Thursday, October 29, 2009

Getting the most out of bibliographic databases

Oh, the soaring boringness of that title. But some careful attention to this part of the process will pay off later in a big way. You've found your way to Web of Science. What now?

Web of Science is part of Web of Knowledge, which also contains lots of other bibliographic databases (they're listed here, in the Web of Science factsheet, which gives the official info on how the system works). A bibliographic database is essentially a massively searchable list of scholarly articles, books, chapters, etc. So what I have to say about Web of Science (the database I am most familiar with) should be mostly applicable to whatever database is relevant to your own discipline.

Here's the basic Web of Science search screen (if you can't get this far, you need to go back to last week's post):



You'll see from the above image that you have got three basic search boxes. Each one can be assigned differently (you can change that 'Topic' to 'Author', for example, using the drop-down menu). Say you're writing an essay on the evolution on the human female body shape, as some of my students are doing this term. Just type some or all of those words into the 'Topic' search box, and hit 'Search'. You're away.

When I did that just now, I got 104 relevant results. How am I to know which of these articles is important? Well, I can use the other search boxes to limit my search in various ways. Use the 'Year Published' criterion to pick a particular date or date range. Back in the 'Topic' box, try putting female body shape in double quotes, so that only items containing that exact phrase are returned. 

You've ended up with a list of articles on your topic. The next step is to sort that results list. Look for the 'Sort by' drop-down menu at the top right. This will allow you to sort by date, by relevance, and various other criteria. Possibly the most useful of all is the option to sort by 'Times Cited'. You'll remember that a citation is where one article references another article. This is handy because it allows us to sort our results list according to the number of such references our articles have had. If number of citations is an indicator of how important an article is (debatable, but probably largely true), you have got an instant way of arranging your results list by order of importance, with the most highly cited (most influential) articles at the top.

Now start exploring those items in your results list. Each article in the list will include a live link which takes you to the record for that article. That record will look something like this:



There are several important live links embedded in this screen. Right under the title you'll see the names of the authors. These are clickable, and lead to a separate results list for each author—a list of all the other things they've published. As scientists tend to stick to the same areas, it's highly likely that your author will have written other relevant things on the topic. Click on him/her, and then sort the results by 'Times Cited'. Then you'll find out what your author is really famous for.

Just below the author names, you've got links for Times Cited and References. These lead to lists of which papers have cited this particular article (Times Cited) and which papers this article itself cites (References). Explore both of these to follow the trail in different directions. You'll also see some details on citing articles in the blue box to the right of the screen.

Now it's time to actually read something. You've decided that you like the look of this article, and you want to track it down (preferably through e-journals, so that you don't have to traipse to the library). The Connexions button, right under the title, will tell you immediately what access you have to that article, and will hopefully take you right to it. If Connexions doesn't work, go back into the library catalogue and search E-journals. Failing that, the library may have a paper copy of the journal, so you'll need to get your shoes on.

Web of Science can of course do much more than this. Do you remember the name of an author mentioned in the lecture, but no other details? Search on the name using the 'Author' search box, and add in a keyword in the 'Topic' box. Make use of the fact that Web of Science is updated every week to find the very latest articles in your topic area. Use the 'Add to Marked List' tool to keep track of the articles that look interesting (retrieve your list at any time by clicking on 'Marked List' at the top). Keep an eye out for recent review articles (comprehensive reviews of a topic published in journals such as Psychological Bulletin, Annual Review of Psychology or Developmental Review). These will give you an expert review of the topic which should be a valuable way of directing your reading.

Using these tools wisely will mean that you should be able to include lots of relevant references that go beyond the info in your lecture notes. And that, as we saw last week, is a Good Thing.

Reblog this post [with Zemanta]

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.