Readability and Page Layout
Design standards that apply to page layout in the print medium do not necessarily apply to the Web.
Readability and the Web
The researchers also used traditional readability measures (Gunning Fog and Flesch-Kincaid), and found that ...
- Users experienced greater success with sites that were the least readable
- User impressions of sites that were less-readable were that they were more authoritative, clear, complete, satisfying, and useful than the more readable sites
- However, users also perceived less-readable sites as being "overwhelming" and "contains too much detail"
It's noted that less-readable sites contained fewer conjunctions and standard grammatical structures. For example, a list is interpreted as a compound sentence by the metrics. However, the advantage was that the layout of these sites was, by design, easier to scan than to read, which enabled users to find the information they sought more readily.
EN: The Fog/Flesch metrics were designed to assess text that is formatted only into paragraphs. Headers and lists, which are very common on the Web, foul the measurements, in that a header is interpreted as a short sentence (when it is not a sentence at all, merely a "tab" for reference) and a list is interpreted as a long compound sentence (when it would be just as valid to interpret it as a series of short sentences). Both are ill-suited to the Web.
The White-Space Dilemma
Regarding white-space: there was no indication that the use of white space impacted success on the Web. When testing a page with a tight design with the same information laid out with greater white-space, users were less successful with the latter. They also perceived sites with greater white space as being harder to use, harder to read, difficult to scan, and less visually appealing.
EN: This is also questionable: how did the study define "white space"? It's been found that too wide a column of text confounds readers, on screen more so than on paper, so wide margins are inevitable (and filling them with junk, just to fill the empty space, has been found to be distracting). Since the concept is vague, it's difficult to derive practical information from this finding.
Scrolling and "the Fold"
Conventional wisdom dictates that users "don't like" to scroll, and that good design means placing all information above the "fold" (a term borrowed fom newspaper layout).
In their testing, users made no objection to scrolling, and there were no measurable effects (such as users bailing on a task that required scrolling) that would be evidence that this were true.
However, it was found that horizontal rules on a page (a rule that extends the length of the page) were interpreted by users as demarking the end of a page. A significant number of users stopped scrolling when a screen-width HR appeared at the bottom of their screen, even if the scroll-bar indicated there was more content beneath it. It is believed that this is because many sites use an HR at the bottom, beneath which are disclaimers and copyright information that is meaningless to most users.
A later study: users preferred scrolling through content on a long page as opposed to clicking through several short screens to access the same information.
Button Gravity
One phenomenon noticed by the researchers was that, when a page contains multiple HTML forms, users may fill out the information in the topmost form, and then click the bottom-most button on the page to submit it. This happens even more frequently when forms are stacked directly upon one another in a vertical row, with now visual separation in between.