Should we call on non-volunteering students?

Kim Kastens
Author Profile
published Nov 29, 2010
Before the beginning of this semester's "Frontiers of Science" course, the folks in charge of Columbia's Core Curriculum put on a one-day professional development session for instructors new to the course. Among the suggestions for leading a successful seminar was that we should call on non-volunteering students.

This suggestion took me aback. It did not align with my prior practice teaching either undergraduates or graduate students at Columbia. And it certainly did not align with my own experience as a university student. Other than in introductory German class, I cannot recall a single instance in any class I took as an undergraduate at Yale or a graduate student at Scripps Institution of Oceanography, where a professor called on a student who had not telegraphed his or her readiness to answer by raising a hand. The aura seemed to be that it would be ungentlemanly to expose a non-volunteering student's ignorance publicly. (And I do mean "ungentlemanly"; as a science major, I had only two female professors in nine years of university courses.) And finally, my own college-age daughter, who attends a college where almost all courses are taught in seminar format, provided the following advice when I mentioned my new teaching assignment: "Whatever you do, don't call on students who don't raise their hands; that is just totally YYYUCKKK."

However, our workshop leader made the point that students learn through participating actively in discourse, through articulating and defending their ideas, much more so than by listening passively while others talk. This way of looking at the issue resonated with an insight from my own research.

Maps of blue flag data My earliest research-on-learning project explored how children use maps. We put up eight colored flags around the Lamont campus, gave paper maps of the campus to 4th graders, and then asked the kids to put colored stickers on the map to show where they thought each flag was located: "....put the red sticker on the map to show where you think the red flag is located, and put the yellow sticker on the map to show where you think the yellow flag is located....and so on."

Over the years, we used various versions of this task. In one variant, we asked students to write down, after they placed each sticker, what clues they had used to decide where to place that sticker. Our goal was to obtain a glimpse into the reasoning of the students who made seemingly inexplicable sticker placements, such as placing the orange sticker in the middle of the lawn on the map, when the orange flag was in fact on a building. What we found was that the Explaining students' answers clustered more tightly around the correct answers, and the worst sort of errors disappeared almost entirely.

It seems that we had tapped into a phenomenon called the "self-explanation effect." Initially described by M.T.H.Chi in the 1980's, the idea is that when a person is required to generate and articulate an explanation for something, understanding and learning of that content are enhanced. The term "self-explanation" means that the learner generated his or her own explanation. Since Chi's pioneering research, other researchers have documented the self-explanation effect in a wide range of learning and problem-solving contexts, from reading a biology textbook, to designing an experiment, to solving physics problems. The table below summarizes some of these studies:

Table of Self-explanation effect

The apparent power of this "self-explanation effect" is stunning. In education research, researchers have been known to get excited by an intervention that results in a 10% performance improvement. But in these studies, the group who generated and articulated explanations are often seen to be doing twice as well on average as the group who did not self-explain. If you've ever found yourself saying or thinking "It wasn't until I had to teach X that I really came to understand it," you may have been experiencing the self-explanation effect.

popsicle stick with namesarrangement by Dana Chayes
With this background, I interpreted our workshop leader's encouragement to call on non-volunteering students as encouragement to set up opportunities for the self-explanation effect to flourish. I've been trying this for half a semester now, calling on random non-volunteering students. I'm proceeding cautiously, concerned about unintended consequences in the affective domain. So far I'm only calling on non-volunteers when everyone had a chance to prepare in advance, for example when I have posted discussion questions or provided thinking time. I've also adopted a practice I first saw in a teacher education class taught by Professor Suzanne Reynolds at St. Thomas Aquinas College: putting students' names on popsicle sticks and pulling out a name from the pile, so it is clear that I am truly selecting at random and not picking on individuals vindictively.

It seems to be working--they all have had something meaningful to say. And I don't feel the least bit ungentlemanly.


  • Chi, M. T. H., M. Bassok, M. Lewis, P. Reimann and R. Glaser (1989). "Self-explanations: how students study and use examples in learning to solve problems." Cogntive Science 13: 145-182.
  • Chi, M. T. H., N. DeLeeuw, M.-H. Chiu and C. LaVancher (1994). "Eliciting Self-explanations improves understanding." Cognitive Science 18: 439-477.
  • Gagne, R.M., and Smith, E.C., Jr, 1962, A study of the effects of verbalization on problem solving: Journal of Experimental Psychology, v. 63, p. 12-18.
  • Kastens, K. A., D. Kaplan and K. Christie-Blick (2001). "Development and evaluation of Where are We?map-skills software and curriculum." Journal of Geoscience Education49: 249-266.
  • Kastens, K. A. and L. S. Liben (2007). "Eliciting self-explanations improves children's performance on a field-based map skills task." Cognition & Instruction 25(1): 45-74.
  • Lin, X. D. and J. D. Lehman (1999). "Supporting learning of variable control in a computer-based biology environment: Effects of prompting college students to reflect on their own thinking." Journal of Research in Science Teaching 36: 837-858.

Googling "calling on non-volunteering students" brought me to a review paper, Encouraging reticent students' participation in classroom discussion, by William W. Wilen, in a journal about social studies education. This review quotes extensively from a paper that I couldn't get my hands on, but it has a great title: "Calling on Non-Volunteers: Democratic Imperative or Misguided Invasion?" by Tom Kelley.

Should we call on non-volunteering students? --Discussion  

Good topic Kim. On field classes I get discussion going after some data collection by selecting the first person on the basis of birthday or something random, and then allow that person to choose the next one, who can choose the next one and so on. I haven't quantified anything, but very few students miss out on contributing, and it is usually their "friend" who has picked them so the "blame" does not attach to me. Of course, anyone struggling, I give them some helpful hints and encourage the other students to do the same. I haven't tried it indoors.

Alan Boyle
Liverpool, UK.


Share edittextuser=900 post_id=13571 initial_post_id=0 thread_id=3981

This very question has been part of science teaching methods courses since I was first a student teacher, then as a high school science teacher, and then later with both undergraduate and graduate audiences in both content and pedagogy courses. There are a few elements that make it an effective practice, at least for me, each of which deal with student expectations of my behaviors.

First of all, I am up-front with students on Day 1 of class, letting them know that part of my job is to make them slightly uncomfortable, and part of the art of teaching is to constantly adjust what Edward Deci called "optimal incongruity," which provides a motivational boost to learning. On reflection, most people would agree that those times that they really learned something, and felt most proud of that learning, were at those times when they were bothered that they did not know, and were internally pushed to resolve this "bother."

Second, it is necessary to be consistent in your questioning behavior, removing doubt from students' minds on how to respond to instructor questions. They need to have some idea of the local rules - what actions on their part are rewarded, what actions are not rewarded or sanctioned. If the rules are known, then that element of anxiety is removed from the equation. Again, I tell students how I will ask questions, and then stick to it. I do accept volunteered answers, but everyone in the room has the prospect of being called upon, and they know it.

Third, the use of "wait time" in questioning is vital. Waiting 3-5 seconds after asking a question before calling on a student, volunteer or non-volunteer, gives every student the opportunity to think about a response. Not every student can think of a response at the same time, so those students that are not "instant responders" have as much a chance to formulate an answer as those that are not. Corollary to this technique is asking the question first, and then calling on a student. In this manner, students that are not called on do not have the opportunity to "tune out" of a question. Everyone has equal accountability, but also equal opportunity to form a response.

When I call on I a student, I try to maneuver myself in such a manner as to put as many students between myself and the called upon student. My dialogue is with the called upon student, but every other student is at least an implicit member of the dialogue. Wherever possible, I call on that student by name, expressing value for their responses that they can tie back to their self-perceptions.

Finally, feedback is important, but of a varied form. First of all, positive feedback (without being "gushy") for correct responses is good, but "negative" feedback for an incorrect response is suppressed in favor of "sustained" feedback by rephrasing the question, or breaking it down into more discrete elements in succession, scaffolding the question in a manner that allows for talking out a thinking process. So even if a student doesn't believe that they know the answer to the specific question you have just asked, there is an element of trust that they can be guided to a correct response. "Failure" is not an option, and "I don't know" is not an acceptable answer.

I have been teaching since 1986, in both high school as well as higher education, with introductory as well as advanced classes, and I have had my share of negative feedback on course evaluations. Most of the time, these comments are reflections of practices that I have been concerned about myself, and continued to work on. That said, not a single negative comment has been on my questioning techniques and practices. Quite the contrary, students have frequently remarked that I actually cared that they learned the content presented to them, and one inference is that this "caring" is expressed through this use of instructional dialogue.


Share edittextuser=1128 post_id=13572 initial_post_id=0 thread_id=3981

HI Kim, Years ago I helped institute an Honors Seminar for incoming freshmen called "Texts and Critics" in which we read great books and discussed their contributions to our understanding of knowledge, love and power. We commonly encountered topics that made students feel very uncomfortable--in part due to very dense texts, but also because the readings touched on personal values and ethics (and required students to reflect and confront these value systems). To encourage participation by quiet students, we always started the bi-weekly discussions with an "opening question". I often asked a shy student to help prepare the opening question for the next class, and to come with textual citations ready to go so that they would be talking from a position of authority for at least the first few minutes. This was usually a very good ice breaker, and did not put the students in a position of risk by spontaneously asking them a question (often seemingly out of context). Once the conversation started, there was usually no holding back. But, this was an effective way to get quiet students to make a statement in the first instance, and to help instil a sense of confidence in their scholarship.

Many of our opening questions were purposefully ambiguous and oblique to the central theme: In Plato's Republic, one could ask "Why did Thrasymachus blush?"...and this opened the door for students to recreate the arguments made by Socrates, to look deeper into the text for answers, and to discover for themselves the "trap" that was laid that led to the internal inconsistencies in Thrasymachus' arguments. Helping students along the way to discovery can help motivate students and also helps build self-confidence.

I would also note that we should be aware of "active listeners"...It is often the case that students are thinking deeply (and quietly) about a topic...They are indeed actively engaged, just not verbalizing. I think we should be able to honor students who are indeed working through difficult problems, and not succumb necessarily to rewarding those who make a big splash but with little substance.

I think there may also be gender issues here: this may be painting with too broad a brush, but my general observations are that many female students tend to think more holistically and want to more completely understand contexts and connections--and thus appear to be shy or quiet when they are actively processing on many levels. On the other hand, many of my male students tend to "shoot from the hip"--making assertions and giving a false sense of authority of a given subject, even when that assurance is not warranted. Belenky et al. have some insights in their work on Women's Ways of Knowing...


Share edittextuser=7 post_id=13573 initial_post_id=0 thread_id=3981

Hi Kim,
This is a great post, and great responses from Alan, Eric, and Dave. In my introductory geoscience course, I get a lot of upper-level engineering majors (mechanical, civil, environmental) and other seniors in a variety of majors. At the beginning of the quarter, I tell everyone that they are bringing their own expertise to the classroom, and that I will rely on them to share that expertise, because everyone can benefit from a diversity of perspectives. I try hard to learn their majors, and will call on a mechanical engineer when we are talking about elastic properties of the crust, for example. Initially, some students are uncomfortable about being called an "expert" in their major, but it becomes clear to them that I respect what they add to the discussion. Then I'll ask other students to react and add their own perspective - maybe another mechanical engineer, or maybe a computer science major.

I hope this helps students see that class is meant to be a discussion; that deep learning can happen when you have to make connections in real time to things you already know. I don't hesitate to facilitate that discussion by calling on people who haven't raised their hands.


Share edittextuser=888 post_id=13574 initial_post_id=0 thread_id=3981

imo - An absolute key, no matter class size, is to do what Eric says -- discuss with students why you are doing what you do. With that said:

It seems to me that the situation is different in large classes compared with small ones. Many of the suggestions above work very well for small classes. (Define "small" as you wish -- I mean it to refer to a class where you are able to connect personally with all students. Depending on the instructor/students/class, this limit could be 6 students or 60; probably not 600.)

For large classes (probably for small ones, too), I think it helps to scaffold a bit. Start by asking questions and letting students discuss in small groups before choosing a group to answer. Let the alpha person do the talking. Later, do the same but choose a specific individual in a group to answer to get others speaking. Then move on to having students do group discussions sometimes but not other times before you ask for volunteers or call on people. (I have never progressed to eliminating groups altogether.)

As I write this, another thought comes to mind: One thing that also helps is to use clickers or some other system when appropriate. A students then finds out that she/he are not the only one who thinks X. An interesting thing that I have noticed, and encouraged, is that if properly managed, clickers put the entire class on one team. Its sort of a game of "get the best of Dexter." When all or most people get the correct answer, the class wins. They like to win. (And, in my large class, I tell them that if they all got the correct answer to a question -- 100% -- I will buy them all pizza. Unfortunately (?), they did score 100% twice in the past year!)


Share edittextuser=112 post_id=13580 initial_post_id=0 thread_id=3981

I just discovered the Web site, so I'm joining the discussion late.

The interesting thing to me about this question is that it presupposes that a dominant interaction pattern in the classroom is the teachers asking a question of a specific student. In the classic classroom discourse, this is the Initiation-Response-Evaluation IRE pattern. The teacher asks, one student answers, the teacher evaluates. Typically (but not always) the question is at a low cognitive level, and the student's response tends to be short. This pattern is so universal in schools that children understand and adhere to the pattern within their first year of school. Watch kids "playing school' and this is what they do. Interestingly, the only other places in American social interaction where this pattern is common is when parents talk to tiny children, or when lawyers manage cross-examination.

We know from research in language acquisition that students need to use language to acquire both vocabulary and concepts. So maybe the question should really be - how do we promote opportunities for students to talk to each other about science? In IRE discourse there is limited opportunity for students to practice ideas out loud.

Other folks have mentioned letting students talk in small groups. There's a place for students to talk science in a low-risk environment. Think-Pair-Share and a whole host of dialogue protocols provide structures for having students talk to each other - even in big lectures. Students are pretty easily trained to finish their thought and be quiet when you raise your hand.

I think it's always to useful to ask yourself what the goal is in any discourse move in the classroom. As teachers, we tend to ask a lot of questions without considering why we're asking them. Some teachers tell me they do it to keep the class engaged - but there's no real evidence that anyone is engaged except the one student who is talking. Others say they do it to keep the class on their toes. I'm not sure the threat of humiliation is the best way to put your students in a responsive state for learning. I know I sometimes do it just from habit - because that's what teachers do. But it's a good idea for all of us to be more strategic about how we talk to students and how we structure things so students talk as well.


Share edittextuser=4841 post_id=14024 initial_post_id=0 thread_id=3981

Join the Discussion

Log in to reply