Lessons from generating, scoring, and analyzing questions in a Reading Tutor for children Jack Mostow Project LISTEN (www.cs.cmu.edu/~listen)www.cs.cmu.edu/~listen.

Slides:



Advertisements
Similar presentations
Inquiry-Based Instruction
Advertisements

Take a piece of pizza from the counter.
Purpose : To create a fail-safe system of literacy so that all students have equal access to a standards based curriculum Result: Joyful, independent readers,
Chapter 14 Narrative Reading Joe Steele Helping students to recognize the structure inherent in text – and match it to their own cognitive structures –
Department of Mathematics and Science
Learning to Read What separate processes are involved in someone becoming a skilled reader?
Listening Comprehension Instruction
The New English Curriculum
6 Steps to Building Academic Vocabulary Robert J. Marzano and Debra J. Pikening Laredo Independent School District.
Teaching and Monitoring Comprehension in the early grades Leecy Wise
Tutoring and Educational Applications 2pm: Question Generation Based on Numerical Entities in Basque (Itziar Aldabe, Montse Maritxalar, Ander Soraluze)
Teaching Comprehension in the early grades Leecy Wise
 Running are a method of recording a student’s reading behavior. Running Records provide teachers with information that can be analyzed to determine.
26 March Independent STUDIES 100 % by Friday CLICK TO BEGIN Questions: Click Here 5 SOLUTIONS Directions: 1. Click on play. 2. Click to begin. 3. Click.
Close Reading Preparing for the arrival of Common Core Standards in Social Studies.
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.
How to Generate Cloze Questions from Definitions: a Syntactic Approach Donna Gates, Gregory Aist (Iowa State University), Jack Mostow, Margaret McKeown.
Beginning Oral Language and Vocabulary Development
How To Teach Reading To Adults For Teachers and Tutors Edward Fry, PH.D.
Grade 3: Comprehension The material in this Institute has been modified from the Florida Third Grade Teacher Academy which was based upon the original.
April 17,18, ELA Exam Overview Three day exam- April 17,18,19 for Grades 3-8 Day 1- Reading Day 2- Listening Day 3 –Reading/ Writing.
10 Things Every Teacher Should Know About Reading Comprehension 10 Things Every Teacher Should Know About Reading Comprehension Timothy Shanahan University.
Section VI: Comprehension Teaching Reading Sourcebook 2 nd edition.
ESL Teaching and Reading Strategies
Reading in the Upper Grades
Benefits from Formal and Informal Assessments
Narrative Reading By Lorie Sadler. Narrative Reading What Why When How.
The Secrets of Guided Reading (In Lower Elementary) Miss Allison Dalton 1 st Grade Teacher Discovery Elementary School.
English Language Development Assessment (ELDA) Background to ELDA for Test Coordinator and Administrator Training Mike Fast, AIR CCSSO/LEP-SCASS March.
Level 1: Chapter 7.  Add more study strategies to a tutor’s repertoire of skills.  Be able to apply relevant skills to tutoring and academic work.
Prevention to Avoid Intervention Tier 1: the most important tier!
Ideas and Activities to Differentiate Instruction through Strategies
4th & 5th Grade Coffee January 27, Levels are determined by benchmarking, MAP testing, anecdotal notes and MCAS. Assessment informs instruction.
Framework for Diagnostic Teaching. Framework The framework for diagnostic teaching places a premium on tailoring programs that specifically fit all readers.
Comprehension Strategies
Increasing Reading Vocabulary
Kindergarten Reading Comprehension AP3.  Once signed in on the PMRN, the K-2 Demo link is at the left on the main homepage.  The K-2 Demo has first.
Monitoring Comprehension Teaching Comprehension Strategies to Students.
T 7.0 Chapter 7: Questioning for Inquiry Chapter 7: Questioning for Inquiry Central concepts:  Questioning stimulates and guides inquiry  Teachers use.
The New English Curriculum September The new programme of study for English is knowledge-based; this means its focus is on knowing facts. It is.
Chapter 14 Narrative Reading
Improving the Help Selection Policy in a Reading Tutor that Listens Cecily Heiner, Joseph E. Beck, Jack Mostow Project LISTEN
The 5 E’s Science Lesson Inquiry-Based Instruction.
Strategies SIOP Component #4
Reading Comprehension and Vocabulary Development November 3, 2005.
First Grade Reading Workshop
Assessment. Workshop Outline Testing and assessment Why assess? Types of tests Types of assessment Some assessment task types Backwash Qualities of a.
The Power of Reading through SSR. Questions You Might Have What is SSR? What do you want me to do during SSR? Why are we taking time out of the school.
PSAT/NMSQT th Grade Advisement. What is the PSAT? The Preliminary SAT/National Merit Scholarship Qualifying Test (PSAT/NMSQT) The Preliminary.
Key Stage 1 SATs. ‘Old’ national curriculum levels (e.g. Level 3, 4, 5) have now been abolished, as set out in the government guidelines. From 2016, test.
Carnegie Mellon How does the amount of context in which words are practiced affect fluency growth? Experimental results Jack Mostow, Jessica Nelson, Martin.
CREATING AN ACTIVE LEARNING ENVIRONMENT Using Inquiry and Primary Sources.
Goals 1. To understand inquiry 2. To learn about inquiry-based science 3. To compare children’s science and scientists’ science. 4. To compare two methods.
1 Instructing the English Language Learner (ELL) in the Regular Classroom.
What is Inquiry in Science?. Goals 1. To understand nature of science as inquiry 2. To learn about inquiry as a model of teaching 3. To compare inquiry.
OCTOBER 16, 2014 Milton School. Decoding Inferential Comprehension Critical Comprehension Love of Reading Literal Comprehension Word Study, Vocabulary,
Demo.
We will memorize1 multiplication facts.
Improving inference and comprehension skills
Detecting Prosody Improvement in Oral Rereading
How To Teach Reading To Adults
An Embedded Experiment to Evaluate the Effectiveness of Vocabulary Previews in an Automated Reading Tutor Jack Mostow, Joe Beck, Juliet Bey, Andrew Cuneo,
KS2 SATS 2018.
Section VI: Comprehension
Moby Max.
Improving inference and comprehension skills
Educational Data Mining Success Stories
Reading in the Upper Grades
Reading at Lydgate Infant School
Presentation transcript:

Lessons from generating, scoring, and analyzing questions in a Reading Tutor for children Jack Mostow Project LISTEN ( AAAI Symposium on Question Generation keynote, Nov. 5, 2011, Arlington, VA Questions and Answers The research reported here was supported by the Institute of Education Sciences, U.S. Department of Education, through Grants R305A The opinions expressed are those of the authors and do not necessarily represent the views of the Institute or the U.S. Department of Education. about Questions and Answers

Questions about Questions Target: What does it take to answer the questions? Purpose: Why ask the questions? Question type: In what form will the questions be output? Answer type: In what form will responses be input? Generation: How construct questions, answers, distracters? Modality: What channels will convey questions and answers? Assessment: How score answers? How generate feedback? Evaluation: How to tell how well questions serve purpose? 11/5/112Jack Mostow keynote

What can questions target? Reading  Decoding: In “Word Swap,” click on the “misread” word [Zhang ITS 08]  Comprehension: Click on the missing word. [Mostow TICL 04]  Inter-sentential prediction: Which will come next? [Beck ITS 04]  Monitor: Did that make sense?  Self-question: Why was the country mouse surprised? [Chen AIED 09]  Disengagement: hasty guessing [Beck AIED 05] Vocabulary  Recall: Which means most like ? [Aist 01]  Recognize: Which word means ? [TICL 04]  Remind: Definition cloze [Gates QG 11]  Disambiguate: What does mean here? Knowledge  Fact: Noiz arte du nahi duenak iritzia emateko aukera? [Aldabe QG 11]  Skill: How many grams can a worker ant carry? [Williams QG 11]  Concept: … The lid sat in the loft. was Jack Mostow keynote11/5/113

How has Project LISTEN used questions? 1.Assess comprehension [Mostow et al., TICL 04] 2.Help comprehension [Beck et al., ITS 04] 3.Assess engagement [Beck, AIED 05] 4.Teach self-questioning [Mostow & Chen, AIED 09] 5.Model self-questioning [Chen et al., ITS 10] 6.Assess self-questioning [Chen et al., QG 11] 7.Help vocabulary learning [Gates et al., QG 11] Jack Mostow keynote11/5/114

1. Assess comprehension [TICL 04] Target: comprehend sentence Purpose: assess comprehension while reading Source: sentence in text Question type: cloze Answer type: multiple choice among 4 words from text Generation: randomly pick sentence, word, and distracters Modality: play recorded sentence and words; click on one Assessment: original word? immediate correctness feedback Evaluation: correlate against standard comprehension test Jack Mostow keynote11/5/115

Student starts reading a story… Jack Mostow keynote11/5/116

Jack Mostow keynote Now and then during the story… 11/5/117

insert a cloze question…  Reading Tutor reads question and choices aloud  Target and distracters are words from same story Jack Mostow keynote11/5/118

Jack Mostow keynote … just before a sentence. 11/5/119

What do cloze questions test? Jack Mostow keynote11/5/1110

Jack Mostow keynote Oct.01 – Mar.02 evaluation data Reading Tutor asked 69,326 automated cloze questions  364 students in grades 1-9 at 7 schools  questions per student (median 136)  98% of questions answered – else Goodbye, Back, or timeout  24-88% correct (median 61%); 25% = chance How much guessing? Hasty responses [J. Valeri]:  3,078 (4.5%) faster than 3 seconds, only 29% correct  3.9% per-student mean, but below 1% for most students  Guessing rose from 1% in October to 11% by March 11/5/1111

Jack Mostow keynote Reliability: Guttman split-half test Split N responses into two halves  Match by word and story level Reliability increases with N .83 for N  10 (338 students) .95 for N  80 (199 students) Grade = /5/1112

Jack Mostow keynote What affects cloze difficulty? Similarity of distracters to answer  Part of speech [Hensler & Beck, ITS 06]  Semantic class  Consistency with local context  Consistency with inter-sentential context Vocabulary level of answer and distracters  “Sight words” = 225 most frequent words of English [Dolch list]  “Easy words” = 226…3,000  “Hard words” = 3,001…25,000  “Defined words” = marked as warranting explanation Text level of story  Grade K, 1, 2, 3, 4, 5, 6, 7 Cloze performance at 4 word levels x 8 text levels predicts Woodcock Reading Mastery Test comprehension (R =.84) 11/5/1113

2a. Help comprehension [ITS 04] Target: comprehend text by questioning Purpose: scaffold comprehension while reading Source: none Question type: Wh- Answer type: multiple choice Generation: scripted generic questions and choices Modality: play recorded prompt and words; click on one Assessment: none Evaluation: test efficacy on ensuing cloze performance Jack Mostow keynote11/5/1114

Generic Wh- questions: initial Generic prompt (meta-question):  Click on a question you can answer, or click Back to reread the sentence Generic 1-word questions as choices:  Who? What? When? Where? Why? How? So? Evaluation: failed 2002 user test  Meta-question confusing  1-word questions too vague to map to text Jack Mostow keynote11/5/1115

Generic Wh- questions: revised Randomly pick a generic prompt  What has happened so far? When does this take place? … List choices scripted for the prompt  facts were given; a problem is being solved; …  in the present; in the future; in the past; It could happen in the past; I can’t tell Jack Mostow keynote11/5/1116

2b. Help comprehension [ITS 04] Target: comprehend text by predicting Purpose: scaffold comprehension while reading Source: none Question type: Which will come next? Answer type: multiple choice Generation: answer = next sentence; distracters = 2 following Modality: play recorded prompt and sentences; click on one Assessment: original sentence? immediate correctness feedback Evaluation: 41% right; test efficacy on ensuing cloze performance Jack Mostow keynote11/5/1117

Sentence prediction Jack Mostow keynote11/5/1118

2003 evaluation data Reading Tutor asked 23,372 randomly inserted questions  252 students in grades 1-4+? who used for an hour or more  6,720 generic what, when, where  1,865 Which will come next? sentence prediction  15,187 cloze  On average, one question every 10 sentences Jack Mostow keynote11/5/1119

Evaluation results VariableHelps/Hurtsp # 3W questions # sentence predictions # cloze questions = # recent 3W questions  # recent cloze questions = Time since prior question (sec) Time since start of story (sec)  0.14 Jack Mostow keynote11/5/1120

Effect of recent questions Response time < 3 sec indicates disengagement Use to track (dis-)engagement [Beck AIED 05] Time since previous question (sec) Proportion of hasty responses Jack Mostow keynote11/5/1121

Effect of question type Number of prior questions Cloze performance Jack Mostow keynote11/5/1122

Questions about Questions Target: What does it take to answer the questions? Purpose: Why ask the questions? Question type: cloze? wh-/how/so/…? find/compare/…? Answer type: multiple choice? fill-in? open-ended? Generation: How construct questions, answers, distracters? Modality: menu? click? keyboard? speech? graphics? … Assessment: How score answers? How generate feedback? Evaluation: How to tell how well questions serve purpose? Generation Modality Assessment Generation Question Answer Distracter Modality Output Input Assessment Scoring Feedback NoneTrivialSimpleComplexManual Generation QuestionGenericClozeNLPScripted AnswerNoneGiven= textNLPScripted DistracterNoneGenericFrom textNLPScripted Modality OutputMenuTextTTS, visualPrerecorded InputNoneClickKeyboardASR, gestureTranscribed Assessment ScoringNoneM/C= answerNLPManual FeedbackNoneGenericanswerNLPScripted QG costs (lowest  highest) Jack Mostow keynote 11/5/1123

QG costs (lowest  highest) NoneTrivialSimpleComplexManual Generation QuestionGenericClozeNLPScripted AnswerNoneGiven= textNLPScripted DistracterNoneGenericFrom textNLPScripted Modality OutputMenuTextTTS, visualPrerecorded InputNoneClickKeyboardASR, gestureTranscribed Assessment ScoringNoneM/C= answerNLPManual FeedbackNoneGenericanswerNLPScripted 1. M/C cloze 2a. Generic Wh- 2b. Sentence prediction 7. Vocabulary reminders Jack Mostow keynote11/5/1124