David R. Danielson on web navigation and orientation

25Jun08

In a fascinating 2003 paper, David R. Danielson of Stanford University discusses web navigation in terms of ‘transitional volatility’ – that is to say, the extent to which pages change in both navigation display and content within a site.

In Danielson’s conceptual schema, visitors navigation and orientation is the result of three factors:

 

1. Habituation to a particular navigational and layout schema through exposure to a particular ‘patch’ of a website (this seems to correspond well to the ‘hub and spoke’ navigation pattern identified by Andrew Cockburn and Bruce McKenzie).
2. Prediction of the structure and nature of content based on this habituation when clicking on a link
3. Reorientation based on the content actually rendered

Danielson then identifies and develops metrics for three factors afecting this process: trasnitional volatility (the extent to which pages change in terms of navigation options and content), volatility habituation, and volatility predictability. As Danielson points out, habituation within one particular patch will give rise to a perception of enhanced volatility on leaving that patch, while effective prediction will lead to a perception of decreased volatility (and hence need for reorientation).

In D.’s experimental design, test subjects were asked to navigate through an ersatz site with 100 pages of content in three well-formed levels using three different navigational displays:

• Full Overview (with, essentially, a site map constantly displayed on every page)
• Partial Overview (hyperlinks to five top-level pages and all pages within subsite of current node)
• Local Context (links to top-level pages and all siblings and children of current node)

The Full Overview is used purely for purposes of comparison.

Among Danielson’s conclusions are: 

• disoriented users tend to return to the Home page to start all over again, rather than using horizontal subsection links (as I would have anticipated)
•  Local Context design was perceived as less volatile than Partial Overview design: local context navigation makes numerous small and incremental changes, as opposed to the occasional and sweeping character of partial overview models
• Navigational volatility increases the perceived size of a site, but not its complexity
• breadth-first exploration leads to a more accurate mental mapping of a site than approach in depth
• Unexpectedly, high navigational volatility in the local context model led to an increase in the perceived over-all coherence of a site. Danielson’s explanation of this is intriguing: extensive differences in navigational support between distal pages led users to discover connections for themselves (!)
• The page overview model did not cause users to underestimate actual page volatility through predictability. This may, however, have been the result of the extremely high level of predictability the overview model allows in many situations.

Kalbach’s summary of Danielson’s research goes beyond D.’s own conclusions, but seems relevant nonetheless: the small, localized changes afforded by the local context model serve not only to increase predictability and decrease perceived volatility, but also to give a sense of ‘flow’, movement, and progress through a site.

Filed under: Uncategorized   |  Leave a Comment