The Politics of Paper Prototyping - Paper Prototyping: The Fast and Easy Way to Define and Refine User Interfaces - page 231

Figure 13.2: There were several possible reasons why users didn't click the Intro to Psych link,
and only one of them was due to the artificial look of the grayscale paper prototype.

Our observation was, "Users aren't clicking the link." Our premise based on that observation was "The
interface has a usability problem." In discussions with the development team, the question arose
whether this problem really would have happened on a computer screen. Several interesting points
arose about potential biases, some of which acted to strengthen our premise while others weakened it.

Possible Bias: Lack of color

Explanation Direction of

Effect

Magnitude

The paper prototype used screen shots printed in
grayscale, so although the link was underlined, it
didn't appear in blue as it would on a computer
screen. This constituted a potential bias if users
were less likely to notice a link that wasn't in the
standard browser blue.

Weakens the
premise that the
interface has a
problem.

Opinions differed, but
the team was in
agreement that the lack
of color may have had
some effect.

Possible Bias: Artificial data

Explanation Direction of

Effect

Magnitude

One of the developers wondered if users
weren't drawn to the Intro to Psych link
because they weren't psychology teachers
and this wasn't the name of their real course.
This was an excellent point. Artificial data
may affect the way users interact with the
interface. Thus, there was a potential bias
due to the artificial nature of the task.

Weakens the
premise that
the interface
has a problem.

Hard to say, but probably less
than other factors. (However,
the team recognized the value
of using subject matter that
matched what the users
taught in real life and decided
to do this in future tests.)

Possible Bias: Artificial motivation

Explanation Direction of Effect Magnitude

The users' motivation to figure out the
solution might have been stronger in real life
compared with that during a usability test. Or
the opposite argument—users might try
harder in a usability test than in real life
because they're paid to be there.

Because Pearson was
concerned about adoption
rates, the conservative
approach was to treat any
barrier to successful use as a
potential problem.

Probably not
as strong as
other factors.

Possible Bias: Visual design