What makes a Web site effective is an enigma worth investigating. To back up their hypothesis, that “increased interactivity” equals “increased perceived satisfaction” by the user, the authors of this paper conducted a controlled laboratory experiment with three different Web sites, exhibiting varying degrees of interactivity, and 54 subjects. The study examined usability in the context of commercial Web sites.

The study’s conclusion, that high interactivity levels do indeed contribute to higher effectiveness of information delivery on a Web site, is highly debatable, and the authors admit that laboratory settings a priori do not normally lead to valid extrapolation.

Further counterarguments to their hypothesis include: first, that every additional click-through that a user needs to make, in navigating through a commercial online environment, reduces the possibility of a transaction by 50 percent [1]; second, that variables such as response time for loading the Web site have a strong impact on the user; and, third, that the attitudes of a user are very much shaped by his or her personality, and warrant consideration. What if the person is not interested in having choices?

What is impressive is how this paper can serve as an example of how the internal validity of an experiment remains high, due to its stringent methodology, and at the same time, makes the entire problem appear bigger, and not smaller. Equally impressive is the depth with which the authors discuss the project, including about 80 references on the subject. Anyone who wants to evaluate Web users’ attitudes scientifically should start here.