Reading: A Search For Meaning
To understand why textbooks baffle many students, University of Pittsburgh researchers Isabel Beck and Margaret McKeown suggest this exercise: Take a typical 5th grade social studies text on the period preceding the American Revolution and replace such words as "Britain,'' "France,'' and "North America'' with unfamiliar-sounding pseudonyms.
The result might read something like this:
"In 1367, Marain and the settlements ended a seven-year war with the Langurians and Pitoks. As a result of this war, Languria was driven out of East Bacol. Marain would now rule Laman and other lands that belonged to Languria. This brought peace to the Bacolian settlements. The settlers no longer had to fear attacks from Laman. The Bacolians were happy to be part of Marain in 1367. Yet, a dozen years later, these same people would be fighting the Marish for independence, or freedom from United Marain's rule.''
That, these researchers say, is a fair approximation of how a textbook in any subject might appear to a 5th grader. Unlike adults, youngsters often do not have the appropriate background knowledge to make sense of what they read. To them, such names as Britain and France can be almost as unfamiliar as Languria and East Bacol. But even students who already know something about what they are reading have trouble when the ideas in a text are poorly connected or fail to "hang together.''
Now, Beck and McKeown are perfecting a technique that teachers can use to help students derive more meaning from their textbook readings. The technique is known as "Questioning the Author,'' and its methods are deceptively simple. Teachers employ questions designed to encourage students to think of textbook authors as people who may not always present their ideas clearly. "Hmm,'' a teacher might pause and say, "I wonder what the author meant by that?'' Then, the teacher artfully weaves the students' responses together in a way that helps the pupils collectively make sense of what they read.
While this change is a small one in the realm of education reform, it appears to produce some significant effects. In the two classrooms where the technique has been tested so far, the number of studentinitiated questions went up, teacher talk decreased, and students' understanding of the subject matter improved.
But, as the researchers and critics are quick to point out, Questioning the Author is very much a work in progress. The technique is being further tested to see if the results will hold up. "To the question, 'Can it work?' the answer is that we know that it can,'' McKeown says. "But if you ask, 'Will it work?' we can't answer that question yet.''
Cognitive psychologists long ago concluded that students need to actively engage with texts in order to learn from them. Over the years, researchers have come up with a number of promising strategies to help students do that. One is a method called reciprocal teaching, in which students are encouraged to pause at intervals in their reading and "become the teacher.'' They might ask their classmates questions about the text or encourage them to predict what may happen next.
But over the years, as the technique has moved from closely controlled experiments to widespread classroom use, some experts have begun to contend that it may be too mechanistic. Moreover, says Carl Bereiter, a scholar-in-residence at the American Institutes of Research, a think tank based in Palo Alto, Calif., it gives students a "prop'' that might not serve them as well outside the classroom environment. "Eventually,'' he says, "you want all the props removed, and the issue is what are you left with? I think what you're left with with Questioning the Author is good meaningful discussions that can be carried on independently.''
Questioning the Author grew out of years of fine-grained research conducted by Beck and McKeown at the University of Pittsburgh's Learning Research and Development Center (LRDC), which has spawned a number of groundbreaking cognitive studies in recent decades. Recognizing that most social studies texts suffer from a lack of coherence or that they often presuppose background knowledge that a student does not have, the researchers took passages from elementary school texts and rewrote them to fill in background details and make them more coherent. They found that students understood more of what they had read with the revised passages than did other groups of students who were using the original texts.
In later experiments, students were asked to pause as they were reading the revised texts to reflect on what they had just read. Again, comprehension was enhanced.
Questioning the Author combines successful elements of both those early studies. With the technique, students are encouraged to bring a "reviser's eye'' to the task of reading, to grapple with the text so that its meaning is made clearer. They also read as a group and then stop at key points, a practice the researchers call reading "online.''
"What is the author trying to say?'' a teacher might ask, or "How does that connect with what the author already told us?''
"You are changing the goal from reading to understand to reading to try to figure out something that might not be clear,'' Beck says. "If there's a problem, the problem is not with the child. It is with the way it is written. We also think kids don't know they're supposed to work at it, and it doesn't necessarily come easily.''
While the notion may be simple, the technique is not.
"As soon as you say, 'What is the author telling us?' boy, oh, boy, have you opened yourself up,'' Beck continues. "You don't know what the kid is going to say, and, if he tells you something goofy, what are you going to do about it? We have teachers exhausted after one 40-minute lesson.''
Some might say the technique does not go far enough in changing classrooms. But the researchers, admittedly not revolutionaries themselves, say that is the point: The technique can be used within the confines of a traditional classroom. "We're not going to come in and say we will completely change your school,'' Beck says.
For their testing ground, the researchers chose St. Agnes School, an inner-city K-8 Roman Catholic school in Pittsburgh's Oakland neighborhood that has served as host for a number of LRDC experiments over the years. Nearly all of the 214 students in the school are African American, and more than 70 percent come from families poor enough to qualify for the federal free-lunch program.
For all its willingness to participate in cutting-edge experiments in cognition, St. Agnes has a traditional flavor. Students file in single lines from class to class in the turn-of-the-century building. The bell to signal a change in classes is hand rung. And in some of the classes, desks are arranged in straight rows.
Teacher Lisa Donovan remembers the way she used to run her language arts and social studies classes. "I had a teachers' edition,'' she says, "and I had the questions in front of me, and I did not have to do anything else. I was the sole authority in the classroom, and when kids would read independently, they wouldn't always tell us what they didn't understand.''
Teacher talk, Donovan admits, dominated her class; students asked few questions. Studies show, in fact, that that is the case in most classrooms. One 1988 report suggests, for example, that the average number of student-initiated responses in any typical one-hour lesson is only two.
To collect baseline information for their study, the researchers began videotaping Donovan's classes in the spring of 1992 before she began using Questioning the Author herself. At first, Donovan was skeptical that the technique could work. "I thought if I let the students talk to each other freely like that, I'm dead,'' she says. "I was also worried about the amount of time it took.''
Donovan and fellow teacher Kelly Sweeney became converts when they saw a videotape of McKeown using the technique with their own students. "I was astonished at the responses and involvement in the discussion from some of my students who usually never participate,'' Sweeney wrote at the time in a journal the researchers asked her to keep. "I cannot express my astonishment enough.''
Over the following summer, the teachers spent several weeks at the University of Pittsburgh center studying videotapes, learning how to formulate questions and comments, and reviewing texts for sections that their students might have trouble understanding. During the 1992-93 school year, the teachers kept journals and attended weekly meetings with the researchers. In addition, their lessons were videotaped once a week.
Despite all the coaching, change came hard. "I started to feel comfortable with it toward the end of the year,'' Donovan says. "When you go to other teachers, they say, 'That's what I already do,' but I know they can't be doing it.''
In addition to posing the right kinds of questions, teachers must learn how to deal with unexpected student responses. If a student's answer shows a kernel of understanding, for example, the teacher must seize on that and think of a way to develop it or to connect it with what other students in the class are saying.
Donald Weisz, another St. Agnes teacher using the technique, paces around his classroom. His 5th grade students are in the midst of deciphering a textbook history lesson on the pre-Revolutionary War period. The subject is government in the British colonies in North America.
"I think what I'm hearing,'' he says, "is that the king's not the only one who has good ideas and that the people want other people's ideas and opinions, too. Did the authors say the people are smarter than the king?''
A number of hands shoot up.
"They said everything they do they have to check with the king,'' a student volunteers. "First of all, the king is all the way over in another country, so now they get over here, and they made up selfgovernment.''
"But does that mean that's more important than the king?'' another student, Sheree, asks.
"She said the king is all the way over in England, and that's pretty far away,'' Darla says.
"But she said,'' Darla continues, pointing to another classmate, "they all have good ideas, and more than one person can have a good idea.''
"I agree with Jamie Jackson,'' says another student. "I don't think they're smarter than the king; I think they all together have better ideas than the king.''
"Can you connect with what Darla said?'' Weisz asks the student. "Why would people way over here in America have better ideas than the king?''
Eventually, the class arrives at an understanding that is not explicitly spelled out in the text: The British colonists living in America, already familiar with democratic principles of self-government, thought they knew best how to govern matters affecting their home territory-- matters of which a king or Parliament member who had not set foot in America would have little knowledge.
"We just go around and discuss it,'' explains Manda Metzger, a student in the class. "When we didn't have Questioning the Author, sometimes it was harder to figure out what was going on.''
Early findings from the first year of the study bear out her words. In a test of reading comprehension given to students in September 1992, two-thirds of students' responses demonstrated either misunderstanding or isolated repetition of text statements. On a post-test given the following May, however, more than half of the responses were judged to show that students had constructed some meaning from the text.
The percentage of students who could recognize when they did not understand something in a text increased from 30 percent to 80 percent from September to May. And, of the students who said the text did not make sense to them, more than six times as many were able on the post-test to provide reasons why.
There were also differences in the teachers' classroom behavior. In Donovan's language arts class, for example, the ratio of teacher talk to student talk--as measured in lines of dialogue from classroom transcripts--decreased from 71:29 before the experiment to 45:55 the following spring.
Beck and McKeown also analyzed the kinds of questions and comments the teachers were making and found that very few of the questions they asked at the end of the yearlong experiment were simple requests for students to retrieve text information. Instead, their comments were directed, in keeping with the technique, toward eliciting more meaningful responses from students.
This past school year, the researchers expanded the experiment to include two other classrooms. In addition to Donovan's and Weisz's classes at St. Agnes-- Sweeney now teaches in a public school-- the experiment includes two public school classrooms in Pennsylvania's Monongahela Valley. Besides hoping to replicate their results, Beck and McKeown want to study what it takes for classroom teachers to make lasting changes in their behavior. "We've got to do better at understanding why some innovations drop dead after a while,'' Beck says.
In the meantime, Donovan says she plans to continue to use the technique in all her classes. "Now that I've got it, and I've found some success with it,'' she says, "I don't think I could possibly go back.''--Debra Viadero