When searching authors in Web of Science there are often authors with the same surnames and the same first name initials. Sometimes it is suggested to limit your author search to address to get the right person. But that kind of refined search has its limitations mostly due to the implementations in the database. You also must be aware that the person you search for maybe has been publishing articles when working at different universities or even different departments at the same university.

This is an example of the address fields in a Web of Science record:

The first address – with reprint author in brackets – is connected to the first author of the article and reprint refers to author from where you can order a copy of the original article. Nowadays, with lots of electronic articles online, the practice of requesting articles directly from the author is not as widespread as it used to be. Reprint author is also the corresponding author and e-mail address following the addresses is the e-mail of the corresponding author.

This first reprint address is not a part of the address field, though it visually seems to. Sometimes this address could be an administrative address. If the reprint address is the same as one of the author addresses it WILL be duplicated. Otherwise addresses will not be duplicated, except if it’s another department/institution at the same university. Then it will be duplicated.

Because this reluctance to duplicate, the addresses in descending order of the addresses does not correspond to the order of the authors, which makes it difficult to visually determine the residence of each author. For search retrieval purposes it would not be a problem if the implementation would be consistent, which it’s not. Examples follow below:

In this case the first address following reprint author address is Internal Medicine where Asplund K worked at, but Norberg A working at Nursing dep and Eriksson S Ger Dep is not mentioned even if Web of Science says they duplicate addresses if it’s same university but other departments/institutions.
Here’s another example:

Just reprint author address is mentioned but the other authors are they working at the same department as Bjermer L? In this case no regular address field is implemented, just reprint author address and as we already mentioned this is not a regular address field. When the address is not duplicated it’s hard to tell if it’s Eriksson S at same department as Bjermer or Eriksson S at the Ger Dep.

In these examples there would be no problem for search retrieval when limiting to university residence umea, but if we searched the name of the institution there would be. But there are cases when addresses of persons working at other universities are not mentioned along with author addresses from the adresses mentioned. Here’s one example that lacks all addresses:

Which means if you search with limit address you can loose valuable records.

When searching author Astrom S combined with address Umea, the third hit in Web of Science is a record for Astrom S at Dept Ophthalmology beside of records of Astrom S at Nursing Dep:

Also an extreme example if you search Eriksson S Umea you will get both one article by Eriksson S(taffan) and the rest by Eriksson S(ture) both at Dept Community Med & Rehabil (earlier Dept Geriatr Med for Eriksson S[ture]). A good example of the importance of using all first name initials when getting published the first time. Here’s the article by Eriksson S(taffan), confirmed to me by Eriksson S(ture):

And, as mentioned as a problem in the beginning of this blog post, records by Eriksson S(ture) at Dept Community Med & Rehabil have different implementations mainly because of changes from small departments (Dept Geriatr Med) to bigger departments(Dept Community Med & Rehabil). Some variants follow below:

Dept Community Med & Rehabil
Dept Geriatr Med
Dept Geriatr
Dept Community Med & Rehabil Geriatr Med

But in some cases the address field only contains the university and the department is not mentioned:

Umea Univ, Umea, Sweden

Conclusion: To be really sure of retrieving all of an author’s (with usual name) articles you need a publication list from the researcher. I once got a list from a researcher when doing a citation search for him and found two articles in Web of Science he never had on his list(!). Of course some articles may not be retrieved because Web of Science doesn’t index all peer-reviewed journals in the scientific world. Address limitation when searching Authors is really inconsistent in Web of Science.

Advertisements