Drexel University Home Pagewww.drexel.edu DREXEL UNIVERSITY LIBRARIES HOMEPAGE >>

iDEA: Drexel E-repository and Archives > Drexel Theses and Dissertations > Drexel Theses and Dissertations > How information visualization systems change users’ understandings of complex data

Please use this identifier to cite or link to this item: http://hdl.handle.net/1860/3159

Title: How information visualization systems change users’ understandings of complex data
Authors: Allendoerfer, Kenneth Robert
Keywords: Computer science;Information visualization;Human-computer interaction
Issue Date: 5-Jan-2010
Abstract: User-centered evaluations of information systems often focus on the usability of the system rather its usefulness. This study examined how a using an interactive knowledge-domain visualization (KDV) system affected users’ understanding of a domain. Interactive KDVs allow users to create graphical representations of domains that depict important papers, authors, or terms. Interactive KDVs have several potential advantages over other presentation methods, such as making connections explicit, and the ability for users to see the overall structure of the domain. The project examined CiteSpace, an interactive KDV that uses article cocitation analysis and text analysis to create visualizations of important papers and terms in a domain. In this study, participants completed several tasks related to the field of artificial intelligence. Depending on the experimental condition, participants read a review article about the domain, interacted with a CiteSpace visualization containing equivalent information, or both. Participants who neither read the article nor used the KDV system served as a baseline. The participants completed three tasks in which they sorted papers and terms according to their importance and relatedness. The hypotheses predicted that participants who used the KDV, especially in conjunction with the review article, would show a more expert-like understanding of the domain compared to the baseline and to participants who used only the article. The study measured the quality of participants’ understanding by comparing their card sorting responses to benchmark responses obtained from domain experts. Participants who produced judgments of importance and relatedness that were similar to the benchmarks were considered as demonstrating a good understanding of the domain. The card sorting results were analyzed using several statistical techniques, including multidimensional scaling and cluster analysis. The results showed that while participants’ understanding of the domain was influenced by using the KDV, this influence was not in the direction of the benchmarks. The data suggest that a lack of agreement between the benchmarks and the depiction of the field presented in the KDV may have led to these findings. The study discusses several possible reasons for these results and recommends possible changes to KDVs that may increase their usefulness.
URI: http://hdl.handle.net/1860/3159
Appears in Collections:Drexel Theses and Dissertations

Files in This Item:

File Description SizeFormat
Allendoerfer_Kenneth.pdf1.87 MBAdobe PDFView/Open
View Statistics

Items in iDEA are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! iDEA Software Copyright © 2002-2010  Duraspace - Feedback