Setting Testing Goals and Parameters
So now that we know we want to perform a usability test, we know
how we are going to record the users actions for better analysis
and presentation, what next? The next step is to establish the
testing goals and parameters. Some specific areas to be aware
of include: Goals to accomplish, audience, location, length,
and frequency.
The most important issue to nail down is your test goals. What
do you wish to accomplish. Are you looking for broad feedback?
Or are you looking to test one specific portion of your
application like a new system for navigation. One trap to avoid
is trying to test too many things at one time. Sometimes it may
make sense to break up one big usability test into multiple more
focused tests.
Who is your audience? Do you need to set a quota of different
user categories such as age, sex, and profession so that your
test will represent a fair snapshot of your audience? Where can
you find this type of person? Will you pay your testers for
their time? The system that worked best for me was to first
establish quotas for the type of users I needed in the test. I
then could focus on locations that my target audience
frequented. I would then offer random people "$20.00 for 20
minutes of their time to help test a new website and give their
feedback". A couple of things to note. 1) Some may scoff at
having to pay people for their time. Perhaps you can find users
willing to take your tests for free, or perhaps you can offer
non-monetary compensation like a t-shirt or other company swag.
It just seems that a little "something" helps grease the wheels
and makes finding users go much much faster. All things
considered, a couple hundred dollars in testing fees is small
potatoes compared to more formal usability tests that may
require paying finders fees and hourly fees for test users, let
alone the cost of not fixing your application before it goes
live.
What is the target location of your test? Universities are a
gold mine. What student wouldn’t do anything for $20? However,
universities will normally limit you to a set age group. A
large shopping mall is a great location to find a wide sampling
of people, but if you need to focus on a specific audience then
you may need to find a more specialized location.
How long will the test last? Most people do not have very long
attention spans. If the test is going to take more then 15 to
30 minutes, you many want to break it up into multiple more
specialized tests, or look into a more formal test setting.
Frequency. How many tests will you give? To keep it simple I
would recommend at least 2 or 3 sets of tests for each project.
The idea being, to test, adjust based on results, and then
retest. Be sure to leave enough time in between tests to act on
your feedback, and actually work the changes into the project.
You may also want to consider performing your first test as
early as your working user interface mock-ups (on paper or
computerized). That way you can collect feedback before too
much actual work has already gone into the application.
New on the Java Boutique:
New Review:
Time Management Made Easy with the Quartz Enterprise Job Scheduler
Why not just use the Java timer API? This open source scheduling
API boasts simplicity, ease-of-integration, a well-rounded feature
set, and it's free!
New Applet:
Reverse Complement
Reverse Complement is a simple applet that converts DNA or RNA
sequences into three useful formats.
Elsewhere on internet.com:
WebDeveloper Java
Lots of Java information on webdeveloper.com
WDVL Java
Thorough Java resource at the Web Developer's Virtual Library.
ScriptSearch Java
Hundreds of free Java code files to download.
jGuru: Your View of the Java Universe
Customizable portal with online training, FAQs, regular news updates, and tutorials.
