University Webmaster

Back to Web Publishing Policy

Appendix III.

Information Technology Recommended Testing Methodologies for Development of New Web Sites

There are a number of different ways to test to websites. Whichever methodology is used please make sure that it completely tests the site and that the site is a user friendly and accurate as possible. It is the hope of the University that visitors to our website have the best experience possible.

Below are a series of recommending testing approaches that commonly used to test websites. These are recommended to those developing new sites. If you have questions concerning these please contact the University Webmaster. These methodologies are offered as recommendations and should be taken as such.

Types of Testing:

a. Test for Broken Links - Throughout the course of development you should consistently look for links that are not being used or are broken. It is better to find them now than when the site goes live.

b. Browser Compatibility Testing – In compliance with these policies the site must be compatible across multiple Internet browsers. Please test your site to make sure that this is the case. If you need help selecting or obtaining web browser software please contact the University Webmaster or for help using it please contact Computer Support Services.

c. White Box Testing – Let people inside of your department or unit use the new site. Create a list of common tasks or information that is normally completed using the website. See how easy it is to get to that information. This allows you to test internal usability.

d. Black Box Testing – Let people who are not in your department or involved with the development of the site use it. Students would be best for this type of testing. Give them that same common set of tasks and questions as you did with the white box method. Let them give you feedback and input.

e. Regression Testing – Just because you take input and have fixed broken items you might want to test further after you have fixed the previous problems. Sometimes in fixing one problem you can cause another. Try testing the same as you did before now that the changes are in place.