設計資訊網頁的連結架構與 使用者認知風格類型對瀏覽行為之影響

The Effects of Page-Linking Architectures and Cognitive Styles on Browsing Behaviors in the Web Sites of Design Information Services

張貞瑩(Chen-ying Chang)

指導教授 ： 陳立杰

大同大學/設計學院/工業設計學系所/碩士(2005年)

摘要

為了滿足不同類型客戶的需求，近年來有一些網站可在使用者瀏覽網頁時提供即時的客製化網頁。在設計資訊網頁的設計元素中，網頁架構是決定瀏覽效率的一個重要因素。此外，不同的使用者認知風格也可能對網頁瀏覽行為造成影響，進而影響瀏覽時的績效。本研究的目的，為探討在設計資訊網頁中的“連結架構”與“使用者認知風格類型”對網頁瀏覽行為的影響，並進一步訂定適性化設計資訊網網頁架構的設計原則。本研究首先進行GEFT(Group Embedded Figures Test)群體藏圖實驗，受測者經過篩選之後再進行第二階段的階層式與階層線性混合式網頁上機實驗，以分成三種類型的5個搜尋任務，進行實驗。利用雙因子變異數分析檢定不同的認知風格及連結架構對搜尋時間、瀏覽頁數、主觀評量是否有差異，並利用群集分析探討瀏覽路徑之差異。
研究結果顯示：(1)在一個資訊網站中，可依使用者經常執行的任務類型分別製定適性化原則。(2)當搜尋任務為「與日期有關之主題搜尋」類別，使用階層線性混合式網頁會有較少的瀏覽時間與較少的瀏覽頁數。(3)當搜尋任務為「事實搜尋」類別，可以選擇階層線性混合式架構，在瀏覽頁數上會有較佳的表現。當搜尋任務為「事實搜尋」類別的任務，場地依賴者(field dependent (FD))比場地獨立者(field independent (FI))較常使用網頁頁面所給予的連結按鈕，但FD(field dependent)受測者卻會花較多的搜尋時間。本研究亦發現深度或廣度的瀏覽行為與受測者的認知風格沒有很大的關聯性存在。而因為階層式架構會使受測者偏向於深度的瀏覽行為，且階層線性混合式架構可滿足習慣深度及廣度搜尋的使用者在不同行為上之需求，故深度或廣度的瀏覽行為是與連結架構有相關的。整體來說，就設計資訊網頁而言，階層線性混合式架構是比較適合的連結架構。

關鍵字

瀏覽行為、連結架構、使用者認知風格、適性化網頁

並列摘要

In order to satisfy with the demand from different kinds of customers, some websites can provide the real time and adaptive web pages when the users are browsing the web pages. In the design factors of devising information web pages, the architecture of web pages is the important decisive factor of browsing efficiency. Besides, the cognitive styles of different users may also affect the behaviors of web pages browsing and then affect the results of browsing. The purpose of this study is to study the effect of “the link architecture” and “the type of user’s cognitive style” upon the behavior of web pages browsing in the design of information web pages, and further to establish the design rules of the architecture of information web pages. First, the subjects were assigned to take the Group Embedded Figure Test (GEFT). After determining their cognitive styles, i.e. Field Dependent (FD) or Field Independent (FI), they then participated in the experiments of hierarchical and mixed (hierarchical with linear) web site architectures on the computer in the second phase. There were five assignments which are based on the common questions users would like to search in the web site of design information service. After these experiments, the researcher used Analysis of Variance to determine whether two web pages and two cognitive styles have significant effects on the performance of subjects or on subjective evaluation while operating web pages. This study also focused on the browsing behavior that was used to conduct Cluster Analysis to determine the effects of two web pages and two cognitive styles.
The results show that when the searching assignments in an information web site are most of “topic searching concerned with date,” using mixed (hierarchical with linear) web pages yield less browsing time and less browsing pages. If most of the searching assignments are “fact searching,” the designer can use mixed architecture for better performance in browsing pages. When the searching assignment is “fact searching,” FD subjects are more frequently use the link buttons on the web pages. However, FD subjects spend more time in searching. This study also finds breadth or depth browsing behavior does not relate to the subject’s cognitive style. Besides, hierarchical architecture deflects subject to depth browsing behavior, and hierarchical with linear architecture can satisfy customers who with different behaviors, i.e., depth or breadth searching. Therefore, the depth or breadth browsing behavior relate to link architecture. Overall, hierarchical with linear link architecture is more appropriate for web pages for design information services.

並列關鍵字

adaptive web pages ； Behavior of web-page browsing ； user’s cognitive style

參考文獻

4. 黃星峰，2003，運用資料探勘建立個人適性化學習網站，大同大學資訊經營研究所碩士論文。

1. Ayersman, D. J.and Minden, A. V., 1995, Individual differences, computers, and instruction. Computers in Human Behavior, Vol. 11, No. 3-4, pp. 371–390.

3. Brusilovsky, P., 1996, Methods and techniques of adaptive hypermedia. User Modeling and User-adapted Interaction, Vol. 6, pp. 87–129.

5. Carlson, J. R. and Kacmar, C. J., 1999, Increasing link marker effectiveness for WWW and other hypermedia interfaces: an examination of end-user preferences, Journal of the American Society of Information Science, Vol. 50, No. 5, pp. 386-398.

6. Cockburn A. and McKenzie, B., 2001, “What do web users do? An empirical analysis of web use,” International Journal of Human-Computer Studies, Vol. 54, pp. 903-922.