|
|
|
Writing Questionnaires That Work
Being able to write a good questionnaire is essential if you’re going to
become involved in business development or conduct research for your
firm or for a client. You might already know the questions you want
answered, but getting the right answers in a useful format takes more
than just asking them as you would in a conversation with someone you
know.
Questionnaires that are administered by professional interviewers are
usually phrased very differently than those intended to be
self-administered. There’s also an art to the placement of questions –
some should go up front and others are better suited to the end of the
questionnaire.
The exact wording of each question must be carefully constructed, and as
a general rule questionnaires should be pared down to the shortest
possible length. Even the number of subject areas covered needs to be
carefully considered.
The first step in writing any questionnaire is to define precisely the
information needed and be committed to using the fewest questions to
obtain it. Questions off the topic or those that seek to find out
relatively unimportant information must be avoided. Here are some more
of the essentials of writing an effective questionnaire.
1. The Self-administered Questionnaire
It’s important to structure self-administered questionnaires logically.
They need to be kept simple and have each question flowing from the one
before it.
Start with an interesting question or two. This will get the subject
involved with the questionnaire and make them want to see what else is
going to be asked. Remember that the issues raised in one question can
influence how people think about subsequent questions.
It is best to begin each topic or section with a broad question and then
ask more specific questions relating to it. This will serve to introduce
a topic and then direct the subject’s thoughts to more details of that
area.
Self-administered questionnaires need to be worded as simply as
possible. Avoid long words or words with more than one meaning.
Questions that can be answered ‘yes’ or ‘no’ or by placing an ‘x’ in a
box are far more likely to be answered than those requiring a long
self-generated answer.
2. Questionnaires Given by an Interviewer
Although these can be more complex and involved than a self-administered
questionnaire, the subject should still be able to understand the
questions easily, without having to try to work out exactly what’s being
asked.
Although it’s not as important when an interviewer’s asking the
questions to have each question following from the one before it, you
should try to create a logical flow to the questions to make it easier
for both the subject and the interviewer.
Anticipate questions subjects may ask the interviewer and provide a set
of notes that can serve as a guide for them while they’re conducting the
interviews.
3. The Design of a Questionnaire
The design is particularly important for self-administered
questionnaires. Page layout should be kept simple with text size a
minimum of 10-point so it’s easily read by most people. The appearance
of the printed sheet(s) should be uncomplicated, with graphics only
where needed to assist the subject in answering the questions.
Whether self-administered or administered by an interviewer,
questionnaires should always include instructions to the subject no
matter how easy to complete you may think the questionnaire is. Keep
them simple and put them at the beginning so the subject doesn’t start
answering questions the wrong way.
Also, wait until the end of the questionnaire to tell the subject who is
asking the questions and why they’re being asked. Otherwise you’ll give
the subject a mental ‘target’ for the answers they give and this can
sway their answers towards a particular viewpoint or bias.
4. The Order of the Questions
A good questionnaire will have a flow to it that keeps the subject’s
focus on a single area and doesn’t require them to jump from one topic
to another. Start with the easiest-to-answer questions up front and
gradually get deeper into the topic. This also gives the subject a
feeling of confidence that they’ll be able to answer all the questions
without difficulty.
If some of your questions require a great deal of thought or analysis,
put those in the middle of the questionnaire. If some of the questions
are of an especially sensitive nature, leave those for the last so the
subject isn’t put off before the questionnaire can be completed.
5. The Way Questions are Worded
Keep each question a single question that will get a single answer. “Did
you like or dislike anything in particular that you ate?” and “Where did
you go for lunch and dinner on your holiday?” are really compound
questions with more than one possible answer. Rephrase them and break
each of them down into separate questions.
This principle also applies to a question like: “Do you want to go to
Greece or India on your next holiday?” Even though the answer might be
the same, this should be two separate questions when asked in the
questionnaire.
You want to structure each question in such a way that you get the
answers you need without leading the subject towards a particular
answer. For example, the questions “What did you like most about your
holiday?” and “What did you like least about your holiday?” lead the
subject in the direction of finding a positive and a negative about an
experience that may not have had a real positive or negative aspect for
them.
To remove the bias in the question, make it the subject’s option to come
up with a definitive answer only if they want to. “Is there anything you
feel you liked most about your holiday?” and “Is there anything you feel
you liked least about your holiday?” overcome the problem and give the
subject the option of answering “No” if it’s applicable.
Avoid open-ended questions. The results will vary from one subject to
another, and you’re likely to get fewer or unsatisfactory responses from
subjects who are less able to articulate their feelings.
As much as possible minimize the amount of recall needed to complete the
questionnaire. For example, most people don’t have perfect memory so
instead of asking for specific numbers ask for a range of numbers if
possible.
6. Always Pre-test your Questionnaire
It is better to identify any problems before you’ve sent out
questionnaires than to find out when you get them back. Test your
proposed questionnaire on a small number of people, and conduct your
test self-administered or interviewer-administered just as the final one
will be.
After they have completed the questionnaire, ask them if they had
problems answering any questions. Be sure that the understanding they
have of each question is the same as the one you want them to have, even
if their answers make it seem that there were no misunderstandings.
Some questionnaires give the impression that the author tried to ask
every conceivable question about the general topic of interest. Others
look like a committee designed them with each member asking questions
about whatever subject came to mind. Very long questionnaires populated
with questions that have little or no relationship to each other will be
the first ones rejected by their subjects. They create annoyance and
frustration and generally yield inadequate or useless results.
|
|
