

On February 9, 2015 Ryan Baker delivered a compelling lecture on predicting student outcomes using automated predictors of engagement and affect. The lecture was delivered as part of the QuanTM Learning Analytics Speaker Series.
On February 9, 2015 Ryan Baker delivered a compelling lecture on predicting student outcomes using automated predictors of engagement and affect. The lecture was delivered as part of the QuanTM Learning Analytics Speaker Series, and will be made availalbe via the Emory University YouTube channel in the coming weeks.


Higher Ed

Game Design Helps Engage Students in Classroom
Press and Guide | 13 February 2015

University of Michigan Ann Arbor is in the midst of a pilot program in which classes are taught using a new gamified LMS called Gradecraft. “GradeCraft is a game-inspired LMS designed to facilitate autonomy while building student competence in the subject area. It not only motivates students in nontraditional ways, but also leverages learning analytics to better inform the instructor on a student’s strengths, weaknesses and interests so that the instructor can best understand how to help students succeed.” Students are provided with analytics which chart their progress and identify ways to earn additional ‘points.’ GradeCraft was designed by Caitlin Holman, Stephen Agular, and Barry Fishman, who presented their initial findings at LAK’13.


Analytics Creating Whole Host of Privacy Issues, Claims University of Derby’s IT Director
Computing | 10 February 2015

IT director of the University of Derby, Neil Williams, comments on the opportunities and challenges associated with the use of big data analytics in university environments. “The challenge for the university, which isn’t the same as for other organisations, centres on the ethics of such ‘Big Brother’ big data efforts. We are building capabilities but we have to look through these internally and ask what it means and go through the appropriate governance methods and understand what our stakeholders feel. We have to make sure what we do is right. In contrast, if you look at a website and somebody is tracking what you’re looking at, you expect to then be targeted [with advertisements], but these [students] are members of our institution – not customers, so it’s a different situation.”



University of Nottingham Announces Learning Analytics Program
University of Nottingham | 8 February 2015

The University of Nottingham has initiated a learning analytics project that will roll out in multiple phases. The first phase will involve identifying correlations between students’ use of Moodle, module scores, and satisfaction levels in order to optimize their LMS environment. Future goals for the project involve creating a recommender system and embedding analytics to create interactive learning environments.



Uncovering Security Flaws in Digital Education Products for Schoolchildren
New York Times | 8 February 2015

Tony Porterfield has alerted the makers of nearly 20 digital education products to serious security flaws which are symptomatic of “widespread lapses in student data protection across the education technology sector.” Although some companies, including Pearson and Class Dojo, took immediate action in the face of Porterfield’s concerns, many have not.



DIY Learning Analytics Workshop at Emory University
Timothy D. Harfield | Analytics for Learning at Emory | 13 February 2015

Microsoft Education evangelist, Patrick Leblanc, visited Emory University on February 11 to facilitate a workshop on the use of MS Excel, Power Query, Power Pivot, and Power BI to work with educational data sets. He effectively demonstrated the power and versatility of Microsoft products that many currently already license.


LACE initiative Resulted in New Asian SIG on Learning Analytics
Weiqin Chen | Learning Analytics Community Exchange (LACE) | 8 February 2015

A successful pre-conference learning analytics workshop held in conjunction with the International Conference on Computers in Education (ICCE) has led to the formation of learning analytics special interest group within the Asia-Pacific Society for Computers in Education (APSCE)



Journal Papers

Data Mining in Higher Education: University Student Dropout Case Study
Ghadeer S. Abu-Oda & Alaa M. El-Halees |
International Journal of Data Mining & Knowledge Management Process (IJDKP) | 5 (1)

ABSTRACT: In this paper, we apply different data mining approaches for the purpose of examining and predicting students’ dropouts through their university programs. For the subject of the study we select a total of 1290 records of computer science students Graduated from ALAQSA University between 2005 and 2011. The collected data included student study history and transcript for courses taught in the first two years of computer science major in addition to student GPA , high school average , and class label of (yes ,No) to indicate whether the student graduated from the chosen major or not. In order to classify and predict dropout students, different classifiers have been trained on our data sets including Decision Tree (DT), Naive Bayes (NB). These methods were tested using 10-fold cross validation. The accuracy of DT, and NlB classifiers were 98.14% and 96.86% respectively. The study also includes discovering hidden relationships between student dropout status and enrolment persistence by mining a frequent cases using FP-growth algorithm.


Conference Proceedings

A Novel Similarity Measure Between Two Probability Distributions For Course Establishment
Aijiao Liu, Yiping Zhang, Min Chen | International Conference on Education, Management, Commerce and Society (EMCS 2015)

ABSTRACT: In this paper, in order to obtain the optimized analysis of clustering for the probability distributions, the increment of the description length is proposed to instead the relative entropy as the similarity measure between two probability distributions. Its corresponding features are also discussed in detail in this paper. As the improvement, the increment of description satisfies the symmetrical feature. On the basis of this similarity measure, K-means algorithm is employed to analysis the police training data and to influence the corresponding course establishment. The experiment results indicate that the proposed similarity measure can lead to better clustering results than some other previous similarity measure.



Best of Blogs

Not In the Clear: Libraries and Privacy
Barbara Fister | Inide Higher Ed | 12 February 2015

Barbara Fister (librarian and self-identified privacy nut) discusses the tension between (1) an interest in promoting student success through learning analytics and (2) a commitment to protect student privacy. She observes that, in spite of the latter, “libraries are terrible at privacy! … in an era when everybody’s doing data dragnets, it’s alarming to see how leaky our library websites are, how revealing our catalogs and databases are, and how cavalier we have been with patron data that we swear we will protect.”


Learning Analytics: On Silver Bullets and White Rabbits
Simon Buckingham Shum | Medium | 8 February 2015

With a poignancy so often lacking in discussions of learning analytics, Simon Buckingham Shum asks a question that should be top of mind for anyone before, during, and after working with big data in education: “In the very process of trying to value certain learning qualities by tracking them, will we in fact distort or even destroy a living, organic system, through clumsy efforts to categorise and quantify?


Learning Analytics in Practice

Tips and Tricks

7 Ways to Get Started with Analytics & Reports in Moodle
Sean Marx | iLite | 7 February 2015

Reviews seven plugins and reports that give users access to view trends, analytics and data Moodle sites.


Co-curricular Module Trends in Moodle, Using Learning Analytics
Charles Kasule | Kings College London, Centre for Technology Enhanced Learning | 6 February 2015

The Kings College London, Centre for Technology Enhanced Learning recently developed a learning analytics tool using Microsoft Excel that could visualize raw Moodle data and produce reports that are viewable offline. “In comparison to what is currently offered by Moodle reporting option, the Learning Analytics tool offers a better way of navigating, comparing, analysing and tracking how users use the VLE materials from the data that the Moodle report provides.”


Presentations & Webinars

What is the Tin Can API and how does it enable the flow of data?
Megan Bowe | SoLAR Storm | 7 February 2015