Data Driven Instruction for Personalized Learning Sagar Kamarthi
- Slides: 32
Data Driven Instruction for Personalized Learning Sagar Kamarthi, Ph. D Northeastern University NAE FOEE, Irvine, CA, Sept. 26 -28, 2016 Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 1
Basic Principles of Data Driven Instruction p Focus: Are students learning? p Philosophy: Goal-directed actions with periodic feedback enhances student learning Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 2
Steps for Data Driven Instructions p Assessment Periodic assessment to gather meaningful data p Analysis Identify causes of strengths and weakness p Action Teach to address barrier to learning and fill knowledge gaps p Culture Institutionalize data driven instructional practices Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 3
Personalized Learning p Personalized learning results from tailored instruction aligned with students’ aptitude, background, knowledge, and interests p NAE has recognized personalized learning as one of the fourteen Grand Challenges Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 4
Need for Personalized Learning in Engineering Courses p Changing demographics p Increasing numbers of Under-Represented Minority (URM) and woman students in engineering p Transfer students from community colleges p International students p Diversity of student preferences and aptitudes Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 5
Challenges to Tailored Instruction p Scalability is a main challenge for tailored instructions Enormous time and effort are required to create teaching material for differentiated instruction Difficult to elicit information regarding individual student characteristics Traditional resources are inadequate to collect assessment data, analyze data, and offer individualized instruction Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 6
Mass Customized Instruction (MCI) Model p I borrowed strategies from the field of mass customization to address scalability issues associated with differentiated instruction Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 7
Mass Customization p Mass customization offers individually tailored products and services on a large scale as opposed to offering one standard solution to all, or offering expensive custom solutions to a few Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 8
Example of Mass Customization p Dell Computers p Fro. Yo yogurt p Nikei. D custom shoes p Lands’ End custom apparel Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 9
Key Strategies of Mass Customization p Solution space development p Robust process design p Choice navigation Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 10
Solution Space for Personalized Learning p Identifying student attributes along which their learning needs diverge the most Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 11
Solution Space for Personalized Learning Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 12
Robust Process Design for Personalized Learning p Seamless and dynamic integration of different instructional materials and resources Classroom lectures Video documentaries Multimedia interactive learning tools Hands on activities Supplementary reading material Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 13
Choice Navigation for Personalized Learning p Tools to help students determine their own learning needs and means with manageable number of choices Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 14
Implementation of MCI Model p. I partially implemented MCI model in Manufacturing Systems (IE 4530) course Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 15
Dimensions for Solution Space p Dimensions along which the student needs differ the most Prior knowledge Motivation level Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 16
Student’s Motivation Level (y) 2 D Solution Space Considered for Manufacturing Systems Course Student’s Prior Knowledge (x) Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 17
Student’s Motivation Level (y) Discretized 2 D Solution Space x 1 = Students with good prior knowledge x 2= Students with some gaps in their prior knowledge x 3 = Students with serious deficiencies in their prior knowledge y 1 = Students with high motivation to learn subject y 2 = Students with neutral motivation to learning subject y 3 = Students with poor motivation to learn subject y 3 y 2 y 1 x 2 x 3 Student’s Prior Knowledge (x) Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 18
Customized Instruction x 1 = Students with good prior knowledge y 1 = Students with high motivation to learn subject Read an interesting case study and discuss it with his/her team mate Build a digital simulation model to observe variation in the production line efficiency under various scenarios (with or without storage buffers) Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 19
Customized Instruction x 3 = Students with serious deficiencies in their prior knowledge y 3 = Students with poor motivation to learn subject Design and build a candle stand using turning machine in the lab Conducted a physical simulation of manual assembly line to observe variation in the production line efficiency under various scenarios (with or without storage buffers) Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 20
Customized Instruction x 3 = Students with serious deficiencies in their prior knowledge y 3 = Students with poor motivation to learn subject Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 21
Prior Knowledge Components Engineering Materials Manufacturing Processes and Systems Simulation Manufacturing System Design and Analysis Engineering Echonomics Optimimization Models Statistics Northeastern Performance Metrics of Manufacturing System Manufacturing Outsourcing Manufacturing Systems and Techniques Course (IE 4530) Automation and Numerical Control NAE FOEE Irvine, CA 9 /26 -28/ 2016 22
Instructional Materials to Address Prior Knowledge Gaps Instruction feature Teaching materials Teaching modes Instructional Material Supportive tools Collaboration tools Text book Class handouts Lecture using blackboard Power-point presentation Simulation models Show-and tell physical models Email Blackboard Excel documents Record Keeping Northeastern Cases studies Video, vides streams of lectures Cases discussion Hands-on lab experiment Out-of-class assignments Plant tours Digital discussion rooms File sharing/exchange tools To communicate student specific recommendation of instruction material and activities To post grades and feedback NAE FOEE Irvine, CA 9 /26 -28/ 2016 23
To Improve Motivation for Manufacturing p Students are assigned a term paper to research the State of manufacturing in the U. S. Factors influencing the emigration of manufacturing from the U. S. Barriers to return manufacturing to the U. S. Imperatives and strategies to keep manufacturing in the U. S. Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 24
Result of Motivation Activities p Once students complete the term paper they are typically more motivated to learn about manufacturing and willing to dedicate their energy to learn the subject Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 25
Data Driven Approach to Personalized Learning Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 26
Score for “I Learned a Lot” in Last Four Offerings of Mfg. Sys. Course 4. 55 4. 45 4. 35 4. 25 4. 15 4. 05 3. 95 0 Northeastern 1 2 3 4 5 NAE FOEE Irvine, CA 9 /26 -28/ 2016 27
Challenges and Solutions to Implementing MCI Model p Requires a lot of effort to create an array of instructional material Solution: Supplementary material can be develop collaboratively as an open source effort by the engineering education community Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 28
Challenges and Solutions to Implementing MCI Model p It is not easy to design and create effective assessment and feedback instruments Solution: Engineering education research community can develop feedback instruments through a research grant Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 29
Challenges and Solutions to Implementing MCI Model p Require sophisticated information technology tools to track individual student assessment and prescription data Solution: In collaboration with information systems experts and engineering education researchers can develop IT tools for tracking individual student assessment and prescription data Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 30
Data Analytics for Personalized Learning Predictive Analytics p Affinity Analysis p Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 31
Data Analytics for Personalized Learning p Common words in comments for high performance students Grit Creativity Curiosity p Common words in comments for poor performance students Consistency Sufficiency Focus Northeastern NAE FOEE Irvine, CA 9 /26 -28/ 2016 32
- Sagar kamarthi
- Data driven instruction cycle
- Data driven instruction cycle
- Personalized learning learner profile
- Institute for personalized learning
- Ravi raj sagar
- Chaki 1
- Sagar institute of research and technology
- Heather sagar noaa
- Sagar samtani
- Sagar chaki
- Sagar block
- Sagar recruitment
- Sagar dhekne
- Individualized instruction vs differentiated instruction
- Site:slidetodoc.com
- Adversarial personalized ranking for recommendation
- Personalised mobile search engine
- Pentaho business analytics
- Personalized patient education
- Contextual bandits for personalized recommendation
- Personalized recommendations
- Personalized navigation
- Personalized statement messaging
- All-to-all personalized communication
- Limitations of trait theory of leadership
- Social stories questionnaire
- Medicine
- Oyunlarm
- Cuadro comparativo e-learning y b-learning
- Self-taught learning: transfer learning from unlabeled data
- Api driven ddi
- Des fleet operations