Methodology I: Task Analysis - Foundations for Designing User-Centered Systems - page 320

understanding of the general context in relation to work and activities. Methods
like activity theory address these issues (e.g., see Bertelsen and Bødker 2003 for an
introductory review). Activity theory is a very high level method that analyses the
types of user activities, generally using a very informal representation. It uses these
descriptions to suggest how to design the context in which the activities take place.
For example, identifying children's activities (play activities, painting, block
stacking, and doll-house play, for example), would inform the design of a kin-
dergarten room. Similar analyses make strong suggestions about the design of
research labs (collaboration spaces, mailing rooms, teaching rooms) and interfaces
where goals are less important than activities such as drawing, designing, inter-
acting, team-building, exploring, or enjoying.
Most TA methods do not note and are not sensitive to context, including
physical context such as lighting and heating, social norms such as doing what
helps others in the environment, and not using others space or resources. It is
especially common in work places to discover that the physical or social context
produces changes in the way people do tasks from how they would ideally do
them, or from the way they are supposed to do them. Casey ( 1998 , 2006 ) provides
numerous examples where not understanding the context on task analyses and on
design led to problems, such as where other people are when the user is doing a
task (not in the path of the rocket!), or the effect of weather on how to do tasks and
which tasks to do on a ship.
11.8 Summary
Task analysis in all its various guises is a very useful technique, and usually fairly
easy to perform when designing systems, which may help explain why it is so
widely used. The results of a task analysis can be used to help designers create
easier to use interfaces by highlighting the structure of tasks, and the time it takes
to perform tasks, as well as making it easier to compare alternative methods for
carrying out those tasks. Task analysis can also be used to represent the trade-offs
in designs, helping designers to make informed decisions. TA can also be used
directly to improve interfaces, and Table 11.7 notes a few common ways in which
TA can help improve interfaces.
There are many different uses for task analysis, and the method that you choose
will depend to some extent on how you intend to use the results. Choosing any one
is better than none, that you should choose one that is easy for you to use because
you are a user as well. The choice will also be affected by the context in which you
will use the method, the limitations, and possibly the experience of the analyst(s)
(several of the methods work at similar levels of abstraction, so they can be used
somewhat interchangeably).
Using task analyses to guide system design has a good return on investment
(RoI), ranging from 7 to 20 to 100 in one survey (Booher and Minninger 2003 )to
1,000 in another more focused HCI study (Nielsen and Phillips 1993 ). Given the