Web Searches Often Overwhelm Young Researchers
New Search Engines Aimed at Children's Needs Seek to Clear Confusion
Students in many schools have Web access that approaches the ideal of “anytime, anywhere.” But some experts say the academic payoff often is minimal.
“Lots of kids are ‘googling’ everything today, which throws them into this ocean” of information, said David Loertscher, a professor of library sciences at San Diego State University, referring to the ubiquitous Google online-search engine. “The major problem kids have on the Internet is judging quality; they don’t understand the difference between data smog, or advertising, or whatever.”
Mr. Loertscher, who trains future school librarians, points to the crude nature of Web search tools as a big part of the problem.
In search of better approaches, a growing number of educators are now experimenting with a new generation of search tools that presort results using simple, visual formats, rather than the endless lists of Web hits that often confuse students, and send them off on searches that waste valuable learning time.
Dania Bilal, a researcher who studies how children find information on the Web, said that even search engines marketed expressly for children have serious limitations.
The associate professor at the school of information sciences at the University of Tennessee, Knoxville, said child-tailored search engines do yield useful Web resources, and they insulate children fairly well from noxious Web pages. But the problem is they don’t allow the wide-ranging exploration that is a great part of the potential of the World Wide Web.
That’s the balance the developers of the new search engines need to find, Ms. Bilal said.
In one of two studies on middle-school students that Ms. Bilal has published in the Journal of the American Society for Information Science & Technology, children from ages 11 to 13 were given searches for science information to perform using Yahooligans! (www.yahooligans.yahoo.com), a search tool for children that is part of the Yahoo! Web directory and search engine owned by Sunnyvale, Calif.-based Yahoo! Inc.
The approaches taken by the children, who were 7th graders, to the search assignment were recorded using a computer screen camera.
After they completed the assignment, the students were interviewed by graduate-level researchers.
Ms. Bilal found that, sometimes, even when students used good online-search strategies, they found nothing. “The problem is that the indexing [of the search engine] was not comprehensive, and no vocabulary was used specifically for children,” she said.
Ms. Bilal said that, according to cognitive psychologists, children have less ability to recall from memory words that might make good search terms, compared to adults, but they do recognize such words when they see them.
Most students, if their assigned search failed to produce the expected information, did not try using a synonym or a term related to the one they had been given. They might have done so, however, if the search engine had an online thesaurus that offered a selection of alternative search words from which children could choose, Ms. Bilal said.
An even more basic problem, she pointed out, is that the Yahooligans! search tool does not suggest spellings of terms that the child may have misspelled, a feature that Google provides.
Yahooligans! offers topical directories as an alternative way to search, but that approach, while helpful, also has its problems.
The directories, even though they’re tailored for children, use categories that reflect grown-up understandings of the topic, Ms. Bilal said. In her latest study, which has not yet been published, researchers asked children to sort words themselves that were listed in online directories at Yahooligans! and KidsClick! (www.kidsclick.org), a child-oriented search tool that is maintained by the Colorado State Library in Denver.
In organizing terms from KidsClick!’s health and family category, for example, children placed “disabilities” under “family life” or “hospitals,” reflecting their awareness of disabilities as a circumstance affecting a family or requiring hospitalization.
But the directory, reflecting an adult perspective, classified “disabilities” under the relevant medical disciplines—a decision that would confuse many children.
“Kids organize concepts differently from the ways the search engines organize concepts,” Ms. Bilal said.
Even when children find good Web sites, she said, they easily become lost when perusing them or moving from site to site.
“The Web by its structure encourages discovery and exploration, but there are no landmarks,” she said.
One online tool offers some features that Ms. Bilal and other experts would like to see for the Web. The search tool is for the International Children’s Digital Library, a free online collection of fiction for 3- to 13-year-olds. The library is a research project of the Internet Archive and the Human-Computer Interaction Lab at the University of Maryland, College Park.
In the site’s “simple search,” children undertake a selection process by clicking on illustrated buttons indicating subjects that include real or imaginary animals; reality vs. make-believe; fairy tales; different feelings; and simply colors. Clicking on any button leads to a selection of online books.
Not only does this help children who do not read, experts said, it lets them base their choices on their learning styles or emotional connections to the content. Another search method used in the digital library—choosing books based on geography by spinning and clicking on a virtual globe—is the most popular, said Allison Druin, an assistant professor at the lab who serves as the project director of the library.
But she pointed out that adapting to the Web the features of the library’s search tool, which was designed with input from children, would be difficult.
“We’re better off right now, limiting the content to specific content, than giving kids general access” to the Web, she said.
One search tool already is helping some students in middle school and high school get better results.
Grokking the Web
Grokker2, developed by San Francisco-based Groxis Inc., is a software program that resides on a user’s desktop computer. It is used to initiate searches using Google or other Web search engines or databases.
The name comes from Robert A. Heinlein’s 1961 novel Stranger in a Strange Land, in which “to grok” is a Martian word meaning “to drink” and metaphorically “to be one with.”
Grokker2, which costs $49, aims to give users similar intimate and exhaustive knowledge of their topic. It operates by taking the results of the search—which can run through Google or a medley of popular Web search tools—and sorting them into a visual map of circles, which represent groups of Web pages that have more qualities in common than was determined in the original search. Each of the three-dimensional circles contains a portion of the search results, and the links to tens or hundreds of Web pages, depending on the search.
Using a computer mouse, the user can focus on any of the circles—and look at circles within circles—but the screen never presents more than 10 or so, so the user is not overwhelmed. Key words and basic information about the Web pages appear as the user’s mouse arrow passes over each circle, and circles that do not seem germane can be swiftly deleted by the user. The visual map can be presented on one-half of the screen, while specific Web pages are shown on the other.
Jason Epstein, a 7th grade science teacher in Palo Alto, Calif., started experimenting with Grokker2 this past summer. This fall, he installed copies of it throughout the computer labs of the 350-student Gideon Hausner Jewish Day School, which serves grades K-8.
He said students in grade 4 and higher have been using Grokker2 throughout the semester, replacing the general use of Google, which they now use only to search for images.
“Grokker’s pretty much taken over the middle school level—and the elementary school goes back and forth between Grokker and Yahooligans!, depending on the project,” Mr. Epstein said.
Vol. 24, Issue 14, Page 8Published in Print: December 1, 2004, as Web Searches Often Overwhelm Young Researchers