17
Applying Automated Analysis to Develop a Cost-Effective Measure of Science Teacher Pedagogical Content Knowledge NARST Baltimore, 16 April 2016 Zoë Buck Bracey, Molly Stuhlsatz, Chris Wilson (BSCS) Mark Urban-Lurain, John Merrill, Kevin Haudek (MSU)

Applying Automated Analysis to Develop a Cost … Automated Analysis to Develop a Cost -Effective Measure of Science Teacher Pedagogical Content Knowledge NARST Baltimore, 16 April

Embed Size (px)

Citation preview

Applying Automated Analysis to Develop a Cost-Effective Measure of Science Teacher Pedagogical Content Knowledge

NARSTBaltimore, 16 April 2016

Zoë Buck Bracey, Molly Stuhlsatz, Chris Wilson (BSCS)Mark Urban-Lurain, John Merrill, Kevin Haudek (MSU)

DefiningPCK(PedagogicalContentKnowledge)

• [K]nowledgeof,reasoningbehind,and planning forteachinginaparticular topic inaparticularway foraparticularpurpose toparticularstudents forenhancedstudentoutcomes(Gess-Newsomeetal.,2015,p.36).

• PCKhasbeenshowntobeasignificantpredictorofbothreform-basedclassroompracticeandstudentachievementinscience(Rothetal.,2011).

ResearchQuestion

Howcanlexicalanalysisandmachinelearningtechniquesbeappliedtodevelopinganefficient,valid,andreliablemeasureofteacherpedagogicalcontentknowledge?

TheAssessmentTriangle

Cognition:FrameworkforPCKconstruct

Observation:Teacher

responsestoclassroomvideo

AssessmentTriangle

Interpretation:Automatedanalysisofteacher

responses

Adaptedfrom:NationalResearchCouncil (NRC).(2001).Knowingwhatstudentsknow:Thescienceanddesignofeducationalassessment.Washington,DC:NationalAcademyPress.

TheAssessmentTriangle

Cognition:FrameworkforPCKconstruct

Observation:Teacher

responsestoclassroomvideo

AssessmentTriangle

Interpretation:Automatedanalysisofteacher

responses

Observation:Teacherresponsestoclassroomvideo

• Innovativevideo-basedlessonanalysistask

• Teacherswatchvideoclipsofsciencelessons

• Clipscarefullyselectedfromauthenticclassroomvideotoincludearangeofstudentactivityandteachermoves

• Teachersrespondtopromptstomakeanalyticalcommentsaboutwhattheysee

TheAssessmentTriangle

Cognition:FrameworkforPCKconstruct

Observation:Teacher

responsestoclassroomvideo

AssessmentTriangle

Interpretation:Automatedanalysisofteacher

responses

Cognition:FrameworkforPCKconstructDimension OverviewContextualized analysis of student thinking (CAST)

noticing student thinking re: domain content and/or bringing in knowledge of common patterns of student thinking in the domain

Contextualized analysis of instructional coherence (CAIC)

noticing how the teacher does/does not maintain a coherent content storyline

Contextualized analysis of constructivist instructional strategies (CACIS)

noticing how the teacher does/does not invoke teaching strategies that allow students to construct their own understanding individually or in groups

Contextualized analysis of responsive teaching (CART)

noticing how the teacher does/does not respond to student thinking through formative assessment, “teachable moments,” or allowing inquiry to be student led

Contextualized analysis of NGSS practices (CANP)

noticing how classroom activity does/does not align with any of the eight NGSS practices (does not need to explicitly mention NGSS)

Contextualized analysis of scientific discourse and language in the classroom (CASD)

noticing how science/everyday language is/is not used/scaffolded in the classroom

TheAssessmentTriangle

Cognition:FrameworkforPCKconstruct

Observation:Teacher

responsestoclassroomvideo

AssessmentTriangle

Interpretation:Automatedanalysisofteacher

responses

Examplesfromthedata

Theteacherinthisclipseemedtohaveanicecollaborativeclassroom.Studentswereontaskandaskingquestionsiftheydidn’tunderstandsomething.Theteacherinteractedwithonegroupandwalkedthemthroughthecondensationprocess.Studentsalsostoodupwhentheyspokeout.TheonlythingIquestionhereis,shestoodwithonegroupthewholetimeduringtheclip.Notsureifshewentaroundmonitoringothergroups.

StudentThinking=0InstructionalCoherence=0ConstructivistStrategies=0ResponsiveTeaching=0NGSSPractices=0ScientificDiscourse=0

Therearesomescienceideasthatthestudentsunderstandandsomemisconceptionsthatthestudentshaveaboutwatermolecules.At02:36:00,thestudentsaysthatyoucan’tseewatervapor.Thatisascienceideathattheyunderstand.Andat05:07:00,thestudentsaysthatmoremoleculearecomingtogetherandthenitturnsintowater. HereIwouldprobethestudent’sthinkingbyaskinghimwhatwatervaporisthen.Isitnotwatertoo?At05:25.22,theteacheraskshowaretheygoingtobecomeliquidwater,Ilikethatsheusedthewordliquidwater.It’simportanttousethevocabularythatyouwantyourstudentsusing.

Examplesfromthedata

Therearesomescienceideasthatthestudentsunderstandandsomemisconceptionsthatthestudentshaveaboutwatermolecules.At02:36:00,thestudentsaysthatyoucan’tseewatervapor.Thatisascienceideathattheyunderstand.Andat05:07:00,thestudentsaysthatmoremoleculearecomingtogetherandthenitturnsintowater. HereIwouldprobethestudent’sthinkingbyaskinghimwhatwatervaporisthen.Isitnotwatertoo?At05:25.22,theteacheraskshowaretheygoingtobecomeliquidwater,Ilikethatsheusedthewordliquidwater.It’simportanttousethevocabularythatyouwantyourstudentsusing.

Examplesfromthedata

StudentThinking=1

Therearesomescienceideasthatthestudentsunderstandandsomemisconceptionsthatthestudentshaveaboutwatermolecules.At02:36:00,thestudentsaysthatyoucan’tseewatervapor.Thatisascienceideathattheyunderstand.Andat05:07:00,thestudentsaysthatmoremoleculearecomingtogetherandthenitturnsintowater. HereIwouldprobethestudent’sthinkingbyaskinghimwhatwatervaporisthen.Isitnotwatertoo?At05:25.22,theteacheraskshowaretheygoingtobecomeliquidwater,Ilikethatsheusedthewordliquidwater.It’simportanttousethevocabularythatyouwantyourstudentsusing.

Examplesfromthedata

StudentThinking=1InstructionalCoherence=0ConstructivistStrategies=0ResponsiveTeaching=1

Therearesomescienceideasthatthestudentsunderstandandsomemisconceptionsthatthestudentshaveaboutwatermolecules.At02:36:00,thestudentsaysthatyoucan’tseewatervapor.Thatisascienceideathattheyunderstand.Andat05:07:00,thestudentsaysthatmoremoleculearecomingtogetherandthenitturnsintowater. HereIwouldprobethestudent’sthinkingbyaskinghimwhatwatervaporisthen.Isitnotwatertoo?At05:25.22,theteacheraskshowaretheygoingtobecomeliquidwater,Ilikethatsheusedthewordliquidwater.It’simportanttousethevocabularythatyouwantyourstudentsusing.

Examplesfromthedata

StudentThinking=1InstructionalCoherence=0ConstructivistStrategies=0ResponsiveTeaching=1NGSSPractices=0ScientificDiscourse=1

TestBlueprintStudent Thinking

Instructional Coherence

Constructivist Strategies

Responsive Teaching

NGSS Practices

Scientific Discourse

Video 1 x x

Video 2 x x

Video 3 x x

Video 4 x x

Video 5 x x

. . .

Video 12 x x

• NextstepsintheQDCdevelopmentcycle:LexicalResourceDevelopment

MSUAACRQuestionDevelopmentCycle

ContactandAcknowledgement

ThismaterialisbaseduponworksupportedbytheNationalScienceFoundation underGrant1437173.Anyopinions,findings,andconclusionsorrecommendationsexpressedinthismaterialarethoseoftheauthor(s)anddonotnecessarilyreflecttheviewsoftheNationalScienceFoundation

Zoë[email protected]

[email protected]

[email protected]

[email protected]

[email protected]

[email protected]