TASK-TECHNOLOGY FIT - Human-Computer Interaction and Management Information Systems: Foundations - page 183

Dynamic complexity : the degree to which relationships between task inputs and task prod-

ucts change over time.

For a collection of possible execution sequences that contain the same types of action steps, the
computational complexity of each could be estimated using the framework above. Defined in this
way, complexity is an objective characteristic of an execution sequence, independent of the task
doer's characteristics.
The construct of difficulty suggested by Campbell (1988) also allows us to account for individual
factors. Campbell proposed a useful distinction between task complexity and task difficulty , in which
difficulty captures the commonsense observation that the same execution sequence may appear hard
for some and easy for others because of their different levels of skills, abilities, and experience.
All this suggests a new way of thinking about task-technology fit. When considering applying
a new technology to an existing underlying task, we need to consider the execution sequences cur-
rently being used, and the new execution sequences that will be enabled by the new technology.
These new execution sequences can be characterized by their impact on the quality of the ultimate
outcome, as well as by the cost born by the task doer in terms of complexity and difficulty. Task
doers presumably choose an execution sequence based on their perceptions and weightings of
these various factors.
As far as predicting performance impacts (as opposed to predicting use), a new technology has
higher task-technology fit than the old technology when it enables an execution sequence that has
some combination of lower complexity, lower difficulty, and an a priori prediction of higher out-
put quality. The “ a priori prediction of higher output quality” deserves a few word of explanation.
In other words, it must be possible to recognize some difference in the execution sequence that
logically ought to lead to higher output quality. It is important to stress the fact that one cannot
define high TTF by an a posteriori experienced higher output quality—that would in effect define
TTF as that which leads to higher quality output, creating a tautology.
Clearly, changing the weights attached to the three characteristics ( a priori quality, complex-
ity, difficulty) can change the determination of which execution sequence has the highest task-
technology fit. This might seem like a disadvantage of looking at task-technology fit in this way,
but in truth it is far more realistic and accurate. It is not possible to assess task-technology fit
without taking into account the relative importance of output quality and effort. Altogether this
moves us to a much more concrete conceptualization of task-technology fit than terms such as
correspondence, congruence, or cognitive fit.

SUMMARY

The key concept of task-technology fit is that the value of a technology with a particular func-
tionality depends on how much users need that functionality in the tasks they are doing. Task-
technology fit is an important construct in the chain from technology to performance that is often
ignored by MIS researchers and practitioners. This paper describes a model that includes two
important but very different ways in which TTF can affect performance. First, by affecting user
beliefs about the consequences of using a technology, TTF can change the likelihood that a tech-
nology will be used. Secondly, regardless of the reasons the technology is used, at any level of use
(beyond no use!) greater TTF will deliver more performance impacts. This model is quite differ-
ent from the most cited MIS models of technology impact on individual performance.
Finally, a new way of looking at task-technology fit suggests that technologies enable different
execution sequences for carrying out an underlying task. Each execution sequence has a certain