diversity measures

Annex to:

Within-Journal Self-citations and the Pinski-Narin Influence Weights

Gangan Prathap^a and Loet Leydesdorff^b

^aA P J Abdul Kalam Technological University, Thiruvananthapuram, Kerala, India 695016; e-mail: gangan_prathap@hotmail.com

^bAmsterdam School of Communication Research (ASCoR), University of Amsterdam, PO Box 15793, 1001 NG Amsterdam, The Netherlands; loet@leydesdorff.net

The routine on this page (available at http://www.leydesdorff.net/iw/vector.exe ) enables the user to compute influence weights as defined by Pinski & Narin (1976). The argument is developed in the article.

Abstract:

The Journal Impact Factor (JIF) is almost linearly sensitive to self-citations because each self-citation adds to the numerator, whereas the denominator is not affected. Pinski & Narin (1976) derived the Influence Weight (IW) as an alternative to Garfield’s JIF. Are IWs insensitive to self-citations? This is a very desirable property of a measure of quality or impact. Whereas the JIF is based on raw citation counts normalized by the number of publications, IWs are based on the eigenvectors in the matrix of aggregated journal-journal citations without a reference to size: the cited and citing sides are combined by a matrix approach which implies a recursively iteration of the multiplication by itself. IW emerges as a vector after recursive iteration of the normalized matrix. Before recursion, IW is also a (non-network) indicator of impact, but after recursion (i.e. repeated improvement by iteration), IWs are a network measure of prestige among the journals in the (sub)graph as a representation of the field. We illustrate the concepts using datasets from various journal ecosystems.

1. The program vector.exe computes Influence Weight on the basis of a 1-mode matrix. The user is prompted for the size. There are no data limitations.

Input has to be a file named "text.csv" (for example, text.csv from Price (1981, p. 59).)
that is, comma-separated variables in ASCII format.
This file should contain only the matrix
(without a header at the first line).
Each record is closed with CR + LF (CSV MS-DOS in Excel).

Output is found in the file "vectors.dbf".

2. The program power.exe can be used with the same input (“text.csv”). It generates power2.dbf, power3.dbf, …, power15.dbf containing the powers of the original matrix M^1, M^2, M^2, etc.

3. Influence weights: one can first run vector.exe in order to obtain the normalized matrix in the file narin1.dbf (using Pinski & Narin’s normalization) and the file vectors.dbf with the recursive vectors p(k). In a second step, Narin.dbf (use Excel, SPSS, or Open Office for opening) can be saved as text.csv and be input into power.exe .

Amsterdam, 28 April 2019.

References:

Prathap, G., & Leydesdorff, L. (2019). Within-Journal Self-citations and the Pinski-Narin Influence Weights. Scientometrics, under submission.

Pinski, G., & Narin, F. (1976). Citation Influence for Journal Aggregates of Scientific Publications: Theory, with Application to the Literature of Physics. Information Processing and Management, 12(5), 297-312.

Price, D. J. de Solla (1981). The Analysis of Square Matrices of Scientometric Transactions. Scientometrics, 3(1), 55-63.