4. Meet the interviewee in his/her typical
environment, so you can see them in context.
In-context interviews give the participant greater
ease and allow you to see the objects, spaces, and
people that they talk about in the interview.
6. Ideally no more than three people should attend
any single interview so as to not overwhelm the
interviewee(s).
7.
8. Conversations should be conducted without an
audience, since the presence of neighbors, friends,
or others can sway what the person says or what
they are able to reveal. Privacy can often be
difficult to accomplish, however.
However, sometimes a second person helps the
interviewee be more confident.
9.
10. Guidelines for the interviewer
•
Enter the interview without an agenda
•
Refrain from passing judgments
•
Ask the interviewee for permission to ask certain
questions
•
Ask for clarification when in doubt
•
Let them do the thinking – do not lead them to the
solution
•
Be aware of both yours and their body language
•
Listen actively; allow for silent pauses & let them think
•
Remind the interviewee that there is no “right” answer
11. ME
THOD
ME
THOD
Int erview Prep
arat ion
Int erview f or E p t hy
ma
E
xplore
E otions
m
E
voke
S
tories
Intro
Y
ourself
Intro
P
roject
Build
R
apport
Question
S
tatem
ents
Thank &
W
rap-up
WHY interview
WHY prepare for an interview
W want to understand a person’s thoughts, em
e
otions, and m
otivations, so that we can determ how to
ine
innovate for himor her. By understanding the choices that person m
akes and the behaviors that person
engages in, we can identify their needs and design for those needs.
Tim with users is precious, we need to m the m of it! While we always m allow roomfor the
e
ake
ost
ust
spontaneous, blissful serendipity of a user-guided conversation, we should never abdicate our responsibility
to prepare for interviews. Especially in following up with users (after testing, etc.), it is im
perative to plan
your interviews. Y m not get to every question you prepare, but you should com in with a plan for
ou ay
e
engagem
ent.
HOW to interview
HOW to prepare for an interview
B
rainstormques
tions
W down all of the potential questions your teamcan generate. T to build on one another’s ideas in
rite
ry
order to flesh out m
eaningful subject areas.
Identify and order them
es
S ilar to “grouping” in synthesis, have your teamidentify them or subject areas into which m
im
es
ost
questions fall; once you’ve identified the them of your question-pool, determ the order that would
es
ine
allow the conversation to flow m naturally. This will enable you to structure the flow of your interview
ost
,
decreasing the potential for hosting a seem
ingly-scattershot interaction with your user.
R
efine ques
tions
Once you have all the questions grouped by them and order, you m find that there are som redundant
e
ay
e
areas of conversation, or questions that seemstrangely out of place. T a few m ents to m sure that
ake
om
ake
you leave roomin your planning to ask plenty of “why?” questions, plenty of “tell m about the last tim you
e
e
___ questions, and plenty of questions that are directed at how the user FE LS
__?”
E .
:: 9 ::
As why E when you think you know the answer, ask people why they do or say things. The answers
k
. ven
will som
etim surprise you. A conversation started fromone question should go on as long as it needs to.
es
Never s “us
ay ually” when as
kinga ques
tion. Instead, ask about a specific instance or occurrence, such as
“tell m about the last tim you ___”
e
e
___
Encourage s
tories Whether or not the stories people tell are true, they reveal how they think about the
.
world. Ask questions that get people telling stories.
Look for incons
istencies S etim what people say and what they do are different. These
. om
es
inconsistencies often hide interesting insights.
Pay attention to nonverbal cues Be aware of body language and em
.
otions.
Don’t be afraid of s
ilence. Interviewers often feel the need to ask another question when there is a pause.
If you allow for silence, a person can reflect on what they’ve just said and m reveal som
ay
ething deeper.
Don’t s es answersto your ques
ugg t
tions E if they pause before answering, don’t help themby
. ven
suggesting an answer. This can unintentionally get people to say things that agree with your expectations.
As ques
k
tionsneutrally. “What do you think about buying gifts for your spouse?” is a better question than
“Don’t you think shopping is great?” because the first question doesn’t im that there is a right answer.
ply
Don’t as binary ques
k
tions Binary questions can be answered in a word; you want to host a conversation
.
built upon stories.
Only ten wordsto a ques
tion. Y user will get lost inside long questions.
our
Only as one ques
k
tion at a tim one pers at a tim R
e,
on
e. esist the urge to am
bush your user.
Make s you’re prepared to capture. Always interview in pairs. If this is not possible, you should use a
ure
voice recorder— is im
it
possible to engage a user and take detailed notes at the sam tim
e e.
:: 1 :: Visual adapted fromMichael Barry, P
0
oint Forward
12. Goals:
• Prepare enough questions for 1520 minute interview with at least
one person
• Test questions by asking them to
other team
• Create a plan for interviewing
13. When interviewing, assign the following
roles:
• 1-2 people to lead the interview(s)
• 2-3 note takers
• 2 photographers
• 2 “Storytellers” and quote-capturers
**Capture as much data as possible!
14. Interview Practice
Using the Interview Guidelines, students ask each other
about life before ALA.
Pair up and interview for 5 minutes each way.