Is usability testing as we know it about to radically change?
This article, by Dr. Susan Weinschenk of CUA, explores the current practice of usability testing and its possibility of changing directions in the near future. Apparently, newly conducted research regarding usability testing has offered some results that may potentially point usability testing in a new direction. The first aspect researched dealt with the question of the need for a usability engineer to be present during the testing. Research demonstrated that more usability problems were discovered when an engineer was physically present. The second factor studied was the need of testing with lo-fidelity or hi-fidelity prototypes. It turns out that mixed fidelity prototypes should be used based on the particular type of usability test you are conducting.
Another area researched concerned the dilemma of testing one design or multiple designs. Traditional methods of usability testing prefer testing one design, but researched proved that testing multiple design solutions provided more accurate and genuine results. A fourth aspect studied regarded the traditional practice of the think-aloud technique which encourages users to narrate their actions and voice their decisions. Research found that this method of testing provided more results than the newly proposed method of talking about actions after they were completed. Lastly, the study examined the validity of think-aloud techniques and the fact that most of our decisions come from non-conscious processes.
In conclusion, the research demonstrated that usability testing is most likely headed in a new direction in the years to come. I, for one, think this is a good idea for usability testing as a whole. Virtually all aspects of technology and online processes are rapidly advancing with each new day. Usability testing should follow suite and adapt along with the rest of the technological realm. Personally, if I were to conduct usability testing, I would want to explore new and improved methods of testing users.