Database Reference
In-Depth Information
train trip with a friend. Gazing at the passing scenery, the friend remarks, "Look, that flock of
sheep has just been sheared." The scientist scrutinizes the flock, and cautiously assents, "So
they appear to have, been... on this side."
Taking Notes
There is no one right way to take notes—each observer will have his or her own conventions and
shorthand. This is fine unless you plan to read each other's notes in their original format, in which case
it can be confusing unless you know what each other's conventions are.
Table 11.1
shows some notes
I took from a usability test of a travel Web site. Other than spell checking and some minor editing for
clarity, these are typical of the notes I take. As shown in the Discussion column, subtle wording can
mask some of the differences between observations and inferences, and further underscores the risks
of interpreting someone else's notes.
Table 11.1: Example Test Observations from a Travel Web Site
Observation (from my notes)
Discussion
Next trip he's taking? Going to Chicago in
September for a wedding in Barrington. A friend
had told him a particular place—Amerisuites.
What other factors would he consider? Says he
prefers a pool. Wants a restaurant since he's not
going to have a car. Shuttle service to the airport
would also be good. I ask about price: "I like to
be under $100, unless I'm on vacation."
This is the sort of speech I tend to para-
phrase—the factors the user is listing give
us some background on how he'll
approach this task, but in this case his
exact words aren't crucial.
Clicks NW suburbs. "It gave me a list of 43
properties, hotels in the northwest region of
Illinois. Since I know I'm going to the
Amerisuites, I'll probably do a search off the first
page." He backs up to try that approach.
Saying that the user backs up is an
observation; the reason why is an
inference. In this case the user's
subsequent actions happened to confirm
it, but before the fact it can only be an
inference.
Types barrington in city, Hotel
name—amerisuites. "That would be a problem if
I don't know how to spell it." He knows the exact
dates of his trip, enters them in the search.
The phrase "he knows the exact dates"
appears to be an inference. However,
earlier in the test the user had said when
his trip was, so in the larger context of the
test this is actually an observation. If I
were rewriting this for a report, I would
clarify by phrasing it as two observations:
"User had mentioned exact dates earlier;
he entered those dates in the search."
No matches found. Says he thinks he spelled
amerisuites wrong ... or maybe it's not in
Barrington. Gets rid of the hotel name and it
returns one property in Barrington.
If the user didn't say what he was
thinking, I might write "(Not sure of
spelling???)" to indicate that I'm making
an inference. If I had asked him, my notes
would say, "I ask about spelling."
Laptop versus Hand-Writing
I tell observers to take notes in whatever form is most comfortable for them, and most people bring a
notebook and pen. Some teams write their observations directly onto sticky notes, which as I'll explain
can give you a leg up on analyzing the results. I prefer to use a laptop for note taking because I type
quickly and can capture far more information. The notes I take by hand don't have as much detail, and
thus it's easier to blur the difference between observations and inferences.
One potential concern about a laptop is that it can be intimidating or distracting to users, but I haven't
found this to be the case. It helps if you:
Search WWH ::
Custom Search