Author-address search

Web of Science have introduced a feature called Refine your results which is displayed in top of each search. Check this screenshot:

There are several options for refining, for example institutions. When searching for Haglin L as author you get 11 hits and when clicking Institutions 8 alternative addresses:

The problem is:

1) You won’t always see the name of the department in the overview, even if it’s in the record. You have to click the record to view it.

2) You will get all affiliations for all authors of all articles, not just for the author you searched or the first author name. Addresses are mixed together for all authors.
Positive is you get a better overview of different spelled addresses for authors.

Refining option Subject categories is not a subject search limit I recommend. It’s built upon the subject categories of the journals where articles is published. Sometimes you can see an option for Concept Codes or Descriptors or Controlled Index. It’s connected to content from the database BIOSIS and should not be used as an multidisciplinary subject search. These are options in development and will see in future what it can bring.

Conclusion: A better, faster overview, though same problems exist with lacking address/author information.

Web of Science have launched Author Finder and Scopus has of course launched Author Identifier. Both tools are supposed to assist when there are lots of variants in author listings. In this news flash Authors, Authors: Thomson Scientific and Elsevier Scopus Search Them Out by Barbara Quint at 24 July 2006 you can read more about these tools.

So let’s do some evaluations of these tools by using examples from my other Author-address search evaluations in Web of Science and Scopus. Let’s try my extreme author search example on Rantapaa-Dahlqvist, S. This is a screenshot from Web of Science when searching Rantapaa in Author Index which I usually suggest to use to get more complete author searches:

Here’s a screenshot of the first step in using Author Finder in Web of Science entering the last name and first initials Rantapaa-Dahlqvist, S.:

Step 2 displays author variants. But in this case Author Finder can’t find misspelled variants of the last name, just variants of first name initials. Both 51 records of Rantapaa-Dahlqvist, S and 1 record of Rantapaa-Dahlqvist, SSRD is retrieved:

Step 3 gives you the option to choose subject categories. Those categories are based on the subject of the journal published in, not the article itself:

Step 4 is a valuable refine option called Select Institution. Here you can choose the author affiliation, but it’s not always possible to refine on department level, even if the department exist on the record. The confusing thing here is that you get all the affiliations of all authors of all articles, not just for the author you’re searching. One benefit with this refine option (which also exist in Refining you results) is you get a list of possible synonym addresses and misspelled or incomplete addresses also. If you use this option to refine your search you must be aware that all records are not complete with author addresses and refining can make you loose important records:

Let’s try searching for Stegmayr, B. You get options for Stegmayr BG and Stegmayr BK also. If you choose all variants: Stegmayr, B, Stegmayr, BG, Stegmayr, BK you get 236 records. Trying to refine in step 3 by address is just such a mess because all the author addresses from the articles are displayed. Both Stegmayr are active at the same university department. Stegmayr, BK has written one article with address DEPT INTERNAL MED, but when refining with that department name you get 2 other records instead.
Let’s search for Haglin, L with Author Finder and refine to address. In this screenshot we choose UNIV UMEA HOSP, UMEA UNIV, CTY COUNCIL VASTERBOTTEN, UNIV HOSP:

That search returns 7 hits which means 4 older records are not returned. Compare results here:

This is one of the records that will be missing because Haglin’s affiliation at that time was in Uppsala:

When searching Astrom, S in Author Finder in Web of Science there are lots of options for refining on affiliation level. Refining to UNIV HOSP will also retrieve Astrom, Siv at the Department of ophthalmology, not just Astrom, Sture at Nursing Deparment. There is no option to choose just UMEA or if you take UMEA UNIV you miss Astrom, Siv. But most strange is why DEPT OPHTHALMOL isn’t optional, though other deparments like DEPT MICROBIOL is displayed. It stands clear this refining option on Institution name is just a mess!!

Searching for author Bernspang B in Cited Ref search returns a list with one entry Bernsprang B also:

The entry Bernsprang B is a misspelled citation for her thesis:

When searching Author Finder you will not retrieve the misspelled Bernsprang when searching Bernspang:

Searching Lundin Olsson or Lundin-Olsson returns no hits either in General Search or Cited ref search. Searching LundinOlsson in Author Index retrieves both LundinOlsson ,L and LundinOlsson, I:

Checking the variant with I as first initial indicates it’s a incorrect citation:

Checking Scopus shows the record is not incorrect in their reference list:

So let’s check the Author Indentifier in Scopus. It’s integrated in Author Search. A search for Rantapaa-Dahlqvist returns three variants:

But truncated Rantapa* S* returns boths variants of Rantapaa-Dahlqvist and Rantapaa-Dahlquist:

But you also need to search Dahlqvist* s*:

An advantage in Scopus compared to Web of Science is the possibility to search fullength first name. But that doesn’t even help always. Searching Eriksson, Sture returns following screenshot:

Choosing first hit gives 66 records and three records are associated via Eriksson, S-E. It’s not the same person as Eriksson, Sture at Umea university. Eriksson, S-E affiliation 2001 was Falun hospital:

Hit 9 doesn’t show affiliation but when checking the hit the record has affiliation connected in the record:

Hit 2 when checking has also affiliation at the record and shows up being Eriksson, Sture at Umea university and hit 3 and 10 are Eriksson, Sture at Umea university but none of them are connected to the first hit of Eriksson, Sture with 66 records. That’s not an improvement, that’s confusing.

The good thing is though that Eriksson, Staffan (same first initial as Sture) at the same department as Eriksson, Sture is not connected here in the search for Eriksson, Sture, which is a problem in Web of Science, as they are not using fulllength first name.

Conclusion: Author finder in Web of Science could be hazardous for detecting possible misspellings of author names. The refining option of Institution Name in Author finder presents a nice overview but is really hazardous because the addresses don’t refer just to the author, instead all co-authors also! All address information is not always possible to include and excluding options are not available. I suggest using Author Index instead.

Scopus Author Identifier has some improvements, though there still are serious flaws existing. Misspellings are still a problem and not solved in this new algorithm.

Of course, Google Scholar also has problems similar to Web of Science and Scopus when indexing author Rantapaa Dahlqvist S. When searching via advanced scholar search in author field, the following advanced search operators are returned: author:Rantapaa Dahlqvist author:s and 41 hits.

One of the 41 hits have both first name initials SB. The result also includes 7 hits on author variant Dahlqvist SR. A search on Dahlqvist SR shows on the contrary 28 hits on that variant. And here we find the article that was published in Lancet and was misspelled in Web of Science as Dahlwvist SR.

In contrast, a search on Rantapaa-Dahlqvist S without ää returns just 35 hits(!).

Trying to search misspelling Rantapaa-Dahlquist S returns no hits while same misspelling with ää Rantapää-Dahlquist S returns 2 hits which are not included in the Rantapaa Dahlqvist S search of 41 hits.

But searching Rantapää S returns 43 hits included misspelled rantapää-dahlquist.

But this is not all. Searching Rantapaa SB gives one more hit not included in the 43 hits or 41 hits mentioned before. Such a mess!

Let’s try some other authors. Searching author:sojka author:p also returns hits of P Jakubus Z Sojka. This means that all first name initials with P and all surnames called Sojka are searched with this syntax.

To get refined matching, use quotation marks like this: author:”p sojka”. But still the problem exists that you have both pa and pe sojka. To exclude Paul E Sojka you could write: author:”p sojka” –author:”pe sojka”, but still you can’t be sure that all p Sojka records include just PA Sojka. And to find records by PA Sojka you can’t add it like an OR-statement. Didn’t work for me. You need to do a separate search: author:”PA Sojka”

And it’s not possible to a make limited search to address. But, as you may already have learnt from previous postings, author address search in Web of Science and Scopus is too inconsistent to suggest as a valuable method for refining your author search.
It is also worth noting that if you don’t restrict an author search to the author field in advanced search or with advanced search operators you will get hits where the author name exists in the fulltext of the articles which Google Scholar indexes parts of.

Conclusion: The same problems with author search as in Web of Science and Scopus exists in Google Scholar. And as I said in a previous posting, a proper publication list from the author is the best way to be sure to find every important article.

When searching for an author in Cited Ref Search, my recommendation is to use the Cited author index which makes it possible to browse authors. When searching Rantapaa in the Cited author index you get 7 author variants as seen in the screenshot:

As we learnt from the previous posting Rantapaa Dahlqvist search test in Scopus we also have to try the author variant Dahlqvist SR. Two additional author variants are found: Dahlqvist SR and Dahlqvist SRR.

In contrast, browsing Author Index via General Search gives 8 author variants when searching Rantapaa.

And browsing Dahlqvist S gives three author variants: Dahlqvist SR, Dahlqvist SBR and Dahlqvist SRA.

Searching all these author variants both in Cited Ref Search or General Search doesn't retrieve the article published by Rantapaa Dahlqvist S in Lancet. Why? Searching the article title "New concept in Echocardoigraphy" gives the explanation. As seen in this screen shot the author is misspelled Dahlwvist SR.

And as we know from the same author test in Scopus, both PubMed and the original electronical article from Science Direct have interpreted the author correctly as Rantapaa Dahlqvist S(olbritt).

These are all author variants found in Web of Science:

From Citations:

Rantapaa S
Rantapaa SB
Rantapaadahlqvi S
Rantapaadahlqvi SR
Rantapaadahlqvist S
Dahlqvist SR
Dahlqvist SRR

From articles:

Rantapaa Dahlquist S
Rantapaa Dahlquist SB
Rantapaa Dahlqvist S
Rantapaa Dahlqvist SSRD
Rantapaa Dahlqvst S
RantapaaDahlqvist S
RantapaaDahlqvist SB
RantapeDahlquist S
Dahlqvist SR
Dahlqvist SRA
Dahlqvist SBR
Dahlwvist SR

Conclusion: Although this is an extreme example I suggest to use author index to check for possible misspellings. Having a proper publication list is recommendable.

In Scopus you have a separate bar for Author search. You could also choose to limit to Author in Basic search or search operators in Advanced search. When searching rantapaa s or Rantapää s with Swedish/Finnish diacritic characters ä, you receive 23 hits including the article indexed with the full first name Solbritt.

But if you choose to search with Rantapaa-Dahlqvist S or Rantapaa Dahlqvist you loose the hit with first name Solbritt because that record is spelled Dahlquist.

But searching Rantapaa Dahlquist S doesn’t find all dalhquist-spelled records. Neither does a search for Rantapaa Dahlquist SB or with dots S.B. or with space S B.

When you instead search Dahlquist s, you will find the dahlquist-spelled records.

So when I click on the 20 hits from my initial search Rantapaa Dahlqvist S I get 69 (!) records instead of 20. How come?
Even stranger: When clicking the author name in one of the records, 71 records are retrieved. But the number of records retrieved in the initial search was 72. Could it be more confusing? View this video [AVI]

Still none of the 72 records mentioned includes the article by Rantapää Dahlqvist S she published in Lancet 1998. Searching that article by title shows that her name is implemented as Dahlqvist SR. Clicking on her name under author(s) gives 28 hits. 4 of them give hits on other spelling variants of her name but not all 72 articles from the initial search.

If you check this article “New concept in echocardiography” in PubMed you will see it’s implemented with Rantapaa Dahlqvist S. Checking the article in the electronic source of Lancet via Science Direct (Yes, it’s owned by Elsevier who owns Scopus!) shows the name implemented as Rantapaa Dahlqvist, Solbritt.

So how to find all articles by Rantapaa Dahlqvist S and author variants? The initial search Rantapaa S gives 72 records. Searching Dahlquist S gives 2 records. Searching Dahlqvist SR gives 24 records. Together 98.
Conclusion: Though this author search is an extreme example, it shows the problem of the implementations of author names. You have to search all possible misspellings of a name. Preferably, you should have a proper publication list from the author. And at last WHY is the reference from Lancet discussed above consistent in one Elsevier product (Science Direct) and not in the other (Scopus)?

When searching with author limited to address, you sometimes loose articles because of inconsistent implementation of addresses as we already have shown in a previous posting. We tested Web of Science and Scopus. In Google Scholar there is no option for address limit.

By searching for example au=haglin l NOT ad=umea in Advanced Search in Web of Science and haglin l (author field) AND NOT umea (affiliations field) in Scopus, the articles lacking correct addresses were easily found. Then we counted the citations in these articles that would be lost when limited to address search.

Note: The example is based on searching in author in General Search, not Cited ref search.

Elsevier told me the following concerning implementation of addresses in Scopus: “For recent and ongoing records we include all author affiliations, if they were available in the original source title. However, we have not been able to source affiliation data for all archive material. In some cases only the first author address is included”.
Scopus is labelling author search field affiliations as seen on this screenshot:

When Scopus has implemented all address information,
contrary to Web of Sciecne, they connect each author
with the exact affiliation/address by using letters
alphabetically from a, b, c etc. Look at this screenshot:

As you see in the screenshot they also use two icons right behind
the last author.

The letter icon points to the corresponding e-mail address and
the pencil-on-paper icon links to the corresponding/reprint
author address and e-mail again.
If searching, like in the Web of Science, for authors Haglin, L and
Rantapaa Dahlqvist, S and then limited search for address umea you
get the following result as seen in the screenshot. Be aware that

Rantapaa Dahlqvist is searched both with single first name 
initial S and then both initials S B.

This means you loose 9 records from Rantapaa Dahlqvist, S(B) and 5

from Haglin, L. 3 of the

records from the search on Rantapaa Dahlqvist, S(B) are from 2002,
which does indicate that missing

addresses is not just a problem for older records.

This screenshot shows a sample record
from journal Rheumatology 2002 lacking addresses for

Rantapaa Dahlqvist, S(B).

Checking the original electronical article in the following screenshot
shows that address for Rantapaa Dahlqvist, S(B) and co-authors
at same institution are included.

Next Page »