Information Technology Reference
In-Depth Information
keystrokes needed to perform tasks, TAM sought to predict user acceptance from measures captured
based on low- to medium-fidelity mockups of a system created before writing a single line of code.
A major software vendor (henceforth, XYZ Corporation) agreed to support my dissertation
research into developing and validating TAM. XYZ had withdrawn a recently introduced end-user
software product from the marketplace after a dismal launch. Although the product had success-
fully passed the state-of-the-art usability testing process that XYZ applied to all of its end-user
products, its poor reception in the marketplace caused them to realize that traditional usability test-
ing, by itself, was not enough to assure user acceptance. In the aftermath of a seven-figure finan-
cial loss related to this failed product, XYZ needed new metrics for evaluating its products that
could better gauge user acceptance. Moreover, they sought measures that could be applied early in
a product's development life cycle in order to determine if a product concept was sufficiently
promising to justify the full funding needed to bring it to market. XYZ had begun using videotape
mockups of software product functionality as a prototyping and design tool.
EMPIRICAL VALIDATION OF TAM
XYZ Corporation wanted my research to answer two questions. First, can validated and reliable
measures of perceived usefulness, perceived ease of use, and intention accurately predict workplace
usage of new systems? And, second, can videotape mockups be relied upon as substitutes for work-
ing prototypes of end-user systems that have not yet been developed? In 1984, following psycho-
metric measure development procedures, I used the definitions of perceived usefulness and ease of
use to develop and pretest questionnaire items, and conducted a survey to validate the scales and per-
form item analyses. The survey confirmed that the perceived usefulness and ease of use constructs
were not only conceptually distinct, but also empirically distinct (Davis, 1986). It also supported the
idea that usefulness was a stronger determinant of usage than ease of use (Davis, 1986).
XYZ Corporation created brief video mockups for two existing business graphics software prod-
ucts, demonstrating their functionality and showing examples of how they can be used. These video
mockups were designed to parallel the detail and realism of video mockups that XYZ was currently
using to evaluate concepts for new products. I conducted an experiment on forty executive MBA stu-
dents that measured their perceived usefulness, perceived ease of use, and usage intentions twice for
each system, once after they had viewed the videotape mockup, and again, several weeks later, after
they had had an hour-long hands-on demonstration of the software. As hypothesized, the video-
based measures of usefulness accurately predicted the hands-on-based measure of usefulness, but
the video-based ease of use measure did not predict its hands-on counterpart. Direct hands-on inter-
action with a product is generally needed to form an accurate perception of its ease of use, but non-
interactive video is sufficient to form an accurate perception of its usefulness. Also supported was
the hypothesis that usefulness would have a stronger influence on usage intentions than would ease
of use. Because usefulness was stable between the video and hands-on conditions, and it had a
stronger effect on intention, intention was also stable across the video and hands-on conditions
(Davis, 1986). This experiment provided evidence that TAM could provide realistic evaluations of
new software systems based on videotape mockups before a single line of program code is written.
XYZ Corporation began to use TAM routinely in its new product development processes.
EVOLUTION OF TAM
My dissertation (Davis, 1986) was defended in December 1985. Two journal articles that intro-
duced TAM to the literature were Davis (1989), which reported on the theoretical definition of
Search WWH ::
Custom Search