Dissertation Information for Charlie D. Hurt III
- Charlie D. Hurt III
- (Alias) Charlie D. Hurt
- (Alias) Charles D. Hurt
- Library and Information Science
- University of Wisconsin, Madison (USA) (1981)
- Richard D. Walker
- Jack A. Clarke
- Charles A. Bunge
- Warren Olaf Hagstrom
- Daniel Menahem Siegel
MPACT Status: Fully Complete
Title: IDENTIFICATION OF IMPORTANT LITERATURE IN QUANTUM MECHANICS: INVESTIGATION OF A BIBLIOMETRIC AND AN HISTORICAL APPROACH
Abstract: This study examined the problem of identification of important literature in a specific scientific area, quantum mechanics. An examination was conducted on literature identified as important using two means of selection. Four histories of quantum mechanics were examined and the references each author used to support his historical analysis were analyzed. The second means of selecting important literature was a bibliometric analysis, using citation frequency, of literature spanning the years prior to 1937. For each piece of literature cited, the references each contained were analyzed. Both the Historical and the Bibliometric files were ranked by frequency.
The hypothesis tested in this study was that of no association between pairs of ranks in the two files. The statistical test employed was a gamma test of association. Under decision rules preset before the analysis, the gamma value observed, -.11472, was not significant and was interpreted as lack of association between the rankings of the two files of literature. Lack of association implies statistical independence.
Specific characteristics of the literature were examined for predictive ability. In all cases, the particular attributes examined showed no symmetric predictive ability.
The two files were partitioned into Old and New Quantum Theory periods and the same statistical procedures employed to test for association in the composite files were used again. Neither partitioned file showed significant association. Ceiling and floor effects were controlled for by restricting both files to the years 1900 through 1935. No significant association was found for the restricted files using a gamma statistic.
A large number of items were contained in one file and not in the other file. To determine if a statistical artifact had been produced, a test of association was initiated using only those items common to both files. Statistical association was not found for the common items using a gamma test of association.
The results of the analyses led to the following conclusions. (1) Association between the rankings of cited items using a citation analysis approach and the ranking of cited items using an approach which pools the references used by historians in their reconstruction of the period is statistically nonsignificant. From no association, statistical independence may be inferred. (2) Association between the attributes chosen to investigate particular areas of potential difference in the two literature distributions is nonsignificant and supports the conclusion above. (3) The use of citation analysis alone or historical analysis alone will not result in the same set of literature being produced than would result if both methods were used. Using citation analysis to predict important literature in a scientific area appears to be risky since another method which would obstensively produce the same literature of importance did not.
Further research into the possibility of combined approaches to the problem of identification of important literature is indicated by this study. Additionally, validation of the present study in the area of quantum mechanics and in other scientific areas is suggested.
MPACT Scores for Charlie D. Hurt III
A = 3
C = 1
A+C = 4
T = 3
G = 1
W = 3
TD = 3
TA = 0
calculated 2008-01-31 06:27:09
Advisors and Advisees Graph