finds Web quality time
Technology Research News
Ever swear at a computer because the two
of you were not communicating very well?
The communications interface between computer and human has always been
a sore spot, at least for humans. The increasing use of the Web,
which has shifted human-computer communications out of the work realm
and into our leisure time as well, makes it even more apparent that interfaces
Researchers from the U.S. Air Force have published a study that takes
some initial steps in quantifying exactly what Web applications need to
do to provide users with a better experience. The research could help
developers keep to a minimum those experiences that lead to swearing at
The difficulty of judging the quality of a Web experience is that the
possibilities are so broad, according to Jason Turner, a computer network
countermeasures engineer at the Air Force Information Warfare Center at
the Kelly Air Force Base.
"Unlike a traditional information technology system or network that might
be designed or acquired for a specific function [like] banking, office
automation [or] academic study, the Internet is simply a collection of
connections with little task specificity," he said. "Everyone's experience
with the Internet is undoubtedly influenced by many factors: browser,
means of access, reason for use, just to name a few."
The researchers drew on established theories of human-computer interaction
and information technology to conduct their study. The human-computer
interaction theory they used defines ease-of-use as a function of three
factors, said Turner. They are "utility -- whether the system does what
is needed functionally; usability -- whether the users can actually work
with the system successfully; and likability -- whether the users feel
the system is suitable," he said. These three factors are all balanced
against cost, which includes capital and operating expenses as well as
social consequences, he said.
The researchers also took into account regret theory from behavioral science,
which places importance on factors that prevent us from achieving our
goals, especially after discomfort arises from a failure to do so.
The researchers surveyed Internet users in order to find what factors
mattered to them the most in terms of having a good experience on the
Internet. In the study, 148 users of varying ages and Internet experience
levels identified key circumstances or issues that they associated with
the best and worst experiences they had with the Internet.
The researchers found that users' perceptions of and feelings towards
Internet use were predominantly determined by a relatively small, seemingly
stable and apparently consistent set of conditions, said Turner. These
ten critical quality-of-experience items "may typify the events or factors
which influence the relative success with which the sample of Internet
users were able to accomplish their goals," he said.
The critical factors were the relevance of information to a task, the
availability of search engine options like multiple criteria, the amount
of information available, the organization of information, the ease-of-use
of the browser, the reliability of the connection, the clarity of directions
for Web site navigation, whether links were up-to-date, the availability
of information at remote servers, and perceptions of security and/or privacy.
How efficient we believe our Internet experiences will be "center[s] around
our assessment of how successful we think we will be at using the Internet
to accomplish our goals," said Turner.
The researchers also found that several factors which are often thought
to enhance Web experiences fade from view if the interface as a whole
doesn't provide a quality experience, said Turner. "Heavy graphics, sound,
animation, et cetera might not be so important to the user if the interface
doesn't first and foremost help the user achieve his or her goals by providing
for those factors which relate to a high-quality usage experience," said
The researchers' study also suggested that the critical factors do not
vary as users gain experience on the systems, said Turner. This means
that investment in making systems more usable would apply to both novice
and expert users.
Interface developers could probably save time and money by focusing efforts
on features that enhance the quality of the interface experience and on
those aspects of the interface that helped users attain their goals, said
Turner. "It might help to ensure we design ... from the start those ...
functions which will ensure [that] the network or application itself is
not abandoned or underutilized, regardless of what task or role that network
or application will eventually play," he said.
Studies that quantify in some way the quality of users' experiences are
badly needed, said Albert Badre, a computer science professor at the Georgia
Institute of Technology.
One of the major problems is those studies often focus on usability to
the exclusion of user satisfaction, he said. They measure things like
the time it takes to perform a task, and make an assumption that "usability
in terms of performance -- time performance, number of letters and so
on -- correlates positively with users' pleasantness of experience," said
Badre. "That is, in my opinion, a wrong assumption," he said.
In order to test that assumption, "we really need to be able to answer
questions like did [users] perform tests the way they expected, were they
successful at achieving their goals, and how do they feel as they were
performing -- do they have a nice life, so to speak," he said.
For instance, "if you provide me with an aesthetically appealing Web site
that is exactly the same thing as one that's very dry I will probably
enjoy looking at the nice colors and nice pictures even though it might
make it a little less efficient because the time to... download graphics
is longer," he said.
Usability studies seldom directly measure the user's comfort, preferences,
and pleasantness of experience in using any kind of technology, said Badre.
This is because before the Web, "the main preoccupation of usability people
was business software. And there you do want efficiency because that's
what you focus on, productivity," he said.
But productivity is not the main goal when you're sitting at home trying
to enjoy yourself surfing the Web, Badre said. "We did a study of that
specific issue and [found] that when people have a different purpose they
have a different idea of what they'd like to experience. I'm not saying
that we don't enjoy good efficient interaction, but it could be there's
more to it than simply that. And this is the kind of thing we need to
The Air Force researchers are "trying to get at quantifying that quality
of experience. The objective... is a good one," Badre said.
The research is mainly meant to be a springboard for follow-on research,
said Turner. The findings could be used to develop a practical application
in two or three years, he said.
Turner's research colleague was Michael Morris, formerly of the Air Force
Institute of Technology and now that the University of Virginia. They
published the research in the June 1, 2001 issue of the International
Journal of Human Computer Studies. The research was funded by the researchers.
Timeline: 2-3 years
TRN Categories: Human-Computer Interaction
Story Type: News
Related Elements: Technical paper, "Assessing Users' Subjective
Quality of Experience with the World Wide Web: an Exploratory Examination
of Temporal Changes in Technology Acceptance," International Journal of
Human Computer Studies, June 1, 2001.
Statistics sniff out
Quantum bit withstands
Image search sorts by
Study finds Web quality
Powerless memory gains
Research News Roundup
Research Watch blog
View from the High Ground Q&A
How It Works
News | Blog
Buy an ad link