For the Honors Chemistry Web Search Unit.....

To determine the reliability of web documents, Tom O'Haver (http://www.inform.umd.edu/

UMS+State/UMD-Projects/MCTP/Courses/PhysicalScience/Chem122Spring96/Informatics.html) suggests following these guidelines. In order to be able to use a web document as a source, you should determine the author of the page, his affiliation, and a date for the material. In ideal cases, the author provides this information on the page you wish to reference. Most accredited universities (ending in .edu) and official government agencies (ending in .gov) give this information either on the page itself or on the next higher level page via a link. In contrast, many pages produced by individuals do not contain the name of the author, the institution, or a date on the page, nor in the next higher levels of the web account. You can usually identify individual homepages: after the first slash in the http address, you see a ~name: for example, http://www.teleport.com/~hemphill.

http://sedwww.cr.usgs.gov:8080/radon/georadon.html. A geologic service page on radon and how it enters homes. Can we locate an author, institution, or date......

In contrast, see

http://www.lclark.edu/~ruggiano/netc/toc.html. This is a good, current resource, but it is not citable. Can we find an author, institution, or date? We do not even know WHERE or WHAT the institution is (educational because it ends in .edu, but that alone is not a guarantee of authenticity or reliability).

To try to locate additional information, you peel away the layers of the http address starting at the end. The / slashes separating parts of the http address simply divide directories and sub directories. For example


At the National Computer Center computer at the University of Illinois at Champaign in the SDG directory, there is a Software subdirec tory. Nested within that subdirectory, there is a Mosaic subdirectory. In the Mosaic subdirectory is the NCSAMosaicHome.html document on the use of Mosaic.

If you are at the NCSAMosiacHome.html document, you are in the Mosaic directory. To get to the next higher directory, simply highlight AND delete the Mosaic/NCSAMosaicHome.html part of the URL. The browser should return a page with the Software directory. You will notice that the Mosaic subdirectory is highlighted a different color since you just visited it. You can keep deleting subdirectories (next you would delete Software/ ) until you get to the "root" directory where you should be able to find information about the host organization. You will get to try this during the lab time, using an evaluation model suggested by Dr. O'Haver.

Remember that a journal article from a professional print journal has been reviewed and edited by anywhere from 2 to 5 editors or peers, and then rewritten more than once, usually over 6 to 18 month period before it is published. A person citing the journal article has fair assurance that the data and information in the article is not simply opinion, but is based on sound data. Many professional journals provide some or all of their refereed articles online; these articles come with the same editorial review assurances that the print version does. However, on the web, articles from NON-refereed sources, can easily be published or self-published. One has no assurance that the information in such articles is true or that the source is reliable-because the article has not been peer-reviewed or edited. The burden of deciding whether to "trust" and use as reliable a web-published article falls on you, the user. We will go through a few exercises to help you distinguish citable from non-citable and reliable from non-reliable URL sources.

For each of the assignment URL sites and for your Internet citation for your Background, you will search for citation information: the author, the date, the publishing organization (eg., University), the title of the page (as it appears on the page), and the URL Title (as it appears in the uppermost Title Bar of the Browser). You must also note the URL, AND the date you accessed the page.

For each site, you will check the server type (see the next page for help discerning the types of servers: .edu, .com, .net, etc). .Edu and .gov types generally indicate you may have a page with reliable information (but not always-student and personal pages, on educational servers usually remain in the province of personal opinion). Personal pages, generally (but again, not always) marked with a tilde (~) are not usually considered reliable or citable. As authors change servers, the original URL may no longer be valid or the information in the page may be personal opinion, speculation, or spurious misinformation. One example would be the http://www.kzoo.edu:80/~k93hs01/chemlinks/ page, which contained information on natural medicines. Today, the page is no longer available. The information itself might have been true, but the source was not reliable and cannot be cited. Information on .com, .net, .org servers is often not citable and may be unreliable. Many .org sites are companies selling their wares, publishing non-refereed "data" to support their sales claims. The information may be sound, but you must subject the information to closer scrutiny than for information for the online version of the Journal of Biological Chemistry or information published on formal .gov pages.

You will encounter sites that seem to have all the citation information you need, but that should not be cited except in exercises like this one: http://lib.nmsu.edu/staff/susabeck/cat.html on "Feline Reactions to Bearded Men." The citation information lists Harvard and other prestigious institutions in the citation information. Your first clue that this site does not qualify for your internet citation is the .com server ending. The second should be your common sense.

One of the best sites in helping to evaluate web pages is the EBSCO Publishing page on Information Literacy: http://www.epnet.com/lrc_ft/infolit.html.

Another is the Widener University site, http://www2.widener.edu/Wolfgram-Memorial-Library/examples.htm.

Actually, the http://www.lclark.edu/~ruggiano/netc/toc.html page contains excellent sources. If you really wanted to cite this page, you could peel away directories to find the citation information. Http://www.lclark.edu/~ruggiano/netc/ tells you that the toc.html page was written on September 20, 1998. Http://www.lclark.edu/~ruggiano/ tells you that the author of the pages is Vince Ruggiano. Finally, http://www.lclark.edu/ identifies the institution as Lewis and Clark College. Without this information, you should not cite this page. In your information about the reliability of the information on this page, the ~ruggiano should give you a clue that the information is provided by an individual at the college and may or may not be true or accurate. Ruggiano's page lists links to fairly reliable university sources and might indeed be a good resource. The usual warning about personal opinion, non-refereed material that may be gone when the student leaves the institution apply.

R. Hemphill

last updated 9/1999