|
Journal of the American Society for Information Science (JASIS) -- Table of Contents
Contributed by
Richard Hill
American Society for Information Science
Silver Spring, Maryland, USA
[email protected]
VOLUME 51, NUMBER 11
[Note: This table of Contents includes a new feature.
You will note that entries include a new item, "Published online 6 July
2000" after the page number. This refers to Wiley's "Early View" feature,
posting articles as soon as they are approved. ASIS members who wish
electronic access but didn't so elect can contact <[email protected]>.]
CONTENTS
In this issue
Bert R. Boyce
Page 969
Research
- A Usability Assessment of Online Indexing Structures in the Networked
Environment
Carol A. Hert, Elin K. Jacob, and Patrick Dawson
Page 971. Published online 6 July 2000
Hert et al. used metrics developed from Liddy and Jorgensen's back of the
book index work to compare variations of the FedStats web site index and to
attempt to gain an understanding of user's difficulties with such indexes.
For comparison a similar index with HTML title information added to each
index term was used (group 2), as were two newly created indices, one where
an indexer added see references to the original A-Z index for the pages
identified as relevant (group 3), and a second where synonymous terms were
added as entry points (group 4). Twenty students at Indiana University were
video taped as they used the index structures and asked to think aloud so
that they could be recorded. Five were assigned to each index variation. A
demographic survey preceded the searching and a satisfaction questionnaire
and a semi-structured interview followed. Five search tasks were given to
the subjects to perform in varying orders. Number of clicks in each index
was collected as was number of terms selected for a search, and time spent
in the index. Success was marked by the researchers when pages with
appropriate information were identified on a 0 to 1 scale. Subjects also
rated their perception of success on a 1 to 5 scale, and their reactions to
the system on a scale of one to nine. In terms of clicks, terms used, and
time in index, the Group 4 index was strikingly superior although no
statistical analysis is done. Group 4 was also most successful in
researcher evaluated success and in user perceived success although no
level of significance is provided. Group four also has a numeric lead
position in satisfaction measures. This would appear to once again indicate
that increased depth of indexing increases performance and satisfaction.
- Interactive Query Expansion: A User-Based Evaluation in a Relevance
Feedback Environment
Efthimis N. Efthimiadis
Page 989. Published online 6 July 2000
Using questionnaires, search logs, and evaluated printouts of results of
25 searches by students and faculty Efthimiadis investigates interactive
query expansion in a user centered environment. Users made topical
relevance judgements on a relevant, partially not relevant (not included as
relevant), partially relevant (included as relevant), and not relevant
scale. Precision and number of relevant and retrieved were recorded.
After the initial search was run, and the on-line results judged, terms
for expansion were extracted from the top 5 relevant documents and
presented as a ranked list. Users evaluated the list choosing good terms,
the top 5 of which were used for expansion. After a research, questionnaire
data was collected and offline printouts evaluated. Reasons for term
selection were solicited. Terms were chosen 88% of the time that were
thought to be related in some way, and in 64% of the time as actual
synonyms. New ideas not related to the original query were chosen 44% of
the time. About 75% of term associations fall within a hierarchical
relationship. Single posted terms were chosen by users in 42%of the
searches although they had come from an already retrieved document record.
About one third of user chosen terms are potentially useful. On the average
expansion produced nine highly relevant documents beyond the three highly
relevant ones found in the initial search. Precision increased slightly.
- Aging, Obsolescence, Impact, Growth, and Utilization: Definitions and
Relations
Leo Egghe and Ronald Rousseau
Page 1004. Published online 6 July 2000
Egghe and Rousseau clarify the relationships among several traditional
bibliometric concepts. Growth can mean production in a year, cumulative
production, or the increase in production between time periods. Aging
implies that citation rates change over time. Since impact is citation
count relative to number of sources, impact can be viewed as relative
aging. Growth and aging are modeled in a similar fashion but growth
influences aging in that more articles provide more sources and competition
becomes heavier to appear in the reference lists of new articles. Growth,
however, does not cause aging.
- Network Organizational Development in the Public Sector: A Case Study of
the Federal Emergency Management Administration (FEMA)
Robert Ward, Gary Wamsley, Aaron Schroeder, and David B. Robins
Page 1018. Published online 6 July 2000
Ward et al. examine the view that Information Technology is an agent for
change within the organizational hierarchy that will lead to more effective
and responsive administration. Existing organizational theory research
shows that management uses technology to preserve existing control and
centralization. An examination of the Federal Emergency Management
Administration's (FEMA) experience in adopting an information network
reinforces this view. The National Preparedness Directorate, one of the
five major areas of FEMA, did not support the other areas which were left
without technology guidance or integration. The NPD produced a very
advanced telecommunications network for disaster response which was not
available for dealing with civilian disasters. The hurricane Hugo and
Andrew disasters and the Loma Prieta earthquake swamped the civilian paper
based systems and led to sever criticism. Clearly the IT effort was used by
upper management to maintain their own objectives. More than sophisticated
IT is necessary to provide a more effective and responsive administration.
Later the Information Technology Services Directorate was removed from NPD,
centralized, and new models and geographical information systems were
designed under contract. By 1998 a real time system was in place
integrating available data with direct feed from the disaster area and
providing the location of food supplies and shelter and contact with
supporting agencies. Clearly management change had driven the use of
technology, not the technology by its own existence.
- Using Kintsch's Discourse Comprehension Theory to Model the User's Coding
of an Informative Message from an Enabling Information Retrieval System
Charles Cole and Bertie Mandelblatt
Page 1033. Published online 11 July 2000
Cole and Mandelblatt model the retrieval interaction as a Shannon like
process where the integration function of the destination is considered to
involve perception and comprehension understood in terms of Kintsch's
four-layer propositional-language theory. To stimulate the comprehension
process a form requires the subject to list four questions about her topic,
answer these questions, label four concepts for each question answer pair,
and then supply a weight indicating the probability the this alternative
will be the final thesis used. It is claimed that this form's use brings to
the fore the user's tacit knowledge and enables the user to create a real
thesis for the essay to be produced.
- Translingual Alteration of Conceptual Information in Medical Translation: A
Crosslanguage Analysis between English and Chinese
Shaoyi He
Page 1047. Published online 11 July 2000
Using 200 English to Chinese title translations from JAMA and Archives of
Ophthalmology, and 200 Chinese to English title translations from the
National Medical Journal of China and the Chinese Journal of Ophthalmology,
Shaoyi He had two bilingual MDs identify concept pairs and unlinked
concepts, generating a list of concept pairs. Differences in the pairs were
noted and rated on a scale of 1 (identical) to 5 (quite different). Cohen's
Kappa indicated a strong reliability of inter-judge agreement. 3.66
concepts per title were identified in English and 3.54 in Chinese.
Conceptual addition as well as conceptual omission takes place in both
directions but addition is more frequent. Conceptual alteration is almost
twice as common in Chinese to English as in English to Chinese, and in both
cases general rather than medical concepts seem to be those more often
altered.
Book Reviews
- New Organizational Designs: Information Aspects,, by Bob Travica
Patricia F. Katopol
- Information Visualization: Perception for Design,
by
Colin Ware
Terrence A. Brooks
- Information Retrieval: Algorithms and Heuristics, by
David A. Grossman and
Ophir Frieder
Hugh E. Williams
- The Internet Public Library Handbook, by Joseph Janes, et al.
Billie E. Walker
|
