Lecturas de clase, Ejercicios de Psicología Educacional. Universidad de Granada (UGR)
gretah-2
gretah-2

Lecturas de clase, Ejercicios de Psicología Educacional. Universidad de Granada (UGR)

20 páginas
5Número de visitas
Descripción
Asignatura: psicologia de la educación, Profesor: Francisco Cano, Carrera: Psicología, Universidad: UGR
20 Puntos
Puntos necesarios para descargar
este documento
Descarga el documento
Vista previa3 páginas / 20
Esta solo es una vista previa
3 páginas mostradas de 20 páginas totales
Descarga el documento
Esta solo es una vista previa
3 páginas mostradas de 20 páginas totales
Descarga el documento
Esta solo es una vista previa
3 páginas mostradas de 20 páginas totales
Descarga el documento
Esta solo es una vista previa
3 páginas mostradas de 20 páginas totales
Descarga el documento

Implementing Reciprocal Teaching in the Classroom: Overcoming Obstacles and Making Modifications

Douglas J. Hacker University of Utah

Arnette Tenent Memphis City Schools

The purpose of this qualitative study was to investigate teachers’ implementation and practice of reciprocal teaching (RT) in 2 elementary schools. Over a 3-year period, 17 elementary school teachers participated in the implementation of RT. The obstacles they encountered and modifications made to RT were examined in vivo. Teachers modified their practice of RT, and the authors examined their modifications using 3 elements of RT: strategy use, dialogue, and scaffolded instruction. The focus was on whether these 3 essential elements remained in the teachers’ constructions of RT. The authors also focused on whether teachers added anything new to RT. Theory and guidelines that can be used to help teachers with the implementation and practice of RT are developed.

Reciprocal teaching (RT) is an instructional procedure in which small groups of students learn to improve their reading compre- hension through “scaffolded instruction” of comprehension- fostering and comprehension-monitoring strategies (Brown & Palincsar, 1989; Palincsar & Brown, 1984). These strategies are predicting, clarifying, questioning, and summarizing. There are many good sources that can be consulted for details concerning the classroom procedures of RT (see Brown & Palincsar, 1989; Marks et al., 1993; Palincsar & Brown, 1984; Palincsar, David, & Brown, 1989; Rosenshine & Meister, 1994). Suffice it to say that RT involves the use of the four comprehension strategies in ongoing dialogues between a dialogue leader and students in small groups. The dialogue leader, who can be a teacher or student, models the strategies by asking questions about a text, summarizing the text, clarifying misunderstandings, and asking students to predict up- coming text. Dialogue leaders fade their involvement, and other students in the groups take turns as leaders. The overall goal is to create, through collaboration, the self-regulated and flexible strat- egy use that is necessary for students to gain greater meaning from their reading (Palincsar, David, Winn, & Stevens, 1991).

Many studies since Palincsar and Brown’s (1984) seminal work have been conducted to test RT’s effectiveness. These studies have included at-risk readers, remedial readers, and good, average, and poor comprehenders, and their ages have ranged from 7 years of age to older adults (e.g., Brown, 1997; Hart & Speece, 1998; Lysynchuk, Pressley, & Vye, 1990; Rosenshine & Meister, 1994). Although results differ according to the kinds of measures used to evaluate instructional effectiveness, reading comprehension has consistently increased using RT. Across 16 studies examined by

Rosenshine and Meister (1994), there was an overall effect size of .14 on standardized tests favoring RT over control programs, but median effect sizes were as high as .34 to .60, depending on whether students were taught the strategies in conjunction with reading texts or as an explicit component prior to reading texts, respectively (Shuell, 1996).

However, even the best instructional programs result in limited gains if teachers find them difficult to implement or antithetical to their established practices (Duffy, 1993; El-Dinary & Schuder, 1993). All teachers, novice and experienced, rely strongly on their beliefs and knowledge about instruction when considering new practices (Borko & Putnam, 1996). An oversight of researchers who have been advocating constructivist practices for students has been to ask teachers to implement new curricula and use new instructional methods without considering that teachers, too, need to take ownership of their learning by constructing understanding of new curricula and methods using their prior knowledge. Instruc- tional methods change with each teacher, and perhaps they need to change at least to some extent to become part of a teacher’s constructed practice.

An important question that must be addressed by researchers who advocate new programs is as follows: Will the essence of what makes an effective program work survive a teacher’s con- struction of it? Researchers have given attention to reading pro- cesses and programs, but there continues to be little attention given to the implementation and practice of programs (Hiebert & Ra- phael, 1996). As far as RT is concerned, although data indicating the number of teachers who use RT are scarce, reported difficulties with implementation and practice are not uncommon (e.g., Marks et al., 1993). In contrast to most studies of RT, which have involved mainly quantitative analyses of the effectiveness of RT versus other reading instruction, the present qualitative study in- vestigates teachers’ implementation and practice of RT.

Over a 3-year period, 17 teachers from two elementary schools participated in the implementation of RT. The obstacles they encountered and the modifications they made to RT were exam-

Douglas J. Hacker, Department of Educational Psychology, University of Utah; Arnette Tenent, Memphis City Schools, Memphis, Tennessee.

Correspondence concerning this article should be addressed to Douglas J. Hacker, Department of Educational Psychology, University of Utah, 1705 East Campus Drive, Room 327, Salt Lake City, Utah 84112-9255. E-mail: hacker_d@ed.utah.edu

Journal of Educational Psychology Copyright 2002 by the American Psychological Association, Inc. 2002, Vol. 94, No. 4, 699–718 0022-0663/02/$5.00 DOI: 10.1037//0022-0663.94.4.699

699

1 LINE SHORT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

ined in vivo within each teacher’s classroom (Dunbar, 1997). Teachers on their own initiative modified their practice of RT, but most retained the use of the reading strategies in RT. We have generalized across the teachers’ practices and the modifications they made to develop theory about how RT can be more effec- tively practiced. Using this theory, we have provided guidelines for teachers to use in their implementation and practice of RT.

Essential Elements of RT

At the outset of this study, we identified three essential elements of RT as described by Palincsar and Brown (1984): (a) instruc- tion and application of four comprehension-fostering and comprehension-monitoring strategies (i.e., predicting, questioning, clarifying, and summarizing), (b) instruction and application of the four strategies using rich and meaningful student dialogues, and (c) providing scaffolded instruction of the strategies during which teachers gradually fade their modeling of the strategies and relin- quish greater control to the students. Our focus was on whether these three essential elements remained in the teachers’ construc- tions of RT, and, if so, what form the elements took in each teacher’s practice and how teachers differed from one another. We also focused on whether teachers had added anything new to the practice of RT. Most teachers have a variety of pedagogical approaches that they have used, tested, and believe to be effective, and it seemed reasonable to assume that these approaches would be incorporated, at least in part, into the practice of RT.

The comprehension-fostering and comprehension-monitoring strategies at the heart of RT are still the focus of much debate about how to increase readers’ comprehension. Rosenshine and Meister (1994) have raised several important considerations about which strategies should be taught in RT, how many strategies should be taught, and how these strategies should be taught. In their analysis of RT, Rosenshine and Meister found that significant gains in student achievement were made when 2, 3, 4, or 10 strategies were used, implying that perhaps the kinds and number of strategies are not as important as the kind of cognitive process- ing that is needed to increase comprehension. Some speculation is that the critical function of a comprehension strategy is to increase elaboration and inferential processing (Rosenshine & Meister, 1994; Rosenshine, Meister, & Chapman, 1996). Both elaboration and inferential processing help readers to integrate textual infor- mation with background knowledge (Kintsch, 1998).

However, in addition to considering which of the RT strategies effectively increase comprehension, other equally important con- siderations include which strategies teachers continue to use in their practice and how teachers use these strategies. An effective strategy rarely or poorly used does not lead to increased reading comprehension. In each year of the present study, we instructed teachers in the use of the four RT strategies and then observed them in their practice of RT. We were interested to find out which strategies teachers continued to use and of those strategies still in use, how teachers were using them.

Dialogue is also an essential element of RT (Palincsar, 1986). Through highly social, interactive, and holistic dialogues, students acquire the use and generalization of the four reading strategies (Palincsar & Klenk, 1992). Using Vygotskian psychology as a theoretical foundation for this social component of RT, Palincsar

and Brown (1984) have argued that participation in high-quality group dialogue helps students to internalize the use of the strate- gies and to learn monitoring and regulation of their own compre- hension. Learners internalize higher cognitive functions, such as monitoring comprehension, by moving from an interpersonal plane, in which social interactions serve as modeling for new behaviors, to an intrapersonal plane, in which those modeled behaviors become the bases for thinking and learning (Vygotsky, 1978). An underlying assumption of RT is that by using the four strategies in a group process, students make this shift from the interpersonal to the intrapersonal.

However, whether students make this shift hinges on whether they engage in high-quality group dialogue. Superficial or unfo- cused discussions of content do little to help students internalize the four strategies. Therefore, our second interest was to find out whether high-quality dialogue was occurring among students and how the strategies were being used to stimulate this dialogue. We expected that teachers would use dialogue in different ways, but our question was whether their uses of dialogue led to high-quality interactions among students.

Finally, the reciprocal nature of RT requires the teacher to scaffold the use of the strategies and to provide support only on an as-needed basis (Palincsar & Brown, 1984). With scaffolding, the teacher models how the strategies can be applied to text and the students gradually acquire the modeled behaviors and thoughts. The teacher fades his or her modeling of the strategies and relin- quishes control to the students. The students also can act as models for one another as they apply the strategies to their reading and reason about the text. Eventually, the students assert their own interpretations of the text without the potential constrictions im- posed by the teacher as inquisitor or evaluator (Almasi, 1995). By gaining greater control of the reading process, students potentially become better self-regulators of their reading.

To facilitate students’ self-regulatory processes, teachers must reduce their scaffolding and place students in control of the pro- cesses they are eventually to self-regulate. Arguably, by maintain- ing the role as cognitive monitor and providing more scaffolding for longer periods, teachers may inhibit the very self-regulatory processes that are necessary for reading comprehension. However, more research is showing that metacognitive monitoring and con- trol of strategy use are qualities of the learner that take long periods of time to develop and are effectively developed only through extensive practice of skills that are made visible by teach- ers thinking aloud or modeling (Pressley & McCormick, 1995). Our third interest, therefore, was to examine over time how each teacher adjusted the amount and duration of scaffolding they provided for students to learn and apply the strategies.

Method Participating Schools

The two schools that were the sites for the present study were elementary schools located in a large urban area of the mid-southern region of the United States. The schools had no obligation to participate in the study, and their implementation of RT was completely voluntary. The principals at both schools had independently contacted Douglas J. Hacker (D.J.H.) and asked whether he could provide assistance in developing literacy programs.

Each school served students from kindergarten to sixth grade, with a total population at each of approximately 300–400 students. School 1

700 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

served an entirely African American population, and School 2 served a population that was approximately 55% Caucasian, 38% African Ameri- can, and 7% Hispanic or Asian. School 1 had a teacher population that was about 65% African American and 35% Caucasian, and School 2 had a teacher population that was about 56% Caucasian, 38% African American, and 6% Hispanic. Both schools were site-based managed.

We acknowledge that a necessary ingredient for teachers to construct and reconstruct their practice is motivation and that without motivation to change practice, instructional reform will be minimal (Borko & Putnam, 1996; Richardson, 1996). We believe that the teachers at both schools had several reasons for considering RT. Foremost, standardized reading scores at the two participating schools were far below national averages. Teachers were interested in a reading program that could get students actively involved in reading and that had a strong research base demonstrating its effectiveness. Also, because the two schools were involved in schoolwide reform initiatives that required the implementation of student-centered approaches, the teachers were interested in the group format of RT. Moreover, the reform initiatives did not contain an explicit reading pro- gram; therefore, both schools needed a reading program that could be implemented throughout the curriculum and used across content areas. In general, the participating teachers believed that the four strategies in RT could be sufficiently flexible to generalize to a variety of content areas. Finally, because teacher adoption of RT was voluntary, motivation for change was intrinsically driven rather than externally dictated by the administration at each school.

Data Collection

The primary sources of data over the 3 years of the study consisted of observations of 17 teachers practicing RT and surveys from the teachers. D.J.H. observed and wrote field notes that were used along with the survey data to identify the obstacles that teachers had encountered and the mod- ifications they had made to RT. D.J.H. observed teachers in their class- rooms several times throughout the school year. The number of observa- tions for each teacher varied but ranged from two to five. Each observation was about 30–45 min. For all observations, D.J.H. assumed a privileged position (Wolcott, 1988) and observed while on the peripheries of the classrooms, engaging in little or no interaction with the students or teach- ers. Occasionally, brief verbal exchanges occurred between the teacher or students and D.J.H.; however, exchanges were limited to comments about the content of the texts being read. At the end of each observation, D.J.H. gave oral or written feedback to teachers on their RT instruction, and teachers gave comments and suggestions about their practice. Teachers’ comments and suggestions served as a source of information that D.J.H. used to further modify guidelines for implementation and practice of RT.

In addition to teacher observations, at the end of the second and third years of implementation, D.J.H. administered a teacher survey that asked teachers about their practice of RT, their perceived strengths and weak- nesses of RT, and their recommendations for improvements. At the end of the third year of implementation, students and parents were surveyed for their impressions of RT and the reading program. Moreover, during the third year, students were pretested at the beginning of the year and posttested at the end of the year using in-house measures of reading that included both a measure of vocabulary knowledge and comprehension. Gain scores were calculated for each student by taking the difference between pre- and posttest scores. At the end of the second and third years, students from Grades 3, 4, and 5 were tested with the reading portions of the Comprehensive Test of Basic Skills (CTBS; 1997). Administration of the CTBS was part of a districtwide accountability measure.

The various kinds of data collected and the methods of data collection helped to ensure the validity of the study (Denzin & Lincoln, 1994). Data triangulation was accomplished by collecting data from 17 different teach-

ers on multiple visits over an extended period of time. Methodological triangulation was accomplished by collecting data using observations, surveys, test scores, and discussions with teachers. Moreover, with 3 years of prolonged field engagement, D.J.H. had ample time to learn the class- rooms and school cultures and build trust with the teachers.

Finally, member checking (Gall, Gall, & Borg, 1999) was accomplished at various times throughout the 3 years. D.J.H. used the multiple sources of data to analyze consistencies and differences in the teachers’ practice of RT and then used this analysis to modify the RT implementation and practice guidelines proposed by Palincsar, David, and Brown (1989). The modified guidelines were provided to teachers for their review and comment, and their comments served as additional data that were used to further modify the guidelines. Thus, using a method similar to the constant comparative method (Strauss & Corbin, 1994), practice and the underlying theory were elaborated and modified several times during the 3 years as new data were compared against them.

Data Analysis

For each year of the study, D.J.H. analyzed the various sources of data to identify patterns in teachers’ implementation and practice of RT. The patterns were organized around four themes: (a) the four RT strategies and how they were being used, (b) student dialogue, (c) teacher scaffolding, and (d) added elements to RT. The collective evidence for each year was used to categorize the teachers’ practice of RT as either a version of RT that closely matched the RT described by Palincsar and Brown (1984; i.e., traditional RT) or a version that used RT in ways that modified strategy use, dialogue, or scaffolding from the traditional RT practice. By the end of the third year, three categories of RT practice emerged: traditional RT, traditional RT that incorporated some use of whole-class instruction, and RT that used whole-class instruction exclusively.

In addition, the teachers’ comments about their practice of RT were used as input to the ongoing modification to the implementation and practice guidelines. At the end of each school year, D.J.H. compiled the modifica- tions that each teacher had made to RT, looked for consistencies and differences, and generated new guidelines for the implementation and practice of RT. Thus, the new guidelines reflected not only consistencies in practice across the teachers but also specific practices that a teacher had adopted with good effect. These guidelines were shared with the teachers, who provided feedback. At the end of the first year, the collaborative efforts of the 2 teachers and D.J.H. resulted in a five-step implementation procedure (see Appendix A), which served to inform the implementation of RT the following 2 academic years at School 2. The collaborative efforts of D.J.H. and teachers at the end of the second and third years resulted in a final revised implementation procedure (see Appendix B).

Year 1: Pilot Implementation at School 1

Procedure

During the first year, D.J.H. worked with 1 third- and 1 fourth- grade teacher from School 1. Each classroom consisted of about 22 African American students. Both teachers volunteered to partici- pate in implementing RT in their classrooms. The 2 teachers were free to make whatever changes to RT that they believed were necessary to increase students’ literacy skills, but before they made changes, the teachers agreed to first discuss them with D.J.H.

Teacher training began at the beginning of the calendar year, with the 2 teachers reading Using Reciprocal Teaching in the Classroom: A Guide for Teachers (Palincsar et al., 1989). This was followed by a 1-day in-service, during which D.J.H. described

701IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

each step of the RT process and discussed the theory behind RT and its potential strengths and problems. To further demonstrate the use of RT, D.J.H. acted as dialogue leader with the 2 teachers and led them through a short story using the four RT strategies according to Palincsar et al.’s guidelines.

In their classrooms, the 2 teachers followed the RT guidelines, using the first 4 days to explicitly instruct and discuss the four strategies, with each day devoted to one of the strategies. Explicit instruction of the four strategies prior to starting the RT dialogues has been supported by a variety of studies (see Rosenshine & Meister, 1994). On the fifth and sixth days of implementation, D.J.H. visited the two classrooms to demonstrate RT. Prior to the demonstration, both teachers had organized their students into small reading groups of about five students each. Using a narrative text in which preselected points had been chosen to stop for questions, clarifications, summaries, and predictions, D.J.H. dem- onstrated to teachers and students the use of the strategies. The demonstrations lasted about 1 hr, after which D.J.H. further dis- cussed implementation with the 2 teachers.

Over the next 4 months, D.J.H. observed each teacher three or four times in their implementation and practice of RT and took field notes of the observations. At the end of each observation, D.J.H. discussed with the teacher the difficulties and strengths of RT that she had encountered, and the two agreed on the next steps in implementation. In addition, D.J.H. also met twice with both teachers together to discuss their progress with RT and to address their concerns. Field notes on these two group meetings were also taken.

Obstacles Encountered

With the very first observations in each classroom, both teachers had encountered several obstacles. The obstacles were similar across the two classrooms, but the manner in which teachers responded to them differed. The obstacles are organized around the four themes that were mentioned above: strategy use, dialogue, scaffolding, and added elements. Modifications that the 2 teachers made to RT in response to these obstacles are discussed in fol- lowing sections.

Strategy use. In both classrooms, the RT groups were not using all four of the strategies, and the strategies that were being used (mostly questioning and summarizing) were often being used inadequately. Many of the questions and summaries of the texts were superficial and literal and expanded only on propositional or sentential levels of the texts rather than more global understanding of them. For example, on the basis of a story about gravity, a student generated the question, “What pulls our bodies down toward the Earth?” which was a literal extraction from the text, “Gravity pulls our bodies down toward the Earth.” Rather than making elaborations and inferences that tie text information to students’ background knowledge, the superficial and literal kinds of questions and summaries that students generated left them with surface-level representations of the texts.

Also, many of the questions bore a remarkable similarity to one another. For example, students asked the following questions about a text they were reading: “Why did Benny tell his friends to keep it a secret that he could play the drum?” “Why did Benny ask his friends if anyone else was trying out on the drums?” “Why did

Benny try out for the drums?” and “Why was Benny so nervous?” The impression given was that the first question asked in a group established a prototype that was to be followed by the other students.

The clarifying strategy was rarely used, and students frequently passed over difficult words or sections of text that were not fully understood. Finally, predicting was used frequently and effectively for narrative text; however, there was an awkwardness using prediction with expository text. In these cases, the students would often pass over prediction and go directly to questioning.

Dialogue. Both teachers had difficulties trying to stimulate high-quality dialogues within their student RT groups. Many stu- dent interactions were constrained and remained at superficial levels throughout the RT sessions. Although the students were third and fourth graders, many of them lacked knowledge of the basic classroom rules of group discourse; therefore, they could not engage or were uncertain about engaging in even basic discourse turns. Some students who were assigned to be group leaders lacked knowledge and motivation to lead the group, and the group mem- bers quickly figured out that the leader would not or could not keep them on task. Many students became passive in their learning and relinquished responsibility for the discussions to a few members of the group. Groups occasionally got sidetracked with the procedural matters of RT, arguing whether it was time to question, summa- rize, or predict. Also, the discussions frequently led to irrelevant or tangential topics or into personal matters having nothing to do with the texts.

Scaffolding. Even after 2 months of implementation, many RT groups continued to need substantial amounts of teacher help using the strategies. During this time, the 2 teachers occasionally had difficulties devoting sufficient time to each of the four or five groups in their classrooms to provide the necessary scaffolded instruction. Thus, rather than gradually fading modeling of the strategies, per Palincsar et al.’s (1989) guidelines, the teachers found it necessary to maintain highly scaffolded instruction.

Moreover, the students did provide modeling for one another in their RT groups, but often the students modeled incorrect use of the strategies. Without the teachers’ direct intervention to fix incorrect strategy use, some students were forming incorrect no- tions of the strategies, the texts, or both. Also, without the teach- ers’ consistent intervention to scaffold students’ knowledge con- struction, the kinds of elaborations and inferences necessary for deeper understanding of the texts were rarely being modeled by students.

Additional concerns. Finally, the 2 teachers had difficulties assessing students’ progress with the strategies and assessing whether their reading comprehension was improving with the strategies. Both teachers were uncertain about whether their stu- dents had truly internalized usage of the four strategies or whether they were simply imitating strategy use by following group mem- bers. Furthermore, both teachers needed some type of assessment that could provide formal measures of reading comprehension. The teachers needed comprehension measures to satisfy accountability goals within the school, but, more importantly, they needed as- sessments to be able to give more constructive and personalized feedback to students about their reading progress.

702 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

Modifications Made

Although the obstacles to the practice of RT were similar across the two classrooms, the teachers differed in their responses to them. Because the teachers were free to make whatever changes they thought were necessary to help their students become better readers, we inferred that the changes they made to RT reflected each teacher’s own unique beliefs about practice. The unique variations on RT that eventually resulted in each classroom dif- fered from traditional RT; still, each teacher’s practice retained a strong semblance to the strategic processing recommended by Palincsar and Brown (1984).

Scaffolding of strategy use and dialogue. By far, the most challenging problem the 2 teachers experienced was getting stu- dents to maintain meaningful dialogues. Student dialogues were hampered because of the students’ poor group discourse skills and also because the students were using the strategies so inadequately or superficially that there really was little for them to discuss. The 2 teachers addressed these difficulties by maintaining highly scaffolded instruction of the RT strategies for extended periods of time. Both teachers extended whole-class instruction of the strat- egies to at least 2 months. Although both teachers believed that their students had learned the strategies during the first several weeks, students’ abilities to effectively apply the strategies to text was the primary problem. Thus, the greater scaffolding that teach- ers provided was primarily intended to help students apply what they knew about the RT strategies to different kinds of texts. At the same time, teachers provided greater scaffolding of the collabora- tive processes necessary for group dialogue. Both teachers devoted a great deal of time and effort to helping students becoming better listeners, responders, and contributors to a group process.

Traditional RT plus whole-class instruction. Both teachers developed a similar approach to using whole-class instruction of RT. Each teacher first introduced to the class a text that was to be read. Often, the teacher explicitly tied the content of the text to a topic that had been or was to be the focus of study in the classroom. She also would bring attention to pictures or other graphics in the text. Next, the teacher asked the students to make predictions about what they thought the text was about. The students then took turns reading the text aloud, and the students who were listening generated questions. To model deeper ques- tioning skills, the teachers often focused greater attention on those questions that required more complex processing, and if such questions were not asked, the teachers generated their own for the class. The teachers directed each question back to the class, calling on several students to provide an answer or to elaborate on a previous one. To encourage students to use the clarifying strategy, the teachers asked their students to circle any words or sentences that did not make sense. During questioning, the teachers made sure that the circled material was clarified. Finally, the teachers asked students to summarize in their own words what they had read. The summaries were done collaboratively with the entire class, and the teachers called on students to elaborate on the evolving summaries of the text. By being more directive (i.e., providing more highly scaffolded instruction), the teachers found that students maintained their discussions of the text longer and more seriously, were more active in their reading, and were pro-

vided with good models for summarizing, clarifying, predicting, and questioning.

After approximately 2 months of whole-class instruction of the strategies, both teachers began to reduce their scaffolding and got students working in RT groups. One teacher immediately began RT groups of about four to five students and followed a more traditional style of RT, as set forth in Palincsar et al.’s (1989) guidelines. However, the other teacher, although reducing the amount of support she gave to students, provided students with one additional step of scaffolded instruction by introducing reading partnerships before moving to RT groups. In the reading partner- ships, the teacher had two students work together, with one student reading a section of text and the other generating questions. Once all the pairs had read and questioned a section of text, the teacher called together the entire class so that each reading pair’s questions could be answered in the class. After several weeks of the reading partnerships working together, the teacher expanded the partner- ships into RT groups of about six students each. The RT groups then collaboratively read and generated questions, summarized, and clarified text on their own. After the texts were read, the teacher would again call all the students together in a whole-class instruction to review the questions–answers, summaries, and predictions.

Additions to RT. To encourage more active processing of text and to foster writing skills along with reading skills, both teachers introduced writing into their use of RT. Both teachers required students to write their questions, answers, and summaries. The teachers also found that students’ questions and answers could be used to help assess their command of the content that they were covering and how well they were reading.

Although the 2 teachers required written summaries, the way they used the summaries differed. The teacher who had begun RT groups immediately after the extended period of whole-class in- struction used summary writing as part of a process writing exer- cise. She first had students within each RT group collectively write their summaries. Once the summaries were completed, the groups then exchanged their summaries so that each group could have their work peer reviewed by at least one other group. The sum- maries were then returned to the original group for revision. Finally, the teacher collected the revised summaries and used them to assess student progress and provide additional feedback to students.

The teacher who began reading partnerships used summary writing in a whole-class instruction. Rather than asking individual students or groups of students to generate summaries, she led the class in the generation of a single summary using the collective input from the class. She first asked students to summarize in their own words what they thought the passage was about and then wrote their responses on the blackboard. After writing each sen- tence, she would ask students to add, change, or elaborate on what had been written. Occasionally, she would use this exercise to instruct grammar and spelling. On these occasions, she would intentionally introduce spelling and grammatical errors to the summary and would either have students spontaneously correct them or call on students to correct them. Once the summary was completed, she would ask students to read the summary aloud and to judge whether the summary was complete and to add anything they thought was missing.

703IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

By the end of the year, both teachers were using various com- ponents of their RT instruction as a means to informally assess student progress (e.g., collecting written questions or summaries, monitoring the quality of questions and answers); however, both teachers agreed that much more needed to be done about assess- ment. Students’ answers to their self-generated questions and their written summaries provided good approximations of students’ progress, but formal measures of reading ability and reading com- prehension were lacking. Both teachers agreed that systematic assessment of reading ability was needed throughout the academic year.

Year 2: Implementation at School 2

Procedure

During the second year, we worked with 7 teachers from School 2. Arnette Tenent was the full-time curriculum coordinator for School 2. Six teachers (2 from each grade level) were from Grades 3, 4, and 6, and the 7th teacher taught a Grade 4–5–6 split. Class sizes varied from 20 to 25 students, and the demographics of the students were representative of the school’s overall population (i.e., approximately 55% Caucasian, 38% African American, and 7% Hispanic or Asian). All 7 of the teachers volunteered to participate in the RT training and implementation. As with the procedure established the previous year, the teachers were free to make whatever changes to RT that they believed were necessary to increase students’ literacy skills, but, before they made changes, the teachers agreed to first discuss them with D.J.H.

D.J.H. followed the same implementation procedures that had been used the previous year. At the beginning of the school year, D.J.H. provided a 1-day RT in-service to teachers. Prior to the in-service, the teachers were asked to read Using Reciprocal Teaching in the Classroom: A Guide for Teachers (Palincsar et al., 1989). During the in-service, D.J.H. and the teachers discussed the RT process, the theory behind RT, and the results from the previ- ous year. Particular attention was given to the obstacles teachers had encountered and, more importantly, to how they had modified RT to fit their individual practices while still maintaining use of the RT strategies (see Appendix A).

For the implementation of RT, each of the 7 teachers used 4 consecutive days to explicitly instruct the four strategies, with each day devoted to one of the strategies. After the 4 days of strategy instruction, D.J.H. visited each classroom to model RT. Using a narrative text in which preselected points had been chosen to stop for questions, clarifications, summaries, and predictions, D.J.H. demonstrated to teachers and students the use of the strategies. The demonstrations lasted about 1 hr.

Over the next 2–3 months, D.J.H. conducted regular observa- tions of the teachers. At the end of each observation, D.J.H. and the teacher discussed the difficulties and strengths of RT that had been encountered, and the next steps in implementation were mutually agreed on. In February, D.J.H. scheduled with each teacher a formal observation of her practice of RT. The primary purpose of the formal observation was to examine how the teacher had uniquely modified RT. In these observations, no evaluation was made as to the effectiveness of the teachers’ practice of RT. Rather, the focus was on describing how the teachers had modified

RT in respect to the four themes that had been discussed above (i.e., the four RT strategies and how they were being used, student dialogue, teacher scaffolding, and added elements to RT). For the formal observation, written field notes were taken, and written feedback was provided to each teacher. At the end of the year, the teachers completed a survey about their experiences with RT. Among other things, the survey contained questions about the frequency and duration of RT use, whether RT was used with groups or whole-class instruction, how effective RT had been, whether RT worked better for some students, and how RT could be improved.

Obstacles Encountered

The obstacles that the 7 teachers encountered were similar to the obstacles that had been encountered by the 2 teachers the previous year; however, because of the greater number of teachers during the second year, the variety of obstacles increased. We again used the four themes discussed above to describe these obstacles. The following descriptions are based on data collected during the observations of each teacher and from the teacher surveys.

Strategy use. D.J.H. observed in several of the classrooms that maintaining meaningful dialogue was problematic and that a major contributor to the lack of meaningful dialogue was the mechanical use of the four strategies by some students. Often, the RT groups placed greater value on the product of the group processes rather than the process of learning about the texts. Students would make a prediction, generate a question, produce a summary, and be done with the strategies, regardless of the depth of understanding they gained. One of the third-grade teachers said, “When they work with me I ask the questions. Alone, they don’t ask each other. They read the paragraph, then move on. They are satisfied even if they do have questions.” Thus, the students perceived little need to engage in extended dialogue when the products of the dialogue (i.e., the prediction, question, or summary) were already accomplished.

Also, even after 5 months of practice using the strategies, 5 of the 7 teachers continued to find that many student questions were at superficial levels of understanding. For example, a fourth-grade teacher said, “I need to find more time to work with them more and spend more time giving direction and feedback on their questions.” Clarifications were rarely pursued, summaries missed major themes or concepts, and predictions were rarely followed up to determine if they were correct or incorrect.

Dialogue. Maintaining meaningful collaborative dialogues was again the primary challenge for teachers. Although all 7 of the teachers enjoyed using RT groups and the use of collaborative learning fit well with their overall teaching practices, the teachers encountered difficulties getting their students engaged in group discourse and directing the discourse toward greater comprehen- sion of the texts. Disruptions in the flow of dialogue were regular occurrences in the RT groups. One sixth-grade teacher stated, “They work in groups pretty well, but there always seems to be one ‘trouble maker’ in every group who disrupts the others.” A third- grade teacher noted, “Many groups argue for various reasons while others play around.” Some students often lacked motivation to engage in the group process and drifted off task, bringing a halt to the dialogue or shifting the responsibilities of reading to those who

704 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

stayed on task. The other third-grade teacher said, “But the low [reading ability] kids really do not participate like they should. They let the others do it for them. They’re not sure of themselves.” Some students needed greater direction about what to do when questions were not generated, when answers were incomplete or incorrect, or when summaries were word-for-word renditions of the text.

Teachers tried to arrange their groups so those students who tended to be off task or needed greater direction were placed with more disciplined students or students who could work indepen- dently. However, students who lacked motivation and self- regulation often had a debilitating effect on the entire group, and students who needed direction often became passive and allowed the other students in the groups to engage in the text. One of the third-grade teachers commented, “Those with little self-control have difficulty paying attention to other students in the group. Some allow one person to do all the work.”

Scaffolding. Using RT as a whole-class instruction was a recommendation generated the previous year in response to the difficulties that teachers faced getting students to maintain mean- ingful dialogues in their groups and to work cooperatively with one another. Whole-class instruction was intended to provide greater scaffolding for students during their acquisition of the RT strategies and collaboration skills. As per Palincsar et al.’s (1989) guidelines, teacher scaffolding should be gradually reduced as students become more self-directed with the strategies and begin to work collaboratively in their RT groups. However, after 5 months of practicing RT, 5 of the 7 teachers found that their students still needed highly scaffolded instruction of the four strategies: “They still need teacher direction to remain focused,” “All students do not work well in groups,” “The students do not function well yet as a group. They are not staying on the topic,” and “I wouldn’t say doesn’t work. But it takes a little longer for my slower students to participate with the group.” Overall, therefore, most of the teachers felt a need to maintain highly scaffolded instruction for longer periods of time.

Additional concerns. Similar to the previous year, the teachers voiced a need for formal assessments of reading ability. As 1 teacher noted, “RT does not provide much opportunity for objec- tive grades, and we are required to take grades each week.” The teachers believed that their practice of RT allowed for informal assessments of students’ reading progress, but because of a strong accountability movement throughout the school district, each teacher was responsible for providing multiple measures of student achievement. In addition to scores from standardized tests, teach- ers were expected to collect a variety of achievement measures in several domains, particularly literacy.

Modifications Made

Over 5 months of practice, teachers developed unique ways of responding to the obstacles that they had encountered and made a variety of modifications to RT. Overall, 3 teachers developed versions of RT that closely followed the RT guidelines described by Palincsar et al. (1989), and 3 teachers developed versions of RT that more closely followed the revised RT guidelines that had been generated the previous year. The 7th teacher eventually stopped using RT and chose other reading instruction for her class. The

following descriptions of the modifications are based on data collected during the formal observation of each teacher and from the teacher surveys. Evidence relevant to the four themes was used to categorize the teachers’ practice of RT as either traditional RT or traditional RT with whole-class instruction.

Traditional RT. The 2 fourth-grade teachers and 1 sixth-grade teacher followed the RT guidelines described by Palincsar et al. (1989; see Table 1). Their traditional approaches to RT were well received by students, who demonstrated many positive aspects of reading. During the formal observation of the classrooms, D.J.H. noted that the quality of student questions and answers was typi- cally high, and students were engaged in a great deal of discussion about the questions. One of the fourth-grade teachers noted, “It [RT] makes them slow down and think about what they’re reading. The questions are imperative to retaining knowledge.” There was a good deal of group collaboration in the generation of summaries for each section of text, and students appeared to be developing a good understanding of the texts. The sixth-grade teacher said, “It [RT] makes students go back and think about what they’ve read. It provides a good check for comprehension.” The sixth-grade teacher introduced writing into RT and required her students to write their questions and answers, which were reviewed for their content and quality by the teacher.

As a slight variation on traditional RT, on the day before reading a text, the other fourth-grade teacher assigned as homework the task of reading the text and generating five questions. On the day of discussion, the teacher then divided the students into their RT groups and had them engage in traditional RT: Group leaders first asked their groups to recall what had been read the previous class,

Table 1 Grades and Practices of Teachers Who Used Traditional Reciprocal Teaching (RT) Versus Traditional RT Plus Whole-Class Instruction During Year 2

Traditional RT Traditional plus whole-class RT

Grade Grade 4 (2 teachers) 3 (1 teacher) 6 (1 teacher) 4-5-6 (1 teacher)

6 (1 teacher)

Practices Practices RT groups (group leader) 1. Whole-class instruction

Take turns reading Introduce text Strategy use Discuss topic of text

Questioning Predicting 2. RT groups (group leader or dialogic) Clarifying Take turns reading Summarizing

Additions to RT Writing

Questions–answers Summaries

Question generation prior to RT groups

Strategy use Questioning: Write questions–

answers Predicting Clarifying Summarizing

3. Whole-class instruction Discussion of text Teacher questioning Teacher assessment of

comprehension Write answers to text questions

705IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

students took turns reading the text and asking their questions that had been generated the night before, and each RT group then summarized their reading. The teacher found that because students had read the texts and generated questions the night before, they were better able to engage in higher quality dialogues during RT:

It [RT] challenges them more to find answers to numerous questions. Since they take the role of asking the questions, it makes them more ready to have the answers in their minds or on paper for their class notes. This encourages them to do more reading since they want to be prepared since they take on the role as a leader.

Also, this teacher asked that each RT group collaboratively write the summary for the day’s reading. She then placed the written summaries and questions in each student’s portfolio as products for later assessment.

Traditional RT plus whole-class instruction. One third-grade teacher, 1 sixth-grade teacher, and the teacher who taught the fourth–fifth–sixth-grade split developed a version of RT that more closely followed the revised RT guidelines in which teachers maintained more direct control of the classroom by using whole-class instruction. Table 1 shows the sequence of reading activities developed across the 3 teachers. Because of the difficulties with small groups that were noted above, these 3 teachers felt that the whole-class format provided them with a better means to monitor the level of students’ questions and summaries, to keep students on task, and to assess students’ comprehension of text. The sixth-grade teacher commented, “As a small group students have more opportunity to participate but also more opportunity to play and not stay on task,” and the third-grade teacher said, “I use the components of RT but not the RT format. I use the parts with the whole class as a way, tool, to evaluate their understanding. I don’t particularly like leaving the kids ‘in charge’ every time.”

Each of the 3 teachers used whole-class instruction at the beginning of a reading assignment as a way to introduce the readings, to engage students in a discussion about the topics, and to activate students’ prior knowledge. After the introduc- tion to the readings, the teachers broke the students out into their RT groups for reading, questioning, predicting, clarifying, and summarizing the text. One of the 3 teachers followed Palincsar et al.’s (1989) guidelines and had a leader in each RT group facilitate the discussion of the text. However, the other 2 teachers did not use group leaders. Instead, RT groups in these two classrooms distributed the responsibility for reading and using the strategies across all group members in more of a dialogic fashion. Once the RT groups had completed their reading, the 3 teachers brought their students back into a whole-class instruction to discuss their understanding of the texts. During these discussions, the teachers asked the students additional questions that served to elaborate on understanding of the texts, to provide additional modeling of deeper level questions, and to informally assess students’ comprehension of the reading. Finally, 2 of the 3 teachers had incorporated writing into their versions of RT: One teacher used writing at the end of the RT sessions by asking her students to write the answers to the questions that appeared at the end of the text, and the other teacher used writing during the RT sessions by re-

quiring her students to write their questions and answers as a component of their group work.

The 7th teacher found RT to be ineffective with her students and stopped using it after a few months. Her primary criticism was that because her students had difficulties with the cooperation and collaboration necessary for RT groups to work, they needed an instructional method that provided greater teacher direction. She had found that some students could work well in groups; however, the majority of her students showed little self-control and did not stay focused during the RT sessions. She did find that some components of RT were useful, namely, the questioning and pre- dicting strategies, and these two strategies became part of her whole-class reading instruction that developed around literature circles.

Year 3: Schoolwide Reading Program at School 2

Procedure

For the third year, we worked with all 17 teachers at School 2 to implement a schoolwide reading program. Of those 17 teachers, the 15 full-time teachers had voluntarily agreed to participate in the RT training and to use RT in their classrooms. Five of the 15 teachers had used RT the previous year at School 2. Once again, the teachers were free to make whatever changes to RT that they believed were necessary to increase students’ literacy skills, but, before they made changes, the teachers agreed to first discuss them with D.J.H. The school’s reading committee, which consisted of teachers from the school, a consultant from the school district, and us, planned a schoolwide daily reading period, during which all teachers were to focus on building literacy skills. For the first 50-min period of each day, students attended classrooms that differed from their regular classroom placements and consisted of students grouped together on the basis of reading ability following the Joplin Plan (Floyd, 1954). The Joplin Plan is a between-grade ability grouping procedure for reading instruction. Its effectiveness in increasing reading skill has been demonstrated in numerous studies (for a review, see Slavin, 1987).

Students from Grades 1–3 and 4–6 were grouped by reading ability. Four principles were adhered to in establishing the ability groups: (a) Students would be in their regular classrooms through- out the day and grouped for only the 50-min reading period; (b) students were grouped on the basis of individual reading ability as determined by the previous year’s standardized test scores on reading and by teachers’ judgments of each student’s reading ability; (c) students would be assessed quarterly and reassigned to other reading ability groups, or, if a teacher judged that a place- ment in a reading group was inappropriate, the student could be reassigned prior to the end of a quarter; and (d) teachers were to vary the level and pace of their instruction to match the level of the students within each ability group (Slavin, 1987).

By adopting the Joplin Plan, at least two of the obstacles that teachers had encountered the previous year were addressed. Be- cause students in their regularly assigned classrooms often varied greatly in their reading abilities, the composition of RT groups within those classrooms represented a broad heterogeneity of abilities that some teachers often found led to passiveness in students with lower reading abilities. In some cases, the students in

706 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

a single classroom were separated by 3 years in reading ability. Therefore, in contrast to the recommendations of Palincsar et al. (1989), who recommend heterogeneous groupings, the homoge- neous groupings recommended by the Joplin Plan were hoped to place students in groups that were more closely matched on reading skills. Students in each reading classroom still represented a spectrum of reading abilities, but the spectrum was restricted to approximately a single grade level rather than to multiple grade levels.

The second obstacle that was addressed with the adoption of the Joplin Plan was to provide teachers with methods for formally assessing students’ reading ability. With the regular assessment component of the Joplin Plan, each student would be assessed at least once every quarter of the school year in addition to the standardized testing that every student was required to participate in at the end of the year.

D.J.H. followed the same implementation procedures that had been used for the first and second years of implementation. At the beginning of the academic year, D.J.H. gave a 1-day in-service to teachers on RT, and prior to the in-service, the teachers were asked to read Using Reciprocal Teaching in the Classroom: A Guide for Teachers (Palincsar et al., 1989). During the in-service, D.J.H. and the teachers discussed the RT process, the theory behind RT, and the results from the previous 2 years, emphasizing the obstacles that teachers had encountered and how they had modified RT to fit their personal teaching styles and classroom demands. Again, the modified RT guidelines and the rationales for the modifications were presented and discussed with teachers (see Appendix A).

The 15 teachers who had agreed to use RT devoted their first four 50-min reading periods on 4 consecutive days to explicitly instruct the four RT strategies, with each day devoted to one of the strategies. Once the four strategies had been instructed, D.J.H. visited classrooms to model RT for the students and teachers. D.J.H. used the reading materials that teachers had selected for use in their reading groups. The demonstrations of RT lasted about 50 min.

Similar to the previous year, during the first several months of school, D.J.H. observed the teachers practicing RT. At the end of each observation, D.J.H. and the teachers discussed the difficulties and strengths of RT and agreed on the next steps in implementa- tion. The teachers’ comments again were used as input to the ongoing modification of RT implementation and practice guide- lines (see Appendix B). In late February and early March, D.J.H. scheduled with each teacher a formal observation of her practice of RT. Like the previous year, the primary purpose of the formal observation was to examine how each teacher had uniquely mod- ified RT. No evaluation was made as to the effectiveness of the teachers’ practice of RT. Rather, the focus was on how the teachers had modified RT with respect to the four themes discussed previ- ously (i.e., strategy use, student dialogue, teacher scaffolding, and added elements to RT).

At the end of the school year, D.J.H. administered the same teacher survey that had been administered the previous year, and, this year, students and parents also were surveyed for their im- pressions of the reading program. To measure reading gains across the year, we tested students at the beginning of the year and at the end of the year using in-house measures of reading, and at the end

of the year, students were tested with the reading portions of the CTBS.

Obstacles Encountered

With the greater number of teachers who participated during the 2nd year at School 2, there was a greater range in the kinds of obstacles encountered. Nonetheless, the obstacles were still de- scribable using the themes that had been developed the previous 2 years. The following descriptions of the obstacles are based on data collected during the observations of each teacher and from the teacher surveys.

Strategy use. Some of the teachers reported that using the strategies in RT groups was often problematic. Several of the same problems had been reported the previous year. For example, ques- tioning was at superficial levels. A fourth-grade teacher com- mented, “My students seem to rush through the assignments when they work in small groups. The questions they write are usually very literal.” Also, as noted by a third-grade teacher, there were few clarifications: “I find my students not being able to clarify, and I end up doing all the clarifying for them.” In addition, this year, predictions were noted by a third-grade teacher as problematic: “Predictions are awfully imaginative at this age and not based upon what’s just been read.”

Some teachers also claimed that the use of the four strategies could eventually lead to a monotonous routine of predict, read, clarify, question, and summarize that, as 1 fourth-grade teacher noted, “[S]ometimes my students get bored with.” One of the fifth–sixth-grade teachers believed that once students had learned the RT strategies, other reading activities were necessary to sup- plement the strategies or to combine the RT strategies with other strategies to keep student interest and motivation high: “I would break it up more and do different skills and activities.”

Dialogue. Similar to the findings from the previous year, students’ lack of collaboration skills sometimes hindered RT groups from engaging in meaningful dialogues. One third-grade teacher noted that “I often find them unmotivated and off task,” and a fifth–sixth-grade teacher commented, “I do not believe all my students participate like they should.” Also, in response to a request from D.J.H. to describe those students for whom RT does not work, a fourth–fifth–sixth-grade teacher answered, “Those students who lack cooperative learning skills.”

Moreover, some students assumed passive roles when placed in groups. A third-grade teacher said that, “Some students who seem to be somewhat shy and won’t speak in front of a group do not respond during RT;” another teacher said, “I do RT in my reading class of second and third graders and they get ‘shy’ about asking questions.” Finally, the teacher who taught a fourth–fifth–sixth- grade split commented, “However, my more shy students eventu- ally joined in and became successful with the RT strategy.”

Scaffolding. Similar to the previous year, several of the teach- ers felt a need to maintain greater scaffolding of instruction for longer periods of time than what was recommended by Palincsar et al.’s (1989) guidelines. After 5 months of practicing RT, 5 of the 13 teachers who continued to use RT found that continuing highly scaffolded instruction was necessary for students to effec- tively use the RT strategies. A fifth–sixth-grade teacher felt that, “Some students need more structure,” and that “They [students of

707IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

lower reading ability] still do not quite understand ‘thinking ques- tions.’ It is hard for them to apply concepts.” This was similar to a comment from a third-grade teacher, who noted that without additional help, her students “can look up words in the dictionary but have trouble applying the meaning of the words with the context.” A first-grade teacher commented that without greater direction from the teacher, it was difficult to keep her students attentive. Finally, another fifth–sixth-grade teacher said, “Some students play if they are on their own.” By providing greater assistance to students, this teacher believed that students could be kept on task.

Modifications Made

Of the 15 teachers who began the third year of this study using RT in their classrooms, 13 continued to do so throughout the year. Two teachers discontinued RT and chose other reading instruction for their students. The following descriptions of teachers’ practices of RT are based on data collected during the formal observation of each teacher and from the teacher surveys. During these observa- tions, the teachers practiced versions of RT that could be placed in one of three categories, two of which had appeared the previous year: (a) traditional RT that followed Palincsar et al.’s (1989) guidelines, (b) a modified RT that used RT groups along with whole-class instruction, and (c) a modified RT that used whole- class instruction exclusively (new this year). Because the teachers used RT primarily for the reading period, during which students in Grades 1–3 and 4–6 were grouped by reading ability, the follow- ing practices of RT are described by student reading ability and whether the students were lower or upper elementary.

Traditional RT. Three teachers followed Palincsar et al.’s (1989) guidelines and developed a traditional form of RT in their classrooms (see Table 2). Two of the 3 teachers were upper elementary teachers: One taught a regular education classroom with students of high reading ability, and the other was a special education teacher who taught a resource room with students of low reading ability. The third teacher taught lower elementary students with high reading ability.

The students in the upper elementary, regular education class- room made good use of all four strategies. Small groups of students first made predictions about upcoming text, then ques- tioned and wrote their questions, clarified, and summarized text together in dialogues that were led by a group member. When observed by D.J.H., the students in each group demonstrated good ability to cooperatively engage in dialogue, and the quality of the predictions, questions, answers, and summaries was high. The teacher commented after the observation that, “The groups work well with a little monitoring.”

The seven students in the resource room worked together as a single RT group following a traditional form of RT, with the teacher serving as dialogue leader. Although the teacher provided more direction for her students than a peer group leader likely would have, she engaged all her students in applying each of the strategies to their reading. The teacher added one small modifica- tion to RT by asking the students to read each portion of text twice: The first time students read the text silently, and immediately after, they took turns reading portions of the same text aloud. By reading

twice, the students not only spent more time with the texts, but they also tended to become more confident in their oral reading.

In contrast, the students in the lower elementary classroom made little use of the predicting, clarifying, and summarizing strategies. However, the questioning strategy was used extensively. The teacher closely scaffolded and monitored the students’ use of questioning and answering in their RT groups.

Traditional RT plus whole-class instruction. Three upper ele- mentary and 2 lower elementary teachers used RT in a combina- tion of whole-class instruction and RT groups. One of the upper elementary teachers, who worked with average-ability readers, said, “I use RT both ways [RT groups and whole-class], but I often feel more productive when I use it with the whole group.” This teacher made extensive use of RT groups, but before breaking into groups, she had her students in a whole-class format generate ideas about the topic of the text that was about to be read (see Table 2). Following a discussion of the students’ ideas, the students broke out into their groups, where they read the text aloud, wrote their questions and summaries, and periodically walked to the front blackboard to write the words that they needed clarified. After the students completed the reading, the teacher brought the groups back into a whole-class format so that students could ask their questions of one another, and the teacher could help the students with the words that were written on the blackboard. The period ended with each RT group reading its summary to the class, and the students selected those summaries that they believed were the best representations of the text.

The 2nd of the 3 upper elementary teachers, who also taught average-ability readers, used RT similarly as both a small group and whole-class instruction (see Table 2, Upper elementary, aver- age, starting at Step 2). She liked the RT groups because she believed that “RT is good for student involvement and to facilitate [sic] the concept of ‘student responsibility’ for learning.” Rather than starting her class in whole-class instruction, as did the previ- ous teacher, this teacher began with her students participating in a traditional form of RT in which each RT group read the text, clarified meanings, generated questions, and wrote their predic- tions for the next text section. After the students completed the assigned reading, the teacher brought the groups together in a whole-class format to discuss the content and clarify any words. The teacher augmented students’ questions with her own to check students’ comprehension. This teacher felt that by using the whole- class format in this way, she could “work with students and assess their progression.” Thus, the teacher-generated questions served not only as an informal assessment of students’ comprehension but also as models for students’ future questioning.

The 3rd upper elementary teacher who used RT as both a group and whole-class instruction worked with low-ability readers (see Table 2). She differed from the other 2 in that she only occasion- ally used RT groups and more often used RT in a whole-class format, during which she led the discussions about the text. She felt that the groups were not meeting her expectations because, “My class is so low and reads below grade level, I often find them unmotivated and off task.” She began her whole-class instruction by asking the students to tell in their own words what they had read the previous day. The teacher helped them to elaborate their recall by asking questions, which served as good modeling of questions. Following this, the teacher read the text paragraph by paragraph,

708 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

Table 2 Grades, Ability Levels, and Practices of Teachers Who Used Traditional Reciprocal Teaching (RT) Versus Traditional RT Plus Whole-Class Instruction Versus Whole-Class RT Instruction During Year 3

Traditional RT Traditional RT plus whole-class instruction Whole-class RT

Grade, reading ability Upper elementary, high (1 teacher) Upper elementary, low (1 teacher) Lower elementary, high (1 teacher)

Practices RT groups (group leader)

Take turns reading Strategy use

Questioning Predicting Clarifying Summarizing

Additions to RT Writing questions Read text silently and orally

Grade, reading ability Upper elementary, average (2 teachers) Upper elementary, low (1 teacher) Lower elementary, high (2 teachers)

Practices: Upper elementary, average 1. Whole-class instruction

Introduce text Discuss topic of text

2. RT groups (dialogic) Take turns reading Strategy use

Questioning Predicting Clarifying Summarizing

Write predictions, questions, summaries, and clarifications

3. Whole-class instruction Discussion of text Students ask questions Teacher asks questions Teacher clarifies words Students read summaries

Practices: Upper elementary, low 1. Whole-class instruction

Students review and elaborate on previous day’s reading Teacher reads text Teacher-led strategy use

Questioning Clarifying Predicting: Write predictions

2. RT groups: Occasionally, breakout groups engage in reading texts and using strategies

Practices: Lower elementary, high 1. Whole-class instruction with student pairs

Strategy use Predicting Discuss predictions Teacher writes predictions

2. RT groups (pairs) Take turns reading Strategy use

Clarifying Questioning Summarizing: Write summaries

3. Whole-class instruction: Discussion of summaries

Practices: Lower elementary, high 1. RT groups

Take turns reading Strategy use

Questioning Clarifying Answer teacher-generated questions Predicting

Writing predictions, answers, and words to clarify

2. Whole-class instruction: Discussion of questions– answers, predictions, and clarifiying words

Grade, reading ability Upper elementary, low (1 teacher) Lower elementary, high (1 teacher) Lower elementary, average (1 teacher) Lower elementary, low (2 teachers)

Practices: Upper elementary, low Take turns reading Strategy use

Questioning in class Summarizing in class Individuals summarize entire text and

write summaries

Practices: Lower elementary, high 1. Predicting

Write predictions Discuss predictions

2. Read page silently and generate questions 3. Read page aloud and ask questions 4. Summarize text

Practices: Lower elementary, average 1. Introduce text 2. Predicting

Write predictions Discuss predictions

3. Take turns reading 4. Clarifying text 5. Student pairs generate questions and ask

class to answer 6. Student pairs summarize

Write summaries Read summaries aloud to class

Practices: Lower elementary, low 1. Introduce text 2. Teacher questioning 3. Students predicting on each page of text 4. Teacher reads each page 5. Student questioning on each page with

other students answering 6. Teacher models questions 7. Teacher questioning at end of text 8. Students summarizing text together 9. Students individually write about text

709IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

with the students generating questions and answers for each para- graph. The questioning generated additional discussion of the text, and students had opportunities to clarify unknown words or por- tions of the story. Once finished with a chapter, the teacher asked the class to predict the content of the next chapter. Students wrote their predictions, and the teacher led the students in a discussion of their predictions.

One of the 2 lower elementary teachers who worked with high-reading-ability students used a modified RT that began with students working in RT groups and ended with whole-class in- struction (see Table 2). This teacher was convinced RT was a good approach to reading: “I love it [RT]. I don’t use it all the time, but I think it’s a great strategy.” In their groups, the students took turns reading while using the questioning and clarifying strategies. Be- cause some students had ongoing difficulties generating deeper level questions, the teacher regularly modeled good questioning by displaying on the blackboard a collection of questions that were specific to the assigned reading. After completing the assigned reading, the students answered these questions and wrote their answers. During reading, the teacher encouraged the students to ask any clarifying questions about unknown words or phrases and to write them down. When the reading and questioning were finished, the teacher asked the students to write a prediction for the upcoming part of the text. At the end of the lesson, the teacher brought the students into a whole-class discussion of the questions, their answers, the words or phrases that needed clarification, and their predictions.

The other lower elementary teacher who also worked with high-reading-ability students used a modified RT in which stu- dents worked primarily as pairs who were periodically brought together in a whole-class format (see Table 2). The teacher first prompted the student pairs to look at the title and pictures on the first page and to predict what they believed the story would be about. After the reading pairs discussed their predictions, the teacher asked each pair for a prediction, which she wrote on the blackboard. Students in each pair then took turns reading the first page, during which they clarified with one another any unknown words and generated questions for each other to answer. The teacher monitored each reading pair’s clarifications, questions, and answers. Each page of the text was read following this same procedure. Once the entire story had been read, the teacher di- rected each student pair to summarize the story and to write their summaries. At the end of the lesson, the teacher brought the whole class together and led them in a discussion of the summaries.

Whole-class instruction. Five teachers, 4 lower and 1 upper elementary, developed a version of RT that used whole-class instruction exclusively. The upper elementary teacher, who worked with low-ability readers, began her reading period by asking students as a whole class to take turns reading aloud (see Table 2). She periodically stopped students as they read so that they could generate questions about what they had just read. The teacher then called on other students to answer the questions. The students who read aloud then summarized the text that they had just read. At the end of the reading period, all of the students were asked to individually summarize the entire text that had been read during the period and to write their summaries. After class, the teacher used the students’ summaries to assess their comprehen- sion of the text, and she later gave them feedback on their writing.

Of the 4 lower elementary teachers who used RT exclusively in a whole-class format, 2 developed similar procedures with their students of low reading ability. One of the 2 teachers stated that using RT with the whole-class was better than groups “because students can feel free to respond and not pressured, and they learn from each other as well as from the teacher’s instruction.” Both of these teachers began their reading classes by introducing a story to the students and questioning them about general topics related to the story (see Table 2). One of the 2 teachers wrote some of the students’ ideas on a chart in the front of the room. The teachers then asked their students to predict what they thought the story would be about. After the predictions, the teachers then read the first page aloud and asked students to generate a question about the content. The teachers directed the questions back to other students to answer. Because the students in one class continued to have difficulties generating deeper level questions, the teacher posted on one wall a collection of question stems that the students could use. The other teacher expanded on her RT session by asking students to identify different parts of speech on each page of text. When the entire story was read following these procedures, 1 of the teachers summarized the text and asked her students additional questions necessary to complete the summary. The teacher who had students identify parts of speech ended her lesson also by asking students additional questions; however, she generated the questions by using a game board with a spinner that pointed to specific ques- tions she had written prior to the lesson. Each student took a turn and responded to a question. She then led the class in generating a summary of the story by calling on students to tell in their own words what the story was about. As a final exercise, both teachers incorporated writing into their RT sessions by asking their students to individually write a short story or letter to a friend about a theme that had occurred in the story just read.

The lower elementary teacher who taught students of high reading ability started her whole-class lesson by asking students to make predictions about the text that had been selected for the day (see Table 2). The students wrote their predictions and then dis- cussed them, with the teacher directing the discussion. Once the predictions had been thoroughly discussed, the teacher directed the students to read the first page silently and to generate a question as they read. When everyone finished with the first page, the students took turns reading the same page aloud, after which the students then asked one another the questions they had generated. The next several pages were read following this procedure. The teacher then asked the students to summarize in their own words what the story was about. Once the summaries were shared with one another, the students read the next few pages of the story following the same procedure (i.e., read silently, generate questions, read aloud, ask questions, summarize).

The remaining lower elementary teacher used RT in a whole- class format with her students of average reading ability. She began her class by introducing the text to her students and asking them to write a prediction about what they thought the text would be about (see Table 2). The teacher asked the students to read their predictions aloud, and she led the class in a discussion of the predictions. Students then took turns reading the story aloud, and at the end of each paragraph, the teacher prompted the students to clarify any words they did not understand. Once the reading was completed, the students paired off with one another to write a question about what had just been read, and these questions were

710 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

asked of one another in a whole-class format. After questioning, the teacher asked the student pairs to summarize the text and to write their summaries. The teacher visited with each pair to mon- itor their progress. Once the student pairs finished with their summaries, the teacher had them read the summaries aloud to the class.

Program Impact on Students’ Reading

Throughout the 2nd year at School 2, several sources of data were collected to measure the impact of the reading program on students. We calculated gain scores in reading for each student by pretesting them at the beginning of the academic year and post- testing them at the end of the year using in-house measures that included both a measure of vocabulary knowledge (i.e., the San Diego State College Quick Assessment; LaPray & Ross, 1969) and comprehension (i.e., a cloze procedure from the Harcourt Brace Treasury of Literature Series; Farr, 1995). At the end of the 2nd and 3rd years of implementation, students from Grades 3–5 were tested with the reading portions of the CTBS. Also, students, teachers, and parents–guardians were surveyed for their impres- sions of the reading program.

Gain scores from the in-house measures of reading showed that of the 224 students tested, 8 (4%) realized reading gains that exceeded the upper limits of the tests, 58 (26%) realized gains of 2 or more years, 23 (10%) realized gains between 1 year and 6 months, 74 (33%) realized gains of 1 full year, 30 (13%) realized gains of 6 months, and 31 (14%) experienced no measurable gains. Of the 61 students who realized gains of 6 months or no gains, 15 were far below grade level at the beginning of the school year, 22 were English as a second language students, 6 were resource children, and 18 began the year at or slightly above grade level but still realized no measurable gains during the year. Thus, 73% of the students met or exceeded expectations for reading growth during this year of the program.

Results from the CTBS showed that for 1997–1998, 33% of the students in Grades 3–5 were above the 50th percentile on the reading composite measure, and for 1998–1999, 52% were above the 50th percentile. For 1997–1998, 41% of the students in Grades 3–5 were above the 50th percentile on the language measure, and for 1998–1999, 60% were above the 50th percentile. Thus, over the 2 years at School 2, there was an increase of approximately 20% in the number of students reading above the 50th percentile.

On the student survey, we found that of the 246 students surveyed, 201 (82%) liked the reading program, 210 (85%) thought that the reading program helped them become a better reader, and 173 (70%) reported that they read more books during the year. Using a 10-point scale indicating how they would rate the reading program, with 10 being the highest, 202 (82%) students rated the program 8 or higher.

Of the 166 parents or guardians who responded to our survey, 134 (84%) noticed that their children exhibited an increased inter- est in reading, 139 (87%) thought that their children read more often, 145 (88%) reported that their children talked with them about books, and 154 (96%) felt that the reading program had been beneficial to their children. Some of the parents’ written comments included, “I have noticed a great improvement in [student’s name] reading but I was not aware of a new reading program, but it works,” “I feel that the reading class needs to continue this

summer. I feel that she has done so well, I hate to stop just when she is working so hard,” “He has been reading much better,” “The reading program is very beneficial to [student’s name],” and “I feel this program is a good idea. It helps my child.”

Finally, of the 15 full-time teachers, 11 believed the reading program was beneficial for their students and 3 indicated that it was helpful for some students but not for all, 14 felt that most students increased in reading skills and that grouping on reading ability was beneficial, and 13 were willing to continue the reading program, with 1 indicating that he or she would be willing to continue if certain changes were made.

Discussion

The purpose of this qualitative study was to investigate teachers’ implementation and practice of RT in the classroom. We con- ducted the study over 3 academic years, during which we in- structed teachers to use RT, modeled RT in their classrooms, monitored their classroom practice of RT, consulted with them concerning their practice, and observed and recorded the obstacles they encountered and the modifications they made. We are con- vinced that only by observing teachers over extended periods of time can we come to understand how instruction changes and to know the forces or conditions that motivate those changes. Some teachers practiced RT closely following the guidelines recom- mended by Palincsar et al. (1989); however, a majority of teachers modified the practices embedded in RT, resulting in personalized constructions of RT.

At the outset of this study, we identified three essential elements of RT that likely contribute to its demonstrated effectiveness: instruction and application of the four comprehension-fostering and comprehension-monitoring strategies, using rich and meaning- ful student dialogue as the means to instruct and apply the RT strategies, and providing scaffolded instruction during which the teacher fades his or her modeling of the strategies and relinquishes greater control to the students. Looking across the classroom experiences of 17 teachers, we found numerous ways in which these three elements differed. Moreover, teachers added several new elements to RT, with writing being the most common addi- tion. In the following discussion, we provide a summary of the obstacles that teachers encountered with respect to these elements and the ways teachers modified their practices of RT.

Strategy Use

Of the four RT strategies, questioning played the most dominant role in teachers’ practice of RT. In fact, questioning was the one strategy that was never omitted by teachers, although this may be due, at least in part, to the general prevalence of questioning in the typical classroom. (Gage & Berliner, 1998, estimate that the num- ber of questions asked in primary grades is around 150 questions per hour.) The dominant role that questioning played is also in conformance with findings from Rosenshine and Meister’s (1994) review. In each of the 16 studies of RT that they reviewed, not all four of the RT strategies were evident, but all 16 of the studies did include questioning.

We observed that many students’ questions rarely got beyond literal levels of the text, a finding that was also evident in Marks et al.’s (1993) study in which teachers “found that students had

711IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

difficulties generating anything but literal questions” (p. 278). Therefore, similar to Marks et al.’s study, many teachers in the present study devoted a great deal of time and effort to teaching students how to question. Some teachers modeled questioning during whole-class or group RT sessions. Other teachers provided explicit instruction in questioning. Still other teachers provided question stems to students from which they could select and practice their questioning skills. Finally, 1 teacher had her students prepare questions and answers the night before discussing them in class to help stimulate classroom dialogue.

Of the other three strategies, summarization was the one most often observed, and predicting and clarifying were used sporadi- cally, with clarifying often being omitted. Some teachers tried to encourage students to use clarifying more often by asking them to circle words or sentences that were not understood, another teacher asked her students to write unknown words on the board so that the class could discuss them as part of the reading instruction, and 1 teacher combined the clarifying strategy with the questioning strategy and encouraged her students to generate clarifying ques- tions. Many teachers tried to encourage greater use of these three strategies by having students write their summaries, predictions, and clarifications. In some cases, the summaries and predictions were written collaboratively and then used to facilitate whole-class discussions of the texts.

Researchers and practitioners are in agreement that strategic processing of text is critical to reading comprehension. However, just how each of the RT strategies contributes to reading compre- hension is a question that still needs to be addressed. Rosenshine and Meister (1994) found gains in comprehension when students used 2, 3, 4, or 10 strategies, implying that the kinds and number of strategies are not as important as the kind of cognitive process- ing that is needed to increase comprehension. Perhaps questioning, which is found in all studies of RT as well as many other reading comprehension programs, is the single most important strategy. Or, perhaps the strategies themselves play only a secondary role in improving comprehension, and the primary contributor is in- creased reading time during which readers can engage in multiple readings of a text. Recent studies of rereading have shown that the kinds of reading in which readers engage change with rereading, and with these changes, comprehension can be increased (Mills, Simon, & tenBroek, 1998). Before instructional theory can be advanced, rigorous examinations of how and why the RT strategies improve comprehension need to be conducted.

Dialogue

Palincsar (1986) and others (e.g., Marks et al., 1993) have acknowledged that dialogue plays a critical role in promoting deeper levels of text comprehension. Given the critical role of dialogue, the success of RT, therefore, hinges on the quality of the dialogues that students can achieve. Across the 3 years, however, the most pervasive problem that teachers faced with RT was getting students to learn and use the RT strategies in group dia- logues. In almost all cases, the teachers were convinced that the four strategies could effectively increase reading comprehension, but the main difficulty was using the group dynamic as the me- dium through which to instruct and apply the strategies.

Some RT groups were able to engage in rich discussions of the text, discussions that were spurred by predicting, questioning,

clarifying, and summarizing. However, many RT groups failed to achieve meaningful dialogues. Such difficulties, which stemmed, in part, from students’ poor cooperative learning skills, are not uncommon and have been experienced by many teachers who are interested in using cooperative learning in their classrooms (e.g., Druckman & Bjork, 1994; McCaslin & Good, 1996). Although the potential for group learning is great, recent sources have indicated that cooperative learning “is not a panacea that always provides outcomes superior or even equivalent to those of individual train- ing” (Anderson, Reder, & Simon, 1996, p. 10).

The quality of classroom dialogue can be compromised for a variety of reasons: inadequate knowledge, misunderstandings among participants, widely disparate knowledge of the partici- pants, and lack of discourse skills. Although all of these were likely evident to some degree in the present study, the most salient was students’ lack of discourse skills. Many students had little or poor prior experience working in groups and, therefore, were uncertain about even the basic rules of group discourse. Some teachers directly addressed this inadequacy by teaching their stu- dents how to be good listeners, how to negotiate compromise, how to take turns, and how to give constructive feedback in positive ways.

Social constructivist theories place a great deal of importance on the role of student–student dialogue in learning, and in RT, there is an underlying assumption that by participating in dialogue, students will internalize the four strategies and learn to monitor and regulate their own comprehension. If researchers and practi- tioners are motivated by these theoretical perspectives, then pro- viding training for students in basic discourse skills may be a necessary prerequisite for the success of RT. The social milieu of the classroom is much different from any environment most stu- dents experience. We cannot expect students’ social interactions in the classroom to serve as models for new behaviors when those social interactions are foreign to students and without meaning. For students to move from the interpersonal to the intrapersonal, the interpersonal must first be made meaningful, and this, in some cases, may be the primary objective of classroom dialogue.

Scaffolding

When teachers perceived weaknesses in student dialogue, their response, in general, was to become more directive. Thus, rather than reducing the amount of scaffolding they provided for stu- dents, many teachers maintained highly scaffolded instruction for several months before beginning to reduce the scaffolding and allow groups more independence. Some teachers found it neces- sary to maintain high levels of scaffolding throughout the school year. Overall, the amount of scaffolding and the duration of scaffolding were associated with the age and reading ability of the student, with amount and duration inversely related with age and reading ability. Other strategy instruction research also has indi- cated that extended periods of time are required to ensure the practice and durability of strategy use (e.g., Marks et al., 1993; Pressley & McCormick, 1995).

Teachers’ scaffolding took many forms: whole-class instruction, reading partnerships, explicit modeling of each strategy, extended periods of time to learn the strategies, and direct guidance with basic discourse and group skills. Perhaps even the ability grouping that was used in the reading program could be interpreted as a type

712 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

of scaffolding (i.e., with less heterogeneity in their reading groups, teachers could provide more focused support of activities). By being more directive and using one or more of these approaches, the teachers developed a reading program that in many ways resembled direct explanation (see Duffy & Roehler, 1989; Roehler & Duffy, 1984). Direct explanation shares many commonalities with RT, including the use of strategies and scaffolding of instruc- tion; however, one important difference is the role of the teacher. Direct explanation places the teacher in a more directive role in which he or she explains the strategies, models them, and guides practice until students can demonstrate proficiency with the strat- egies, whereas in RT, the teacher and students develop continuous and mutually responsive interactions (Pressley, Goodchild, Fleet, Zajchowski, & Evans, 1989). Many teachers in the present study believed that by maintaining a more directive role, students en- gaged in dialogues longer and more seriously, were more active in their reading, and were provided with better models for summa- rizing, clarifying, predicting, and particularly questioning.

An important element of RT proposed by Palincsar and Brown (1984) is that students must internalize the four strategies to be able to monitor their strategy use and become better self-regulators of their own reading comprehension. To facilitate students’ inter- nalization of the four strategies, teachers need to reduce their scaffolding and place students in control of the processes they are eventually to self-regulate. A strong constructivist argument could be made that by maintaining the role as cognitive monitor and providing more scaffolding for longer periods, many teachers in the present study were inhibiting students’ self-regulatory pro- cesses necessary for reading comprehension. However, more re- search is showing that metacognitive monitoring and control of strategy use are qualities of the learner that take a great deal of time to develop and are effectively developed through extensive practice of skills that are made visible by teachers thinking aloud or modeling (Pressley & McCormick, 1995). Novice readers do not become expert simply by acquiring more knowledge. The novice becomes expert by acquiring knowledge and restructuring that knowledge through repeated use across a wide variety of contexts, and that takes time. Having this process guided by experts (i.e., the teachers) rather than novices (i.e., the students) may have a stronger impact on learning.

Writing

Nearly all teachers had incorporated the use of writing into their versions of RT, albeit the ways in which they used writing varied widely. Numerous studies have examined the common ground shared by reading and writing processes (e.g., Martin, 1987; Nystrand, 1986; Petrosky, 1982; Tierney & Shanahan, 1991), and there are potentially great benefits to learning by integrating read- ing and writing processes (e.g., Klein, 1999; Schumacher & Nash, 1991; Spivey, 1990). Although there is some disagreement about the ways in which reading and writing overlap, there is a general consensus that both involve cognitive acts of meaning making. Therefore, just as the RT strategies are intended to build and elaborate on text meaning, writing about one’s reading can build and elaborate on the meaning that a person derives from a text.

What must be considered, however, is that not all kinds of writing lead to learning gains (Klein, 1999). Different kinds of writing invoke different kinds of cognitive operations, resulting in

different kinds of learning (Schumacher & Nash, 1991). Therefore, the purpose for writing and the kind of writing must be considered when evaluating writing’s potential for learning. Writing for the purpose of simply expressing one’s thoughts may have positive effects on recall and recognition of those thoughts but modest effects on more complex thinking (e.g., Stotsky, 1995), whereas, writing to compare or contrast views or to analyze a position may have very strong impact on learning at complex levels of thinking (e.g., Langer & Applebee, 1987).

Teachers in the present study had varying purposes for incor- porating writing into their practice of RT. Teachers in the early elementary grades used writing mainly as a way to learn and strengthen basic writing skills. Their writing tasks often had su- perficial ties to the texts and focused more on lexical and syntactic processing of text. One teacher used writing as a means for recording unknown words that would be clarified at a later time. In this case, the teacher’s purpose for writing had little to do with fostering more complex learning. Several teachers had students write their predictions and questions as they were reading, and the predictions and questions were then discussed and answered. The discussions often led to more complex thinking about the text, but whether the activity of writing, per se, added to this is uncertain.

Finally, some teachers in the upper elementary grades had their students write their questions, answers, and summaries. The im- pact on more complex processing of text exerted by answering questions must be considered in terms of the kinds of questions asked. Many student questions touched on superficial levels of the text and could be answered by writing a few words; however, some questions delved into deep levels of understanding, requiring syn- thesis, integration, or evaluation of text information. Similarly, summary writing could lead to more complex processing, depend- ing on the quality of the summaries. Some student summaries were word-for-word renditions of the text and required only copying words. However, many students summarized by selecting main ideas, integrating information from across the text, and creating gist by synthesizing ideas into concise statements. Thus, writing can be a key addition to RT by focusing readers–writers on deeper levels of text.

Conclusion

Teachers’ practice of RT changed with time, and it changed in response to many variables. Some of these changes represented minor variations on traditional RT as proposed by Palincsar and Brown (1984), and some changes represented wide divergences from traditional RT. Within this spectrum of differences, most teachers continued to place strong emphasis on strategic process- ing of text, and the results at School 2 at the end of Year 3 suggest that the strategic processing that was emphasized had a positive impact on students. In the present study, we did not attempt to determine what portion of these positive results could be attributed to RT, individual teachers, student differences, or the Joplin Plan. Like most pedagogy, a likely scenario is that increased learning occurred as a complex interaction among teacher, student, and instructional variables. Future research needs to examine this com- plex dynamic of RT.

Finally, an implication that we hope to draw from this study is that researchers and educators who advocate new programs must be aware of the ways in which programs change with each teacher

713IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

as he or she works to construct a new practice. Future attempts to disseminate RT may meet with even greater success if teachers are first given adequate training in traditional RT, are encouraged to construct their practice of RT using their prior knowledge about practice, and are provided with sustained feedback on their prac- tice throughout the construction process from people knowledge- able of RT and the schools in which they work.

References

Almasi, J. F. (1995). The nature of fourth graders’ sociocognitive conflicts in peer-led and teacher-led discussions of literature. Reading Research Quarterly, 30, 314–351.

Anderson, J. R., Reder, L. M., & Simon, H. A. (1996). Situated learning and education. Educational Researcher, 25, 5–11.

Borko, H., & Putnam, R. T. (1996). Learning to teach. In D. C. Berliner & R. C. Calfree (Eds.), Handbook of educational psychology (pp. 673– 708). New York: Simon & Schuster.

Brown, A. (1997). Transforming schools into communities of thinking and learning about serious matters. American Psychologist, 4, 399–413.

Brown, A. L., & Palincsar, A. S. (1989). Guided cooperative learning and individual knowledge acquisition. In L. B. Resnick (Ed.), Cognition and instruction: Issues and agendas (pp. 393–451). Hillsdale, NJ: Erlbaum.

CTBS. (1997). Comprehensive Test of Basic Skills. Monterey, CA: CTB/ McGraw-Hill.

Denzin, N. K., & Lincoln, Y. S. (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage.

Druckman, D., & Bjork, R. A. (Eds.). (1994). Learning, remembering, believing: Enhancing team and individual performance. Washington, DC: National Academy Press.

Duffy, G. G. (1993). Teachers’ progress toward becoming expert strategy teachers. The Elementary School Journal, 94, 109–120.

Duffy, G. G., & Roehler, L. R. (1989). Why strategy instruction is so difficult and what we need to do about it. In C. B. McCormick, G. Miller, & M. Pressley (Eds.), Cognitive strategy research: From basic research to educational applications (pp. 133–154). New York: Springer-Verlag.

Dunbar, K. (1997). How scientists think: On-line creativity and conceptual change in science. In T. B. Ward, S. M. Smith, & J. Vaid (Eds.), Creative thought: An investigation of conceptual structures and processes (pp. 461–493). Washington, DC: American Psychological Association.

El-Dinary, P. B., & Schuder, T. (1993). Seven teachers’ acceptance of transactional strategies instruction during their first year using it. The Elementary School Journal, 94, 207–219.

Farr, R. C. (1995). Individual inventory for reading and writing. San Diego, CA: Harcourt Brace.

Floyd, C. (1954). Meeting children’s reading needs in the middle grades: A preliminary report. The Elementary School Journal, 55, 99–103.

Gage, N. L., & Berliner, D. C. (1998). Educational psychology (6th ed.). New York: Houghton Mifflin.

Gall, J. P., Gall, M. D., & Borg, W. R. (1999). Applying educational research. New York: Longman.

Hart, E. R., & Speece, D. L. (1998). Reciprocal teaching goes to college: Effects for postsecondary students at risk for academic failure. Journal of Educational Psychology, 90, 670–681.

Hiebert, E. H., & Raphael, T. E. (1996). Psychological perspectives on literacy and extensions to education practice. In D. C. Berliner & R. C. Calfree (Eds.), Handbook of educational psychology (pp. 550–602). New York: Simon & Schuster.

Kintsch, W. (1998). Comprehension: A paradigm for cognition. New York: Cambridge University Press.

Klein, P. D. (1999). Reopening inquiry into cognitive processes in writing- to-learn. Educational Psychology Review, 11, 203–270.

Langer, J. A., & Applebee, A. N. (1987). How writing shapes thinking: A study of teaching and learning. Urbana, IL: National Council of Teach- ers of English.

LaPray, M., & Ross, R. (1969). The graded word list: Quick gauge of reading ability. Journal of Reading, 12, 305–307.

Lysynchuk, L. M., Pressley, M., & Vye, N. J. (1990). Reciprocal teaching improves standardized rewarding-comprehension performance in poor comprehenders. The Elementary School Journal, 90, 469–489.

Marks, M., Pressley, M., Coley, J. D., Craig, S., Gardner, R., DePinto, W., & Rose, W. (1993). Three teachers’ adaptations of reciprocal teaching in comparison to traditional reciprocal teaching. The Elementary School Journal, 94, 267–283.

Martin, S. (1987, December). The meaning-making strategies reported by provident readers and writers. Paper presented at the meeting of the National Reading Conference, St. Petersburg, FL.

McCaslin, M., & Good, T. L. (1996). The informal curriculum. In D. C. Berliner & R. C. Calfree (Eds.), Handbook of educational psychology (pp. 622–673). New York: Simon & Schuster.

Mills, K. K., Simon, S., & tenBroek, N. S. (1998). Resource allocation during the rereading of scientific texts. Memory & Cognition, 26, 232– 246.

Nystrand, M. (1986). The structure of written communication: Studies in reciprocity between writers and readers. Orlando, FL: Academic Press.

Palincsar, A. S. (1986). The role of dialogue in providing scaffolded instruction. Educational Psychologist, 21, 73–98.

Palincsar, A. S., & Brown, A. L. (1984). Reciprocal teaching of comprehension-fostering and comprehension-monitoring activities. Cognition and Instruction, 1, 117–175.

Palincsar, A. S., David, Y. M., & Brown, A. L. (1989). Using reciprocal teaching in the classroom: A guide for teachers. Unpublished manual.

Palincsar, A. S., David, Y. M., Winn, J. A., & Stevens, D. D. (1991). Examining the context of strategy instruction. Remedial and Special Education, 12, 43–53.

Palincsar, A. S., & Klenk, L. (1992). Fostering literacy learning in sup- portive contexts. Journal of Learning Disabilities, 25, 211–225.

Petrosky, A. (1982). From story to essay: Reading and writing. College Composition and Communication, 33, 19–35.

Pressley, M., Goodchild, F., Fleet, J., Zajchowski, R., & Evans, E. D. (1989). The challenges of classroom strategy instruction. The Elemen- tary School Journal, 89, 301–342.

Pressley, M., & McCormick, C. B. (1995). Advanced educational psychol- ogy for educators, researchers, and policymakers. New York: Harper- Collins.

Richardson, V. (1996). The role of attitudes and beliefs in learning to teach. In J. Sikula (Ed.), Handbook of research on teacher education (2nd ed., pp. 102–119). New York: Simon & Schuster.

Roehler, L. R., & Duffy, G. G. (1984). Direct explanation of comprehen- sion processes. In G. G. Duffy, L. R. Roehler, & J. Mason (Eds.), Comprehension instruction: Perspectives and suggestions (pp. 265– 280). New York: Longman.

Rosenshine, B., & Meister, C. (1994). Reciprocal teaching: A review of the research. Review of Educational Research, 64, 479–530.

Rosenshine, B., Meister, C., & Chapman, S. (1996). Teaching students to generate questions: A review of the intervention studies. Review of Educational Research, 66, 181–221.

Schumacher, G. M., & Nash, J. G. (1991). Conceptualizing and measuring knowledge change due to writing. Research in the Teaching of En- glish, 25, 67–96.

Shuell, T. J. (1996). Teaching and learning in a classroom context. In D. C. Berliner & R. C. Calfree (Eds.), Handbook of educational psychology (pp. 726–764). New York: Simon & Schuster.

Slavin, R. E. (1987). Ability grouping and student achievement in elemen- tary schools: A best-evidence synthesis. Review of Educational Re- search, 57, 293–336.

714 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

Spivey, N. N. (1990). Transforming texts: Constructive processes in read- ing and writing. Written Communication, 7, 256–287.

Stotsky, S. (1995). The uses and limits of personal or personalized writing in writing theory, research, and instruction. Reading Research Quar- terly, 30, 758–776.

Strauss, A., & Corbin, J. (1994). Grounded theory methodology. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 273–285). Thousand Oaks, CA: Sage.

Tierney, R. J., & Shanahan, T. (1991). Research on the reading–writing

relationship: Interactions, transactions, and outcomes. In R. Barr, M. L. Kamil, P. B. Mosenthal, & P. D. Pearson (Eds.), Handbook of reading research, Vol. II (pp. 246–280). New York: Longman.

Vygotsky, L. S. (1978). Mind in society. Cambridge, MA: Harvard Uni- versity Press.

Wolcott, H. F. (1988). Ethnographic research in education. In R. M. Jaeger (Ed.), Complementary methods for research in education (pp. 187–249). Washington, DC: American Educational Research Association.

Appendix A

Starting Reciprocal Teaching (RT) in Your Classroom: Program Implementation Guidelines

Step 1.0: Establishing a foundation for RT. Teachers must read Using Reciprocal Teaching in the Classroom: A

Guide for Teachers (Palincsar, David, & Brown, 1989). Step 2.0: Whole-class instruction of the four RT strategies.

Palincsar et al. (1989) recommends that at least 25 class sessions be allocated to teach RT and that the first 5 days be spent introducing the four strategies. Therefore, a class time of approximately 40 min should be reserved on 5 consecutive school days to instruct the strategies, one strategy per day, with a review of all four on the 5th day. The following 20 class sessions do not necessarily need to be placed consecutively (Palincsar et al., 1989). However, throughout those 20 days, periodic reinstruction of the strategies should take place.

Instruction of the four strategies can be done using whole-class instruc- tion or groups, whatever format is working best in your classroom at present. Be sure to prominently display in the classroom the strategies (e.g., posters on the wall or hanging from the ceiling). For suggestions on how to instruct the strategies, see pages 7–8 and Appendix A (pp. 30–35) of Guide for Teachers. My suggestion for their presentation is predicting first, then questioning, followed by summarizing, and ending with clarifying. On the 5th day, a short text or parts of a text should be used by the teacher to demonstrate how to apply the strategies to it. Step 3.0: Forming RT groups.

The optimal size for the RT groups is about six students. The groups should represent a range of reading abilities (i.e., good readers, average readers, and poor readers). Of course, if behavioral problems arise within a group, members should be shifted to other groups. Other than that, however, membership in each group should remain fairly constant until the students become proficient with the strategies. Once they become profi- cient, changing the makeup of groups will likely further enhance learning and use of the strategies.

Also, within each group, form reading partners. Pair up each child with a reading partner of similar reading ability. As with the groups, reading partnerships should remain fairly constant until the students become pro- ficient with RT. Step 4.0: Getting the RT groups started.

Research of RT has shown that often teachers have difficulty getting five or six RT groups started at the same time. Many demands are placed on teachers as they attempt to attend to each group to monitor and direct progress. And, if students have not yet acquired the skills necessary for group work, there is little motivation for them to spontaneously engage in the RT sessions. Therefore, it may be necessary for teachers to take a more directive role in getting the RT groups started. Once students become more familiar with RT and the teacher’s expectations, teachers can begin to

decrease their involvement and allow students to assume greater control over the groups. Having students assume greater control of their reading is a critical component of RT.

On the 6th day of instruction, after the four strategies have been taught, have the students assemble in their RT groups. Select for this day a short text (one or two pages) that is related to a topic currently being covered in the class. It may be best to work initially with narrative text to help maintain student motivation; however, keep in mind that RT eventually should be used with all kinds of text. Also, have the story sectioned into four or five parts, each part having a discrete meaning contained within it.

4.1: Teachers introduce the story and describe how the story ties in with material already being covered in class.

4.2: Talk about the title and illustrations, if any accompany the story, and ask students to predict what they think the story will be about. Encour- aging predictions from all the groups will help to engage all students. Once students’ predictions have been exhausted, teachers should encourage students to look for clues as they read the story as to what the story is about and whether their initial predictions were correct.

4.3: Direct the students to form their reading partnerships and have them begin to read the story. For each section of the text, one partner should read and the other should write one question about that section. The reading partners then should switch off reading and writing tasks for each subse- quent section of the text. The reading partners also should be told to circle any words or any sentences that they do not understand so that they can be clarified.

4.4: After the first section of the text has been read and a question generated, the students should assemble into their RT groups to answer the questions that had been generated by their group. After RT groups have had sufficient time to answer their questions, the teacher should ask a group for one of its questions and direct that question to another group. The teacher may wish to confer with the RT group who generated the question to determine whether the question was answered correctly. This questioning activity should continue until all groups have had a chance to respond to at least one question. This is also a good opportunity for the teacher to clarify any student misunderstandings of words or sentences.

4.5: Once questioning of the first section has been exhausted, the teacher should ask all students to help summarize the first section. Calling on each group to participate, the teacher can then write the collectively produced summary on the board.

4.6: Before the students reconvene in their reading partnerships, the teacher should ask them to predict what will happen next in the text based on what they have already read in the first section. Calling on each group will help to encourage engagement.

(Appendixes continue)

715IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

4.7: The reading partners once again get together to read the next section of the text. The same procedure can then be followed for each section. At the end, the teacher can have students read the entire summary, clarify any lingering misunderstandings, and revise the summary if needed. Step 5.0: Getting the RT groups working independently.

Always keep in mind that the goal of reciprocal teaching is to get students reading independently in their groups and then to get them reading independently by themselves. This highly directed instruction described so far should be continued only for as long as it is necessary to get the students familiarized with the RT method. As groups gain greater and greater proficiency with the RT method, less and less direction needs to come from the teacher.

How much independence the students can assume and how fast they can assume it will vary by grade and by classroom. Rather than simply handing the task over to students, what might work better is to gradually remove the directed role of the teacher. Start by having each RT group discuss collaboratively a summary for each section. Once they have discussed how best to summarize each section, the group members should produce a clear and concise summary, which should be written by one of the group members. It is important that all students share the responsibility for writing the summaries during the class period. Perhaps a routine can be

established at the beginning of each class period in which students rotate responsibility for each section.

The teacher can hand over further responsibility to students by assigning an RT group leader who will assume the role that has been modeled by the teacher. The group leaders can direct their groups through predicting, clarifying, questioning, and summarizing just the way that the teacher did with the entire class. Because it is important that all students assume the role of group leader, the teacher needs to consider how the position can be rotated regularly. Once RT groups are working independently, the role of the teacher becomes more of a facilitator who goes from group to group helping to clarify misunderstandings or new concepts and to make sure the groups are running smoothly. Step 6.0: Assessment of comprehension.

After a text has been completely read and thoroughly discussed, the teacher can assess students’ comprehension of it by asking them to provide answers to higher level comprehension questions. The members of each RT group can collaborate to generate their best answers to each. The responses can be written by each group and handed in for teacher evaluation. After all groups have had an opportunity to respond to each question and have written their answers, the teacher may wish to discuss their answers in a whole-class setting.

Appendix B

Starting Reciprocal Teaching (RT) in Your Classroom: Revised Program Implementation Guidelines

Step 1.0: Establishing a foundation for RT. Teachers must read Using Reciprocal Teaching in the Classroom: A

Guide for Teachers (Palincsar, David, & Brown, 1989). Step 2.0: Whole-class or group instruction of the four RT strategies.

Palincsar et al. (1989) recommends that at least 25 class sessions be allocated to teach RT and that the first 5 days be spent introducing the four strategies. Therefore, a class time of approximately 40 min should be reserved on 5 consecutive school days to instruct the strategies, one strategy per day, with a review of all four on the 5th day. The following 20 class sessions do not necessarily need to be placed consecutively (Palincsar et al., 1989). However, throughout those 20 days, periodic reinstruction of the strategies should take place.

Instruction of the four strategies can be done using whole-class instruc- tion or groups, whatever format is working best in your classroom. Be sure to prominently display in the classroom the strategies (e.g., posters on the wall or hanging from the ceiling). For suggestions on how to instruct the strategies, see pages 7–8 and Appendix A (pp. 30–35) of Guide for Teachers. My suggestion for their presentation is predicting first, then questioning, followed by summarizing, and ending with clarifying. On the 5th day, a short text or parts of a text should be used by the teacher to demonstrate how to apply the strategies to it. Step 3.0: Forming RT groups.

The optimal size for the RT groups is about six students. The groups should represent a range of reading abilities (i.e., good readers, average readers, and poor readers). However, keep in mind that if the differences in reading ability are grossly exaggerated (e.g., 3 or more years), the students of low reading ability may become passive in their interactions or have difficulties interacting with the students of high reading ability. If behav- ioral problems arise within a group, members should be shifted to other groups. Other than that, however, membership in each group should remain fairly constant until the students become proficient with the strategies. Once they become proficient, changing the makeup of groups will likely further enhance learning and use of the strategies. Step 4.0: Getting the RT groups started.

Research of RT has shown that often teachers have difficulty getting five or six RT groups started at the same time. Many demands are placed on teachers as they attempt to attend to each group to monitor and direct progress. And, if students have not yet acquired the skills necessary for group work, there is little motivation for them to spontaneously engage in the RT sessions. Therefore, it may be necessary for teachers to take a more directive role in getting the RT groups started and to provide highly scaffolded RT instruction for extended periods of time. Depending on the group skills of your students and how thoroughly they learn the RT strategies, getting your RT groups working effectively could take as long as 4 months. Once students become familiar with RT and the teacher’s expectations, teachers can begin to decrease their involvement and allow students to assume greater control over the groups. Having students assume greater control of their reading is a critical component of RT.

Your implementation and practice of RT may follow one of three trajectories, varying according to the level of teacher directiveness and amount of scaffolding. None of these trajectories should be considered as rigid and inflexible. Rather, teachers may develop their own trajectories that are variations of these three. In deciding which trajectory to follow, you need to consider two components: ability of students to participate in group process and ability of students to effectively practice the four strategies. If your students have developed good group skills (e.g., good listening skills, ability to give constructive feedback and to negotiate compromise) and have learned the four strategies, then adopting a more traditional approach to RT, as described in the Palincsar et al. (1989) guidelines, may be the best trajectory to follow (Trajectory A). If either one or both of these two components are weak, then a more directive approach may be considered in which you provide more highly scaffolded instruc- tion in a combination of whole-class instruction and RT groups (Trajectory B). If both components are considerably weak, you may wish to consider an even more directive and more highly scaffolded approach in which only whole-class instruction is used (Trajectory C).

Trajectory A: Follow the Palincsar et al. (1989) guidelines.

716 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

Trajectory B: The following steps describe a general RT procedure that is more directive in nature than traditional RT and is designed for use with a combination of whole-class and group instruction.

B.1: Teachers introduce the text to the class and describe how the text ties in with material already being covered in class.

B.2: Talk about the title and illustrations, if any accompany the text, and ask students to predict what they think the text will be about. Encouraging predictions from all the RT groups will help to engage all students. Teachers may want to model a few predictions and talk out for students the process of predicting and checking the predictions. Once students’ predic- tions have been exhausted, teachers may wish to encourage students to look for clues as they read the text as to what the text is about and whether their initial predictions were correct.

B.3: Direct the RT groups to begin to read the first section of the text in their groups. Have the students in each group circle any words or any sentences that they do not understand so that the words or sentences can be clarified.

B.4: After the first section of the text has been read, each student within a group generates a question, and the students within the group answer each. After RT groups have had sufficient time to answer their questions, the teacher should ask a group for one of its questions and direct that question to another group. The teacher may wish to confer with the RT group that generated the question to determine whether the question was answered correctly. Once again, the teacher may wish to model good questioning and talk out for students the process of questioning. The questioning activity should continue until all groups have had a chance to respond to at least one question. This is also a good opportunity for the teacher to clarify any student misunderstandings of words or sentences, or the teacher could ask students in the classroom to clarify.

B.5: Once questioning of the first section has been exhausted, the teacher then asks each RT group to summarize the first section. After sufficient time has passed for the RT groups to summarize, the teacher can call on each group to contribute to a collective summary of the text section just read. At this point, the teacher can further model summarizing by talking out for students the summarization process.

B.6: The teacher then asks the RT groups to predict what will happen next in the text based on what they have already read in the first section. After sufficient time has passed for the RT groups to predict, the teacher can again call on each group to contribute to a class discussion of their predictions.

B.7: The same procedure can then be followed for each section. At the end of the text, the teacher can have students read their summaries, generate an overall summary for the entire text, and clarify any lingering misunderstandings.

B.8: Once students demonstrate proficiency with the combination of whole-class and group instruction, the teacher may wish to move instruc- tion along the path described by Trajectory A.

Trajectory C: The following steps describe a general RT procedure that is more directive in nature than Trajectory B and is designed to provide highly scaffolded instruction through a whole-class format. This more directive trajectory may be helpful if the students need more extensive training with basic group discourse skills and with the four RT strategies.

C.1: Along with the strategy training that occurs during the 20 days after initially instructing students about the RT strategies (see Step 2.0), teachers will also instruct basic group discussion skills. How to be a good listener, how to negotiate compromise, and how to give constructive feedback are just a few of the main topics students should learn and practice.

C.2: Because students tend to remain at superficial levels of questioning, they may also need additional help with the questioning strategy. A strong

focus on different kinds of questions and the kinds of text processing necessary to answer each kind of question may be necessary. Providing students with question stems or questioning strategies may be helpful. The teacher may wish to explicitly model good questioning and talk out for students the process of questioning. Spending time and effort on good modeling of questioning and on practice with questioning will be time and effort well spent.

C.3: Teachers introduce the text and describe how the text ties in with material already being covered in class.

C.4: Talk about the title and illustrations, if any accompany the text, and ask students to predict what they think the text will be about. Encouraging predictions from all the students will help to engage them all in the task. Teachers may want to model many predictions and talk out for students the process of predicting and checking the predictions. Once students’ predic- tions have been exhausted, teachers should encourage students to look for clues as they read the text as to what the text is about and whether their initial predictions were correct. More explicit discussion of what kinds of clues to look for and where to find the clues may be helpful.

C.5: Direct the students to read the first section of the text. Have students take turns reading orally. Also, have the students circle any words or any sentences that they do not understand so that the words or sentences can be clarified.

C.6: After the first section of the text has been read, each student generates a question, and the teacher or student who generates the question should direct the question to another student to answer. The teacher could use this question generation period as a way to further model good questioning and to emphasize to the students the qualities and character- istics of good questions. More explicit instruction and modeling of ques- tioning again may be helpful. This is also a good opportunity for the teacher to clarify any student misunderstandings of words or sentences, or the teacher could ask students in the classroom to clarify.

C.7: Once questioning of the first section has been exhausted, the teacher should ask the students to summarize the first section. The teacher can work with all the students to generate a classroom summary of the text section just read. The teacher can explicitly demonstrate how in writing a summary that only the most important elements from the text are selected. If the teacher writes the collective summary on the board or overhead, the students can see how the summary evolves and changes with students’ contributions.

C.8: With the summary written and with general agreement from the students that the summary accurately represents the text just read, the teacher should ask the students to predict what will happen next in the text based on what they have already read in the first section.

C.9: The same procedure can then be followed for each section. At the end of the text, the teacher can have students generate an overall summary for the entire text and clarify any lingering misunderstandings.

C.10: Once students demonstrate proficiency with the RT strategies in the whole-class format, the teacher can then move instruction along the path described by Trajectory B. Or, if a great deal of progress with the RT strategies and discourse skills is demonstrated, perhaps the teacher could move instruction along Trajectory A. Step 5.0: Getting the RT groups working independently.

Always keep in mind that the goal of RT is to get students reading independently in their groups and then to get them reading independently by themselves. How much independence the students can assume and how fast they can assume it will vary by grade and by classroom. Trajectories B and C are highly directed instruction and should be continued only for as long as it is necessary for the students to learn the RT strategies and group

(Appendixes continue)

717IMPLEMENTING RECIPROCAL TEACHING

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

discussion skills. As students gain greater proficiency with the RT method and with the give-and-take of working with one another in groups, teachers can begin to reduce their levels of scaffolding, and students can assume greater control in their reading. The role of the teacher at this point becomes more of a facilitator who goes from group to group helping to clarify misunderstandings or new concepts and to make sure the groups are running smoothly. Step 6.0: Use of writing.

Introducing writing into RT can be a great way to help students practice basic writing skills and/or acquire complex levels of learning from text, depending on the purpose for writing and the kind of writing. Learning and strengthening basic writing skills, which may be an objective for the lower rather than upper elementary grades, could be introduced in every step of RT: writing words that need clarification, writing questions, predictions, and summaries. The teacher could provide editorial feedback on each written component. More complex learning could be fostered by asking students to write their answers to the questions, provided the questions are asking for deeper levels of comprehension, or by writing their summaries of each text section and the text as a whole. Discussing the answers and

summaries and then revising them could add considerably to student learning. Step 7.0: Assessment of comprehension.

After a text has been completely read and thoroughly discussed, the teacher can assess students’ comprehension of it by asking them to provide answers to higher level comprehension questions. The members of each RT group can collaborate to generate their best answers to each, and the responses can be written by each group and handed in for teacher evalu- ation. Or, students could provide their own individual answers. After all groups have had an opportunity to respond to each question and have written their answers, the teacher may wish to discuss their answers in a whole-class setting. Providing regularly scheduled formal and informal assessments can be helpful for the student to assess his or her reading progress and for the teacher to assess his or her teaching effectiveness.

Received September 5, 2001 Revision received February 14, 2002

Accepted February 16, 2002 

718 HACKER AND TENENT

Th is

d oc

um en

t i s c

op yr

ig ht

ed b

y th

e A

m er

ic an

P sy

ch ol

og ic

al A

ss oc

ia tio

n or

o ne

o f i

ts a

lli ed

p ub

lis he

rs .

Th is

a rti

cl e

is in

te nd

ed so

le ly

fo r t

he p

er so

na l u

se o

f t he

in di

vi du

al u

se r a

nd is

n ot

to b

e di

ss em

in at

ed b

ro ad

ly .

No hay comentarios
Esta solo es una vista previa
3 páginas mostradas de 20 páginas totales
Descarga el documento