Introduction: Motivation and Goals

The growing global information infrastructure places increasing demands for materials to be available in digital format for research, business, and learning. While much information today is digitally produced, there still remains a considerable amount of potential digital library content in the form of primary or historical source materials (such as manuscripts, official records, oral histories, moving images, research data, and three-dimensional objects) that needs to be transferred into digital formats and further transformed or enhanced to be used effectively by diverse groups with widely ranging information and educational needs.

This project will develop a model for, and then use case studies to examine many of the educational, technological, and content issues connected with such digital transformation -- how primary source materials in the natural and health sciences might be transformed into digital library content and then used to enrich K-12 science education. There are several motivating factors underlying this research:

Science educators have stressed over and over again the crucial role that knowledge of scientific history and the changing nature of the boundaries between scientific disciplines plays in building a foundation for scientific literacy (Duschl, 1993; Wandersee, 1986). Students need not only to understand key concepts and principles of science and know how to apply scientific knowledge, but also to understand the cultural and social contexts within which advances in science were achieved. Hess writes that "the great strength of the cultural approach to issues of science and technology is in challenging people to consider how other cultures and groups may have new and different ways of defining the true and the useful" (Hess, 1994).

The development of the digital primary content necessary for such learning activities is a complex undertaking. While a considerable number of research efforts have been dedicated to the design of digital library architectures, less efforts have been concerned with the building and transformation of unpublished or non-bibliographic content for pedagogical uses. Content holders or information providers, such as archives and museums professionals, need to understand better how to prioritize their collections for digitization, and the most effective means for describing and visually representing digital versions of primary content for use by teachers and students. Digital library developers need to understand better the design and process issues associated with digitizing, storing, and retrieving contextualized and authenticated primary content.

From the perspective of teachers, the availability of primary source materials in digital format, previously inaccessible outside an archival or museum repository because of their unique, brittle, or valuable nature, presents a wonderful opportunity to enrich classroom activities and to place scientific endeavors into a wider social and cultural context (Hess, 1994). This availability also presents teachers with new challenges, however. For example, many teachers and students are unfamiliar with how to assess critically the quality and origination of unpublished or non-bibliographic content of varying quality and origins that they access through digital means such as the World Wide Web.

This project builds upon pilot studies underway at the Department of Library and Information Science (DIS) and the Department of Education within the Graduate School of Education and Information Studies (GSE&IS); collaborative relationships between GSE&IS and the UCLA University Research Library, the Los Angeles County Museum of Art, and the Smithsonian Institutions; as well as upon issues raised by the report generated by the National Science Foundation-sponsored Invitational Workshop on the Social Aspects of Digital Aspects held by DIS in January 1996.

We propose to develop a process model derived from an expanded concept of the life cycle of digital library content, one where the functionalities of systems developers, content creators, information providers, and end-users begin to merge as the digital library grows and becomes integral to learning activities. This model will be applied in a series of case studies in natural and health sciences education in an elementary school setting that will execute the full process of digital content transformation as follows:

Non-digital primary sources held by archives and museums will be transformed into richly described and highly structured content for a prototype digital library.

The prototype digital library will be accessed through a World Wide Web interface by teachers who will be able to select and incorporate some of the content and its descriptions into personalized information systems or "Digital Portfolio Archives" (DPAs). These DPAs will assist the teachers in preparing innovative curricular content that incorporates multimedia learning opportunities.

Students undertaking individual or team-based classroom projects will be able to access and selectively incorporate components of both their teachers' DPAs and the prototype digital library, as well as additional project-related materials that they have either located or created into their own DPAs.

All or any of the DPAs can optionally be incorporated into the prototype digital library, but as a distinct layer of user-created content distinguished from the base content of the digital library by an embedded description of its provenance.

This cyclical process model is also of value in that it provides an evaluative structure that facilitates feedback to systems designers and content holders/information providers about the most effective ways to identify and transform their holdings into richly described, user-accessible, digital library content. This, in turn, facilitates the identification of digital primary content by teachers, and its incorporation into innovative classroom activities, and the personalization of students' learning and use of digital information sources.

The content for this project will be drawn from the seven world-renowned archival, manuscript, and museum repositories at UCLA. Project investigators will work with these repositories and with teachers to select, describe, and digitize subject-appropriate digital library content for the case studies. Project investigators will also design and develop a prototype digital library architecture structured around an emerging SGML standard for description of archival materials, and incorporating the following functionalities:

contextual linking of related content;

strict tracking of versions and provenance of content;

hierarchies of content from broad descriptions of historical collections down to facsimiles of individual items;

manipulability by users of individual items of content; and
multiple browsing, retrieval, and display options available through a World Wide Web interface.

Teachers from UCLA's laboratory school, the Corinne A. Seeds University Elementary School (UES), will build DPAs by accessing the prototype digital library and selecting, and incorporating content they wish to use in the classroom, and then augmenting it with their own digital annotations and additional relevant materials. They will then use their DPAs in their classroom activities. The teachers' DPAs, or selected aspects thereof, will be made available to students, together with the full digital library, to use on class projects and developed into student DPAs.

The goals of this exploratory CRLT grant are to generate the following knowledge and processes that would allow us to build a foundation for larger research projects in the areas of digital library technology, inclusion of primary content in digital libraries, and the impact of both of these on K-12 teaching and learning:

Generalizable processes for structuring digital libraries and enriching content descriptions that meet the needs of primary content creators, providers, and non-traditional users of historical materials such as teachers and students (technical and content outcomes);

Knowledge of how digitized primary sources might contribute to the enhancement of students' understanding of, and engagement in the health and natural sciences and their cultural and social contexts (learning outcomes); and

Knowledge of how non-traditional users of historical or primary materials, such as teachers, might select primary content and integrate it into classroom activities if that content were digitally available (teaching outcomes).

Some of the key technological activities in reaching these goals are the following:

Develop a small prototype digital library designed especially to explore the design issues associated with storing and retrieving reliable digitized unpublished historical material.

Test the effectiveness of digital descriptions (finding aids) for primary content that are marked up according to an experimental Standard Generalized Mark-up Language (SGML) standard called Encoded Archival Description (EAD) in assisting elementary school teachers in retrieving and understanding the nature of, primary source materials. EAD was initiated at the University of California-Berkeley and is being developed by a team that also includes individuals from the Library of Congress, the National Archives and Records Administration, UCLA (Gilliland-Swetland), University of California-Irvine, Dreamworks SKG, Duke University, Electronic Book Technologies, the Research Libraries Group, and the Minnesota Historical Society (Pitti, 1994; Hensen, 1995; McClung, 1995).

Explore ways to enrich the EAD-encoded descriptions through the inclusion of teacher and student annotations of the primary content and how it was used in classroom activities.

Draw conclusions about the strengths and weaknesses of EAD as a potential infrastructure for digital libraries that have primary source content.

Develop an understanding of what elementary teachers and students' needs might be for locating, manipulating, and understanding primary sources in digital form; draw conclusions about how the content and retrieval needs of teachers and students might be better anticipated and incorporated into digital library and DPA development.

PROBLEM STATEMENT

With the growing accessibility of the Internet in classrooms, the potential for students' use of primary data sources increases proportionately. Many recent research efforts (e.g., Gordimer, Polman & Pea, 1994; Songer, 1995; Soloway et al., 1995) have taken advantage of this situation and provided students with access to weather data including analysis, communication, and documentation tools. Within the context of these learning enterprises, students often compile research reports that pull together digital data, information collected from books, and other resources. CSILE is one of the very few projects, however, that makes an explicit attempt not only to collect students' designs and annotations about biological knowledge but also to provide technical features to link them together (Scardamalia & Bereiter, 1995). There remains a strong need, therefore, to examine the potential architectural and intellectual relationships that might exist between student-created content and digital library content.

While previous research efforts such as those mentioned above have focused on important features of students' science learning, we propose, through the development of the prototype digital library and the personalized DPAs, to address two additional and relatively unexplored issues:

the processes by which historical or primary source materials can most effectively be incorporated into digital libraries; and
the importance of such materials as information and learning resources, and how they might be integrated into the learning process for building science understanding.

Considerable research efforts have focused on studying children's informal conceptions in science (Driver, Guesne & Tiberghien, 1985; Smith, diSessa & Roschelle, 1993/1994) and devised instructional interventions and technological tools. But less attention has been placed on the role of a knowledge of the historical context to science even though it has been recognized as an essential component in science learning (Duschl, 1993; Wandersee, 1986). By focusing on primary sources in the natural and health sciences, the proposed project will provide opportunities to make these aspects an integral part of children's science learning. The potential benefits of this approach include confrontation with conceptions held in earlier times based on then current knowledge; examination of societal pressures as driving questions for research agenda; understanding of tool development as an extension of human perceptions; and appreciation of different cultural approaches to, and attitudes toward science.

The integration of primary sources into the learning process raises another important issue, educational users are accustomed to working with materials that have already been selected, edited, or otherwise synthesized. With the increasing accessibility of the World Wide Web into classrooms, what intellectual skill set and digital toolkit are not only teachers but also students going to need in order to be able to select materials; consider their validity, and integrate them into their classroom activities?

From the standpoint of the development of digital library architectures and content that will lead to the enhancement of learning environments, especially for elementary levels, there are several more critical areas that need further investigation. For example,

what are the legal, technological, and economic constraints on primary content selection, and what limitations, perhaps even biases do these introduce into digital library content?

what are the most effective ways to present (intellectually and physically) authentic primary content to teachers and students (e.g., extensive files of digitized content or digital facsimiles; pre-packaged selections of digitized content, or descriptions only of non-digital content held by repositories)?

what mechanisms and heuristics will teachers employ to select digital content, and how will they use this content in the classroom? Can these mechanisms and heuristics be captured and incorporated into digital library design and descriptive structures in ways that might improve intellectual access by other teachers to the same body of digital content?

what ways might elementary students use digital content in their learning activities? How do they perceive the materials? Are there activities they would like to be able to perform that are not currently possible with digital library content?

can we empower teachers and students to become content creators and engage actively in the building and modification of digital libraries? If we give them the opportunity to create personalized information systems, what will they those systems look like, and what functionality will they need to have?

Elementary school students are truly the generation that will grow up using and creating digital libraries, and yet they have not often been the subject of study in digital library research. This project represents an important opportunity to introduce the digital library to these students, and also to learn more about their learning and content needs of a digital library.

By making students and teachers part of the full information cycle, we turn them into information creators and holders and use their portfolio development and the assessments they contain as feedback for a larger social goal‹to inform digital library designers, content providers, and other users about use, usability, and relevance. With this approach, aspects of information science-related work such as providing annotations and descriptions is made an essential part of the learning and teaching process. In fact, categorizing and describing are part of many elementary classroom activities, they are just not placed with a larger context of creating digital portfolios which are both personal and able to be shared beyond the parameters of the classroom. In other words, what we are proposing is a systematic approach that connects systems designers, content holders, and content users and informs each group about the needs of the other.

METHODOLOGY: Implementing and evaluating the DPA process model

The participating school site, the Corinne Seeds University Elementary School (UES) at UCLA is ideally situated for implementing the case studies for the following reasons: (a) its location on the UCLA campus facilitates the teachers' access to internationally renowned campus archives, manuscript, and museum repositories with strong science and health science holdings; (b) the availability in each classroom of computer equipment with Internet access and within the school of additional multimedia equipment facilitates project implementation; and (c) UES students are admitted on the basis of maintaining an ideal laboratory study population. As such, they are ethically and socio-economically diverse, and also include a bilingual (Spanish-English) cohort. This project will work with two upper elementary classes (4th and 5th grade), thus ensuring that the students are of an age where they are likely to have the necessary motor skills and grasp of intellectual concepts such as time, to be able to participate in the proposed learning activities.

The digital library and DPAs will be maintained on equipment located within the Department of Information Studies (DIS) (adjacent to UES), although they may be accessed, and content contributed remotely using GSE&IS and UCLA network capabilities. DIS maintains a systems laboratory which is a Type 1 10-megabyte ethernet network. The network is comprised of a DEC Alpha XP 150 machine running UNIX, a DEC Alpha Server 400 running NTS 3.51, and a 486 IBM compatible PC running Linux, a 2-gigabyte tape drive, and more than 40 multi-platformed workstations, several of which are dedicated for multimedia development. The network is connected to the UCLA Fiber Optic Backbone Network so that users have direct access to the Office of Administrative Computing's UNIX cluster and the Internet via the TCP/IP network protocol. The laboratory also maintains other digital multimedia and video equipment. Professional staff support consists of a Laboratory Librarian, a Digital Resources Librarian, and several graduate laboratory assistants. The DIS laboratory is one of two laboratories maintained by GSE&IS' Educational Technology Unit (ETU). ETU is also responsible for maintenance of and instruction in technology at UES.

We propose the following iterative design cycle that is divided into the following phases to capture the development and use of the digital library and DPAs:

Phase 1: Preparation of Digital Library Content
Content Selection and Description (by CONTENT PROVIDERS) (Gilliland-Swetland)
Participating content holders/information providers (UCLA museums, archives, and manuscript repositories) will be asked to compile a range of materials that meet the following criteria: (a) relate to the natural or health sciences, as broadly defined; (b) fall outside current copyright restrictions, or where copyright is held by UCLA; (c) are in sufficiently good physical condition that they are able to be digitized without sustaining damage; (d) represent relatively small, qualitatively dense, and self-contained collections of materials, or comprise individual items of exemplary value; and (e) are not subject to confidentiality restrictions. Providers will be supplied with a selection decision matrix that will go into these and additional criteria in more detail. Content providers may also employ additional criteria to make their selections, and will notify project investigators thereof.
Assisted by project investigators, the content providers will describe the materials they have selected using the Encoded Archival Description (EAD) standard. These EAD descriptions will be made accessible at a dedicated World Wide Web (WWW) site.

Content Selection, Description and Review (by TEACHERS) (Gilliland-Swetland, Kafai)
Teachers will review on the WWW (either from classroom or home) the EAD descriptions and select those materials that appear to be useful or interesting for instruction. In a template form provided on the WWW, teachers will provide annotations to the EAD descriptions on (a) why they chose the materials; (b) what they expect to find when they examine the actual materials; and (c) how they envision using the materials in the classroom.
Teachers will travel to the appropriate repository and review, in person, the materials they requested. They will confirm or revise their initial selections (perhaps eliminating some, following up new leads that emerge, or selecting only discrete items from a collection) and will provide additional annotations to the EAD descriptions that address the following questions: (a) what drew them to the materials they picked? (b) were the collections what they expected from reading the EAD descriptions? and (c) what other issues were generated by their selections?

Digitizing and Incorporation of Materials into Prototype Digital Library (Gilliland-Swetland, Kafai, Maddox)
All the materials selected by the teachers will be digitized at a minimum of 600 d.p.i.. A higher resolution takes longer to scan and requires more storage space but would allow for more subsequent flexibility in how content is digitally presented and disseminated.
EAD descriptions will be used as the primary infrastructure for the prototype digital library. Digital representations of primary content and teacher-contributed annotations will be linked as digital objects to the EAD descriptions in hierarchies that go from broad descriptions of historical collections down to facsimiles of individual items. Project investigators will also build in contextual links to additional related content; tracking of versions and provenance of content; manipulability by users of individual items of content; and multiple browsing, retrieval, and display options available through a World Wide Web interface.

Phase 2: Creation of Portfolio Digital Archives (DPA) by Teachers (Kafai, Maddox)
Teachers will download the digitized content that they wish to include in their teaching activities in the areas of the natural and health sciences. A research assistant will assist teachers in digitizing additional materials such as photographs, maps, drawings, or audio materials for inclusion in their DPAs. The use of production packages such as PageMill will allow teachers to circumvent the additional learning of the HTML formatting language. At this point teacher DPAs will consist of: (1) digitized representations of materials; (2) EAD descriptions; (3) Teacher's own notes (including their annotations to EAD); and (4) digital or digitized materials drawn from other sources that they decide to include.

Phase 3: Use of Teachers' Digital Portfolio Archives in the Classroom, and Creation of Students' DPAs (Kafai, Maddox)
Teachers will conduct classroom intervention in which students will write research reports. For that purpose, students will be asked to create their own DPAs that will contain the following items: (1) diary describing the information resources they used; (2) digital content from teacher DPAs and from the original digital library content; (3) annotations to EAD why they choose this one (3) any other digital or digitized materials that they choose to include (e.g., personal photographs, drawings, graphs). A research assistant will assist students in digitizing additional materials for inclusion in their DPAs. The use of production packages such as PageMill will allow students to circumvent the additional learning of the HTML formatting language.

Phase 4: Final Evaluation of Digital Portfolio Archives
Teachers will reflect on the success of their teaching intervention, on the ease of use and effectiveness of DPA in enhancing science learning, and technology integration and use in the classroom.
Teachers and students will reflect on their experiences with, the functionality of, and the potential of DPAs as personal information systems, as digital library content of value to a wider group of users, and as teaching, learning, and assessment aids.
Teachers and students will have the option of contributing their DPAs to the digital library where they will become a layer of value-added content distinguished on the basis of provenance.
Teachers and students will be able to retain and maintain their DPAs as "works-in-progress." DPAs will then become sources for personal reflection and self-assessment during the academic year, and resources for other projects and exhibits.

The benefits of this approach are manifold: (1) it draws on the diverse research strengths of all three project investigators; (2) it addresses the full information life cycle, from creation, through description, retrieval, use, and resulting creation of new materials that are incorporated back into an ever-expanding digital library; (3) it implements and tests an emergent national and international standard for the description of digital archival material; and (4) it includes assessment of all participants in these processes.

EVALUATION FOCUS

The following aspects will be investigated in more detail: