{\rtf1\ansi\ansicpg1252\cocoartf1138\cocoasubrtf470
{\fonttbl\f0\fswiss\fcharset0 Helvetica;}
{\colortbl;\red255\green255\blue255;}
\margl1440\margr1440\vieww14880\viewh17020\viewkind0
\pard\tx720\tx1440\tx2160\tx2880\tx3600\tx4320\tx5040\tx5760\tx6480\tx7200\tx7920\tx8640\pardirnatural

\f0\fs24 \cf0 Analytics\
detailed notes on session: {\field{\*\fldinst{HYPERLINK "https://docs.google.com/document/d/1ErMZWAKlJlc1cJ0KWk9e068YvqUROmPczpOIX-h6Xgo/edit?pli=1"}}{\fldrslt https://docs.google.com/document/d/1ErMZWAKlJlc1cJ0KWk9e068YvqUROmPczpOIX-h6Xgo/edit?pli=1}}\
\
Market encouraging greater "flexibility" to meet differing student needs - say for 2 year or part time 6 year course.\
\
Learning analytics: mea, collection, analysis and rporting of data about learners and their contexts, for purposes of understanding and optimizing learning\'85\
\
- indicators of actual behavior of students (particularly in online context)\
- indicators of learning\
- why students succeed or fail\
- formative evaluation and feedback during the course\
- opportunity to personalize students' experience\
- how is the course performing\
\
UK activities\
much is about retention; many are library based\
so far, small groups with specific goal or charge rather than institution-wide\
sustainability of these efforts after initial funding is questionable\
Hull U: eCommons - learning and teaching and communications digital platform\
	analytics look at retention and utilization of library resources\
	in Chemistry, 1st year benchmarking of students, looking for signs of underperformance;\
	intervention to increase retention by 25%\
\
UMBC use: students in BbLearn get feedback on their activities compared to the average in class\
\
Learning Analytics 101\
Steve Lonn, U Mich & Josh Baron, Marist College\
\
"Big data" in education - from many "silos" of data (LMS, ERP, \'85)\
\
LA= \
understand entires systems and support human decision making\
applies known methods & models\
\
Examples of LA projects\
Purdue "Course Signals" college wide\
U Mich E**2Coach - course specific\
UMBC "Check My Activity" tool - student-centered \
\
Course Signals: build predictive model form LMS and SIS & demographic data\
leverage model to create early-alert: identify students at risk to not complete course; deploy intervention\
systems automates intervention: students get a "traffic light" in LMS and messages to student suggesting corrective action\
\
data based on those using Course Signals: increase grades, and increase retention\
\
Ellucian product now; integrates with BbLearn\
\
E**2Coach - specific physics courses\
pre-course survey uses academic, psycho-social factors and performance\
Michigan Tailoring System (MTS) provides highly customized messages to students\
\
UMBC Check My Activity Tool\
students earning Ds and Fs us Bb Learn 39% less than higher graded students\
Provide data on LMS activity to course averages; can also compare average usage to grade outcomes\
Feedback from students positive\
\
LA Research Questions\
Data Mining vs. Learning Science approach.  Build models from large data sets or from understanding of learning sciences?\
Scaling across higher ed: how portable are predictive models?  do we need an open standard for LA?\
\
Ethical questions\
Are we obligated to use such tools to improve student learning?\
Who owns the data?  Should students be allowed to "opt out"?  Suppose opting out reduces overall predictive power of model!\
What should be revealed to students? to instructors? will instructors "profile" students? or will some students be de-motivated?\
}