Addictive links, Keynote talk at WWW 2014 workshop
of 51
/51
-
Upload
peter-brusilovsky -
Category
Internet
-
view
256 -
download
3
Embed Size (px)
description
Empirical studies of adaptive annotation in the educational context have demonstrated that it can help students to acquire knowledge faster, improve learning outcomes, reduce navigational overhead, and encourage non-sequential navigation. Over the last 8 years we have explored a lesser known effect of adaptive annotation – its ability to significantly increase student engagement in working with non-mandatory educational content. In the presence of adaptive link annotation, students tend to access significantly more learning content; they stay with it longer, return to it more often and explore a wider variety of learning resources. This talk will present an overview of our exploration of the addictive links effect in many course-long studies, which we ran in several domains (C, SQL and Java programming), for several types of learning content (quizzes, problems, interactive examples). The first part of the talk will review our exploration of a more traditional knowledge-based personalization approach and the second part will focus on more recent studies of social navigation and open social student modeling
Transcript of Addictive links, Keynote talk at WWW 2014 workshop
Addictive Links: Engaging Students through Adaptive Navigation Support and Open Social Student Modeling
Peter Brusilovsky with: Sergey Sosnovsky, Michael Yudelson, Sharon Hsiao
School of Information Sciences, University of Pittsburgh
MOOC
Massive Open Online Course
Completion Rate
MOOC Completion Rate
Classic loop user modeling - adaptation in adaptive systems
http://www.katyjordan.com/MOOCproject.html
What Else These Students Need?
• Top colleges
– Stanford, CalTech, Princeton, GATech, Penn, Duke..
• Great faculty – top guns in their fields
• Great content
• Top online platform – Coursera
• FREE!
The Problem of Engagement
• Great free content and top teachers is not enough to engage students
• Peter Norvig: Motivation and Engagement are key problems for MOOCs
• The problem is not new
• A lot of great advanced content – Works perfectly in lab studies, great gains – Released to students to enhance learning – No impact – students do not use it
The Case of QuizPACK • QuizPACK: Quizzes for
Parameterized Assessment of C Knowledge
• Each question is a pattern of a simple C program. When it is delivered to a student the special parameter is dynamically instantiated by a random value within the pre-assigned borders.
• Used mostly as a self-assessment tool in two C-programming courses
QuizPACK: Value and Problems
• Good news: – activity with QuizPACK significantly correlated with
student performance in classroom quizzes – Knowledge gain rose from 1.94 to 5.37
• But: – Low success rate - below 40% – The system is under-used (used less than it deserves)
• Less than 10 sessions at average • Average Course Coverage below 40%
Adding Motivation
• Students need some better motivation to work with non-mandatory educational content…
• Added classroom quizzes: – Five randomly initialized questions out of 20-30 questions
assigned each week
• Good results - activity, percentage of active questions, course coverage - all increased 2-3 times! But still not as much as we want. Could we do better?
• Maybe students bump into wrong questions? Too easy? Too complicated? Discouraging…
• Let’s try something that worked in the past adaptive hypermedia that can guide students to the right content
User Model
Collects information about individual user
Provides adaptation effect
Adaptive System
User Modeling side
Adaptation side
User-Adaptive Systems
Classic loop user modeling - adaptation in adaptive systems
Adaptive Link Annotation: InterBook
1. Concept role
2. Current concept state
3. Current section state 4. Linked sections state
4
3
2
1
√"
Metadata-based mechanism
The Value of ANS
• Lower navigation overhead – Access the content at the right time – Find relevant information faster
• Better learning outcomes – Achieve the same level of knowledge faster – Better results with fixed time
• Encourages non-sequential navigation
Questions of the current quiz, served by QuizPACK
List of annotated links to all quizzes available for a student in the current course
Refresh and help icons
QuizGuide = QuizPACK+ANS
Topic-Based Adaptation
Concept A
Concept B
Concept C
n Each topic is associated with a number of educational activities to learn about this topic
n Each activity classified under 1 topic
QuizGuide: Adaptive Annotations • Target-arrow abstraction:
– Number of arrows – level of knowledge for the specific topic (from 0 to 3). Individual, event-based adaptation.
– Color Intensity – learning goal (current, prerequisite for current, not-relevant, not-ready). Group, time-based adaptation.
n Topic–quiz organization:
QuizGuide: Success Rate
n It works! n One-way ANOVA shows
that mean success value for QuizGuide is significantly larger then the one for QuizPACK: F(1, 43) = 5.07 (p-value = 0.03).
QuizGuide: Motivation
• Adaptive navigation support increased student's activity and persistence of using the system
Average activity
050
100150200250300
2002 2003 2004
Average num. of sessions
0
5
10
15
20
2002 2003 2004
Average course coverage
0%10%20%30%40%50%60%
2002 2003 2004
Active students
0%
20%
40%
60%
80%
100%
2002 2003 2004
n Within the same class QuizGuide session were much longer than QuizPACK sessions: 24 vs. 14 question attempts at average.
n Average Knowledge Gain for the class rose from 5.1 to 6.5
A new value of ANS?
• The scale of the effect is too large… May be just a good luck?
• New effect after 15 years of research? • Maybe the effect could only be
discovered in full-scale classroom studies – while past studies were lab-based?
Round 2: Let’s Try it Again…
• Another study with the same system – QuizGuide+QuizPACK vs. QuizPACK
• A study with another system using similar kinds of adaptive navigation support – NavEx+WebEx vs. WebEx
• NavEx - a value-added ANS front-end for WebEx - interactive example exploration system
WebEx - Code Examples
Concept-based student modeling
Example 2 Example M
Example 1
Problem 1
Problem 2 Problem K
Concept 1
Concept 2
Concept 3
Concept 4
Concept 5
Concept N
Examples
Problems
Concepts
NavEx = WebEx + ANS
Does it work?
• The increase of the amount of work for the course
Clicks - Overall
0
50
100
150
200
250
300
Non-adaptive Adaptive
Examples
Quizzes
Lectures - Overall
0
2
4
6
8
10
12
Non-adaptive Adaptive
Examples
Quizzes
Learning Objects - Overall
0
5
10
15
20
25
30
Non-adaptive Adaptive
Examples
Quizzes
Is It Really Addictive?
• Are they coming more often? Mostly, but there is no stable effect
• But when they come, they stay… like with an addictive game
Clicks - Per Session
0
5
10
15
20
Non-adaptive Adaptive
Examples
Quizzes
Learning Objects - Per
Session
0
1
2
3
4
Non-adaptive Adaptive
Examples
Quizzes
Why It Is Working?
• Progress-based annotation – Displays the progress achieved so far – Does it work as a reward mechanism? – Open Student Modeling
• State-based annotation – Not useful, ready, not ready – Access activities in the right time – Appropriate difficulty, keep motivation
A Deeper Look
The Diversity of Work
• C-Ratio: Measures the breadth of exploration
• Goal distance: Measures the depth
Self-motivated Work - C-Ratio
(%)
0
0.2
0.4
0.6
Non-adaptive Adaptive
Quizzes
Examples
Self-motivated Work - Goal
Distance (LO's)
0
5
10
15
20
Non-adaptive Adaptive
Quizzes
Examples
Round 3: Trying another domain…
• Is it something relevant to C programming or to simple kind of content?
• New changes: – SQL Programming instead of C – Programming problems (code writing) instead of
questions (code evaluation) – Comparison of concept-based and topic-based
mechanisms in the same domain and with the same kind of content
• SQL-KnoT delivers online SQL problems, checks student’s answers and provides a corrective feedback
• Every problem is dynamically generated using a template and a set of databases
• All problems have been assigned to 1 of the course topics and indexed with concepts from the SQL ontology
SQL Knowledge Tester
• To investigate possible influence of concept-based adaptation in the present of topic-based adaptation we developed two versions of QuizGuide:
Topic-based Topic-based+Concept-Based
Concept-based vs Topic-based ANS
• Two Database Courses (Fall 2007): § Undergraduate (36 students) § Graduate (38 students)
• Each course divided into two groups: § Topic-based navigation § Topic-based + Concept-Based Navigation
• All students had access to the same set of SQL-KnoT problems available in adaptive (QuizGuide) and in non-adaptive mode (Portal)
Study Design
• Total number of attempts made by all students: in adaptive mode (4081), in non-adaptive mode (1218)
• Students in general were much more willing to access the adaptive version of the system, explored more content with it and to stayed with it longer:
Questions
0255075100
Quizzes
0510152025 Topics
0123456
Sessions
012345 Session Length
0510152025
Adaptive Non-adaptive
It works again! Like magic…
Round 4: The Issue of Complexity • Let’s now try it for Java…
• What is the research goal?
• Java is a more sophisticated domain than C – OOP versus Procedural – Higher complexity
• Will it work for complex questions?
• Will it work similarly? 0% 20% 40% 60% 80% 100%
C
Java
language complexity
Easy
Moderate
Hard
Meet QuizJET!
Naviga&on Area Presenta&on Area
JavaGuide
!! !! JavaGuide
(Fall 2008) QuizJET
(Spring 2008) !! parameters (n=22) (n=31)
Overall User Statistics
Attempts 125.50 41.71 Success Rate 58.31% 42.63% Distinct Topics 11.77 4.94 Distinct Questions 46.18 17.23
Average User Session Statistics
Attempts 30.34 21.50 Distinct Topics 2.85 2.55 Distinct Questions 11.16 8.88
Magic… Here We Go Again!
Round 5: Social Navigation
• Concept-based and topic-based navigation support work well to increase success and motivation
• Knowledge-based approaches require some knowledge engineering – concept/topic models, prerequisites, time schedule
• In our past work we learned that social navigation – guidance extracted from the work of a community of learners – might replace knowledge-based guidance
• Social wisdom vs. knowledge engineering
Open Social Student Modeling
• Key ideas – Assume simple topic-based design – No prerequsites or concept modeling – Show topic- and content- level knowledge progress of
a student in contrast to the same progress of the class
• Main challenge – How to design the interface to show student and class
progress over topics? – We went through several attempts
Parallel Introspective Views
40
0
40
80
120
160
QuizJET+IV QuizJET+Portal JavaGuide
Attempts
Attempts
Results: Progress
F(1,32)= 11.303, p<.01
71.35%
42.63% 58.31%
0.00%
20.00%
40.00%
60.00%
80.00%
100.00%
QuizJET+IV QuizJET+Portal JavaGuide
Success Rate
Success Rate
Results: Success
Class vs. Peers
• Peer progress was important, students frequently accessed content using peer models
• The more the students compared to their peers, the higher post-quiz scores they received (r= 0.34 p=0.004)
• Parallel IV didn’t allow to recognized good peers before opening the model
• Progressor added clear peer progress
Progressor
44
The Value of Peers
205.73
113.05
80.81
125.5
0
50
100
150
200
250
Attempts
Progressor
QuizJET+IV
QuizJET+Portal
JavaGuide
68.39% 71.35%
42.63%
58.31%
0.00%
20.00%
40.00%
60.00%
80.00%
Success Rate
Progressor
QuizJET+IV
QuizJET+Portal
JavaGuide
The Secret
Take-home messages
• A combination of progress-based and state-based adaptive link annotation increases the amount and the diversity of student work with non-mandatory educational content
• The effect is stable and the scale of it is quite large
• Properly organized Social Navigation might be at least as successful as the knowledge-based
• Requires a long-term classroom study to observe
Why It Is Important?
• Many systems demonstrated their educational effectiveness in a lab-like settings: once the students are pushed to use it - it benefits their learning
• However, once released to real classes, these systems are under-used - most of them offer additional non-mandatory learning opportunities
• “Students are only interested in points and grades” • Convert all tools into credit-bearing activities?
• Or use alternative approaches to increase motivation
What we are doing now?
• Exploring new generation of open social modeling tools in wide variety if classes and domains from US to Nigeria – Interested to be a pilot site?
• Exploring more advanced guidance and modeling approaches based on large volume of social data
• Applying open social modeling to motivate readings
Acknowledgements
• Joint work with – Sergey Sosnovsky – Michael Yudelson – Sharon Hsiao
• Pitt “Innovation in Education” grant
• NSF Grants – EHR 0310576 – IIS 0426021 – CAREER 0447083
Try It!
• http://adapt2.sis.pitt.edu/kt/
• Brusilovsky, P., Sosnovsky, S., and Yudelson, M. (2009) Addictive links: The motivational value of adaptive link annotation. New Review of Hypermedia and Multimedia 15 (1), 97-118.
• Hsiao, I.-H., Sosnovsky, S., and Brusilovsky, P. (2010) Guiding students to the right questions: adaptive navigation support in an E-Learning system for Java programming. Journal of Computer Assisted Learning 26 (4), 270-283.
• Hsiao, I.-H., Bakalov, F., Brusilovsky, P., and König-Ries, B. (2013) Progressor: social navigation support through open social student modeling. New Review of Hypermedia and Multimedia [PDF]
Read About It!
