The nature of groups: implications for learning design.

Considering recent claims that groups are important to teaching and learning, the authors examine the differences between groups and collections of people to see how one might create learning groups when designing learning and instruction. Based on literature in the fields of group dynamics and education, characteristics of groups are listed and explored. Strategies for creating learning groups are synthesized. Implications for practice are discussed.

**********

There have been a number of claims that groups are important to teaching and learning. For example, Shaban and Head (2003) have called the use of group projects an essential tool for learning. In addition, Bhattacharya and Chatterjee (2000) claimed there were several "pedagogical advantages" (p. 298) to using groups.

Much of this interest in groups stems from the work of Piaget (1928, 1973), Vygotsky (1978), and, more recently, Jonassen (1990). Crook (1998) argued that, though these works have often talked about social learning or collaboration, it may not be accurate to interpret them as group work. For example:</p> <pre> Students learn to work in groups to solve a realistic and authentic problem, thus gaining collaborative learning experience. (Neo & Neo, 2002, p. 144)

The students work together in groups (collaboration) ... (Black &

McClintok, 1996, p.26) You must provide as much opportunity as possible for individual reflection (for example, analysis of errors) and social interaction (for example, group labs). Clearly, each educator must decide how to apply these aphorisms in a concrete situation. (Ben-Ari, 2001, p. 68) </pre> <p>If using groups of people is important to teaching and learning, we need first to answer the question, "What is a group?" "More than one learner working together" may not be a sufficient definition. Tudge (1999) argued that merely asking learners to work together in educational research, or even pairing learners based on ability, is insufficient to foster the types of learning Piaget and Vygotsky discussed. In the fields of social psychology and group dynamics, Lewin (1948) and Forsyth (1999) argued that, even though people often find themselves gathered together, not all gatherings of people qualify as groups. Thus, while creating learning "groups" may appear simple, it is, in fact, more complex than it appears. Learning designers (teachers and instructional designers) will likely need to do more than simply cluster learners and assign them shared tasks.

To begin to address what learning designers should consider in creating learning groups, it may be beneficial to look at the possible definitions of groups to understand what makes a collection of people a group. This is not a new or novel idea. We often define the tools we are interested in using. [For example: What is the Internet? (Parritt, 2003) or What is a browser? (Ehley, 2003)]. We often seek to answer the same questions: What is this tool? What does it do? How does it work?

With familiar items such as chalkboards, it is easy to assume users will understand them and their uses, based on past experience. However, relying only on our past experience with groups to define them as a tool may be problematic. Kagan, Haveman, and Segal (1984) argued that relying on past experience may produce persistence of set. That is, when we rely on past experience with things, we build up a habitual way of approaching them (p. 192). This may lead us to functional fixedness. That is, once we see a tool as useful in one way, we ignore other potential uses and techniques for the tool (Bruning, Schraw, & Ronning, 1999). Further, according to Nelson (1990), relying only on our experiential understanding of a type of item may cause us to misinterpret an object's attributes and purpose. Nelson cited the example of one coin that was similar to another coin, and often mistaken for the other, despite key differences in value and use.

Applying this to groups, it is quite possible we may take our experiential understanding of groups and misperceive their familiar operational qualities without fully understanding them. This could lead to the misapplication of groups or perpetuation of conceptual flaws in learning designs. Therefore, for the beginning teacher or instructional designer, it is important to understand just what a group is. This article explores different definitions of groups to give the beginning instructor or instructional designer some insight into what groups are. The article then identifies factors that contribute to the realization of these definitions, and poses questions learning designers might consider when designing group learning experiences.

WHAT MAKES A COLLECTION OF PEOPLE A GROUP?

Researchers vary in their opinion of what makes a collection of individuals a group. Cartwright and Zander (1968) asserted differing definitions of what a group is derived from the complexity of groups and the different aspects of groups and their context. Three ways of defining a group follow. While each is slightly different, they do not necessarily conflict. In fact, they often support one another and might be viewed as interlocking perspectives on a single definition.

A Group Perceives Itself to be a Group

Sometimes we find certain classroom groups interact more like a pile of conflicting individuals; other times the members seem to gel together into what feels like a group. As teachers, instructional designers, and learners, we can perceive that "groupness." This is the basis for social psychology's definition of a group: "A collection of persons who are perceived to be bonded together in a coherent unit to some degree" (Baron & Byrne, 2003, p. 475).

This idea of perception defining the reality of a group goes back to the early days of social psychology, when researchers argued whether a group was "real," or just the sum of the individual action of its members. Much of this debate was fostered by Triplett's (1898) claim that racers perform better when racing against each other than when racing individually against the clock. According to Triplett, this appeared to support the claims of some theorists, who suggested the existence of a "group mind" or "group intelligence" distinctly separate from the individual (Le Bon, 1896). In response to the idea that groups might have intelligence beyond that of individual members, Allport (1924) claimed (a) researchers need to stop relying on concepts from the past and start relying on scientific experimentation; (b) there is no group mind; and (c) group actions are simply the sum of individual members' actions.

Thomas (1928) addressed the crux of the argument when he explained that, even if groups are not real entities, when people perceive them to be real, they are real in their consequences. The idea is that people who perceive groups as "real" seem to act differently. Campbell (1958) coined the term entitativity to refer to the members' perception of the group as a real entity. According to Campbell, some collections of individuals are perceived to be a group more than other collections. Based on Campbell's idea of entitativity, Lickel et al. (2000), ran one study in which 40 different types of groups were rated for their entitativity by 199 U.S. college students, and a second study in which the same groups were rated by 154 Polish college students. The results are interesting, showing students questioned do indeed perceive a difference in whether the collection of people is a "group."

Lickel et. al. (2000) claimed that there was substantial variation among the group ratings. For example, in study one the ratings varied between an 8.27 and a 2.4. To show significance in perception, they used clustering analyses to segment the groups into categories. First, intimacy groups with emotional bonds--such as members of a family, friends, or people in a relationship. Second, task groups that had a job to do--such as an airline flight crew, cast of a play, or members of a jury. Third, social categories of people were recognized--such as citizens of a country, women, or racial groups. Fourth, weak social associations were identified--such as people who live in the same neighborhood, or students at a university. Fifth, transitory associations that cause people to come together throughout the day were seen--such as people at a bus stop, or people in line at a bank. All pair-wise ANOVA comparisons of these clusters (in study one) were found to be significantly different (at p < .05) in their entitativity rating with one exception--there was no significant difference between social categories and groups with weak social relationships. In study two, the difference between the intimacy and task clusters was not significant when using ANOVA and posthoc tests, but both were significantly higher than the social categories and weak social relationships clusters. Again, the difference between social categories and weak social relationships clusters was not significant; however, all groups were still significantly higher than the transitory cluster.

As Table 1 shows, people in groups with higher entitativity ratings tend to be referred to as members. This, however, introduces a problem: Does the label employed (for instance, "member," "team," "family," or "crew") affect our perception of whether these collections are groups, or does our perception cause us to use differing labels?

A Group Has Interactive and Interdependent Members.

Lewin (1948) argued social interaction and interdependence among individuals is the key element that defines groups. Based on Lewin's premise, several prominent researchers in the field of group dynamics have offered concise and similar definitions. Cartwright and Zander (1968, p. 46) contended "... a group is a collection of individuals who have relationships to one another that make them interdependent to some significant degree" and Forsyth (1999, p. 5) defined a group as "... two or more interdependent individuals who influence each other through social interaction." Thus, in these researchers' terms, groups are aggregates of people who affect one another. Since the field of group dynamics seeks to understand relationships between the individuals in groups, it is understandable that their definition of a group is based on the interaction of group members.

In a more technical fashion, Cattell (1953) asserted, "... a group is a collection of organisms in which the existence of all (in their given relationships) is necessary to the satisfaction of certain individual needs in each" (p. 20). Cattell argued individuals belong to the group to achieve certain goals that may be different from the goals of other members, and even from the group itself. To be a group, though, they are dependent on interaction with other individuals to achieve, or fail to achieve, desired goals for either the group or the individuals.

In a way, this relationship of effect between members appears to have some connection to the idea of entitativity. Notice how the listed "groups" in Lickel et al.'s (2000) study had a higher entitativity rating if the members of the group expected to be affected by the actions of others. For example, members of a flight crew or sports team can have a dramatic effect on each other through their actions. The stronger the effect of individuals on others in the group, the stronger the perception of the group seems to become.

A Group Is Both Task-Related and Social-Emotional in Nature

Some writers argue assigning a collection of individuals a task to complete together is not enough--groups cannot be groups without a socialemotional exchange between the members. For example, Pierce (1962) conjectured that, while groups often form to complete a task, emotional involvement is an underlying reason for belonging to the group. According to Pierce, the task focus of groups is merely a medium for the emotional exchange that humans need. Further, Pierce speculated that, if a group cannot address the emotional needs of its members, that group will fall apart. He argued the role of the group is to balance emotional and task activity.

In his study of task groups, Bales (1965) asserted an increase in task-related activities of a group appeared to bring a resulting increase in socialemotional activities. According to Bales, focusing on task effort builds tension among members, which is then met by social efforts to repair strained bonds between group members. Bales argued successful group function requires its members to cycle through periods of both task and social-emotional activity. According to Forsyth (1999), group attention shifts back and forth between task and the socio-emotional needs of its members. Forsyth offered a case study on the animators at Disney Studios as an example of how times of intense work were then countered by intense focus on "off-task" social activity.

Ellis and Fisher (1993) also contended these two dimensions are a part of every group. They argued social-emotional bonds affect group cohesion, which in turn affects the ability of the group to work on a task. Task accomplishments then affect the social emotional bonds of the group, all interconnected in a circle of cause and effect. Thus, group task and social-emotional forces influence each other, and are inseparable in group function.

For these researchers and scholars, it is not enough to merely define a task for a collection of people to complete to make a group. Group members need to relate on a more emotional level as well. This supports, in part, the previous two definitions. First, when looking at perception of the group and entitativity, people with whom we form emotional groups seem to be high on the list, but not as high as groups that may cause people to relate on both the task and social-emotional planes. For instance, family rates high on Lickel et al.'s (2000) list, but not as high as sports teams. Further, when looking at interdependence and the way people can affect others in the group, interdependence may increase when people can affect each other not only through task-goal realizations, but also through social-emotional interchanges.

FACTORS THAT PROMOTE THE DEFINITION OF REAL LEARNING GROUPS

Given these three definitions, what are the key factors that might prove most useful in designing effective learning groups?

Factors Promoting the Perception of a Group as "Real"

According to Campbell (1958), the perception of a group as "real" is related to several things: (a) Common fate: How are the destinies of its members related?, (b) Similarity: How are the members alike?, and (c) Proximity: Are the members located physically near one another? More recent works also suggest (d) time and (e) perceptions of those outside the group may also be factors. This section examines these five factors.

Common fate. The idea that the fate of the entire group is dependent on the actions of its individual members is a strong force in determining whether the group is perceived as real. This often manifests itself in the idea of a group mark (Slavin, 1991a) or group accountability (Johnson & Johnson, 1996). While assigning a group mark for completion of a learning task may heighten perceptions of "groupness," it also begs the question of how the common fate of the group changes the interaction of the group and affects the psychological strain upon the members.

After looking at groups involved in the Cuban missile crisis (Kennedy's advising team--Janis, 1983) and an Antarctic expedition (Peri, Barbarito, Barattoni, & Abraham, 2000), the researchers argued highly stressful environments may shift group behavior toward one of two extremes: heightened conflict or heightened cohesion. It is doubtful learners in a most learning situation face comparable pressures. It is, however, difficult to separate what is actual pressure from what the learner perceives as pressure. Most of us can recall anxiety and pressure we have experiences in learning situations that affected the ways we behaved and interacted with others. What pressures might a student with a 4.0 GPA feel if he or she were to fail a course because someone in his group made a mistake? Clearly, using a group mark (common fate) has implications.

While common fate is often interpreted as assigning a group mark, it may not need to be. Slavin (1990) discussed the use of group rewards, not necessarily group marks. In some professional training courses, marks may not be assigned at all. Instructional designers need, therefore, to consider how the fates of the learners are tied together and the nature of group "rewards."

Similiarity. While physical similarity has been one way to group people and items, it may have a downside. Fixation on superficial attributes may lead to attribution error (Baron & Byrne, 2003). This occurs when people group others based on surface attributes, such as appearance, and assign certain attributes and expectations to all members of the group. This can result in unjustified stereotypes that often come into play in situations where gender, age, and/or race are prominent. Campbell's idea of similarity is related to perceived commonalities that are not always superficial. These deeper commonalities between members may help to foster feelings of "groupness."

Stressing common goals of group members may be one way to foster this "deeper similarity" among group members. For instance, members of the Humane Society have a common interest in the humane treatment of animals. They may vary in their profession, age, race, and gender. On a deeper level of difference, they may also differ in how one should treat animals. However, it is a common goal that brings them together and joins them in a group.

In contrast to this point of view, Cattell (1953) argued the goals and motives for each person in the group may be very different, and are based on individual goals. A person may join a group for what others might deem "ulterior" motives. For instance, a celebrity might join the Humane Society to improve his or her image, not because he or she cares about animals. Likewise, one might join a course-related group to receive a good mark, rather than to learn.

Proximity. According to Campbell, physical closeness may be an important criterion for grouping. It is unclear whether perceived proximity has been reduced by new technologies (such as the interstate highway system or the Internet), technologies that enable us to redefine "closeness." In the 1850s, many people died within 50 miles of where they were born. Today many people travel that distance in less than an hour each day. It is similarly possible today to communicate over the Internet with people with whom one is unlikely ever to be in physical proximity (such as those in foreign lands). At the same time, Kiesler and Kraut (1999) contended interpersonal ties across the Internet are still not as strong as those formed face-to-face. If one applies this to learning groups, one would conclude distance and e-learning may be ways to produce learning groups, but online groups may intrinsically be weaker than groups of individuals that meet face-to-face.

Time. Tuckman (1965) and Tuckman and Jensen (1977) identified five group stages: (a) forming: getting to know each other and individual values, (b) storming: conflict over how the group should operate and the values it should have, (c) norming: establishing standards for operation in the group, (d) performing: working to accomplish tasks, and (e) adjourning: relinquishing group roles and dependency. As Wheelan, Davidson, and Tilin (2003) observed, groups in the earlier stages of development often expend a great deal of energy in the forming and norming struggle, attempting to define how the group will work. As further reinforcement of influence of time on group function, groups in the earlier cited Lickel et al. (2000) study that tended to stay together for a longer time seemed to have a higher degree of entitativity.

Wheelan et al. (2003), studied varying numbers of workgroups--sometimes 24 and sometimes 26--at both profit and nonprofit organizations, with group members ranging in age from 20 to 65. The researchers concluded member perceptions of the group change over time and groups tend to argue less and become more productive as they spend more time together. They also found the time it took to move to the fourth (performing) stage of development was 8.5 months. Given this finding, one would have to wonder whether how likely it is that groups in a course or classroom will be together long enough to get to the performing stage.

In contrast, Johnson, Suriya, Yoon, Berrett, and La Fleur's (2002) findings with graduate students seem to argue that group formation adapts to the time allotted to a group's task. In their study, groups formed quickly. The graduate students involved skipped the storming stage and completed the task in two weeks. However, one must question whether the group ever really "formed," or if the individuals merely completed the task. The absence of storming suggests norms and roles in the group may never have really developed, leading to absence of group leaders. Also, Johnson et al. noted there were issues with trust among team members that could be due to lack of human contact.

Recognized outside the group. Rupert Brown (1999) argued that, for a group to exist, it must also be recognized as a group by at least one person who is not a member. According to Brown, making the group visible to others outside the group aids its perception as legitimate. However, it might be argued there have been several secret societies throughout history, such as Skull and Bones at Yale, which have functioned as a group, yet were unknown to those outside the group.

Factors Promoting Interdependence Among Group Members

Wageman (1999) argued interdependence of group members is dependent on four influences: (a) Responsibility: Is the team collectively responsible for the task or is each individual responsible for certain identified parts of the task? (b) Process: What are the rules and restrictions for how the team members are required to operate? (c) Physical and technical demands: Does the task require or prevent learners from working together? (d) Resources: What talent, time, and materials exist across group members and how are they distributed? There is also one other key influence: Is the nature of the interaction (e) competitive or cooperative?

Responsibility. When considering how much of the task will be generically assigned to the group, and how much will specifically be assigned to each individual, one must also consider how the responsibility will be rewarded or ensured. As noted earlier, Slavin (1990, 1991a, 1991b) argued group rewards and grades are important to the group learning process. At the same time, Jackson and Williams (1985) asserted individual accountability when measuring group performance could positively affect the performance of the individuals in the group, a contention supported by Slavin (1990). Remembering common fate and Cattell's idea of individual motives and goals, one must ask if only group rewards or assessments are enough. Should there be individual assessments and rewards as well, or perhaps only individual assessments and rewards, if responsibility is only assigned directly to individuals?

Process. Kagan (1990) postulated cooperative learning was based on structures, or ways of organizing social interaction within the learning environment. According to Baloche (1998), these structures include the Think-Pair-Share, Think-Pair-Square, Pens in the Middle, Numbered Heads Together, Roundrobin, Roundtable with Roundrobin, and Group Interview structures.

These structures define how the members of the group will interact. For example, Jigsaw--yet another structure suggested by Aronson and Patnoe (1997) and Baloche (1998)--asks groups to divide, and have each member study one portion of the topic. In groups with four learners, each learner might count off one to four. All the "ones" would get together and study part A of the process, all the "twos" part B and so on. The original groups would reconvene, and the "ones" would become the "expert" on part A for their group, and assume responsibilities of teaching (or performing) part A to (or for) the rest of the group. Despite a lack of statistically significant findings, Gillies (2003) still concluded 8th-10th grade students with high structure and training in use of collaborative learning processes provided more help and assistance to one another and found the experience more enjoyable than students not provided with such structure.

Task demands. Determining how the physical and technical demands of the task affect interaction is partly related to the tools the learners need to use, and partly determined by the task itself. Wageman gave two examples of how these technical demands may affect interaction: He noted the technical definition of performing a musical quartet piece requires four people and requires those four people to work together to play music. Thus, interaction is required to meet this definition. He cited as his second example that typing a report on the computer presents a technical (tool) limitation that prevents more than one person from using the keyboard at a time. This, in turn, may discourage interaction.

Steiner (1972, 1976) created a taxonomy of task types to help demonstrate how the task itself may produce different types of interaction. Steiner argued some tasks are (a) unitary (had one action to be performed, such as shoveling snow), while other tasks are (b) divisible (had many steps that could be divided into several subtasks to be distributed among its members). These two classifications determine which tasks need the cooperation and special abilities of the group members (unitary) and which merely need everyone to do some of the same work (divisible).

In terms of divisible tasks, Steiner offered four possible ways we might look at how member inputs are combined to complete such tasks: (a) Additive: A simple task goes faster when more people assist in its completion (for example, shoveling, picking apples, weeding, stapling packets of papers); (b) Disjunctive: The group selects one of the options provided by individual members; (c) Compensatory: The group product is the average of the input of the individual group members; (d) Conjunctive: The actions of group members all contribute to the outcome, but are slowed by the weakest member (for instance, workers on an assembly line cannot go faster than the slowest member. If the slowest worker can be moved to another part of the line, performance may improve).

Resources. As noted earlier by Wageman (1999), making sure the talents and knowledge to complete a task are distributed among group members may promote interaction between the members as a necessary means of getting the job done. This applies in several ways: First, one must assure the group has all the talents it needs. Second, one needs to confirm talents are distributed so that every team member has something to contribute and each needs, in turn, to rely on the other members for their contributory talents.

Crook (1998) also argued external resources are important to the group. His idea of external resources includes items that will make the meaning of the content available to learners, and permit the social construction of new knowledge. According to Crook, this could include computers to provide representations of the content that could be manipulated.

Cooperation or competition. At some point the members define--either through words or actions--whether the interdependence of its members is based on cooperation or competition. Deutsch (1949a, 1949b) discussed this idea of cooperative versus competitive interdependence in groups. In Deutsch's terms, contrient interdependence means that for one person to win, the other person must fail, while promotive interdependence means that the success of one will increase the chances of success for the other members. In a similar definition from game theory, Mero (1998) observed individuals either decide to work together to maximize everyone's achievement (cooperate) or decide to maximize their personal achievement and minimize others personal achievement (compete). Whichever type of interdependence is present, group members are affected by other members of the group, and, in turn, affect other members in the group, whether they intend to or not.

Mero (1998) argued game theory can inform our decision of whether to cooperate or compete. According to game theory, (a) greater gains will be made by everyone if cooperation is used and (b) if it is unclear whether the other group members are competing, the "tit-for-tat" principle suggests individuals will be most successful in a group situation if they start off being cooperative and switch to competition if they find other members competing.

Deutsch (1949a, 1949b) also contended groups of college students that cooperate are better off than groups with members that compete. However, Steiner (1976) argued Deutsch's study was set up with individual cooperative groups that may, in fact, have actually been competing with one another. Thus, any benefits Deutsch observed may, in fact, be attributable to competition among groups, rather than the cooperation of members within a group.

Going back to Mero's game theory, those who compete gain an advantage for a short period of time, since tit-for-tat needs time for the reaction of others to balance things out. One must consider how attractive it may be for students who are in class together for a short period of time to switch to competition to gain a needed edge on a project. In contrast, are the students in a position where they may need to face these students again, and aware that current actions may have future ramifications?

Factors Promoting Socio-Emotional and Task Balance

In considering this dual nature of task and social-emotional function, one must consider the phrase "on-task" and how the inseparable nature of these two group dimensions have ramifications for productivity. The "old school" idea of making sure learners are spending time working on the functional task goals of the group seems to have an air of academic legitimacy--learners are not spending time "playing" or "goofing off." When on task, learners appear to be studious and productive. However, does a teacher who rigorously enforces time-on-task actually reduce the productivity of the group?

The social-emotional exchange is extremely important for the members to learn each other's principles and expectations to decide on what they want to do about problems, and set guidelines for future action (group norms). When looking at time and perception, Johnson et al. (2002) found when groups on a tight time line skipped the storming phase, they failed to resolve these issues. Johnson et al.'s findings may infer that iterative cycling through the conflict stage, group trust problems, and leadership issues in the group are results of not having time to resolve the social-emotional needs of the individuals.

Several factors seem to affect the social-emotional ties to the group and between group members. The most prominent appear to be (a) advantages, (b) communication, (c) friendships, and (d) accomplishment.

Advantages. Ramifications of belonging to a group can affect how individuals act in a group, or whether they want to belong to a group. Festinger (1953) theorized that attraction to group relates to the privileges given to recognized members of the group, and whether the group offers relationships and activities that meet the personal needs and interests of its members. Festinger argued that these things work in combination to bring people to groups, and that any one by itself is not usually enough to create attraction.

When belonging to the group creates a perceived disadvantage in either material or non-material needs (such as prestige), Thibaut (1953) claimed the peripheral members (those not seen as central or popular) tend to split into a faction. Conversely, when members feel there is an advantage belonging to a group, they tend to band together. Thibaut theorized this relates to the perceived attractiveness of belonging to the group.

Moreover, Thibaut found this behavior became more complicated when considering inter-group relations. He asserted that while cohesion in the underprivileged group would decline if underprivileged groups were successful in gaining access to privileged treatment, members of the underprivileged group could band together in their hostility toward the privileged group if they could not find a way to gain access to privileged status.

Communication. To relate to other members, individuals need to learn how others communicate and overcome differences in communication styles. Resolving these differences in communication is a large part of the storming and norming phases of group development (Tuckman, 1965; Tuckman & Jensen, 1977).

There may be several ways to speed the storming and norming stages, and to aid communication within the group. For instance, in studying online learning groups at the college level, Caropreso and Chen (2003) suggested personality type plays a role in group interaction, and that certain mixes of personalities may affect the type and pattern of online communication. In another approach, formal training in communication and learning procedures may help improve group communication. For example, Wentzel (1991) argued such training can only improve the situation of learners. Similarly, Johnson and Johnson (1990) called for, and outlined, such a process for training in communication and conflict resolution.

Organizational development offers one basis from which to help learners develop communication skills and better understand one another. Bolman and Deal (1997) proposed a model for getting things done in organizations. This model lays out four frames in which every organization operates: (a) The structural frame: Who is the leader and how do communications and requests flow through the group? What is the formal chain of command?; (b) The human resources frame: Who has the know-how and skill to get the job done?; (c) The political frame: Who has the connections and pull to get the job done? How can the actions of the group affect the perceptions and treatment of those in the group both positively and negatively?; and (d) The symbolic frame: What is important to the group? How does the group want to be viewed?

Friendship. Crook (1998) argued that whether or not learners are friends with one another is an important aspect of groups that needs to be studied more. In addition, Hare (2003) suggested friendships foster exchange and support among individuals in the group and friendship is a reason people may join a group. Likewise, Wright (1943) argued friends in a frustration situation show more cooperation and spend less time in conflict.

Accomplishment. Based on Ellis and Fisher's (1993) argument that accomplishment builds cohesion that, in turn, brings group members emotionally closer, leaving groups with a history of accomplishments together may help build social-emotional ties among its members. Wheelan et al.'s (2003) theory of time affecting group perception and communication appears to lend further support to this argument. Keeping the group together and giving them early accomplishments may help bind them together.

SYNTHESIS

We have presented a somewhat comprehensive review of key literature. This section identifies three major questions that may help teachers or instructional designers use learning groups to greater effect. Each question is followed by four to six questions that help identify operational issues (and perhaps practices) that might help answer the major question in the direction desired. For each question, the writings or research behind it are cited.

1. Why would the members perceive themselves as a group?

a. Common Fate: Will you assign a group mark, and to what extent will that mark affect overall assessment? (Campbell, 1958, common fate; Johnson & Johnson, 1996; Slavin, 1990, 1991a, 1991b)

b. Similarity: Do you group learners that have a common interest or likeness beyond the assignment to speed the group formation and norming process? (Campbell, 1958, similarity; Wheelan et al., 2003) or Do you group individuals with different views to foster growth? (Piaget, 1928, 1973; Vygotsky, 1978?

c. Proximity: Are group members seated together or do they see each other regularly? (Campbell, 1958, proximity)

d. Time: Will the group have enough time to form? Will you leave the group together for an extended period of time (Ellis & Fisher, 1993; Wheelan et al., 2003) or change groups to expose learners to different ideas? (Piaget, 1928, 1973; Vygotsky, 1978)

e. Recognized as a Group: Are groups or their accomplishments known to others outside the group or class? (Brown, 1999; Ellis & Fisher, 1993) Is there a group identity or "name" that binds them together? (based on the findings of Lickel et al., 2000)

2. How are the group members interdependent?

a. Responsibility (and Common Fate): Do learners get group, individual marks, or a mix (Campbell, 1958, common fate; Cattell, 1953, individual goals)? How will you hold individuals accountable for their effort? (Johnson & Johnson, 1996; Slavin, 1990, 1991a, 1991b)

b. Responsibility: Do learners need to rely on the group to accomplish their individual and group tasks, or can one person do them alone? (Cattell, 1953; Steiner, 1972, 1976; Wageman, 1999)

c. Process & Task Demands: Does the task ask them to work together or merely work alongside one another? (Steiner, 1972, 1976) Are they working on a task that would ask them to be interdependent?

d. Resources: Do the members of the group provide the necessary expertise? Is certain expertise possessed by certain group members, or do they share similar expertise? Do external resources permit group members to work together or force them to work individually?

e. Cooperation/Competition: Do group members cooperate or compete? (Deutsch, 1949a, 1949b; Mero, 1998) How have you arranged the group to be sure?

3. Have you designed for task and social-emotional bonding time?

a. Time: Is there off-task time to build--and rebuild--social bonds, and also to complete tasks? (Bales, 1965)

b. Advantages: Is there a relative advantage in some way to belonging to the group? Does the group improve the status, prestige, or connections of the individuals involved, or does it impair them? (Festinger, 1953; Thibaut, 1953)

c. Communication: Have the members looked at the ways they communicate, the skills they have, what they want to get out of the group, and how they want the group to work? (Bolman & Deal, 1997; Thibaut, 1953)

d. Friendship-enhanced or Resource-enhanced: Are the groups constructed with focus on providing access to task resources or social compatibility? (Caropreso & Chen, 2003; Crook, 1998; Hare, 2003) In other words, are the learning groups engineered to provide a diverse set of resources, abilities, and perspectives; or to pair friends; or to bring together those who have enough in common that they could become friends? If deciding on diverse talent, will there be enough time for the group to form and bond? If deciding on friends, will they have the necessary diversity of skills and expertise not only to complete the task, but also to require interaction? (Crook, 1998; Wright, 1943)

e. Accomplishment: Are the group projects arranged in a way to build a feeling of joint accomplishments among team members? (Ellis & Fischer, 1993, cohesion and accomplishment; Steiner, 1972, 1976, types of tasks) Does (or will) the group have a history of accomplishment that will bring them closer together? (Ellis & Fischer, 1993; Wheelan et al., 2003)

DISCUSSION

For those of you who came seeking answers, but found only questions, you may be wondering where to go from here. What are the answers to these questions? What does all of this mean to the field of learning design?

First, designing groups for learning is difficult, and there are no sweeping and easy answers to apply to any situation or question. As this article demonstrates, literature in the field is often vague (at best) or contradictory (at worst) about how to handle grouping for learning. Even on the most basic issues, such as heterogeneous versus homogeneous grouping, there is room for debate. Though some studies show what may work, there is still a great deal of research to be done in the field.

Second, while there is a temptation to use the previous section's list as a checklist, methodically addressing each theoretical concern in turn, learning designers should avoid this. Many groups have been quite successful while only being influenced by some of the factors here. Some of the questions listed may apply to the groups you deal with and others may not. Perhaps only some of them need to be addressed for your particular situation. Don't give in to the "checklist mentality"; the key here is thoughtful consideration, not simply addressing all possibilities in turn and then moving on. A simple analogy: A boat that incorporated all possible safety features would likely be too crowded to work on, too ladened to maneuver easily, and perhaps too heavy to float. Of course, boats incorporating too few safety features have led to historic disasters. Thus, judgment, prudence, and experience are important in design.

Third, even though only some of the questions you now know may be applicable to your situation, it is important to realize we still don't know all the questions. These questions only begin to hint at the complexities faced by learning designers who plan on using groups. If you wish to explore group learning further, we suggest you investigate work in the areas of social facilitation (why and when performance is improved by working with others), social loafing (why and when people put in less effort when working in groups than when working alone), conformity, cohesion (both too much and too little are problems), and factors that cause the isolation of group members. These would be good starting points for further exploration of this topic.

References

Allport, F. H. (1924). Social psychology. New York: Houghton Mifflin.

Aronson, E., & Patnoe, S. (1997). The jigsaw classroom: Building cooperation in the classroom. New York: Longman.

Bales, R.F. (1965). The equilibrium problem in small groups. In A. P. Hare, E. F. Borgatta, & R. F. Bales (Eds.), Small groups: Studies in social interaction (pp. 444-476). New York: Knopf.

Baloche, L. A. (1998). The cooperative classroom: Empowering learning. Upper Saddle River, NJ: Prentice Hall.

Baron, R. A., & Byrne, D. (2003). Social psychology. New York: Allyn & Bacon.

Ben-Ari, M. (2001). Constructivism in computer science education, Journal of Computers in Mathematics and Science Teaching, 20(1), 45-73.

Bhattacharya, M., & Chatterjee, R. (2000). Collaborative innovation as a process for cognitive development. Journal of Interactive Learning Research, 11(3), 295-312.

Black, J. B., & McClintock, R. O. (1996). An interpretation construction approach to constructivist design. In B. G. Wilson (Ed.), Constructivist learning environments: Case studies in instructional design (pp. 25-31). Englewood Cliffs, NJ: Educational Technology Publications.

Bolman, L., & Deal, T. (1997) Reframing organizations: Artistry, choice, leadership. San Francico: Jossey-Bass.

Brown, R. (1999) Group processes: Dynamics within and between groups. Malden, MA: Blackwell.

Bruning, R., Schraw, G., & Ronning, H. (1999). Cognitive psychology and instruction. Columbus, OH.: Merrill, Prentice Hall.

Campbell, D. T. (1958). Common fate, similarity, and other indices of the status of aggregates of persons as social entities. Behavioral Science, 3, 14-25.

Caropreso, E., & Chen, S. (2003, October). Personality as a grouping strategy for online collaborative learning. Poster session presented at the annual meeting of the Association for Education Communications and Technology, Anaheim, CA.

Cartwright D., & Zander A. (Eds.). (1968). Group dynamics: Research and theory (3rd ed.). New York: Harper & Row.

Cattell, R. B. (1953). New concepts for measuring leadership, in terms of group synality. In D. Cartwright & A. Zander (Eds.), Group dynamics: Research and theory (pp. 14-27). Evanston, IL: Row, Peterson.

Crook, C. (1998). Children as computer users: The case of collaborative learning. Computers & Education, 30, 237-247.

Deutsch, M. (1949a). A theory of cooperation and competition. Human Relations, 2, 129-152.

Deutsch, M. (1949b). An experimental study of the effects of cooperation and competition upon group processes. Human Relations, 2, 199-231.

Ehley, L. (2003). What is a browser? Retrieved October 10, 2003, from http://www.depts.alverno.edu/cil/mod1/webtutorial/browser.html

Ellis, D. G., & Fisher, B. A. (1993). Small group decision making. New York: McGraw-Hill.

Festinger, L. (1953). Group attraction and membership. In D. Cartwright & A. Zander (Eds.), Group dynamics: Research and theory (pp. 92-101). Evanston, IL: Row, Peterson.

Forsyth, D. (1999). Group dynamics. Pacific Grove, CA: Brooks/Cole.

Gillies, R. M. (2003). The behaviors, interactions, and perceptions of junior high school students during small-group learning. Journal of Educational Psychology, 95, 137-147.

Hare, A. P. (2003). Roles, relationships, and groups in organizations: Some conclusions and recommendations. Small Group Research, 34(2), 123-154.

Jackson, J. M., & Williams, K. D. (1985). Social loafing on difficult tasks: Working collaboratively can improve performance. Journal of Personality and Social Psychology, 10, 446-454.

Janis, I. L. (1983) Groupthink. In H. H. Blumberg, A. P. Hare, V. Kent, & M. Davies (Eds.), Small groups and social interaction: Vol. 2 (pp. 39-46). New York: Wiley & Sons.

Johnson, D. W., & Johnson, R. T. (1990). Social skills for successful group work. Educational Leadership, 47(4), 29-33.

Johnson, D. W., & Johnson, R. T. (1996). Cooperation and the use of technology. In D. H. Jonassen (Ed.), Handbook of research for educational communications and technology (pp. 1017-1044). New York: Simon and Schuster Macmillan.

Johnson, S. D., Suriya, C., Yoon, S. W., Berrett, J. V., & La Fleur, J. (2002). Team development and group processes of virtual learning teams. Computers & Education, 39, 379-393.

Jonassen, D. H. (1990). Thinking technology: Toward a constructivist view of instructional design. Educational Technology, 30(9), 32-34.

Jonassen, D. H., Peck, K., & Wilson B. G. (1999). Learning with technology: A constructivist perspective. Upper Saddle River, NJ: Prentice Hall.

Kagan, S. (1990). The structural approach to cooperative learning. Educational Leadership, 47(4), 12-15.

Kagan, J., Havemann, E., & Segal, J. (1984). Psychology: An introduction. New York: Harcourt Brace Jovanovich.

Kiesler, S., & Kraut, R. (1999). Internet use and ties that bind. American Psychologist, 54, 783-784.

Le Bon, G. (1896). The crowd: A study of the popular mind. New York: Macmillan. Retrieved September 9, 2005 from http://etext.lib.virginia.edu/etcbin/toccer-new2?id=BonCrow.sgm&images=images/modeng&data=/texts/english/modeng/parsed&tag=public&part=all

Lewin, K. (1948). Resolving social conflicts: Selected papers on group dynamics. New York: Harper.

Lickel, B., Hamilton, D. L., Uhles, A. N., Wieczorkowska, G., Lewis, A., & Sherman, S. J. (2000). Varieties of groups and the perception of group entitativity. Journal of Personality and Social Psychology, 78, 223-246.

Mero, L. (1998). Moral calculations: Game theory, logic, and human frailty. New York: Springer-Verlag.

Nelson, D. A. (1990) The design of everyday things. New York: Currency.

Neo, M., & Neo, K. (2002). Building a constructivist learning environment using a multimedia design project--a Malaysian experience. Journal of Educational Multimedia and Hypermedia, 11, 141-153.

Parritt, H. J. (2003). What is the Internet? Retrieved September 9, 2005, from http://www.kentlaw.edu/cyberlaw/resources/whatis.html

Peri, A., Barbarito, M., Barattoni, M., & Abraham, A. (2000). The dynamics and the interpersonal and intrapersonal relations within an isolated group in extreme environments. Small Group Research, 31, 251-274.

Piaget, J. (1928). Judgment and reasoning in the child. New York, Harcourt.

Piaget, J. (1973). To understand is to invent: The future of education. New York: Grossman.

Pierce, D. R. (1962). The dual function of the task group. Journal of Educational Sociology, 36, 159-162.

Shaban, S., & Head, C. (2003). E-learning classroom environment: Description, objectives, considerations and example implementation. International Journal on E-Learning 2(3), 29-35.

Slavin, R. E. (1990). Research on cooperative learning: Consensus and controversy. Educational Leadership, 47(4), 52-54.

Slavin, R. E. (1991a). Group grades make groupwork work: Response to Kohn. Educational Leadership, February, 89-91.

Slavin, R. E. (1991b). Synthesis of research on cooperative learning. Educational Leadership, 48(5), 71-82.

Steiner, I. D. (1972). Group process and productivity. New York: Academic Press.

Steiner, I. D. (1976). Task performing groups. In J. W. Thibaut, J. T. Spence, & R. C. Carson (Eds.), Contemporary topics in social psychology (pp. 393-422). Morristown, NJ: General Learning Press.

Thibaut, J. (1953). An experimental study of the cohesiveness of underprivileged groups. In D. Cartwright & A. Zander (Eds.), Group dynamics: Research and theory (pp. 102-120). Evanston, IL: Row, Peterson.

Thomas, W. (1928). The child in America. New York: Knopf.

Triplett, N. (1898). The dynamogenic factors in pacemaking and competition. American Journal of Psychology, 9, 507-533.

Tuckman, B. W. (1965). Developmental sequences in small groups. Psychological Bulletin, 63, 384-399.

Tuckman, B. W., & Jensen, M. A. C. (1977). Stages in small group development revisited. Group and Organizational Studies, 2, 419-427.

Tudge, J. (1999). Vygotsky, the zone of proximal development, and peer collaboration: Implications for classroom practice. In L. C. Moll (Ed.), Vygotsky in education: Instructional implications and applications of socio-historical psychology (pp. 155-172). New York: Cambridge University Press.

Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.

Wageman, R. (1999). Task design, outcome interdependence, and individual differences: Their joint affect on effort in task-performing teams. Group Dynamics: Research, Theory, and Practice, 3, 132-137.

Wentzell, K. R. (1991). Classroom competence may require more than intellectual ability: Reply to Jussim (1991). Journal of Educational Psychology, 83, 156-158.

Wheelan, S. A., Davidson, B., & Tilin, F. (2003). Group development across time: Reality or illusion? Small Group Research, 34, 223-245.

Wright, M. E. (1943). The influence of frustration upon the social relations of young children. Character and Personality, 12, 111-122.

TODD OHL 

Agilefox, USA

[e-mail]

WARD CATES

Lehigh University, USA

[e-mail]

Table 1 Group Entitativity Ratings Excerpted from Lickel, et al. (2000,
p. 227). Possible mean (M) ratings range from 1 (low) to 9 (high).
Standard deviation (SD) is also reported for each group.

                                       Entitativity Ratings
                                         U.S.       Poland
Group                                  M     SD    M     SD

Members of a professional sports team  8.27  1.07  7.49  1.83
Members of a family                    8.16  1.62  7.60  2.07
Friends who do things together         7.75  1.72  7.18  2.17
Members of an airline flight crew      6.54  1.91  6.71  2.24
Members of the same political party    5.59  2.31  5.94  2.07
Students enrolled in a class           5.18  2.10  6.20  2.00
Students at a university               4.75  2.31  4.70  2.20
People in the audience at a movie      3.27  2.10  2.58  1.75
People in line at the bank             2.40  1.82  2.23  1.53