Johan Bollen, Marko A. Rodriguez, and Herbert Van de Sompel "Journal Status" arXiv:cs.GL/0601030 v1 9 Jan 2006

[Loet Leydesdorff]

Dear Frank:

Nancy Geller (1974) elegantly formulates the problem which is also valid for
IFs:

"We also observe that the Markov Chain context explains why caution is
necessary in comparing influence weights which arise from separate
aggregates of journals (etc). Comparing one physics journal that has
influence weight four among physics journals to another physics journal
which has influence weight three among the same physics journals is not the
same as comparing a physics journal which has influence weight four among
physics journals with a chemistry journal which has influence weight three
among chemistry journals. One explanation of this is that an influence
weight is a function of the long-run probability of being cited within the
aggregate itself. Thus a journal's influence weight depends on which
journals it is aggregated with, and different aggregations will ordinarily
lead to different influence weights for the same journal."

In other words, the zero's in the citation matrix make the system nearly
decomposable. This is reinforced by ISI's decision to lump all ones under
the category "All others". This makes the matrix even more sparse. Thus, one
first needs the decomposition before one can make comparisons.

With best wishes,


Loet

________________________________
Loet Leydesdorff
Amsterdam School of Communications Research (ASCoR),
Kloveniersburgwal 48, 1012 CX Amsterdam.
Tel.: +31-20- 525 6598; fax: +31-20- 525 3681;
loet at leydesdorff.net ; http://www.leydesdorff.net/



> -----Original Message-----
> From: ASIS&T Special Interest Group on Metrics
> [mailto:SIGMETRICS at LISTSERV.UTK.EDU] On Behalf Of Frank Havemann
> Sent: Wednesday, March 08, 2006 4:31 PM
> To: SIGMETRICS at LISTSERV.UTK.EDU
> Subject: Re: [SIGMETRICS] Johan Bollen, Marko A. Rodriguez,
> and Herbert Van de Sompel "Journal Status"
> arXiv:cs.GL/0601030 v1 9 Jan 2006
>
> Adminstrative info for SIGMETRICS (for example unsubscribe):
> http://web.utk.edu/~gwhitney/sigmetrics.html
>
> The main problem I have with this paper is:
>
> The weighted "PageRank" algorithm was proposed already in
> 1978 for citation networks of journals by Nancy L. Geller (see below).
>
> The difference of Geller and PageRank is only that in the
> case of journals normally a damping factor is not needed (lambda = 1).
>
> In the paper by Bollen et al. I have not found any hint for
> which reason they need damping and how different damping
> factors influence their results.
>
> @Article{Geller1978citation,
> author = {Nancy L. Geller},
> title = {{On the citation influence methodology of Pinski and
> Narin}}, journal = {Information Processing \& Management},
> year = {1978}, volume = {14}, number = {2}, pages = {93--95}, }
>
> Frank Havemann
>
>
> ***************************
> Dr. Frank Havemann
> Department of Library and Information Science Humboldt
> University Dorotheenstr. 26
> D-10099 Berlin
> Germany
>
> tel.:  (0049) (030) 2093 4228
> http://www.ib.hu-berlin.de/inf/havemann.html
>