Is Recitation Reinforcing?
by Mary Alice White - 1974
The author describes four routes by which a teacher can find out if her pupils are learning: 1. By asking them questions in class; 2. By checking their homework; 3. By scoring their tests which she devises; 4. By computing their scores on standardized tests.
Teachers have been criticized repeatedly for the ways in which they ask pupils to recite. They have been told that the questions they ask are singularly dull, trivial, or simple; and that too infrequently they ask pupils questions that involve critical thinking or which require pupils to apply a principle. Teachers have been told they dominate classroom recitation too much, often as high as 80 percent of all classroom discourse. Educational critics have urged instead a more Socratic dialogue or more discussion. "Recitation," in fact, has become a term for mindless repetition of details.
This form of recitation is very common in a number of classroom studies and may have existed in American classrooms for the past hundred years, according to at least one study. The conclusions one reaches from conventional evaluation of this evidence of a widespread and long-standing practice are twofold: (1) many teachers teach badly much of the time, and (2) this behavior is persistent.
If this style of recitation is so bad (an unproved assumption, however), why does it persist? Why do teachers ask relatively simple questions, which often require only a discrete item of information for an answer? Why do they ask a lot of these questions? Why do they ask them of so many pupils? When actions are as persistent as this behavior, some would interpret the behavior in the light of E. L. Thorndike's theories and those of his followers—reinforcement theory— which states that behavior which is persistent and widespread is being reinforced. Let us assume that teachers are human beings and are subject to the same behavioral regularities demonstrated in other human beings.
To think of recitation in this light seems incredulous at first. Teachers are not pigeons or rats. No principal is standing inside a classroom cage dispersing M&M's to the teacher for each such question. What reinforcement? From whom?
In a typical classroom (if there are any left), there are two sets of human beings, both of whom respond, as do all normal human beings, to social reinforcement. Social reinforcement involves attention from another human, whether verbal responding or physical responding or even just facial gestures. William James said that each person wants above all to feel appreciated, which is an elegant way to describe social reinforcement.
It seems highly probable then that the two sets of human beings, the teacher and the pupils, are likely to seek social reinforcement from each other. (Much social reinforcement takes place among the pupils, and sometimes there is more of that going on than between teacher and pupil.) The teacher would like the pupils to think well of her (to appreciate her) and to respond to her positively. The pupils feel the same way. They want the teacher to like them, to approve of them, and to respond positively to them.
The teacher wants the reward of being a good teacher, however she may define it. One way most people define good teaching is in its outcome, good learning. If the pupils are learning, the teacher is teaching (which may not necessarily be true, but it is commonly assumed). There are at least four routes by which a teacher can find out if her pupils are learning:
1. By asking them questions in class
2. By checking their homework
3. By scoring their tests which she devises
4. By computing their scores on standardized tests.
Homework as Evaluation
Holding aside the first route for a minute, let us examine the other three. Checking homework is an immensely time-consuming procedure, to begin with. Those who have corrected homework can also testify that very frequently it is an unpleasant experience because of the errors which pupils make. It is common to hear teachers at all levels, including graduate school, complain about the number and type of errors in their students' papers, implying that they came as an unpleasant surprise. One explanation for the numerous errors may be that homework is a much larger sample of a pupil's progress. For example, David may answer two, possibly three, questions in mathematics class, and answer them correctly, so his teacher feels assured about his progress. In his homework that night, he may answer fifteen problems, and his teacher discovers to her horror that he has failed to master an operation that was taught last week. This is very discouraging to any teacher, and would be called a punishment in plain English, or an aversive experience in the language of reinforcement theory. The more intensively or extensively that pupils perform, or the more examples they do, the more errors that will show up. One could argue that more correct answers will also show up under these conditions, which is also true. But the teacher has assigned homework (presumably) for additional practice or for evaluation of progress, or both, and she approaches it with an error-oriented attitude: Where are the mistakes? What do 1 need to go over in class? How many need extra practice or individual attention? The errors are the guideline for the next teaching. (Perhaps they should not be, and the measurement toward criterion-referenced measurement may place more needed emphasis on what a pupil has done correctly.) So homework is one mirror of one's teaching, and it is too discouraging too often, too full of drudgery, and too time-consuming to make correcting papers an enjoyable experience for the teacher. Thus, most teachers do not find homework a rewarding way of finding out if their pupils are learning.
Teacher-made Tests as Evaluation
If the teacher takes the third route by making up her own tests to check her pupils' progress, she encounters two problems. First, a really good test takes a team of experts months to construct. Many studies have repeated the criticism that teacher-made tests ask for detailed facts only, and often in a very mechanical way. (That sounds like the same criticism that has been made of classroom recitation.) To write a question (1) that can be answered with an "X" mark out of several choices and (2) which demands reflection, analysis, or application of a complex idea or principle is an extremely difficult matter. If the reader finds this hard to understand, let him try to write only one multiple choice item (four choices, one of which shall be correct) which will require the pupil to evaluate a major cause of the Revolutionary War in terms of the heritage of self-government experiences of colonists in New England and the South. This type of item can be written successfully, but it takes considerable sophistication and training. It is not surprising that teachers write simple items.
The second problem with teacher-made tests as a form of evaluation stems from the first, in that the tests are constructed individually by the teacher for her own class, and will not give her any information about her class' progress compared to another class.
Standardized Tests as Evaluation
It is this comparison that she can get from the fourth route of standardized tests, for here she can find out how her class is doing compared to other classes in her building, system, region, and nation. These tests have problems too, particularly because they are used nationally and therefore reflect the communalities of classrooms.
Teachers tend to dislike everything about standardized tests, except perhaps high school teachers who often react positively to a Regents type of high school examination, which is a New York State mandated testing procedure in the content of most high school subjects. The differences in attitude are enlightening about reinforcement. Elementary school teachers dislike the lowas, the Stanfords, and the other widely used standardized group tests for many reasons. Often they have to administer them, which is both a dull and complicated procedure. Then they usually wait months before the scores are returned, so that the scores do not help them plan instruction. Often the original answer sheets are not returned, so teachers have no way of finding out exactly where Sarah or Billie may be making their mistakes. Lastly, it is a source of considerable discouragement if their pupils do not do well on these tests compared to other classes in the school. If they do not do well, school lore says that it is because she is a "poor" teacher; if the pupils do well, it is because she has a "good" class. Teachers never feel they win (are reinforced) from the results of standardized group tests—and they are quite right. One teacher may have done a brilliant job of teaching the Revolutionary War, for example, but she emphasized regional differences in participation and attitude because she planned to go on to the problems surrounding the Constitution that stemmed from these same regional differences. Unfortunately for her, that very special unit over which she had slaved was represented by one item on the standardized group test, under "map skills" where pupils were to label three regions. If a teacher is truly creative and unique, her efforts may not show because the tests are devised for the national classrooms, to test that which is most commonly taught.
This brings us to the Regents, and why many teachers find them rewarding. The Regents type of examination is one of the few instances where a teacher can test her teaching. By using typical Regents questions of previous examinations, a teacher can exert her skills as a teacher to prepare her students to do well. The teacher knows what will be covered, if not the exact questions. From her experience she may be a good predictor of the questions. As a teacher she can use every technique to help her students do well. The reward (= reinforcement) is that if they do indeed do well, everyone will know it almost immediately—the principal, the students, the parents, and other teachers. In any high school where Regents are given, the students and faculty underground circulate the same comment: "You can say what you like about _______, but her students always do well on the Regents." That is teaching that shows, and shows where it counts (= very reinforcing). Everyone knows about her performance as a teacher. (Some teachers, of course, resent having "to teach to the Regents" because their prerogative to select what to teach has been preempted.)
These comparisons of how to find out if pupils are learning showed the following:
Teachers find learning evaluation pleasant (= rewarding) when it:
1. Tests what the teacher has actually taught
2. Provides as immediate feedback as possible
3. Reports results about individual pupils in sufficient detail to be of teaching use
4. Compares the teacher to other teachers teaching exactly the same material
5. Provides pleasant information about the teacher's teaching
Teachers find learning evaluation unpleasant (= aversive) when it:
1. Is very time-consuming for the teacher
2. Is very tedious for the teacher
3. Reports errors of which the teacher had not previously been aware
4. Requires test-construction skills the teacher does not have
5. Makes unpleasant inferences about the teacher's skills
6. Involves complicated administration or reporting
7. Provides delayed feedback
8. Does not test what the teacher has taught.
Recitation is one form of learning evaluation which has these virtues from the list above:
1. It tests what the teacher has actually taught
2. It provides immediate feedback
Recitation avoids these unpleasant attributes:
1. It is not time-consuming
2. It is not tedious
3. It does not require test-construction skills the teacher does not have
4. It does not make unpleasant inferences about the teacher's skills.
Recitation or Evaluation
Recitation provides the teacher with more immediate information about how her pupils are learning for less trouble and effort than any other evaluation method. The Regents type of examination requires more effort, but is more rewarding for those teachers whose classes do well. Recitation has a lot to recommend itself to teachers as a method of evaluation. No wonder so many teachers use it.
Recitation has a lot to recommend it also as a source of social reinforcement for the teacher. It is important to teachers that pupils respond verbally to what she says or asks. (It is to anyone.) A pupil response to a teacher question is a very important social reinforcer, one that the pupils control. So well do they control the social reinforcers that pupils have altered teacher behavior experimentally by behaving more responsively to certain actions by the teacher.
When a teacher asks a question about the subject matter, there are at least two ends she has in mind:
1. To see if the pupil is learning, i.e., can give the correct answer
2. To elicit a social response from the class.
The worst thing (= most aversive) that can happen to a teacher is that the class pay no attention to her. She gives directions, orders behavior to change, states the lesson for the day and finds no pupil minding her. Instead, the pupils are attending to each other. This is the kind of aversive situation many teachers encounter in urban disadvantaged classrooms.
In nondisadvantaged classrooms, the same pupil reaction can happen. The teacher may ask a complex question, such as, "How would you interpret the American experience in terms of land use?" and a reply is not forthcoming. The class is silent and embarrassed because the class understands full well that they are withholding social approval (reinforcement) by their silence. Finally, a pupil ends the silence by asking instead, "May we put on our sneakers before the gym period?" Result? Teacher receives a social reinforcer (they responded verbally) but an aversive response (they did not understand the question, did not respond to it, and asked an irrelevant question).
If a teacher wants not only a response, but a correct response, then she will have to tailor her questions to the answerable level of her class. The more a teacher wants to hear correct responses, the lower the level of complexity of her questions; the lower the level, the more probable it is that more of the class will respond. If a teacher wants to determine if each pupil in her class is learning, and uses recitation as an evaluation tool, then it is clear that the questions she will use will have to include so-called lower level questions which call for the recitation of a single fact which has been taught or read. Questions calling for single fact responses have one other advantage for the teacher. They can tell her if a specific passage has been read by the pupil. What would be called a higher level question, one perhaps that involves pulling together several pieces of information to make a generalization, may be a better question only in the sense that it requires, presumably, a more complex cognitive procedure in order to answer it. From a teacher's point of view, such a question may not help her to determine if a pupil has done his work because the answer may stem from information acquired from nonclassroom sources. Some would think it desirable that a student use his memory, regardless of the source of the information. There are teachers who see that an important part of their job is to make sure the material is "covered," that pupils read the assignments and do whatever work is assigned. In this sense, teachers are like work supervisors who use recitation questions to check out the productivity of their workers.
This view of how many teachers see their responsibility will not sit well with those who think classrooms should involve Socratic dialogues, prepare children to question, and practice pupils in such processes as integration, inference, evaluation, and generalization. Critics of the schools often score the lack of creativity (whatever that is) and intellectual curiosity exhibited in the classrooms they visit.
If one looks at a teacher in terms of how a teacher's work is rewarded (= reinforced), the reasons for her classroom recitation behavior are soon understandable. Unless a teacher covers the material as outlined in the curriculum, her pupils will not be prepared for the standardized tests, nor will they be prepared for the next grade's curriculum. If her pupils do not do well on the standardized tests, or if next year's teacher finds them unprepared, she is likely to receive disapproval (punishment) from the principal and from other teachers. If they do well, she is regarded in a better light, even though the credit is often given to the pupils as being a "good class," i.e., quick to learn and therefore easy to teach.
What does the principal reward in teachers? In most public schools he rewards those teachers' behaviors which most affect his performance as the manager of a building unit in a bureaucracy. The teachers' behaviors that most affect his performance are the keeping of records; the efficient and smooth movement of people in time and space within the building; the preservation of property around the building; the lack of incidents that reflect badly on the principal's ability to operate a smoothly running unit; and the deployment of personnel so that there are no empty or unsupervised places, which could lead to lawsuits or insurance claims. These are the bread-and-butter behaviors to which the principal attends; and, in turn, is obliged to attend to because he, in turn, is rewarded or criticized (punished) for the way he and his building unit perform in these same areas. True, once a year his building's test results may be reported and commented upon, and that reflects an educational performance as opposed to a purely bureaucratic performance. But it is only once a year, which is a very infrequent rate of reinforcement, when everyday, and every week, which are high rates, there are all these checks being made on his unit's bureaucratic performance. It is not surprising that such a principal approves of (rewards) teachers who keep accurate and up-to-date records; who run quiet, incident-free classes; who move their class to and from the auditorium with a minimum of disruption; who keep their classroom at a good level of maintenance; and who willingly fill-in for an absent teacher for cafeteria or corridor duty.
These are the teacher behaviors that are most central to his reinforcement as a principal in a larger bureaucracy. As a consequence, he attends to and observes those behaviors more often because they affect his rate of reward. What the principal rarely attends to is the teacher teaching in the classroom. What he will attend to is whether the shades are drawn or whether the bulletin board is neat. Teachers report how ridiculous this is—yet the behavior persists, so it must be reinforced. The answer has to be that anyone in a position superior to the principal rarely comes into his school, looks at a teacher teaching, and says to the principal, "What a wonderful job of teaching! You must be proud to have her in your building." What such a bureaucratic superior observes most frequently is the organization of the building, its maintenance, and the level of classroom noise to see if the unit is operating smoothly and is not likely to create problems (= aversive responses) for higher supervisory levels.
The view presented here is that a teacher and a principal follow the same laws of reinforcement that other human beings are known to follow, i.e., we increase the frequencies of those behaviors that are rewarded; we decrease those that are ignored or punished. Let a school be found where teachers are rewarded most often for their teaching, and where principals are rewarded most often for being a "principal teacher" and there will be found classrooms that sparkle with curiosity and intellectual excitement and innovations and special units and hard work and systematic evaluation—and all those other things that principals and teachers are rewarded for in this school where teaching is the focus of reinforcement. In the meantime, let us refrain from criticizing teachers who use "recitation" or any other teaching technique so persistently and ask the important question instead: "Where are the reinforcers for this behavior coming from?"
Bellack, A.A., Kliebard. H.M., Hyman, R.T., and Smith, F.L., The Language of the Classroom. New York: Teachers College Press, 1966.
Gall, M.D., "The Use of Questions in Teaching," Review of Educational Research, Vol. 40, 1970, pp. 707-721.
Hoetker, J. and Ahlbrand, W.P., "The Persistence of the Recitation," American Educational Research Journal, Vol. 6, 1969, pp. 145-167.
Rappaport, Sandra, "Teacher-Student Questioning and Approval/Disapproval Behavior in High School Social Studies," unpublished Ph.D. dissertation, Teachers College, Columbia University, New York, 1974.
Skinner, B.F., "The Technology of Education," in Skinner, B.F., ed. Cumulative Record. New York: Appleton-Century-Crofts, 1961.
Smith, Maureen, "Teacher-Pupil Questioning and Approval/Disapproval Behavior in Elementary Social Studies," unpublished Ph.D. dissertation, Teachers College, Columbia University, New York, 1973.
Stevens, R., The Question as a Measure of Efficiency in Instruction: A Critical Study of Classroom Practice. New York: Teachers College, Columbia University, 1912.
Thorndike, E.L., The Psychology of Learning. New York: Bureau of Publications, Teachers College, Columbia University, 1913.