Dissertation Information for Marilyn Slater
- Marilyn Slater
- Library and Information Science
- University of California, Los Angeles (USA) (1988)
- Harold Borko
- Marcia J. Bates
- William H. Fisher
- Arthur Charles Gibson
- Eric Williamson Holman
MPACT Status: Fully Complete
Title: Bibliographic coupling among scientific papers in biological research specialties
Abstract: Bibliographic coupling strength among scientific papers was studied as an information variable and evaluated as an indicator of specialties in biology. Three specialties were identified from corresponding review articles and were characterized along four facets: systems, conditions, observations, and methods. Approximately 110 papers were collected for each specialty. Papers in each sample were assigned a degree of membership in the specialty depending upon facet representation by the papers.
Specialty samples were used to evaluate two bibliographic coupling measures, author and reference coupling, with regard to (1) the relation of coupling strength between papers to degree of membership in a specialty, (2) the distribution of coupling strengths among pairs of papers, and (3) the power of each measure to resolve a mixed sample into constituent specialties by cluster analysis. Reference coupling is a traditional bibliographic coupling measure and has been studied as a subject indicator. It is the percent overlap of references between two papers. Author coupling, a new bibliographic coupling measure, was defined as the percentage of author names common to two reference lists. Reference lists of sample papers were used to construct two types of data files (author and reference) for subsequent bibliographic coupling and cluster analyses.
How well each measure (author and reference coupling) indicated degree of specialty membership was calculated by averaging coupling strengths among papers that had been segregated into various degrees of specialty membership.
Members of specialties generally exhibited higher bibliographic coupling strengths than did non-members; however, the variability in coupling strengths was high for all degrees of membership. Standard deviations exceeded average coupling strengths in all cases.
The distribution of coupling strengths among pairs of papers followed a power law, similar to the laws of Lotka and Zipf. Author and reference coupling performed well as similarity measures in cluster analyses when combined with group average hierarchical agglomerative clustering. Resulting cluster patterns showed good resolution of specialties, but author coupling performed somewhat better than reference coupling as a similarity measure.
MPACT Scores for Marilyn Slater
A = 0
C = 0
A+C = 0
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2008-01-31 06:14:35
Advisors and Advisees Graph