Carnegie Mellon Project LISTEN16/29/2004 Which Help Helps? Effects of Various Types of Help on Word Learning in an Automated Reading Tutor that Listens.

Slides:



Advertisements
Similar presentations
Using the Four Block Literacy Framework for Students with Disabilities
Advertisements

PGCE Lecture Systematic Synthetic Phonics. Teachers’ Standards PART ONE: TEACHING A teacher must: TS3 Demonstrate good subject and curriculum knowledge.
Chapter 6—Phonics Kendra McLaren Doug McLaren
LITERACY IN PRIMARY/JUNIOR DIANE NEWMAN PROFESSIONAL DEVELOPMENT OECTA.
Learning decomposition WARNING. Goals Understand what learning decomposition is –And basic intuition See how it was applied to a variety of problems Think.
Continuous Improvement in Teaching and Learning Candace Thille Director, Open Learning Initiative.
Carnegie Mellon 1. Toward Exploiting EEG Input in a Reading Tutor Jack Mostow, Kai-min Chang, and Jessica Nelson Project LISTEN
Mining Data from Randomized Within-Subject Experiments in an Automated Reading Tutor Joseph E. Beck and Jack Mostow Project LISTEN (
Detecting Prosody Improvement in Oral Rereading Minh Duong and Jack Mostow Project LISTEN Carnegie Mellon University The research.
Carnegie Mellon Project LISTEN 17/22/2004 Some Useful Design Tactics for Mining ITS Data Jack Mostow Project LISTEN (
1 Are all questions created equal?: Factors that influence cloze question difficulty. Brooke Soden Hensler Carnegie Mellon University (starting graduate.
Lessons from generating, scoring, and analyzing questions in a Reading Tutor for children Jack Mostow Project LISTEN (
Carnegie Mellon Project LISTEN17/22/2004 If I Have a Hammer: Computational Linguistics in a Reading Tutor that Listens Jack Mostow Project LISTEN (
How to Generate Cloze Questions from Definitions: a Syntactic Approach Donna Gates, Gregory Aist (Iowa State University), Jack Mostow, Margaret McKeown.
Megha Cynthia Shyam Hannah Jen. What is it? Carnegie Learning, Inc. Math curriculum Cognitive tutor is software For middle & high schools as well as homeschooling.
Social Action Plan Objectives: The main aim of the project in to enable student to identify and raise awareness about the autism and inculcate love, affection.
Recommendations for Morgan’s Instruction Instruction for improving reading fluency Instruction for improving word recognition, word decoding, and encoding.
Early Identification and Intervention to Prevent Reading Difficulties Linda Siegel University of British Columbia Vancouver, CANADA
1 SPACE Reading Tutor: Design and Functionality Leen Cleuren Jacques Duchateau Pol Ghesquière Hugo Van hamme.
Foundational Skills Module 4. English Language Arts Common Core State Standards.
Part Three: Try It Out! Team member’s name: Melissa MayoName of strategy implemented: TPSI With whom the strategy was implemented (grade-level/content.
Capstone Experience at UNH Manchester Student Guided Mentoring for an Undergraduate Research Group in Speech Capstone Objectives Challenges Technology.
SmartGraphs CyTSE Conference – March 8, 2011 This material is based upon work supported by the National Science Foundation under Grant No. DRL ==≠≠
MS STATE UNIVERSITY DELVRISHA MAGEE EDU 505: FUTURE OF EDUCATION DR. LINDA KAISER Vision for Education
EVALUATION OF SOCIAL SCIENCE 5TH GRADE TEXTBOOK BY THE STUDENTS Akdoğan Dr. Fazıl Küçük PREPARED BY: CLASS 5.
Welcome et Bienvenue Introduction of Literacy Support Team: Mme Pam, Mme Robyn and Mme Cristina.
DIBELS: Dynamic Indicators of Basic Early Literacy Skills 6 th Edition A guide for Parents.
1 Computational Linguistics Ling 200 Spring 2006.
Being Smart with Graphs This material is based upon work supported by the National Science Foundation under Grant No. DRL Any opinions, findings,
Challenges of Mathematics in Urban Schools Opportunities of Technology and Research Partnerships OliveAnn D. Slotta, Ph.D. Denver Public Schools Denver,
Improving the Help Selection Policy in a Reading Tutor that Listens Cecily Heiner, Joseph E. Beck, Jack Mostow Project LISTEN
AUTHORS: María Eugenia Guerrero Andrade Martha Catalina Puga Cevallos ADVISORS: Director: MS. Miguel Ponce Medina Co-Director: MG. Néstor Bonilla Bonilla.
Being Smart with Graphs This material is based upon work supported by the National Science Foundation under Grant No. DRL ==≠≠ == Any opinions,
Developing an automated assessment tool for children’s oral reading Leen Cleuren March
Carnegie Mellon Mostow 12/7/2015, p. 1 The Sounds of Silence: Towards Automated Evaluation of Student Learning in a Reading Tutor that Listens Jack Mostow.
Transforming lives through learning POLAAR Marion Cochrane, Development Officer for English and Literacy, Dyslexia Conference,
Practice-Based Professional Development Daniel C. Edelson GEODE Initiative School of Education and Social Policy Northwestern University.
Carnegie Mellon How does the amount of context in which words are practiced affect fluency growth? Experimental results Jack Mostow, Jessica Nelson, Martin.
Phonics Instruction by Chuck Branch. Phonics Instruction While the National Reading Panel found it essential that a planned sequence be taught explicitly,
What are my next steps? Incorporating universal design into your own work and advocating for accessibility in your own institution NISE Net Universal Design.
Action Research Chantal Smith Liberton Christian School Is Blogging a Useful Tool for Supporting and Monitoring the Personal Reading of my Year 6-8 students?
Sketched-Truss Recognition Tutoring System: Improved Student Learning through Active Learning and Immediate Student Feedback, NSF DUE , 3/10-2/13.
Phonics Training for Parents October What is phonics?  Letter sounds  Government initiative  Good phonics = good reading, writing and spelling.
LST. Collective Responsibility Concentrated Instruction Convergent Assessment Certain Access RTI and Reading.
Engineering programs must demonstrate that their graduates have the following: Accreditation Board for Engineering and Technology (ABET) ETP 2005.
OCTOBER 16, 2014 Milton School. Decoding Inferential Comprehension Critical Comprehension Love of Reading Literal Comprehension Word Study, Vocabulary,
State Board of Education Achievement and Graduation Requirements Committee October 19, 2015.
Trial Teaching Strategies: Linking Testing to Teaching Mary Beth Curtis Center for Special Education March 31, 2009.
Welcome to ‘Supporting your child with Reading’
Wei Chen, Jack Mostow Gregory Aist Project LISTEN
Turning Around 1,000 Schools: The Story of Success for All
Discussion and Conclusion
Micro-analysis of Fluency Gains in a Reading Tutor that Listens:
Detecting Prosody Improvement in Oral Rereading
Warm-Up: Take a ¼ sheet of paper.
Lessons from Project LISTEN: What have we learned from a Reading Tutor that listens? Jack Mostow, Director Project LISTEN ( Carnegie.
Critical - thinking Assessment Test (CAT)
Introduction to SAGE 2YC
Independent versus Computer-Guided Oral Reading:
An Embedded Experiment to Evaluate the Effectiveness of Vocabulary Previews in an Automated Reading Tutor Jack Mostow, Joe Beck, Juliet Bey, Andrew Cuneo,
Neil T. Heffernan, Joseph E. Beck & Kenneth R. Koedinger
Jack Mostow* and Joseph Beck Project LISTEN (
Warm-Up: Take a ¼ sheet of paper.
Educational Data Mining Success Stories
Experimenter-defined measures in a Reading Tutor that Listens
Data-Driven Decision-Making
IERI educational data mining panel
The 4 systems that “clue” us into making meaning!
This material is based upon work supported by the National Science Foundation under Grant #XXXXXX. Any opinions, findings, and conclusions or recommendations.
Project Title: I. Research Overview and Outcome
Presentation transcript:

Carnegie Mellon Project LISTEN16/29/2004 Which Help Helps? Effects of Various Types of Help on Word Learning in an Automated Reading Tutor that Listens Jack Mostow, Joseph E. Beck, Cecily Heiner Project LISTEN ( Carnegie Mellon University Presented 6/29/2004 at 11 th Annual Meeting of the Society for the Scientific Study of Reading, Amsterdam. This work was supported in part by the National Science Foundation under ITR/IERI Grant No. REC Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the National Science Foundation or the official policies, either expressed or implied, of the sponsors or of the United States Government.

Carnegie Mellon Project LISTEN26/29/2004 Previous findings on word help Spaai (1991):  Whole word feedback beat sounding out Wise (1992):  Whole word, syllables, sub-syllables beat sounding out  Onset + Rime (c+at) best for less severely disabled  Syllables best for more severely disabled But:  Limited samples of students  Limited data per student

Carnegie Mellon Project LISTEN36/29/2004 Project LISTEN’s Reading Tutor: A rich source of experimental data The Reading Tutor beats independent practice…  Effect sizes up to 1.3 (Mostow SSSR02, Poulsen 04)Mostow SSSR02Poulsen 04 … but how? Embed an experiment to find out!  9 schools ( )  200 computers  600 students, age 6-12  Average 9.2 hours each  Detailed interaction logs  460,000 words helped  5 million words read aloud  Scored using automatic speech recognition (ASR)  “Invisible experiments”  Randomized trials Frey, K. (October 17, 2003). CMU's Automated Reading Tutor Helps Children (2m16s). On Evening News. Pittsburgh, PA: WTAE. At

Carnegie Mellon Project LISTEN46/29/2004 Within-subject experiment design: 270 students, 180,909 randomized trials Outcome: success = ASR accepts word within first utterance (How) does the type of help affect the next encounter? Randomized choice among feasible types Student clicks “read.” “I love to read stories.” “People sit down and …” “… read a book.” Student is reading a story Student needs help on a word Tutor chooses what help to give Student continues reading Student sees word in a later sentence Time passes…

Carnegie Mellon Project LISTEN56/29/ ,909 word hints (average success rate 66.1%) Whole word:  56,791 Say WordSay Word  24,841 Say In ContextSay In Context Decomposition:  6,280 SyllabifySyllabify  14,223 Onset RimeOnset Rime  19,677 Sound OutSound Out  22,933 One GraphemeOne Grapheme Analogy:  13,165 Rhymes WithRhymes With  13,671 Starts LikeStarts Like Semantic:  14,685 RecueRecue  2,285 Show PictureShow Picture  488 Sound EffectSound Effect Which types stood out?  Best: Rhymes With 69.2% ± 0.4%Rhymes With  Worst: Recue 55.6% ± 0.4%Recue Example: People sit down and read a book.

Carnegie Mellon Project LISTEN66/29/2004 What help did students prefer? Students interrupted 30% of help to get other help. Interruption varied by help type.  Least interrupted (14%): Say WordSay Word  Most interrupted (67%): RecueRecue Success was lower when students interrupted.  If interrupted: 58.5% ± 0.1%  If not: 69.4% ± 0.4%  Difference presumably due to student effects

Carnegie Mellon Project LISTEN76/29/2004 What helped the same day vs. later? Success rates varied with time till next encounter.  67.4% ± 0.1% for 105,424 on same day  64.3% ± 0.2% for 75,485 on later day Time till next encounter varied by help type.  Soonest (72% same day): Sound Effect (frequent words)Sound Effect  Latest (48% same day): Syllabify (rarer words)Syllabify Best help types varied with time till next encounter.  Same day: Say In Context 71.5% ± 0.4%Say In Context  Later day: Rhymes With 67.3% ± 0.6%Rhymes With Caveats:  Same day outcomes include recency effects.  Later day outcomes skew toward harder words.

Carnegie Mellon Project LISTEN86/29/2004 What helped which words best? Same day:Later day: Grade 1 words:Say In ContextSay In Context, Onset Rime Grade 2 words:Rhymes WithRhymes With, Say In Context Say In Context Rhymes With Grade 3 words:Say In ContextOne GraphemeOne Grapheme, Rhymes With Rhymes With Compare within level to control for word difficulty.

Carnegie Mellon Project LISTEN96/29/2004 Conclusions Reading Tutor is a powerful research platform  Hundreds of thousands of randomized trials  Millions of read words  Detected subtle effects despite imperfect ASR Embedded experiments have tricky confounds  Student behavior, recency, word difficulty  So design and analyze with caution!  See Joe Beck’s poster and Type of word help mattered measurably  Recue took longest and helped least Recue  Say In Context helped in the short term Say In Context  Rhymes With had longer lasting benefits Rhymes With

Carnegie Mellon Project LISTEN106/29/2004 The Project LISTEN team  Director:  Jack Mostow  Tutoring:  Dr. Joseph Beck, mining tutorial data  Prof. Albert Corbett, cognitive tutors  Prof. Rollanda O’Connor, reading  Prof. Kathy Ayres, stories for children  Joe Valeri, activities and interventions  Becky Kennedy, linguist  Listening:  Dr. Mosur Ravishankar, recognizer  Dr. Evandro Gouvea, acoustic training  John Helman, transcriber  Programmers:  Andrew Cuneo, application  Karen Wong, Teacher Tool  Field staff:  Dr. Roy Taylor  Kristin Bagwell  Julie Sleasman  Grad students:  Hao Cen, HCI  Cecily Heiner, MCALL  Peter Kant, Education  Shanna Tellerman, ETC  Plus:  Research partners  DePaul  UBC  U. Toronto  Advisory board  Teachers  Children