35.9 Cooperative Learning and Technology-Based Instruction

In order to enhance learning, technology must promote cooperation among students and create a shared experience. Technology-assisted cooperative learning exists when the instructional use of technology is combined with the use of cooperative learning groups. Students, for example, may be assigned to cooperative groups of two or three members and given a cooperative assignment to complete a task for which a technology is to be utilized. Positive interdependence is typically established at the terminal so that students are aware of their dependence on other group members in accomplishing their learning goals.

Adding technology to a lesson inherently increases the lesson's complexity. When students participate in technology-assisted instruction, they have the dual tasks of (a) learning how to use the technology (i.e., the hardware and software required by the lesson) and (b) mastering the information, skills, procedures, and processes being presented within the technology. When cooperative learning groups are used, students have the additional task of learning teamwork procedures and skills. The complexity may be worth it. Technology-assisted cooperative learning tends to be a cost-effective way of teaching students how to use technology, increasing academic achievement, giving learners control over their learning, creating positive attitudes toward technology-based instruction and cooperative learning, promoting cognitive development, and increasing social skills. Computers themselves promote cooperative interaction among learners. The composition of the group and the gender of the learners are factors that have been hypothesized to affect the success of technology-assisted cooperative learning. Through technology, individuals in different settings can be network-ed into electronic cooperative learning groups.

35.9.1 Cost Effectiveness

The use of cooperative learning increases the cost effectiveness of technology. Although the range of technology that could be used in schools is increasing yearly (Hancock & Betts, 1994), the cost of adopting new technologies is an inhibiting factor to its use. Ensuring that every student is Provided with the latest technology is beyond the financial resources of most school districts. Giving each cooperative learning group access to the latest technology is much more cost effective. An historical example is the adoption of

computers by schools. By having groups work at computers (instead of individuals), schools were able to reduce significantly the cost of obtaining and maintaining computers (Johnson & Johnson, 1985; Wizer, 1987).

35.9.2 Learning How to Use Technology

Cooperative learning may reduce hardware and software problems that decrease achievement when students work alone (Hativa, 1988). Students naturally form groups when learning how to use a new technology or software program (Becker, 1984). When technology-assisted lessons require complex procedures (such as learner-controlled lessons), cooperative learning promotes greater mastery of the procedures than does individualistic learning. Trowbridge and Durnin (1984) found that students working in groups of two or three seemed more likely to interpret program questions as the authors of the materials intended. Discussions of multiple interpretations tended to converge on the correct interpretation. Hooper (1992) reported that students were frustrated and could not master the computer-assisted lesson when they worked alone with a learner-controlled lesson. Dyer (1993) compared structured cooperative pairs, unstructured cooperative pairs, and individuals completing a computer-assisted series of math problem-solving lessons. Structured cooperative pairs communicated more frequently and used the computer more efficiently and skillfully than did the unstructured cooperative pairs or students in the individualistic condition. McDonald (1993) found that students in the learner-controlled/cooperative learning condition selected more options during the lesson and spent more time interacting with the tutorial than did the learnercontrolled/individual learning condition. When teachers wish to introduce new technology and new software programs of some complexity, they will be well advised to use cooperative learning. Hooper, Temiyakam, and Williams (1995) found that cooperative learning established a mutually supportive learning environment among group members in which both cognitive difficulties and navigational disorientation were overcome in using the computer to complete a symbolic-reasoning task. Students studying alone had greater difficulty reading and understanding lesson directions, used the help option more often, and required more attempts to master embedded quizzes than did students in cooperative learning groups. Generally, this evidence indicates that students will learn how to use hardware and software more quickly and effectively when they learn in cooperative groups rather than alone.

In learning how to use computers, Webb (1984a) and Webb, Ender, and Lewis (1986) found that, in cooperative groups, explaining how to do computer programming was not related to skill in doing so (see also 24.5.3), and receiving explanations only influenced the learning of basic commands (not the interpretation of programs or the ability to generate programs). Fletcher (1985), on the other hand, investigating cognitive facilitation, found on a computer task calling for solving equations in an Earth spaceship game that individuals who were told to verbalize their decisions did as well in problem-solving performance on the game as groups told to come to consensus (both of which had superior results to individuals working silently). King (1989) asked groups of fourth-graders to reproduce a stimulus design using LOGO computer graphics (see 24.5.1.3) after they had watched a videotape modeling of "think-aloud problem solving." The groups were instructed to think aloud as they performed their task. More successful groups asked more task-related questions, spent more time on strategy, and reached higher levels of strategy elaboration than did groups who were less successful on the task.

We conducted several studies examining the use of cooperative, competitive, and individualistic learning activities at the computer (D. Johnson, Johnson, Stanne & Garibaldi, 1989, 1990; R. Johnson, Johnson & Stanne, 1985, 1986; R. Johnson, Johnson, Stanne, Smizak & Avon, 1987; Richards, Johnson & Johnson, 1986). The studies included students from the eighth grade through college freshmen and lasted from 3 to 30 instructional hours. The tasks were a computerized navigational and map-reading problem-solving task and word-processing assignments. Computer-assisted cooperative learning, compared with competitive and individualistic efforts at the computer, promoted (a) higher quantity of daily achievement, (b) higher quality of daily achievement, (c) greater mastery of factual information, (d) greater ability to apply one's factual knowledge in test questions requiring application of facts, (e) greater ability to use factual information to answer problem-solving questions, and (f) greater success in problem solving. Cooperation at the computer promoted greater motivation to persist on problem-solving tasks. Students in the cooperative condition were more successful in operating computer programs. In terms of oral participation, students in the cooperative condition, compared with students in the competitive and individualistic conditions, made fewer statements to the teacher and, more to each other, made more task-oriented statements and fewer social statements, and generally engaged in more positive, task-oriented interaction with each other (especially when the social skill responsibilities were specified and group processing was conducted). Finally, the studies provided evidence that females were perceived to be of higher status in the cooperative than in the competitive or individualistic conditions.

In addition to our work, there are a number of studies that have found that students using a combination of cooperative learning and computer-based instruction learn better than do students using computer-based instruction while working individually (Cox & Berger, 1985; Dalton, 1990; Dalton, Hannafin & Hooper, 1987; Hooper, 1992; Hooper, Ten-dyakam & Williams, 1995; Hythecker, Rocklin, Dansereau, Lambiotte, Larson & O'Donnell, 1985; Love, 1969; Mevarech, Stem & Levita, 1987; Okey & Majer, 1976; Repman, 1993; Rocklin, O'Donnell, Dansereau, Lambiotte, Hythecker & Larson, 1985; Shlecter, 1990; Stephenson, 1992; Webb, 1984; Yueh & Alessi, 1988). There are also a number of studies that found no statistically significant differences in achievement between subjects who worked in groups and subjects who worked alone (Carrier & Sales, 1987; Cosden & English, 1987; Hooper & Hannafin, 1988; Trowbridge & Dumin, 1984). No study has reported significantly greater learning when students work alone.

Simon Hooper and his colleagues have conducted a series of studies on technology-assisted cooperative learning involving fifth- to eighth-graders and college students (Dyer, 1993; Hooper, 1991; Hooper & Hannafin, 1988a, 1988b, 1991; Hooper, Ward, Hannafin & Clark, 1989; Huang, 1993; McDonald, 1993). They found that: (a) Cooperative group members achieved significantly higher than did students working under individualistic conditions; (b) cooperative learning groups in which individual accountability was carefully structured achieved higher than did cooperative learning groups in which no individual accountability was structured; (c) the achievement of lowability students in heterogeneous cooperative groups was consistently higher than the achievement of low-ability students in homogeneous groups; (d) a positive and significant correlation was found between achievement and helping behaviors, and increases in achievement and cooperation to be significantly related within heterogeneous groups; and (e) cooperative (compared with individualistic) learning resulted in greater willingness to learn the material, options selection, time on task, perceived interdependence, and supportiveness for partners. Carlson and Falk (1989) and Noell and Camine (1989) found that students in cooperative groups perform higher than students working alone on leaming tasks involving interactive videodiscs. Adams, Carson, and Hamm (1990) suggest that cooperative learning can influence attention, motivation, and achievement when students use the medium of television.

Combining cooperative learning and technology-assisted instruction results in students having more control over their learning (see 33.1 to 33.3). Simon Hooper and his associates (Hooper, 1992; Hooper, Temiyakam & Williams, 1993) note that three forms of lesson control are used in the design of technology-based instruction: learner, program, and adaptive control. Learner control involves delegating instructional decisions to learners so that they can determine what help they need (see 33.2), what difficulty level or content density of material they wish to study, in what sequence they wish to learn material, and how much they want to learn. Program or linear control prescribes an identical instructional sequence for all students regardless of interest or need. Adaptive control modifies lesson features according to student aptitude (see 22.1; Snow, 1980), prior performance (see 22.3.4.4; Tobias, 1987), or ongoing lesson needs (e.g., Tennyson, Christensen & Park, 1984).

Of the three, learner control may be the most important, as Hooper (1992) notes that the field of technology-assisted instruction seems to be moving toward learner-controlled environments, such as simulations, hypermedia (see Chapter 21), and on-line databases. He suggests that as learner control increases so does instructional effectiveness and efficiency (Reigeluth & Stein, 1983) and learner independence, efficiency, mental effort, and motivation (Federico, 1980; Salomon, 1983, 1985; Steinberg, 1984). On the other hand, linear or program control may impose an inappropriate lesson sequence on learners and thereby lower their motivation, and adaptive instruction may foster learner dependence (Hannafin & Rieber, 1989).

Technology-assisted cooperative learning tends to increase the effectiveness of learner control. When students work alone, in isolation from their peers, they tend not to control the learning situation productively, making ineffective instructional decisions and leaving instruction prematurely (Carrier, 1984; Hannafin, 1984; Milheim & Martin, 1991; Steinberg, 1977, 1989). Students working cooperatively tend to motivate each other to seek elaborative feedback to their responses to practice items during learning control and to seek a greater variety of feedback types more frequently than did those working alone (Carrier & Sales, 1987). Additionally, the cooperative pairs spent longer times inspecting information on the computer screen as they discussed which level of feedback they needed and what the answers were to practice items. McDonald (1993) found that students in the learner-controlled/cooperative learning condition selected more options during the lesson, and spent more time interacting with the tutorial, than did students in the learner-controlled/individual learning condition. Hooper, Temiyakam, and Williams (1995) found that students in the prognam-control conditions attempted more than 4 times as many examples and nearly twice as many practice questions as did the students in the learner-control conditions. The LOGO computer environment (see 12.3.2. 1, 24.5.1.3) tends to promote more actual learner control over the task structure and the making of rules to govern it than does the CAI computer environment (Battista & Clements, 1986; Clements & Nastasi, 1985, 1988; Nastasi, Clements & Battista, 1990). Learner control seems to be most effective when prior knowledge is high or when students possess well-developed metacognitive abilities (Garhart & Hannafin, 1986). What these studies imply is that cooperative learning is an important variable in improving the effectiveness of learner-controlled environments.

35.9.5 Attitudes Toward Technology-Based Instruction

Cooperative learning tends to promote positive attitudes toward technology-based instruction. A key aspect of technology-assisted instruction is the student attitudes generated by the experience. Students are more likely to learn from 2nd to use technology-based instruction in the future when their self-efficacy toward technology and attitudes about technology-based instruction are positive (Sutton, 1991). Hooper, Temiyakarn, and Williams (1995) found that students developed more positive attitudes toward the computer-based instructional lesson when they worked in cooperative learning groups than when they worked individually. McDonald (1993) found that students developed more positive attitudes toward learning with a computer in the cooperative conditions than in the individualistic conditions. Huang (1993) found that students working cooperatively had more positive attitudes toward the computer-based lesson than did students working individually. Students appear to enjoy using the computer to engage in cooperative activities (Bonk, Southerly, Brantmayer & Smith, 1991).

35.9.6 Attitudes Toward Cooperative Learning

Closely related to the attitudes toward technology (see 34.6) are students' attitudes toward cooperative learning. Students with negative attitudes about cooperative learning may be less likely to invest effort in the group process and to engage in actions that mediate achievement. Mevarech et al. (1985) found that students who learned in pairs were more positive in their attitudes toward cooperative learning than were students who worked individually with the computer. Evaluations obtained by Rocklin et al, (1985) from students involved in computer-based cooperative learning were more positive towards cooperative learning and how it affected them personally than were subjects who worked individually- Hooper, Temiyakam, and Williams (1995) found that students working in cooperative pairs developed more positive attitudes toward cooperative learning than did students working alone. Hooper et al. (1993) found that students rated cooperative learning in a computer-assisted lesson almost a point higher on a five-point scale than did students who worked alone. Dyer (1993) found that students in the structured cooperative learning conditions developed more positive attitudes toward working cooperatively than did students in the unstructured cooperative learning or the individualistic learning condition. McDonald (1993) and Huang (1993) both found that students in the cooperative conditions developed more positive attitudes toward working cooperatively than students working alone. Thus, when technology-assisted instruction is used, students' attitudes toward the instructional experience will be more positive when cooperative learning is an inherent part of the lesson.

Social-cognitive theory posits that cognitive development is facilitated by individuals (Bearison, 1982; Johnson & Johnson, 1979, 1992; Perret-Clermont, 1980): (a) working cooperatively with peers on tasks that require coordination of actions or thoughts, (b) collaborators contradicting and challenging each other's intuitively derived concepts and points of view (i.e., engaging in academic controversy), thereby creating cognitive conflict within and among group members, and (c) the successful and equitable (members contributing approximately equally) resolution of those conflicts (leamers have to go beyond mere disagreement to benefit from cognitive conflict [Bearison, Magzament & Filardi, 1986; Damon & Killen, 1982]). In order to create the conditions under which cognitive development takes place, students must work cooperatively, challenge each other's points of view, and resolve the resulting cognitive conflicts. Douglas Clements and Bonnie Nastasi have conducted a series of studies on the occurrence of cooperation and controversy in technology-assisted instruction (Battista & Clements, 1986; Clements & Nastasi, 1985, 1988; Nastasi & Clements, 1992; Nastasi, dements & Battista, 1990). They have found that both LOGO (see 12.3.2.1, 24.5.1.3) and CAI/CBI-W computer environments promoted considerable cooperative work and conflict (both social and cognitive). The LOGO environment (compared to CAI/CBI-W computer and traditional classroom tasks environments) promoted (a) more peer interaction focused on leaming and problem solving, (b) self-directed problem solving (i.e., learners solve problems they themselves have posed) in which there is mutual "ownership" of the problem, (c) more frequent occurrence and resolution of cognitive conflicts, (d) greater development of executive level problem-solving skills (planning, monitoring, decision making), higher-level reasoning, and cognitive development. The development of higher-level cognitive processes seemed to be facilitated by the resolution of cognitive conflict that arises out of cooperating. They also found that the LOGO (compared with the CAI) computer environment resulted in more learner satisfaction and expressions of pleasure at the discovery of new information and their work, variables reflective of intrinsic and competence motivation. Clements and Nastasi conclude that the LOGO environment generally promotes the development of motivated, self-directed learners who seek to validate their ideas not only through their own reasoning but also through meaningful communication with others.

If students are to work effectively in cooperative groups, they must have the teamwork skills to do so. In order to examine the importance of social skills training on the productiveness of cooperative groups, it is possible to compare studies that have included cooperative skills training and those that have not. A number of studies have found that when teamwork procedures and skills are present, cooperative learning results in higher achievement, in technology-assisted instructional lessons than individualistic learning (Hooper & Hannafin, 1991; Hooper & Hannafin, 1988a; Hooper, 1991; Johnson R., Johnson & Stanne, 1985, 1986). In studies where teamwork procedures and skills were not emphasized, reliable differences in achievement i cooperative and individualistic technology-assisted instruction were not found (Meuarech, Stem & Leuita, 1987; Underwood & McCaffrey, 1990; Hooper, Ward, Hannafin & Clark, 1989).

Software designers may be able to facilitate the development use of the interpersonal and small-group skills required for teamwork in several ways:

1. Before students engage in the actual instruction, they might first be required to complete a tutorial activity designed to introduce or refresh their understanding of cooperative skills. This could include a discussion of each member's role and its value in determining the overall group success.
2. Teachers' guides can suggest roles to assign to each group member to perform in the group (keyboarder, recorder, checker for understanding, encourager of participation).
3. Allow time for group processing to analyze and discuss how effectively they are working together and how they may work more effectively together in the future. Software could be designed to include pauses during which group members are directed to focus on their progress, discuss the records they are keeping, or reflect on improvements or changes they might make to increase performance.
4. The software could periodically remind students to monitor their own performance and to assist in optimizing group performance.
5. Yueh and Alessi (1988) suggest that group reward is crucial to provide a group goal motivating everyone to work well together, and individual accountability is needed to create a feeling of fairness among group members. Tangible prizes are recognition for individual successes, and group achievement offers motivation to succeed on both levels. One computer-generated reward would be a printout of collective characters, coupons, or certificates that are assigned points or a relative value, or are valued based on the number accumulated. These items could be displayed by students where they would be acknowledged by the teacher and other classmates.

35.9.9 Preference for Using Technnology Cooperatively

There is a natural partnership between technology and cooperation. There is evidence that individuals prefer to work cooperatively at the computer (Hawkins, Sheingold, Gearhart & Berger, 1982; Levin & Kareev, 1980; Muller & Perlmutter, 1985). The introduction of computers into classrooms increases cooperative behavior and taskoriented verbal interaction (Chemick & White, 1981, 1983; Hawkins, Sheingold, Gearhart & Berger, 1982; Levin & Kareev, 1980; Rubin, 1983; Webb, 1984). Working at a computer collaboratively with classmates seems to be more fun and enjoyable, as well as more effective, to most students. Students are more likely to seek each other out at the computer than they normally would for other school work. Even when students play electronic games, they prefer to have partners and associates. The computer may not only be a good place to cooperate but may also be a good place to introduce cooperative learning groups in schools.

A factor hypothesized to effect the success of technologyassisted cooperative learning is whether cooperative learning groups are composed homogeneously or heterogeneously. There is considerable disagreement. Advocates of heterogeneous grouping point out that students are more likely to gain sophistication and preparation for life in a heterogeneous society by working cooperatively with classmates from diverse cultures, attitudes, and perspectives rather than by learning in homogeneous groups or studying alone. Proponents of heterogeneous ability grouping point out that (a) high-achieving students benefit by the cognitive restructuring that occurs when providing in-depth explanations to peers, and (b) less academically successful students benefit from the extra attention, alternative knowledge representations, and modeling that more academically successful students provide (Johnson & Johnson, 1989; Webb, 1989). Students in heterogeneous-ability groups learned morelthan did students in homogeneous-ability groups (Yager, Johnson & Johnson, 1985; Yager, Johnson, Johnson & Snider, 1986). Beane and Lemke (1971) found that high-ability students benefited more from heterogeneous than homogeneous grouping. The academic discussion and peer interaction in heterogeneous groups promotes the discovery of more effective reasoning strategies than would occur in homogeneous groups (Johnson & Johnson, 1979; Berndt, Perry & Miller, 1988).

Proponents of homogeneous-ability grouping, however, state that heterogeneous-ability grouping may fail to challenge high-ability students (Willis, 1990) and that less academically successful students benefit at the expense of their more successful partners (Mills & Durden, 1992; Robinson, 1990). Many of the most carefully conducted studies aimed at resolving this controversy have been focused on ability grouping in technologically-assisted instruction. Webb (1982) and Swing and Peterson (1982) reported that heterogeneous grouping hinders the performance of average-~performing students when groups include a wide range of student performance levels.

In a week-long study on the learning of LOGO, Webb (1984) investigated whether the higher-ability students in cooperative groups of three would try to monopolize the computer. She found that (a) student ability did not relate to contact time with the computer, and (b) student success in programming was predicted by different profiles of abilities and by group process variables such as verbal interaction. Yue-h and Alessi (1988) used group ability composition as one of their treatments for students utilizing the computer to learn three topics in algebra. They formed groups of medium-ability students and groups of mixed-ability students, and found that group composition had no significant effect on achievement.

In a study with 40 eighth-grade students, Hooper and Hannafin (1988a) had students work in cooperative groups of three or four which were classified as homogeneous low, homogeneous high, or heterogeneous. Students worked on a computer task. Low-ability students working with highability partners achieved higher than did low-ability students studying in homogeneous groups or alone, and the achievement of high-ability students was basically the same whether they worked with a low-achieving partner, a high-achieving partner, or studied alone.

Hooper and Hannafin (199 1) conducted a study involving 125 sixth- and seventh-grade students. Subjects were randomly assigned to homogeneous or heterogeneous pairs, and the pairs were randomly assigned to cooperative or individualistic conditions. The high-ability students interacted equally across treatments, but low-ability students interacted 30% more when placed in heterogeneous pairs. Increases in achievement and cooperation were significantly related within heterogeneous groups.

Simsek and Hooper (1992) compared the effects of cooperative and individual learning on student performance and attitudes during interactive videodisc instruction. Thirty fifth- and sixth-grade students were classified as high or low ability and randomly assigned to cooperative or individual treatments. Students completed a level 11 interactive videodisc science lesson. The achievement, attitudes, and time-on-task of high- and low-ability students working alone or in cooperative groups were compared. Results indicated that both high- and low-ability students performed better on the posttest when they learned in cooperative groups than did their counterparts who learned alone. Students who worked individually spent less time-on-task. Members of cooperative groups developed more positive attitudes toward instruction, teamwork, and peers than did students studying alone.

Simsek and Tsai (1992) compared the effects of homogeneous- versus heterogeneous-ability grouping on performance and attitudes of students working cooperatively during interactive videodisc instruction. After two cooperative training sessions, 80 fourth- through sixth-grade students, classified as high and low ability, were randomly assigned to treatments. Students completed a level 11 interactive videodisc science lesson. The amount of instructional time for each group was also recorded. Homogeneous lowability groups scored significantly lower than the other three groups, while the difference between achievement of high-ability students in homogeneous and heterogeneous groups was not statistically significant. Homogeneous low-ability groups consistently used the least amount of time. Low-ability students in heterogeneous groups had significantly more positive attitudes than did their highability groupma.tes.

Hooper (1992) compared individual and cooperative learning in an investigation of the effects of ability grouping on achievement, instructional efficacy, and discourse during computer-based mathematics instruction. A total of 115 fifth- and sixth-grade students were classified as having high or average ability and were randomly assigned to group or individual treatments. Students in the cooperative condition were assigned to either heterogeneous or homogeneous dyads, according to ability. Results indicated that students completed the instruction more effectively in groups than alone. In groups, achievement and efficiency were highest for high-ability homogeneously grouped students and lowest for average-ability homogeneously grouped students. Generating and receiving help were significant predictors of achievement, and average-ability students generated and received significantly more help in heterogeneous groups than in homogeneous ones.

Hooper, Temiyakam, and Williams (1995) compared cooperative and individualistic learning on academically high- and average/low-performing students. They classified 175 fourth-grade students as high or average/low performing academically and randomly assigned them to paired or individual conditions strategies by performance level. Performance level was determined by scores on the mathematics subscale of the California Achievement Test. High-performing students scored at or above the median, and average/low-ability students scored below the median score of all fourth-grade students in the school. All cooperative pairs consisted of one high and one average/low-performing student. They found that the students in the cooperative conditions performed higher on a cornputer-assisted symbolic reasoning task than did the students in the individualistic conditions. The greatest benefactors from the group learning experience appeared to be the highest-performing students. Overall achievement increased by almost 20% for high-academic-ability students, but only 4% for average-ability students. High-ability students may have benefited from generating explanations for their less-able partners. Less-able partners might have adopted more passive roles. Mulryan (1992) found that the highestachieving students adopted the more active roles in cooperative learning groups and the least-able students demonstrated high levels of passive behavior, a pattern that according to Webb (1989) further decreases the achievement of the passive students.

Siann and Macleod (1986) found that mixed-gender pairs working on a LOGO programming exercise were dominated by the males (females were less motivated and successful). Underwood and McCaffrey (1990) studied pairs of students (10 and 11 years of age) on a computer task filling in missing letters from words. They were not told how to work together. Single-sex pairs were more productive than mixed-sex pairs, who did not improve in performance over the individual performance of group members. Single-sex pairs worked by discussion and agreement, with each member of the pair contributing and both sharing keyboard control. In contrast, the mixed-gender pairs tended simply to divide the labor, with one taking over the keyboard and the other instructing the typist, with little discussion of or negotiation about alternative solutions.

The results of these studies indicate that cooperative learning may be used effectively with both homogeneous and heterogeneous groups, but that the greatest educational benefits may.be derived when heterogeneous groups work with technology-assisted instruction. In heterogeneous cooperative learning groups, low-ability students increased their achievement considerably, and high-ability students generally either increased their achievement or achieved at the same levels as did their counterparts in homogeneously high groups.

The gender of group members has been hypothesized to be an important factor in determining the success of technology-assisted cooperative learning. Johnson, Johnson, Richards, and Buckman (1986) found that computer-assisted cooperative learning, compared with competitive and individualistic computer-assisted learning, increased the positiveness of female students' attitudes toward computers, equalized the status and respect among group members regardless of gender, and resulted in a more equal participation pattern between male and female members. While females in cooperative groups liked working with the computer more than males did, there was no significant difference in oral interactions between males and females. Dalton et al. (1987) examined interactions between instructional method and gender and found that cooperative learning was rated more favorably by low-ability females than by low-ability males. Other studies noted no significant differences in performance between males and females in computer-based instruction cooperative learning settings (Mevarech, Stem & Levita, 1987; Webb, 1984). Carrier and Sales (1987) compared female pairs, male pairs, and mixed pairs among college juniors and noted that female pairs verbalized the most while male pairs verbalized the least, and that male-female pairs demonstrated the most off-task behavior. Lee (1993) found that males tended to become more verbally active and females tended to become less verbally active in equal-ratio, mixed-gender groups.

A study that looked at mixed-gender groups versus singlegender groups was done by Underwood and McCaffrey (1990) in England. Two classes of students between the ages of 10.5 years and 11.4 years from a single school were the subjects in this study. Forty girls and 40 boys were randomly assigned to three types of pairs: male/male, female/female, and male/female. Tie study was divided into three sessions. The first session had the subjects working individually. In the second session, subjects worked in pairs. The third session also involved pairs, but subjects who were in mixed pairs were shifted to single-gender pairs, and single-gender pairs were assigned to mixed pairsThe subjects worked with a computer program in language tasks that required them to place missing letters into text. The results showed that single-gender pairs completed more stories and had more correct responses than did niixed-gender pairs. When subjects were shifted from singlegender pairs to mixed-gender pairs, their level of activity decreased, but there was no change in their overall performance. The study found no overall differences for gender on any of the measures. No cooperative training of subjects was undertaken for this study, and it was found that mixed pairs rarely discussed their answers. Rather, one subject operated the keyboard and the other gave directions.

Overall, there is mixed evidence concerning the impact of technology-assisted instruction on males and females. A conservative interpretation of the existing research is that there will be no performance differences between males and females on technology-assisted cooperative learning, but females will have more positive attitudes toward using technology when they learn in cooperative groups.

35.9.12 Networking into Teams

Technology such as electronic mail, bulletin boards, and conferences can be used to create teams made up of individuals who are widely separated geographically. In an electronically networked team, interaction no longer has to be face to face; team members can be anywhere in the world. In electronically networked teams, members, may depend on one another differently than they do in face-toface teams, Meetings only require that members be at their terminals. Communication between meetings can be asynchronous and extremely fast in comparison with telephone conversations and interoffice mail. Participation may be more equalized and less affected by prestige and status (McGuire, Kiesler & Siegel, 1987; Siegel, Dubrovsky, Kiesler & McGuire, 1986). Electronic communication, however, relies almost entirely on plain text for conveying messages, text that is often ephemeral, appearing on and disappearing from a screen without any necessary tangible artifacts. It becomes easy for a sender to be out of touch with his or her audience. And it is easy for the sender to be less constrained by conventional norms and rules for behavior in composing messages. Communicators can feel a greater sense of anonymity, detect less individuality in others, feel less empathy, feel less guilt, be less concerned over how they compare with others, and be less influenced by social conventions (Kiesler, Siegel & McGuire, 1984; Short, Williams & Christie, 1976). Such influences can lead both to more honesty and more "flaming" (name calling and epithets).

35.9.13 The Need for Groupware

For cooperation to take place, students must have a joint workspace. One of the promises of the computer is to allow students all over the world to create powerful shared spaces: super blackboards and super models. Instead of sharing a blackboard or a worktable, people from a wide variety of locations can share a computer screen. The future of technology-assisted cooperative learning will be greatly enhanced by developing both appropriate software and hardware to create workspaces that may be shared by all members of a group, all groups within the same classroom (or school), and all groups in a network that stretches throughout the world. Increasingly, work is being done in self-managing teams, networked electronically with other teams throughout the company and the world. The ability of the hardware to allow or even require people to work cooperatively is an important design issue. Developers of hardware need to think seriously about how technology can increase human cooperation within education and within the workplace. In addition, a challenge facing softwiire programmers is to write groupware to support group rather than individual work. The availability of groupware will increase the productivity of joint efforts. In order to write such software, programmers need to understand the nature of cooperation and the five basic elements that mediate its effectiveness.