In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • RefViz 2
  • Marilyn Domas White
RefViz 2Philadelphia: Thomson ISI ResearchSoft, 2003–2005. $299.95 (System requirements: Microsoft Windows 2000/ XP/Vista, 256MB RAM, 1 GB free disk space; Mac OS X (10.3 or later), Power Mac G5, JAVA 1.4.2 or later, 256MB RAM, 1 GB free disk space)

Any researcher who has faced the daunting task of perusing over one thousand bibliographic records retrieved in response to a search query will be thrilled with the capabilities of RefViz for saving time and effort. RefViz is a textual analysis and visualization package intended to be used with bibliographic records and moves beyond normal searching software to allow the searcher to analyze search results and identify major themes and patterns within a body of literature. It is particularly useful for working with large bodies of literature, that is, those with one thousand or more items. A brief description of the steps usually followed is noted to illustrate its functions:

  • • The data for analysis consist usually of bibliographic records with abstracts. These may be input from an existing personal database or from a search in one or more commercial databases. RefViz’s searching component, Reference Retriever, allows for searching multiple databases simultaneously and removing duplicate records. RefViz also allows for importing text files or files from EndNote, ProCite, and Reference Manager, all Thomson Reuters bibliographic management packages.

  • • Using computational linguistic analysis, RefViz clusters the data, disaggregating the larger set of records into more specific reference groups that are similar and showing the proximity of these topic-related [End Page 343] clusters in a network representing the broad topic. The researcher can see the analysis in two views: the Galaxy View visually shows the clusters as nodes in a network, and the Matrix View identifies the major concepts and shows those discussed together.

  • • Besides providing the global picture, RefViz characterizes each individual cluster in several ways. A mouse-over of an individual cluster briefly identifies the number of references in the cluster and the “top topics” in the cluster. In a is also its greatest limita separate window, RefViz shows the bibliographic citations for the items in that cluster (the researcher can specify the bibliographic data items to be shown). In yet another window, it lists the major and minor topics (with number of hits) used to group the references and the other descriptive terms in the cluster.

  • • The analysis can be sent to others, which is especially useful for working with collaborators, and the bibliographic items can be output to bibliographic management packages or as text.

The program is highly flexible. It does not limit input to items searched via RefViz’ Reference Retriever but accepts any type of bibliographic item in a wide range of formats, including the RIS export format available at many academic libraries. In user-generated databases, the notes field can include any type of text, not just an abstract. For example, for Library of Congress records, which usually have no abstract, the researcher can substitute tables of contents. The researcher can modify the initial analysis by promoting or demoting terms or eliminating some completely. The researcher can also customize the thesaurus to input equivalent terms not immediately generated, for example, acronyms for organizational names. RefViz allows for focusing on the overall set of references or on the individual reference clusters. In Reference Retriever, although RefViz provides configurations for some commonly used databases and vendors, it also allows the researcher to configure additional online sources and makes additional configurations available at the RefViz Web site as they become available.

RefViz is an effective tool for secondary analysis of search results. Perhaps its greatest benefitis also its greatest limitation in that a researcher has to be able to think visually to optimize its use. Understanding the Galaxy View, for example, calls for being sensitive to density, proximity of reference groups, blank spaces between groups, the overall shape of the network, and outliers. Initially, this different way of thinking in relation to literature may require considerable effort, but the researcher can work interactively with the analysis to understand the implications of different modifications of information on the output. RefViz also provides good online help and a...


Additional Information

Print ISSN
pp. 343-344
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.