TUlib

CR regionEN, Wysiwyg textEN

Evaluating Search Results

You are writing a paper about:

Developments in noise pollution from airplanes

Your search has produced a list of results. How can you make a quick decision about which documents will be useful for your paper?

You have to evaluate the relevance and reliability of your results in order to select the best literature to read and use:

• Relevance depends on your topic.
  A document can be scientifically sound and written by a very good author, but if the topic is not what you are looking for, it is still useless.
   
• Reliability depends for a large part on the source of the information.
  If you use the TU Delft Library catalogue or a bibliographic database from a reputable publisher, you can safely assume that the information it includes is reliable.
  If you use websites or other freely available information where the source is unknown, you will need to check if the information is reliable.

How do you judge relevance and reliability? Here are some examples using different sources.

Search results in the TU Delft Library catalogue

What can a catalogue record tell you?

• Title: is the title relevant for your project?
  Is it about noise pollution from airplanes, and not about other pollution or pollution from other sources? A more general work about noise pollution may be relevant as background information.
• Year of publication: when was the document published?
  Do you only want very recent material or can older material be useful as well?
• Author/corporate author/publisher: is the author a well-known authority in this field or a respected research organisation?
  Is the publisher a reputable scientific publisher?
• Document type: a book, dissertation or report may all be useful?
  If the document you found is a journal or conference proceedings and the title seems very relevant, you will have to look at some recent issues or the table of contents to see if the articles are relevant.
• Classification code: the classification code gives an indication of the subject.
  Is the subject relevant for your paper?

Search Results in a Bibliographic Database

In the results list of a bibliographic database, you can first look at the title, year of publication and author, but a database provides extra information.

Three important extra criteria for evaluating results in a bibliographic database are:

• Citation information: has the article been cited by others and if so, how many times?
  Of course, it takes time for an article to be cited, so very recent articles will not have been cited yet.
• Source: which journal did the article appear in?
  Is it a well-respected journal? Is the journal perhaps included in the publication list of your department?
• Abstract and keywords: the abstract and keywords offer a good insight into the contents of the article.

Websites

Searching the internet is different, because unlike a library catalogue or a bibliographic database, information on the internet is not selected or regulated by a reliable organisation. Assessing search results in Google or other search engines is almost entirely a question of evaluating the reliability of sources.

• Objectivity: who is responsible for the website?
  Was it set up by a university, a government department, a company, a pressure group? What is the purpose of the site? Is it a sponsored site? 
• Author: what information can you find about the author?
  Who is the author? Can you find the author’s affiliation? Is the author an expert or a lay person? 
• Accuracy: is the information correct and can you find cited references?
• Comprehensiveness: is the information complete, does it give a full picture?
• Date: when was the website set up or last revised?