Cognition: Human–Computer Communication - Foundations for Designing User-Centered Systems - page 212

When developing the IA, it is important to consider the information at several
levels of abstraction. At the lowest level, this may be as simple as the way that
items are ordered on a page. At the highest level it could involve how to organize
the content appropriately to support the user's tasks. At a strategic level it could
involve making decisions about how articles and metadata are placed into a
content management system.
7.5.3 Creating Content
Getting the content strategy and information architecture correct are both impor-
tant, but it is equally important to make sure that your content is usable too.
Writing for printed material is a skill in itself, but creating content for web pages is
also an important skill in its own right and one which often involves more than just
writing text. The way that users read web pages is usually not the same as the way
that they read printed pages (unless they are reading an online text article from
start to finish).
In most cases, web content is written for a general-purpose audience, which
means that it should be readable and understandable by a large proportion of the
population. The usual guide rule is that you should try to write text that requires a
reading age of around 12 years, and that can be understood by students in the
eighth grade in the US (this equates to an age range of about 12-14 years). The
readability of text can be assessed using the Flesch reading ease score (Flesch
1948 ): generally, you should be aiming for a score of 60-70. Comprehensibility is
assessed using the Flesch-Kincaid grade level score (Kincaid et al. 1975 ), and here
you should generally be aiming for a score of 7.0-8.0. Many word processors
include tools that will let you calculate both of these scores, and there are several
online calculators that can be used too.
These scores should be used as guidance, not as requirements that have to be
pursued at all costs. The scores provide a relatively simple but useful assessment
of the clarity and directness of the language that is used in the text.
7.5.4 Structuring Content
We know that users organize and associate concepts using something like a graph
structure (e.g., Collins and Quillian 1969 ; Klahr, Chase, and Lovelace 1983 ;
Woods 1975 ). This led to the suggestion that knowledge should therefore be
presented to users as a hypergraph structure because if it goes into the eyes like a
graph (like hypertext or the web) then it can simply be put into corresponding
graph structure inside the user's head. This example of a homeopathic fallacy—
that ''like causes like''—may have arisen from a misinterpretation of Vannevar
Bush's ( 1945 , reprinted in numerous places) view of how information is organized
by people and machines.