Noreen M Webb. 21st Century Education: A Reference Handbook. Editor: Thomas L Good. Sage Publication. 2008.
It is hard to exaggerate the interest in small-group learning in today’s schools. Recognizing that students can learn by working with and helping each other, school districts, state departments of education, national research organizations, and curriculum specialists recommend, and even mandate, the use of peer-based learning. Teachers can choose from many commercially available guides and programs designed to help them plan, implement, and manage small-group learning. Whole schools have even been organized around students cooperating with each other as the primary mode of instruction. Under-girding these activities is a large body of research that shows the positive effects of small-group methods on student achievement, especially compared to other forms of instruction that involve less interaction between students (e.g., Slavin, 1995).
In spite of the potential for small-group work to foster student learning, most researchers agree that simply placing students in small groups does not guarantee that learning will occur. Rather, the extent to which students benefit from working with other students depends on the group dynamics and the nature of students’ participation in group work. This chapter explores how group work is thought to benefit student learning and the many ways in which teachers might orchestrate small-group work to achieve those benefits. It does not catalog differences among small-group approaches, which often have such labels as cooperative learning, collaborative learning, peer tutoring, and peer-based or peer-directed learning, but considers all small-group contexts in which students are encouraged, expected, or required to work with each other to improve student learning.
Mechanisms for Learning from Peers
The view that cognitive conflict leads to higher levels of reasoning and learning derives principally from the work of Piaget (1932) and his followers. Cognitive conflict arises when a learner perceives a contradiction between her or his existing understanding and what the learner hears or sees in the course of interacting with others. This contradiction and its disequilibrating effect lead the learner to reexamine and question his or her own ideas and beliefs, to seek additional information in order to reconcile the conflicting viewpoints, and to try new ideas. Piaget suggested that exchanges with peers are more likely than exchanges with adults to promote children’s cognitive development because children are likely to cooperate as equals, to speak at a level that other students understand, to challenge each other, and to share each other’s point of view.
Numerous studies have examined the cognitive development of pairs of children working on conservation tasks that ask them to discuss whether some characteristics of objects (e.g., the volume of liquid) remain the same when others (e.g., the width or height of the container) change (De Lisi & Golbeck, 1999). When children who have not yet learned the principle of conservation are paired with children who have mastered it, the former often gain conservation knowledge whereas the latter rarely regress to giving incorrect responses. Students with and students without a knowledge of conservation have been shown to behave in different ways, with conservers asserting and justifying their answers, as well as producing counterarguments, and non-conservers restating their original responses without any elaboration. Even non-conservers working together progress when they hold conflicting, but still incorrect, ideas, but not when they hold the same incorrect conception.
Vygotsky (1978) posited that the benefits of collaboration occur when a more expert person helps a less expert person. With the help of a more skilled person, a process of negotiation and transformation enables the less competent person to carry out a task that the student could not perform without assistance, a process sometimes called scaffolding or guided participation. Through practice and intemalization, new skills and knowledge become part of the student’s individual repertoire (Tudge, 1990).
In peer-directed settings, students can often provide effective scaffolding and support due to their understanding of each other’s misunderstandings and their ability to explain concepts in familiar terms (Noddings, 1985). Students may be especially able to provide each other assistance that is at the right level, rather than assistance that is beyond their teammates’ understanding or assistance that is not necessary.
Effective scaffolding requires several conditions. Help provided must be relevant to the less proficient student’s need for help, and it must be timely, correct, comprehensible, and at the level of help needed. Further, students receiving help must have the opportunity to use explanations to solve the problem or carry out the tasks and must use that opportunity for practice by attempting to apply the explanations received to the problem at hand (Vedder, 1985). Students must both have and take advantage of the opportunity to apply the information contained in the explanation. Empirical studies of the relationship between these behaviors and achievement support the power of the latter conditions. Receiving explanations has sometimes been found to be more strongly related to achievement than is receiving less elaborated responses, such as the answer to a problem without suggestions about how to solve it, or receiving no help at all; however, a stronger predictor of learning outcomes is whether students use the help they receive to solve problems on their own without assistance (Webb & Palincsar, 1996). This relationship is consistent with the Vygotskian notion that the process of learning is more than simply the transfer of knowledge from expert to novice. This perspective holds that the active participation of the learner is critical; the more competent partner helps the less competent partner not only by informing, correcting, and explaining, but also by forcing the latter partner to explain (Tudge, 1990).
Co-Construction of Knowledge
Students can also learn by co-constructing knowledge with their peers. Hatano (1993) showed how students can contribute different pieces of information or build upon others’ explanations to jointly create a complete idea or solution. Researchers have documented how students can collaboratively build knowledge and problem-solving strategies that no group member has at the start by acknowledging, clarifying, correcting, adding to, building upon, and connecting each other’s ideas and suggestions (e.g., Hogan, Nastasi, & Pressley, 2000). Such interactions help children co-construct and internalize strategies and concepts in many areas including identifying chemicals in chemical reactions, understanding of place value, learning principles of transformational geometry, learning how to multiply whole numbers, and so forth.
Co-construction requires a high degree of coordination among group members. Barron (2003) noted that highly coordinated groups acknowledge each other’s ideas, repeat others’ suggestions, and elaborate on others’ proposals. Speakers’ turns are tightly connected, with group members paying close attention to and responding to what other members do and say, giving space for others’ contributions, and monitoring how the unfolding contributions relate to the problem-solving goal. Proposals are directly linked to the prior conversation and are acknowledged and discussed, not rejected or ignored. Interaction in such groups is marked by a high degree of joint attention and respect. According to Barron, highly coordinated groups simultaneously and successfully negotiate a dual-problem space consisting of a content space and a relational space.
Cognitive Elaboration Perspective
From a cognitive elaboration perspective, interacting with others may encourage students to engage in cognitive restructuring, by which they restructure their own knowledge and understanding. This occurs when students elaborate on their thinking during conversations with others. Specifically, explaining the material to others may promote learning by encouraging students to rehearse information, reorganize and clarify material, recognize their own misconceptions, fill gaps in their own understanding, strengthen connections between new information and previously learned information, internalize and acquire new strategies and knowledge, and develop new perspectives and understanding (Bargh & Schul, 1980). In the process of formulating an explanation, students may think about the salient features of the problem and generate self-explanations that help them to internalize principles, construct specific inference rules for solving the problem, and repair imperfect mental models (Chi, 2000). This process may help them develop a better awareness of what they do and do not understand. In addition, tailoring explanations to the difficulties of other students may push helpers to construct more elaborate conceptualizations than they would otherwise. Giving non-elaborated help (such as the answer to a problem without any accompanying explanation of how to solve it), on the other hand, is expected to have fewer benefits because it may involve less cognitive restructuring by the helper.
Research documents the strong relationship between giving explanations and achievement in small groups (Webb & Palincsar, 1996). Research on peer tutoring also demonstrates the effectiveness of giving explanations, with tutors who give more elaborate explanations (e.g., describing the relationship between the current problem and real-life examples; showing connections between ideas in the material being discussed) learning more than tutors who give less elaborate explanations (King, 1999).
The cognitive elaboration perspective is not necessarily independent of the Piagetian and Vygotskian perspectives, or of the co-construction model, but, instead, may be an integral part of them. In Piaget’s model of sociocognitive conflict and learning, confronting others’ contradictory ideas may involve explaining and justifying one’s own position, with the potentially positive effects on learning described above. Similarly, the foregoing description of scaffolding in the Vygotskian perspective has already highlighted the active role that the less competent student plays. Specifically, explaining one’s own thinking and understanding constitutes part of the process through which the less capable person constructs his or her knowledge. And during co-construction, students may justify their own ideas and elaborate on others’ ideas, with new knowledge and understanding resulting.
According to the motivational theory of Deutsch (1949), when groups work toward a common goal, students will praise, encourage, and support each other’s contributions, resulting in more effort, a greater liking of the task and of other students. In a cooperative goal structure, group members can attain their own personal goals only if the group is successful. One way to create such a group goal is through the use of rewards or incentives. In Slavin’s (1995) Student Teams Achievement Divisions (STAD), for example, teams work in a peer tutoring format to master information presented to the whole class and then take quizzes to assess individual achievement. Teams are rewarded based on the improvement of individuals in the team, thus holding both groups and individuals accountable. By making the team’s success dependent on the individual learning of all members of the team, these group rewards ensure individual accountability and a feeling of personal responsibility for what happens in the group. This motivates students to work hard toward the group goal, to encourage others to do the same, and to help each other in order to ensure the group’s success. These processes increase the chance of a favorable group outcome and consequently a reward for each group member.
The use of group rewards is not without controversy. Lepper (1983) posited that extrinsic rewards can undermine intrinsic motivation and interest in the classroom, and that students may begin to pay more attention to obtaining the reward than learning the material. Students may invest the least effort necessary to obtain rewards or may choose to perform less challenging tasks. Students may help each other only as the means to an external reward rather than valuing helping each other and learning for their own sakes. When groups fail to obtain the reward, students may feel dissatisfied with themselves or others or may blame low performers for the group’s failure.
Features of cooperative methods such as Slavin’s STAD are intended to prevent these debilitating processes. To limit the focus on the reward, the incentive (e.g., a team certificate or mention in a classroom newsletter) is not designed to be highly desirable or to have high stakes. To ensure student effort, task difficulty is tailored to the capabilities of individual team members and no one is assigned overly difficult tasks. Finally, to reduce pressure on less capable group members, rewards are based on individuals’ improvements in scores, not on their absolute performance. Cooperative learning methods that use both group rewards and individual accountability generally produce greater achievement than cooperative learning methods that include only one or neither of these components (Slavin, 1996).
Social Cohesion Perspective
Deutsch (1949) also described how students may be motivated to help each other because they care about the group and its members. To develop a sense of group identification and concern for others, a number of cooperative learning methods include team building and the development of social skills in order to build group cohesion instead of (or sometimes in addition to) using group rewards; this approach is the foundation of Johnson and Johnson’s (1994) Learning Together. Team-building activities help team members to get to know each other and to experience success as a team. Social skills training focuses on interpersonal skills such as active listening, stating ideas freely, and accepting responsibility for one’s behaviors, as well as small-group skills such as taking turns, sharing tasks, making decisions democratically, trying to understand others’ perspectives, and clarifying differences (Gillies, 2004). Developing and using these skills helps group members to trust, accept, and support each other, to communicate accurately and effectively, and to resolve conflicts constructively.
Feedback on how well the group is functioning also helps to build cohesion. Discussing their group’s interaction and how they might improve it (sometimes called group processing) helps groups to function more effectively. Having students share their thoughts and feelings about their own and their teammates’ participation also helps groups to identify, understand, and solve general communication problems (disruptive or bullying behavior), provides positive reinforcement for student collaboration, and helps students to maintain good working relationships (Johnson & Johnson, 1994).
Groups do not always function in ways that are optimal for learning. Researchers have documented a number of detrimental processes.
Individuals don’t always have equal opportunities to participate in groups. Some members are much more active and influential than others. Personality characteristics may explain some effects such as extroverted, outgoing, and energetic members doing most of the talking. Status characteristics may also determine relative influence in the group (Cohen & Lotan, 1995). High-status students, especially those with high academic standing or peer status characteristics (perceived attractiveness or popularity), tend to be more active and influential than low-status individuals, while low-status individuals tend to be less assertive and more anxious, talk less, and give fewer suggestions and less information than high-status individuals. Social characteristics, such as gender or race, may also operate as status characteristics in heterogeneous small groups; boys and White students may be more active than girls and students of color.
Even artificially created status differences (such as classifying students’ competence on the basis of fictitious test scores; Dembo & McAuliffe, 1987) can create imbalances in individual participation and influence.
Whereas some students may be shut out of interaction, other students may choose not to participate. Students may engage in social loafing, or diffusion of responsibility, which arises when one or more group members sit back and let others do the work, possibly because they believe that their efforts can’t or won’t be identified or are dispensable (Salomon & Globerson, 1989). This free-rider effect may turn into a sucker effect when the group members who complete all of the work discover that they had been taken for a free ride and start to contribute less to the group work in order to avoid being a sucker.
Students who choose not to be involved or who are excluded from group interaction will not experience the benefits of active participation described in the previous sections of this chapter. And the students who do participate will not benefit from the knowledge and perspectives of the passive students, and may even lead the group off track by pursuing the wrong task or suggesting incorrect solutions that are not challenged.
The goal of promoting the participation of all group members was the foundation of one of the earliest cooperative learning methods, jigsaw (Aronson, Blaney, Stephin, Sikes, & Snapp, 1978), which was designed to bring together students from different cultural backgrounds. In the jigsaw classroom, students are assigned responsibility for mastering a portion of the material, they meet to discuss that material with other students assigned the same topic, and then they return to their heterogeneous groups to teach their topic to the other members of their groups. Another early cooperative learning approach that also built in participation by all students is group investigation (Sharan & Hertz-Lazarowitz, 1980), in which students carry out research on their piece of a group project and then come together as a team to integrate their findings and plan a class presentation.
Failure to Seek and Obtain Effective Help
Another set of debilitating processes concerns the failure of students to seek help when they need it and to obtain effective help when they seek it. Students who do not seek help may never correct their misconceptions. Students may fail to seek help for many reasons (Nelson-Le Gall, 1992). Students may lack the metacognitive skills necessary to monitor their own comprehension and so may not realize that they don’t understand the material or can’t perform the task without assistance, or they may watch their teammates solve a problem or accomplish a task and assume that they can do it too.
Even if students are aware that they need help, they may decide not to seek it for fear of being judged incompetent and undesirable as a workmate, or they may not want to feel indebted to those giving the help (Newman, 1998). Or students may believe that help seeking is undesirable (as a result of classroom norms that call for students to remain quiet and work alone, or sex-typed norms that view help seeking as more appropriate for females than males) or they may have received antagonistic or unsatisfactory responses to previous help-seeking attempts. Students may also believe that no one in the group has the competence or resources to help, or that they themselves lack the competence to benefit from help provided.
When students seek help, they may select helpers who are nice or kind, or who have high status, rather than those who have task-relevant skills (Nelson-Le Gall, 1992). Students may not have effective strategies for eliciting help. In particular, the kinds of questions students ask often have important consequences for the kinds of responses they receive. Requests for help that are explicit, precise, direct, and targeted to a specific aspect of the problem or task are more likely to elicit explanations than unfocused questions or general statements of confusion. Asking precise questions makes it easier for other group members to identify the student’s misconceptions or areas of confusion and to formulate effective help. General questions, in contrast, may signal a lack of ability, a lack of effort, or both. Students who appear to be loafing are typically less likely to receive help than those who appear to be working hard.
Even if group members are willing to help, they may not have the skills to provide effective explanations. Help-givers may have misconceptions themselves, may not be able to translate their thinking into appropriate or understandable language, may not provide enough detail or detail relevant to the help-seeker’s particular difficulty or level of understanding, or may not be able to identify the help-seeker’s problem.
Finally, whether a student obtains help may depend on the group’s composition and the student’s relative position within that group. For example, in groups that are heterogeneous in terms of student proficiency, middle-ability students may be left out of a teacher-learner relationship that develops between high-ability and low-ability members of the group; whereas the same middle-ability students may participate more actively in homogeneous groups, especially with regard to receiving answers to their questions (Webb & Palincsar, 1996).
Too Little or Too Much Cognitive Conflict
Although students can learn by resolving discrepancies in ideas, too little or too much conflict can be detrimental (Bearison, Magzamen, & Filardo, 1986). Infrequent conflict may reflect suppression of disagreements, or pseudoconsensus or pseudoagreement, in which students minimize disagreements or pretend they don’t exist. In these cases, incorrect ideas may go unchallenged. Too much conflict, on the other hand, may prevent group members from seeking new information to resolve their disagreements. If they spend all of their time arguing (especially if their aim is to win the argument regardless if they are right or wrong), they may never develop new insights.
Lack of Coordination
Group functioning and individual learning may also suffer from a lack of coordination of group members’ efforts and participation (Barron, 2003). Group progress may be inhibited by low levels of attention to and uptake of members’ suggestions, even when those suggestions are correct and potentially productive. In incoherent conversations, students advocate and repeat their own positions and ideas, ignore others’ suggestions, reject others’ proposals without elaboration or justification, and interrupt others or talk over them. Such lack of coordination and joint attention may undermine all of the processes through which individuals can gain by collaborating with others, including resolving conflicts, co-constructing knowledge and understanding, participating in scaffolding, giving and receiving help, and engaging in cognitive elaboration.
Other Negative Socioemotional Processes
Other negative socioemotional processes, such as rudeness, hostility, and unresponsiveness, can also impede group members’ participation, cause groups to reject correct ideas, and prevent groups from solving problems correctly (Chiù & Khoo, 2003). Such processes can also suppress help seeking, especially when students are insulted when they seek help, receive sarcastic responses, or have their requests rejected or ignored. Students who carry out negative behavior may have their requests for help rejected. Even students who are knowledgeable may withhold information or refuse to provide help when insulted or taunted by others.
Approaches to Promoting Beneficial Group Processes
Researchers have designed a great variety of collaborative learning methods in order to promote beneficial group processes and inhibit detrimental group dynamics. Many, if not most, small-group approaches incorporate one or more of the components described here.
Preparing Students for Collaborative Work
Altering Expectations and Status Relationships
To promote equal participation among students in heterogeneous (especially multiracial) groups, Cohen and Lotan (1995) developed methods of minimizing status effects by altering high-status students’ expectations about low-status students’ competence. In expectation training, low-status students receive special training on both academic and nonacademic tasks, and then teach high-status students how to do the tasks. All students then work on unrelated activities in their heterogeneous groups. The important feature of this training program is changing high-status students’ perceptions about low-status students’ competence. Increasing the competence of low-status students without also manipulating the high-status students’ expectations of the low-status students’ performance (as it is when the low-status students teach the high-status students) is not sufficient to change the usual pattern of high-status students’ dominating group interactions. A less time-intensive approach is the multi-ability intervention, which raises students’ awareness of the multiple skills necessary to do the task. The teacher discusses with students the multiple abilities needed to solve complex problems (e.g., visual thinking, intuitive thinking, and reasoning), and when groups work on these tasks the teacher comments on particular contributions of low-status students and identifies the importance and value of their contributions.
All of these programs have shown success in reducing the relationship between status (based on language background, race, socioeconomic status, or academic ability, for example) and behavior in small groups. The more frequently teachers talk about the multiple abilities needed for a task (and emphasize that no one has all of the abilities) and comment on the value of low-status students’ contributions, the greater the participation rate of low-status students and the smaller the gap between high-status and low-status students’ participation rates.
Instruction in Explaining and Group Reasoning Skills
Many studies have incorporated instruction in academic helping and explaining behaviors (e.g., Gillies, 2004). Students are taught help-seeking and help-giving skills, such as asking clear and precise questions, giving explanations instead of answers, justifying their views, checking for understanding, checking others’ answers, challenging others, and giving specific feedback on how teammates solve the problem. Compared to untrained groups, trained groups exhibit more explaining and often higher achievement.
Structuring Group Interaction
Some small-group approaches structure group interaction in specific ways or implement activities to guide groups’ cooperation. Features of these methods include requiring groups to carry out certain strategies or activities, assigning students certain roles to play, or both.
In teacher scaffolded instruction, referred to as reciprocal teaching, teachers help students carry out certain strategies designed to improve comprehension of text: generating questions about the text they have read, clarifying what they don’t understand, summarizing the text, and generating predictions (Palincsar & Brown, 1984). Teachers initially take the leadership by explaining the strategies and modeling their use in making sense of the text. Then teachers ask students to demonstrate the strategies, but give them considerable support. For example, in order to help a student generate questions to ask groupmates, the teacher might probe what information the student gleaned from the text and help the student phrase a specific question using that information. The teacher gradually assumes the less active role of coach, giving students feedback and encouraging them. Students then carry out the text comprehension strategies in their small groups. With this method, students show considerable gains in reading comprehension and sometimes greater progress than with other cooperative learning approaches or in instruction without peer interaction.
Some peer learning approaches give students specific prompts to encourage them to give elaborated explanations. Students may be prompted to describe what happens in science experiments, to find patterns in their results, to explain why results occur, to justify answers or opinions, or to connect real-world experiences to classroom learning. Such explanation prompts have successfully produced conceptually advanced explaining as well as more complete and deeper understanding of the material.
Guided Reciprocal Questioning
In guided reciprocal questioning, students ask each other high-level questions about the material. These questions are intended to help students monitor their own and each other’s comprehension and to encourage students to describe and elaborate on their thinking. Students may be given “how” and “why” question stems to guide their discussions of text, for example, Why is … important? How are … and … similar (King, 1992, p. 113)? Or students may be given questions to help them co-construct and explain strategies for solving problems, such as What is the problem? What do we know about the problem so far? What information is given to us? and What is our plan (King, 1999, p. 101)? Groups’ use of these high-level questions increases the frequency of elaborated explaining and produces greater learning than individuals using such questions outside of a collaborative context or engaging in group discussions without such question prompts.
In order to promote the benefits that arise when students try to resolve conflicting ideas, Johnson and Johnson (1995) built controversy into the group’s task by subdividing groups into teams and requiring the teams to master material on different sides of an issue (e.g., should there be more or fewer regulations governing hazardous waste disposal). Then groups presented their views to the other team and synthesized the two positions. Compared to groups required to seek concurrence by working cooperatively and compromising, groups required to discuss opposing ideas often carried out more high-level discussion of the material and less description of the facts and information; they also attained higher achievement.
Cognitive Role Specialization
To facilitate high-level discourse and learning in groups, a number of approaches give students specific roles to play (usually alternated or rotated) in which they carry out specific cognitive activities. Students may be assigned such roles as recaller (also called learning leader or summarizer) and listener (also called active listener, learning listener, or listener/facilitator) that can be incorporated into scripts for groups to follow. The recaller summarizes the material and the listener is responsible for detecting errors, identifying omissions, and seeking clarification. Students then work together to elaborate on the material; they change roles for the next part of the task. Scripted cooperation usually produces greater student achievement than unstructured cooperation, as a result of students’ explicit use of elaboration (O’Donnell, 1999). These scripts may be even more productive when students teach each other different material than when students have studied the same material; students may be less likely to worry about how others will evaluate their summaries, and may be less likely to make inferences about what others are saying and more likely to ask them to clarify their summaries.
In reciprocal peer tutoring, students receive training in how to model strategies such as summarizing text and in how to give explanations, corrections, and feedback about other students’ work. They then alternate tutor and tutee roles during pair work. Some approaches pair more skilled and less skilled learners, whereas others pair students randomly or pair those with similar proficiency. Special attention is often paid to promoting high-level discourse during paired discussions, such as training tutors to give highly elaborated conceptual rather than algorithmic explanations to their partners (e.g., using real-life examples, discussing why an answer does or does not make sense; Fuchs et al., 1997). Reflecting the importance of the activity of the help-receiver, some peer tutoring models guide the tutor in helping the tutee to give high-level explanations (King, 1999). The tutor asks questions designed to encourage the tutee to provide explanations of the material, asks further questions to push the tutee to elaborate upon or justify their explanations as well as to correct incomplete or incorrect explanations, and asks questions to push tutees to make connections among ideas and to link new material to their prior knowledge. Observations of tutor-tutee discussions show that these strategies for increasing the level of pair discourse are usually highly successful with corresponding positive effects on achievement.
Manipulating the Group-Work Task
To maximize the participation of all group members, Cohen (1994) recommended that groups be given complex tasks or open-ended problems without clear-cut answers or procedures that cannot be completed very well by a single individual and that require the combined expertise of everyone in the group. Such tasks encourage groups to recognize the multiple skills and perspectives needed to complete the task and to value the different contributions that each student makes. Tasks or problems that can be completed by one student with the requisite skills, on the other hand, are more likely to limit the participation of students without those skills.
The Teacher’s Role
Developing Classroom Goal Structures and Norms
The teacher can play an important role by working with students to mutually construct norms for how students should engage with each other during group work. The teacher can monitor and intervene in small-group dialogues in order to remind students about their obligations (e.g., to make sure that everyone participates, to describe their interpretations of problems and solutions, to share their thinking and solution methods with others, to justify their own answers, and to challenge each other’s solutions) and to make specific suggestions for how students should behave in particular circumstances (e.g., stop another student and ask for help, be sure that everyone understands each other’s problem-solving strategies; Yackel, Cobb, & Wood, 1991). The teacher can discuss with the whole class expected small-group behaviors and can use specific, as well as hypothetical, situations to initiate discussions with the class about students’ responsibilities in collaborative work and to show examples of genuine dialogue between students.
Modeling Desired Discourse
Teachers also communicate their expectations for students’ behavior through their own behavior. Two recent studies showed how students’ behavior mirrored the very different styles of teacher interaction with those students. In one study in which teachers received little specific instruction in how to engage with students around the instructional material, the teachers tended to give unla-beled calculations, procedures, or answers instead of labeled explanations; they often instructed using a recitation approach in which they assumed the primary responsibility for solving the problem, having students only provide answers to discrete steps, and they rarely encouraged students to verbalize their thinking or to ask questions (Webb, Nemer, & Ing, 2006). In their cooperative groups, students adopting the role of help-giver showed behavior very similar to that of their teachers: doing most of the work, providing mostly low-level help, and infrequently monitoring other students’ level of understanding. Students needing help did not use effective help-seeking skills nor did they often persist in seeking help, which reflected the expectations communicated by the teacher about the learner as a fairly passive recipient of the teacher’s transmitted knowledge. In another study, the teachers were trained to use specific skills to challenge students’ thinking and to scaffold their learning (e.g., asking questions that probed or attempted to clarify student thinking, challenging discrepancies in students’ ideas, offering suggestions tentatively instead of directing students to follow certain procedures or strategies; Gillies, 2004). In these small groups, the students mirrored their teachers’ behaviors (e.g., attempting to relate new information to ideas previously discussed).
The past several decades have produced a great deal of research investigating how and when students learn in small groups, as well as how small-group work might be orchestrated to produce desired learning outcomes. Moving considerably beyond characterizations of early research as “black box” in nature (Bossert, 1988-1989), research in this area has uncovered much information about the mediational processes that promote as well as inhibit learning in small groups.
Despite considerable progress in the field, however, there are a number of topics that need more attention. First, not enough is known about the role of student characteristics (e.g., ability, gender, cultural or racial background, personality) in group functioning to make recommendations to teachers about how to compose small groups. This issue is further complicated because the same student may behave and have quite different experiences in different groups; there is still debate about which characteristic(s) are most salient to groups when group members differ on multiple characteristics simultaneously. Second, although working in groups has positive effects on interpersonal attitudes and liking of the task (Solomon, Watson, Schaps, Battistich, & Solomon, 1990), little is known about whether the processes that promote these outcomes are consonant with those that promote learning. Third is the possible incompatibility of different goals of group work, such as individual learning versus group productivity. For example, depending on the nature of the task, processes that benefit individual students’ learning may impede group productivity, such as when producing a high-quality group product is best and most efficiently accomplished when only the most skilled students contribute (Webb, 1995). How to balance these competing aims needs further study. Fourth, with the explosion of interest in online communication comes the natural question of whether collaborating with others in a remote environment functions similarly to the face-to-face group dynamics considered in this chapter. There is much interest in computer-mediated communication (Lou, Abrami, & d’Apollonia, 2001), but researchers have only recently begun to systematically study the issues of learning within this context.
Finally, there is a pressing need for more research on how to help teachers implement small-group learning in ways that fit their own classroom needs (O’Donnell & O’Kelly, 1994). Given the relentless demands on teachers and their time, a major challenge is how to enable teachers to create classroom learning environments that encourage beneficial small-group work, to prepare their students to interact collaboratively, to create small-group structures and tasks that are most appropriate for their own classrooms, to monitor and evaluate groups’ progress, and to facilitate students’ interaction with each other in ways that benefit all of their students.