TCDL Bulletin
Current 2005
Volume 2   Issue 1

Kirsten Butcher and Sonal Bhushan
University Corporation for Atmospheric Research:
DLESE Program Center
P.O. Box 3000; Boulder, CO 80307-3000
{kbutcher, sonal}@ucar.edu

Abstract

Our research examined whether using strand maps as an interface for digital library search tasks would change learners' cognitive processes when seeking educational resources. Results demonstrated that strand maps can help learners engage with science content and that semantic-spatial interfaces can support meaningful search processes.

Introduction

Digital libraries face a significant challenge in connecting users to relevant content. Especially in science, users often lack the varied and specialized vocabulary necessary to find relevant resources using textual search [1]. Learners lacking adequate domain knowledge tend to engage in superficial Search behaviors by focusing on creation and revision of queries during information seeking [2].

This research explored whether strand maps – search interfaces that include semantic and spatial information – would better support student engagement with science content during digital library search for educational materials when compared to an existing textual search interface.

The Existing DLESE Interface

The existing DLESE search interface (http://www.dlese.org) is a good example of a successful textual interface onto a digital library discovery system. DLESE supports textual search in two ways:

• Searching on keywords or phrases that typed into the textbook.
• Using optional terms to retrieve materials and/or limit search results. For example, a user can search only for resources including images or other multimedia.

For a larger viewof Figure 1, click here.

For a larger view of Figure 2, click here.

Strand Maps for Conceptual Browsing

Strand maps [3] are node-link diagrams for science, math, and technology, where nodes represent learning goals and links represent conceptual relationships between learning goals. The links illustrate the ways that students' knowledge should progress as their education advances.

Strand maps are based on the educational benchmarks developed by the AAAS [3]; they describe what learners should know or be able to do at key stages in their education across the sciences, etc.

For a larger view of Figure 3, click here.

The Strand Maps Interface

Users can explore learning goals by navigating through different topics and grade levels within a map.

• Users can explore via direct manipulation; selecting strand names or grade-levels lets them 'drill down' and explore smaller portions of the map.
• The interface supports more hierarchical concept browsing – learners & teachers can explore the map by manipulating the folders on the left.

When users identify a relevant learning goal, they can click it to retrieve related digital library resources.

For a larger view of Figure 4, click here.

For a larger view of Figure 5, click here.

User Study

Goals:

• Determine whether different search interfaces would influence the cognitive search processes of digital library users.
• Contrast the multimedia strand maps interface with the DLESE textual search interface.

Hypothesis:

Because the stand map interface provides a visual representation of rich semantic content, its interface should promote increased use of content-based (science-focused) processes when searching for information.

Methodology:

• 12 undergraduates completed tasks describing common information needs faced by science educators (example below).
• Half searched for resources with the SM interface, half with DLESE's text search
• During tasks, students produced think-alouds that were transcribed and coded for cognitive processes identified in the verbal data.

Example Task

Jack is a 10th grade science teacher who has volunteered to fill in for Jan, a 7th grade science teacher. Jan was supposed to teach about changes in the Earth's surface. She suggests Jack come up with a classroom activity based on changes in the Earth's surface.

Jack teaches his 10th graders about earthquakes, and wants to teach the 7th graders about them too. He often uses DLESE to find activities and text, so he decides see what it has for 7th graders. He wants to find out which concepts he needs to teach and find classroom activities to support them.

Results of User Study

• A repeated measures analysis of variance tested for between-group differences (strand maps interface vs. textual search) in the number of statements related to science content made during search performance.
• Digital library users given the strand maps interface were significantly more likely to engage with science content (F = 6.03, p < .04) than those who used in the textual search interface.

For a larger view of Figure 6, click here.

• Compared to students using the text search, those using the multimedia strand map interface engaged more with science content when searching for digital resources.
• Students using the text-based search interface tended to focus on choosing keywords and experimenting with the combination of search options that would return the "correct" resource. They spent very little time talking about science content during search.

For a larger view of Figure 7, click here.

Conclusions

• Educational digital libraries can support engagement with important domain knowledge through search interfaces that provide conceptually organized semantic information.
• Students who use conceptual representations to pursue digital library materials are more likely to engage with science content during search compared to students using a text-based digital library search interface.
• Engaging learners with relevant science knowledge is seen as an important first step towards supporting learning in a digital library environment.

References

[1] Vakkari, P. A theory of the task-based information retrieval process: A summary and generalization of a longitudinal study. Journal of Documentation 57, 2001, 44-60.

[2] Marchionini, G. Information seeking in electronic environments. Cambridge: Cambridge Univ. Press, 1995.

[3] AAAS Project 2061. Atlas of Science Literacy. Washington DC: American Assoc. for the Advancement of Science (AAAS) and National Science Teachers Association, 2001.

Acknowledgements

Based upon work supported by the National Science Foundation Grant #0215640 to the University Corporation for Atmospheric Research (UCAR), and Grant #0226286 to the University of Colorado. Opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF.

© Copyright 2005 Kirsten Butcher and Sonal Bhushan
Some or all of these materials were previously published in the Proceedings of the 5th ACM/IEEE-CS Joint Conference on Digital libraries, ACM 1-58113-876-8/05/0006.

Top | Contents
Previous Article
Next Article
Home | E-mail the Editor