1

APrimeronProductiveClassroomConversations

ccomplished musicians and master scienceteachershavesomethingincommon—theycanboth make complex performances look

effortless. The great jazz pianist, Thelonious Monk,would take song requests from the audience thenreinventthepieceasheplayed itbychangingthekey,tempo, and mood of the tune. At the right time hewouldbackoff themelody to letanotherplayer inhisensemble takecharge, thenlistenforsubtlerhythmiccuesthatitwashisturnagaintotakethelead;hisfingerswould dance over five octaves on the keyboardwhile he gazed out at the crowd,smiling.Monkhadadeepunderstandingofthefundamentalstructureofthemusic,butknewhow to shape and improvise the experience for the audience aswell as share theproductionwithothersinhisband.

The experienced science educator is no less an artistwhen it comes to classroom conversations. Consider a“simple”highschool laboratoryactivitythatbeginswiththeteacherplacingamassonascaleatthefrontoftheclassroom. The scale reads “10 kilograms.” He thenproducesalargebelljarwhichheplacesovertheentirescale and attaches the jar to a vacuum pump. “Can

anyone share their thinking about what the scalemight read if I pump all the airout?” Over the next 20 minutes he orchestrates a flow of discourse among hisstudentsthatcompelsthemtohypothesizeabouttheeffectsofairontheweightofobjects, tosuggest thoughtexperiments, tomakereasonedconnections, to tryoutand justify explanations with one another—in other words, to  think. He posesquestions thatprobe thementalmodelshis studentsarebeginningwith,assessinghowelaborate thesemodels are, howgeneralizable,whether they are referring toobservations or to theory. During this time the teacher constantly judgeswhetherthediscussion ismovingthestudents towardascientificwayof thinkingabout thephenomena.Hemustdecidewhohas“pieces”ofthescientificexplanationandhowto help students put these partial understandings together for themselves. Inadditiontoallthis,hemonitorswhetherstudentsarefollowingclassroomnormsforcivil conversation and the degree of involvement, puzzlement, or frustration ofindividualstudents.

A

2

Helpingkidsmakesenseofscienceisdone,inlargepart,by getting  students  to  talk. There are two reasons forthis.First,talkingistheprimarymodeofsense‐makinginhuman beings. Second, hearing kids’ talk, gives youaccess to their thinking and allows you to adaptinstruction to their current understandings. What kids

thinkisoftenvastlydifferentfromwhatyoubelievetheyarethinking.Howdoes onedevelop expertise like this? For any teacher, there are principles ofclassroom discourse that can be studied, practiced, and refined. The followingsections of this document will acquaint you with some fundamental ideas aboutproductivediscourse inclassrooms.Theycanserveasconceptual tools tohelpyouanalyze theclassroomdiscourseofother teachersand tobeginplanningyourownformsofconversationswithkids.Thebeginningsetofdiscourseideasweintroduceare:

•maintainingasafeclassroomenvironmentfordiscussions•primingyourselffordifferentconversations•cognitivedemandofquestionsandtasks•askingmeta‐cognitivequestions•usingwaittime•productivediscoursemoves•scaffoldingstudents’useofacademiclanguage•encouragingpeer‐to‐peertalk

Idea #1. Maintaining a safe classroom for student conversation

A safe classroom isone inwhich students feel that theywill nothavetheirideasridiculed,andthattheteacherandtheirpeerswillvalue what they have to say. Productive conversations requirestudentstotakerisksinpublic—tohypothesizeaboutthingstheyare only partially familiar with, to comment on the ideas ofclassmates, or to ask questions that may reveal a lack ofunderstanding. Because of this, the most basic pre‐requite for

productiveconversationsisthatallstudentsfeel safeinspeakingpublicly.Classrooms should have norms for civil discussions that are developed—with helpfromyourstudents—fromthefirstdayofschool,thatareexplicitlymodeledbytheteacher (i.e. the teacher “names” thenormas she/heuses it) and reinforcedon adailybasis.Herearesomesamplesofthesenorms—therearemanypossibilities:

•Anyonecanaskquestionsiftheydon’tunderstandanideathatisbeingtalkedabout.

3

•We(studentsandteacher)cancritiqueideasofothers,butpersonalattacksareoutofbounds.•Don’ttalkoveryourclassmates.•Theteacherwillgive“thinktime”beforeaskingforstudents’ideas.•Insmallgroupwork,everyonewillcontributetotheconversations.

Itishelpfultore‐visitthesenormsperiodically,askingyourstudents:“Howdidwedotodayinourdiscussion?Whatdoweneedtoworkon?”

Idea #2. “Priming” yourself for classroom conversationsNovice teachers often flounder in the middle of classroom conversations becausetheyhaven’t imaginedwhat the specific goalsof the conversationare. It is vital to“pre‐think”whereyou’dliketoend upatthefinishofaconversation.Doyouwantalist of students’ initial ideas about a scientificphenomenon?Doyouwant them tomakesenseofanactivity?Doyouwantthemtocritiqueanexplanatorymodel?Theseconsiderations“prime”youtohearcertaintypesoftalkfromstudentsandprepareyoutorespondwithouthavingtoimproviseeverywordyousay.Westartbyconsideringthree major conversation types thatweusetoorganizeourthinking—eachhasadifferentpurpose,andeachusesdifferentinstructionalmovestoaccomplishthatpurpose.Theseare:

• Eliciting  students’  initial  scientific  hypotheses  in  order  to  plan  for  further instruction.Thegoalofthisdiscourseistodrawoutstudents’understandingsofaphenomenon(e.g.abicyclerusting inthebackyard)that isrelatedtoanimportantscientificidea(inthiscasechemicalchangeorconservationofmass)and then to analyze students’ ways of talking about it in order to adaptupcominglessons.• Connecting  activities  with  scientific  ideas. The goal of this practice is tocombine hands‐on work with readings and conversation in order to buildcontent knowledge and to advance students’ understanding of a naturalphenomenon.•Pressing students for evidence‐based causal explanations.Thisdiscourseisdesignedtohappenattheendofaunit,butelementsofthisconversationcanalsohappenwhentheteacheristryingtogetstudentstotalkaboutevidence.Thegoalofthisdiscourseistoassiststudentsinco‐constructingevidence‐basedexplanatorymodelsforthenaturalphenomenonthathavebeenthefocusoftheunit.

If you are clear aboutwhat kindsof talk youwant to foster—even to thepoint ofhavingdifferentkindsofnamesfor theseconversations—itbecomeseasier foryouto anticipate student contributions and to plan how students can becomemeaningfullyinvolvedinthetalk.

4

Idea #3. The cognitive demand of questions/tasks

We share this example because all too often, we forget that sense‐making is astudent‐drivenprocess,notsomethinginherentinanactivityoralecture.Itrequiresactiveprocessingonthepartofstudentstohelpmakeconnectionsamongideasandtounderstandexplanations.Sense‐making has a lot to do with the types of questions that get asked in class.Questionsandtasks inclassroomscanbe thoughtof in termsofwhat they requirelearners to do intellectually. These can roughly be divided into those with low‐cognitive  demand  and high‐cognitive  demand (Henningson& Stein, 1997; Leach &Scott,2002).Lower cognitive demand questions/tasksThese typically focus on either memorization (recall), on vocabulary‐levelunderstanding only, or on procedural tasks that ask students to follow prescribedstepsorplugnumbers into formulae.There isnothing inherentlywrongabout lowcognitive demand questions. They can be challenging for students to answer, butthey don’t involve much intellectual work. If they becomethedefaultmodeofyourinstructionandyoufailtoaskhighercognitive demand questions, your students will only rise tothelevelofwhatyouareaskingthem.Higher cognitive demand questions/tasks 

Thesetypicallyfocusonsense‐makingbythestudents.Thesequestions/tasksdemandmoreintellectualworkandmaynothavediscreteanswers—this iswhytheyareoftenreferredto as “authentic questions or tasks.” They are much like whatprofessionalsdealwithineverydaylife.

Followingaresomecomparisonsbetweenlowandhighcognitivedemandquestions.

Whileworkingwithagroupof6thgradersduringalesson on solubility, a teacher asked students toworkwithapartnertohelpthemmakesenseofareading. One young girl, Emmy, wasdumbfounded, she approached the teacher andsaid“You mean  I have  to make  it make sense?—doesn’t it make sense by itself?”

Worthnoting:Theideaofhighandlowcognitivedemandappliestowrittenaswellasspokenquestionsandtasks.

5

Askingstudentstoworkwithinformation

LowerCognitiveDemand HigherCognitiveDemand

Recalling and reproducing ideas 

• Task/questionrequiresonlytherecallofpreviouslylearnedmaterial—sometimesaone‐wordorone‐phraseresponse.Inonephysicstextbook,aquestionisposed“Whatismeantbyfreefall?”Inanothersection,“Whattwounitsofmeasurementarenecessaryfordescribingspeed?”Inbiology,astudentmaybeaskedto“Definenaturalselection.”

• Task/questioninvolvesreproducinganexplanationpreviouslyseeninwrittenmaterialorgivenbytheteacher.Anexampleshereis:“WhereisDNAstoredinacell?”Anotherexample:“What2particlesaccountforalmostallofanelement’satomicmass?”

Processing ideas 

•Tasksorquestionsrequirestudentstouse(notregurgitate)ideasandinformationinwaysthatexpandunderstanding,suchasto:

‐makeconnectionsbetweendifferentkindsofrepresentationsofideas(e.g.asrepresentedinvisualdiagrams,graphs,drawings,analogies,manipulatives,symbols,problemsituations)‐recognizeanduseevidencetosupportexplanatoryclaims‐distinguishbetween“what”,“how”,and“why”explanations‐createandcritiqueexplanatorymodels‐applyknowledgeincontextsdifferentfromthosepreviouslydiscussedinclass

Anexamplemaybe:“Afrog’shearthas3chambers(2atriaand1ventricle),andtheirbloodflowsfromtherightatrium,totheventricle,tothelungs,totheleftatrium,andthenbacktothesameventricle.Becauseofthisstructure,whymightitbepossibleforhumanstohavemoreendurancethanafrog?”Anexamplefromchemistry:“Imagineyouliveinawater‐proof(butnotpressure‐proof)house20metersunderwater.Howmightyourcookinghabitschangeifyouhadtoboilwater?”

Anexamplefromphysics:“Whenridinginacar,rolldownthewindowandholdyourhandlikeaflatwingintheair.Slightlytiltthefrontedgeofyourhandupandpayattentiontohowtheairmovesagainstyourhand.HowcanNewton’sLawsofMotionhelpyoutoexplaintheliftingeffect?Aretheselawsenoughforacompleteexplanationordoyouneedotherideas?”

 

Questionsaboutclassroomactivities

LowerCognitiveDemand HigherCognitiveDemand

Failing to connect activities to ideas

• Taskislikearecipetofollow;itleaveslittleambiguityaboutwhatneedstobedoneandhowtodoit.Manylabactivities,forexample,areoftenmadeintoroboticexercisesinfollowingdirections.Usuallyitisgoodtobeexplicitaboutwhatyouwantstudentstodoinatask,butstudentsshouldbegiventheopportunitytoproblem‐solveontheirownratherthanrotelyfollowprocedures.

•Sometimesquestions/tasksinvolvemanipulatingnumbersandsymbols.Again,thesearenotbadquestions,unlesstheybecomethestaplesofyourinstruction.Anexampleinphysics:“Whatistheaccelerationofavehiclethatchangesitsvelocityfrom100km/hrtoadeadstopin10seconds?”

Achemistryexample:“Pleasebalancethefollowingequation:Zn+HCl‐>ZnCl2+H2”

Connecting activity with ideas

• Task/questionrequireshands‐onactivitytobeintheserviceofdevelopingortestingascienceidea.

• Taskrequiressomethinking:althoughmayuseaprocedure,itcannotbefollowedmindlessly.Studentsneedtoengagewithconceptualideas(understandwhattheymean)inordertosuccessfullycompletethetask.

•Tasksolutionnotself‐evidenttostudent,duetonatureofsolutionprocessrequired.

Anexampleinchemistry:“Youandyourfriendliketocook.Yourfriendthinksthatpurewater(H2O)willboilfasterthansaltwater.Youdisagree.Whoisright?Designanexperimenttotestyourrespectivehypothesesaboutwaterboiling,andprovideevidenceyoucouldusetosupportyourclaims.

6

 

Askingforexplanations

LowerCognitiveDemand HigherCognitiveDemand

Seeking a “what” explanation •Taskorquestionrequiresonlya“what”explanation”ofthetargetphenomenon,not“how”or“why”explanations.A“what”explanationisnot an explanation at all—itismerelyadetaileddescriptionofsomethingobservedorreadabout.Forexample,a“what”explanationinchemistrymightbe“Explainwhatthedifferencesarebetweenacidsandbases.”Or“Explainwhathappenedwhenwemixedbakingsodaandvinegar.”Hiding behind vocabulary • Insomecasesaquestionisphrasedasa“why”buttheteacherissatisfiedwhenstudentgivesavocabularytermastheanswer.Anexamplehereis:“Whydoarcticfoxhavewhitefur?”Studentanswer:“Becauseofevolution.”

Anotherexampleis:“Whydoesdyedispersefasterinwarmwaterthanincoldwater?”Studentanswer:“Becauseofkineticmolecularmotion.”

Studentsmaybeabletoreproduceorrecognizesuchbriefresponses,buttheyshouldbepressedfurther:“Whatdoyoumeanbythat?Canyouexplain?”

Seeking “why” explanations:•Task/questionrequiresawhyexplanation.Bya“why”explanation,wemeanthatthestudentcanuseevidence,information,andlogictotellacausal storyforthetargetphenomenon.Thiscausalstoryalways involves unseen, underlying events and processes that have to be connected in a logical way to explain observable events.Thiscausalstoryisoftenreferredtoasanexplanatory modelbecauseitcanbeusedtoexplainawholerangeofphenomenainthenaturalworld.Anexamplefrombiologymightbe:“Thebirdflu(avirusfoundinbirds)hasbeeninthenewsrecentlybecauseseveralpeoplehavediedfromit.However,theinfectedpeopledidnottransmitthebirdflutoanyotherpeople.Usingyourunderstandingofevolution,pleaseexplaina)Whypeoplecanbecomeillfromavirusthatinfectsbirds,andb)Isitpossibleforsomeoneinfectedwiththebirdflutotransmitthevirustoanotherperson?Ifso,why?”Fromchemistry:“Explainatthemolecularlevelwhybakingsodaandvinegarproducegaswhenmixed.”

Cognitive demand: Avoiding I‐R‐E patterns of talk 

 Onecommonpatternoftalkinclassroomregularlyuseslow cognitive demand questions. This pattern, I‐R‐E,stands for initiation‐response‐evaluation. It representsdiscoursethat is notconducivetostudentthinking,andisactuallyusedbyteacherstoconstrainstudents’talkintheclassroom.Theinitiation(bytheteacher)istypically

aquestionthathasaknown,“correct”response,andrequiresonlyrecallorasimplecalculationonthepartofthestudent.Theresponse(bythestudent)isusuallyaoneortwowordphrase.Theevaluation(bytheteacher)isacommentthatthestudentiseitherrightorwrong.Anexamplewouldbe:

Teacher:Whatarethethreedifferentkindsofrocks?Student:Sedimentary,metamorphic,andigneous?Teacher:Yes,that’sgood.

7

Someteachershavea“fillintheblank”or“readmymind”speechpattern,butitstillqualifiesasI‐R‐E.Anexamplewouldbe:

Teacher:Sowemeasuremassin________?(withrisingintonationinthevoice)Student 1:Weight?Teacher:Noooo…(againtherisingintonationofvoice)Student 2:Pounds?Teacher:Almost…cananyonehelp?Student 3:Newtons?Teacher:That’sright.

Therearemanyvariationsonthistheme.Studentsmaygivepartialanswersandtheteacher may have other students fill in what is missing from a initial studentresponse.Someteachersaremorepolitethanothersaboutincorrectanswers,somearemoreterse.Thissortofdialoguecanbecomearunningquizthatputsmostkidsonedge,allowsonlycertainkidstoparticipate,focusesonlyonthelowestlevelsofthinking, and in the process, drastically undershoots what kids are capable of.Teachers using I‐R‐E typically cherry‐pick the right answers from student and thenassumeeveryonehas that sharedunderstanding. The teacher thenmoves forwardunderthisfaultyassumption.The “guess‐what’s‐in‐my‐head” dialogue is so common that it’s been called thedefaultpatternoftalkinschools.Itsdubiousdistinctionisthatitisthemostschool‐likeformoftalk.This type of discourse is also one of themost difficult for students with differentcultural and linguistic backgrounds. They do not know how this artificial speechpattern is part of the “game of school” and find it a bizarre exchange betweenteacherandlearner.WewriteaboutI‐R‐Eherebecause,aswementioned,itissocommoninschools.Wehaveseenentire50minuteclassperiods inwhichstudentshaveenduredoneI‐R‐Esequenceafteranother.Ifyourecognizethisformoftalkinyourownclassroom,youshouldmakeeveryefforttofigureoutwhyyouareusingitandhowyoucanshifttomore productive forms of discourse. One place to start is to raise the cognitivedemand of the questions you’re asking students. Another is to consider how youcouldbuilduponorcreateanotherquestion fromstudents’ responses, rather thansimplyjudgethemascorrectorincorrect(seethesectioncomingup:Using discourse moves).Idea #4. Meta‐cognitive questions in the classroom 

Meta‐cognitivequestionsaskkidstothinkabouttheirthinking.Thesekindsofquestionsaimtogetkidstoself‐monitortheirthinkingtojudgeforthemselveswhethertheyareunderstandinganidea.Meta‐cognitivequestionsalsocanhelpa

8

studentself‐regulatetheirprogresstowardsimportantprojectgoals.Forexample,ateachermayaskstudentsworkingtogetheronaprojecttowriteortalkabout:

WhatprogressamI(arewe)makingonthisproblem?HowwillIknowifIam(weare)successful?WhatadditionalinformationorexperiencesdoI(we)needtobesuccessful?WhatgapsdoI(we)haveinmy(our)thinking?

Usingmeta‐cognitivequestionshelpmake students abitmore independentof theteacher—rather than needing the teacher to tell them about the quality of theirthinking,kidslearntofairlyjudgeitforthemselves.Theyarenotusedtothinkingthisway!Italsohelpsthemmanagetheirtimeandeffortacrossclassactivitiesthattakeawhiletocomplete.Idea #5. Using wait timeInwhole class discussion, kids need time to think. Not all kidscan spontaneously interpret what a teacher’s question meansand respond to it within a couple of seconds. Rapid firequestioning privileges those few kids who have masteredEnglish, who are familiar with the “game of school”, who cananticipate thetypesofquestions the teacherwillask,andwhocan recall facts easily. The majority of kids, however, aresilencedbythisinequitabletypeofdiscourse.Onewaytomakeconversationsmoreequitableistopayattentiontowait time.Thisis the amount of time between when a teacher poses a question, and when theteachereithercallsonastudent,rephrasesthequestion,givesahint,oranswersthequestion himself or herself—essentially the amount of time the teacher gives thestudents to think.Researchhas shownthat thewait time teachersgive students isremarkably short. Believe it or not, the average wait time for most teachers isapproximately one  second! This is because teachers are almost immediatelyuncomfortablewithsilenceinaclassroomconversation,andseektofillthevoidwithastudent’svoice,ortheirown.This same research has shown that when wait time is kept short, only a smallminority of kids respond, and their responses are very brief. Some teachers havepurposelylengthenedtheirwaittimeto5,10,or20seconds,togiveallkidstimetothink. In these classrooms, a far  greater  percentage of kids responded to theteachers’questions,andtheresponseswerelonger and more thoughtful.Waittimeworksbestwhenthecognitivedemandofquestionsisatamediumorhighlevel.There are other strategies to give all kids time to think before joining theconversation.Think‐pair‐shareisamovewhereateacherposesaquestion,thenasksstudentstoconsidersilentlyhowtorespondforabout30seconds,thenjoinwitha

9

peer to compare their responses, then return to the whole class conversation toshare their ideas. An even simpler way to give kids time to respond is to pose aquestion,requestthateveryonekeeptheirhandsdownfor30seconds,thenaskforresponsesafterthat.In summary,extendingwait time isoneof thesimplestbutmosteffectiveways toencourageequitableandhigherqualityparticipationinclassroomdiscourse.Idea #6. Using different discourse moves 

By discourse  moves, we mean the specific conversationalstrategies that teachers use to foster the development ofideas in the classroom. Thesemoves can be used either inwhole class conversations or when the teacher talks withstudents in small groups. Discourse moves serve severalpurposes. They elicit student thinking, model how onethinks,encourageallstudentstoparticipate,emphasizekeyideas, and ultimately help students appropriate scientific

discourse themselves. Here are four categories ofmoves that skilled teachers use.Therearenotpresentedinanyparticularorder.Probing Probing questions or prompts get students tomake publicmore of their thinking.Thisisperhapsthemostimportantfunctionofclassroomconversation.Usuallythesequestionsorprompts areprecededby someactivity/situation/reading/videoaboutwhichstudentscanshareinitialideas.Commonprobesinclude:

Whatexperienceshaveyouhadwith…?Canyoutellmemoreaboutthat…Canyouexplainitinadifferentway?

One specific kind of probing is the reflective toss (van Zee&Minstrell, 1997). Thereflectivetossreferstotheteachercatchingthemeaningofastudents’remarkandthrowing responsibility for thinking back to the student. A sample of this kind ofthree‐partexchangemightgothisway:

T: How might we find the best representative of three differentmeasurementsofthesamething?S:Averagethem.T: Okay, we might average them. Now what do you mean by “average”here?

Usuallya reflective tossaddresses themost recent remarkbya student,but it canaddressanearlierremark,ortheteachercanaddresstheremarkfortheentireclasstorespondto.

10

Re‐voicing Re‐voicingmeansthattheteacherrepeatsorparaphraseswhatastudenthassaid,inordertoachieveaninstructionalgoal.Herearesomeexamplesofre‐voicingmovesandthereasonsyouwouldusethem:

Marking/Amplifying:Thisiswhenateacherselectsaspecificportionofstudents’commentsthattheteacherbelieveswillcontributetotheclasses’developmentofascientificidea.Theteacherwillrepeatthat“piece”ofanideatotheclass.Itmay sometimes be only a word the student has said, but it can be a fullhypothesis, observation, or a question. This is used when there may be anumberof ideas“flyingaroundtheroom”thatcouldconfusestudents,orthatdivert from the main idea that the class is working on. Here is a commonsentence stem a teachermight use after a student has given a legitimate butlongandoccasionallydisconnectedinterpretationofaclassroomdemonstration:“So[nameofstudent],whatIhearyousayingisthat[heathassomethingtodowiththemotionofthemoleculesofwaterinourfooddyedemonstration]?”

Repairing/clarifying  an  idea: This is a teacher’s re‐statement of a studentcontributioninwhichtheteacherjudiciouslycorrectsoneaspectofanotherwisevalid statement. This is done to prevent confusion by students when suchstatementsmight otherwise be takenwithout comment by the teachers. Thisdoesnotmeantheteacher“corrects”statementsonaroutinebasisorevaluatesthemovertly,butitdoesmeanthatclarificationsaremadeinasensitiveway.Asentence starter here might be “I understand your explanation, but did youmeantosay____?”Connecting students’ everyday language with academic language:Thisisalsotobe used judiciously. Students need to hear how some forms of everydaylanguageareconnectedwithscientific language.Scientific language isvaluablebecauseitallowsstudentstothinkinconceptualtermsaboutideas,andbecauseitallowsacommonreferencetotalkaboutscientificpractices(i.e.whatcountsasan“explanation”,orwhatcountsasa“model”).Anexampleheremightbe:“Sowhenyoutalkaboutacceleration,youusuallymeantospeeduplikeyoudowhenyoupressthegasonacar.Scientiststhoughusethatterminadifferentway—tomeananychange in speedordirection.”Anotherkindofconnectionbetweeneverydayandscientific(oracademic)languageiswhenastudentusesaneverydaytermandtheteacherre‐voicesbysubstitutingascientificterm.Forexample, substituting “convection” for a student’s description of “warm airrising while cold air sinks.” The teacher here, however, must take care tomaintainthestudents’ownershipofthatidea.

 Pressing Teachersaregenerallysensitivenottoembarrasstheirstudentsinpublicdiscussions.Thisisunderstandable.Students,however,cancometoexpectthatiftheygiveone‐

11

wordanswers toquestions, sometimesnoteven finishing their sentences, that theteacherwill not ask them to finish their thought or to thinkmore deeply about aquestion.Becauseofthis,therearetimeswhenateachermust“press”studentsinbothwholegroup and small group conversations. This is a reasonable way of holding kidsaccountableforthinking.Bypresswemeanthattheteacherdoesnotallowstudentstooffershortcutresponses,unsupportedclaims,orrespondwith“youknow.”Whena student offers an initial idea, the teacher for example, can ask in return: “Whatevidencedoyouhave for thatclaim?”, “Isn’t thatacontradiction towhatyousaidearlier?”or“Whydoyouthinkthat?” Youcantellwhenyouarepressingstudentsbecausetheywilloftenvisiblysquirmwhenyouwon’tgiveupontheirthinking.Pressingstudents isverydifferentfromjusthearingeveryone’svoiceduringaclassperiod(i.e.itisdifferentfromprobing).Itdoeslittlegoodtogetdozensofone‐wordresponsesfromavarietyofstudents.Gettingonewordresponsesfromkidsdoesnotmeantheyareinvolved.As youcan imagine, knowingwhoandwhen topress requires that youknowyourkids and that you’ve established a safe classroom environment for theseconversations. Youwill needextrapatiencewith this conversational expectation intheclassroom,since“press”issomethingthatveryfewkidsexperienceintheirotherclassesorintheirentirehistoryasstudents. Putting an idea “on hold” Intheenthusiasmofwholeclassdiscussions,studentsoftenmake statements that can be off‐topic, or that are betteraddressed later on. In these cases, teachers need politeways of acknowledging the students’ contributions, whilemarkingitassomethingthatisnotgoingtobetalkedaboutat this point. A teacher might say: “That’s an interestingidea,anditissomethingthatwewilltalkabouttomorrow,butfornow…”or“I likeyourthinking,but let’sholdontothatthought…”Some teachers have a sectionof theirwall spacedevoted to genuinequestionsorcommentsthatstudentshavewhichmaynotbethefocusonthecurrentlesson.Thishas been called the “Parking Lot”, and it signals to students that their ideas havevalue,butmaynotfitthecurrentdiscussion. 

Cautionarynote:Manyofourbeginningteachershaveusedthisconversationalmovetoavoidreallyinterestingandrelevantquestionsbystudents.Donotusethisstrategytosuppresskidswhowanttoexploretheboundariesofthelesson!

12

Idea #7. Scaffolding academic language Academiclanguagecanbedefinedasthelanguageusedintheclassroomorschool‐talk.Itisneededbystudentstodotheworkinschools.Academiclanguageincludessuch things as specialized vocabulary, grammarandpunctuation, conventional textstructures within a field (e.g., essays, lab reports), particular forms of talk (e.g.,storytelling, talk about evidence), and other language‐related activities typical ofclassrooms, (e.g., expressing disagreement, discussing an issue, asking forclarification). For some students, academic language seems to come naturallybecause it is similar to the kinds of language used by peers and familymembers.However, for many students, academic languagerepresents an entirely new language. Proficiency inconversational English does not necessarily mean thatstudentswilleasilytakeupthis“new”academiclanguage.It is a teacher’s responsibility to help students developfluency in the academic language of schools and thespecificlanguageofsciencesothatstudentscanbecome“multilingual”participantsinscienceatschool. When thinking about how to scaffold academic language for students wehave to consider all of the ways that students communicate to others throughproducing language, (e.g., speaking, writing, drawing) and all of the ways thatstudentsreceivecommunicationfromothersbyconsuminglanguage,(e.g.,listening,reading,viewing).Tolearnalanguageanddevelopfluency,itiscrucialthatstudentsengageinregularpracticeinallofthewaysofproducingandconsuminglanguage.Itisalsoimportanttoprovidescaffolding–temporaryhelpintheformoftoolsortasksthat give students a boost when trying something that is challenging or new. Themain purpose of scaffolding academic language is to make all of the elements oflanguagemorevisibleforstudentsincludingthemodesofcommunication,formsoftalk, language structures and functions, and specialized vocabulary. The followingtableprovidesadditionaldescriptionsofeachoftheseaswellassuggestionsaboutpossiblescaffolding.Youwillneedtoidentifyfurtherresourcesbeyondthisprimertohelpyoudesignthisscaffolding.

5ElementsofLanguage

AskYourself… PossibleScaffolding

 

Modes of Communication 

Languageconsistsofinputs(listening,reading,viewing)andoutputs(speaking,writing,drawing).

  

Howislanguagebeingusedbystudentsintheclassroom?

Areyouprovidingopportunitiesforstudentstoengageinspeaking,writing,reading,drawing,listening,andviewingintheacademiclanguageofscience?

Providesentencestartersandconnectingwordstohelpstudentstalkandwriteinacademiclanguage.

Usepre‐readingandduring‐readingstrategiestohelpstudentsreadacademictexts.

Worthnoting:LearninganacademiclanguageisnotjustforEnglishLanguageLearners.Allyourstudentswillneedhelplearningtospeak“school”and“science.”

13

Forms of Talk

Languageincludesformsoftalkspecializedforparticularfieldswhichcanseemunusualtopeopleoutsideafield.Theseformsarepartofthe“culture”ofafieldsuchasarguingwithevidenceinscienceorlaw.

Howarespecializedformsofsciencetalknecessaryintheclassroom?

Arestudentsmakingexplanatoryclaims,providingevidence,designingexperiments,anddiscussingmeritsofmodels?

Provideexplicitexpectationsaboutthecontentof“good”explanationsandargumentsandgeneraterubricswithstudentsinclass.

Bearolemodelby‘thinking’outloudsostudentscanhearacademicformsoftalk.

Language Structures 

Languageisorganizedintostructuressuchasthetypeof“why…because”questionandanswerstructurecommoninschools.Structuresareoften“invisible”andcanseemmysteriousuntiltheyareexplicitlyidentified.

Whatkindsofstructuresoflanguageareusedintheclassroom?

Arestudentsexpectedtomakecomparisons,answer“why”questions,usesymbolsandequations,oruseconventionalrepresentationslikegraphs?

Explicitlyidentifythestructureoflanguageinthetaskandhelpstudentsseewhattheyaresupposedtodotocommunicate.

Introducetransitionsbetweenstructuresstep‐by‐step(i.e.,movefromanidea,towords,tosymbols,tographs,andback).

Language Functions 

Languageservesspecializedfunctionsuniquetoafieldoracontextlikeschool.Likelanguagestructures,thefunctionsoflanguagecanseem“invisible”untilsomeoneexplicitlyidentifiesthefunctionsofcertainlanguagepractices. 

Whatkindsoffunctionsareservedbylanguageintheclassroom?Whyarecertainlanguagepracticesbeingusedincertaincircumstances?

Arestudentsproposingtentativeexplanations,formingspecifichypotheses,describing,explaining,inferring,relatingeventsacrosstime,interpretingdata,orgeneralizingtonewcontexts?

Worktogetherwithstudentsinclasstoclarifywhycertainlanguagepracticesareusedindifferentcircumstances(e.g.,observationsareusedindescriptions,butexplanationsrequiretheoryandinferences).

Constructexampleswithstudentsandworktogethertoevaluate“good”,“better,”and“great”exemplars.

Specialized Vocabulary 

Everyfieldandeverycontexthasspecializedvocabularythatisusedonlywithinthatfieldorcontext. 

Whatspecializedacademicand/orscientificvocabularyisrequiredforparticipationinthisactivity?Whatacademic/scientificvocabularyisrequiredtomakesenseofaclasstext?

Howcanstudentsslowlygainvocabularytoenrichtheirparticipationinacademicandscientificcommunities?

Workwitheverydaywordsandacademicorscientifictermsinparalleluntilstudentscanmakevocabularytransitions(e.g.,physicsstudentsmightsay“drag”insteadofsaying“friction”whichisfineinitially).

Providetransitionwordsforstudentstouseastoolswhenwriting:“Forexample,”and“Inaddition,”helpstudentselaboratetheirparagraphsinanacademictone.

14

 Idea #8. Encouraging peer‐to‐peer talk for students  Productive conversations are not only about teachers’ questions, they arealso about students’  questions and how the teacher helps  students  talk  to  one anotherusingthe languageandrhetoricofscience.One long‐termgoalyoushouldhaveregardingclassroomconversationsisthateventuallyyourstudentsshouldtakeover someof the responsibilities of guiding the discourse. Thedifferencebetweenthefirstandlastweekofschoolshouldbethat,bytheendoftheyear,yourstudentshavebegun toask thequestions thatyou askedearlierand that your studentsaredoing the probing, the comparing of ideas and the critiquing of peers’ ideas. Yourstudents shouldbedeveloping the civility needed toelaborateonand critique theideasofothers inapublic setting—withoutyouactingasan intermediarybetweeneveryturnoftalk. To start this process, you need to tell studentsexplicitlythatyouwantthemtoaddresseachother’sideas.Also, you should regularly provide responses to studentsthatyouwantthemtosaytoeachother.Hereareexamplesentence‐starters,someofthesecanevenbeputoncardsthatstudentsuseinsmallgroups,youmayrecalltheseareformsofprobing,reflectivetoss,andpressing:“Cananyoneaddto______’sidea?”

“Whatisthedifferencebetweenwhatyou’vesaidandwhat_____hassaid?”

“Doesyourideamakeyouquestionsomethingthat_____hassaid?”

“So, _______, it sounds like your claim is _______ and one piece of evidence is______.But_____hasthisotherpieceofevidencewhichconflictswithyours,whatdoyouthink?”

Developing peer‐to‐peer talk is difficult work that will take weeks or months ofencouragement.Mostteachersexperiencedinclassroomdiscourselookforstudentstobeginengagingwithoneanotherafteraboutthreemonths.Yes,that’salongtime,sopersistenceisthekey.

Worthnoting:Youmayliterallyhavetoturnawayfromyourstudentstoforcethemtoaddressoneanotherwiththeircomments.

15

In summary: Developing a theory of action  Althoughwe’vetalkedaboutanumberofideas,itismosthelpfulforteachersto relate these ideas together ina “theoryof action”—awayoforganizingwhat isgoing to be done in a classroom andwhy.We provide one example here, butweencourageyoutoreorganizethisframework,addtoit,inordertomakeityourown.

TheoryofActionAnotherway to thinkabout the ideas in thispaper isacheck‐listyoumightuseasyouplanoryourownclassroomdiscourse.

Theclassroomenvironmentwillbesafeforstudentstoexpresstheirideas.Goalsofclassroomconversations/discussionsregardingwillbeanticipatedbytheteacherandmadecleartostudents.Questionsandtaskswillbepredominantlyofhighcognitivedemandformakingsenseofscienceideasandphenomena.

Importantthatallkidshaveregularopportunitiestoparticipatein

meaningfulscienceconversations

Productiveformsoftalkmediatesense‐making,

andsense‐makingisnecessaryforstudentlearning

Learningscienceislikelearninganewlanguage,notjustforELLstudents,butfor

everyone

Provideasafeclassroomenvironment—clearnorms,expectations,routinesoftalk

Primeyourselffordifferentkindsofconversations

Includehigh‐cognitivedemand

questions

Use“discoursemoves”likeprobing,pressing,re‐voicing

Encouragemeta‐cognitivereflection

Usewaittimeandthink‐pair‐share

Encouragepeer‐to‐peertalk

Scaffoldacademiclanguage

Workingassum

ptions

Waystoachievegoals

16

Waittimewillbeusedafteraskingquestionsrequiringhighcognitivedemandandaftereachstudentresponds.Avarietyofdiscoursemoveswillbeusedtomanagetheinitiationanddevelopmentofideaswhileatthesametimehonoringthethinkingofallmembersoftheclass.Meta‐cognitivequestionswillbepartofalllessonssothatstudentslearntoself‐critiquetheirthinkingandmonitortheirprogresstowardlongertermgoals.Studentslanguageandformsofcommunicationwillbescaffoldedfromwhattheybringtoclasstowardmoreacademicwaysofspeaking.Theteacherandstudentswillmodelinteractionsthatfostersbeingcriticalofunsupportedideaswhileencouragingsharingofideasandrespectforthosewhoaresharing.

Classroom talk matters for student learning. This general guide to classroomdiscourse is only a starter document, but you can use the ideas here for severalpurposes:

•tointerpretclassroomdiscourseyouhearfromotherteachers,•tohelpyouevaluateyourownclassroomdiscoursepatterns,•toserveasaframeworktodesignyourownsequencesofconversationalmoves,•tohelpyouestablishlong‐termdiscoursegoalsforyourclassroomcommunity.

Using thisdocumentasan interpretive frameworkcanhopefully convinceyou that“teacher storytelling” or “quizzing” is not howmost students learn, and that goodteaching is the product of having specific goals and enacting specific patterns ofverbalinteractionwithlearners. Some beginning teachers have a natural aptitude for fostering meaningfulconversations, but no novice has the all the skills to artfully design conversations.Everyteacher,however,whotakesaprincipledapproachtoclassroomdiscoursecaneventually develop an interactional expertise with student conversations that willleadtolearning.Wehaveseenthishappeninfirstyearteachers—anditisinspiringtoseehowkids fromallcultural, linguisticandsocioeconomicbackgrounds intheirclassroomsaregiventhechancetoparticipateandtoachieveathighlevels.

ReferencesHenningson,M.&Stein,M.K.(1997).Mathematicaltasksandstudentcognition:Classroombasedfactorsthatsupportandinhibitmathematicalthinkingandreasoning. Journal for Research in Mathematics Education, 28(5),524‐549.Leach,J,&Scott,P(2002).Designingandevaluatingscienceteachingsequences:Anapproachdrawingupontheconceptoflearningdemandandasocialconstructivistperspectiveonlearning. Studies in Science Education,38,115–142.VanZee,E.&Minstrell,J.(1997).Usingquestioningtoguidestudentthinking.TheJournal of the Learning Sciences, 6(2),227‐269.