In the July issue of C&RL, Thomas Nisonger & Charles Davis have a paper titled “The Perception of Library and Information Science Journals by LIS Education Deans and ARL Library Directors: A Replication of the Kohl-Davis Study.” This is, well, a replication of the 1985 Kohl-Davis study, which ranked LIS journals based on the perceptions of LIS school deans and research library directors. Mostly because I was just curious (and because I’m hung up on IF lately — must be my upcoming 3rd year review), I decided to compare Nisonger & Davis’ findings with the latest IF rankings. Conveniently, both are based on 2003 data. So I created this table, comparing the latest IFs (which you can download from Web of Science) & the K-D results (but only the data from deans, since that’s what I personally care about). I only have the top 20 rankings here, though IF goes to 55 & K-D goes to 71.

Rank IF K-D
2 MIS Quarterly
3 J Amer Medical Informatics Ass’n ARIST / LISR
4 Info Syst Res
5 Info Manag – Amsterdam JDoc
6 JDoc Library Trends
8 C&RL
9 Scientometrics ASIST Proceedings
10 J Manage Info Syst RUSQ
11 IP&M C&RL
12 J Info Sci JELIS
13 Int J Geogr Inf Sci Libraries & Culture
14 J Health Commun J of the Medical Library Association
15 Lib Resources & Tech Svcs Lib Resources & Tech Svcs
16 Gov Info Q School Library Media Research
17 Int J Info Manage J Info Sci / Libri
18 Telecommun. Policy
19 Libr Inform Sci J Amer Medical Informatics Ass’n
20 LISR School Library Journal

Interesting that in the top 10, only 3 appear on both lists, and only 9 appear in both top 20s.

Bollen et al. have an article in the December 2005 issue of IP&M (don’t even ask me how that’s out yet) titled “Toward alternative metrics of journal impact: A comparison of download and citation data.” In it they propose this taxonomy of impact measures:

Bollen et al 2005 Fig 5

The first dimension corresponds to whether an impact measure is based on frequency-based metrics as opposed to structural metrics. The second dimension concerns whether a measure is based on author- or reader-defined data sets.

Clearly the K-D study measures reader-defined ranking. But it seems to me that the X-axis really needs 3 points (thus making it a plane, not an axis, yes, I know): frequency-based, structural, & perceptual metrics.