Searching for literature: why Google Scholar is a blunt instrument

If you’re going to use a tool to help you with an important part of your research, it helps if you know a bit about how it works.

Searching for literature is a major, time consuming, and vital part of any PhD, so your choice of search tool matters.

There are some major drawbacks to using Google Scholar. That doesn’t mean you shouldn’t use it, but you should know what it’s weaknesses are.

The Google Search Algorithm

When you do any kind of search through google, their search algorithm decides in what order to show you the results.

It’s an incredibly sophisticated system, taking into account all kinds of measures of importance and relevance for each search result. But here’s the problem…

Nobody knows what the algorithm is. So you don’t know how they’ve sorted them, and you have no control beyond selecting the search terms and what years to search.

An example

If I search for “scanning tunneling microscopy”, then I recognise some of the top results (including the nobel-prize-winning inventors of the technique). There are papers there which have been cited hundreds or even thousands of times. So far so good.

But there are articles there that (with all due respect to the authors) have no business being in the top 10.

Why Google ranks a paper with 3300 citations at number at number 6, and a paper with 21 citations (from 1990, so it’s had plenty of time to have more of an impact) at number 7, is anyone’s guess.

This stuff matters. There are over 200,000 search results, so if google is filtering and sorting the results for me, I’d first I’d like to know how, and second I’d like to be able to play with the settings to sort results the way I want.

The advantages of Google Scholar

Well it’s free, so anyone anywhere can use it (even if you have to pay for access to some of the results).

And of course you can change the search terms you enter to get more specific results (but that’s not really an advantage as you can do that with any search engine).

Alternative Tools

I always used Web of Knowledge, which gives far greater control and transparency over search results. It requires a subscription, but if your institution is registered then definitely use it.

The key thing is being able to control how search results are presented to you. Leaving it up to Google is not PhD-level thinking.

I’m going to throw this one over to you in the comments section. What tools do you use to search? And why?


13 thoughts on “Searching for literature: why Google Scholar is a blunt instrument”

  1. Heya i am for the first time here. I found this board and I in finding It truly helpful & it helped me out much. I am hoping to offer something back and help others such as you aided me.

  2. Two more drawback of google scholar are: 1) it do not indexes all data bases and 2) sometime most relevant results are in the 70th or 80th pages.

  3. The thing that really bugs me about google scholar is that you can be searching for an exact book for citation, and it will show every. Paper that has mentioned the book, but not the book itself!
    I use it because I just use uni computers and it can link into uni refworks account.
    I’m generally quite computer illiterate, I’m old enough to have missed the computer revolution at school, so even though I would like to try the things you recommend it seems a bit scary.

    • Hi Gill, why not go and see the librarians at your university library? They are the best people to recommend which sources you should search and can give you search tips to help you get the best results possible. I’m biased of course, because I’m a librarian (‘Information Specialist’) working in a university library!

  4. The feature I like is the Create email alert button at the bottom of the search results. It will send a digest of “new” links it has found that match your search query to a Gmail account. You get a title, 3 lines of abstract, the authors and a link. You have to be very careful that it doesn’t take too much of your time sifting through the results. I found that it presented many results where the search terms were just mentioned in passing and not relevant. It’s better for following an author, say the big names in the field or someone who’s work you’re making comparisons with. To make the alerts work well, you need to be _very_ selective.

  5. Hi Guys, I am quite new to this blog. I am doing PhD. I was wondering, while searching literature, if there is a latest review that has summarized the latest findings drawbacks and gaps in the literature, do I need to go through all the papers references in the review or can put the review only as the reference and go through the most important ones.

    Thank you.

    • It’s a good idea to read some of the original papers, but not all. It really depends how important each one looks to your research project.

  6. can I ask you advices about papers organization? (sorry James if you already talked about it and I missed it!)
    I downloaded Mendeley and I was really enthusiastic at the beginning, I thought it would have changed my life. It didn’t, and my papers are still in the same mess as before.
    Paper mess = panic just to think about them (even if they are really cool…they get me in mess-panic mood)

    • The low-tech solution is the best, I think. Print papers, and file in folders by subject. Write notes and highlight key sections on the hard copy of the paper so you can reach out and grab everything you need on a particular topic.

      Also, focus most of your effort on the most important papers. There will usually be a handful of papers which are far more useful than any others, make sure you fully understand them!

  7. Agree with all of you. The key is to look widely but be discriminating in what you read. Any number of algorithms can’t replace your own critical thinking. My problem is knowing when to stop!

  8. Well, Scopus is my favorite. Their author search works quite well to find more literature written by the same author.

  9. I’m trying to write my PhD Dissertation while living in Nigeria (though doing the PhD at a university in the U.S.) so Google Scholar (as a free resource) is absolutely essential to my research. I generally use it to look up resources found in the bibliographies of other papers. It’s the best way to find out if the source is available for free before going through the longer, more tedious process of signing into my university’s library website, which can be difficult on slow internet. For searching topics, yes, it doesn’t always bring up the most helpful information initially. I have been impressed by how many resources you can find for free–including entire books, which being an ocean away from my university library, I would not otherwise be able to access.

  10. totally agree with your writing… but let us not limit our searching to google scholar only, maybe for a kick start it would be great jump.. to reach higher peak.. but i believe, for literature one must explore whole… example.. windows live academic search,… and scopus.. ieee… but… in the end.. ISI web of knowledge is a bless…. all the best to all…

Comments are closed.