Vol. 28, No. 11, 15 September 2006, pp. 1315–1346
RESEARCH REPORT
Classroom Interaction in Science: Teacher questioning and feedback to students’ responses
Christine Chin*
Nanyang Technological University, Singapore
0The purpose of this study was to (a) develop an analytical framework that represents classroomtalk and questioning in science, (b) find out how teachers use questioning to engage their studentsin thinking about conceptual content that enables the construction of knowledge, and (c) identifythe various forms of feedback provided by teachers in the follow-up move of the initiation–response–follow-up format of teaching exchange. Several lessons from Year 7 classes wereobserved across a variety of lesson structures such as expository teaching, whole-class discussions,laboratory demonstration, and hands-on practical work. The lessons were audiotaped and video-taped. Transcripts of the lessons were made and analysed, with particular attention paid to interac-tions that involved questions. Using the “Questioning-based Discourse” analytical frameworkdeveloped in this study, four different types of feedback were identified. Interactional issues relatedto ways of speaking and questioning that encourage student responses and thinking are addressed.This information provides a description of what constitutes effective discourse in science teachingand learning, and will also be useful for both teachers and teacher-educators in identifying anappropriate repertoire of skills for subsequent teacher education and professional development.
Introduction
When students learn science in a classroom setting, a primary source of informationinput comes from teacher talk and teacher–student interactions, as the processes andtransactions involved in the construction of meanings are mediated throughlanguage. Given the important role of verbal discourse in meaning-making bystudents and its significance for teaching and learning, classroom discourse andinteraction has been the subject of interest of several researchers (e.g., Cazden,2001; Edwards & Mercer, 1987; Edwards & Westgate, 1994).
*Natural Sciences and Science Education, National Institute of Education, NanyangTechnological University, Singapore. Email: hlcchin@nie.edu.sgISSN 0950-0693 (print)/ISSN 1464-5289 (online)/06/111315–32© 2006 Taylor & Francis
DOI: 10.1080/09500690600621100
1316C. Chin
In particular, the three-part exchange structure known as “triadic dialogue”(Lemke, 1990) or recitation has been found to be pervasive in classrooms. Thisdiscourse format typically consists of three moves—initiation (often via a teacherquestion), student response, and teacher evaluation—and has been commonlyreferred to as “IRE” (Mehan, 1979). The teacher asks a closed question that is basi-cally information-seeking, that requires a predetermined short answer, and that isusually pitched at the recall or lower-order cognitive level. He/she then praisescorrect answers and corrects those that are wrong. Sometimes, it is also known as“IRF”—initiation, response, and follow-up or feedback (Sinclair & Coulthard,1975), as the third move may not necessarily be an explicit evaluation. Wells (1986),for example, has discussed ways in which teachers may provide feedback by encour-aging students to externalize ideas, generate hypotheses, and test them.
The triadic dialogue, which is typical of traditional teaching, is often perceived tohave restrictive effects on students’ thinking as students’ responses remain brief andteacher-framed, thus minimizing their role in the co-construction of meaning.Although such conventional teacher-questioning practices based on this discourseformat have been criticized (e.g., Lemke, 1990), some authors have accorded it acertain functionality that is consistent with educational goals. For example,Newman, Griffin, and Cole (1989) argued that the three-part exchange has “a built-in repair structure in the teacher’s last turn so that incorrect information can bereplaced with the right answers” (p. 127). Such a view is appropriate if we view theresponsibility of teachers as ensuring that students appropriate the knowledge that isnormative within a particular culture.
Similarly, Wells (1993) has argued that, when used effectively, “it is in this thirdstep in the co-construction of meaning that the next cycle of the learning-and-teach-ing spiral has its point of departure” (p. 35). Thus, the triadic dialogue could havemerit if teachers can scaffold students’ extension of knowledge through furthersupportive dialogue (Bruner, 1986; Vygotsky, 1978). An instance of this would bewhen teachers pose a question that stimulates further productive thought, based ontheir evaluation of students’ previous responses. In such a case, teachers would beguiding the development of students’ ideas by successively building on their contri-butions in a reciprocal manner. This suggests that the triadic dialogue does not existas a homogeneous format. To achieve a more adequate understanding of classroomdiscourse, we could study the variations that stem from the IRF format.
This study investigated questioning-based discourse practices in science class-rooms through the interaction between teacher and students across a number ofactivities. It aimed to identify the different ways in which teachers follow up onstudents’ responses to their questions.
Classroom Interaction and Discourse in Science
Knowledge is constructed in the social context of the classroom through languageand other semiotic means. Central to Vygotsky’s (1978) sociocultural theory oflearning is the idea that conceptual knowledge first appears between people on an
Classroom Interaction in Science1317
interpsychological plane, and then inside the learner on an intrapsychological plane.The notion of the teacher assisting student performance through the “zone of proxi-mal development” also suggests that teachers can guide the discourse on theinterpsychological plane to support student learning. This recognizes the importanceof teacher–student discourse in the classroom, which may be considered as a form ofscaffolding (Bruner, 1986; Wood, Bruner, & Ross, 1976), although the latter wasoriginally conceived in the context of tutoring a single individual in problem solving.Edwards and Mercer (1987) identified the following features of classroomdiscourse at an increasing level of teacher control: elicitation of students’ contribu-tions, significance markers, joint-knowledge markers, cued elicitation of students’responses (which is similar to the IRF structure), paraphrastic interpretations ofstudents’ contributions, reconstructive recaps, and direct lecturing. Lemke (1990)identified several “thematic development” strategies used by teachers in scienceclassrooms. These include “dialogue and monologue” strategies. Dialogue strategiesinclude the Teacher Question Series (similar to the triadic IRF), selection and modi-fication of student answers, retroactive recontextualization of student answers, andjoint construction. Monologue strategies include logical exposition, narrative, selec-tive summary, and foregrounding and backgrounding.
Scott (1998) characterized authoritative and dialogic discourse based on thegeneral features of the discourse, the nature of teacher utterances, and the nature ofstudent utterances. While authoritative discourse focuses on the “information trans-mitting” voice and has a fixed intent and outcome, dialogic discourse involvesseveral voices and has a generative intent. In authoritative discourse, the teacherconveys information and his/her utterances often involve instructional questions,factual statements, and reviews. However, dialogic discourse encourages challengeand debate, and is often based on open or genuine questions. For authoritativediscourse, student utterances are often given in response to teacher questions, andconsist of single, detached words interspersed in teacher delivery. In contrast, theyare often spontaneous, expressed in whole phrases or sentences, and are tentativesuggestions in dialogic discourse.
While dialogic discourse allows students to argue and justify their ideas, theauthoritative discourse also has its place in the classroom, particularly when thealready constructed shared knowledge needs to be emphasized. Indeed, an alterna-tion between these two types of discourse is important for developing conceptualthinking on the intrapsychological plane (Mortimer, 1998). Scott (1998) referred tothe alternation between these two types of discourse as “rhythm of the discourse”,and suggested that learning will be enhanced through a balance between presentinginformation and allowing exploration of ideas.
In their “flow of discourse” analytical framework, Mortimer and Scott (2000,2003) addressed aspects of classroom discourse including (a) teaching purposes, (b)the content of the discourse regarding whether a student utterance matches theintended learning goal, (c) the form of the utterance in terms of whether it is adescription, explanation, or generalization, (d) the communicative approach (inter-active vs non-interactive, authoritative vs dialogic), (e) the patterns in the flow of the
1318C. Chin
discourse, and (f) teacher interventions. Central to their framework is the communi-cative approach, which focuses on whether or not the teacher interacts with students,and whether he/she takes account of students’ ideas. The dialogic approach recog-nizes more than one point of view, while the authoritative approach focuses on justone (the school science) point of view. The interactive approach allows for the partic-ipation of other people, but the non-interaction approach excludes them.
Thus, for the interactive/authoritative communicative approach, the teacherinvites responses from students but discounts their ideas, as he/she focuses solely onthe scientific idea. He/she typically leads students through a sequence of questionsand answers with the aim of reaching one specific point of view. In contrast, for theinteractive/dialogic approach, the teacher explores students’ views and takes accountof them, even though they may be quite different from the scientific one. The non-interactive/authoritative approach is best represented by the formal lecture where theteacher presents normative ideas in a monologue. As for the non-interactive/dialogicapproach, the teacher does not invite any turn-taking interaction with students,but makes statements that addresses other points of view in addition to the formalscientific one.
As for patterns of discourse, Mortimer and Scott (2003) expanded on the IRE orIRF structure by identifying the IRFRF chain where the elaborative feedback fromthe teacher is followed by a further response from a student. This form is typical ofdiscourse that supports a dialogic interaction. As part of the feedback, the teachercould repeat a student’s comment to encourage the student to continue, elaborateon the comment, or ask for elaboration. By establishing this pattern of discourse, theteacher is able to explore students’ ideas.
van Zee and Minstrell (1997a) examined ways of speaking that were characteristicof “reflective discourse”. In such interactions, students articulated their own ideasand posed questions; and teachers and students engaged in an extended series ofquestioning exchanges. Teachers helped students develop understandings through aprocess of negotiation rather than transmission or confrontation of misconceptions.Teaching strategies included soliciting students’ conceptions, restating studentutterances in a neutral manner, using reflective questioning, and invoking silence tofoster student thinking.Teacher Questioning
Teacher questioning is a prominent feature of classroom talk (Wellington &Osborne, 2001). Early studies on teacher questioning focused on the IRE pattern ofdiscourse (Mehan, 1979; Lemke, 1990), the lack of student active engagement whenteachers asked too many questions based on the IRE format (Dillon, 1985), and theimportance of wait time in increasing students’ thoughtfulness (Rowe, 1986; Tobin,1987). More recent studies, however, have focused on the characteristics of teachertalk that encourage students’ construction of knowledge.
Unlike teacher questioning in traditional lessons where the purpose is to evalu-ate what students know, the nature of questioning in constructivist-based or
Classroom Interaction in Science1319
inquiry-oriented lessons is different. In such lessons, the teacher’s intent is to elicitwhat students think, to encourage them to elaborate on their previous answers andideas, and to help students construct conceptual knowledge. Thus, questioning isused to diagnose and extend students’ ideas and to scaffold students’ thinking.Such questions are open, requiring one-sentence or two-sentence answers, and theteacher engages students in higher-order thinking (Baird & Northfield, 1992).
Flexibility in questioning is needed, the teacher adjusts questioning to accommo-date students’ contributions and responds to students’ thinking in a neutral ratherthan evaluative manner. For example, the feedback step of the IRF sequence couldbe in the form of a “reflective toss” (van Zee & Minstrell, 1997a), where the teacherthrows the responsibility for thinking back to a student by asking a question inresponse to a prior utterance, thereby shifting toward more reflective discourse. Areflective toss sequence typically consists of a three-part structure: a student state-ment, a teacher question, and additional student statements.
van Zee and Minstrell (1997b) found that the teacher used reflective tosses toserve a series of subgoals. These included using questions to help students (a) maketheir meanings clear (e.g., clarifying the meaning of what had just been said, bring-ing student knowledge into public view, prompting articulation of the focal issue bya student, and emphasizing a procedure), (b) consider a variety of views, and (c)monitor the discussion and their own thinking. The authors further proposed thatthis form of questioning may help teachers shift toward more reflective discoursethat help students to clarify their meanings, consider various points of view, andmonitor their own thinking.
In another study on teacher questioning during conversations about science, vanZee, Iwasyk, Kurose, Simpson, and Wild (2001) found that teachers elicited studentthinking by asking questions that developed conceptual understanding, and practis-ing quietness through long wait times, attentive silence, and reticence. The teachers’questions included those that elicited students’ experiences, diagnosed, and refinedstudent ideas, as well those that helped students to clarify, explore, and monitortheir various points of view and thinking.
Roth (1996) described a case study where the teacher’s questioning was designedto “draw out” students’ knowledge, scaffold students’ discursive activity to lead toindependent accounts and student-centred discussions. The students’ answers werenot evaluated against the external standard of canonical knowledge. Although theteacher’s discourse contributions did not have an evaluative function, her authorityas a teacher was undisputed. Instead, teacher authority was asserted and maintainedby means other than the IRE sequence often linked to control (Lemke, 1990). Bymeans of contingent queries, the teacher was able to ultimately lead the students tothe canonical knowledge that was aligned to her lesson objectives.Purpose of Study
Carlsen (1991) proposed a sociolinguistic framework for research into teacherquestioning that would illuminate contextual issues that could not be addressed by
1320C. Chin
studies based on a process–product paradigm. This framework consists of threefeatures: the context of questions, the content of questions, and the responses andreactions to questions.
In view of Wells’ (1993) suggestion that the third step of the IRF questioningsequence might offer potential for productive discourse and of Carlsen’s (1991)proposal for the use of a sociolinguistic framework to research into teacher question-ing, this study was conducted to investigate the communicative and cognitive func-tions of teacher questions and the variety of feedback moves employed by teachers.Given the important role of discourse in meaning-making by students, there is also aneed to characterize the positive kinds of “talk-scaffolding” in some way (Westgate& Hughes, 1997). The latter authors have suggested some potential areas for furtherresearch that involve discourse–cognition relationships. This includes developingsome form of cognitive coding for identifying and classifying cognitive functionscarried out in talk, and which might incorporate elements and hierarchies withinsuch categories. In connection with this aim, one area that might be fruitful toexplore is the variety of feedback moves employed by teachers.
Accordingly, the purpose of this study was to (a) develop an analytical frameworkthat represents classroom talk and questioning in science, (b) find out how teachersuse questioning to engage their students in thinking about conceptual content thatenables the construction of knowledge, and (c) identify the various forms of feed-back provided by teachers in the follow-up move of the IRF (initiation-response-follow-up) format of teaching exchange.Methods
This study was carried out in Singapore. It was part of a larger study thatinvolved six teachers from four schools teaching Year 7 (12–13 year olds) science.However, the present study on which this paper is based involved only two ofthese teachers, who came from different schools. The lessons of these two teach-ers were selected for more detailed analysis because, when compared with theothers, there was a relatively larger amount of rich, interactive questioning in theirclassrooms. This selection criterion was important as the focus of the study wason questioning-based practices and the feedback moves employed. Thus, purpose-ful sampling was used. The average class size was 40 students per class. Thestudents were generally motivated, on-task, and ranged from average to above-average ability.
Because of large class sizes, time constraints to cover a prescribed national sciencecurriculum, and accountability pressures on teachers for students to succeed onexaminations, teaching was implemented predominantly via direct instruction orguided discussions in whole-class contexts. However, small group discussions andhands-on practical work in the science laboratory were also carried out on a regularbasis. Class activities included expository lectures, whole-class guided discussions,teacher demonstrations, small-group hands-on tasks, paired discussions, and labora-tory experiments carried out in pairs or individually.
Classroom Interaction in Science1321
Fourteen lessons (seven from each of the two teachers), which comprised a totalof about 14 h, were observed; these were either audiotaped or videotaped, or both.The lessons covered a range of topics included in the science syllabus. Theseincluded mass, volume, and density; elements, mixtures, and compounds; photosyn-thesis; and respiration. Because of manpower constraints and the limited availabilityof audio-recorders for use, only classroom discourse in whole-class settings and insome cases small groups were taped. The latter occurred whenever the teacher circu-lated among groups to talk to individual students. The audio-recorder was strappedto the teacher and so recorded whatever she said during the lessons. The videocamera was set up at the back of the classroom and was directed at the teacher andstudents. Besides visually recording the transactions occurring in the classroom, italso helped to capture the voices of students who were seated at the back of theroom. Data sources included audiofiles and videotapes of science lessons, copies oflesson handouts given to students, and notes of meetings with the teachers. Theaudiofiles of recorded classroom talk were transcribed verbatim. Video-clips of thelessons were observed and interpretive notes were made.
Verbal data from the transcripts served as primary sources of data. They wereanalysed interpretively, with a focus on teacher–student interactions and question-ing. Video-clips of the lessons, lesson handouts, and students’ written work providedadditional information about the classroom contexts. The discourse was analysedbased on the scientific content of the talk, type of utterance, type of thinking associ-ated with students’ responses, and interaction patterns.
The unit of analysis was the IRF exchange. Each pair of the teacher’s initiatingquestion and the corresponding student’s response that it elicited was analysed,with a focus on the type of question posed, how it was asked, and the relationshipbetween the cognitive level of the question and student’s responses. This ques-tion–answer pair was coded using cognitive categories, according to the type ofresponses generated. These cognitive categories reflected the type of thinking thatwas elicited. These included mere recall, as well as the higher-order cognitiveprocesses such as hypothesizing, predicting, explaining, interpreting, and makingconclusions.
To identify the different kinds and patterns of interaction in classroom talk, Itraced the questions asked, the responses that they triggered, and how the teacherfollowed-up on these responses. In particular, I examined the impact of precedingutterances on later ones. By examining student utterances before and after ateacher’s question, I traced how the question influenced what students said andwhether it elicited further thinking. I identified episodes of dialogues that seemed toprompt deeper thinking, move thinking forward, or lead to productive responses,and interpreted the questioning and follow-up that occurred within these. I looked atboth the content and patterns of interaction in the flow of discourse in the class-room, paying particular attention to how the ideas evolved and progressed over time.Key strategic moves or questions that appeared to change or influence the directionand content of the talk were noted. Analysis thus focused on systematically analysingwhat was observable, in terms of turns or moves, and then on whether any emerging
1322C. Chin
patterns in the forms or functions of the discourse could be discerned, especially inassociation with the teachers’ input (Westgate & Hughes, 1997).
Results
The “Questioning-based Discourse” Analytical Framework
Four aspects of classroom discourse (namely, content, type of utterance, thinkingelicited, and interaction pattern) constitute the elements of the “questioning-baseddiscourse” analytical framework. Content refers to the scientific ideas and conceptsaddressed in the discourse. The type of utterance refers to whether it is a question,answer, statement, or comment. In the context of this framework, a statement refersto further content-related propositions made by the teacher, whereas a comment isan evaluative or neutral utterance given by the teacher in response to a student’sreply to her question. The latter could also take the form of a restatement or refor-mulation of the student’s response. The thinking elicited was coded according to thetype of cognitive process associated with a student’s response. Finally, the interac-tion pattern takes into account, the nature of students’ responses and reactions toquestions initiated by the teacher, the type of feedback given in relation to thepurpose of the question, as well as the form and function of the utterance.
An analysis of the follow-up or feedback given by the teachers in the IRF sequenceshowed that this was typically in the form of a comment or statement followed byeither another question, or further statements that expounded more scientificcontent. Thus, the “F” part of the three-part exchange could comprise a “comment–question” (C–Q) or “statement–question” (S–Q) couplet where the questioncomponent of the couplet may be regarded as overlapping with the initiation or “I”move of the next IRF sequence. However, if no questions were asked, it took theform of a “comment–statement” (C–S) couplet. At times, feedback consisted of onlycomments (C) or statements (S). An example of a questioning sequence based onthis framework is represented visually in Table 1.
In Table 1, the column titled “Move” indicates the form of the utterance (I, R, orF) while the column labelled “Purpose of utterance” represents the purpose or func-tion in that discourse move (e.g., elicit, reply, probe, extend). Entries in the columntitled “Type of utterance” indicate whether the utterance is in the form of a ques-tion, answer, statement, comment, or a combination of more than one type. Takentogether, these three components (namely, move, purpose, and type of utterance)represent the “interactive” aspect of the discourse. The final column, entitled“cognitive process”, indicates the thinking processes associated with students’ utter-ances. Since it was not possible to gain direct access to the minds of the students,this analysis was inferential in nature, and based on what was known about the class-room context. Analysis of the relationship between the interactive and cognitiveaspects of the discourse helped to identify patterns embedded in the talk, and toidentify any specific teacher discourse-moves that facilitate productive responses instudents.
Table 1.Sample excerpt based on the “Questioning-based Discourse” frameworkMoveIRF–IRF–IRF–IAC–QAC–QAS–QReplyAccept, elicitReplyAccept, elicitReplyFocus, elicitQElicit–Hypothesize/recall–Hypothesize/recall–Hypothesize/recall–Type of utterancePurpose of utteranceCognitive processSpeakerUtteranceTeacherStudent 1TeacherStudent 2TeacherStudent 3TeacherStudentsTeacherStudentIRF–IRF–IRF–IRF–IQAC–QAQAC–QAS–QRF–IRAC–QAReplyProbeReplyProbeReplyProbeReply, justifyExtendReplyAccept, elicitReplyFocus, elicitRecall–Observe–Evaluate–Explain–Compare–Hypothesize–TeacherStudentsTeacherClassroom Interaction in Science1323
StudentTeacherStudentTeacherStudentTeacherWhat are the factors that affect the rate of dissolving? … What do you think …Temperature of solvent.Temperature of solvent. What elseThe rate of stirring.The rate of stirring. How fast you stir it. And also YesThe volume of the solvent.Yes. To be more specific, we are talking about size of solute … surface area … What do you observe in daily life that has a relation to the … size of solute?. … Let’s say, in the morning when you make a cup of Milo or a cup of coffee … You want to make it sweet. What do you useSugar.You think of sugar. And does … the size of sugar have an effect? …Yes. (The student then shared his observation that sugar is sold in both fine and coarse grained forms).What type of sugar would you like to use?Fine grain sugar.Fine grain sugar. Why would you like to use fine grain sugar? Yes, anyone wants to share?Because size of solute affects the rate of dissolving.So you find that, uh, does it dissolve faster or slower?Faster.Faster. Okay, what is the theory behind here? Can anyone propose? …Because the size of solute is smaller and covers more surface area …Let’s say, we have this cube A represents sugar. And we take this cube A and break it into many, many tiny cubes … There is a lot of it and all of them will have the same volume as this. So which has larger surface area? A or B?Table 1.(continued)MoveRF–IAC–QReplyAccept, elicitHypothesize–Type of utterancePurpose of utteranceCognitive process1324C. Chin
SpeakerUtteranceStudentsTeacherStudentTeacherRAReplyRF–IAC–QReplyProbePredict–PredictStudentBB. As you know, if you cut them into several pieces, the surface area exposed to the surrounding increases or decreases … What was your prediction, CharlesMy prediction was smaller size, bigger surface area.The smaller the size, the larger surface area. Therefore, what happens to the rate of dissolvingIncrease.Note: I, initiation; R, response; F, follow-up; Q, teacher question; A, student answer; C, teacher comment; S, teacher statement (for type of utterance).Classroom Interaction in Science1325
This excerpt in Table 1 occurred during a lesson on designing an investigation tofind out the effect of surface area on the rate of dissolving sugar in water. Theexcerpt was taken from the beginning of the lesson where there was a whole-classdiscussion on the factors that affect the rate of dissolving. In the excerpt, the teacherfirst elicited students’ ideas about the factors that affect how fast a solute dissolves ina solvent. Students hypothesized that variables such as the temperature of solvent,rate of stirring and volume of solvent would affect the rate of dissolving. It is proba-ble that students were engaging in hypothetical thinking as this was the first timethat the topic was formally introduced in secondary school. However, it is also possi-ble that the students’ responses were based on recall from primary school sciencelessons or of previous daily life experiences with dissolving sugar or salt in water.Since it is not possible to tell the exact nature of the cognitive process at play in theseinstances, the cognitive processes are labelled as “hypothesize/recall”.
The teacher then posed questions pertaining to the size of the solute. Using thecontext of dissolving sugar in a beverage, she then asked students whether theywould prefer to use fine or coarse sugar, and to give their reasons. Her discursiveinteractions with students consisted of a series of C–Q couplets where her questionsbuilt on students’ earlier responses. Typically, she remained neutral in her responsesto students’ replies and offered explicit evaluation only occasionally. In replying tothe teachers’ questions, the students were engaged in the cognitive processes ofhypothesizing, recalling, observing, evaluating, explaining, deducing, and predicting.Using this framework, verbal data from transcripts of classroom discourse wereanalysed to identify the various ways in which the teachers gave feedback followingstudents’ responses to their questions.Types of Teacher Feedback to Students’ Responses
In response to a teacher’s question that solicits factual information or that taps intoconceptual understanding or reasoning, a student’s answer could be either scientifi-cally correct or incorrect. After evaluating the student’s answer mentally, the teachercould verbalize this evaluation publicly to the class by providing a comment.Alternatively, he/she may not articulate this overtly but keep this evaluation silentlyto himself or herself, thus remaining neutral in his/her response. In the case of a correctanswer, this evaluation could be in the form of a praise or acknowledgement; whereasfor a wrong answer, the teacher might either issue a put-down or remain neutral.Thus, as a follow-up to a student’s correct answer, a teacher could proceed ineither of two ways: (a) affirm the answer, reinforce it, and then move on to furtherexpository talk via direct instruction; or (b) accept the answer and then ask anotherrelated question or series of questions that build on the previous ones to extend theline of conceptual thought. On the other hand, in response to a student’s answer thatis incorrect or that deviates from the scientific norm, corrective feedback could bevia (c) explicit correction followed by further expounding of the normative ideas, or(d) evaluative or neutral comments followed by reformulation of the question orchallenge via another question.
1326C. Chin
Unlike feedback types (a) and (c), which did not encourage student input beyondthe initial solicited answers, feedback types (b) and (d) further elicited students’responses, stimulated productive thinking, and extended lines of conceptual thoughtin students. The different types of teacher feedback are referred to as Affirmation-cum-Direct Instruction, Focusing and Zooming, Explicit Correction–DirectInstruction, and Constructive Challenge. These different types of feedback aresummarized in Table 2, and illustrated in the following by specific examples fromthe dataset.
Feedback in Response to a Student’s Correct Answer
Example 1: Affirmation-cum-Direct Instruction. The following excerpt illustrates acase (comprising feedback type (a)) where the teacher affirmed and reinforcedstudents’ correct answers and then moved on to expound further scientific informa-tion via direct instruction. While the discourse was dialogic in nature during theinitial stage, teacher talk was authoritative and had a transmissive function duringthe latter expository phase. This type of feedback is referred to as “Affirmation-cum-Direct Instruction”.
The lesson was on the topic “Respiration”. The teacher had given the studentsseveral “thinking tasks”, one of which required students to imagine that a givenpiece of plasticine was an organism. Students had to first work in groups to discusstheir answers to given questions on a handout. Following this, a representativefrom each group presented their findings. The questions posed on the handoutincluded:
Table 2.Nature of student’s responseCorrect
Types of feedback to students’ responses in the ‘F’ move of the IRF exchange
Predominant nature of key utterancesC–Q, C–S
Type of feedback(a) Affirmation–Direct instruction
(b) Extension by responsive
questioning: Focusing and Zooming(c) Explicit
correction–Direct instruction
(d) Constructive challenge
Description
Affirm and reinforce response followed by further exposition and direct instruction
Accept response followed by a series of related questions that build on previous ones to probe or extend conceptual thinkingExplicit correction followed by further expounding of the normative ideas
Evaluative or neutral comment followed by reformulation of the question or challenge via another question
Mixture of correct and incorrectIncorrect
C–Q
S–Q, S
C–Q, Q
Note: C, teacher comment; Q, teacher question; S, teacher statement.
Classroom Interaction in Science1327
●
●
●
How would you shape it such that there is a maximum exchange of gases betweenthis organism and its surroundings?
What are two factors observed that can help to ensure the maximum exchange ofgases taking place?
Where can these features be found in our bodies?
These questions were also displayed on a PowerPoint slide on a screen at the front ofthe classroom. Four students then went up to the front of the classroom to showtheir moulded products. Table 3 shows the conversation that followed, based on theproposed framework given.
In this excerpt, the teacher got the students to observe the features of their moul-ded products by asking questions such as “What would the shape be?” and “Whatcan you say about the thickness?”. She also told the students to act as “judges” andto evaluate which sample would be “the best shape for the maximum exchange ofgases”, as well as to explain and justify their choice of sample. After this, she stimu-lated analogical thinking in students by asking “Which organ in your body has suchfeatures?”. The students were then further posed the question “How are you goingto contain it?” where they had to figure out how to get the plasticine into a smallcontainer, without changing its shape. In trying to solve the latter problem, studentshad to generate a variety of ideas including “rolling”, “flattening”, and “bending”.During the conversation described, the teacher encouraged students’ thinking byaccepting responses in a neutral manner, affirming the responses with a commentsuch as “okay”, or restating students’ responses. This was followed by a questionthat built on the student’s previous response. Thus, the teacher’s replies to students’responses consisted of a series of C–Q couplets. Subsequently, the teacher consoli-dated the key points embedded in the students’ responses by introducing the appro-priate scientific vocabulary (such as lungs, air sacs, large surface area, bloodcapillaries, large network [of blood capillaries], one cell thick, rapid exchange ofgases) in the form of direct instruction. During this latter phase, feedback consistedof a series of teacher statements, and was predominantly of the C–S couplet type.Feedback in Response to a Mixture of Correct and Incorrect Answers
Example 2: Extension by responsive questioning (Focusing and Zooming). In this exam-ple, the teacher remained neutral in her evaluation of students’ responses (even whenthe responses were inconsistent with the scientific norm) but asked a series of furtherrelated questions that extended student thinking (feedback type (b), whichcomprised mainly C–Q couplets). These questions were used to elicit, probe, extend,and elaborate students’ thinking, with a view to helping students construct concep-tual knowledge. There was “responsive questioning” in that the teacher adjusted herquestions to students’ responses, with each subsequent question building on to theprevious one(s) to help students progressively construct a related and integratedframework of ideas. The teacher also invited other students to respond to a givenanswer. Discourse was dialogic and occasionally authoritative, and had a facilitative
Table 3.Example of excerpt for Affirmation-cum-Direct InstructionMoveIRF–IRF–IRF–IRF–IRF–IAC–QAC–QAC–QAC–QReplyAccept, probeReply, justifyAccept, extendReplyExtendReplyAccept, extendQAC–QElicitReplyAccept, extendType of utterance–Describe–Evaluate–Explain–Describe–Apply/recall–Purpose of utteranceCognitive process1328C. Chin
TurnSpeakerUtterance123TeacherStudentTeacher45StudentsTeacher67CharlesTeacher891011StudentsTeacherStudentsTeacher121314151617StudentTeacherStudentStudentTeacherStudentWhat would the shape beFlattened …Okay, which do you think is the best shape for the maximum exchange of gases? … You will be judges. Which one would be bestCharles’s …Okay, Charles, can you tell the class why your friends think this is the best? …This has the largest surface area.Okay, this has largest surface … What can you say about the thickness?ThinThin … Which organ on your body has such features?LungsOkay. Very good. Lungs … I have another more mind boggling thing. Imagine I have this thing [holding a container]. Don’t change the shape. Try and contain it into this vessel. How are you going to contain it? You may do anything but don’t change the shape.Roll it upRoll it up, okay.Try flattening it. [Laughter from the class]Roll another way.Any other way? …Bend it, bend it. [Students tried different ways, some without success.][Finally, one student “succeeded” in getting his plasticine into the vessel without altering its shape and held up his sample for the class to see. The class cheered.]RFRRIRACAAQAReplyAcceptReplyReplyElicitReplyHypothesize–HypothesizeHypothesize–HypothesizeTable 3.(continued)MoveFC–SAccept, expound–Type of utterancePurpose of utteranceCognitive processTurnSpeakerUtterance18TeacherOkay. What he did was he rolled it up! And when you roll it up, what happens to the features Do you realize it’s thrown into folds Have you ever wondered why our lungs are in this shape … [showing a slide] Lungs are organs. Lungs are actually organs. If you were to look into the organs, that’s the tissue. You will see that at the end of tissue, you have air sacs. If you look at the cross section, do you see that it is thrown into folds? … I want you to put down whatever you are doing. Listen, you see they are thrown into folds. There were two factors that you said. First, the large surface area. That’s correct. Let’s look further in. You see large surface area. You can get large surface area if these tissues are arranged into ball-like structures … And to further increase the large surface area, your lungs have blood capillaries, a large network of it.... That’s the first point. Second point, in order to be very thin, it is one cell thick. Why one cell thick So that there is rapid exchange of gases taking place.Classroom Interaction in Science1329
1330C. Chin
function. This approach to questioning is termed “Focusing and Zooming” as thequestions zoomed “in and out”—alternating between a big, broad question andmore specifically focused, narrow, subordinate questions. The teacher used her“questioning lens” to adjust and refocus the nature of her questions, as appropriate.As a prelude to learning about density, the students had just learnt how to find thevolume of a solid such as a marble, by the method of displacement of water using ameasuring cylinder. The teacher then posed the question of how to determine thevolume of sugar and a piece of sponge. An excerpt is presented in Table 4 for thecase of sugar.
The overarching, macroscopic question here related to how the volume of sugardetermined by the displacement method would be affected by its dissolution inwater. The teacher first elicited students’ ideas by asking students to give theirpredictions. Students’ different answers included “the same” and “it will be less”.The teacher then probed students to explain their reasoning by asking furthersubsidiary questions such as “can you elaborate on why the volume will be less?”which required them to think at the submicroscopic or molecular level (zooming-in).After focusing on the case of sugar, the teacher moved on to the case of a piece ofsponge (zooming-out) using a similar questioning approach. Finally, she posedanother overarching question: “If you want to determine the volume of the materialby this method of displacement of water, what must be the nature of this material?”.Through a series of responsive “focusing and zooming” questions, she guidedstudents to generate their own inferences and conclusions, where they gaveresponses such as the material must be “non-porous, it does not absorb water”,“must not dissolve in water”, and “it must not float”.Feedback in Response to Student’s Incorrect Answer
Example 3: Explicit Correction–Direct Instruction. An example based on feedbacktype (c) is now given in a lesson on density where the teacher elicited students’ ideason how fishes sink and rise in water. A student proposed an answer and the teacherthen invited peer evaluation from the class. After explicitly pointing out the student’smistake, the teacher proceeded to give the correct answer and then carried on withtelling students more information. Table 5 presents a description of this excerpt.The teacher’s feedback was in the form of an overt correction followed by a seriesof statements alluding to experiences with swimming. By having the correct answertold to her directly, the student, Yi Le, who had inappropriate conceptions about thetopic, did not have to do the cognitive work to reason out the answer herself.Example 4: Constructive Challenge
Here, the teacher did not articulate the student’s mistake in the form of an evaluativecomment. Instead, she remained neutral but challenged the student by posinganother question, thereby throwing the responsibility of thinking back to the studentin the form of a reflective toss (van Zee & Minstrell, 1997a). She did this by further
–Cognitive processPredict–Predict–Explain–Hypothesize–Predict/guess–Classroom Interaction in Science1331
Explainti fctioilc eeilec s,eony ,fpartyyeyeyeyilillllrecprpbpbpbputitlealeoeoeoePuERCRrPRrPRrPR efcon eapreytQQQQQ----Tt–uSACACACACAevoIIIIM----IRFRFRFRFR ttacrs ’heo ffenett ie s tave rcio ar o ao tsit,h hLd nym ) cat.aipi…ramah eth g bk d ,wrs eueeOtrnu’ naisei ,ds o?do dnhdm usbohklg tasoitliwloeu ns thyovloen tii tcw h e rewo lu gsmg,b g,cnenmraeeg rnieiol iltulelneritwoeD mubaswaap?sow e aar ewrmpa sepnht i m.Drau(ah Hsime lsliu…ehth hel? tav TlWo…gus i ks st ?vsn? aueg o e eilhsnsnss…ui…i me dehS hetirhw s d?nura tgatb wa …, is l lwWd?arg n tificeuieap .ey ts hhw?eec rausvdlua dMttee e rlooosgeusbywkn cAahimhptcsyi n fo tiuws l eaefiior?C?d yo t ettfkav.mhtwldn sA oanohg…tcns eo.r ecrfe?fep iomlke i eherha tu ugta ome?ffdata peercnioemr ,e scluuy awsemetevll curlenniuoaoohpp ekuaoolwlottm ,yus hat Yw o mssi? pva wre, nspt tv dor tns eietmh.?eu ni …adhnieausso e afn sseksif t ds taehi hego r.,eetktttrunoewf s c oeeeeeella i a rso nnwdivlmhtmmeeo]art wud seesnesaoaabbbat B ceeromou rlssllagI] laemmY ?en llu?m.rlh hrssoOtteiv eeiie shoetteetd ?UWme eshhww[rapt ooe u usycncdthsitelTTttotFieeaoIIyIH[WdasbhppsNg111 rrn irtrtrraeekhMehneneteuac ehehnheaacdcdcecH pehaaeueuadakteueuSTCTtSTSTtSTSnruT01234567891ebyloprePRQAIR ne erewttaewb tnui os.bescelaau gpcnse likerolaam t ererreetehawTw e.eWd?hegt kn ?inecnecaeeappp pwsy lateeershb oosmliec tc paauoshp egws ek,anightent itbi vp lutto ’ossnnsesi esdkoa ad nt i s ryeoahlwuSgc .u delms noeeAmhhtTrethnceadeuTtS1211Table 4.Example of excerpt for Focusing and ZoomingTable 4.(continued)MoveF-IC-QAccept, elicit–Type of utterancePurpose of utteranceCognitive processTurnSpeakerUtterance1332C. Chin
13Teacher1415161718Suk HuaTeacherHuiyiTeacherHuiyiRF-IRF-IRAC-QAC-QAReplyElicitReplyProbeReply, justifyPredict–Predict–Explain, theorize19TeacherOkay, so the sugar takes up the space between the water molecules, which were like empty spaces originally. So how then, does that affect the volume of water … If these water molecules are all closely packed together and if there is no space for this sugar molecule, it will take up all this space. And it will displace volume of water. Right? But the water molecules have a lot of empty spaces in between, [the sugar molecules] squeeze in between, and take up the space, which was originally empty in that sense. So therefore, the volume of the water displaced by these sugar molecules, is it still the same as or less than what it should displace [be]?Same.Same, uh You think it will be the same. Huiyi, what do you thinkLessIt will be less. And why do you think it will be less?The water displaced will be less. Because the sugar molecules will seep inside the [inaudible] … like take up the space in the spread-out water molecules. So there will be less. And then those that cannot [inaudible] … they have no space already, they will have some displacement. Then that [volume of water displaced] will be less.Uh-huh … Those [sugar molecules] that don’t squeeze in between will be the ones that displace the water upwards. It’s just like if you have a box … If the box is all taped up and you have some materials inside the box, but the box is not fully packed … The box is quite empty so you squeeze in some more things … When you close up the box again, the box is still occupying the same space as the volume of the box … It is very similar to this idea. (Drawing diagram on board) … So the volume that you determine would have been less, as the molecules of rock sugar can … move in between the spaces of water molecules … as a result, displacing a smaller volume of liquid than its own.F-IC-SAccept, consolidateTable 5.Example of excerpt for Explicit correction–Direct instructionMoveIQElicit–Type of utteranceCognitive processPurpose of utteranceTurnSpeakerUtterance1Teacher23RF–IRF–IAS–QAS–QReplyCheck, elicitReplyInform, linkYi LeTeacherRF–IAC–QReplyClarify, focusHypothesize–Hypothesize–Evaluate–45Yi LeTeacher67StudentsTeacherClassroom Interaction in Science1333
89Yi Le TeacherHow do fish sink and rise in water? Anybody wants to suggest an answer What do fish do when they want to go down into the water? And what do they do when they want to rise to the surface of water? Yi Le?The fish releases air. It can also absorb air.Yes. It releases or absorbs air When does it need to release airWhen it wants to float. It rises.When it wants to float, it needs to release air. Is that correct?No. [Class laughed and teacher also laughed].When it wants to sink, it needs to release air. Just like when you want [pause] those of you who have gone swimming. If you want to make yourself sink, it’s not that easy. You need to take in air or blow out air? You need to blow out lots of air before you can sink, right? Same thing. And when you want to rise, you must [pause]?Absorb air.You must take in more air.RFASReplyRestate, reinforceHypothesize, deduce–1334C. Chin
elicitation as well as reformulating her question in the form of a recast, which forcedthe student to reflect on and reconsider her answer (feedback type (d)).
In a lesson on density, the teacher posed the problem of how to find the density ofone’s body. A student had estimated that it would be a figure close to that of wateras “our bodies have 70% of water”. The teacher then asked students how they woulddetermine their own volume so as to calculate their body density. After some groupdiscussion, a student presented her group’s ideas while the teacher drew on theboard as the student spoke. The drawing showed a stick figure of a personsubmerged in a tank of water with two water levels indicated by V1 and V2. Table 6shows the interaction involved in the conversation between the teacher and studentpresenter.
Upon detecting the error in the student’s thinking that the markings on the tankrepresented heights and not volumes, the teacher did not explicitly tell the studentthat she had made a mistake. Instead, she posed a series of questions, articulatingthem in several different ways, with each question cast in a slightly different wording.These questions provided the student with cues to draw on her own conceptualresources. They prompted the student to self-evaluate her own thinking, reflect onher incorrect assumption made earlier, discover the fallacy in her reasoning, and torectify her mistake. Subsequently, the teacher also stimulated the student to evaluateher proposed method to see whether there were possible sources of error. Thestudent became aware that the presence of clothes on a person’s body would affectthe accuracy of the volume determined.Discussion
Questioning is a significant part of teaching and science talk. As Tsui, Marton, Mok,and Ng (2004) have pointed out, questions posed at critical junctures of a lesson canfocus students’ attention on the critical aspects of the object of learning, and openup the space for further inquiry and learning. This study was undertaken to betterunderstand how teacher questioning is woven into everyday instruction and how itinfluences subsequent student responses. A “questioning-based discourse” analyti-cal framework was developed for the description and analysis of classroom discoursein science, with a focus on questioning-based practices. It was used to analyse waysin which teachers used questions to frame and guide classroom discourse, andprovides a fine-grained description of the subtleties and nuances for a range of possi-bilities in the IRF teaching exchange. This characterization of teacher questioning inthe classroom contributes to an understanding of how questions can stimulateproductive thinking in students as part of a teaching sequence.
Analysis of classroom discourse using the aforementioned framework revealedsome recurrent patterns in the teachers’ interactional strategies whenever productiveresponses were generated by students. Where the F-move in the customary IRFcycle or “essential teaching exchange” (Edwards & Westgate, 1994) was not justevaluative but also supportive in that it embedded a further question that provokeddeeper thinking beyond simple recall, it engaged students in more cognitively active
Cognitive processProblem-solve, apply, generate–Reflect, self-correct–Generate, apply–Classroom Interaction in Science1335
Analyse, identify– ,e ftoal eeecugsonnparymreyllrepollputfeeaePtuRRhcR efcon eapreytTtuAQAevoMRIR s,f’td e]anruo?. r n. heotyeegt….onstA a , mani?t efwaer. wrtoe)wt u hhg f o t1y ndidemuldgiosrfetatie nhuotlv aaieeeeeiAV(hhhttg . sehs’ ir.nht pttfeena eo mmrhh eetsuontvg nu ortoh tnsrohtf saeirll tdeeieeAroioAde rf etuhtuv eeennoriegarmh .hpeppc st tmkiae…tut es itaetmd uresleiersh auarltt? rtdeshnaokr hcehtm ebo ,itsg suevl tnt awvalr lf e ieedteTifkmagaf oodo Io t)y auchl ntt dhe o s h .w s s osi iete’sertdTtn ee l.ekedwuysnmkimmtuvidnrgenAo harettooteiaulasuuegi .yd e asrflnkroei eAhwsrpttnl i oodidrtr n fevn ovvufa…t ? a rmeoyeemh en es e wou t mhd eTehhwetohhehomo Bu… tyh t . th u Hd .oreetYe we teedd Tet gwl?am.lbwd fpkne .bnnt otrssiev ihbuenooooaaAi]l sh eltgillda rhb rehLr eiru hiipsinsT . .t .e)a tewo eaosdrdnt enho dr2 hcv sdeetieditrneyhet eaeedgi atotsVw derba(f ohn] waen ahc]etai dutofuwni r etaTa.gtrrkrgrt gelun ssleC foehht.dnniba snbio ubec ogtp. 2ei aaott phlsktleeVtkws2e[dducnat irhV m geas teedtap cnnligdad uauwo teto me rthd gdtiuniar hheootr oe eenohet k ggnhtrdr ha ytbre irs imnoo gfeht slakesanbas reia rs i rn1eh on1ruatiVeha o aneeitccesd pVhdt ahtc o es s cegrinah da pa ltacI aeeeereeucegn?Ona Yrer t?atS tnni eer .euh ey?eeymssnothc tia eetlnrdmdel m nthuup trhliearefuniuetdlce ndstUToosiTaohfetiroll[bvdWeewhnebtagu[s[cP(dlaawYyohcvTrtreetknhnaeceedadputeuSSTtSnruT123yficeptesis c,iaelenet u,,ttt-ypypellnpecpeciececFRARAQ–CACACIFRFR–F l a ensrhoot irftrceoe syy-ntsssAeo dr?csot. ,nsmaee ermhetha mot ltesoc uacso ibyhb ,n a aeAgnw i.r?roaayae l?eH rets.ahnrws tee o ehsrrnhrotteefes o lsmfitreidmt uec ttiop gibenth gdiahirti hedsTagh s ei nu]w]enoce d.isssneth iuh eredaohgn sgici up reenp oaeu hceyrnyrpal nneefAdeestif ifhs ri.eedtDa Tal cff h nieg?g]Cedhidng[ ir ten?, Ree .vdinlldifhfakaoiDeaevomosrYTranNSei[[prretetrhnhnececehadadceueuaTtteSTST45678Table 6.Example of excerpt for Constructive Challenge1336C. Chin
roles. In such instances, the teacher’s subsequent questions stimulated students toformulate hypotheses, predict outcomes, brainstorm ideas, generate explanations,make inferences and conclusions, as well as to self-evaluate and reflect on their ownthinking. Thus, the thrust of the teachers’ utterances in the F-move consisted notjust of an evaluative comment and further statements (i.e., C–S couplet), but rathera comment and/or a further “productive” question (Elstgeest, 1985), in the form ofa C–Q couplet, that took students forward in their thinking.
The teacher moves, in such cases, did not appear to be overly evaluative in eitheran explicitly positive or negative way (e.g., in the form of praises or put-downs).Instead, they were often neutral or accepting of students’ responses (in the form ofsubtle affirmations such as “okay”) if they were appropriate. As Hogan, Nastasi, andPressley (2000) reported in their study on discourse patterns in teacher-guided smallgroup discussions, “although the roles of teacher as questioner and student asanswerer are similar to those found in traditional recitation-based classroomexchanges, the crucial differentiating feature of this question-answer exchange is thelack of evaluative statements by the teacher” (p. 411).
Another common feature observed in the teachers’ comments in the F-move wasthat the teachers often restated students’ responses following their questions. Thisphenomenon, termed “revoicing” by Chapin, O’Connor, and Anderson (2003),served not only to affirm students’ responses, but also to make their ideas availableto all in the class, thereby making it “common knowledge” (Edwards & Mercer,1987). In addition, teacher feedback in the form of paraphrasing students’ responsesmay allow students, particularly those with weak language abilities and who mayhave difficulties in verbalizing their thoughts, the opportunity to co-construct aresponse with their teacher and peers. In doing this, the teacher provides not onlyconceptual but also linguistic scaffolding, adjusting both the cognitive and linguisticloads of students.
A further characteristic of facilitative F-moves was that teacher questions immedi-ately following teacher feedback built upon students’ previous contributions,supporting students’ cognitive activity, and allowing progress towards a jointconstruction of the concepts concerned. These questions were responsive tostudents’ answers, followed from, and were closely linked to the utterance in thefeedback from the preceding F-move. They were integrated with each other as partof a coherent framework in that they addressed various types of thinking related tothe scientific concepts being discussed. Common purposes associated with teacherutterances include those of “accept” and “elicit”. The former affirmed students’contributions, while the latter used students’ knowledge as starting points and raisednew questions to take students’ thinking forward. These “elicit” moves were thusnot of the “closed” kind but were genuine questions, which often appeared to havethe sole intention of “drawing out”, probing, and extending what the students werereally thinking.
Figure 1 represents the possible range of purposes underlying teachers’ utterancesduring the initiation and feedback moves in facilitative IRF iterations. Three relatedsets of purposes are depicted, based on the findings in this study. In the initiation
Classroom Interaction in Science1337
1nitiationDrawoutElicit,probe,extendCueandProvokeClarify,prompt,challenge4esponseStudentsÕResponsesEvaluateStudentsÕResponses.eedbackReinforceAffirm,restate,consolidateFigure 1.Purpose of teachers’ utterances during facilitative IRF iterationsmoves, the set termed “Draw out” consists of teacher questions that aim to elicit,probe, and extend students’ thinking. These work in tandem with “Cue andprovoke”, where questions are designed to clarify, prompt, and challenge students’responses. As for the feedback move, teachers’ utterances that affirm, restate, andconsolidate students’ correct ideas have the overarching purpose of reinforcing thekey scientific concepts involved in the lesson. The solid arrows in the diagram depictthe two-way interaction between the initiation and feedback moves that are oftenclosely linked. As teacher evaluation is central to the teaching exchange and wasobserved to be typically neutral and covert in nature, it is placed in the middle of thediagram and linked to the other purposes by dotted lines.
What might be some factors that determine the nature of students’ thinking andresponses, and what might prompt each of the four types of teacher feedback? Tosome extent, the type of cognitive processing that students engage in depends on thedemand level at which the teacher’s question is pitched (e.g., whether the questionrequires students to recall, hypothesize, explain, or predict). In turn, the form of theteacher’s feedback and follow-up may be triggered by the kind of response given bythe student.
The Structure of the Observed Learning Outcome (SOLO) taxonomy (Biggs,2003; Biggs & Collis, 1982) describes the nature of students’ responses in terms oftheir sophistication and structural complexity. At the lowest level of “pre-structural”responses, there is little evidence of understanding. “Uni-structural” responses focus
1338C. Chin
on one conceptual aspect in a complex case, while “multi-structural” responsesshow several components that are discrete and unrelated to each other. Evidence ofdeeper understanding is manifested in “relational” responses that show integrationof concepts. At the highest level, “extended abstract” responses show conceptualiza-tion beyond what has been dealt with in the teaching, as well as application to newdomains.
As an illustration, consider the case of Example 2 earlier (Focusing andZooming). When asked how the value for the volume of sugar would be affected,Cha Ming’s answer of “the same” (turn 2) would be pre-structural since little under-standing was shown. The answer given by Student 1 (turn 4) “it will be less” wouldbe uni-structural as it was only a minimal response. Subsequently, when Student 1went on to elaborate that “it [sugar] takes up space” (turn 6) and “[from] within thewater” (turn 8), features of a multi-structural response are evident as the studentattempted to articulate more than one idea, although these were given in twodiscrete pieces. Upon further probing by the teacher, another student gave aresponse “the sugar takes up the space between the water molecules” (turn 12) thatwas relational, as this showed the integration of two concepts (namely, that there isspace between the water molecules and that the sugar molecules take up this space).Towards the end of this given excerpt, Huiyi’s attempt at theorizing the mechanisminvolved at the submicroscopic level (turn 18) would be indicative of an extendedabstract response.
From this example, it appears that one factor influencing the form of a teacher’sfeedback and follow-up questions is the nature of the response made by the student.In the example, when the student, Cha Ming, gave an incorrect answer, the teacherrestated her question and redirected it at another student, Student 1 (turns 3 and 4).Then, when Student 1 was able to give the correct answer, albeit at the uni-struc-tural level, the teacher posed further questions that nudged students towards multi-structural, relational, and extended abstract responses at an ascending order ofcognitive complexity.
Apart from the nature of students’ responses, there are other conceivable factorsthat determine the form of a teacher’s feedback and follow-up, although these arebeyond the scope of this study. These include the nature or difficulty level of thetopic being addressed, the ability level of students, the curriculum time available,and the teacher’s epistemology and preferred style of teaching. In table 7, I postulatethe different conditions under which each type of feedback might be used.Implications for Instructional Practice
The findings of this study showed turn-taking and discourse patterns other than thetraditional IRE sequences. Some exchanges were of the IRFRF pattern (Mortimer &Scott, 2003) or an IDRF type where students discussed their ideas in pairs or groups(hence, the “D” for discussion) following a teacher’s initiating question, beforeresponding to the question. Depending on the nature of the activity, some question-ing sequences had the quality of exploratory and facilitative rather than evaluative
Table 7.Conditions under which each type of feedback may be usedType of feedbackAffirmation–Direct instructionFocusing and ZoomingExplicit correction–Direct instructionReconstructive challengeConditions under which feedback may be used✓✓✓✓✓✓✓✓✓✓✓✓✓✓✓✓✓✓✓✓Nature of students’ responsesWhen students’ answers are scientifically correctWhen students’ responses contain both correct and incorrect ideasWhen students’ answers are scientifically incorrect or insufficiently detailedNature or difficulty level of topicWhen new scientific vocabulary needs to be introducedWhen concepts addressed are familiar to or within grasp of the studentsWhen concepts addressed may be too difficult for students to handle on their ownAbility level of studentsWhen teacher believes students may not be capable of formulating or reasoning out the answers on their ownWhen teacher believes students are capable of thinking at progressively higher levels (e.g. relational and extended abstract levels of the SOLO taxonomy)Time availableWhen there are constraints in curriculum timeWhen time is available for extended classroom discourseTeacher’s epistemology, skills, and preferred teaching styleWhen teacher views that dispensing information or teaching via exposition is appropriateWhen teacher subscribes to a constructivist epistemologyClassroom Interaction in Science1339
1340C. Chin
talk. In such discursive episodes, teachers’ questions served to scaffold students’thinking and nudge students toward conceptual development instead of just assess-ing the correctness of their responses. The findings also revealed that the follow-up(F) component of the IRF structure could take various forms. This suggests that bychanging the third turn of an IRF questioning sequence from an explicit evaluationto one that includes “responsive questioning”, teachers can make their classroomdiscourse more thought-provoking and stimulate more elaborate and productivestudent responses.
An implication of this is that an initiating I-move should be closely tied andaligned to the preceding follow-up or F-move, to ensure meaningful and progres-sively links among the IRF chains. Also, teacher initiations do not have to be realizedthrough closed or directive questions belonging to a strongly-framed teachingagenda, but should be responsive to the preceding students’ responses. As Mortimerand Machado (2000) have pointed out, the IRF pattern of discourse is authoritativeor univocal as long as the feedback from the teacher is an evaluation. However,where the feedback is elaborative, in that it allows for a further extension of theresponse by students, or elicits new ideas and contributions from them, the IRFpattern corresponds to a dialogic function.
In a study on managing the conclusion phases in teacher–student interactions,Morge (2005) differentiated between the Question–Answer–Evaluation (QAE)pattern in recalling knowledge, and the QAE pattern in inquiry. When recallingknowledge, the question asked by the teacher is for an answer the student alreadyknows, but this is not the case for inquiry. Because the conclusion stage of an inter-action during an inquiry-based lesson is the point at which the teacher has to decidewhether to accept or refuse a production provided by a student in response to agiven task, Morge pointed out that the teacher must avoid using authoritative argu-ments if he or she is to move towards constructivist management of the conclusionphase. Thus, a second implication is that instead of judging a student’s response assimply right or wrong (which corresponds to a dogmatic view of science and theteaching of science), the teacher could instead delegate the control of students’productions to other students, as well as ascertain its validity by determiningwhether the response answers the question asked, and whether it is relevant to and isconsistent with reference knowledge. As Morge (2005) argued:
On an epistemological level, this mode of teacher-student interaction would becompatible with the idea of scientific knowledge that is constructed and which is aresult of human activity … On the pedagogical level, this mode of interaction isconsistent with a socio-constructivist conception of learning … for the knowledge isconstructed collectively in the interaction between the pupils and the teacher, whichthe theory maintains facilitates the pupils’ individual construction of knowledge.(pp. 942–943)
When students give scientifically incorrect answers to teachers’ questions, correc-tive feedback can be explicit in the form of an overt correction, or implicit. Implicitfeedback can take the form of a constructive challenge where the teacher posesfurther challenging questions or recasts her questions, although other manifestations
Classroom Interaction in Science1341
such as facial expressions or gestures are also possible. Feedback that does notprovide the correct answer explicitly may encourage learners to use their ownresources in eliciting self-correction and repair. This was seen in Example 4 earlier.In this regard, a third implication is that implicit feedback may improve students’ability to monitor their own thinking, and, under the appropriate conditions, couldbe more beneficial than simply providing them with the correct form.
However, because constructive challenge could be potentially threatening forsome students, it may not work well for all students. Zohar and Aharon-Kravetsky(2005) found that cognitive conflict had dissimilar effects for students of differentacademic levels. In a study on teaching the control of variables strategy thinking,students with high academic achievements benefited from a teaching method thatinduced cognitive conflict while direct teaching hindered their progress. In contrast,students with low academic achievement benefited from the direct teaching methodwhile the induced cognitive conflict teaching method hindered their progress.Similar findings were reported by Dreyfus, Jungwirth, and Eliovitch (1990), whereacademically successful students reacted enthusiastically to the “flabbergastingeffect” of cognitive conflict and confrontation, but students with low academicachievement did not, perhaps because of negative self-images and high levels ofanxiety.
The effectiveness of different types of feedback is determined by whether or notthe feedback results in productive uptake, and if it does whether it results in success-ful repair. Uptake refers to the student’s utterance or response that immediatelyfollows a teacher’s feedback and that constitutes a reaction in some way to theteacher’s intention to draw attention to some aspect of the student’s initial utterance(Lyster & Ranta, 1997). Future research could look into the differential effect ofdifferent types of feedback, the conditions under which different types of feedbackare most effective in mobilizing students’ ideas and triggering change, and the issueof uptake.
Teacher questioning that elicits information about students’ understanding andprovokes classroom dialogue is an important instrument for formative assessment.Because the quality of teacher’s questions can influence the degree to which thequestions do or do not extend students’ thinking and draw out their ideas, both theactual content of the questions and the ways of following up on the responsesbecome important (Black & Harrison, 2001). Thus, a fourth implication is thatteachers need to think about how questions can be constructed and used to developstudents’ learning.
Learning to talk science is more than simply being able to verbalize the appropri-ate words, phrases, and scientific terminology. Lines of discourse need to be devel-oped where students are continually engaged in various cognitive processes such ascomparing, hypothesizing, predicting, explaining, interpreting, inferring, and reflect-ing. In moving through a planned agenda of instruction, the teacher needs to decidewhat questions to ask and the sequence in which to ask them, to move studentsforward in their thinking. She also needs to decide how to adjust her questioning toaccommodate student contributions and respond to students’ thinking when guiding
1342C. Chin
students through inquiry-based discussions. In particular, if teachers are clear aboutthe kind of cognitive processes that they want to elicit in their students, then theycan craft questions that would stimulate such responses.Implications for Teacher Education and Professional Development
In contributing to an understanding of the role of language and the discursive inter-actions involved in the elaboration of scientific concepts, this study also providesimportant elements for teacher reflection in preservice and inservice education.Teachers could, for example, record teaching sessions, analyse the various interac-tions involving teacher questioning and follow-up, critique them from an epistemo-logical and pedagogical point of view, and reflect on alternative modes ofinteraction. This would be learning from posteriori self-analysis, which is in keepingwith the paradigm of a reflective practitioner (Schon, 1983).
This study provides specific examples of questioning-based practices that may beuseful to teachers who are interested in honing their discursive skills and in adoptingways of talking and classroom interaction that foster productive student responses.The “questioning-based discourse” analytical framework developed here helps makesalient and visible, the moves in classroom questioning sequences that can takestudents forward in their thinking. Because such discursive teaching strategies havebeen tacitly employed by teachers, they have generally been invisible to others. Also,unlike other approaches to discourse analysis that tend to focus on general aspects ofclassroom interaction irrespective of subject matter, this framework pays attention tothe content of science. It can thus serve as a useful heuristic for science teachers whoare interested in shifting their classroom discourse toward more reflective andinquiry-based practices. The findings from this study also have potential in translat-ing research insights into practical advice for teachers regarding tactical moves inclassroom discourse.Limitations of Study
One limitation of this study is that, in the analysis and interpretation of classroomdiscourse, the linguistic form (verbal data) was used as a predominant marker ofinteractional or cognitive function, and is thus at best inferential. This methodologi-cal issue was raised by Barnes and Todd (1977, 1995). A second limitation relates tothe generalisability of each respondent’s utterance to the rest of the students. Muchof the data in this study were derived from discourse in whole-class settings.However, at any moment in time there can only be one person responding to theteacher, except in the case of chorus answers. The analysis and interpretation of datawere based on the utterances and responses of individual members who participatedin the verbal exchanges, but collectively extended to the class as a whole. Theassumption was made that whatever applied to the individual respondents alsoapplied to the other students in class. This assumption has limitations as the processof internalization does not simply involve direct transfer from social to personal
Classroom Interaction in Science1343
planes and it is not possible to know for sure the extent to which individual studentswere able to internalize and make sense of the concepts addressed.
A third limitation of this study is that, because of the large class size andpredominantly whole-class teaching context, students seldom spoke up in class.Consequently, students’ responses were limited and non-elaborative. Also, inSingaporean classrooms, students are generally not so forthcoming with verbalizingtheir ideas publicly in front of their classmates or volunteering unsolicited informa-tion unless they are called upon to do so. Thus, there was not much verbal studentparticipation in classroom interaction. This is partly because of the average 12-year-old to 13-year-old student’s lack of fluency in spoken English. Although the mediumof instruction is English in all schools, more than 50% of students do not speakEnglish as first language at home. Thus, these students may find it difficult to articu-late and verbalize their thoughts in English although they may be actively engaged inconceptual thinking. Another reason might be the influence of Confucian views ofteaching and learning on education (Marton & Tsui, 2004; Watkins & Biggs, 1996),which is part of the cultural heritage in the country, where students are expected toshow respect for the teacher and where it is less socially acceptable for students tocompete with the teacher in dominating the conversational floor, lest this be inter-preted as their being disrespectful. In such a context, however, the teacher’s“responsive questioning and feedback” skills in eliciting, probing, and extendingstudents’ thinking during the IRF cycles become all the more important in providinglinguistic scaffolding for students as they are guided towards successively higherlevels of cognitive processing.
Finally, the four types of feedback reported in this study were based on an inter-pretive analysis of data from the lessons of two teachers used as case studies. Thisapproach was used to understand the particular in depth. Thus, the findings arepresented as grounded hypotheses (Glaser & Strauss, 1967) rather than generalis-able findings across universal contexts.Conclusion
This study contributes to the characterization of teacher–student interactions, inparticular, during the follow-up or F-move of the IRF teaching exchange. Throughan analysis of the relationship between the interactive and cognitive aspects of bothteacher and student moves, some apparently enabling strategies related to teacherquestioning and feedback were identified. Interactionally, a teacher’s avoidance ofexplicit evaluation or put-downs, acknowledgement of students’ contributions,restatements of students’ responses, and, more importantly, her ability to posesubsequent questions that build on students’ earlier responses and that stimulate useof various cognitive processes, all appear to promote productive talk activity instudents at a level beyond mere recall.
Students can be stretched mentally through sensitive teacher-led but not teacher-dominated discourse. As orchestrators of classroom discourse in shaping students’learning, teachers need to position themselves as enablers of talk for thinking. One
1344C. Chin
way of doing this is to pay particular attention to the follow-up move of the perva-sive IRF exchanges in teacher–student talk, and consciously pose a series of mean-ingfully related questions that stimulate students to tap into higher-order thinkingprocesses.
Acknowledgements
This study was supported by the Centre for Research in Pedagogy and Practice atthe National Institute of Education, Singapore, under research grant CRP 12/03CHL. The author is grateful to the teachers and students who participated in thisstudy. Thanks also to the Editor and reviewers for their valuable comments on aearlier draft of this paper.References
Baird, J. R., & Northfield, J. R. (Eds.). (1992). Learning from the PEEL experience. Melbourne,
Australia: Monash University Printing.
Barnes, D., & Todd, F. (1977). Communication and learning in small groups. London, UK: Routledge.Barnes, D., & Todd, F. (1995). Communication and learning revisited: Making meaning through talk.
Portsmouth, NH: Boynton Cook.
Biggs, J. (2003). Teaching for quality learning at university. Buckingham, UK: The Society for
Research in Higher Education & Open University Press.
Biggs, J. B., & Collis, K. F. (1982). Evaluating the quality of learning: The SOLO taxonomy. New
York: Academic Press.
Black, P., & Harrison, C. (2001). Feedback in questioning and marking: The science teacher’s role
in formative assessment. School Science Review, 82(301), 55–61.
Bruner, J. S. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press.
Carlsen, W. S. (1991). Questioning in classrooms: A sociolinguistic perspective. Review of Educa-tional Research, 61, 157–178.
Cazden, C. (2001). Classroom discourse: The language of teaching and learning (2nd ed.).
Portsmouth, NH: Heinemann.
Chapin, S. H., O’Connor, C., & Anderson, N. C. (2003). Classroom discussions: Using math talk to
help students learn. Sausalito, CA: Math Solutions Publications.
Dillon, J. T. (1985). Using questions to foil discussion. Teaching and Teacher Education, 1,
109–121.
Dreyfus, A., Jungwirth, E., & Eliovitch, R. (1990). Applying the “cognitive conflict” strategy for
conceptual change—some implications, difficulties, and problems. Science Education, 74, 555–569.
Edwards, D., & Mercer, N. (1987). Common knowledge: The development of understanding in the
classroom. London, UK: Methuen.
Edwards, A. D., & Westgate, D. P. G. (1994). Investigating classroom talk. London, UK: Falmer Press.Elstgeest, J. (1985). The right question at the right time. In W. Harlen (Ed.), Primary science:
Taking the plunge (pp. 36–46). Oxford, UK: Heinemann.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative
research. New York: Aldine de Gruyter.
Hogan, K., Nastasi, B. K., & Pressley, M. (2000). Discourse patterns and collaborative
scientific reasoning in peer and teacher-guided discussions. Cognition and Instruction, 17(4),379–432.
Classroom Interaction in Science1345
Lemke, J. L. (1990). Talking science: Language, learning and values. Norwoord, NJ: Ablex.
Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake. Studies in Second Language
Acquisition, 19(1), 37–66.
Marton, F., & Tsui, A.B.M. (2004). Classroom discourse and the space of learning. Mahwah, NJ:
Lawrence Erlbaum Associates.
Mehan, H. (1979). Learning lessons. Cambridge, MA: Harvard University Press.
Morge, L. (2005). Teacher–pupils interaction: A study of hidden beliefs in conclusion phases.
International Journal of Science Education, 27(8), 935–956.
Mortimer, E., & Scott, P. (2000). Analysing discourse in the science classroom. In R. Millar, J.
Leach, & J. Osborne (Eds.), Improving science education: The contribution of research.Buckingham, UK: Open University Press.
Mortimer, E. F. (1998). Multivoicedness and univocality in classroom discourse: An example
from theory of matter. International Journal of Science Education, 20(1), 67–82.
Mortimer, E. F., & Machado, A. H. (2000). Anomalies and conflicts in classroom discourse.
Science Education, 84, 429–444.
Mortimer, E. F., & Scott, P. H. (2003). Meaning making in secondary science classrooms. Maidenhead,
UK: Open University Press.
Newman, D., Griffin, P., & Cole, M. (1989). The construction zone: Working for cognitive change in
school. Cambridge, UK: Cambridge University Press.
Roth, W.-M. (1996). Teacher questioning in an open-inquiry learning environment:
Interactions of context, content, and student responses. Journal of Research in Science Teach-ing, 33(7), 709–736.
Rowe, M. B. (1986). Wait time: Slowing down may be a way of speeding up! Journal of Teacher
Education, 37(1), 43–50.
Schon, D. A. (1983). The reflective practitioner. New York: Basic Books.
Scott, P. (1998). Teacher talk and meaning making in science classrooms: A Vygotskian analysis
and review. Studies in Science Education, 32, 45–80.
Sinclair, J., & Coulthard, M. (1975). Towards an analysis of discourse. London, UK: Oxford University
Press.
Tobin, K. (1987). The role of wait time in higher cognitive level learning. Review of Educational
Research, 57, 69–95.
Tsui, A. B. M., Marton, F., Mok, I. A. C., & Ng, D. F. P. (2004). Questions and the space of
learning. In F. Marton & A. B. M. Tsui (Eds.), Classroom discourse and the space of learning (pp.113–137). Mahwah, NJ: Lawrence Erlbaum Associates.
van Zee, E. H., Iwasyk, M., Kurose, A., Simpson, D., & Wild, J. (2001). Student and teacher
questioning during conversations about science. Journal of Research in Science Teaching, 38(2),159–190.
van Zee, E. H., & Minstrell, J. (1997a). Reflective discourse: Developing shared understandings in
a physics classroom. International Journal of Science Education, 19(2), 209–228.
van Zee, E. H., & Minstrell, J. (1997b). Using questioning to guide student thinking. The Journal
of the Learning Sciences, 6, 229–271.
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge,
MA: Harvard University Press.
Watkins, D. A., & Biggs, J. B. (Eds.). (1996). The Chinese learner: Cultural, psychological, and
contextual influences. Hong Kong: Comparative Education Research Centre and AustralianCouncil for Educational Research.
Wellington, J., & Osborne, J. (2001). Talk of the classroom: Language interactions between teach-ers and pupils. In J. Wellington & J. Osborne (Eds.), Language and literacy in science education(pp. 24–40). Buckingham, UK: Open University Press.
Wells. G. (1986). The meaning makers: Children learning language and using language to learn.
London, UK: Hodder and Stoughton.
1346C. Chin
Wells, G. (1993). Reevaluating the IRF sequence: A proposal for the articulation of theories of
activity and discourse for the analysis of teaching and learning in the classroom. Linguistics andEducation, 5, 1–37.
Westgate, D., & Hughes, M. (1997). Identifying “quality” in classroom talk: An enduring research
task. Language and Education, 11(2), 125–139.
Wood, D. J., Bruner, J. S., & Ross, G. (1976). The role of tutoring in problem solving. Journal of
Psychology and Psychiatry, 17, 89–100.
Zohar, A., & Aharon-Kravertsky, S. (2005). Exploring the effects of cognitive conflict and direct
teaching for students of different academic levels. Journal of Research in Science Teaching, 42(7),829–855.
因篇幅问题不能全部显示，请点此查看更多更全内容