Also published as a technical report (#TR-154) of the
Advanced Technology Group, Apple Computer, Inc.

Building a Foundation for Information Literacy:
Creating an Annotated WWW-Index by Children for Children

Y. Kafai, M. Bates, P. D. Braxton, D. Childs*, P. Ender, H. H. Lo, M. Martin, K. Rose** & L. Yarnall

UCLA Graduate School of Education & Information Studies
* UCLA University Research Library
** Apple Computer, Inc.


ABSTRACT

The growing Internet accessibility for educational purposes has raised a range of issues that concern not only children's emotional safety and the beneficial use of limited resources but also their development of critical skills in searching, retrieving, and using information media. As interactive electronic media become more pervasive in children's school and home lives, it is especially important that even young children are developing these skills so that they can build a solid foundation for their later learning. As a case in point, the report describes and discusses an educational intervention with over 200 elementary school children ranging from grade 1 to grade 6. During a five week period, students from six separate classrooms in the West Los Angeles area built together a collaborative WorldWideWeb-directory. In each class, the teacher and students would decide on one search topic related to their other classroom work. They would then search relevant sites, review them, and annotate their selections so that others could understand their choices. The resulting WWW "SnapDragon"-site is an annoted index by children for children (http://www.gseis.ucla.edu/SNAP/snapdragon.html). One goal of this intervention was to help children develop criteria of "information relevancy" as they were reviewing sites and writing their evaluations. A preliminary analysis of the evaluations indicates that many children placed a primacy on "visual relevancy" in contrast to "text relevancy"-a result which reflects two trends: (1) many WWW-sites are not written in child-adequate language thus making it difficult especially for young children to gather pertinent information, and (2) children's growing acquaintance with visual media for collecting information about their world and culture. We furthermore discuss issues regarding children's information searching skills and collaborative models in classrooms with limited Internet access.


INTRODUCTION

The rush to connect schools to the Internet is in full swing. In the report "America's Children & The Information SuperHighway" Lazarus & Lipper (1994) concluded that it is critical that all children have access to the net and furthermore that children's needs must be given a high priority as technology strategies are developed and implemented. Netday'96, a non-profit initiative, aimed to connect twelve thousand California schools to the Internet in March 1996. While many schools have begun to give students Internet access, allow them to browse and search information, and to build their own WWW-pages, it is still an open question in which ways Internet activities can effectively be used and integrated with other classroom learning.

The "SNAPdragon"-project was created to investigate how children can interface with the Internet by asking them to build an annotated directory of web sites for other kids. The project's general goal was to build children's information literacy skills and to develop children's understanding of the value and relevancy of the information they gathered from various sites. For that purpose we asked them to write annotations to the sites they had visited.

Furthermore, we wanted to provide children with a context in which the search and retrieval of information would no longer be an isolated experience but connected to their classroom learning and to a larger social goal-to share their insights with other children. Instead of keeping the annotated search results confined to individual classrooms, we implemented a collaborative index that could be accessed by other children through the WWW. We also saw the construction of an information structure such as an index as another powerful way for children to gain better understanding of existing WWW-directories such as Yahoo etc.

By observing students' efforts we could start to answer questions such as: Can children effectively use the search engines currently available? Can they find appropriate resources in a directed search? Can they evaluate and use the selected resources? The SNAPdragon project was a first step and only a few classrooms were involved in the preliminary research. But we hope that the insights gained in this study will help teachers and researchers to design better learning environments in regard to classroom management, children's information search and evaluations, and students' motivation and interest.


PROJECT OVERVIEW

For the implementation of the SNAPdragon project, we took advantage of an existing consortium of West Los Angeles schools, called The School Networking Action Project (SNAP). SNAP was started by the schools and Apple Computer. It was funded largely by Pacific Bell's CalREN (California Research and Education Network) initiative to afford California schools high speed networking capabilities. Pacific Bell and a series of other business and industry partners provided the participating schools with high speed phone lines, servers and Internet access. At the beginning of the project in January 1996 most of the schools (but not all their classrooms) had been connected to the Internet. But there had not been extensive use of this connectivity due to a series of technical problems.

To begin the collaboration, a call went out to all the SNAP schools in November 1995, asking teachers and technology coordinators to send in applications indicating their interest. Schools were informed that the SNAPdragon project would run concurrently with an UCLA graduate seminar in which students would not only examine and discuss the educational benefits of telecommunication activities but would also be required to hold an internship in a local school. By January we had received over 20 applications. Teachers, UCLA graduate students and instructors met once in January to get to know each other and to select the sites. For this project, only six classrooms participated because we wanted to match each classroom with one UCLA graduate student or researcher. The classrooms were selected in such a fashion that at least one classroom from each SNAP school was represented. The following table provides an overview of the grade levels, number of students, number of computers with Internet access in each classroom (see Table 1).


TABLE 1
Overview of Participating SNAP classrooms


School Grade Students Internet Classroom Curriculum
Name Computers Search Topic Integration

-------------------------------------------------------------------------------------------
Open School
1 3/4 64 1 CityBuilding Commissions yes

Westwood Charter School
2 5 9 1 National Parks no

Corinne Seeds UES
3 1/2 I 33 6 Airports/Iditarod yes

4 1/2 II 31 6 Ocean Animals yes

5 3/4 30 15 Black HistoryPoetry yes

Hawthorne School
6 6 29 1 Ancient Egypt yes



Each classroom had a different setup, different number of students, and a different search topic. Inspite of all these differences, we hoped that the collaborative activity would provide enough "glue" to bring all the students and their teachers together. The following summaries provide an overview of how the SNAPdragon-project was implemented at each site.


1. The Open Charter School (Louise Yarnall) is located in an ethnically and socioeconomically diverse section of the city of Los Angeles. The school has enjoyed a high degree of academic autonomy over the years because in the 1980s it received support from Apple Computer to install computers in every classroom. The SNAP classroom, a combined third- and fourth-grade class involved in this project, had two teachers, 64 children, and one Power Mac 6100 linked to the Internet via a 14.4 baud modem. We dialed into a server at UCLA for access.

The class was involved in a yearlong project of building a city of the future, called City Building Education (Nelson, 1984). Since the children were already organized into various commissions for building the city, we structured the Internet search sessions around these commission schedules and themes:

Historical Commission museum web sites
Building and Safety Commission landmark web sites
Imagination Commission artists/composers/filmmakers
web sites
Transportation Commission mass transit web sites
Environmental Commission environmental web sites
Social Services Commission web sites related to homelessness,
unemployment; poverty, and
disabilities
Communication Commission web sites related to E-mail and
sharing information over the Internet
Agricultural Commission web sites related to food, gardens
and cooking

On most Mondays and Fridays in February 1996, two hours of the school day were set aside for Internet time. We managed the Internet searches by having four of the commissions take half-hour turns sitting in a circle of chairs around the computer. The search session was structured by the researcher, who came to class equipped with URL address lists and then looked up the Web sites as the children observed. This strategy was adopted because it seemed the most efficient way to describe how the Web worked to large groups of children. As the sessions evolved, the students took turns working different "jobs" related to the search. There was a "mouse master," a "reader," and a "scribe" to record each commission's annotations. These jobs were rotated for each Web page search.

Internet search sessions were focused on critiquing the Web pages. These sessions challenged students to move beyond simple "I like it" and "I don't like it" answers and toward offering evidence for their opinions. The sessions also gave students a chance to practice reading and writing. Issues of vocabulary came up repeatedly because many Web sites included difficult words.

The sessions also introduced the students to basic concepts about the Internet: how it is a network of computers linked by a spider's web of telephone lines; how the data from these computers far, far away are broken down into little bits for the journey through the telephone lines and then reassembled inside their computer; and how there are certain tools for browsing the Web: scroll bars, back keys, forward keys, hypertext links and bookmarks. This lesson, although far from complete, seemed to be as much new information as the students could stand. We will have to wait for a later date to introduce them to search engines and techniques of selecting search terms.

The Web searching process revealed a dearth of "child-friendly" sites, particularly concerning subjects considered too technical for kids, like new transportation technology. We found that including the word "kids" in searches usually turned up more colorful, interactive sites in plain English. By far, these were the children's favorites. In total, the children received 9.5 hours of Internet time. The preparation time outside of the lesson was about equal to in-class time, divided equally between searching for good sites and troubleshooting technical problems.

The highlights of our findings were that: 1. children prefer Web sites with high visual content and short, simple textual content; 2. children are eager to create their own work and put it on the Internet; 3. children want to use the Internet for recreation, long-distance communication and class project research; 4. children are inspired to talk about their social views and surroundings when they view Internet sites featuring children's artwork and photographs from other places; 5. children would like to see more animation and interactivity on the Internet; 6. children have a low tolerance for long download times; and 7. children require support in their search process, particularly with scanning text and using hypertext links.


2. The Westwood Charter School (Marlene Martin) is a public West Los Angeles City School located in a middle class, mainly single family dwelling neighborhood. Nine students from a fifth grade classroom participated. The ethnically diverse group consisted of 3 girls and 6 boys all of whom were ten or eleven years old. The children were comfortable using computers and had adequate typing and spelling skills. Many of them had access to computers at home. The children's teacher encouraged the use of technology in the classroom and was herself quite knowledgeable. The school was fortunate to also have a very knowledgeable and enthusiastic technology coordinator who had previously worked in the software industry.

The children met for two hours once a week for six weeks. Generally the first part of each session was used for discussion, mini-lessons and problem solving. Since there was only one computer available, the children were divided into self selected collaborative groups of three. The children selected the topic "National Parks" and the three groups took turns searching the net for sites containing information about National Parks. Halfway through the project, the children were asked to find six National Parks as follows: two in Europe, two in the United States, one in either Africa or Australia, and an additional one not in the United States. This task was designed to encourage the children to perform more directed searches and to formulate alternative search arguments.

In the first two sessions the children explored the net. They learned how to navigate, perform searches, use bookmarks, visit sites and use hot buttons. As the children became more skilled, they were exposed to additional search techniques and helped to perform more directed searching. Finding European parks proved to be particularly difficult since the term 'European' did not appear in the titles or descriptions of the parks located in Europe and therefore the search results were poor. The children had to be helped to come up with better search arguments such as searching for parks in individual countries. Selecting potentially good sites from the list returned by the search engines also proved to be difficult, especially at first, because the children were attracted to the marketing and sales promotion sites. During the later sessions, the children demonstrated improved searching skills and were better at selecting sites for inclusion in the WWW directory for kids.

The children preferred sites with pictures and colorful graphics. Sites composed of only text or with more than 1 or 2 pages were left unexplored and in general children rarely spent more than a few minutes at any one site. The most popular site was the Hawaii Volcanoes National Park site. It contained a map of the caldera superimposed over a photographic image of the area. The colors were vivid and most of the names and locations on the map were hot buttons that linked the visitor to a brief description of the selected feature or area. These hot button encouraged the children to explore at a greater depth than if the information had been contained in a single site. While the children were enthusiastic about visiting Internet sites, they resisted writing annotations for the directory. Only after they saw other children's contributions were they willing to put effort into constructing the annotations.

Limited observation of these fifth graders suggests that children, age 10 and above, are capable of finding resources on the net. Their typing, spelling, vocabulary, and Boolean logic skills do come into play and can limit their ability to find appropriate resources. However, with direction, they are able to formulate search arguments, narrow or broaden a search depending on the previous search results; and they are capable of comprehending hierarchies when they have adequate base knowledge. Just as in the library, where the children often need the assistance of a librarian to find the materials they need (and so do most adults), children searching the net will also need an intermediary to help them track down appropriate sites. The level of comprehension and interest would suggest that the Internet may be a valuable resource for children.


3. The Corinne Seeds University Elementary School-Primary Grade I (Diane Childs) is a laboratory school located on the UCLA campus. Miss Dennis's primary class has 33 students, ages six through eight. Some are still learning to read; others are fluent readers; while at least one is quite advanced in reading and comprehension. About one third of the students have computers at home.

There are six computers (Macintosh LC II) arranged along one wall. Students have been assigned to particular computers. In addition to software for word processing and other applications, each is networked to the UCLA campus backbone and has access to Netscape. While e-mail is technically a possibility, it is currently enabled only for the computer Miss Dennis uses. There is also a laser printer that is shared by the six computers.

A rug surrounded by chairs is the area where students meet as a class. There are several tables designated as work stations for small group activities as well as a reading corner. All wall space is used effectively to highlight current projects and student work. For example, one wall is devoted to the "Community Building" theme which is a major focus of the class. One area currently showcases the Iditarod-unit which has been in progress since early March. Students have been working on construction of a community since last fall. The community has many components, including a farm, school, hospital, police station, harbor, and an airport. It is laid out, using wooden blocks, on a large area of the floor and has roads as well as structures.

Community Building: Airports. As a field trip to an airport was anticipated, Miss Dennis and Mrs. Childs decided to use airports as the theme for Internet exploration. Airport Internet sites were identified and two rounds of work with the students took place. Students were introduced to the Internet by working in pairs (two pairs at a time) at one computer. Students were coached to type in the URL for the Big Bear Airport. They were shown how to move around on the page and follow links. They were asked to "rate" the Big Bear page. They were also given lists of other airports with their URLs. It was anticipated that students could explore on their own those airport pages in which they might have some personal interest. This did not turn out to be a realistic expectation, however. In the second round, groups of four to five students were scheduled to work at individual computers exploring airport sites. Bookmarks had previously been put on each computer--a different selection for each in addition to the Big Bear Airport page that had been explored previously. Students were asked to select their favorite airport site and print off a copy of the page. Last, individual students worked with Mrs. Childs to select one site and critique it for other students. The results were posted to the SNAPdragon page.

Iditarod. Miss Dennis decided to incorporate the Internet into the unit on the Iditarod Dog Sled Race she had planned. There are several interesting sites. Miss Dennis gave each student the biography of a "musher" (dog sled driver) that she had obtained from the Internet and made a homework writing assignment. The students would then be following "their" musher throughout the race. The official Iditarod page includes frequent updates of the race standings, which Miss Dennis printed and reported to the class each day or so. She also posted a map of the route so that the mushers' progress could be tracked with flags. Even students without well-developed reading skills were able to find the Iditarod pages and recognize the status of "their" musher.


4. The Corinne Seeds University Elementary School-Primary Grade II (Kim Rose) of Mr. Raul Alarcon has 31 bilingual (English/Spanish) children. In his classroom are six networked Macintosh LCIIs which have direct Internet access from UCLA. Netscape's Navigator (version 1.1) Web browser is installed on each computer. The children spent a total of 10 sessions using the World Wide Web. In addition to me and the classroom teacher, a school librarian was also present and available for help at each session. Each of the sessions was approximately 30 minutes long. Some children casually used the computers in the morning before school started or at other free times throughout the week. A few children also had computers available to them at home.

The children were studying "ocean life" over the course of the semester. Prior to the introduction and use of the Internet, the class discussed sea life, read about fish and ocean mammals, did some library research, and collected appropriate books to have available in the classroom throughout the duration of the project. Each student selected one ocean fish or mammal to learn more about. The children planned to use the World Wide Web to find more information for their sea animal research reports.

The classroom teacher and I made the following two decisions prior to the children using the Internet: (1) Because the total project time was short and the children were so young, we decided that the adults would search the Web in advance to find subject matter and age-appropriate content. The teacher believed that a prepared list of appropriate sites would best serve as an initial starting point for the children as they began to explore the World Wide Web. Once the children downloaded the content, we wanted them to think carefully about the information they retrieved. We hoped to get the children to think about how some information might be more valuable or relevant to their needs than other. (2) We also agreed that we would not bookmark the sites within Netscape, but instead write the URLs on a poster-sized chart for the children's reference. I also created and distributed printed handouts listing a number of URLs which had pertinent ocean information for the children to explore. The teacher felt is was important for the children to type the URLs themselves to gain a broader familiarity with the keyboard as well as the Web browser's interface.

The children spent several sessions exploring the Web sites the adults had found to determine each site's relevance to their particular ocean animal. The children were very eager to share their findings with one another. If one child found an interesting site which did not contain information on the animal they were researching, but thought the material would be of interest to a friend, they would tell the other child. They also printed and exchanged relevant pages with one another. Some children had more success than others in finding information depending on what fish or mammal they picked. There was far more age appropriate information on dolphins, sharks and whales than for seahorses, star fish or eels for example.

The children learned that the information they found on the Web could be submitted by other adults or by children and that the information came from all around the world. As Mya (age 7) and I waited for one Web page to download I asked her what she thought about waiting. She said, "Well, since there's just one world -- its gotta be coming from somewhere in this world, but it could be coming from another city or country."

We wanted to get the children to think critically about the information or perhaps the misinformation they might be receiving. As with books and other "authoritative" media, the children were quick to assume everything they found about their sea animal on the Internet was correct, just because it was there. Many of the information providers may not have accurate information or the ability to communicate it effectively to children, however.

The children's annotations of the Web sites played a secondary role in our project. We did not begin to evaluate the sites until the third week into the study. Because of the children's young age we could not ask them to write up annotations on their own. I sat with the children one at a time to get their selections for the "best site" relevant to their sea animal. In general, the children had difficulty articulating why they picked one site above the others. The resounding reason was that "it had lots of information", or "it had good information".

We talked about whether the information was easy to read (it generally was not) and if there was both text and pictures at the site. If there were pictures we discussed how the pictures may have added to the written information. The children were generally highly accepting of all of the Web sites and believed they were all helpful and interesting. Their favorite selections were generally Web pages created by other children. It was interesting to observe that the children were most critical of a drawing they found at one site which was made by another child.

When the Snap Dragon site became "live" the children were delighted to see their own comments posted on the Web. It gave them a great sense of pride to know that children and adults from around the world could read their comments and also benefit the information provided in their "favorite" sites.


5. The Corinne Seeds University Elementary School-Middle Elementary Grade (Phyllis Braxton). In contrast to the UES primary grades, the Internet experience for the middle elementary students (Grade 3 and 4) was part of a computer lab directed by a teacher, Mr. Jaime Calderon. All the middle elementary students participate in a two week-long computer class, of which one week was dedicated to the Internet. Each computer class consisted of approximately 25-30 students. The classroom in which I worked consisted of 15 Power Macintoshes arranged mostly in table clusters of 2-4 computers. All of the computers had direct connections to the Internet. The browser used was Netscape, version 1.1. Three classes were part of the SNAPdragon experience.

Each new Internet class was structured by the Mr. Calderon giving an introduction of the Internet, providing a question-and-answer period and describing the goals of the class. Students then worked mostly in pairs on the computer. In the first round of Mr. Calderon's classes, students focused on the Black History Month as a theme for their Internet searchers. In the later classes, the teacher decided to change his focus to poetry, which all of the students were working on in their home room classes. The teacher provided a specific site most of the time for the students to visit and to evaluate for their annotations. In the final class I observed, he allowed advanced students use the search engines to find other types of information.

One of the primary issues was the obvious disparity between students who had prior access and experience with the Internet and those who did not. The students with experience using the Internet at home were more knowledgeable about the features, capabilities, and search engines of the Internet and tended to dominate the interactions at the computer. They often decided which sites to visit and how long to stay on a page before moving on. The students who were expert users tended to be boys.

Gender differences also expressed themselves in the beginning of the activity when the teacher asked students to describe what the Internet is and what one can do. Only boys raised their hands and provided answers such as "the information highway" or "places where you can find information"-descriptions they might have picked up from public media or adult conversations. In the last class, we introduced an activity that allowed all students to participate. We asked students to close their eyes for two minutes and to think of all the things and pictures that came to their mind when considering the word "Internet". Here nearly two thirds of the students had something to contribute. Students offered pictures such as the "spiderweb" or "telephone poles." One student even offered to draw a picture on the blackboard.

In the end, the time constraints (approximately five hours) were probably the most limiting factor of this Internet experience for the students and the teacher. Because students came for this class only to the computer lab, it took several minutes to assemble all the students, before the teacher could start the Internet activity. During the first sessions, students got acquainted with features of the Internet. Only in the last sessions students dedicated more time to the evaluation and annotation part of the WWW-sites. As in other classrooms, once students had visited personally the SNAPdragon site and read other students' evaluation, they became more involved in the activity.


6. Hawthorne Elementary School (Howard Lo) is a middle size public school in the Beverly Hills Unified Public School district with about 750 students in grades K-8. A class of 29 sixth-grade students participated in the SNAPdragon activity. The classroom of Mrs. Janet Galeko was equipped with one Macintosh LC computer that had Internet access via modem and a Netscape 1.1 browser. The computer was set up in front of the class on the teachers' desk and could be connected to a large TV monitor thus allowing to display the screen contents to the rest of the class. Most of students in this class own a computer at home and have seen their parents or siblings using it; 18 of them have experience on the Internet themselves. Furthermore, there are two experts, HTML programmers, in their class.

In their social science unit, students were covering ancient Egypt. In the discussion with the teacher, we decided that the SNAPdragon project would become part of students' project reports. In addition to reading in their textbook or consulting the library, students would use the WWW for finding additional information on topics of their interest in ancient Egypt. In a first conversation, students expressed interest in topics such as weapons and arms, fashion, and animals that they intended to follow up.

Because this was also the teacher's first contact with the Internet, we decided to take advantage of the large number of Internet experienced students in the class. Students were divided up into teams with 5-6 members each. A team member who was considered to be an "Internet" expert became then the tutor for the team that would guide the other students in their searches. In the first meeting, we gave with the teacher a general overview of the SNAPdragon project and showed students the "empty" SNAPdragon WWWsite. We conduced together some WWW searches on topics of ancient egypt using different search engines. Then we asked all the tutors to convene for a brief meeting at the class computer. We discussed how they would introduce the WWW to their team members, how they would assign and rotate different "roles" (such as mouse master or reporter). We also discussed how to formulate and constrain their searcher by using combination terms such as "ancient egypt" and "weapons" and how to bookmark sites.

For the remainder of the project, the teacher set up specific times at which each team could convene and conduct their searchers. Also, they were asked to record the URLs of the interesting and valuable WWWsites related to their topics and write down an annotation for each recorded site according to their own opinions. Some of these can be found in the SNAPdragon index. In sum, all students and the teacher found this a valuable activity and intend to incorporate it into future classroom activities.


SUMMARY

In March 1996, the SNAPdragon-project was brought to a temporary completion. We have appended to this report a printout of all the annotations provided by the students up to that point. In addition, the current WWW-site features samples of digitized drawings provided by the students. We concluded our intervention with another meeting in March to which all the participating teachers, even some students and other interested researchers and visitors came to share their experiences. While the SNAPdragon-project was not a systematic study of different classroom settings and students' information literacy skills, we nevertheless gathered some valuable insights for future Internet activities.

Internet Access. As noted in many policy reports and discussions, the access to the Internet is an important issue, if only for the limited resources of computers, phone lines and wiring. While providing all students with computers might not be possible, we found that working with even one Internet-accessible computer turned out to be feasible given certain classroom arrangements. In one case (Open School) we took advantage of existing team structures, i.e., the different city building commissions, and used the commissions and their topical interests to direct the Internet searches. In another case (Hawthorne School), we adopted a "tutoring" model in which the most Internet-experienced students became the tutors to teams of their classmates guiding them along their searches. In other classrooms, we build teams of students that searched together. In our effort to integrate the SNAPdragon activity with existing classroom activities, we found that (with some exceptions) these collaborative arrangements worked well and amplified students' learning of how to work together.

In our classroom observations we found considerable differences between those students who had opportunities at home or other places to familiarize themselves with computers and different tools and those for who it was a first experience. These differences appeared as early as in the primary elementary grades with students of ages 6 and 7. Expertise in manipulating the mouse, identifying interface elements such as buttons and menus, and finding keys on the keyboard varied among students.

We also were hesitant at first about the inclusion of primary elementary school students in the SNAPdragon activity. It was unclear how these young students would be able to use the WWW for their learning purposes because of their beginning writing and reading experience. One situation concerned the typing of the long string of HTML addresses. The children often typed the URLs incorrectly and were prevented from connecting to their desired Web site. The typing activity also took valuable time, leaving less time for the children to spend with the content material they required for their reports. Some of us felt it was a waste of time for the children to labor over the cumbersome, meaningless strings of letters and symbols. Our recommendation is to have children use bookmarks for either directing their searches or marking "valuable" sites.

Information Searching. One of the main goals of the SNAPdragon project was to provide a context for children to exercise their information searching activities for a larger social goal-to build a collaborative annotated WWW-directory for other children. While the younger students were not engaged directly in searching sites on the WWW, the older elementary school students conducted some searches and used different available Internet search engines. In initial efforts, students simply put in terms such as "National Parks" and expected the results to be their end of their searches. We found that asking students to pick different search engines, to input the same terms, and to compare and to discuss the different search findings resulted in a better understanding of search functionality. These sessions seemed to help best when introduced after students had experienced some frustrating search results. As some previous research on young students' information search abilities has found, students have difficulties of developing extending search strategies and developing criteria for satisfactory end results.

Information Selection. Selecting good sites was also difficult for students. Titles and descriptions returned by search engines were sometimes misleading and difficult for the children to evaluate. After some discussion, students improved their ability to identify sales and marketing sites versus informational sites. This problem was also compounded by their reluctance to read or scan the list of results. It appeared that only titles were used to decide which site to visit. Since the system was often slow, picking the wrong site precluded visiting additional sites.

The Web searching process revealed a dearth of "child-friendly" sites, particulary subjects concerning too technical for kids, like new transportation technology. We found that including the word "kids" in searches usually turned up more colorful, interactive sites in plain English. By far these were the students' favorites, but even they didn't really measure up to students' high standards. The students demanded higher production values: quality audio, video and heightened interactivity. The attractiveness for the later aspect showed itself in a site which student found in their searches about "ancient Egypt": they could submit their name to the site and have it returned in hieroglyphs. This was by far one of the most popular sites. Sites without graphics or pictures were almost always deemed boring and uninteresting and text-only sites were not explored. It was from reading the evaluations written by the students (see Appendix) it became evident that students placed a primacy on visual information on the WWW-sites and had more diffiuclties to evaluate the text-based content.

Information Evaluation. Creating annotations for this index, dubbed the SNAPdragon page, introduced students to the practice of critiqing other people' work. In critique sessions, students were challenged to move beyond simple "I like it" or "I don't like it" answers and offering evidence of their opinions. Their initial annotations were for the most part poor and it required a significant effort to persuade them to actually evaluate the site. They did not see the point or value of writing annotations. Two factors changed this. Often the researcher or teacher reminded students that their efforts would be seen on the WWW and that they were representing their school. The second factor had a greater effect. When the students were actually able to visit the SNAPdragon site and read other children's write ups the task became important. In one case, a student wanted to write "this site sucks"; only after he became aware that other students from all over the world would be able to read this annotation, he decided to expand his commentary. These sessions also gave students a chance to practice reading and writing. Issues of vocabulary came up repeatedly because many Web sites included difficult words.


CONCLUSIONS

In general, students' interest in Internet activities was high. The students, of course, were enthusiastic since "surfing the net" is the cool thing to do. They all expected that they would use the net in the future probably from home and several students talked about their parents giving them access to the net. For future activities, we recommend to build on children's interest in the net by providing them with opportunities to represent themselves to this community. SNAPdragon became one way to do so. We would probably start out by asking children to write something about themselves, their class or school and use it as a first submission to a site. Having an "Internet" identity seems to be of great value to students.

While the project in its first installment was initiated and organized by adults, we think that having the students handle all aspects, from creating the site, collecting annotations to indexing the entries will prove to be the best learning experience. In fact, one of the most valuable learning experiences for the members of the university course appeared when we were confronted with the task of categorizing the first SNAPdragon submissions. Due to the different search areas, content ranged from national parks to transportation and poetry. We had extensive debates which category names to choose and how to index entries-tasks faced by any professional information specialist. By default of any better solution, we adopted the Dewey classification system. While the categorization of musical instruments, animals and plants is a common activity in elementary classrooms, students usually do not deal with the wide range of issues that appeared in SNAPdragon indexing, or for that matter with any directory. We wished in retrospect that students had had this opportunity. Becoming information literate, or better, fluent, means more than knowing how to search for the right information. It means understanding all issues ranging from the making, finding and using of information.


ACKNOWLEDGEMENTS

We wish to thank all the teachers and their over 200 students who participated in this project: Raul Alarcon, UES; Sara Boulton, Westwood Charter School; Jaime Calderon, UES; Denise Cole, Open Charter School; Craig Davis, Bevery Hills Unified Public Schools; Faith Dennis, UES; Jane Geletko, Bevery Hills Unified Public Schools; Dolores Patton, Open Charter School; and Cathy Rodriguez, Westwood Charter School. We thank Doreen Nelson who joined the class meetings to keep us alert with her questions about educational purposes. We also thank Phil Ender for handling all the technical details of implementing the SNAPdragon site.


FOR FURTHER INFORMATION

Lazarus, W. & Lipper, L. (1994). America's Childrens and the SuperInformation Highway. The Children's Partnership, Santa Monica, CA.

Nelson D. (1984). City Building Education. Transformations: Process and Theory. Center for City Building Education, Santa Monica, CA.


APPENDIX

SNAP dragon WWW-site: http://www.gseis.ucla.edu/SNAP/snapdragon.html



updated