J. EDUCATIONAL COMPUTING RESEARCH, Vol. 20(4) 329-343,1999

DOES COMPUTER TECHNOLOGY IMPROVE STUDENT LEARNING AND ACHIEVEMENT? HOW, WHEN, AND UNDER WHAT CONDITIONS? JOHN SCHACTER CHERYL FAGNANO Milken Family Foundation

ABSTRACT

There have been widespread claims that technology can make learning easier, more efficient, and more motivating. This article argues that ease and efficiency should not be the leading criteria for advocating and implementing computer technology in schools. The authors assert that to produce more meaningful learning, computer technologies need to be designed according to sound leaning theories and pedagogy. When administrators, teachers, and parents understand that different computer technologies serve and augment different learning experiences, they can make informed judgments about which technologies are best suited to enhance student learning and achievement.

For the past four years, education has been captivated with computer technology. Advocates claim that technology makes learning easier, more efficient, and more motivating. They argue that telecommunications and Internet technologies will revolutionize teaching and learning because these devices widen the audience of the insular classroom to a community of “real world” learners. These advocates further rely on public opinion polls that demonstrate that politicians, administrators, teachers, and parents support the funding for and use of computer technology in America’s schools [ 11. While some of these calls for computers in schools are well intentioned and perhaps are necessary to secure popular public support for education technology, it is essential that educators and policymakers recognize and understand the complex nature of technology’s impact on student outcomes. It is not a one-to-one equation where the public provides funding for school-based technologies and today’s students become tomorrow’s “knowledge workers.” Further, the public 329 0 1999,Baywood Publishing Co., Inc.

doi: 10.2190/VQ8V-8VYB-RKFB-Y5RU http://baywood.com

330 / SCHACTER AND FAGNANO

should not accept the rhetoric that technology makes learning easier and more efficient because ease and efficiency are not prerequisite conditions for deep and meaningful learning. John Holt reminds us that children need to get an accurate sense of the long and effortful processes by which good learning occurs [2]. The only “way that children can learn how much time and effort it takes to build something, say, a table, is to be able to see someone building a table from start to finish. Or painting a picture. Or repairing a bicycle, or writing a story, or what ever it may be” [2, p. 1301. Neil Postman adds that we “learn more by failing, by trial and error, by making mistakes, correcting them, making more mistakes, correcting them, and so on. We are all in need of remedial work, all the time” [3, p. 1191. Finally, K. Anders Ericsson’s studies of expertise show that it takes ten years to become an expert, that over the course of those ten years that one needs to deliberately practice for four hours each day (14,600 hours), and, that during that practice one needs to be tutored by the best experts in the field [4]. These theorists and social critics tell us that learning is not easy or efficient, and that making it seem like it is, is a disservice to educating young people. For schools to produce “knowledge workers” (e.g., people who can plan, analyze, and design in complex problem domains [S, p. 41, future learning technologies need to support the development of these knowledge leaming processes. In this article we demonstrate that Computer Based Instruction (CBI), the most widely implemented and studied computer technology, moderately improves student learning. We then make the more important distinction that computer technologies, when designed according to sound learning theory and pedagogy, have and can substantially improve student learning. We conclude that teachers, administrators, policymakers, and parents need to understand the learning theories and principles around which the technology is designed in order to select and implement appropriate technologies that will have a significant impact on student achievement.

COMPUTER BASED INSTRUCTION Computer Based Instruction (CBI) individualizes the educational process to accommodate the needs, interests, proclivities, current knowledge, and learning styles of the student. CBI has been shown to moderately improve student learning and achievement. Over the past fifteen years, several meta-analytic research studies have demonstrated that CBI improves student achievement across multiple leaming domains [6]. Meta-analytic research combines data from multiple similar single research studies (in this case CBI studies) to generate a single effect size coefficient that illustrates the treatment effect across all the studies [7]. Table 1 reports the results from twelve CBI meta-analytic studies based on a total of 546 individual studies. In the studies reported in Table 1, effect sizes range from .25 to .57. Translating these effect sizes, one can see that students in CBI conditions performed at the

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 331

Table 1. Findings from Twelve Meta-Analyses on Computer-Based Instruction Number of Studies Analyzed

Average Effect Size

Meta-Analysis

InstructionalLevel

Bangert-Drowns, J. Kulik, & C. Kulik (1985)

Secondary

51

.25

Burns & Bozeman (1981)

Elementary & Secondary School

44

.36

Cohen & Dacanay (1991)

Health Professions Education

38

.46

Hartley (1978)

Elementary & Secondary Math

33

.41

Fletcher (1990)

Higher Education & Adult Training

28

.50

C. Kulik & J. Kulik (1986)

College

119

.29

C. Kulik, J. Kulik, & Shwalb (1986)

Adult Education

30

.38

J. Kulik, C. Kulik, & Bangert-Drowns (1985)

Elementary

44

.40

Niemiec & Walbert (1985)

Elementary

48

.37

Roblyer (1988)

Elementary to Adult Education

82

.31

Schmidt, Special Education Weinstein, Niemiec, & Walberg (1985)

18

.57

Willett, Yamashita, & Anderson (1983)

11

.22

Pre-College Science

Note: Table excepted from James A. Kulik [S].

332 I SCHACTER AND FAGNANO

56th to 72nd percentile compared to students in the control who performed at the 50th percentile. These are moderate positive gains that show that students who engaged in Computer Based Instruction performed significantly better than students who did not. In a mofe recent review, Jay Sivin-Kachala analyzed 219 research studies from 1990 to 1997 to assess the effect of computer technology on learning and achievement across all learning domains and all ages of learners [8]. His analysis revealed that students in technology rich environments experienced positive effects on achievement in all major subject areas. Further, students in technology rich environments showed increased achievement in preschool through higher education, and their attitudes toward learning improved consistently when computers were used for instruction. Finally, Dale Mann’s study of the state of West Virginia’sBasic SkillsKomputer Education (BS/CE) program analyzed a representative sample of 950 fifth grade students’ achievement from eighteen elementary schools across the state [9]. These fifth grade students had been participating in West Virginia’s BS/CE program since 1991-92. Data was also collected from 290 teachers to show the influence that West Virginia’s Computer Based Instructional technology had on student achievement.The technology focused its teaching on spelling, vocabulary, reading, and mathematics. Several variables were collected and analyzed (i.e., intensiveness of Basic Skills/ComputerEducation (BS/CE), student prior achievement and sociodemography,teacher training, teacher and student attitudes toward BS/CE). The findings for West Virginia’s statewide initiative were as follows: 1) the more students participated in BS/CE, the more their test scores rose on the Stanford 9; 2) consistent student access to the technology, positive attitudes toward the technology (by both teachers and students), and teacher training in the technology led to the greatest student achievement gains; 3) all students’ test scores rose on the Stanford 9 because of BS/CE, with lower achieving student scores rising the most; and 4) half of the teachers in the sample thought that technology had helped a lot with West Virginia’s instructional goals and objectives. These teachers also reported that they became more enthusiastic about BS/CE as time passed. Yet, these moderate positive gains have come under much criticism from the educational community. One criticism is that Computer Based Instruction is one among many kinds of computer technologies refemng mostly to drill and practice and simple tutorial programs. There are numerous educational computer technologies that are studied far less, but may have far greater learning potential and impact. Today, most learning and technology researchers would argue that Web based collaborative technologies, authoring and programming applications, intelligent tutoring systems, simulations, modeling programs, and productivity tools provide a richer computer environment for learning [lo-141. These newer technologies differ from CBI applications because instead of being based strictly on behaviorist

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 333

theories of learning, they are based on socio-cultural theories, constructivist theories, and cognitive science. We discuss below how theories from the cognitive revolution have shaped newer technology designs.

THEORY DRIVEN COMPUTER TECHNOLOGIES When computer technologies are designed around principles gleaned from learning theories and implemented systematically, one can argue that the effects that these technologies have on student learning and achievement are both powerful and transfonnative. Technologies designed around educational and psychological theory compare favorably to other education refom efforts because they have embedded proven teaching principles into the technology. Thus, one gets the effects of both the teaching reform and the technology. We evidence this point by briefly reviewing three learning theories and then reviewing research on technologies that were designed around these theories. Because of space constraints, we recognize that our theory reviews may in some instances oversimplify some theoretical underpinnings.

Socio-Cultural LearningTheory Jean Piaget and Lev Vygotsky have been credited as the forefathers of sociocultural theories of learning. Piaget posited that learning and development occur during cooperative socialization between peers [ 151. When children interact with other children, they are agents and recipients of instruction. Meanings that each child brings to an event are negotiated, and the interactive procedures for presenting, listening, and compromising result in children’s ceconstructed production of meaning [16, 171. Children, according to Piaget, discover meaning based on their shared similarities with, as well as their differences from, other children. To a large extent, Piaget’s theoretical observations have been well supported in the cooperative group learning literature. It is a widely accepted fact that cooperative learning boosts achievement [18-20]. The reasons underlying successful learning in cooperative group conditions are that students are exposed to and have the opportunity to discuss each other’s ideas, opinions, and beliefs. During these interactions, conflict may occur. That conflict, in turn, may drive the child to question his or her beliefs and to seek and generate alternative explanations to help reshape his or her understanding [15, 21, 221. Conflict has been found to also trigger students in giving explanations. Giving explanations in small-group learning has been widely documented as a behavior that improves student achievement and understanding in a number of academic domains [23-261. Through both conflict and giving and receiving explanations, children build on each other’s ideas. As Roschelle argues, these conversational interactions provide a means for students to develop increasingly sophisticated approximations of concepts collaboratively by gradual refining concepts and principles that are ambiguous at first [27].

334 / SCHACTER AND FAGNANO

Lev Vygotsky’s [28] work also stressed that learning occurs in social contexts. His work differed from Piaget in that he argued that learning is best when it takes place with the guidance of more capable partners. Vygotsky maintained that instruction should emphasize 1) developing goaldirected learning environments, with 2) mediational tools, and 3) the opportunity for interaction with an expert [29]. Vygotskian theory has been applied to apprenticeship instructional techniques and in situations where students collaborate with more capable partners. Research in apprenticeship learning has demonstrated that learners taught by more experienced and knowledgeable adults 1) receive more total information than learners taught by children [30]; 2) are more concerned with problem definition and efficiency than children planning alone or with a peer [31]; 3) are twice as likely to expIore the problem before tackling it; and 4) are significantly more likely to state optimal planning strategies than peer dyads [32, 331. Under apprenticeship conditions, similar to those in cooperative learning, students achieve more than students working individually or students receiving traditional whole class instruction [23,28,34-371. Although Vygotsky’s theories differed from Piaget’s in terms of who one should collaborate with, both positions emphasized collaborative learning contexts, and both theories emphasized the various collaborative process behaviors through which increased knowledge and understanding was fostered.

Computer-SupportedCollaborative Learning (CSCL) Driven by the theoretical ideas of Piaget and Vygotsky and the success of cooperative learning and apprenticeship models of instruction, researchers who work in a Computer Supported Collaborative Learning framework have developed several successful collaborative technologies that enhance student achievement. CSCL contexts typically involve multiple methods of computer-based communication combined with the capacity for students to look at and manipulate digital text and objects collectively [13]. For instance, students at their computers can manipulate text, a three dimensional object, graphics, sound, or video. While manipulating these artifacts, students can simultaneously or reflectively engage in discussion, debate, and present and argue about ideas. Scardamalia and Bereiter’s Computer Supported Intentional Learning Environments (CSILE), the most widely studied CSCL application in schools today, has entire classrooms of children conceive, respond to, and reframe what is said and written over time [14]. CSILE students ask questions, search for other students’ answers to their questions, comment on and review other students’ work, and then restructure and formulate answers to their original inquiries. This knowledgebuilding model provides a mechanism by which a student can pursue hidher own learning goal, diagnose his or her learning needs, collaborate with others, and then identify the next steps based on other students’ criticism and help. Eight years of

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 335

research on CSILE has demonstrated that CSILE maximizes student reflection and encourages progressive thought, taking multiple perspectives, and independent thinking. CSILE students surpass students in control classrooms on measures of depth of understanding, reflection, and also on standardized reading, language, and vocabulary tests. Researchers Frederiksen and White have developed collaborative computational tools (Thinkertools) for middle school students that introduce a method for promoting reflective conversation during collaborative scientific research [38]. These researchers reasoned that having students engage in computer-based selfand peer-assessment of their work, reflect on the goals of their work, and reflect on the intellectual processes involved in implementing their goals were critical elements of learning. Empirical evidence collected by Frederiksen and White has shown that students who used their Thinkertools software evidenced significant improvement in the inquiry cycle compared to control groups; produced better research projects than students in the control groups; scored higher on post-test assessment of genetics learning than students in the control groups; and were better able to judge quality research projects than their control group counterparts [38]. In the development of his software called SenseMaker, Philip Bell set out to create a collaborative computing environment in which students could express and reflect on their conceptual ideas about phenomena, explore and compare their ideas to those of others, and make sound discriminations between the set of models under consideration [39]. SenseMaker allowed groups of students to organize and annotate their research and conversations by electronic means. Middle school students using SenseMaker showed significant progress toward appropriating correct scientific conceptions. They were better able to relate evidence into their arguments than students not exposed to SenseMaker. Finally, students who used SenseMaker were able to construct more scientific conjecture in their evidence explanations, accrued more evidence to defend their arguments, and were better able to communicate their understanding than students in the control group. Collectively, this body of empirical research shows that CSCL software that is based on socio-cultural theories of learning improves student learning, student reflection, student projects, and students’ ability to judge other students’ and their own work.

Constructivist Theories Driven by the Dewinian imperative that knowledge cannot be transmitted or conveyed ready-made, constructivist theories state that children need to actively and personally construct meaning from knowledge [a]. Learning is seen to occur when the learners’ expectations are not met, and he or she must resolve the discrepancy between what was expected and what was actually encountered. Thus

336 / SCHACTER AND FAGNANO

the learning is in the individual’s construction as he or she attempts to resolve the ambiguity or conflict [41]. Learning is thus the active process of iterative cognitive reorganization. That reorganization takes place when one builds representations or conceptual models, designs something, establishes connections between what one knows and what one is learning, generates analogies, and creates and develops projects. During constructivist learning, the child engages with the learning material in a meaningful and personally relevant way, and through engaging builds his or her knowledge over time [42]. Learning by design and project-based learning have been popular methods of applying constructivist theory to successful practice. Learning by designing something fosters critical thinking, judgment, and personal involvement [43-451. Designing requires one to ask what is the design’s purpose? What is its function? What are models and examples, and does the design work? Designing puts students in charge and engages them in a continuous dialogue with their own ideas and with the ideas of the intended user [42, 461. In design activities, students assume control of their learning through putting the pieces of the design together. And as their learning grows, the complexity of their expression grows. Project-based learning requires a question or problem that serves to organize and drive activities. These activities result in a series of artifacts or products that culminate the final project and address the driving question. During project-based learning, students refine questions, make predictions, design plans, collect and analyze research, draw conclusions, communicate findings, etc. Student learning is contextualized through doing a meaningful project [44,47-501.

The Computer as a Tool for Design and Project-BasedLearning The Epistemology and Learning Group at MIT have employed constructivist ideas to educational technology by having students become creators and designers of educational software. These researchers use the computer as the machine to be acted upon and students as the actors. Thus, children learn through design activities by programming computers to create applications that other children use and learn from. Designing software for other students makes learning instrumental to a larger intellectual and social community [44]. Research by Idit Hare1 had children program in Logo (a computer language written specifically for children) to design games to teach mathematics to younger students. Students had to structure their program, maintain connections between content and functionality, and design the user interface and activities. In addition, students needed to consider different ideas about how to teach fractions to younger students. As opposed to previous studies which had used Logo to teach programming, Harel’s study used Logo to teach students how to design educational software [43]. Her research demonstrated that students simultaneously learned Logo programming, fractions, and design principles within one integrated context.

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 337

A follow up study by Harel and Papert showed that students who designed fraction software for other students using Logo learned fractions better than students taught fractions using conventional methods “1. Harel and Papert also found that students who used Logo to design software learned Logo better than students who received Logo programming instruction only. The University of Michigan’s Hi-C research group Foundations of Science (FOS) program was a three-year project-based learning science cumculum that used technology on a daily basis to support students as they engaged in authentic science inquiry [52, 531. One project of the FOS program had students focus on stream water quality by collecting biological organism data, performing chemical tests on water samples, and doing physical assessments of the soil, vegetation, and topology near the stream. Students carried out this research from beginning to end through the use of technology such as MBL (microcomputer-based laboratory) probes and hand-held Apple Newton computers. At the completion of the first year of water quality tests, students presented their findings via the Web and broadcast television to the Huron River Watershed Council for inclusion in their report to the Michigan Department of Natural Resources. After hearing of the water quality testing with FOS students were involved in, the Ann Arbor Transportation Authority (AATA) requested the help of FOS students in assessing the water quality of a drainage pond outside of their headquarters. Along with all the benefits of conducting real research, analyzing data, and publishing and presenting their results, FOS students were further compared to students in a traditional biology course through both a standardized test and a performance based biology test. Although they had covered the same material, the students in the biology class were given lectures, while students in FOS worked on projects. The FOS students demonstrated their deeper knowledge of the material by performing equally well on the multiple choice questions, and creating essays which were far more sophisticated with regard to interpreting data and reaching conclusions than those of students in the regular Biology class. Using computers as tools to design with and also as instruments to collect, analyze, report and publish results of projects has been a particularly interesting way that technology can increase student learning and motivation to learn.

Cognitive Science At the heart of theories of cognitive science is the drive to “describe with precision the behavior or thought processes of an organism engaged in learning” [54, p. 181. To do this, cognitive scientists have focused on how the mind processes symbols, how information is encoded, manipulated, stored, retrieved, and then used. John Bruer has argued that cognitive science research has led to the identification of numerous learning processes including those needed to read, write, and develop mathematical and scientific understanding [SS. 561. We agree,

338 / SCHACTER AND FAGNANO

but cognitive science’s shortcoming become evident when one tries to instruct students in how and when to activate these various processes to improve their learning. One solution to these learning and instruction issues has been to examine experts’ representations as well as students’ (novices’) representations, and then remediate the novice’s learning by teaching them expert strategies and exposing them to expert representations. In their book, The Nature of Experhe, Michelene Chi, Robert Glaser, and Marshall Fan analyzed expert and novice learning across multiple domains and multiple problem solving scenarios [57]. They found the following: Experts perceive large meaningful patterns and can then use those patterns to reduce the cognitive load to solve problems. Experts solve problems faster and with less error than novices. Experts have superior short-term memory. Experts represent problems at deeper more principled levels than novices. Experts organize based on principles whereas novices organize based on literal objects. Furthermore, experts run mental models of complex systems and can define component part interrelations as well as the systemic whole. Experts spend a great deal of time analyzing the problem, novices tend to plunge in immediately. Experts have s m n g monitoring skills. They are more aware than novices of when they make errors, why they fail to comprehend, and when they need to check their solutions. To teach students these expert ways of thinking, computer scientists and educational technology researchers have developed Intelligent Tutoring Systems (ITS).

Intelligent Tutoring Systems Intelligent Tutoring Systems are designed around three principles: 1) the knowledge of the domain (expert model), 2) knowledge of the student (student model), and 3) knowledge of teaching strategies (tutor) [58-611. An ITS application starts by assessing what the student knows, the student model. The system concurrently considers what the student needs to know, the content domain or the expert domain. Finally, the system must decide which content and how that content is to be taught, the tutor [62]. From all of these considerations, the computer application selects or generates a problem, the student then works on the problem as the ITS compares the student’s solution to the expert solution in real time. ITS then provides appropriate and “intelligent” feedback to the student to help that student appropriate the expert model. After the feedback, the program updates the student skills model and increments the learning progress indicators. These updating activities modify the student model and the entire cycle is repeated. Perhaps the most widely studied Intelligent Tutoring Systems were developed by John Anderson and his colleagues. Anderson reports that ITS applications have been successful at improving student learning across a variety of mathematical subject areas [63]. The geometry proof tutor (GP Tutor), developed by Anderson’s

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 339

group, provides direct evidence of this assertion. During a two-year period, observations of seven different classrooms in which the tutor was’ used were conducted in order to study the effects of a large-scale implementation of ITS within actual classrooms. Both qualitative and quantitative data were collected. These data showed that students who used Geometry Tutor learned more geometry than control groups who were instructed via whole classroom methods. GP Tutor also increased student motivation and decreased the amount of discipline problems. Several other controlled evaluations have been conducted on ITS software. In these studies, students either received the ITS software or learned the domain through whole class instruction. The ITS application reviewed here include: the LISP Tutor (for instructing LISP programming), Smithtown (a discovery world that teaches scientific inquiry skills), Sherlock (a tutor for Avionics instruction), Bridge (which teaches Pascal programming), and Stat Lady (a tutor that teaches introductory statistics). Taken collectively, these Intelligent Tutoring Systems showed that treatment groups receiving ITS learn significantly more than the control groups, gained deeper understanding than the control groups, learned the materials faster than control groups, and that that accelerated learning had no degradation on final outcome performance [62].

CONCLUDING REMARKS

In this article we have made an effort to show that computer technologies are most effective when they are designed according to different educational and psychological theories and principles. We have shown that different technologies designed around different learning theories can each be effective in improving student achievement. This argument, although sensible, receives far less attention than reports that focus on one type of computer technology and the silver bullet answer that technology will reform America’s schools. We conclude with a vignette from Charles MacKay’s 1841 classic Extraordinary Popular Delusions: The Madness of Crowds [MI. MacKay documented delusional behaviors that swept up entire societies. We use his historical analysis of the crowd mentality as a warning against the same fate for education technology. MacKay’s example of the South Sea Trading Company illustrated how thousands of people invested in stocks where no companies existed. The people of England in the seventeenth century were fascinated with the notion of earning a fortune without considering or caring how that fortune would be built and maintained. England’s South Sea Trading Company turned into such mass madness that in 1734, one thousand people owned stock in a company called: “For carrying on an undertaking of great advantage; but nobody to know what it is.”

340 / SCHACTER AND FAGNANO

Let us not “carry on an undertaking of great advantage; but nobody to know what it is” with computers in education. Let us implement computer technologies that endorse and exhibit the learning that we want all children to engage in.

REFERENCES 1. V.Edwards (ed.). Technology Counts, Education Week 1 8 5 , 1998. 2. J. Holt, Learning All the Time, Addison-Wesley,New York. 1989. 3. N. Postman, The End of Education: Redefining the Value of School, Knopf. New York, 1995. 4. K. A. Ericsson and A. C. Lehmann. Expert and Exceptional performance: Evidence

on Maximal Adaptations on Task Constraints, Annual Review of Psychology, 47, pp. 273-305,19%. 5 . D. C. Engelbart, The Augmentation Papers: A Collection Since 1960, Bootstrap Institute, Fremont, California, October 1994. 6. J. A. Kulik, Meta-Analytic Studies of Findings on Computer Based Instruction, in Technology Assessment in Education and Training, E. L. Baker and H. F. ONeil, Jr. (eds.), Lawrence Erlbaum. Hillsdale. New Jersey, 1994. 7. G. V. Glass, B. McGaw, and M. L. Smith, Meta-Analysis in Social Research, Sage Publications, Beverly Hills, California. 198 1. 8. J. Sivin-Kachala, Report on the Effectiveness of Technology in Schools. 1990-1997, Software Publisher’s Association, 1997. 9. D. Mann, C. Shakeshaft, J. Becker, and R. Kottkamp, West Virginia’s Basic SkilldComputer Education Program: An Analysis of Student Achievement, Milken Family Foundation, Santa Monica, California, 1999. 10. A. Kay, Observations about Children and Computers, paper presented at the meeting of For a New Social Science, Boca Raton, Florida, May 1994. 11. M. Riel, Cross Classroom Collaboration: Communication and Education, in CSCL: Theory and Practice of an Emerging Paradigm, T. Koschmann (ed.), Erlbaum, Mahwah, New Jersey, 1996. 12. S. Papert, The Connected Family: Bridging the Digital Generation Gap, Longstreet Press, Atlanta, Georgia, 1996. 13. R. D. Pea, Seeing What We Build TogethK Distributed Multimedia Learning Environments for TransformativeCommunications (Special Issue), The Journal of the Learning Sciences, 3:3, pp. 285-299, 1993. 14. M. Scardamalia and C. Bereiter. Computer Support for Knowledge-Building Communities, in CSCL: Theory and Practice of M Emerging Paradigm, T. Koschmann (ed.), Erlbaum, Mahwah, New Jersey, 1996. 15. J. Piaget, TheMoral Judgment offhe Child Free Press, New York, 1965. 16. W. Damon and M. Killen, Peer Interaction and the Process of Change in Children’s Moral Reasoning, Menill-Palmer Quarterly, 28, pp. 347-367, 1982. 17. J. Youniss, Parents and Peers in Social Development, University of Chicago Press, Chicago, 1980. 18. D. W. Johnson and R. T. Johnson, Cooperative Learning and Achievement, in Cooperative Learning: Theory and Research, S . Shlomo (ed.), Praeger, New York, 1990.

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 341

19. S. Sharan and A. Shaulov, Cooperative Learning. Motivation to Learn. and Academic Achievement, in Cooperative Learning: Theory and Research, S . Shlomo (ed.), Praeger. New York, 1990. 20. R. E. Slavin, Effects of Student Teams and Peer ‘Tutoring on Academic Achievement and Time-On-Task, Journal of Experimental Education, 48, pp. 252257,1980. 21. P. De Paolis, W. Doise, and G. Mugney, Social markings in Cognitive Operations, in Current Issues in European Social Psychology, Vol. 2, W. Doise and S. Moscovici (eds.), Cambridge University Press, Cambridge, England, pp. 1-45, 1987. 22. E. A. Forman and C. B. Cazden, Exploring Vygotskian Perspectives in Education: The Cognitive Value of Peer Interaction, in Culture, Communication and Cognition: Vygotskianperspectives, J. V. Wertsch (ed.). Cambridge University Press, New York, pp. 323-347, 1985. 23. A. L. Brown and A. S. Palinscar. Guided, Cooperative Learning and Individual Knowledge Acquisition, in Knowing, Learning and Instruction: Essays in Honor of Robert Glaser, L. B. Resnick (ed.),Erlbaum, Hillsdale, New Jersey, pp. 393-451, 1989. 24. G. B. Saxe, Studying Children’s Learning in Context: Problems and Prospects, Journal of the LRarning Sciences, 2, pp. 215-235, 1992. 25. N. M. Webb, Task-Related Verbal Interaction and Mathematics Learning in Small Groups, Journal for Research in Mathematics Education, 22, pp. 366-389, 1991. 26. N. M. Webb, J. D. Troper, and R. Fall, Constructive Activity and Learning in Collaborative Small Groups, Journal of Educational Psychology, 87, pp. 406-423, 1995. 27. J. Roschelle. Learning by Collaborating: Convergent Conceptual Change, in CSCL: Theory and Practice of an Emerging Paradigm, T. Koschmann (ed.), Erlbaum, Mahwah, New Jersey, pp. 209-248, 19%. 28. L. S . Vygotsky. Mind in Society: The Development of Higher Psychological Processes, Harvard University Press, Cambridge, Massachusetts, 1978. 29. J. V. Wertsch, Vygotsky and the Social Formation of the Mind, Harvard University Press, Cambridge, Massachusetts, 1985. 30. S. Ellis and B. Rogoff, The Strategies and Efficacy of Child versus Adult Teachers, Child Development, 53, pp. 730-735, 1982. 31. M. Gauvain and B. Rogoff, Collaborative Problem Solving and Children’s Planning Skills, Developmental Psychology, 25, pp. 139-151, 1989. 32. B. Radziszewska and B. Rogoff, Influence of Adult and Peer Collaborators on Children’s Planning Skills, Developmental Psychology, 24, pp. 840-848. 1988. 33. B. Radziszewska and B. Rogoff, Children’s Guided Participation in Planning Errands with Skilled Adult or Peer Partners, Developmental Psychology, 27, pp. 381-389, 1991. 34. J. S. Brown, A. Collins, and P. Duguid, Situated Cognition and the Culture of Learning, Educational Researcher, 18, pp. 32-42, 1989. 35. B. Rogoff. Apprenticeship in Thinking: Cognitive Development in Sociocultural Activity, Oxford University Press, New York, 1990. 36. B. Rogoff, Developing Understanding of the Idea of Communities of Learners, Mind, Culture, anddctivity, I , pp. 209-229, 1994.

342 / SCHACTER AND FAGNANO

37. J. V. Wertsch and C. A. Stone, A Social Interactwnal Analysis of Learning Disabilities Remediation, paper presented at the International Conference of the Association for Children with Learning Disabilities. San Francisco, 1979. 38. J. R. Frederiksen and B. Y.White, Cognitive Facilitation: A Method for Promoting Reflective Collaboration, in Proceedings of CSCL '97,R. Hall, N. Miyake, and N. Enyedy (eds.). University of Toronto Press, Toronto, pp. 53-62, 1997. 39. P. Bell, Using Argument Representations to Make Thinking Visible for Individuals and Groups, in Proceedings of CSCL '97,R. Hall, N. Miyake, and N. Enyedy (eds.), University of Toronto Press, Toronto, pp. 10-19, 1997. 40. J. Dewey, Democracy andEducation, Macmillan, New York, 1916. 41. T. M. Duffy and D. J. Cunningham, Constructivism: Implications for the Design and Delivery of Instruction, in Handbook of Research for Educational Communications and Technology, D. H. Jonassen (ed.), Macmillan, New York, pp. 170-198. 1996. 42. S . Papert, The Children's Machine: Rethinking School in the Age of the Computer, Basic Books. New York, 1993. 43. I. Harel, Software Designed for Learning: Children's Construction of Meaning for Factions and Logo Programming, unpublished Ph.D. dissertation, Massachusetts Institute of Technology, Cambridge, 1988. 44. D. N. Perkins. Knowledge as Design: Critical and Creative Thinkingfor Teachers and Learners, Erlbaum, Hillsdale, New Jersey, 1986. 45. Y. Kafai. Minds in Play: Computer Game Design as a Contextfor Children's Learning, Erlbaum, Hillsdale, New Jersey, 1995. 46. M. Resnick, A. Bruckman, and F. Martin, Pianos not Stereos: Creating Computational Construction Kits, Interactions, 3, p. 6. 1996. 47. P. Blumenfeld, E. Soloway. R. Marx. J. Krajcik, M. Guzdial, and A. Palinscar, Motivating Project-Based Learning: Sustaining the Doing, Supporting the Learning, Educational Psychologist, 26, pp. 369-398. 1991. 48. B. J. S. Barmn, D. L. Schwartz, N. J. Vye, A. Moore, A. Petrosino, L. Zech, J. Bransford, and The Cognition and Technology Group at Vanderbilt, Doing with Understanding: Lessons from Research on Problem- and Project-Based Learning, The Journal of the Learning Sciences, 7314, pp. 271-31 1, 1998. 49. J. Erikson and R. Lehrer, The Evolution of Critical Standards as Students Design Hypermedia Documents, The Journal of the Learning Sciences, 7314, pp. 351-386, 1998. 50. J. Krajcik, P. C. Blumenfeld, R. W.Marx, K.M. Bass, J. Fredricks, and E. Soloway. Inquiry in Project-Based Science Classrooms: Initial Attempts Made by Middle School Students, The J o u m l of the Learning Sciences, 2314, pp. 313-350, 1998. 51. J. Harel and S. Papert, Software Design as a Learning Environment, in Constructionkm, I. Harel and S. Papert (eds.), Ablex, Norwood, New Jersey, pp. 41-84, 1991. 52. M. Huebel-Drake, M. Mouradian, E. Stem, and L. Finkel. Foundations of Science: A Three Year Integrated High School Science Curriculum, unpublished manuscript, Center for Highly Interactive Computing in Education, University of Michigan, 1997. 53. A. Pryor, E. Soloway, and The Hi-C Research Group, Foundations of Science: Using Technology to Support Authentic Science Learning, unpublished manuscript, Center for Highly Interactive Computing in Education, University of Michigan, 1997.

DOES COMPUTER TECHNOLOGY IMPROVE LEARNING? / 343

54. H. Gardner. The Mind’s New Science: A History of the Cognitive Revolution, Basic Books, New York. 1985. 55. J. T.Bruer, Schoolsfor Thought: A Science of Learning in the Classroom, MIT Press, Cambridge, 1993. 56. J. T. Bruer. Education and the Brain: A Bridge Too Far, Eiiucational Researcher, 268, pp. 4-16, 1997. 57. M. T. H. Chi, R. Glaser, and M. J. Farr (eds.), The Nature of Expertise, Erlbaum, Hillsdale, New Jersey, 1988. 58. S. P.Lajoie and S . J. Deny, Computers as Cognitive Tools, Erlbaum, Hillsdale, New Jersey, 1993. 59. M. C. Polson and J. J. Richardson, Foundations of Intelligent Tutoring Systems, Erlbaum, Hillsdale, New Jersey, 1988. 60. J. Psotka, L. D. Massey. and S. A. Mutter, Intelligent Tutoring Systems: Lesson Learned, Erlbaum, Hillsdale, New Jersey, 1988. 61. J. W. Regian and V. J. Shute (eds.), Cognitive Approaches to Automated Instruction, Erlbaum, Hillsdale, New Jersey, 1992. 62. V. J. Shute and J. Psotka, Intelligent Tutoring Systems: Past, Present, and Future, in Handbook of Research for Educational Communications and Technology, D. H. Jonassen (ed.), Simon and Schuster, New York, 19%. 63. J. R.Anderson, C. Boyle, and G. Yost, The Geometry Tutor, Proceedings of IJCAL-85, Los Angeles. California, pp. 1-7, 1985. 64. C. MacKay, Exrraordinary Popular Delusions and the Madness of Crowds, Crown Trade, New York, 1841.

Direct reprint requests to: Dr. John Schacter Milken Family Foundation 1250 Fourth Street, 4th Floor Santa Monica, CA 90401-1353