Use of personas for user interface design: Results of field and experimental studies.

摘要

Persona is a fictitious user serving as a role model for designers, helping them make difficult decisions from the users' perspective during the interface design process. This concept, which was introduced by Alan Cooper in 1999, has been found to be a powerful tool for engaging the designers with the users. Although personas do not replace the rich data acquired through user testing, they enhance the interface development process by focusing the designers' attention on the target user until the interface qualifies for testing. Recognizing the benefits of persona, many companies, including the Fortune 500 have embraced this concept as a way to enhance their customer experience with computer interfaces.;As this new technique originated in the practitioner's world, no experimental validation has been conducted on the impact of implementing this technique in the interface design process. Although Pruitt and Adlin investigated methods for developing personas and using them in the design of interface, publishing their results in March 2006, the benefits of personas have yet to be experimentally validated. To address this need, the current research presents two studies focusing on the use of personas. The first details a case study developing an intranet application intended to be used by various user categories for analyzing data from aviation maintenance processes. The second experimentally validates the effect of persona-based user interface design on the performance of the respective user types.;The SUMI data and performance specifications gathered through the field studies indicated that the web applications developed to cater to various user categories from the aircraft maintenance industry were highly satisfactory to the end users. The empirical study conducted to validate use of personas showed that the users performed significantly better when they used persona interfaces than while using non-persona interfaces. The results also indicated that the secondary user categories liked the interfaces which were primarily targeted to the primary user category better than those which did not distinguish between the two user categories.