Evaluating Information

How to evaluate information, from social media to scholarly articles.

Info on the Internet

This document will give you a variety of ways to look for each kind of information. Always remember that there are other, nonelectronic, methods of getting much of the information discussed in this document. Visit the library and ask a librarian for help. Of the five evaluative criteria listed in Evaluating Information found on the Internet, three may be investigated by electronic means:

  • Authorship
  • Publishing body
  • Currency (of the document itself)

Update: This page cited by The New York Times! See "Whales in the Minnesota River? Only on the Web, where skepticism is a required navigational aid", by Tina Kelley (New York Times, March 4, 1999; available online at http://www.nytimes.com/library/tech/99/03/circuits/articles/04trut.html)

What do we need to know about the author? When the author is someone unknown to you, ask the following questions:

  1. Is the document signed?
  2.  Can I get more information on the author by linking from this page to other documents?
  3.  Was there information about the author on the page from which I linked to this one?

If you can answer "Yes" to the second or third question, it's possible that you will have enough information for evaluative purposes, or at least enough information to help you find the author's telephone number or e-mail address so that you may contact him or her with questions.

If you can answer "Yes" to the first question only, you may need to find further information on the author. There are a number of ways in which you might do this:

  1. Go to the home page of the web site where the document lives and search for the author's name using any available internal search engine or directory (works best for academic web sites). This may help establish affiliation.
  2.  Try searching the author's name, enclosed in quotation marks, in Google. This may lead to other information on or pages by the same author.
  3. Try using an e-mail address finder. A compendium of finding aids and their comparative value may be found at FAQ: How to find people's E-mail addresses by David Alex Lamb, Department of Computing and Information Science, Queen's University, Ontario, Canada. This is what football fans refer to as a Hail Mary: you're tossing a search into the unknown in the hope that it finds someone.

These are not ideal means for finding someone's credentials, even though they may help reveal someone's identity. Evaluating information usually consists of weighing a number of criteria together, so you will need to assess how important authorship is on a case-by-case basis.

If no information on the author can be found, or there is no signature or attribution on the page itself, go directly to...

OK, so where are we, in the geography of cyberspace? Look at the web page you are trying to evaluate. Does it include any of the following?

   1. A header or footer that shows its affiliation as part of a larger web site.
   2. A watermark or wallpaper that fulfills the same function.
   3. A link at the bottom or top of the page that allows you to go to the home page of the web site where the document lives.
   4. A link that allows you to send a message to the site Webmaster.

These features help you judge the "official character" of a web page. They act as an assurance that the page you are evaluating functions within some type of institutional setting. Judging the official nature of a web page is extremely important if the page is not signed. Some web sites do not include attributions to individual authors, so you will have to rely on your ability to evaluate the institution, or domain, where the page lives. Caveat: while the Web is looking better, especially official sites maintained at educational institutions or by scholarly societies, not everyone has caught up with the importance of consistent graphics or "return" links. So you may be looking at a perfectly good page that hasn't got any visual clues to its affiliation. Move on to the next step if this is the case...

If your page gives no clues as to its identity, you will need to focus on the URL, or address.

  1. Can you find the web site's home page by deleting all the information in the URL after the server name?
  2. Can you tell if the page is actually part of someone's personal account, as opposed to being part of an official site? Click for help in evaluating this kind of URL.
  3. If all else fails, can you find information on the server or domain? Try using BetterWhois to get the name of the owner. Type the name of the server (e.g., milton.mse.jhu.edu) in the box and press Enter. BetterWhois should find the identity of a server, as well as contact names and telephone numbers.

Once you find the name of the organization owning the server, you may have enough information to judge its reputation as an information source. Remember, this is only of value for officialpages from a web site. If the page you are evaluating comes from someone's personal account, you really have no idea what their place is within the organization, or if they are in a position to represent the organization. If you are not familiar with the organization, try one of the following:

  1.   If it is an association of some kind, look for it at The Scholarly Societies Project. Is it represented?
  2. For all others, search the name of the organization, enclosed in quotation marks, in Altavista. Does anyone else have information on it?

Why might an author or publishing body want to remain anonymous? Read Information and its Counterfeits: Propaganda, Misinformation and Disinformation to learn why...and to understand why you need to know where "information" comes from.

If you cannot ascertain either the author or publisher of the page you are trying to evaluate, you are looking at information that is as anonymous as a page torn out of a book. You cannot evaluate what you cannot verify. It is unwise to use information of this nature. Look for another source.

Even when you can find information on the author and/or publisher of a web page, you should still consider how "fresh" or "dusty" the document is. This is especially critical if the document discusses time-sensitive information, such as census information or other statistics. Look for internal confirmation of the information:

  1. Does it use a caption such as "Based on 1990 US Census data"?
  2. Does it include information within the document such as "Closing stock prices, September 30, 1996"?
  3. Is the statistical source listed in a bibliography to the page?

If you cannot ascertain where the statistical information comes from, or what its age is, you are once again looking at anonymous information.

It is also valuable to know when a page was last updated. Has it been "pruned" or "dusted" lately, or has it been sitting on the shelf?

  1. Look at the bottom of the page. Does it have a "last updated" date?
  2. Use the "Document info" feature in the "View" menu on Netscape. Does it tell you the date?
  3.  Change the URL by backing up to the last slash (/) in the address. This may allow you to see the details of the directory or subdirectory of the server including your page. Usually the last modification date is included.

It's valuable to know the age and updatedness of your page because you may be looking at an orphaned or superseded document that has been replaced by other information.

Always remember that the best counterfeit looks the most like the real thing. How genuine and trustworthy is your information?