Database Reference
In-Depth Information
form one long page instead of the current ive-page “wizard.” The assumption was
that “wizards” helped keep folks on track and helped them understand where they
were in the process. In addition, the Web is awash in “wizard frenzy;” every Web
shop in town is building some kind of “wizard,” whether it's iguring out your inter-
est rate on your dream house or determining your risk of lung cancer. The “wizards”
work…but, at what cost in total completion time?
Creative Director Andy Moodboard suggests to your designers that they come up
with a one-page prototype of the job posting page. “Kill the ive-page wizard, make
it one page, don't worry about how long it scrolls…let's just see if it's faster. Let's
get the comparison numbers to Hans quick!”
Realizing that “quick” means a standard usability session of 2 days with 10
participants (Moodboard would never give you 4 days to run a test!), you opt to
use the hour of each session to time “virgin” participants as they ill out both the
current form (the one with the ive-page “wizard”) and the new “long scroller,”
one after another. Of course, you'll counterbalance to ameliorate bias. You'll
average the timing for each design and see if the new prototype speeds up the job
posting process.
One group, two designs: enter the world of paired-samples t-tests.
3.3 INTRODUCTION TO PAIRED SAMPLES
As opposed to the scenario in Chapter 2, the fundamental characteristic here is that
each person evaluates both designs. So, in essence, the satisfaction evaluation or the
completion times are “paired.” One person provides two data points and we know
which two came from a given person.
Again, whether we are comparing the mean satisfaction with the two designs or
the mean time it takes to perform a task with the two designs, there is no material
difference in the analysis. That is, the software does not know, nor care, nor can it
distinguish, whether the numbers are representing satisfaction scores, or whether the
numbers represent times to perform a task.
Now, it may not be obvious why it matters whether we have independent samples
or paired samples. While you can rest assured that it does matter (and the software
does the right—but a bit different—process when told whether the samples are inde-
pendent or paired), we will address this issue in more detail later in the chapter to
give you a better sense of why it matters.
3.4 EXAMPLE OF PAIRED (TWO-SAMPLE) T-TEST
When we are comparing two means and the data have the same people providing
measurements of both alternatives (the satisfaction rating of two designs, or the time
for the performance of some task for two designs), we have what are referred to as
“paired data.” The term makes sense, in that a data point from one alternative is
 
Search WWH ::




Custom Search