Lesson 2

Designing a
Survey
 Objectives

? AA
Identify the Design a survey
qualities of a questionnaire.
good survey
questionnaire.
 What is a Survey?
A survey is a method of gathering
information from a group of people,
typically using questionnaires, to
understand their opinions, behaviors, or
characteristics.
 Qualities of a Good
Questionnaire/Survey
valid
reliable
interesting
succinct
 Valid
A valid questionnaire should ask for precisely what
it intends to learn. In other words, the questions
should be written in a way that the respondent will
understand what the objective of the question is.
To do this, the questionnaire should be reviewed
by an expert on the topic during a pilot test. For
example, if the target respondent is a person with
disability (PWD), then you should check with a
PWD if the questions make sense and are
understandable. Any clarifications should be
made until the question is clearly understood.
 Reliable
A reliable questionnaire should yield generally
the same answer if the same question is
repeatedly posed to the same respondent
within a short amount of time. This can be
achieved by performing retests, meaning you
administer the same questionnaire to a
respondent more than once and check the
consistency of his or her answers. If there are
any discrepancies in the answers, it could be
because the questions are unclear or the
instructions are vague. If so, these should be
reviewed and revised.
 Interesting
An interesting questionnaire is
more likely to be completed by the
respondent, thus providing a higher
response rate. This means that the
researcher should put some
thought into making sure that the
questions are relevant to the
respondent and are arranged in a
logical sequence.
 Succinct
A succinct questionnaire asks questions that are
straight to the point and only answer the research
objectives. Any questions that are beyond the scope
of the research should be excluded. Some
researchers tend to “cast the net wider” by asking
more questions so that they can collect more data,
regardless of whether the data is directly relevant to
their research or not. However, this usually
happens when the researcher has not yet properly
fleshed out his or her objectives. Be careful in this
case, because it runs the risk of asking too many
questions and the questionnaire becomes too
lengthy and difficult to complete.
 Dependent and Independent Factors

setting the objectives of your survey is always the first thing

you do when you are preparing a survey questionnaire. Once
you have your objectives, you also have to make sure that
these are aligned with your research question. You must
also take note of the “dependent” and “independent”
factors you intend to investigate.
For example, look at this research question: “What are the biggest distractions that
prevent students in the senior high school from studying at home?” You should begin
with initial background research to get information on this topic, like what teachers and
school administrations have said about homework and studying at home. Once you have
done your research, identify the things that you can learn via survey, and these will be
your objectives. To continue from the example, you can ask in the survey what students
do at home that prevent them from studying, and how much time they are able to spend
actually studying during their time at home. If you are able to ask a high percentage of
the student population in your school, that could yield a pattern that you can analyze.
The independent factor here are the forms of distractions available to students, like
smartphones or video game consoles, because this is what changes. The dependent
variable is the amount of time the students spend studying, as this is what is affected by
the distractions. The importance of establishing these pieces of information is to ensure
that the research covers all relevant variables and any irrelevant variables are excluded.
This will answer the commonly asked questions, like: “Did I miss any important
questions in the questionnaire?” or “Should I include/exclude this particular question?”
 Asking the “Right”
Questions
Once you have established your
question, objectives, and factors, it
is time for you to design the
questions in such a way that it is
valid and reliable. The researchers
have to brainstorm and come up
with the preliminary questions.
 Structured or Non-
structured Questions
Decide using your objectives and
factors if you will be using structured
questions, non-structured questions,
or a combination of both. To refresh
your memory, structured questions
offer a set list of items, while non-
structured questions are open-ended
and allow respondents to write their
own answers. The answers to open-
ended questions should be grouped
together for analysis.
Remember Choose structured
questions if you are already
confident that your range of
answers is well-known and you
would like to focus on those
answers. Choose non-structured
questions if you are still unsure
of the information you have and
would still like to explore more
options through the survey.
 Options
When using structured questions, the options given to the
respondents for each question should be as exhaustive as possible.
This will ensure that they can pick an option that best suits their
answer. Preliminary research and analysis is important to create a list
of options that can grasp every aspect of the question. To allow other
possible options, the researcher can include “Other: please specify
________” as one of the options.When asking a factual question
instead of a subjective question, it is important to include “Don’t
know” as one of the answers, as not all respondents may know the
answer to the question. If you do not provide this option, you are
forcing respondents to answer by taking a guess, which would not be
an accurate representation of what the respondent knows.
 Filtering
Some questions in a survey questionnaire do not have
to be answered by the respondents, depending on
the information they provided or option they chose in
a previous question. In these cases, filtering is used to
guide the respondents to answer only the questions
that are relevant to them. However, you should avoid
filtering too much as this may confuse the
respondents and make the questionnaire too
complicated.
Example 1 Here is an example of a question
that is followed by filtering:
1. During the last time that you had a fever,
did you consult a doctor?
[ ] Yes
[ ] No

If No, proceed to question 4.

 Order of
Questions
The order of the questions should follow a logical
sequence. A common rule is to start with simple
questions before moving on to more complicated
ones. Generally, you can start with the socio-
demography of the respondents, but if you will
include more personal questions such as
household income, education level, and religion,
then you can opt to leave these questions for last.
Sometimes, it helps by explaining to the
respondent what your reason is for asking a
personal question or by making a general
statement to normalize the “sensitive” question.
 Double-barrelled
Questions
One mistake to avoid is asking a double-
barrelled question. Avoid asking two things in
one question. This will make it more difficult for
you to interpret and analyze the data you have
collected later on.
Example 2 Here is an example of a double-barrelled
question that you should avoid:

Have you ever stopped studying to check your phone or

play a video game?
[ ] Yes

[ ] No

Instead of this question, it would be better to divide it into

two separate questions.
Have you ever stopped Have you ever stopped
studying to check your studying to play a video
phone? game?

[ ] Yes [ ] Yes

[ ] No [ ] No
 Ambiguous Questions
Avoid ambiguous questions and be as specific as
possible when asking a question. For example,
terms such as “seldom,” “sometimes,” and “often”
may mean different things to different people.
Keep questionnaire items short, preferably less
than 20 words. When you are reviewing your
questions, ask yourself, “Is this question clear? Can
it be more specific?”
Example 3 Here is an example of an ambiguous question
that should also be avoided:
How often do you visit a doctor when you have a fever?

[ ] Not at all
[ ] Not often
[ ] Sometimes
[ ] Quite often
[ ] Very often

Instead, give a question like this more specific options

How often did you visit a doctor when
you had a fever in the past year?
[ ] Never
[ ] 1 time
[ ] 2–3 times
[ ] 4–5 times
[ ] More than 6 times
 Keep Analysis in
Mind
When preparing your survey questionnaire, always remember
that the data you collect here will have to be analyzed
afterward. For example, let’s say one of your factors for your
research is age. If your objective for asking the age is to find
out the average age of your respondents, then your question
should ask the exact age (e.g. “What is your current age?”).
On the other hand, if part of your objective is to categorize your
respondents by age group, then you can use the same
question but add options that give different age groups (e.g.
<18, 18-29, 30-50, etc.).
 Format
The final touch for your questionnaire is to fix its format, and this is
important because the appearance of the questionnaire may decide
whether or not the respondent is going to finish filling it up. This is
especially relevant for postal surveys. The title should be
highlighted and it should give a hint to the main objective of the
research. If possible, divide the questionnaire into sections
according to the content (e.g. boxes with bold headings) and it
should flow smoothly from one section to another with filtering if
necessary. Finally, prepare and include a cover letter stating the
objective of your study, affiliations, and how you are going to use the
information that you will gather though the survey. If appropriate,
ensure confidentiality.
 Pilot Test
Holding a pilot test is a crucial step in the design of a questionnaire
before you start using it to collect data. This can help find any
flaws in the questionnaire in terms of content, grammar, and
format. First, ask your teachers, classmates, family, or friends to
comment on the questionnaire. This will pick up any mistakes in
terms of content, grammar, and format. If possible, this should
then be followed by asking a number of potential respondents to
answer the questionnaire and provide their feedback. For those
questions that you feel may be confusing or sensitive, it is
important to ask the respondents to comment specifically during
the pilot test.
 Wrap Up
A good questionnaire is valid, reliable,
clear, interesting, and succinct.
The research question, survey
objectives, and independent and
dependent factors should be clarified
before designing a survey.
Consider testing the validity and
reliability of your questionnaire by
conducting pilot tests and asking for
specific feedback on grammar, content,
and format.