Login

Publications  •  Project Statistics

Glossary  •  Schools  •  Disciplines
People Search: 
   
Title/Abstract Search: 

Dissertation Information for Michael W. Grunberger

NAME:
- Michael W. Grunberger

DEGREE:
- Ph.D.

DISCIPLINE:
- Library and Information Science

SCHOOL:
- Rutgers University (USA) (1985)

ADVISORS:
- James D. Anderson

COMMITTEE MEMBERS:
- Henry Voos
- John James Regazzi III
- Daniel O'Connor

MPACT Status: Fully Complete

Title: Textual analysis and the assignment of index entries for social science and humanities monographs

Abstract: This dissertation examines, indirectly, the process by which an indexer selects an index entry by focusing upon the results of indexing: the index entry and its relation to indexed text. A sample of index entries from social science and humanities monographs is investigated in order to examine the assumptions which support the naturalist and the formalist theories of indexing. In an effort to discover relationships between the location and frequency of textual references and the generation of corresponding index entries, two related hypotheses are tested: (1) that term location within paragraphs will correlate with indexability, and (2) that term frequency on cited pages will correlate with indexability. Based upon the literature, it was expected that there would be a direct relationship among index entries and term location and frequency.

The data, after analysis by descriptive and inferential methods, did not support either of the two hypotheses. The researcher concluded that: (1) Indexed terms are evenly distributed within the paragraph of social science and humanities monographs and do not significantly cluster in discrete segments of paragraphs, (2) Not-indexed terms are evenly distributed within the paragraph structure of social science and humanities monographs and do not cluster in a significant manner in a discrete segment of a paragraph, (3) There is a statistically significant, but trivial, difference between the location of indexed terms in posted locations and indexed terms in not-posted locations, (4) Indexed term frequencies on posted pages are concentrated in the low and medium frequency ranges, (5) Indexed term frequencies on non-posted pages are concentrated in the low frequency range, and (6) There is a statistically significant, but trivial, difference between the frequency of indexed terms on posted pages and their frequency on not-posted pages.

These conclusions suggest that the assumptions of the naturalist approach to indexing, i.e., that term location and frequency are related to indexability, may not apply to full-text index units. It is suggested that future research in indexing take into account the findings of computational linguistics, artificial intelligence, and cognitive theory in order to better understand the complex process by which a human indexer determines indexability.

MPACT Scores for Michael W. Grunberger

A = 0
C = 0
A+C = 0
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2008-01-31 06:09:42

Advisors and Advisees Graph

generating graph, please reload

Students under Michael W. Grunberger

ADVISEES:
- None

COMMITTEESHIPS:
- None