Sei sulla pagina 1di 12

Li, L.-Y., & Chen, G.-D. (2010).

A Web Browser Interface to Manage the Searching and Organizing of Information on the Web
by Learners. Educational Technology & Society, 13 (4), 86–97.

A Web Browser Interface to Manage the Searching and Organizing of


Information on the Web by Learners
Liang-Yi Li and Gwo-Dong Chen
Department of Computer Science and Information Engineering, National Central University, Taiwan //
lihenry12345@gmail.com // chen@csie.ncu.edu.tw

ABSTRACT
Information Gathering is a knowledge construction process. Web learners make a plan for their Information
Gathering task based on their prior knowledge. The plan is evolved with new information encountered and their
mental model is constructed through continuously assimilating and accommodating new information gathered
from different Web pages. In the process, learners have to maintain many search results for later use and
coordinate information between different topics and different memory aids for understanding. However,
currently widely-used Web browsers provide ineffective support for the process. Learners frequently encounter
these problems, disorientation and cognitive load, and thus impede their learning. This study proposes an
innovative Web browser interface called TopicBrowser, in which learners can make an Information Gathering
plan with a hierarchical concept map. Based on the map, they can search information, group open Web pages
into associated topics, extract information from within Web pages and organize it, and evaluate their
Information Gathering process. An experiment with thirty participants was conducted to compare TopicBrowser
interface with conventional tabbed browser interface. Experimental results show that TopicBrowser can lower
the degree of disorientation, increase learners’ retention, and increase the efficiency in completing an
Information Gathering task. These results suggest that Web browser interfaces should provide effective ways of
extracting and organizing search results. The well-organized information can significantly reduce disorientation
and cognitive load and thus can improve learners’ learning during Information Gathering processes.

Keywords
Information Gathering, Web browser, Structural overview, Cognitive load, Disorientation

Introduction
With the increasing number of published electronic materials, the World Wide Web (WWW) has become a vast
resource for individuals to acquire knowledge, solve problems, and complete tasks that use Web information.
Recently many educators are incorporating Web resources into their curricula to enhance teaching (Lin, Cheng,
Chang, & Hu, 2002; Ruthven, Hennessy, & Deaney, 2005). Students are also increasingly taking the Web resources
to complete course assignments and solve course problems (Chung & Neuman, 2007; Pennanen & Vakkari, 2003).
Therefore, seeking information on the Web has become an important learning activity in current learning
environment.

One of information-seeking tasks often performed by students is Information Gathering, which is the extracting,
evaluating, and organizing relevant information for a given topic (Kellar, Watters, & Shepherd, 2007; Sellen,
Murphy, & Shaw, 2002). For example, students search and gather information for constructing background
knowledge on a research topic.

Information Gathering is an active learning process. In order to create the knowledge, the learners apply the
abstraction mechanism, by choosing those concepts that they are interested in and eliminating the others. Among
these concepts there are the topics, too. Web learners then find and gather information by a self-directed way. Their
mental model or knowledge structure is constructed through continuously assimilating and accommodating new
information found from different Web pages. However, currently widely-used Web browser interfaces do not offer
effective support for the construction process. Learners frequently encounter two problems, disorientation and
cognitive overload, and thus simplify their Information Gathering task to finding an answer or a Website (Wallace,
Kupperman, Krajcik, & Soloway, 2000).

In this paper, we developed a Web browser interface, called TopicBrowser, in which a learner can make an
Information Gathering plan and extend the plan with a hierarchical structure. Based on the structure, they can
organize their search results that consist of Web pages opened in the workspace, information extracted from within
Web pages, and notes taken by themselves into a consistent structure. Using TopicBrowser the learner creates a
ISSN 1436-4522 (online) and 1176-3647 (print). © International Forum of Educational Technology & Society (IFETS). The authors and the forum jointly retain the
86
copyright of the articles. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies
are not made or distributed for profit or commercial advantage and that copies bear the full citation on the first page. Copyrights for components of this work owned by
others than IFETS must be honoured. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior
specific permission and/or a fee. Request permissions from the editors at kinshuk@ieee.org.
hierarchical concept map based on these topics. It intends to reduce learners’ disorientation and cognitive load and
provide support for constructing knowledge with Web information.

Related work
Information Gathering

Information Gathering is a knowledge construction process. Web learners begin this process with recognizing an
anomalous state of knowledge related to a topic (Cole, Leide, Behesht, Large, & Brooks, 2005). This state is the
interest or concern mental state that triggers the information gathering process. Thus, they make an initial search plan
based on their prior knowledge. With each piece of new and useful information encountered giving them new ideas
on their topic, they thus extend or evolve their plan to other relevant topics/subtopics (Lin & Belkin, 2005) or
associate the piece of information with their knowledge structure. Finally, the process is ended up with resolving the
anomalous state.

Information Gathering is a very complex information-seeking task. It can be completed not by a specific answer but
by a series of extractions, comparisons, and syntheses of a broad range of information related to these
topics/subtopics (Morrison, Pirolli, & Card, 2001; Sellen, Murphy, & Shaw, 2002). Learners are frequently required
to maintain many extracted results for later use and reference. However, to keep a huge amount of information in a
human’s mind is difficult because the limitation of working memory (Anderson, 2004). To support the limitation of
memory capacity, learners have to employ external memory aids.

When finding a Web page that is possibly useful, Web users typically use three kinds of memory aids for keeping
information related to this Web page. The first is bookmark collection. A bookmark represents the address of a Web
page as a title or a picture (e.g. the thumbnails of the Web page). A user can easily find a specific Web page from
his/her bookmark collection and return to the Web page using a single mouse click (Aula, Jhaveri, & Käki, 2005).
The second is keeping this Web page opened in the workspace (screen or window). In order to easily access content
of the previously visited Web pages or track search history, users may utilize this method (Aula, Jhaveri, & Käki,
2005; Hawkey & Inkpen, 2005). The third memory aid is to extract useful information (e.g. a graph, a table, or a text
segment) from this Web page and to place it in a text editor (e.g. MS Word) or take notes on the text editor or a
paper. The extracted information and taken notes often play the role of intermediary to help in resolving a user’s
information problem (Spink & Goodrum, 1996).

However, currently widely-used Web browsers, such as Internet Explorer (IE) and FireFox were primarily designed
for navigating with hypertext system and reading content on Web pages. They do not provide effective support for
the management of the search results. Thus, students have several problems when they construct knowledge during
Information Gathering processes.

The first problem is related to coordination of information. To coordinate information kept in the three kinds of
memory aids, students have to frequently change attention among them. The frequently changed on attention make
students easily disoriented. In addition, the structures of information organized in the three memory aids are
inconsistent. For example, students organize bookmarks in a hierarchical structure but keep open Web pages in a
sequential order. To find and recall a piece of information that is previously kept in these memory aids becomes
difficult.

The second problem is related to the switch between topics. When having multiple topics, students may use the
browser’s Back button or leave more Web pages opened in the workspace for coordinating information between
different topics (Spink, Ozmutlu, & Ozmutlu, 2002; Spink, Park, Jansen, & Pedersen, 2006; Spink, Wilson, Ford,
Foster, & Ellis, 2002).To return to a previous topic, students have to identify which Web pages are related to that
topic. However, keeping a large number of Web pages opened may clutter the workspace, making some specific
Web pages associated with a topic difficult for students to identify. Students must remember which Web pages are
associated with that topic and spend more time finding them from this cluttered workspace. The time and effort spent
easily disorient the students.

87
The third problem is related to the organization of information extracted from within Web pages. When finding a
piece of information that is useful within a Web page, students may extract it and place it in an appropriate position
in a text editor. To extract information from within Web pages, students generally apply conventional copy-and-paste
operations. The operation makes students interrupt the primary task, like reading or browsing, and shift their
attention to the destination application (Marshall & Consulting, 2005). Interruptions in reading or attention shifts can
impair learning and retention (DeStefano & LeFevre, 2007). Thus, when returning to continue the primary task,
students have greater difficulty in remembering what has been done before and in continuing the primary task.
Moreover, if students want to save more detailed context information, like the URL and the title of the Web page
from which the text was collected or to take notes on the copied and pasted resources, like for making a summary
and an outline, they have to issue more copy-and-paste operations.

These three problems result in disorientation and increase cognitive load, and increase the time spent completing
Information Gathering tasks, thus reducing the effectiveness of Information Gathering.

Software interfaces for management of Web information

Various software interfaces have been developed to support management of Web information. Some interfaces
support users in evaluating and gathering topic-relevant Web pages, and in organizing them into personally
meaningful collections. For example, Bookmarks are the most popular way to organize topic-relevant Web pages.
Bookmarks consist of lists of URLs of Web pages. Users can organize their bookmarks into a hierarchical category
structure. Data Mountain, which is a software interface, represents documents as thumbnail images in a 3D-desktop
virtual environment. It utilizes spatial memory to find and manage Web pages (Robertson et al., 1998). The
Topicshop software interface can display more detailed information for each kept Web page, including a thumbnail
image, link count and image count. Users can select significantly high-quality Web pages, and organize them into
personal collections quickly and easily (Amento, Terveen, Hill, Hix, & Schulman, 2003). The VisSearch provides an
integrated Web search environment in which students can create a concept map of their Web search results and
organize useful Web resources in the concept map (Lee, 2004, 2005).

Such interfaces can support organization of bookmarks and thumbnails of Web pages. However, they do not support
the organization of open Web pages in the workspace and organization of information extracted from within Web
pages. When a student wants to read or use a piece of information within a Web page previously kept in these
interfaces, he/she has to spend a period of time to reload the Web page, recall why he/she kept the Web page, and
identify which pieces of information are useful in the Web page.

Some Web tools allow users to extract information from within Web pages without leaving their Web browser
window. These tools reduce the impact of the interruption and attention switch in reading or browsing. For example,
ScrapBook (http://amb.vis.ne.jp/mozilla/scrapbook/) is a text editor embedded in FireFox. Users can mark a piece of
Web information and store it in ScrapBook using a shortcut or a toolbar button. Google Notebook
(http://www.google.com/notebook/) is a tool for extracting information from within Web pages. Users can add clips
of text, images, and links from Web pages to a Google Notebook without leaving their Web browser window. Hunter
Gatherer is a Web browser that enables users to extract a piece of information from within Web pages and place it
into an archive as transparently as highlighting (Schraefel, Zhu, Modjeska, Wigdor, & Zhao, 2002). These tools can
help users maintain their focus on the desirable Web pages while extracting information from within Web pages.
However, we did not find any study that has examined the effect of these tools on Information Gathering process.

There are two categories of Web browser interfaces that are currently widely used for keeping open Web pages: the
window Web browser interface, in which a Web page is loaded in one window and is represented as a tile in the
toolbar, and the tabbed Web browser interface, in which a Web page is loaded in one document and is represented as
a tab pane within a window. The tabbed browser interface has been reported to be better in managing open Web
pages than the window browser interface (Aula, Jhaveri, & Käki, 2005). However, Web pages opened in both the
window browser interface and the tabbed browser interface are arranged in a sequential linear order. When students
open many Web pages for different topics in a sequential linear order, identifying a specific Web page or several
Web pages associated with a specific topic becomes difficult.

88
Information Gathering is a very complex information-seeking task. Students have to maintain a broad range of
information related to different topic/subtopic concurrently. Previous studies have focus on developing tools to
support students in managing bookmark collection and information extracted from within Web pages. However, less
study has focused on support students to manage Web pages opened in the workspace. When switching between
topics in a cluttered workspace, students are easily disoriented. In addition, previous studies have focused on
providing support for individual memory aids. They did not provide an integrated approach for the management of
search results kept in different archives. Students have to frequently change focus to maintain information in these
memory aids. Moreover, the structures of information organized in these memory aids are inconsistent. It may
increase students’ cognitive load and disorientation when they maintain information kept in these memory aids.

The description of TopicBrowser


TopicBrowser is a Web browser interface. It hosts a Microsoft WebBrowser control and adopts the functions
provided by this control to mimic the conventional tabbed Web browser interface as closely as possible so that
students can easily learn how to use it. This interface is divided into three sections, toolbar, browsing, and collection
management (Figure. 1). The toolbar section contains seventeen buttons that can be used to perform basic operations,
such as refreshing the active Web page, opening a new topic, and collecting text segments into a topic archive. There
are two tiers in the browsing section. Students can create a new topic in the first tier (above the browsing section)
and open Web pages related to this topic in the second tier (below the browsing section). In the collection
management section, each piece of Web information extracted from within Web pages is represented as a node in a
hierarchical structure. Students can edit any node and organize nodes in this section.

Figure 1. Three sections in TopicBrowser

The conceptual architecture of TopicBrowser is presented in Fig. 2. A student can utilize the provided functions that
include grouping Web pages into topics, collecting information from within Web pages into topics, highlighing and
89
commenting on Web pages to learn on the Web. All information related to TopicBrowser is recorded in a database.
The annotation server processes the requests for retrieving or storing all annotations, highlighted texts, collected text,
and the hierarchical structure with the hypertext.

Figure 2. Conceptual architecture of TopicBrowser

Developing an Information Gathering plan into a hierarchical structure

During an Information Gathering process, students can develop and extend their information plan into a hierarchical
structure. Based on the structure, they can search information for each topic and associate each piece of found
information with the structure.

When clicking the button (new topic), a student can create his/her search plan into a hierarchical structure (a node
in the structure represents a topic/subtopic). They also can extend their plan to other relevant topics/subtopics during
the Information Gathering process. For example, a teacher asked his/her students to construct the background
knowledge related to diabetes by Web information after class. A student firstly created a topic “diabetes” and used
the term “diabetes” to search on a search engine. He/she found that there are three kinds of diabetes: “type one
diabetes”, “type two diabetes”, and gestational diabetes. He/She then created three topics for them and collected and
organized web information for these topics respectively. Figure 1 shows that a student planned and developed his/her
information Gathering plan into three topics for his/her homework assignment (Figure. 1).

Structuring of search processes for handling topic switch

When creating a new topic in the hierarchical structure, a new tab pane that represents this topic is then automatically
generated in the first tier. Students can search information and open Web pages related to this topic in the second
tier. Each open Web page is represented as a tab pane. The operations for opening a Web page are the same as in
conventional tabbed Web browsers.

A Web page also can be moved from the active topic to a pre-existing or new topic. The workspace is well-organized
once all open Web pages have been grouped into topics. Students can switch between their topics by a single mouse
click. For example, a student created three topics for his/her homework assignment. At the same time, three tab
panes entitled “diabetes”, “type one diabetes”, and “type two diabetes”, are present in the first tier (Figure. 1). The
topic “diabetes” is the active topic. When the student searches for information on the topic “diabetes” he/she opened
six Web pages for this topic in the second tier. Later, he/she found that a Web page is related to the topic “type one
diabetes.” Thus, he/she moves the Web page from the topic “diabetes” to the topic “type one diabetes.”
Additionally, the first tier presents the title of every topic and can be regarded as a navigational aid that reminds
students of the topics on which they are working.

90
Structuring of search results for scaffolding knowledge construction with Web information

There are four kinds of tree nodes can be created in the collection management interface: information tree nodes,
which represent extracted text segments; note tree nodes, which represent notes that have been made; Web page tree
nodes, which represent bookmarked Web pages; and folder tree nodes, which categorize collected Web information
according to topics. TopicBrowser employs the “paste-before-copy” model (Apperley, Fletcher, & Rogers, 2002) to
collect information.

When students create a new topic in the hierarchical structure, TopicBrowser automatically generates a folder tree
node with the same title of this topic in the collection management interface. This folder tree node is also
automatically set up as the destination node of any text segment extracted from the Web pages grouped into this
topic. For example, three folder tree nodes entitled as “diabetes”, “type one diabetes”, and “type two diabetes” are
generated in the collection management section and structured hierarchically.

To extract a text segment from a Web page, students first mark this text segment and issue a “highlight” command
by clicking the button or pressing a shortcut key. Then, an information tree node with a title assigned by students
is generated under a corresponding destination node. This title is a simple description of the particular text segment.
It can be seen as a summary of text in this text segment. Each information tree node consist of six elements: (1) the
title of the topic from which the text segment is extracted; (2) the title of the Web page from which the text segment
is extracted; (3) the URL of the Web page from which the text segment is extracted; (4) the content of the text
segment; (5) the location of the marked text segment, and (6) the time when the text segment is extracted.
Meanwhile, this marked text segment is highlighted in yellow. The highlight can be treated as a visual feedback
indicating that this text segment has been extracted successfully. It also can remind students where to continue their
reading and which Web pages he/she has read. Based on these operations, students can extract a text segment from a
Web page and place it in the corresponding folder tree node without leaving the Web browser window. They can
thus maintain continuity of attention in their primary task.

Students also can take notes at any location of a Web page by clicking the button or a shortcut key; and a note
tree node with an assigned title and note is generated under a corresponding destination node. The operation to add a
note tree node is similar with the operation to add an information tree node.

The information tree node related to a marked text segment can be automatically generated under a predetermined
destination node. Students also can select a folder tree node as destination node when extracting a marked text
segment. When marking a text segment within a Web page, a student can select a folder tree node in the collection
management interface and click on the selected node using the right mouse button. A menu is then shown. This
student can click the item “add this marked text segment” in this menu; then an information tree node is thus
generated under this selected folder tree node. Similarity, this student also can click the item “add a note” to generate
a note tree node under this selected node.

Gathering and organizing Web information into the collection management section is a knowledge construction
process. The tree structure in the collection management can be seen as students’ knowledge structure. Students
construct understanding of a topic by finding and associating Web information distributed on different Websites with
their knowledge structure. The tree structure also can be seen as a navigational aid that presents a structural overview
to help students recall and review which topics have been developed, what information has been collected, and the
relationship between topics and the collected information.

Students can create their search plan and extend their plan to other relevant topics/subtopics into a hierarchical
structure. Based on this structure, they can search information for different topics under respective tab panes and
open Web pages into associated topics. In addition, they also can associate Web information found and gathered
from different Websites with the structure. Therefore, they search and organize search results based on a consistent
structure.

Experiment
An experiment was conducted to evaluate the TopicBrowser interface. It aims to test whether the TopicBrowser
interface was more efficient in Information Gathering than currently widely used Web browser interfaces.
91
Participants

Thirty participants were recruited from National Central University in Taiwan. Twenty-four participants were
graduate students, and six participants were undergraduate students. These participants had relatively long histories
of general Web use (Mean=8.3 years, SD=2.09). Almost all the participants could be regarded as experienced Web
users. All of them use tabbed browser as their primary Web browser interface. They have no previous experience in
using Topicbrower before this experiment

Design

The experiment had one independent variable, Web browser interface, with two values: TopicBrowser interface and
the tabbed Web browser interface (they used FireFox in this experiment). The experimental design was within
subject. Thus, a repeated measures tool was used. The dependent variables are the subjects’ Information Gathering
performance measured in terms of these three variables: (1) length of navigational path, (2) text recall, and (3) time
spent to accomplish Information Gathering tasks. The navigational path was employed to test whether TopicBrowser
can improve students’ disorientation. The text recall questionnaire tested whether TopicBrowser can improve
students’ retention. The time spent to accomplish Information Gathering tasks tested whether students can efficiently
complete Information Gathering tasks by using TopicBrowser interface.

Materials

The experiment involved repeated measurement. Some learning and memory effects were generated at post test. To
reduce potential learning and memory effects, two materials were prepared for this experiment. One material was for
the TopicBrowser interface, the other for the tabbed Web browser interface. Each material consisted of a hypertext
system, a paper handout, and a memory test questionnaire. The language used in these materials was Chinese.

A hypertext system comprised eighteen Web pages: three topic Web pages and fifteen content Web pages. Each
topic Web page presented five hyperlinks linking to content Web pages. Three topics in the hypertext system of the
first material were “diabetes”, “type one diabetes”, and “type two diabetes.” Three topics in the hypertext system of
the second material were “stroke”, “embolic stroke”, and “thrombotic stroke.”

The paper handout consisted of twelve task descriptions (T1~T12). Each text description consisted of a part of a text
segment extracted from a content Web Page and an information cue for hinting to the participants about the topic
where the text segment can be found. To force the participants to switch between topics, the tasks presented on the
handout were listed in a specific order.

A memory test paper was composed of 30 questions, classified as topic, title and content questions. Three questions
were topic questions, in which the participants were required to write down titles of the three topics in their hypertext
system. Twelve questions were title questions. The participants were required to recall the titles of the twelve
collected text segments. Fifteen questions were content question. The participants had to recall a blanked-out
keyword in the text segment of a task description.

Procedure

The experiment was carried out in a usability laboratory. The experiment was divided into two parts. One part was
carried out using TopicBrowser interface, and the other was carried out using the conventional tabbed Web browser
interface. We assigned a material for part one and the other material for part two for each participant. The assigned
materials were counterbalanced between the participants

Each part was divided into three phases, named instruction, experiment, and memory test:
 Instruction phase: The whole experimental procedure was explained to the participants, and any questions
regarding the set-up were answered. Each participant was assigned a material. The participants were then given
five minutes to navigate the hypertext system of this material.
92
 Experiment phase: The experiment phase lasted one hour. Each participant was requested to extract the text
segments listed in the twelve task descriptions from the hypertext system, put them into an archive (a Word
document when using conventional tabbed Web browser interface and the collection management interface
when using TopicBrowser), assign a title for each text segment as quickly as possible. All computer activities
were recorded by MS Media Encoder.
 Memory test phase: All participants had to complete a memory test immediately following the experiment
phase.

Results
Navigational path

Navigational path was to test whether TopicBrowser can reduce students’ disorientation during Information
Gathering. The number of nodes a subject visited while finding a text segment for a task was counted as one
navigational path. A longer navigational path means less efficient navigation, as well as a higher degree of
disorientation in hyperspace (Otter & Johnson, 2000; Smith, 1996).

The participants collected information from a single topic during T1 to T4, two topics during T5 to T8, and three
topics during T9 to T12. As shown in Table 1, the participants performed a shorter navigational path with
TopicBrowser interface when involving in a single topic, two topics, or three topics than with the conventional
tabbed Web browser interface. The results show that the participants had lower levels of disorientation when using
TopicBrowser interface to find a text segment than when using the conventional Web browser interface. Therefore,
these results indicate that grouping open Web pages into associated topics can reduce the degree of disorientation.

Table 1. Navigational path value


Tabbed Browser TopicBrowser
Navigational path t-statistic df
Mean SD Mean SD
Total(T1~T12) 92.63 38.83 54.57 20.43 6.60 ** 29
One Topic(T1~T4) 25.03 15.90 14.60 7.26 3.40 ** 29
Two Topics(T5~T8) 35.33 20.88 20.27 13.60 4.90 ** 29
Three Topics(T9~T12) 32.27 18.37 19.70 9.99 3.49 ** 29
** p-value < .01

Additionally, the participants must switch from a current topic to another topic in T5, T8, T9, T10, T11, and
T12.When the participants used the conventional tabbed Web browser interface, the mean navigational path values
between the tasks (mean=53.97) that require topic switching and the tasks (mean=38.67) that do not require topic
switching were significantly (t=−3.11, p=0.004<0.01) different. However, the mean navigational path values were
not significantly (t=-2.34, p=0.026>0.01) different when the participants used TopicBrowser interface (as shown in
Table 2). These results show that the participants had a higher degree of disorientation when finding a text segment
with topic switching than without topic switching when using a conventional tabbed Web browser interface.
However, the participants may have closely the same degree of disorientation in both cases when using
TopicBrowser interface. These results indicate that TopicBrowser provided more effective support for topic
switching than the conventional tabbed Web browser interface.

Table 2. Comparing navigation path values in tasks require to topic switch and in tasks do not require to topic switch
Tasks that do not require Tasks that require topic
Navigational Path topic switching switching t-statistic df
Mean SD Mean SD
Tabbed Browser 38.67 21.81 53.97 25.31 -3.11 ** 29
TopicBrowser 23.20 10.61 31.37 16.70 -2.34 29
** p-value < .01

93
Recall rate in a memory test

The memory test was conducted to test whether TopicBrowser can improve students’ retention. A paired t-test was
conducted by comparing TopicBrowser interface and the conventional tabbed Web browser interface. As shown in
Table 3, the result was significant (t=−3.42, p=0.002<0.01), revealing that using TopicBrowser for Information
Gathering tasks can help the participants recall more information (mean=10.77) than Information Gathering tasks
done using the conventional tabbed Web browser interface (means=8.57). By separating t-tests for the three classes
of questions shows that the mean score of the title question was significantly (t=−3.12, p=0.004<0.01) higher with
TopicBrowser interface (means=3.87) than with the conventional Web browser interface (means=2.80), but the mean
scores of the topic and content questions between both interfaces were not significantly different.

Table 3. Memory test scores


Tabbed Browser TopicBrowser
Memory Test t-statistic df
Mean SD Mean SD
Total 8.57 3.72 10.77 4.53 -3.42 ** 29
Topic question 2.77 0.50 3.00 0.00 -2.54 29
Title question 2.80 2.38 3.87 2.74 -3.12** 29
Content question 3.00 2.10 3.90 2.37 -2.03 29
** p-value < .01

Completed time

The completed time was measured to test whether students could efficiently complete Information Gathering tasks
by using TopicBrowser. A paired t-test was performed by comparing task completion times when using
TopicBrowser and conventional Web browser interface. The result was significant (t=-3.64, p=0.001<0.01),
revealing that the participants completed the twelve tasks more quickly by using TopicBrowser (means=837.90s)
than by using the conventional Web browser interface (means=1052.37s). The result shows that the participants can
complete Information Gathering tasks more efficiently by using TopicBrowser than by using the conventional Web
browser interface.

Discussion
This study has presented an innovative Web browser interface in which students can plan their search process and
organize their search results in a hierarchical structure. The results from the experiment show that TopicBrowser can
reduce the degree of disorientation and can increase students’ retention. Furthermore, the participants could complete
Information Gathering tasks more quickly by using TopicBrowser interface than by using the conventional tabbed
Web browser interface.

The participants significantly recall more title information while performing Information Gathering tasks with
TopicBrowser interface than with the tabbed Web browser interface. There are three possible reasons. First, the
participants assigned a title for each extracted text segment and organized the extracted text segments into a
hierarchical structure. The hierarchical structure can be seen as a structural overview. The participants can easily
remember the relationship between topics and the titles of these collected text segments from the hierarchical
structure than from a sequential list (text editor). Second, the hierarchical structure is presented in the collection
management section. The participants can easily view the hierarchical structure and were reminded of the titles that
had been developed. Third, TopicBrowser can be used to collect information for multiple topics without leaving the
browser window. The participants can maintain their focus on reading and browsing, thus reducing the negative
effect of interruptions when extracting information from within Web pages.

The participants can recall more topic and content information. However, the mean scores of the topic and content
questions between both interfaces were not significantly different. There are two possible reasons to explain these
results. First, the titles of the three topics are easily remembered. Therefore, most of the participants can recall the
three titles when using both interfaces. Second, the participants simply copied a text segment verbatim, assigned a
predefined title for this text segment for each task. They did not need to read the text they extracted in both
94
interfaces. Therefore, the mean score of content question between both interfaces were low and were not
significantly different.

Most participants (29) used shorter navigational path when using TopicBrowser than when using tabbed browser. .
There are two possible reasons to explain this result. First, TopicBrowser can group open Web pages into associated
topics, so the workspace is well-organized. The participants can easily switch between topics and find a specific Web
page from the well-organized workspace. Second, we found that most of the participants (19) use the browser’s back
button to switch between topics when using the tabbed browser interface. Thus, the navigation path is longer when
going to a specific Web page. To keep the workspace clear, these participants left only three Web pages opened in
the workspace when performing their Information Gathering tasks with the tabbed browser interface. Thus, when
returning to a previous Web page no longer currently open, they had to remember where that Web page was and then
use the “Back” button or type the URL of that Web page to revisit it. This navigation strategy was also found in the
other study (Wallace, Kupperman, Krajcik, & Soloway, 2000). However, when using TopicBrowser, in contrast, all
participants opened all Web pages in the browsing section and grouped the Web pages into associated topics when
conducting their Information Gathering tasks. They could switch between topics with a single mouse click and
access any Web page with one or two mouse clicks. Therefore, the number of mouse clicks and intervening pages the
participants experienced were reduced significantly by using TopicBrowser interface.

In addition, we also found that the participants sometimes performed very long navigational path when using tabbed
browser. There are two possible reasons. First, when the participants do not have any idea about where the target is,
they randomly visit the Web pages. Thus the navigational path becomes very long. Second, the participants used the
strategy “Back” button to navigate and thus frequently revisit the same Web pages many times. Because the
participants sometimes performed very long navigational path, thus the standard deviations are very high when using
the tabbed browser interface.

Information Gathering is a common information-seeking task. However, current Web environments do not
effectively support theis task. TopicBrowser allows students to develop their Information Gathering plan into a
hierarchical structure, to group open Web pages into associated topics, and gather Web information for different
topics into the hierarchical structure. It is an innovative browser interface that is different from current widely-used
Web browser interfaces. In the study, we investigated the effect of TopicBrowser in a controlled environment.
Because information-seeking behaviors are effected by many different factors, such as the information seeker, task,
domain, setting, outcomes, and search system (Marchionini, 1995), the understanding of the effect of TopicBrowser
in information-seeking behaviors is limited.

Conclusion
Information Gathering is a very complex information-seeking task. Web learners are required to work on more than
one topic concurrently and to maintain many search results for later use. This study propose a innovative browser
interface TopicBrowser that enabled students to build up their own mental model of the Web information involved,
to gather, organize, and manage Web information in a consistent structure, and to use this structure to evaluate their
Information Gathering process. The experimental results showed that TopicBrowser is more effective for
Information Gathering tasks than the tabbed browser interface. The results suggest that Web browser interfaces
should provide effective ways of extracting and organizing Web information. The well-organized information can
significantly reduce disorientation and cognitive load. Because these improvements, students may put more efforts in
finding valuable Web resources and constructing meaningful knowledge from these Web resources.

In addition, TopicBrowser also can be used for reading on the hypertext document. Because hypertext is a non-linear
document, learners very easily encounter disorientation. TopicBrowser provides note-taking functions. It also can
lower disorientation in a non-linear learning environment. Thus it may support students reading in the hypertext
document. In the future, we will conduct a field study to understand how students use TopicBrowser in their daily
Information Gathering and reading tasks. We also want to conduct a more in-depth testing that is performed with a
wide range of material, task, and target groups.

95
Acknowledgement
This project was supported by the Ministry of Economic Affairs, R.O.C. under contract numbers 98-EC-17-A-02-S2-
0022.

Reference
Amento, B., Terveen, L., Hill, W., Hix, D., & Schulman, R. (2003). Experiments in social data mining: The TopicShop system.
ACM Transactions on Computer-Human Interaction, 10(1), 54-85.
Anderson, J. R. (2004). Cognitive Psychology and Its Implications. New York: W. H. Freeman.
Apperley, M., Fletcher, D., & Rogers, B. (2002). The stretchable selection tool: an alternative to copy and paste. Interacting with
Computers, 14(3), 195-209.
Aula, A., Jhaveri, N., & Käki, M. (2005). Information search and re-access strategies of experienced Web users. Proceedings of
the 14th international conference on World Wide Web (pp.583-592). New York: ACM Press.
Chung, J. S., & Neuman, D. (2007). High school students’ information seeking and use for class projects. Journal of the American
Society for Information Science, 58(20), 1503-1517.
Cole, C., Leide, J., Behesht, J., Large, A., & Brooks, M. (2005). Investigating the anomalous states of knowledge hypothesis in a
real-life problem situation: a study of history and psychology undergraduates seeking information for a course essay. Journal of
the American Society for Information Science and Technology, 56(14), 1544-1554.
DeStefano, D., & LeFevre, J.-A. (2007). Cognitive load in hypertext reading: A review. Computers in Human Behavior, 23(3),
1616-1641.
Hawkey, K., & Inkpen, K. (2005). Web browsing today: the impact of changing contexts on user activity. Proceedings of the CHI
‘05 Extended Abstracts on Human Factors in Computing Systems (pp. 1443–1446). New York: ACM Press.
Kellar, M., Watters, C., & Shepherd, M. (2007). A field study characterizing Web-based information- seeking tasks. Journal of
the American Society for Information Science, 58(7), 999-1018.
Lee, Y. J. (2004). Concept mapping your Web searches: a design rationale and Web-enabled application. Journal of Computer
Assisted Learning, 20(2), 103-113.
Lee, Y. J. (2005). VisSearch: A collaborative Web searching environment. Computers & Education, 44(4), 423-439.
Lin, C. Y., Cheng, Y. J., Chang, Y. T., & Hu, R. P. (2002). The use of Internet-based learning in biology. Innovations in
Education and Teaching International, 39(3), 237-242.
Lin, S.-J., & Belkin, N. (2005). Validation of a model of information seeking over multiple search sessions. Journal of the
American Society for Information Science and Technology, 56(4), 393-415.
Marchionini, G. (1995). Information Seeking in Electronic Environments. Cambridge, England: Cambridge University Press.
Marshall, C. C., & Consulting, S. B. (2005). Saving and using encountered information: implications for electronic periodicals. In
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp.111-120), New York: ACM Press.
Morrison, J. B., Pirolli, P., & Card, S. K. (2001). A taxonomic analysis of what World Wide Web activities significantly impact
people’s decisions and actions. In Proceedings of the CHI ‘01 Extended Abstracts on Human Factors in Computing Systems (pp.
163–164). New York: ACM Press.
Otter, M., & Johnson, H. (2000). Lost in hyperspace: metrics and mental models. Interacting with Computers, 13(1), 1-40.
Pennanen, M., & Vakkari, P. (2003). Students’ conceptual structure, search process, and outcome while preparing a research
proposal: a longitudinal case study. Journal of the American Society for Information Science and Technology, 54(8), 759-770.
Robertson, G., Czerwinski, M., Larson, K., Robbins, D. C., Thiel, D., & Dantzich, M. V. (1998). Data mountain: using spatial
memory for document management. In Proceedings of the 11th annual ACM symposium on User interface software and
technology (pp. 153-162). New York: ACM Press.
Ruthven, K., Hennessy, S., & Deaney, R. (2005). Incorporating Internet resources into classroom practice: Pedagogical
perspectives and strategies of secondary-school subject teachers. Computers & Education, 44(1), 1-34.

96
schraefel, m. c., Zhu, Y., Modjeska, D., Wigdor, D., & Zhao, S. (2002). Hunter Gatherer: Interaction support for the creation and
management of within-Web-page collections. In Proceedings of the 11th international conference on World Wide Web (pp. 172-
181), New York: ACM Press.
Sellen, A. J., Murphy, R., & Shaw, K. L. (2002). How knowledge workers use the Web. In Proceedings of the SIGCHI conference
on Human factors in computing systems (pp. 227-234). New York: ACM Press.
Smith, P. A. (1996). Towards a practical measure of hypertext usability Interacting with Computers, 8(4), 365-381.
Spink, A., & Goodrum, A. (1996). A study of search intermediary working notes: Implications for IR system design. Information
Processing & Management 32(6), 681-695.
Spink, A., Ozmutlu, H. C., & Ozmutlu, S. (2002). Multitasking information seeking and searching processes. Journal of the
American Society for Information Science and Technology, 53(8), 639-652.
Spink, A., Park, M., Jansen, B. J., & Pedersen, J. (2006). Multitasking during Web search sessions. Information Processing and
Management, 42(1), 264-275.
Spink, A., Wilson, T. D., Ford, N., Foster, A., & Ellis, D. (2002). Information seeking and mediated searching study. part 3.
successive searching. Journal of the American Society for Information Science and Technology, 53(9), 716-727.
Wallace, R. M., Kupperman, J., Krajcik, J., & Soloway, E. (2000). Science on the Web: Students online in a sixth-grade
classroom. The Journal of The Learning Science, 9(1), 75-104.

97

Potrebbero piacerti anche