Several web usability authors recommend that website conventions should always be followed and Nielsen even adds that even if another, better design solution exists, the conventional one should be used because if something which is “used on the majority of other sites will be burned into the users’ brains”. They add that if conventions are not followed there will be catastrophic usability problems. This guideline then poses a dilemma for interface designers since sticking to conventions can stall or even in some cases totally block innovation. It also makes one lose hope that a website migrated from the desktop to the small screen (which has to have a different layout due to screen size limitations) will be usable at all.
To complicate matters even further, ‘convention’ is quite a broad term and leaves the person trying to follow the guideline confused. This is understandable as, for example, a certain interface can follow the visual conventions of a website genre but have a navigational structure which violates the conventional one. Thus, when referring to a convention it is essential to be specific.
This muddle has even taken hold of the little empirical evidence available on the subject. Results from empirical studies on conventions have been contradictory (Kalbach & Bosenick, 2006; McCarthy, Sasse, & Riegelsberger, 2003; Vaughan & Dillon, 2006). Some found that violating conventions had no impact on user performance (Kalbach & Bosenick, 2006), some found a small usability problem which was quickly overcome (McCarthy et al., 2003) and others found a serious usability problem (Vaughan & Dillon, 2006).
I believe these conflicting outcomes are a result of an ill definition of the word ‘convention’. I propose that the term ‘convention’, especially regarding digital interfaces (such as the web), be divided into the following subgroups: Structural conventions, Visual conventions, Linguistic conventions and Functional conventions.
Examining the studies above under this light it is possible to observe that:
Kalbach and Bosenick as well as McCarthy, Sasse and Riegelsberger have tested violation of the visual conventions (the position of the menu) of a website while Vaughan and Dillon have violated visual and structural conventions simultaneously. These results seem to suggest that violations of different types of conventions may affect the user experience differently.
In a recent study (Santa-Maria & Dyson, 2008), I had the opportunity of test the violation of visual conventions (conventional position, symbolization and configuration of all the interface elements) of a website genre. Results confirmed that users who are familiarized with the visual conventions of a website genre can easily adapt to a new layout and their performance improves very quickly. The study also showed that the users get less disoriented over time when using the violating website.
Although usability problems caused by violations of visual conventions seem to be short-lived, designers should keep in mind that the users will still face problems in the initial phases of their interaction and therefore designers should balance whether the innovation is worth the initial user problems. Although these initial problems may put users off of using the site, if they are given enough motivation to get through learning stage their performance will quickly improve and equal their previous performance on the conventional website.
For further discussion on the topic see:
Feralabs Blog
Concept7 Blog
Kalbach, J., & Bosenick, T. (2006). Web page layout: A comparison between left- and right-justified site navigation menus. Journal of Digital Information, 4(Vol. 4). Retrieved from http://journals.tdl.org/jodi/rt/printerFriendly/jodi-111/93
McCarthy, J. D., Sasse, A. M., & Riegelsberger, J. (2003, 8th - 12th September). Could I have the Menu Please? An Eye Tracking Study of Design Conventions. Paper presented at the HCI 2003 - Designing for Society, Bath, United Kingdom.
Santa-Maria, L., & Dyson, M. C. (2008). The effect of violating visual conventions of a website on user performance and disorientation: how bad can it be? Paper presented at the 26th annual ACM international conference on Design of communication.
Vaughan, M., & Dillon, A. (2006). Why structure and genre matter for users of digital information: A longitudinal experiment with readers of a web-based newspaper. International Journal of Human-Computer studies, 64, 502-526.
0 comments:
Post a Comment