應用卡片分類法與相似性評定於大學圖書館網站資訊架構之研究

A Study on Applying Card Sorting and Similarity Ratings to Information Architecture of University Library Websites

廖婉竹(Wan-Chu Liao)

指導教授 ： 謝建成

國立臺灣師範大學/教育學院/圖書資訊學研究所/碩士(2013年)

摘要

本研究主要探討將知識結構評量常用的引出方法應用於圖書館網站之資訊架構研究中，即分別使用知識結構引出方法中的相似性評定與傳統資訊架構研究者使用的卡片分類法，比較兩種方法所得之網站架構圖，對圖書館網站內部尋獲度之影響，並以國立臺灣師範大學圖書館作為研究個案。第一階段分別以卡片分類法與相似性評定作為知識結構引出方法，第二階段將兩種知識結構引出方法蒐集得到之資料分別以群集分析及徑路搜尋分析兩種知識結構表徵方法建立網站架構圖，以此兩階段不同的方法組合，期望找到可協助網站架構建置最佳的方法。
研究結果顯示第一階段兩種知識結構引出方法，對於網站尋獲度並無顯著差異，而第二階段兩種知識結構表徵方法則有顯著差異，特別是在同樣採用卡片分類法作為知識結構引出方法時，徑路搜尋分析會比群集分析獲得較好的網站尋獲度。
最後根據研究結果對現行之臺師大圖書館網站提出三點建議：一、設計符合使用者期待之網站資訊架構；二、考量使用者的分類概念設計網站資訊架構；三、從使用者的角度命名網站標籤。

關鍵字

卡片分類法 ； 相似性評定 ； 群集分析 ； 徑路搜尋分析 ； 尋獲度 ； 知識結構

並列摘要

Using the website of National Taiwan Normal University Library as a case study, which applied knowledge structure elicitation methods to information architecture, the purpose of this thesis explored effects on internal findability of websites. Specifically, the researcher compared the results from the two of the knowledge elicitation structure methods, similarity ratings and card sorting, the latter which is used by traditional information architects, then comparing the data of the website architecture diagrams generated from these two methods.
In the first phase, the researcher compared the similarity ratings with card sorting, as knowledge elicitation structure methods to collect data separately. As knowledge structure representation methods in the second phase, cluster analysis and pathfinder analysis were used to construct website architecture diagrams. Combining the two methods in order to find the most effective way facilitates the setup of websites.
The results revealed that there was no significant difference between the knowledge elicitation structure methods in terms of the internal findability in the first phase. However, there were significant differences between the two knowledge structure representation methods. Specifically under the condition of using card sorting as knowledge elicitation structure method, using pathfinder analysis as knowledge structure representation method retrieved better results of website internal findability. Finally, the researcher provided suggestions to the website of National Taiwan Normal University Library based on the experiment results: 1. Development the website information architecture based on users’ expectations. 2. Considerate the users’ categorization schema to design the website. 3. Naming the labels on the website from the users’ perspective.

並列關鍵字

Card Sorting ； Similarity Ratings ； Cluster Analysis ； Pathfinder Analysis ； Findability ； Knowledge Structure

參考文獻

江淑卿、郭生玉（1997）。不同學習過程的概念構圖策略對促進知識結構專家化與理解能力之效果研究。師大學報：教育類，42 期，1-16 頁。

宋德忠、林世華、陳淑芬、張國恩（民87）：知識結構的測量－徑路搜尋法與概念構圖法的比較。教育心理學報，30（2），123-142。

謝建成、丁依玲、陳慧倫（2011）。大學圖書館網站資訊尋獲度之研究。資訊管理學報，18（3），25-49。

Brophy, P. & Craven, J. (2007). Web accessibility. Library Trends, 55(4), 950-973.

Capra, M. G. (2005). Factor analysis of card sort data: an alternative to hierarchical cluster analysis. Paper presented at the Proceedings of the Human Factors and Ergonomics Society 49th Annual Meeting, Orlando, Florida, USA.