What makes one interview flow smoothly and another
painful to carry out? What should you look for in a questionnaire?
I'm probably not the right person to write
this article as my experience of using questionnaires as an interviewer
is limited, and most of my experience comes from writing questionnaires
as a exec (the dreaded exec). So I'm sure that experienced interviewers
will have far more dos and don'ts.
The key areas for me are length, flow, language and relevance. The
three first ones are what every exec is taught. Shorter the better, flow
from the general to the specific, use plain and simple language, but in
my opinion the fourth, relevance, is the most important.
If, as a respondent, I'm asked questions that I find relevant and
interesting I'll happily voice my opinion and tell you how to right the
world, but make it dull and stodgy, or worse don't let me tell you what
I want to say and I'll feel cheated. So the first point for me is to do
your homework and consider the subject from the respondent's point of
view and the pilot the questionnaire to make sure it still meets the
respondent's point of view. I know of some instances where interviewers
happily spent 2 hours on a questionnaire because it was relevant and
interesting, and I know of one particular customer sat questionnaire
where 15 minutes has felt like hours because it never asked the
questions important to me.
Length is always given as the major component for keeping respondents
happy yet there is continual pressure to add a bit more here and there.
Most of the time this suggests a lack of clarity of what the
questionnaire is for - too many bits of interest and too little idea of
what is important. This flabbiness tends to show at interview time too,
often with a questionnaire jumping around like an omnibus survey.
A good flow help you get through the questions faster and means
capturing the line of thought of the respondent and asking the questions
in the best sequence. From an interviewing point of view, I always find
lots of filtered questions and complicated routing difficult. Is it
possible to reduce the amount filtering?
Language is always a challenge. Researchers can be quite ponderous in
the language they use, whereas clients seem to want to use the latest
marketing buzzwords. Plain and simple questions seem to work the best,
and if you're gathering factual information is precise wording that
important? I guess it is for attitudes, but even here you get so many
nots and negatives some of the less well educated respondents just get
tied up in knots.
For me, the best questionnaires are written clearly and simply by
someone who knows how customers think and who knows what they really
need to know.