Moocs_Analytics with Audio

Information about Moocs_Analytics with Audio

Published on July 18, 2014

Author: joedempsey5036



EME6356 Tool Case Study MOOCs & Analytics : EME6356 Tool Case Study MOOCs & Analytics Jiyae Bong, Sharon Gillooly , Monica Surrency What are MOOCs?: What are MOOCs? Online courses delivering learning content to anyone who wants to take courses, with no limit of attendance and low cost Benefits & Challenges MOOC Platforms include statistics or analytics functions Massive Open Online Courses (Chen, Barnett, & Stephens; Pappano , 2012) Content: Content 1. Why use analytics in MOOCs We will review research to provide rational for collecting large datasets. Topic: How are analytics in MOOCs being used for teaching and learning? 2 . Analytics in MOOCs We will explore the data from an individual MOOC, comparing it with a traditional classroom and see how the data can be used for collaboration and shared across multiple platforms. 3 . How analytics are being used in MOOCs . How do we get the data? We will share Monica’s experience of ERAU’s first m MOOC and its analytics and explore an example from m Blackboard CourseSites . Content: Content 4. Who benefits? We we explore the benefits and potential of MOOC analytics for students, instructors, instructional designers and researchers. Topic: How are analytics being used for teaching and learning in MOOCs ? 5. Challenges We will discuss validity concerns and ethical considerations related to privacy and explore the current solutions. 6. Conclusions We will look at some of the results from recent studies in MOOC analytics. 7 . References Analytics & MOOCs: Analytics & MOOCs MOOC data is different from classroom data The magnitude of data Size : Numbers of registrants per course, observations per registrant Types of info : E very click, text-based submission, submission times, IP addresses, context-specific data for each type of interaction with the content. The characteristics of data Diversity of registrants in regard to both background and intent Asynchronous use of course tools and relatively unrestricted sequence (DeBoer et al, 2014) Analytics & MOOCs: Analytics & MOOCs D ata Standards to analyze cross-platform data The data standards development The researchers are developing data standards to allow for both collaboration and comparison across MOOC platforms. It mitigates privacy concerns through the sharing of scripts which analyze, visualize data and prepare data for models, rather than the sharing of data itself. The MOOCViz visualization platform Instructors can download software to create visualization. Researchers can share their visualizations via web-based galleries. They also can refine the scripts. Public can access the visualizations for MOOC data. ( Veeramachaneni & O’Reilly, 2014) Why Use Analytics in MOOCs?: Why Use Analytics in MOOCs? To learn more about who MOOC students are and how they engage with the content Redefine educational variables in a new educational context To inform the MOOC conversation about learner motivation and engagement in order to exploit the potential of the MOOC platform ( DeBoer et al, 2014; Kizilcec , Piech , & Schneider, 2013; Newman & Oh, 2014) Why Use Analytics in MOOCs?: Why Use Analytics in MOOCs? ERAU First MOOC - Content Interaction (Surrency et al., 2014) Why Use Analytics in MOOCs?: Why Use Analytics in MOOCs ? ERAU MOOC Demographic Comparisons (Surrency, 2014) Why Use Analytics in MOOCs?: Why Use Analytics in MOOCs ? ERAU First MOOC (Surrency, 2014) How Do We Get the Data?: How Do We Get the Data? Surveys Grade Center Course Reports Blackboard CourseSites How Do We Get the Data?: How Do We Get the Data? Image Source: Blackboard CourseSites How Do We Get the Data?: How Do We Get the Data? ERAU First MOOC Sample Course Report How Analytics are Used in MOOCs?: How Analytics are Used in MOOCs ? Course Activity Report from ERAU First MOOC Who Benefits?: Who Benefits? ( Cronenweth , 2013; Mayer- Schönberger & Cukier , 2014 ) Challenges - Validity: Challenges - Validity ( Straumsheim , 2014 ) “We spent about 80 to 90 percent of our time on fundamental data transformation” – John Whitmer , Program Manager of Academic Technology and Analytics at California State University System (The data provided) “no insight into learner patterns of behavior over time.” - Higher education consultants examining student data collected by edX “ Just because you have a lot of data doesn’t mean you have good data.” - Laura W. Perna , Professor, Graduate School of Education, University of Pennsylvania “When we have a sample size like we do, it’s really easy to get significance….but we didn’t even get that.” - Whitmer , conducting research on MOOC participants’ engagement with the MOOC data. Challenges - Ethical: Challenges - Ethical Privacy De-identification Ex. of Harvard University & MIT Remove personal identification info Aggregation, anonymization via random identifiers Blurring to reduce individuality of sensitive data fields Unique numeric identifier Ex. of Coursera Use a hashing-based anonymization mechanism that replaces the numeric student identifiers in research data exports with 40-character hexadecimal identifiers. ( Biemiller , 2014; Coursera , 2012; Veeramachaneni & O’Reilly 2014 ) Schema scripts Software transforms data from multiple platforms to the MOOCdb standard eliminating the need to share actual data. PowerPoint Presentation: Conclusions MOOC Analytics are providing useful information for students, instructors, instructional designers and researchers regarding: Who MOOC students are How they engage with the content Educational context PowerPoint Presentation: Conclusions 1. Newman and Oh study, 2014 Majority of students are male MOOCs attract students who already have college degrees Median age is 24.4 1/3 are from North America Nearly half never engage with any content Europeans view the most content Students with a doctorate reviewed more content Serial students are the most engaged 2. Kizilcec , Piech and Schneider, 2013 4 Classifications of MOOC Students: Completing Auditing Disengaging Sampling PowerPoint Presentation: Conclusions 3. De Boer et al 2014 Enrollment times, participation and curricula varies. Achievement needs to be re-conceptualized as the accomplishment of self-defined goals rather than criteria established by instructor. 4. More research is needed regarding validity and student privacy References: References Biemiller , L. (2014, May 30). QuickWire : Harvard and MIT release scrubbed MOOC data [Web blog post]. Retrieved from  http:// Chen, X., Barnett, C. R., & Stephens, C. Fad or future: The advantages and challenges for massive open online courses (MOOCs). Retrieved from https:// . Cronenweth , S. (2013, April 30). Learning analytics and MOOCs [Web blog post]. Retrieved from . DeBoer , J., Ho, A. D., Stumo , G. S., & Breslow , L. (2014). Changing "Course": Reconceptualizing educational variables for massive open online course.  Educational Researcher, 43 (2), 74-84 . Kizilcec , R.,   Piech , C.,  & Schneider, E. (2013). Deconstructing Disengagement:  Analyzing Learner Subpopulations in Massive Open Online Courses, LAK’13 Leuven, Belgium. Mayer- Schönberger , V., & Cukier , K. (2014). Learning with Big Data: The Future of Education . New York, NY: Houghton Mifflin Harcourt. Newman , J. & Oh, S. (2014, June 13). 8 things you should know about MOOCs. The Chronical of Higher Education. Retrieved from   References: References Pappano , L. (2012, November 2). The year of the MOOC. The New York Times. Retrieved from .  Straumsheim , C. (2014, June 10). Data, data everywhere [Web blog post]. Retrieved from Surrency , M., Norris, K, Field, S., Anderson, H. (2014, July). Purposeful MOOCing : Secrets Exposed . Panel presented at the BbWorld conference of Blackboard, Las Vegas. Veeramachaneni , K., O'Reilly, U. (Oct 2013-May 2014) Developing data standards and technology enablers for MOOC data science. Appendix – Research on Analytics in MOOCs: Appendix – Research on Analytics in MOOCs

Related presentations