Networks: An On-line Journal for Teacher Research

Student-Generated Discussion in the Senior Secondary English Classroom

by A.W. Lehmann
A. W. Lehmann is a teacher in School District No. 82, Terrace, BC, Canada. Correspondence concerning this article should be addressed to A. W. Lehmann, English Department, Caledonia Senior Secondary School, 3605 Munroe Street, Terrace, British Columbia, Canada, V8G 3C4.

The purpose of this research study was to devise and test a method of encouraging, and subsequently managing, student-generated discussion of English literature within a senior secondary classroom. The students would provide not only the discussion itself, but also a "client's-eye" evaluation of the process. Accordingly, students were engaged in part of the initial clarification of the study's purposes and procedures, produced the bulk of the discussion which constituted the content for the method being examined, and provided a post-discussion evaluation which could be compared to earlier comments. A simple qualitative analysis of written comments provided by the students and of my own notes and reactions to the discussion allowed me to define more clearly some observations about the dynamics of discussion and to select some directions for further investigation. It would be premature to draw any definitive conclusions based on this study alone, but some of the observations are constructively suggestive.

Preliminary Considerations

Classroom discussions can be wonderfully valuable for constructing and sharing knowledge. Often, however, the origin of these discussions and their movement toward the success described above is haphazard and fortuitous. I hoped to be able to prompt students to generate discussion that accomplished the valuable goals mentioned, in addition to ameliorating two main problems that are common in classroom activity. The first is that in many classroom discussions three or four voices dominate the interchange of ideas, and at least as many students never participate at all. The second is that many discussions are focused upon issues or ideas that the instructor has defined rather than upon considerations and questions which the students might find more significant or personally meaningful. Thus, the method that I adopted was meant to obviate these problems by requiring at least the limited participation of each student at one or more levels of the discussion and by having the students set the questions for the discussion's foci.

There were several questions that I hoped the project would illuminate. First, I desired a clearer profile of student attitudes to discussion in an average class. Second, I wished to discover students' feelings about teacher evaluation of their discussions; that is, should student discussions be graded and if so, how? Third, I wished to supplement my observations of the success or failure of the method of discussion with those of the students. Do they like the method, and do they find it educationally productive? Do their comments as participants describe much the same features that a teacher might observe?

A limiting factor was the brevity of the actual research period. The learning that needs to be negotiated by any class trying a new procedure is one that cannot be mastered immediately. Consequently, the limited success of the method we tried is probably not clearly indicative of the successes or failures that use of such a method might obtain over a longer period. Nonetheless, some constructive information was obtained.

Several authors and researchers provided useful ideas for the direction of the project. Sara Allen's (1992) paper on student-sustained discussion explored the limits to allowing students to manage a discussion without teacher interference. Her method of non-evaluative analysis of her classroom discussions influenced some of my questions.

Douglas Barnes (1991) provided some interesting observations on the methods of and reasons for using small group discussions. Particularly useful were the chapters on learning in small groups (pp. 34-70) and the teacher's control of knowledge (pp. 108-134).

Neil Mercer (1995) makes extensive reference to Barnes. In some ways he recapitulates Barnes' observations, particularly in his support of Barnes' argument that there is a difference between discussion in a classroom and educationally appropriate discussion, the latter being characterized by "learners (a) sharing the same ideas about what is relevant to the discussion; and (b) having a joint conception of what is trying to be achieved by it" (p. 96).

Borich's (1992) comprehensive text includes a chapter on cooperative learning and the collaborative process. It examines in detail the rationale for cooperative learning strategies; discusses the roles of both student and teacher; and outlines methods for specifying the goal of such activities, structuring the student task, teaching and evaluating the process itself, monitoring group performance, and subsequent debriefing (pp. 320-333). These ideas were particularly useful in designing my project. As well, Borich's explanations about the appropriate times for direct instruction supported my own decision about using direct instruction in part of the project's method (p. 187).

Two other texts, Richmond and McCroskey's (1992) Power in the Classroom, and Stewart's (1985) How to Involve the Student in Classroom Decision Making provided insights into problems that might accrue due to disciplinary anomalies. Morgan and Saxton (1994) provided valuable comments on the nature of questioning, whether by instructor or by student. Although I didn't use their specific categories, their text was helpful in clarifying my own approach, which follows.


Before beginning this project, I followed the consent/assent procedures that are standard protocol within the school district (School District #82 has in policy a comprehensive procedure to deal with any research efforts in the district that involve children in any way, to ensure their privacy and safety). University ethics approval procedures were followed, as well. These completed, I turned to the project itself.

I first "floated" the idea of doing research in the classroom with my twelfth grade English class to see how they would feel about having their work made the subject of research. They were a normally inquisitive group of youngsters with a fairly equal spread of ability, ambition, gender, and so forth. They seemed interested in the idea, and when I submitted the information letter and the consent/assent forms for them to examine and sign or have signed, nearly every one of them immediately signed and returned the assent forms. Within about a week and a half we were prepared to proceed. Those students who failed to obtain permission to participate went to the library to finish reading the novel which was subject of the discussion, A Prayer for Owen Meany by John Irving, under the supervision of the librarian.

My first step was to try to obtain an informal overview of student attitudes toward classroom discussion in general, based on their previous experience. I asked them to compose brief responses, three or four sentences long, to the following clusters of questions. (a) How would you characterize yourself as a participant in classroom discussions: As an eager participant; as an occasional participant; as one who prefers to listen and observe; or as one who commonly "tunes out," for whatever reason? (b) Is classroom discussion, in your experience, a good place to develop skills in listening and speaking, and to develop critical tolerance toward others' points of view? Why or why not? (c) If these learning outcomes are important, as the Ministry of Education believes, should student participation in discussions be graded, and if so, how? (d) Is classroom discussion an effective means for taking responsibility for your own learning, that is, for generating questions and seeking out their answers? Why or why not? Students were asked to identify themselves with pseudonyms or to leave the papers unsigned altogether, in the expectation that anonymity would encourage honesty in their responses.

I was moderately surprised at the energy with which the students tackled these questions. Time that I had scheduled for this phase of the project had to be extended so that students could get down all their ideas, and I suspect some of them had more to say but were directed forward onto the next question before they could completely explore their own thinking. Nonetheless, their responses presented an interesting spectrum of student experience, including personal descriptions, confessions and opinions.

I created a summary form loosely based on ideas presented by Miles and Huberman (1994) for compiling the student responses to these questions along with some of their more germane quotes. I read all of the student submissions and transferred, usually by quotation but occasionally by paraphrase, their attitudes and ideas to the summary form. One weakness of this method is that I served as editor of their thoughts, only pulling out their more insightful or well-stated contributions to be included. It may be that something important was omitted, although I went through their responses more than once in order to be reasonably rigorous. As well, had I had the time to interview each of the students orally, I could have asked for clarification of some of the ideas they presented. Some statements were either contradictory or so poorly stated grammatically that they were unusable. I didn't like leaving too much out, but some of the data either repeated things that had previously been said or were so general or clichéd as to be relatively valueless for the purposes of the summary.

I randomly assigned students to numbered small groups. The groups were numbered one through six to match the chapter numbers of our novel with which I wanted each group to deal. Thus group one would work on the first chapter, group two on the second chapter, and so on.

As part of preparation for their discussions, I provided the class with some brief direct instruction on the nature of questions and questioning. I explained to them three categories of questions: literal questions, inferential questions, and elaborative questions. Literal questions are those that can be answered by simple reference to the text. For example, how did Johnny Wheelwright's mother die? An inferential question requires both information from the text and knowledge or speculation on the part of the person seeking an answer in order to deal with it satisfactorily. An example, based on clues from the story, is what might some reasons be for Owen's attitudes toward his parents? Here students might provide ideas from their own family relationships and relate them to evidence available in the text. An elaborative question is one that requires synthesis or evaluation to address it fully. It, too, should have some referents in the text to tie it to the rest of the discussion. A suitable example is, did Dan exhibit the characteristics of a good father? Why or why not? (The three samples above are based upon A Prayer for Owen Meany.) I provided them with examples of those distinctions so that, within the context of the exercise, they could explore several dimensions of questioning, and also so that they could be directed toward creating more complex questions for discussion purposes. I wanted to avoid a discussion based solely or even primarily on literal level, search-type questions, which tend to simplify and deaden discussion rather than to encourage it.

Each student was given the same assignment, which was to compose a question from each level of complexity, based on the chapter his or her group had been assigned. A few students asked brief questions of clarification regarding the levels of question described above, but within a few minutes all seemed to understand. I allowed them a full weekend to read any necessary material (some students were still behind in the reading) and to compose their questions.

On the following Monday, we divided into our groups by chapter. Each small group (the largest had four students, the smallest two) was instructed to select a spokesperson by consensus. Then the groups were to share their questions, one level at a time (literal level first, then inferential, then elaborative). At this stage we ran into a small but annoying difficulty. Only about half the class had actually prepared their questions. Thus, those reliable students who were prepared had to wait patiently while their less reliable group partners hastily cobbled together something to contribute to the day's activity. (This particular hazard of group work, the differential reliability of students, is a common detraction from the success of group activities at the secondary level.) After fifteen minutes or so we were once again under way. Students shared and discussed their questions well, appearing to be on topic and on task, and requiring little in the way of instructor intervention to get the job done. I circulated among the groups, noting the progress of their discussion, which lasted twenty to thirty minutes. At the end of their discussions each group was instructed to select, by consensus, one question from either the inferential or elaborative level to present to the full class group discussion to come. They spent ten minutes or so debating the relative merits of the questions they had discussed and selected their questions for the full class discussion by the end of the period.

The following day we gathered as a large group and began sharing the questions. The discussion went well, with rarely a lapse of energy. Some of the questions were more challenging or appealing to the group than others, and the time devoted to them reflected this. Others were a little obscure and "fizzled out" in the large-group discussion.

During the last twenty minutes of the class we did a post-discussion evaluation, or debriefing, as Borich (1992) would have termed it (p. 331). Once again, this was done through soliciting written answers from the students to specific questions about the discussion as well as the process we used to approach it and carry it out. They were as follows: (a) Do you think a two-level discussion (first small group, then larger group) encourages broader participation by students in the discussion? (b) Are student-generated questions superior or inferior to questions generated by your instructor or that emerge by chance? Explain. (c) Was it useful to you to have a knowledge of various levels of questioning? (d) How would you evaluate the successes or failures of the discussion method we used? (e) If we could eliminate the bureaucratic impedimenta (permission slips, etc.) would you like to use this method again? Why or why not?

The participating students readily addressed these questions, providing on average a page to a page and a half of commentary in response. I analyzed their responses in much the same way as I did the pre-discussion responses. I devised a summary form onto which I mapped student responses into categories. Again, I included quotations or paraphrases which seemed most to characterize or explain the reasons behind their attitudes.


Pre-Discussion Attitudes

The pre-discussion background research led to some useful anecdotal information. The class divided itself almost evenly among three categories of participation in discussion: eager participants, occasional participants and listener/observers. Only one student admitted to habitually tuning out, stating that "wonderland is a good place to be if the topic's boring." Eager participants made such statements as, "I love to argue." "I actively participate and then sit back . . . to get a feel of how people react to my opinions." "It's exciting." Even one student who was a self-described avid participant admitted, "I try to zone out . . . it never lasts very long." Apparently the lure of discussion was far stronger to this student than the appeal of day-dreaming.

Others who described themselves as occasional participants seemed, in some ways, more thoughtful. One stated, "I like to feel I have something really relevant to say before I say anything." Another commented, "I prefer to think out my arguments [so as not] to look stupid." Several would speak based on the immediacy and power of their emotions. "If [my] opinion is challenged I will speak up and defend it." "[I participate] when I have something really important to share." Two students wanted to defend personal identity and beliefs: "I try to be my own voice [rather than] go with what the majority thinks." "I'll speak out quite loudly . . . when it has something to do with my faith."

Listener/observers had a third cluster of perspectives. Some were simply shy. "I don't like the attention I bring myself." "I don't want people to judge me for something I said in class." Others claimed uncertainty or lack of knowledge. "Usually I am quite unsure of myself." "[Sometimes] I don't fully understand what is being discussed." Another group remained quiet because they felt other, more aggressive, students would have covered the topic adequately by the time they felt like contributing. "The issue is usually burnt out before I voice my opinion."

A sizable majority of students, whatever their own proclivities toward contributing to a discussion, felt that the classroom is a good place to learn listening and speaking skills, and critical tolerance toward others' ideas. There were many caveats, though. One observed that classroom discussion works because it is "mostly controlled." Another felt that discussions are excellent, but only useful if the class is "mature enough and tolerant enough to allow this to happen." Discussions could, in this student's view, be quite destructive when they became personal, "cruel," with "insults." A student who professed to love discussion also confessed that "after a discussion I'm often mad, unable to listen, and during and after the discussion I can be quite mean," which rather reinforces the previous student's comments. Some students complained about others' lack of attention. One stated, "I don't think teachers push hard enough to get those few involved and they fall behind." And one blunt student averred that "some people just don't have the know-how to say what they think." Many students, though, recognized that discussions are an excellent forum in which to "widen your understanding of [a] topic" or to be "exposed to many ideas," to get "less nervous," to "develop confidence in speaking and get better at taking 'mental notes.'" One claimed, "That is the way I learn things best" (giving support to theories about preferential learning styles), and another went so far as to make the claim that "in a Utopian classroom all class would be is discussions."

The question about grading discussions generated many interesting and somewhat contradictory ideas. Those in favour of grading them outnumbered those opposed by about three to two, but there were very few constructive ideas about how their contributions to discussion should be graded. There were some astute comments on the problem. As one person put it, "Some [students] are active, some are not; some make themselves heard a lot when they don't have much to say," an observation that shows an awareness not only of the quantity of someone's participation, but also of the quality. Another very clearly pointed out that inopportune timing "makes late speakers' contributions less 'valuable,' thus penalizing shy students," who presumably might speak up later when they got their courage up. One student noted that students are more likely to "speak their minds when they're not worried about a grade," suggesting that some students feel that their opinions might be vulnerable to a low grade due to a teacher who holds a counter-opinion, or that the process of public evaluation inhibits their expressive ability due to a kind of performance anxiety.

Many other unanswered questions about grading discussion activity were raised, either as direct questions or as pithy observations. "How can you put weight on a particular aspect of a discussion [as opposed to another aspect]?" "People [who are] above . . . the activity . . . should be penalized for it." Presumably losing the grade would be the penalty for not participating. "Teacher input would be much more useful [than a number]." Apparently this student would prefer verbal feedback. "It's hard to put a mark on a statement." "A mouthy person should not be marked on the same level as a quiet person." One student made a useful connection between discussion and written activities, suggesting that we "grade essays on the same topic after discussion, or look at notes taken during the discussion," which is quite insightful and pragmatic.

Addressing the problem of personal responsibility for learning and its relation to discussions is a considerably more difficult question. One student expressed the idea that in discussion "your mind doesn't have a chance to wonder [wander?]. This almost forces you to have self-responsibility of your own learning." Another felt it is "good for finding personal answers." A third made the connection that shy students fail to participate, which in turn leads them to devalue the activity because the greatest benefits accrue to those who do participate. "We can ask questions about what we don't understand," said one, indicating a willingness for self-direction. Another acerbically remarked, "The best way to gain knowledge is to shut your mouth and listen, [and] ask questions when you don't understand."

These students showed an alert sensitivity to the problems inherent in classroom discussion. They were able to describe vividly their own strengths and weaknesses as well as to put themselves imaginatively into others' positions, for example, that of another student who might be shy or aggressive, or that of the instructor trying to evaluate the whole process. They also demonstrated strong approval of discussions as learning tools, even if not all students wanted to be actively involved.

Post-Discussion Research

One aim of this research was to evaluate the method we used in our discussion of a novel. Was the division of the discussion into two levels helpful? Was the instruction on levels of questioning useful? How effective was the student-generation of questions? These considerations were to lead to decisions whether or not to try this method again, or to modify it in some ways and then try again.

In answer to the first question regarding whether or not the two-tier discussion process was effective in broadening the discussion, the students were generally of agreement that it was, though three respondents disagreed. One felt that the small groups "pressured [shy students] into stating an opinion; later in the big group their opinion was more likely to come up [than if they had remained silent]." Students felt that in the small group they did not "have to worry about being judged," because "in the small group you formed similar ideas . . . in the large group . . . you felt . . . the small group supported your ideas. It was not like you were alone." Group solidarity can lend confidence. "It gives a chance to organize thoughts," "you can compare and contrast more easily with [the] smaller group before taking the idea to the larger group," and it probably allows "for more involvement." A more avid participant observed that "a couple of shy people in my small group were actively participating." Nonetheless, of the three who felt there was not any real difference between this method and other discussions, one felt that only "the group that made up the question" really benefited, and another felt that often the student "didn't really have [his] own voice."

On the value of student-generated questions, students were fairly evenly divided. Many felt such an approach clearly superior. Students could "focus on the things that interest us" and could "ask about things we did not understand." "Student questioning gives the students a feeling of importance and acknowledgment." However, a considerable number felt student questions were worse, especially if "major points are missed." "Usually the instructor has a firmer grasp on the material in discussion and can therefore focus the questions on the more important aspects." "Sometimes the teacher seems to ask a question on the more difficult concept which most people need to discuss to understand it better." A few felt there was little difference between student-generated and teacher-generated questions, and questions that emerge organically or by hazard from the story were admired by a few. "Chance allows a discussion to flow. If students are thorough, [missing important points] shouldn't be a problem." "Questions due to chance are probably the best. Some teachers feel that they [must] have specific answers and some students cannot make up good questions." The final word by one was, "A question is a question." Does it matter whose question it is if it is a good one, one that generates reflective discussion?

There was a mixed evaluation of the direct instruction on questioning strategies (the division of questions into literal, inferential, and elaborative levels). Several suggested that it was not important. "I don't think it made much of a difference." "My life hasn't been changed too much by the new knowledge." "I don't think you need to have a knowledge of questions." One held the middle ground. "It is useful . . . in a limited way . . . [it] allows one to organize thoughts and questions for writing in a more orderly fashion . . . beyond this [it is] little more than trivia." Positive responses included, "I learned something new." "The knowledge of these levels will affect the end result in the long run." "It expands your knowledge . . . forcing you to really think about the chapter and its connections to the rest of the world."

There was little consensus as to the success of the exercise. Some felt it was very useful; others had serious reservations. On the positive side, students had a chance "to see what things interest others . . . and maybe have things pointed out to us that we might otherwise have missed." It was "an excellent example of two heads being better than one." "Getting in groups was good." In a more ambiguous response, one student rated the process "a 6 out of 10." "Some people weren't done the book," which limited the participation. "It could have been better. Many people . . . weren't really interested in the questions being discussed." "It took two classes and we were still rushing to get through it all, so the efficiency can be questioned. The student questions were not very effective." And as usual, there were complaints about those few "who don't know when to stop or be quiet," or who "ramble incoherently."

Most students wanted to try again, and they provided some interesting suggestions. One was "to post the questions a day or two before the discussion." Another, from a student who likes discussions to be "fast and fight-like," was a request for a little more freedom. "Do the questions have to be about the novel?" asked one. Several pointed out that the bureaucratic problems to do with consent shrank our class, and recommended trying again with "more people." Only one student said, "No. This method just won't work for me, anyway."


It must be noted that a few students chose not to provide assent or neglected to get parental consent for this project; this factor probably skewed my classroom sample to include a greater proportion of students who are actually interested in English discussion or in academic pursuits in general. Nonetheless, for the majority who were involved, the following findings are suggestive.

The purpose of this project was to get a clear profile of student attitudes to discussion in a twelfth grade English class. This was accomplished. Its results were heartening, although they may have been somewhat sanitized by student efforts to please their teacher or to prevent getting a low reputation that might eventually affect their marks. (I made it clear that this project would have nothing to do with the calculation of their English 12 grade, but students are occasionally somewhat skeptical of authority and some may have decided to "play it safe.") I consider the results heartening because of the relatively high level of attention to discussion that was reported by students. There was only one student who confessed to enjoying "tuning out," and even that was in response to topics that are "boring." There may have been a few more such students, but they were not evident in the responses submitted, for whatever reasons.

The second consideration of importance was whether and how to evaluate this activity. The dominant ethos in senior secondary tends toward the evaluation of everything. Anything done for its own sake is usually extra-curricular. It is very common for students to ask, "Do we get marks for this?" in a tone of voice as if to suggest, "If we don't, why should we bother?" And indeed, unless the fear of poor marks threatens, a significant number of students will not participate. Further, as we well know, even this does not motivate a considerable minority of students. As a teacher somewhat habituated to this state of affairs, I would likely try to connect this activity in future to some form of evaluation, and some of the student suggestions mentioned above are good ideas. The students raised some good objections to grading, though, and evaluation of this kind of activity often seems either "fuzzily" inexact or somewhat pointless. It is difficult enough to distinguish reliably between A and B, let alone between 73% and 74%. If the instructor spends all his/her time filling out some kind of checklist monitoring who was talking, how many times, how forcefully, how many comments were questions, how many were answers, and so on, what kind of contribution is he/she likely to make to actually encouraging the discussion through immediately constructive feedback? Perhaps one student comment is most pragmatic. He claimed that we should not mark discussion directly because participation, whether as a speaker or a listener, would lead to better grades anyway, indirectly. Students who "tuned out" would not learn, and their work that was graded (essays and so on) would suffer accordingly. Those who participated would improve their thinking and, by extension, the content and style of their written work.

Finally, did the students find the exercise a successful one, that they would readily repeat? Many students, especially shy ones, found the two-level discussion liberating, and that it enhanced their participation. Less important to them was the theoretical nature of questioning. Rather they preferred to attend to questions that grew organically from their discussions, to focus on their content and meaning as opposed to their level of abstraction. As for the source of the questions, it appears that students were engaged by the process of questioning, but that they also were attuned to the importance of instructor expertise.

Did student perceptions mirror my own? I, like several of my student critics, found some of the questioning to be weak. I, too, was irked by one or two students who had little of significance to say but who "rambled incoherently" before I could tactfully grant someone else the floor. I agree with the youngster who felt that classroom discussion works best when it is controlled to prevent cruelty and insults, and although some increased freedoms from one-speaker-at-a-time decorum might result in greater entertainment value, I think that we should not aspire to model the classroom on Jerry Springer. I do not think that "fast" discussions characterized by "fights" are educationally defensible.

Further Research

Like most of the student respondents, I would like to try this method again. Further useful observations could be made toward refinements that would improve discussion considerably and make it a more predictable process than simply capitalizing on what used to be labeled "the teachable moment."

The comparatively high level of student attention drawn to discussion ,as shown by the student responses in this project, is fascinating in that it is almost surely higher than that given to lecture-style presentations by the instructor or to most presentations by their fellow students. There is something in the spontaneous give-and-take of discussion, an element of unpredictability and challenge, that is absent in the other forms of discourse cited above, and this feature might account for the higher attention. If this is true, then any methods that we can utilize to energize and foster productive discussion in our classrooms should be pursued.

I was particularly struck by the following final message by one of the students.

I really liked this experiment, not quite chemistry, but very fun. I really like the fact that here we get to have a say in the way that we are learning, we get more emotionally involved and therefore we can take it more seriously. It's not just a marks thing, it's us getting to be us and learning through the best way we see fit.

With such a positive endorsement, it would be difficult not to pursue this method further.


  1. Allen, S. (1992). Student-sustained discussion: When students talk and the teacher listens. In N. A. Branscombe, D. Goswami, & J. Schwartz (Eds.), Students teaching: Teachers learning (pp. 81-95). Portsmouth, NH: Boynton/Cook.
  2. Barnes, D. (1991). From communication to curriculum (2nd ed.). Portsmouth, NH: Boynton/Cook.
  3. Borich, G. (1992). Effective teaching methods, (2nd ed.). Toronto: Maxwell MacMillan Canada.
  4. Irving, J. (1989). A Prayer for Owen Meany. New York: Random House.
  5. Mercer, N. (1995). The guided construction of knowledge: Talk amongst teachers and learners. Philadelphia, PA: Multilingual Matters.
  6. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.
  7. Morgan, N., & Saxton, J. (1994). Asking better questions: Models, techniques and classroom activities for engaging students in learning. Markham, ON: Pembroke.
  8. Richmond, V. P., & McCroskey, J. C. (1992). Power in the classroom: Communication, control and concern. Hillsdale, NJ: Lawrence Erlbaum.
  9. Stewart, W. J. (Ed.) (1985). How to involve the student in classroom decision making. Saratoga, CA: R & E Publishers.