Planning for Two Transformations in Education and Learning Technology: Report of a Workshop (2003)

The first session of the workshop focused on the discussion of the first transformation: the challenge of integrating cheap, fast, robust computers into instruction for every student in America. The session began with three presentations. Barbara Allen and Darryl LaGace described the LemonLINK project for integrating computers into instruction in the Lemon Grove School District in California. Steve Rappaport of Advanced Networks and Services discussed some of the requirements for using technology to improve student learning. Geneva Henry of Rice University discussed the Connexions Project for creating a repository of curriculum modules in science, engineering, and mathematics. These presentations were followed by comments by Cheryl Lemke of the Metiri Group and Wanda Bussey of Rufus King High School in Milwaukee.

Barbara Allen, of Project LemonLINK,¹ opened the workshop’s first presentation with the observation that although millions of dollars have

been spent trying to implement technology in classrooms across the country over the past 10 years, too much of the education community is still waiting for it to happen. She and colleague Darryl LaGace proceeded to draw on their experience in the Lemon Grove School District to identify obstacles they encountered and to share what they suggested is a promising approach to realizing the benefits of technology-rich curriculum and instruction that could be applied in other school districts.

Lemon Grove is a community eight miles east of San Diego with 4,600 students in grades K-8, 60 percent of whom are eligible for free or reduced-price lunch. Approximately six years ago the district developed a vision for creating a truly connected learning community, with access to that community from anywhere in Lemon Grove, including classrooms, libraries, homes, and community centers. From the outset the designers of this on-line learning community saw easy and seamless access as pivotal to providing the same type of technology-enabled educational experience across all classrooms and to all students.

Their initial target for access to hardware was a ratio of one conventional computer to four students. After more than a year into the plan, it became clear that the 1:4 computer-to-student ratio was not making a difference in instruction. The computers remained literally and figuratively peripheral, while the amount of time the hardware or software was unusable or required special attention reinforced concerns that this approach to instruction and learning was unreliable. Those involved with developing this learning community concluded that unless they could achieve at least a 1:2 computer-to-student ratio, the traditional model of teacher at the head of the class, lesson-driven education would remain firmly in place. Today Lemon Grove has achieved a ratio of 1:2 and, the presenters contended, a transformed system of teaching and learning. Allen and LaGace proceeded to summarize the multiple organizational, technical, and economic obstacles their community faced and the strategies they adopted to overcome them.

First, Allen identified six challenges to integrating cheap, fast, robust computers into instruction for every student: reducing the cost of ownership; preparing teachers with high-quality, ongoing professional development; providing ready access to educational software linked to standards; involving parents and providing home access, including subsidized access; involving the people and organizations in the greater community whose buy-in is critical to achieve the vision and goals of the learning community; and, perhaps most importantly, justifying the cost of the effort by demonstrating the impact of the project on gains in student learning and achievement.

Some of these challenges relate to school district organization and operations, while others are technical in nature. The critical district-level

issues include the reality that effective district-wide implementation— every school, every classroom with equal access to resources—is rare. Traditional models of use and deployment that view technology as an intellectually and physically separate activity also hamper technology’s potential. Costs are also key: technical support for traditional computer installations is cost-prohibitive for many school districts, as is Internet connectivity. At the same time that costs are rising, most school districts’ dollars for connectivity, equipment, technical support, and professional development are shrinking. As a result, individual schools often are left to implement technology on their own rather than as an integrated district-wide effort.

Next, LaGace introduced the equally difficult technical issues. First, there is no consensus or even a shared vision for what effective use of education technology looks like. Businesses producing technology have not understood the culture of schools well enough to adequately address their needs in the products and services they offer to the education community. Instead, education is expected to tweak equipment designed for other markets and users to make it work for schools. Lack of hardware and software standardization raises costs and creates challenges for effective professional development. Most IT departments in school districts lack expertise for planning, building, and maintaining a robust, cost-effective network and are not client oriented.

Basing his comments on experiences from LemonLINK’s five-year implementation history LaGace turned to the key requirements for reaching the point at which all teachers in a district fully integrate technology into curriculum and instruction for daily use. These include

• equipment that is simple to operate (instant ON, like an appliance);

• fast, dependable connectivity;

• operation that is both reliable and predictable (e.g., technical support is readily available);

• tools that allow teachers to locate quality electronic resources that are aligned with standards; and

• electronic delivery of lessons, instructional materials, and resources that is easy to organize.

Most important, he emphasized, is “access, access, access.” However, if access is defined as a minimum 1:2 computer-to-student ratio, then he acknowledged that access is likely to be cost-prohibitive for the approach to computer use taken by most school districts. This is especially true when the total cost of ownership of hardware, peripherals, and software is taken into account. Associated with such ownership are costs for deploying, operating, and maintaining a computer network over a period of time, including connectivity, network hardware, workstations, technical

support, staff development, repairs, replacement, upgrades, software purchases, and ongoing licensing fees. For LemonLINK, fiscal realities made it essential to drive down both the costs and the complexity of the technology.

LemonLINK approached the cost issues proactively through a broad array of partnerships in order to develop a cost-effective model for computer use in schools that was not already being provided by the market. The design team took the initiative to identify the kinds of capabilities they needed and to develop their technologically based learning community as a business proposition in ways that would appeal to potential partners. Obvious partners included hardware, software, and networking providers, such as Microsoft, Hewlett Packard, and Citrix. The plan also involved working with Cox Communications, the town’s local cable provider, for home connectivity. Finally, higher education institutions, including the University of California, San Diego, helped to round out their partnership strategy. Until the market provides ready access to a cost-effective model for computer use, other school districts may find it worthwhile to pursue such partnerships as well.

In another innovative approach to offsetting costs, LemonLINK has contracted to provide network services to various public-sector organizations in the community. In a prescient move 10 years ago, the district erected a communications tower, which now sends video, voice, and data across its private network, not only to the schools but also to the City of Lemon Grove, local fire departments, the community center, the recreation department, the teen center, the senior center, and the nearby charter high school, which is attended by 60 percent of Lemon Grove’s graduates. Thus the district’s technology budget, only about 1.9 percent of its general fund,² is supplemented by revenues generated from providing these services. And the community benefits from a growing, integrated, and seamless network.

Centralized network design is a key factor in LemonLINK’s cost structure and effectiveness. The district’s technology center allows multiple district organizations to share resources as well as to process and store data that can be accessed across the network. Because everyone’s programs and data reside at the center, students and staff can be anywhere— such as at a school, in different classrooms, the local community center, or

at home—and still access and manipulate their information. The centralized design concept also enables the data center to serve multiple independent organizations without the need to implement or support locally installed servers or network resources. All that is needed at a school or facility is a local area network (LAN) that can connect workstations back to the data center. The center’s high availability is achieved through cluster technologies and mirrored locations. A 30 terabyte storage area network (SAN) allows teachers practically unlimited space to develop and maintain on-line curricula. Each student has a one gigabyte of space to store daily work as well as maintain an ongoing portfolio of final works.

The LemonLINK plan to supply abundant access to technology necessarily moved away from just putting more and more computers in the classroom. Instead, the district began installing smaller, cheaper network appliances known as “thin clients.” By purchasing thin clients at $389 each instead of $1,500 for a multimedia station, the district dramatically increased access. This innovative approach has allowed LemonLINK to install three times more equipment in schools with the same budget allocation, and it was the key factor in attaining both the 1:2 ratio and the instant-on capability that allow teachers to focus on instruction rather than technology. While the interface is the same as that for a PC, a keyboard, a mouse, and a monitor, the thin client workstation doesn’t have all the complex and expensive components of a typical PC. Most thin clients don’t have any moving parts and have instant-on capabilities. Their programs and data come from “slices” of memory and processor power from a terminal services farm located at the data center. Bandwidth-intensive applications, such as streaming video, can also be viewed directly from the thin clients through LAN connections using a locally based web browser and media player. To provide for full computing capabilities, each classroom is also equipped with several multimedia workstations.

LemonLINK made an early strategic decision to get away from having schools or teachers purchase specific applications, many of which were often poorly aligned with district learning standards or were difficult to operate. Instead, LemonLINK adopted a district-wide approach whereby everyone has access to the same library of software. Currently, 15 applications are supported throughout the district. This approach has proven both effective and efficient in terms of training and support. Moreover, teachers easily share how they are using technology across schools in the district and across classrooms in schools. Technical support at the classroom level is provided on site in each school one day per week. Technical support staff participate in professional development meetings so they understand the realities of employing information technology in classroom settings. However, many technical support problems can be solved remotely. Phones in each classroom allow teachers to call the

support center when a technical problem arises and receive remote help immediately, without having to submit work orders.

Staff development was designed to parallel the phased installation of the technology over the past five years. During each year of the installation period, approximately 20 percent of the teachers were provided with 100 hours of initial training, including short workshops on applications, teacher-to-teacher collaborations, observation, and hands-on use of technology in classrooms. Additional professional development is provided on an ongoing basis at the building and district levels.

What have been the results in terms of student learning? About half-way through the implementation phase, LemonLINK’s outside evaluator took a very close look at student performance scores from the Stanford-9 and API state assessment data (Snyder, 2000). Roughly half the schools and students had access to technology at the desired 1:2 ratio, and their teachers were trained to use the equipment and software; the other half had not yet reached these goals. The results, reported Allen, were “astounding”: students who had access to the technology did better across the board than students who did not. In some cases, she reported that the differences were striking. Matched scores on every student also allowed them to observe solid year-to-year improvements for individual students, including various subgroups, again with significant advantage to those in technology-rich classrooms. When teachers saw the results, they became far more interested in learning to use those tools in their own classrooms.

Student engagement has also benefited. In response to a question about whether technology has changed structural practices in the schools, LaGace reported on an extensive series of observations that were conducted in a recent tour of over 60 district classrooms. The object of those observations was not to observe the lessons being conducted, but rather to observe the students and what they were doing. At that time about 70 percent of classrooms were participating fully in the technology, while in about 30 percent of cases, teachers were not quite convinced of the efficacy of integrating technology into their classrooms. LaGace reported that the differences observed were striking: in the technology-rich classrooms students were engaged with their work and progressing at their own rates, collaboration was taking place, and teachers were providing instruction tailored to individual students. In the classrooms in which technology was not being used, the traditional model of teaching and learning was striking by contrast: teachers at the front of the room writing on the white board, students nodding off or otherwise distracted.

Allen also added that the configuration of the classroom has had to change: computers are now on classroom desks, not at the back of the room. The technology is part of the learning process every day, always available for searching information on a topic under discussion in class.

As technology has become a working tool in the classroom, the teacher is no longer up front but working with students as a facilitator, helping them to gain knowledge in many different ways.

Lessons have changed as well. Instead of starting with lessons that are linked to subject matter in textbooks, many teachers begin with large ideas. They then use those larger concepts to bring students to desired levels of learning through designated activities that use technology and related resources to get there. Textbooks become supplements in many cases. ³

Finally and most importantly, according to Allen and LaGace, access to the technology and learning does not end with the school day. Currently 15 percent of district families have on-line access through LemonLINK’s thin clients. With newly developed web-based access (http://mylearningportal.com), LemonLINK expects the percentage of students and families with LAN and Internet access to double as home computers connect to the network. The district also provides every student who scores below the 40th percentile on state standardized tests with a thin client for home at no cost for 12 months as an academic intervention.

The LemonLINK team shared several additional insights in response to questions:

Recruitment of new teachers: Over the five-year implementation period, teachers both retired and left the district. As new teachers have been recruited, LemonLINK is seeing a different kind of candidate. An increasingly important factor of their recruitment efforts is that teacher candidates research the school district on the web and report that they are choosing to apply to Lemon Grove because of its use of technology. One indicator is that these new teachers are themselves more technologically savvy, being able to pick up on the technology and move to a functioning level quickly—in 6 months compared with the 18 months for teachers who had participated in the 100-hour professional development program. Their interview process also emphasizes candidate compatibility with the technology-rich environment.

Experimental process in developing LemonLINK: Another question concerned LemonLINK’s response to experiments in the development process that did not work. Allen was clear:if something is not working, they stop doing it. LemonLINK admits to itself and to its partners when something does not work. She emphasized that doing so is not synonymous with failure. Rather, continuing to spend lots of money without any results constitutes failure. If those who are directly involved with a project

can admit to themselves that that something did not really work, they can try to understand why and learning becomes possible.

Disseminating information about LemonLINK to others: LemonLINK has received numerous visitors from other school districts. The utility of those visits depends on the kinds of people the other districts send. A team that represents different perspectives and is able to look at the challenges from different angles has a much better chance of reliably communicating what they saw and translating what they learned into action in their home districts. LemonLINK tries to focus visitors on the things that have really made a difference—what their presentation at this workshop addressed. Visitors will see the data center and the technical aspects that make the network work. But they will also see that the technical office and the curriculum/staff development/instructional methodologies offices are collocated, allowing for daily conversations. The staffs function as a team, unlike the situation for technology departments in most districts that are located apart from curriculum experts and may never have an opportunity to talk with them. Visitors also observe classrooms. By the time the visitors are finished, they should have the whole picture of what LemonLINK is and the results for teaching and learning. Once visitors return home, it is not uncommon for them to call two months later with more specific questions, especially those of a “How did you do that?” nature. LemonLINK does not have a manual that visitors can take back to their individual school boards. According to LaGace, the hardest thing for visitors to grasp is that LemonLINK is a long-term investment in change. Too often, they want to accomplish this transformation in 12 months.

Curriculum and content in the LemonLINK system: A final question turned to curriculum content. LaGace reported that an increasing amount of content is web-based and the district’s own teachers are creating more and more content. In partnership with some companies, LemonLINK is working on easier ways to locate instructional materials for teachers that are aligned with standards and linked to appropriate grade levels.

In the view of Steve Rappaport of Advanced Networks and Services, the transformation involved with integrating cheap, fast, robust computers into instruction for every student in America and ensuring that technology is integrated in ways that dramatically improve K-12 teaching and learning presents not one but two challenges. The first challenge is making technology widely available in schools and ensuring that the conditions for its effective use exist, especially technical support and professional development for teachers. The second challenge is leveraging those

technological resources effectively in K-12 classrooms so that they achieve the ultimate goal of improving teaching and learning. While the two are related, each has its own issues and outcomes.

The first challenge is making technology widely available and usable by students and teachers, and the principal issues concern the nature of, access to, support for, and cost of technology. Rappaport applauded LemonLINK’s successful formula for meeting his first challenge: making technology sufficiently affordable to be pervasive, reliable, well supported technically, and easy enough to use to be incorporated routinely into educational practice; and educating teachers so that they feel comfortable with technology and, more importantly, understand how to use technology effectively in their classrooms.

He readily agreed that technology has the potential for changing the way we teach and learn. He cited examples including multimedia authoring tools that have been demonstrated to increase students’ means of expression, virtual tours of remote sites, simulations, and on-line collaborations. The ultimate goal, however, is to improve teaching and learning; merely placing technology in schools has a limited impact on student learning. All too often, technology is grafted onto existing teaching practices, so the result is educational practice that is technologically sophisticated but still fundamentally conventional. Rappaport pointed out that using PowerPoint instead of a blackboard or overhead projector for a presentation, for example, does not represent a fundamental shift in educational practice.

Too many policy makers view the potential of technology to improve education through a lens that focuses on efficiency, believing that schools can achieve returns on the investment in technology in education that are similar to what many businesses have realized. He cautioned against confusing efficiency with effectiveness.

Technology can make the education system more efficient in some respects, such as through improving the ability to assess student performance, marshal data in decision making, or communicate with stakeholders. For example, technology can offer significantly improved means of assessment, such as diagnostic instruments on handheld devices that allow ongoing formative assessment in classes in ways and at levels that simply cannot be achieved without technology. In addition, technology makes possible the aggregation and analysis of assessment data and hence evaluations of student performance at the school, district, state, and national levels, as well as the ability to disseminate information to parents and other stakeholders.

As was also discussed by Barbara Allen and Darryl LaGace, Rappaport stated that some evidence also exists that technology can improve student achievement. Some studies, for example, have shown in-

creases in student performance on standardized tests (e. g. , Honey et al., 1999; Mann et al., 1998). Other studies suggest that certain types of educational software can facilitate the acquisition of early literacy skills, such as reading comprehension and vocabulary development, and that other types of software can increase students’ understanding of mathematical and scientific concepts.

But it is the effectiveness of schools—that is, the ability of students to learn in them—that must remain the principal concern; and it is unwarranted to assume that merely introducing technology into educational settings will produce the desired outcome of improvements in learning. Rappaport argued that the education community has failed to demonstrate clearly that technology can improve student learning. Furthermore, he contended, a compelling case has not yet been built because educators and policy makers are asking the wrong question. While the tendency is to focus on technology and ask whether its use is improving student achievement, it is educational practices and processes that determine how well students learn. He emphasized that technology is not a process but a tool through which educational practices are mediated.

He then cited ThinkQuest,⁴ a program that his organization once operated, to illustrate this point. ThinkQuest, in which over 100,000 students from 125 countries have participated, is a large-scale example of project-based learning: several students form a team, intensively study a subject for several months, and then create a web site to reflect the knowledge they’ve acquired. Technology makes certain things possible in ThinkQuest that could not be done otherwise, and it is a powerful motivator for students to engage in their own education. But the educational practice that is at the heart of ThinkQuest is project-based learning: students researching subjects and working on projects reflecting the knowledge they’ve acquired, an educational practice that predates the introduction of technology. In ThinkQuest, students use technology to complete their project, but the end-product could have been a written paper, a play, or a diorama. ThinkQuest’s emphasis on project-based learning is the key, not technology.

If technology is to make a contribution to improving student learning, it must be aligned with educational practices that are most likely to achieve desired learning goals. Unless educational goals are articulated first, policy makers and educators will never understand how to align technology with educational practice to realize the goal of improving student learning. For Rappaport, the key question at this juncture is whether, as a country, the United States wants to preserve educational

practices in essentially their current forms and develop ways to employ technology to increase student achievement on standardized tests, or whether the nation instead wants to take this opportunity, made possible in part by technology, to transform education in ways that will achieve dramatic improvements in student learning.

To address these questions, Rappaport argued that national organizations and the federal government must provide new leadership. A wide range of national organizations concerned with the state of education in the United States has a significant role to play in educating their constituencies and shaping the debate about the role of technology in K-12 education. Only with broad discussion among all stakeholders can a national consensus be established about the proper roles for technology in K-12 education that will allow progress on the required scale. The Consortium for School Networking,⁵ for example, recently formed an emerging technologies committee to educate K-12 school leaders about advances in technologies and innovative applications of them that may enhance teaching and learning as well as school administration and decision making. The committee will also address problems of implementing and owning emerging technologies in schools, including technical issues and the total cost of ownership.

Rappaport emphasized that technology will not in itself change educational practice. It is on educational practice that efforts to improve student learning must focus. That is where the principal challenge lies.

At the heart of Rice University’s Connexions Project is an electronic curriculum repository for concept-driven curriculum modules.⁶ The initial content has covered courses in the sciences, engineering, and mathematics. Newer content includes music and social sciences modules, with additional humanities materials currently in preparation. Most of the existing material has been developed for college-level courses, although some of the content in music has been developed for K-12, and the system would be broadly applicable to K-12 content in other subject areas. As Geneva Henry, executive director for Connexions, explained, the concentration on concepts was a creative response to one professor’s frustration

that his best-performing students had mastered the content of his course in a linear fashion that followed the textbook but missed the big ideas and unifying concepts that were his focus.

To remedy this problem, the Connexions Project was designed to stimulate authors of instructional materials to develop curriculum modules that represent individual concepts, with links included to show how the concepts interrelate. The modules are then placed in the electronic repository so instructors can explore the concepts and use them directly or modify them for use in their own teaching. Students can explore related concepts, yet stay anchored in the course they are studying. By means of open source licensing, which allows users to modify modules created by others, the project is also intended to provide opportunities for collaboration among faculty, teachers, students, and authors of material across disciplines.⁷ The assumption is that many concepts may have relevance beyond individual disciplines and courses and that making them available to others to use will be beneficial.

For teachers at both the high school and postsecondary levels, the repository serves as a significant instructional resource: they can build their own courses entirely around the concepts they create or modify, or they can integrate certain concept modules into existing courses. For teachers in K-12 classrooms, easy access to rich curriculum material and the ability to adapt it to fit their needs are especially important. Aligning what the student is learning with the conceptual knowledge the teacher is trying to get across is the objective. The Connexions repository is freely open to students as well as teachers. Students can engage in general exploration of various concepts or, with the help of a set of tools developed for the purpose, can work through the course that a teacher has assembled out of multiple modules. In both cases, when students find concepts they are interested in pursuing, they will find links to related concepts that allow for easy in-depth exploration, including applications of the concepts in the world beyond the classroom. An electronic roadmap has also been developed that lets them explore without losing connection with their launch points. The objective is to engage student interest in topics that may appear as unrelated or irrelevant or isolated pieces of information when they are presented one course at a time.

The remarks of Cheryl Lemke, president of the Metiri Group, focused on five key points. First, she agreed with Steve Rappaport that the education community must define more clearly what it means by “improving student learning.” Technology is not the issue, in her view; intellectual capital and 21st century skills are. The reality is that technology has changed society so dramatically that people need to develop new skills, skills on which Metiri has been working with the North Central Regional Educational Laboratory (NCREL) in Chicago to define for the last two years. The 21st century skills are shown in Box 3-1. Lemke noted that some of them have to do with technology and some do not. However, they all reflect how technology has changed society.

Lemke then discussed the misalignment between traditional measures of student learning and 21st century skills. The charts used to demonstrate increases in academic achievement on standardized tests address the traditional conception of individual achievement. They stand in marked contrast to the commentary reported from teachers and parents:the mother who said, “My kid’s a self-directed learner”; the teacher who said, “We’re creating our own knowledge” and talked about project-based learning and moving her students away from learning exclusively from the textbook. Was it, Lemke asked, really the computer-aided

instruction, or was it the project-based learning the teachers employed? Relying on only those skills assessed in standardized tests will leave U. S. students behind in this global world. These 21st century skills are important in and of themselves, and they are also an important bridge to higher academic achievement.

Lemke’s second point was that the testing environment is a huge barrier across the country. One of the things teachers report is that the standardized tests in most states actually push teachers away from using technology. The enormous emphasis on student scores on these tests puts a premium on more traditional approaches to curriculum coverage, textbook-based instruction, and test-taking strategies. She contended that the community involved with these transformations has failed to build a compelling case for how technology and 21st century skills can increase student achievement. One aspect of building that case is developing the capacity to assess these 21st century skills, an issue raised by a workshop participant.

In response, Lemke briefly described a project on which the Educational Testing Service (ETS) has been working over two years that may eventually be incorporated into the National Assessment of Educational Progress (NAEP). ETS has developed a web-based, on-line assessment for eighth graders to test their scientific inquiry skills as well as their technological literacy. For the assessment, the students work through a science tutorial on hot air balloons. In the course of the tutorial they can draw objects, observe the balloon going up and down, and graph its motion. Then they are asked three questions that require explanation. As they respond on the computer, every keystroke is tracked and the results are compared with results generated by experts. Student scores are based on their prowess in scientific inquiry and technological literacy. It is of special interest that when this assessment was tested about a year ago in New Jersey, the participating students uniformly said that it was the first time they had ever been tested and learned something at the same time. She suggested that pursuing this approach to assessment could be especially important when many are worried about the instructional time that is taken away by testing.

Roy Pea, workshop cochair, cited another research effort, funded by the National Science Foundation, that confronts some of the same psychometric challenges. Often a big issue is how to develop measures of those things that are defensible in the court of public opinion and scrutiny. The Principled Assessment for Design of Inquiry is looking in detail at component skills and scientific inquiry, trying to develop the kinds of measures that will be found adequate to that task.⁸

In her third point, Lemke agreed with the earlier speakers on the essential conditions that have to be in place for technology to work. She emphasized leadership, an innovative culture, and the support and nurturing of those innovators. She pronounced as “unconscionable” the expectation that teachers adopt technology one at a time rather than as part of a systemic change. In an environment of high-stakes assessment, teachers are held accountable for how their students are progressing. Lemke argued that only a systems approach to making these kinds of necessary changes would yield results, as LemonLINK has shown.

The closely related fourth point concerns whole systems thinking and systems change. She reemphasized Barbara Allen’s conclusion that it is not sufficient for students to have the opportunity to be taught to use technology to enhance their learning or to acquire 21st century skills because they happen to be in Ms. Jones’ classes and not Mr. Smith’s. Fairness demands that these changes be implemented systemically.

Lemke’s final point was the importance of the one-to-one ratio of students to computers. There are some limited cases in which this is being achieved: in Maine, every seventh and eighth grader has been given a laptop and has network access at home; at a Quaker high school in Pennsylvania, the ratio of students to machines is approaching 1:2. The tipping point will occur when teachers—confronted by a classroom of students with laptops or other technology tools on their desks who are able to access a huge knowledge base—conclude they must do things differently. As the developers of LemonLINK learned, pervasive technology changes the learning environment dramatically and pushes everyone associated with the project to really do some of the new things that ought to be done. Lemke brought the message home by challenging participants to imagine having to share their computers with three other professionals. “Do you think we’d ever use it for anything that is mission critical? I don’t think so. And the kids are the same way. ”

The second commentator was Wanda Bussey, a mathematics teacher and department chairperson at Rufus King High School in Milwaukee. She recalled that Milwaukee actually started out very well in the technology world. During the 1980s visionary curriculum specialists “dragged us kicking” into the early stages of the computer revolution. A “stutter step” followed, which corresponded closely to the problems described in Darryl LaGace’s presentation on technical issues and Steve Rappaport’s focus on pedagogical issues. Her conclusion: technology must be simple, reliable, supported, connected, with good resources for lessons and great infrastructure.

The need to address the total costs of ownership is especially important for schools. Early in Milwaukee’s efforts, a good planning process was launched, with a hard-working technology committee that made vis-

its and researched the technology options. They still made quite a lot of mistakes, such as purchasing two sets of laptops that are languishing in corners of classrooms because not enough teachers have figured out how to use them appropriately. These laptops are on a wireless network and could be operating immediately if the teachers could figure out what to do with that capacity. The issues of professional development are critical and Bussey commented that LemonLINK’s phased 20 percent per year rule is an admirable way to approach this need.

She also expressed strong agreement with Steve Rappaport that technology needs to support what is important to the education community. She cited the development of Texas Instruments 80 graphing calculators (now replaced with the TI-73) as an example. Approximately 15 years ago, the calculus community set out to change the focus of calculus education. The leaders of this movement were wise to involve not only people at collegiate levels but also high school teachers and advanced placement (AP) teachers. At about the same time, Texas Instruments was developing a user-friendly technology that did not require a network and could be held in the user’s hand. The two developments coalesced and, over the past decade, calculus, teachers, and the technology have all changed in ways that support and reinforce the others. Today, for instance, the AP calculus test is much more concept-driven. Changes in the technology of the calculator have also greatly enhanced teaching and learning. In the early days, to find an intersection of two curves, users had to “zoom in” in a series of steps. It often required a half hour of classroom time to teach students how to do so. Now the zoom-in function is a command that the technology executes quickly, freeing time for learning and discussing mathematical concepts rather than technical manipulations.

Bussey argued that the critical message of the calculus example is that the teaching methods and the content of instruction have both changed in response to the graphing calculator. Students are now able to do much more work. In calculus courses in the past, there were literally only two kinds of problems that teachers could use to teach such concepts as the length of an arc because they lacked the ability to take antiderivatives. She reminded participants about a specific calculus problem using cubic curves that appeared in most treatments of the topic several decades ago. This specific problem was used because its solution involves a perfect square, resulting in problem that can be solved easily with paper and pencil. As a result, when teachers wanted to teach students this kind of problem solving in calculus, they ended up spending far more time explaining the idiosyncrasies of this particular problem than they did in discussing what was going on in the limiting process of the arc. Graphing calculators have made that dilemma disappear. As a result, students’ use of technology is allowing them and their teachers to focus more on con-

cepts and engage in a free-flowing exchange of information and ideas. Students and their learning are the great beneficiaries.

After the presentations about the first transformation, the participants broke into four groups with an assignment to develop lists of key enablers that could help bring about the first transformation. After elaborating their candidates for key enablers within each of the breakout groups, participants then moved around the conference room, reviewing the lists of all four groups and voting for their top two choices for key enablers. The complete list of key enablers transcribed from the poster board sheets of the breakout groups is included in Appendix B. This section briefly describes the top choices.

Several versions of this key enabler received support from a number of the participants. One version focused on the importance of assembling evidence from the research literature to show that technology enhances student achievement. Another version focused on the importance of making the case to teachers that technology can add value to their own work practices, not only by directly improving the performance of their students but also by helping them prepare lessons, interact with colleagues, and manage routine student work flow. A third version of this key enabler focused on combining these arguments about research evidence and the value of using technology for teachers to build a case for policy makers and industry officials about the types of technology use that can improve K-12 education. A fundamental aspect of this key enabler is the importance of finding ways to effectively communicate arguments about the value of technology that address the needs and expectations of different stakeholders.

Several of the leading candidates for key enablers focused on the importance of addressing the full range of changes required to integrate technology into K-12 education. One of the breakout group facilitators noted the importance of the LemonLINK model in the participants’ thinking about the importance of a systems approach to change. Different versions of this enabler focused on different linkages that a systems approach should bring about: one stressed the linking of curriculum, pedagogy, and technical support while another mentioned the linking of people, community, and technology.

Several of the key enabler candidates focused on the importance of incorporating technology into the education that teachers receive, both in their university education and in professional development activities throughout their careers. One version of this key enabler emphasized the importance of technology’s being embedded in instruction for all university courses that future teachers take, including both education courses and content courses in the arts and sciences.

The next session of the workshop focused on the second transformation: the challenge of combining advances in the science of learning with IT capabilities to dramatically improve student learning. The session included five presentations. Louis Gomez of Northwestern University spoke about some of the capabilities of IT that would allow improved student learning and the necessary research and institutional arrangements to take advantage of those capabilities. Roy Pea of Stanford University discussed the convergence of the research, industry, and teaching communities in finding ways to work together toward using IT to improve learning and the possibility of research partnerships that would involve all three communities. James Pellegrino of the University of Illinois at Chicago spoke about the potential for evidence from the learning sciences to aid in the design of powerful learning environments that take advantage of technology. Edward Lazowska of the University of Washington discussed why past predictions that technology would revolutionize education have not been realized and why the next generation of educational software has the potential to succeed although previous technologies have failed. Robert Tinker of the Concord Consortium discussed the importance of applied research and innovation in education technology and proposed a funding outline for a balanced research agenda. After these presentations, there were additional comments by Nora Sabelli of SRI International and David Vogt of the New Media Innovation Center.

Louis Gomez of Northwestern University was the first speaker to discuss what is required to bring about the second transformation: the challenge of combining advances in the science of learning with IT capabilities to dramatically improve student learning. First, he presented a

vision for the kinds of IT capabilities that would allow improved student learning. Second, he outlined some of the research that is required to make the second transformation a reality. And third, he discussed the type of institutional arrangements that would be necessary to bring about these changes in both research and practice.

With regard to IT capabilities that could improve learning, Gomez first talked about the need to develop a “supportive integrated information infrastructure.” This sort of information infrastructure would relieve teachers of the burden of some routine tasks and therefore allow them to focus more of their time on the activities that form the core of their work. One example of the capabilities of such an information infrastructure would be an IT system that provides students with formative feedback on written materials. This feedback would make it possible for students to have more practice developing their writing skills while reducing the time teachers spend in providing feedback about simpler matters, such as grammatical errors.

Another example of the capabilities of an information infrastructure would be an IT system that provides teachers with information about the social support services that students receive in their communities and at home. These support services often make a large impact on what children are able to do in school, but information about those services can currently be quite time-consuming for teachers to locate. A third example of the capabilities of an information infrastructure is an IT system that encourages collegiality among teachers by making it easy for them to share materials and approaches with their colleagues.

Gomez also spoke about the benefits of “developmentally rational tools.” These are tools that can be used for a wide range of applications, such as visualization tools and spreadsheets. Although such tools are complex and can be ambitious to learn, their breadth of possibilities for increasing the productivity of work makes the investment in learning them worthwhile. In many cases, similar tools are used in other professions as well, and this bridge between school and professional tasks reinforces the importance of both teachers and students learning to use them.

Gomez then turned his attention to the types of research required to bring about the second transformation. This research would explore how new IT tools provide new opportunities for learning. He included several related types of research in this category: (1) understanding how increasing teachers’ opportunities to learn affects the learning of their students, (2) understanding how new tools can create the capacity in an entire school community for students to undertake more ambitious work, and (3) understanding how the new tools themselves can help people understand and can be designed to take advantage of the principles and activities that have been found to work in schools.

A different type of research is related to making sense of numerous anecdotal observations and reports in which researchers and teachers discover that students who are typically not very engaged in learning unexpectedly become deeply engaged when working with an IT-supported learning system. These anecdotes offer the possibility of using better IT-supported learning to close the achievement gap for students in the United States.

Finally, Gomez discussed the issue of the institutional arrangements that are required to bring about better tool development and a new kind of research. Fundamentally, he argued for a switch to demand-side research, with a close coupling of research and practice that allows research and development to be focused on and inspired by the problems of practice. As an example of such a system, he discussed the Chicago Urban Systemic Partnership (CUSP), a city-wide system of coordinated in-service teacher professional development. It sponsors courses that are provided at universities across Chicago with a common approach. Each of these courses provides a joint focus on subject matter, student learning, and pedagogical strategies and includes preparation for integrating technology into the classroom.

Roy Pea, professor of education at Stanford University and cochair of the Committee on Improving Learning with Information Technology, spoke about the importance of “leveraging convergences” in order to advance learning and teaching. The concept of convergence is often used to refer to processes that are merging the different information and communication technologies, including computing, telecommunications, publishing, broadcast media, consumer electronics, photography, video, and music. However, in the context of the work of the committee, Pea emphasized the importance of a second type of convergence: the technical integration that must be pursued by industry, the research and development being advanced by the learning sciences, and the wisdom of practice from K-12 educators. By bringing together these three communities, the committee hopes to be able to stimulate innovations in learning technology that are guided by learning science research with more rapid consequence for education and learning.

As additional context for the committee’s work, Pea noted the conclusions about the power of technology to support learning that came out of the National Research Council report, How People Learn (1999b). The learning that information technology can support includes the provision of (1) real-world problems for learning; (2) connections to experts and communities of learners; (3) visualization and analysis capabilities; (4) scaffolds

FIGURE 3-1 Changes related to IT hardware and services.

for problem solving; (5) opportunities for feedback, reflection, and revision; and (6) opportunities for teacher learning.

He then presented a diagram summarizing changes related to IT hardware and services that he originally presented at the committee’s first workshop in January 2001 (see Figure 3-1).⁹ These changes are driven by steady increases in IT processing power, memory, and connectivity, resulting in a multidimensional explosion in media richness and personalized software services. In the two years since presenting that diagram, there has been both continuity and evolution in the trends that it summarized. The continuity is seen in the continued exponential growth in hardware capabilities, including processor speed by a factor of 400 since 1990, memory by a factor of 120, wireless speed by a factor of 18, and fiber channel bandwidth by a factor of 10,000 (de Ruyter, 2002; National Science Foundation, 2003). One of the most interesting evolutions in these trends is the increasingly important role of the consumer market (as opposed to the business market) as a driver of commercial developments in

electronics and communications. This is seen in the explosive growth of cell phones and wireless networks. The emerging convergence of computers and telephony will soon allow tiny phone-like devices that are always connected and that are more powerful than today’s PCs. These devices offer great potential for use in education, where engaging games and peer-to-peer activities can be designed to incorporate subject matter learning in compelling and adventuresome ways appropriate to the form-factor and use contexts of such devices (e.g., Hoppe et al., 2002; Roschelle and Pea, 2002; also generally see The Journal of Personal and Ubiquitous Computing). As a result, we see many of the big companies in this area, such as Sony, Nokia, and Microsoft, starting to look seriously at learning tools as a potentially very large and profitable marketplace. This is an opportunity that the education community needs to explore further.

Pea identified a number of capabilities that will be possible with future e-learning technologies based on these powerful and evolving trends. The declining cost of the technology will allow a one-to-one computer-student ratio with learning environments that can adapt to the learning needs and styles of individual students from tacit or explicit assessments of their needs. These personalized capabilities also will allow on-demand professional development support for teachers. In addition to allowing personalized learning environments, e-learning technologies will provide learning experiences that are rich in communication, media, and the use of complex simulation models. The technology also will allow learning that is engagement-intensive, using techniques to motivate learning by leveraging gaming strategies and the power of social networks (e.g., Barabasi, 2002). Finally, the technology will allow improved learning and teaching work flow management, including group collaborative learning tasks and embedded assessment with real-time teacher support tools and portable digital learning portfolios. These will make it possible for teachers and technologically enhanced learning environments to personalize the learning experience for students based on a rich understanding of what each student knows and is able to do.

The development of these e-learning technologies will require a broad base of research. Pea stressed that while systematic clinical trials are a worthwhile activity, they represent an end point of a complex and multifaceted research pipeline. This research pipeline includes basic research in the learning sciences, work on the construction of tools and platforms, development of proof-of-concept demonstrations for technical innovations, design research and the development of IT-based curricula and educational applications, and implementation research on the factors that influence the use and success of newly introduced techniques.

He also stressed the importance of developing new approaches to research that will bring together as partners the communities of industry

and K-12 practitioners with the work of researchers in the learning sciences. The concept of finding ways to build partnerships that bridge these communities has been fundamental to the work of the committee. Such joint work will make it possible to develop more coherent approaches to the development of education technology, and it will allow the end users of that technology to make more strategic decisions about what they purchase, how they use it, and what type of professional development support they provide in conjunction with it.

One critical outcome of deeper partnerships will be in encouraging a stronger focus on “use-relevant” research, which Donald Stokes originally propounded in his book Pasteur’s Quadrant (1999). The importance of such research was also emphasized in many of the presentations that were delivered at a workshop organized by Geneva Haertel and Barbara Means (2000) to consider the methods and approaches needed to improve research and development in the area of education technology. Such use-relevant research would involve multiple related studies conducted in networks of test-bed schools. These school sites would be committed to participating in sustained studies of the effects of technology, and they would simultaneously provide evidence of emerging trends in the use of education technology. Such large-scale research using networks of schools would require intermediary organizations to coordinate the research effort by identifying research questions, designing common data collection protocols, and supporting local researchers.

One version of such research efforts using test-bed schools has been described using the term LENS (Learning Expeditions in Networked Systems) partnerships. The term harkens back to expeditions carried out to explore the frontier during the early history of the United States, such as that by Lewis and Clark 200 years ago. The goal of LENS partnerships is to undertake expeditions to scout the future of learning. LENS partnerships would explore systemic approaches to change in education, aligning standards, curriculum, pedagogy, assessment, teacher development, school culture, and school-home connections, in addition to the use of education technology. The partnerships would undertake a continuous innovation cycle for education techniques and technology, in which the design of new prototypes would be followed by observation of the use of those prototypes, which would immediately feed back into modifications in the prototype designs.

Pea finished his remarks by returning to the concern with learning that is at the heart of any effort to improve education. At the center of the mathematics, science, and technology standards that have been developed in recent years is a concern with what some have called 21st century skills. These involve two clusters of skills: one involving seeking, organizing, and evaluating information and the other involving communicating and collaborating with others. One of the most important capabilities that new

education and learning technologies make possible is the ability to provide more meaningful assessment measures for these 21st century skills, which will, in turn, allow their role in the curriculum to be increased.

James Pellegrino of the University of Illinois at Chicago began his talk by discussing the importance of building the links between research and practice. From analyses in previous studies at the National Research Council, such as How People Learn: Bridging Research and Practice (National Research Council, 1999b), there is only a weak relation between research on teaching and learning and actual classroom practice. Usually research affects practice only indirectly through its intermediate effect on educational materials, pre-service and in-service education, policy, and the media. Instead of the current weak and indirect relationship, Pellegrino emphasized that work is needed to create a cumulative knowledge base about classroom teaching and learning that both affects classroom practice directly and is drawn from issues, problems, and studies of classroom practice.

He then turned to discussing research on the development of expertise and competence in particular subject matter domains, such as mathematics, science, and reading. He argued that some of the most important implications for curriculum, assessment, and instruction come out of learning and cognitive research in specific curriculum domains. The fundamental lesson is that expertise is not generic but is essentially related to specific domains. One way to bring about a more coherent alignment of curriculum, assessment, and instruction is to make sure that all three are connected to understandings of how people learn in specific domains rather than to generic models of teaching and learning. However, it is important to note that the current knowledge base in learning science has developed domain-based models for only parts of the K-12 curriculum, with large portions of the curriculum having very little coverage. This underlines the importance of pursuing a vigorous research agenda to develop a full range of domain-based understandings of learning across the K-12 curriculum.

In addition to research on the development of expertise, How People Learn also describes what research in the learning sciences has revealed about the features of powerful learning environments. Pellegrino summarized this in four points. First, a powerful learning environment should be knowledge centered, with a focus on the important things we want people to learn in a given area. Second, it should be learner centered, so that it deals with where the learner is and can thereby respond to the needs of the learner. Third, it should be assessment centered, taking serious account of the processes of assessment to identify what students know and don’t know so that instruction can be

designed appropriately. And fourth, a powerful learning environment should be sensitive to the social aspects of learning, which evidence suggests are extremely important to people’s acquisition of expertise.

He then described different capabilities of education technology that allow it to map onto the various aspects of powerful learning environments. Technology can enable the production of new curricular materials and instructional resources that are more focused on the key knowledge constructs that educators want students to learn. Technology also enables educators to integrate assessment into instruction, manage complex learning environments, and design modular learning and instructional resources that can be used in a variety of contexts.

He next provided more detail about opportunities to use technology to design more powerful and useful assessments that capitalize on new research about cognition and measurement. One opportunity is to use technology to provide problem-solving scenarios that tap into more complex forms of knowledge and reasoning. Another opportunity is to use technology to help interpret observations about complex student performance, such as work with E-rater^®10 and latent semantic analysis, which allow computers to score student writing (e.g., Kintsch et al. , 2000). A third opportunity is to use technology to connect assessment with instruction, with such systems as the Carnegie intelligent tutors, Diagnoser,¹¹ IMMEX,¹² and Summary Street¹³ (e.g., Minstrell, 2000; Vendlinski and Stevens, 2002). He noted recent integrative work by Black

and Wiliam (1998) suggesting that effective formative assessment can improve student learning by 0.4-0.75 standard deviations as determined in a variety of end-of-course achievement measures. In addition to improving student learning, such embedded formative assessment offers the possibility of reengineering current models of assessment so that information is obtained from performances proximal to the teaching and learning process. Under some scenarios in which technology supports the integration of teaching, learning, and assessment processes, it may no longer be necessary to divert attention away from ongoing teaching and learning activities so that students can prepare to be tested on external accountability or so-called drop-in-from-the-sky measures of achievement. Instead, the information needed for various assessment purposes might be derived more directly at the classroom and school levels from data streams derived from technology-based learning activities.

Finally, Pellegrino discussed some of the types of research that are necessary to make full use of technology to improve learning. He mentioned the need for research that differentiates among various technology-based tools with respect to their relative impact on teaching and learning, including attention to their respective costs and payoffs. This includes study of the impact and cost-effectiveness of general tools such as browsers, presentation programs, and spreadsheets versus tools with a content focus of a domain-general (e.g., Geometer’s Sketchpad^®14) or domain-specific (e.g., Worldwatcher¹⁵) nature. There is a need to better understand how much impact we can achieve with tools that do and do not draw on knowledge of domain-specific teaching and learning issues.

He also drew a distinction between research focused on designing technology-based solutions, in which the goal is to understand the impact of the software on what students can learn and understand, and research focused on understanding the conditions impacting implementation and use of those systems, such as teacher knowledge, infrastructure, and organizational constraints. He also discussed the importance of having a

support structure for education research so that each new education research project isn’t forced to reinvent the school-based partnerships necessary to carry out research. The requirements of such a research support structure have been described in the National Research Council’s study for a Strategic Education Research Partnership (National Research Council, 1999c, 2003).

Edward Lazowska of the University of Washington began his talk by discussing the numerous examples of failed predictions that technology would revolutionize education. These include predictions about the impacts of film, radio, and television. In each case, initial hype was followed by a struggle to produce material for the new medium, then by a more mature judgment about the capabilities of the medium, and finally by a sense of disappointment and cynicism. This cycle was renewed with each appearance of a new technology.

Since computers have already passed through several stages of hype about their potential for affecting education, it is reasonable to ask why we should believe that this time will be any different. Lazowska discussed several reasons that he believes make this time particularly promising for the ability of information technology to have a substantial impact on K-12 education. First, he stressed the importance of the progress in the learning sciences over the past few decades. This work has not yet been effectively exploited by education in general or education technology in particular, leaving a huge opportunity for educational gains. Second, there is the tremendous power of the hardware advances that is typified by Moore’s law, which describes the doubling of transistor density on integrated circuits every couple of years. Although that progress is continuous, we tend not to notice it until it suddenly crosses some threshold. He cited the Internet as an example of this process, going through regular doublings since its inception in 1969 but not bursting into public consciousness until it reached a critical mass in about 1993. Third, he stressed the importance of networks and the Internet in connecting people, allowing exploration, interaction, and the creation of communities in ways that previous technologies have not allowed. Fourth, he observed that the education community is moving toward a more widespread understanding that the focus of education technology should be on teaching and learning, not on the technology itself. Finally, he noted that in the current sophisticated media environment, students are accustomed to engaging media and communication technologies. This level of comfort by users with the technology itself will provide a strong demand for learning environments that are more engaging than traditional instruction.

Lazowska then turned to the kind of capabilities that are offered by information technology. First, he listed a number of capabilities that he described as “boring” because their function is straightforward. These capabilities include accessing information, publishing information, collaborating with others, building communities with others, improving class and school administration, adapting materials for learning disabled and physically disabled students, and making use of remote scientific instruments. Although the technologies involved in providing these capabilities are not particularly exotic, they can be used to provide a much more engaging learning experience for students and a broad base of support for teachers. Fundamentally, it is the provision of these “boring” capabilities that is at the heart of the first transformation, which must address huge deployment, integration, and support issues.

He then discussed the more complex and “exciting” capabilities that technology can provide. One of these is the ability to create self-paced and adaptive learning systems. These offer the possibility of simulating the kind of effective intervention and personalized instruction that individual human tutors are able to provide, which has been shown to substantially increase student learning. Such systems also offer the possibility of incorporating ongoing formative assessment that would reduce the need to devote large portions of classroom time to student testing. Finally, technology offers capabilities for complex simulations, exploratories, and clip models. Lazowska illustrated this point with several examples of web-based models providing simulations of phenomena in physics.¹⁶ Later in his talk, he also mentioned the example of the Digital Human project, a sophisticated multilevel simulation of the human body that the Federation of American Scientists is trying to advance.¹⁷

The research in technology that was necessary to develop these more complex technological capabilities is substantial. Lazowska referred to these as “multidisciplinary grand challenge scale problems.” In contrast, it is instructive to compare the level of research in education as a fraction of total expenditures with that in other industries. For example, the semiconductor industry’s research share is over 80 times that of education.¹⁸ In the report on education of the President’s Information Technology

Advisory Committee (2001), the overriding recommendation was to make the effective integration of information technology with education and training a national priority. That report called for the establishment of a major research initiative for information technology in education and training. Lazowska stressed that this step has not yet been taken: the technology offers tremendous opportunities for the next generation, but a serious research and development effort that could realize those opportunities has not yet been achieved.

Robert Tinker of the Concord Consortium focused his presentation on applied research and innovation that meld work in technology, the learning sciences, and educational practice. He argued that these form a crucial missing link between the earlier stages of basic research in cognitive science and new technology and the later stages of dissemination and professional development. This intermediate stage includes the development of educational applications and IT-based curricula, along with research specifically focused on implementation. This intermediate stage is essential to bring about the major advances in education that technology makes possible. However, such efforts currently are both underfunded and often entirely overlooked in policy debates. He stressed that meaningful change in classrooms does not come from a single development but instead from a series of insights and innovations that cascade and evolve from more basic to more applied research.

Tinker described implementation research as being somewhat similar to medical field trials. Done correctly, it should include large numbers of students and teachers and focus on in-school studies. As an example, he described the Concord Consortium’s Modeling Across the Curriculum¹⁹ study of the use of computers to model different areas of science, which involves 13 schools and 10,000 students. The study employs a consistent approach to modeling across high school courses in biology, chemistry, and physics. It also uses random assignment of students into two different versions of the modeling approach, one more open-ended and the other more structured. The study has received $7 million in funding from the Interagency Education Research Initiative, a jointly supported project of the National Science Foundation, the U. S. Department of Education, and the National Institute of Child Health and Human Development. He stressed that this is the first time that this level of support has been supplied for this kind of serious implementation research.

In addition to inadequate support for implementation research, Tinker also noted that innovations themselves are not being funded. He outlined the funding structure and emphases of the Math/Science Partnerships, the various national laboratories, resources and systemic initiatives, and the Centers for Teaching and Learning. Although these efforts are important, they focus on implementing, disseminating, and providing professional development for innovations that already exist, not on creating new innovations. He likened this to funding the construction and employment of a big conveyor belt without offering support for developing the goods that would be placed on the belt. Such innovation research will not come from the researchers who are engaged in basic research or from business or from the schools.

Tinker also described an idea that he called “education accelerators.” These would be interdisciplinary research centers for applied, school-centered research. The accelerators would promote, support, and study large-scale, theory-based change that is supported by existing research evidence. Because such large-scale change is inherently risky, the education accelerators would provide a system of insurance and assurance, with ongoing formative assessment built in to provide an early warning system that would mitigate the risks of change and make sure that any mistakes involving students would be quickly corrected. He envisions that such research centers would receive base funding for staff and for a general research agenda with 5-year renewable grants. However, the bulk of their funding would come primarily through peer-reviewed grants to affiliated institutions.

Tinker concluded his remarks by outlining the level of funding that he believes would represent a balanced research agenda for research related to education technology. This research agenda would span the range from basic cognitive research to innovation in technology, software and curriculum, implementation research, a set of education accelerators, to human resource development at all levels. Box 3-2 reproduces his funding outline, giving an order of magnitude estimate of

funding required for the different types of research. The outline shows that only one-third of the funding would be used to support the cognitive, technology, and software-related research that is typically thought to be the focus of education technology research. Two-thirds of the proposed funding would foster curriculum innovations and different types of implementation research that focus on the use of that technology in classrooms.

Tinker finished his presentation by noting that this level of funding is small compared with the size of the education enterprise itself, and it is about the same order of magnitude as the current efforts being spent to implement, disseminate, and provide professional development for innovations that already exist. Despite the relatively modest size of this proposed investment, he believes it has the power to transform learning in K-12 education.

Nora Sabelli of SRI International spoke as an invited commentator. She stressed the importance of conducting research on the adaptation process that is central to change in education. In this context, she noted that it is unreasonable to expect teachers to aggregate pieces of curriculum from different software developers; such aggregation must be part of the solution that the research and development community provides. She also noted that the adaptation process in schools usually doesn’t involve a single innovation but rather a complex of innovations in curriculum, instructional materials, and pedagogy. As a result, it is important to think about aggregating innovations in ways that are easy for schools to adopt.

In addition, Sabelli talked about the importance of carrying out long-term research to understand the processes of educational change. She argued that typical collaborations between a researcher and a set of teachers are so short and perfunctory that they are over before the researcher and the teachers adequately understand each other’s needs and potential for contributing to solutions to the problems that they should be addressing together.

David Vogt of the New Media Innovation Center also spoke as an invited commentator. In his remarks, he stressed the importance of allowing students to own their learning experiences. He contrasted this “pull” model of education with the current “push” model in which students are not in control of their own learning experiences. He argued that introducing a dynamic push-pull tension into education would make an enormous difference in students’ enthusiasm and participation. As an example

of work that would be relevant to a pull model, he briefly described industry entertainment research in gaming and collaboration that could be adapted to education. He argued that students already have a high level of sophistication with information technologies and that if they are not given control of their own educational experience, they will simply take that control on their own.

As was done for the first transformation, four breakout groups developed lists of key enablers after the presentations dealing with the second transformation. Participants then voted for their top two candidates. The complete list of key enablers transcribed from the poster board sheets of the breakout groups is included in Appendix B. This section briefly describes the top choices.

The discussion in one of the breakout groups identified the fundamental change required for the second transformation as the creation of an ongoing system that allows education to be continually improved through research. A leading candidate for a key enabler of this change was to define a set of targets for research and development that can motivate people, coupled with intermediate milestones to make it clear when progress has occurred. One important aspect of the definition of goals is that it be done in a way that engages the public so that there is broad public and policy support for a vision of the improvement in learning that is possible from research and development in the use of technology.

Several versions of this key enabler received support from a number of the participants. One version referred to targeted test beds that would focus on proof-of-concept support for the first transformation. Another version referred to the LENS partnerships discussed by Roy Pea in his presentation. A third version referred to the creation of technology parks whose mission would be to focus on the use of cognitive science and technology to improve education. These would be similar to university-industry partnerships in science, medicine, and engineering, with open sharing of intellectual property and involvement by teachers and graduate students.

Several groups included versions of this key enabler, which recognizes the driving role played by assessment in the education system. One version mentioned the importance of conducting research and development on formative assessments, while another version mentioned moving beyond paper and pencil assessments. Some of the discussion mentioned the potential to use IT-supported tools to assess more complex 21st century skills, which would in turn allow greater emphasis to be placed on those skills in the curriculum.

Several groups included key enablers addressing issues about the market for education technology that prevent research from being translated into goods and services. One group mentioned possible changes in the tax structure. The discussion in another group focused on creating a forum to reconcile the divergence in views between suppliers who argue that there is no coordination of requirements for purchasing and K-12 practitioners who argue that suppliers do not understand or care about their particular needs. Finding a resolution of this impasse could open a substantial market to industry while providing transformational tools to education practitioners.

The final session of the workshop focused on a discussion of the ways that the National Academies could partner with teachers, industry, learning researchers, and policy groups to help bring about the two transformations in the use of information technology to improve learning. The session began with invited comments by Milton Goldberg of the Education Commission of the States, Marshall Smith of the Hewlett Foundation, Terry Rogers of Advanced Networks and Services, and Michael Feuer of the National Research Council. Following their individual comments, the discussion was opened to all workshop participants.

The following summary of six suggestions integrates the invited comments of the different speakers along with the general discussion. This format brings together related comments that were made at different times by different speakers.

A number of the participants commented that it would be useful for the National Academies to identify effective uses of IT in K-12

education and raise awareness about promising IT tools that have been developed but are not widely known or used by schools.²⁰ Marshall Smith noted that there are a number of high-quality and highly effective IT tools available that educators do not know about and therefore do not use. Henry Kelly of the Federation of American Scientists noted that one of the comparative advantages of the National Academies is in being able to serve as a neutral arbiter in identifying what is new and different about these particular tools. In addition, Smith suggested that the National Academies could conduct design projects related to important areas, such as English-language learning, to describe how existing IT capabilities could be combined to meet pressing educational needs.

Milton Goldberg spoke about the importance of disseminating information from existing National Research Council reports that relate to the use of IT to improve learning. He suggested that it would be useful to form partnerships with constituent groups to explore ways to make the information in such reports more widely accessible. In addition, he underlined the importance in the current budget climate of helping state policy makers understand what technology can do to improve education.

Larry Snowhite of Houghton Mifflin Company suggested that it would be helpful for the National Academies to work jointly with policy makers and industry to facilitate the application of research findings to the development of educational materials. Several other participants argued that the National Academies could play a useful role in identifying the IT tools that are available and defining some criteria for the adoption of those tools. Roy Pea spoke about the possibility of using the convening power of the National Academies to provide a way for the publishing and research communities to work together.

In addition to identifying effective IT uses and tools, some participants noted that the National Academies can help identify policies that facilitate or hinder the use of those IT uses and tools. One aspect of helping to identify policies that promote effective use of IT would be to conduct a cost-benefit analysis of various IT approaches, along with research that demonstrates the effectiveness of employing IT to improve learning. Goldberg argued that this is an important role for the National Acad-

emies to play. As a negative example, Michael Feuer discussed one of the side effects of accountability testing, which is to hinder the ability of teachers to use more creative approaches in their teaching if they aren’t convinced those approaches will lead to direct improvements in test scores. This comment echoed the earlier comment by Cheryl Lemke that the pressure of high-stakes tests often leads teachers to reduce creative uses of technology.

Kelly noted that one of the areas of comparative advantage for the National Academies is in defining a research agenda. There were many other comments that referred to the importance of defining a research agenda while stressing the importance of focusing that agenda on issues of particular concern. Goldberg suggested a focus on the achievement gap as a way of defining a research agenda for the use of technology in education that addresses issues that people care about. Smith noted that providing accommodation in special education is the one area in which technology already has had a large impact.²¹ He argued that a research agenda for the use of IT in K-12 education should be focused on similar targeted areas, such as reducing the achievement gap and using speech and language technologies to help English language learning. Steve Rappaport agreed that there had been far too little emphasis in the workshop discussion on people who have been left behind. Roy Pea discussed the inclusion provisions of the No Child Left Behind Act of 2001, arguing that they provide an opportunity for researchers, industry, and teachers to come together to find ways to use technology to improve the learning of those students who have not been making adequate progress with more traditional approaches to teaching and learning.

Terry Rogers provided a different theme for focusing a research agenda: he argued that it would be helpful to identify the hard questions that must be answered to realize the dream of using IT to transform K-12 education. As one example, he suggested the question of defining the teacher’s role in an educational environment that takes full advantage of technology’s ability to personalize the learning experience for students.

Feuer discussed the current policy focus on scientifically based research in education and suggested that the National Academies could help construct appropriate research designs for demonstrating the effectiveness of IT applications. Goldberg elaborated on this point to note the importance of understanding when a clinical trials approach is appropriate and helping to communicate that importance to local policy makers who would be involved in such trials. Several participants argued that it is important to think carefully about appropriate research designs in relation to the speed of technological change. Pea discussed the difficulty of producing relevant results with long-term research designs when the technology being tested is changing rapidly. Smith noted the difficulties involved in a proposed clinical trial of computer tutors that would not have completed testing until the underlying technology was a decade old.

There is widespread and long-standing concern that the market for education technology is broken in some fundamental ways. Feuer noted that the former Office of Technology Assessment issued a report in 1988 that made this claim. Snowhite spoke about the frustration that publishers feel in dealing with the education market, because of the uncertainty introduced in spending decisions by political pressures. Rogers commented on the wide gulf separating the expectations of practitioners and industry representatives for education technology products. He argued that it would be helpful for the National Academies to carefully investigate this market failure and to broker a new understanding between industry and K-12 education about their respective needs. He referred to the morning’s discussion of the LemonLINK project as an inspiring example because of the project’s decision to negotiate with industry to obtain hardware and services that would work for them. He suggested that the National Academies could play an important role by focusing on difficulties with the market for education technology and identifying solutions that have been proposed.

Rogers discussed the separation of researchers and teachers in K-12 education. In particular he commented on the lack of ownership felt by K-12 practitioners in the current body of education research, which is

perceived as coming from outside the community it is attempting to influence. He contrasted this separation with the organizational research literature on how innovations are developed and used and how organizations evolve and make progress. He stressed that there are important lessons to be learned from this literature, many of which are probably applicable to research in education. In particular, he argued that the literature shows that innovations are unlikely to be successful when the people who implement them are entirely separate from the researchers who design them. Although a gulf between researchers and practitioners can also arise in industry, there are usually management structures in industry that attempt to bridge the gap. No corresponding organizational structure works to bridge the gap between research and practice in education.

In general, the comments discussed during this final session of the workshop indicated that participants believe there is an important ongoing role for the National Academies to play in helping to bring about the two transformations in the use of information technology to improve learning in K-12 education. These comments share an agreement that the convening power of the National Academies can bring clarity to a number of difficult issues related to the use of IT in K-12 education. At the same time, participants were concerned that the National Academies find ways to bring together researchers, teachers, and industry representatives so that the findings from National Research Council studies can be effectively used by the entire community.