Using Google Scholar for African Studies Research
by Hans Zell


--Contents-- people online

Note: this is an extract from the revised and updated version of “Using Google for African studies research: a guide to effective Web searching”, which appears as Chapter 25 in the new and expanded 4th edition of The African Studies Companion: A Guide to Information Sources (print and online). Lochcarron, Scotland: Hans Zell Publishing, 2006; more details can be found at http://www.africanstudiescompanion.com/fourthedition.shtml; after 31 May 2006 at http://www.africanstudiescompanion.com

All examples of search terms are indicated in italics.

Cross-references indicated here with an --> arrow symbol are to other sections, and entries, etc. contained in the new edition of The African Studies Companion.

Still in its Beta version, Google Scholar was launched in November 2004 and can be found at http://scholar.google.com, with the somewhat pretentious slogan “Stand on the shoulders of giants” beneath its search box. The aim of the new service is to provide a search tool for the researcher, enabling them to retrieve quality material from the Web not normally indexed via Web search engines, and allowing searches across many disciplines and sources, including peer-reviewed papers, theses, books, abstracts and articles, from academic publishers, professional societies, preprint repositories, universities and other academic institutions and organizations. Google Scholar has recently expanded its international dimension with the addition of content in two languages, as well as new interfaces for a number of countries.

Google Scholar has attracted a huge amount of interest worldwide, but has had a mixed reception among librarians and the academic community. Some critics – perhaps sometimes a bit unfairly – have compared its search results and search options with other multidisciplinary, citation based or citation-enhanced mega-databases currently on the market. These include Web of Science (WoS) from ISI and Scopus from Elsevier, but both of these are high-priced subscription based products, whereas Google Scholar allows free access. It has to be added, however, that there are other major database services that are freely accessible, such as the scientific digital literature library CiteSeer http://citeseer.ist.psu.edu, where you can search 740,000 documents; or the archive of Highwire Press http://highwire.stanford.edu, a division of Stanford University Libraries, which hosts the largest repository of free, full-text, peer-reviewed content, currently (February 2006) with 915 journals and almost 1.2 million full-text articles online.

Much of the criticism has centred on the fact that there is an apparent secrecy by Google to reveal its sources: first, there are no links to sites where publisher partners and their journals are identified. Secondly, the scope and size of the Google Scholar database, and the breadth of its indexing and archives coverage, is not adequately explained. Thirdly, apart from a generic statement on the “About” pages, as well as one at http://scholar.google.com/scholar/help.html#access6, there is no explicit information about publisher archives, and the precise type of documents that are processed (e.g. major articles vs. shorter items, book reviews, letters to the editor, etc.). The criticism is valid, and it is true that the Google Scholar content and sources disclosure is not very informative.

Google Scholar ranks results in order of relevance, and indicates the number of time the research has been cited by other academics. The number of citations is factored into the Google ranking algorithm, which considers the full text of each article, the author, the publication in which the article appeared, and how often the piece has been cited in other scholarly literature.

Google says that for the time being Google Scholar “indexes only scholarly articles”, but this is a bit misleading as search results in fact also lead to results for a large number of book titles, and/or edited collections and anthologies.

One of the requirements for inclusion in Google Scholar is that at least an abstract must be made available to non-subscribers who come from Google and Google Scholar. It can index any research articles as long as its robot software is able to crawl them online. This includes articles in pdf format as long as they are searchable. Google Scholar also indexes HTML, PostScript, compressed PostScript (ps.gz), and compressed PDF (pdf.gz) files.

How to search Google Scholar:

Google Scholar for African studies:

When first launched in November 2004, initial search tests conducted in the African studies discipline were generally disappointing with highly incomplete results, but since that time the picture has improved considerably, above all in terms of number of search results now displayed. However, there are still some strange anomalies in the number of search results that are difficult to explain.

For example, a search for articles by the distinguished Africanist scholar Professor Ali A. Mazrui generates a somewhat puzzling array of search results.

(i) If you simply enter aa mazrui on the main search pages it will retrieve 634 results.

(ii) If you search for aa mazrui on the Advanced Search pages by clicking on to “Search only in Social Sciences, Arts, and Humanities” you get 478 results, but which then also picks up anything else with “mazrui” and “aa” (including e.g. abbreviations for African Affairs), plus results from --> Google Book Search; while “aa mazrui”, in double quotation marks, finds 194 results in this search mode.

(iii) If on the other hand you enter the search "aa mazrui", either as a general search, or with “Return articles in all subject areas”, it finds 259 results.

(iv) Finally, if you add the “author:” command, author:aa mazrui it will find 176 results, while author:"aa mazrui" displays 160 hits.

Another example:
author:p zeleza shows 108 results, but
author:pt zeleza only 64, or
author:paul zeleza, even less, 51, and which underlines the fact that searching using a single initial is better than entering full initials or full first name.

The search on Paul Tyambe Zeleza also indicated that Google’s policy of a minimum requirement of an abstract is not always strictly enforced, as at least two citations lead to sources, such as periodical articles in the --> African Journals Online/AJOL (554) database, which indicate “no abstract available”.

Some other search examples:
democratisation africa
will deliver almost 10,000 hits, and while it will find the most appropriate articles in the first few hundred results, many of the later results are simply links to articles containing the two words somewhere in the text.

However entering the same search terms as the phrase (in double quotation marks) "democratisation in africa" generates rather better results, a total 411, although this could also mean that you miss out on important papers that do not contain this particular phrase, either in the title of the article or book, or within the text.

This is also an interesting example to demonstrate that using alternative spelling conventions for a word, in this case the word “democratisation” (i.e. –ise or –ize) can make a significant difference in the number of results: "democratization in africa" will retrieve 884 results; and, by adding the OR operator
(see--> Using the OR operator), "democratization OR democratisation in africa" will increase it further to 1,204 results.

You can limit a search by name of journal, but if you search for articles or citations in particular journals, for example search term + “african studies review” it is prudent to put the journal name in double quotation marks, as otherwise you will also get results from other journal titles containing the words “african” or “studies”, and which won’t be relevant.

democratisation "review of african political economy" will display 39 results, or democratisation “journal of modern african studies” retrieves as many as 656 results. However, while in both the above examples the results contain a good number articles on the topic of democratisation in Africa, the rest are simply papers which contain the word “democratisation” somewhere in the text.

The same is the case if you do a country-specific search, about democratisation in a specific African country, aiming to retrieve all available articles, e.g. democratisation uganda While the results page for this search will display as many as 2,440 items, only the first few hundred results will the directly relevant, or at least partly relevant.

A search, without author, article title, or other search terms, but restricted to “african studies review” finds 1,540 results, in citation order. A search restricted to only recent 2005 articles finds 85, but none with citation scores as they are all recently published, and in no apparent sort order (i.e. author, title, or volume/issue number). It shows that, as at February 2006, Google Scholar had indexed the first two issues of African Studies Review of volume 48, 2005. A search for “journal of modern african studies”, with the publication restrict for 2005-2006 issues only, brings up 50 results of articles published in 2005, albeit only two with citation rankings, again due to the fact the articles are very recent. This would seem to indicate that Google Scholar is reasonably current. However, if you want to view table of contents and abstracts for the very latest issues of some of the leading African studies journals you will do much better to go direct to these journals’ Web sites. Most of them – including those of the major journal publishers such as Blackwell’s, Cambridge University Press, or Taylor & Francis – now allow free online access to table of contents of recent issues and/or abstracts. (See also Introduction to section --> 9 Journals and magazines).

The results for “african studies review”, or other Africanist journals (and retrieving all available records) are interesting because they appear in order of number of citations. This tells us, for example, that Jane I. Guyer’s “Household and Community in African Studies” published in African Studies Review (Vol. 24, No. 2/3, Jun. - Sep., 1981) is, according to Google Scholar at least, the most frequently cited article by far, with a total of 63 citations; followed in second place by Karin Barber’s “Popular Arts in Africa," African Studies Review, (Vol. 30, No. 3, Sep. 1987) with 31 citations.

Google’s interpretation of the term “scholarly” is ambiguous, and it is not always clear which papers it considers to be of a “scholarly” nature. For example, when searching for papers on publishing and book development in Africa, it does find a good number of articles and citations published in major journals, but also misses out a very substantial body of the literature on the topic; surprisingly, even some articles that are freely accessible on the Web. Why that should be so is not clear: are they not indexed, (i) because the sources are not crawled by the Google Scholar robots? (ii) The sources are not considered to be scholarly publishing outlets? Or (iii), is it because the topics and/or articles are not considered to be academic or “scholarly” enough?

Some searches in Google Scholar can also be decidedly unhelpful:, e.g. if you enter author:”h zell” Logos one search result will show a reference to “The production and marketing of African books: A Msungu perspective”, an article published in Logos in 1998, but the only two links on offer for this search are one to the British Library Direct service (where you could purchase the article, but does not offer access to an abstract), plus a couple of citations on the Web.

The pros and cons of Google Scholar


What’s good about Google Scholar?

What are the constraints and limitations?

  • There is just a single output format and the results cannot be sorted by the user. Results are sorted by Google Scholar’s citation scores and, in the absence of citations, for example for very recent articles, there doesn’t seem to be an intelligible sort order, e.g. by author, title, or by journal volume/issue number.

  • Google Scholar generally seems to perform better on scientific, technical and medical subjects, rather than in the arts, humanities, and the social sciences.

  • Google Scholar’s claim that it is opening up more content from the “invisible Web” is, at this time at least, not substantiated.

    Google Scholar can certainly be a very useful starting point for a quick overview of a topic, point you in the right direction, and then lead you toward relevant material. However, the results may neither be as current or as comprehensive as you need.

    Google Scholar is no substitute for the various subject-specific databases and indexing and abstracting services, which can provide more focused subject area coverage, and most of which offer sophisticated advanced search features that will allow you to fine-tune your search, and build a highly targeted search strategy.

    Return to Table of Contents

    Next Page