Content area
Full Text
In January 2012, I was attending a library workshop, and trying to set up a Researcher ID profile with data from the Web of Science (WoS), ISI's academic-indexing service. In order to complete my list of publications, I looked for a paper I had published in a Political Science journal (Varela, 2009), without much success, despite the fact that I was sure the journal was fully indexed by ISI. As I could not find my paper, however hard I tried, I decided to list all the papers published by that journal. To my surprise, I found out that none of the papers in the issue I was looking for appeared on the list, whereas the papers in previous and later issues did. I contacted the journal and they in turn contacted the publisher, who promised to research the problem.
I immediately started to think about the potential consequences of a paper having been neglected by this index for nearly 3 years immediately after its publication. On the one hand, I thought that the effect might have been minimal, as there are various other means for researchers to access the journal's contents. This journal in question is also indexed by Scopus and Google Scholar, and has a relatively high self-citation rate of around 25 per cent (compared with 5 per cent for the American Political Science Review ), which may be an indication of a loyal readership that could access the journal directly without the intermediation of indexing and search services. On the other hand, ISI has a reputation for being the leading provider of academic-indexing services, and the citations received in the 2 years immediately after the publication of a paper are considered particularly relevant in the field of bibliometrics, and in fact used for calculating the most common journal impact factor.
I realized that it would not be easy to separate the effect of ISI indexing from other factors that may contribute to the number of citations received by a given paper. On the other hand, the fact that there were several papers in the omitted issue made me think about the possibility of finding some statistically significant differences with the rest of the sample. I was really curious about the potential results of...