Designing a QuestiOnnaire

Designing a questionnaire is not as simple a job as it looks at first sight. A marketing researcher
intending to collect primary data has to be extremely careful in deciding what information is to
be collected, how many questions are to be formulated, what should be their sequence, what
should be the wording of each question, and what should be the layout of the questionnaire. All
these aspects need considerable time and effort of the marketing researcher. If he is able to
develop a questionnaire suitable for his field investigation, he will find that his task of collecting
the data has become much easier than otherwise.
type of information to be Collected
While attempting to design a questionnaire, the marketing researcher has to first ask himself
what type of information he needs from the survey. He should seriously consider this question as
it will have considerable repercussion on the usefulness of the survey. For, if he omits to collect
informa- tion on some relevant and vital aspects of his survey, his research study is unlikely to
be useful. At the same time, if he collects information on some issues not directly relevant to his
study, he not only raises the total cost of the survey but also increases the time factor. This being
the case, the survey will take much more time than is really necessary. It will also lead to greater
inaccuracy as the respondent will have to answer many more questions than are strictly
necessary and he will, therefore, not be sufficiently careful in giving the exact answer. In either
case, the marketing researcher will be the loser. To avoid this situation, he should give serious
thought to the specific information to be sought. It will considerably facilitate him if he looks
ahead to the analysis stage so that he could enlighten himself on the type of tabulation as also the
statistical methods that are to be used.
Crisp4 has very lucidly explained the different types of information which are generally
sought in marketing research. The information could be one or more of the following types: (i)
facts (ii) quasi facts (iii) awareness, or penetration of information (iv) opinions (v) attitudes (vi)
future ac- tion plans and (vii) reasons.
Factual information is perhaps sought most frequently in marketing research. For example,
the question—Do you own a car? is intended to seek such factual information. In addition,
sometimes information collected belongs to the second category, namely, quasi facts. This
implies that the information received from respondents is not factually corrected though it may
appear to be so. Thus, a question in continuation of the earlier one about the ownership of a car,
could be—If yes, when did you buy the car? Here, the respondent may not remember on the spur
of the moment, the year when he bought the car. Apart from this, the inaccuracy may originate
on account of the desire on the part of the respondent to leave a favourable impression on the
interviewer. Thus, for example, if a respondent is asked—Do you eat fruit everyday?, he may
say ‘yes’ even when he does not, simply because he wants to impress upon the interviewer that
he belongs to a well-to-do family and can afford to spend money on fruit on a daily basis.
Regarding factual information, the interviewer has to ensure that the information sought is
available with the respondent and that he is willing to part with it. Questioning respondents
about the distant past is not good as they will not be able to give an accurate answer. Another
point worth noting is that the interviewer should ensure that the respondent has understood his
question correctly and that he in turn has understood the respondent’s answer correctly. This will
avoid miscommunication.
The third category of information sought relates to awareness. Sometimes, the marketing re-
searcher is interested to know whether the respondent is aware of the existence of a certain
product or brand. Such information is particularly sought by a firm soon after it has launched an
advertising campaign, to enable it to know if advertising has contributed to the awareness of the
respondent. Besides, information determining the increase in penetration could be sought
through marketing research. For example, the firm may take up a second study subsequently,
more or less on the same lines as its first study, to ascertain how much increase in the penetration
has taken place between the two dates.
4
Crip, Richard D., “Organisation of the Marketing Research Function” in Handbook of Marketing Research,
New York, McGraw-Hill Book Company, 1973.
 Information is often sought on the opinion of the respondent. He is specifically asked, What
view he holds on a particular subject? and he is free to opine. Similarly, information on the
respondent’s attitude on one or more subjects or things is sought. The distinction between
‘attitude’ and ‘opinion’ is not very clear “A commonly drawn distinction has been to view an
‘attitude’ as a predisposition to act in a certain way and an ‘opinion’ as a verbalisation of the
attitude.”5 However, the distinction between the two terms gets blurred when the terms are used
to predict what the respondent will do. In view of this, some authors have used the terms
‘attitude’ and ‘opinion’ interchangeably.
The question on opinion poses some difficulties for the researcher. To begin with, when an
an- swer to an opinion question is received from the respondent, the researcher is not sure
whether the respondent is well informed about the subject on which his opinion has been sought.
A respondent can give any opinion without knowing the subject at all. This is obviously a great
disadvantage and it is difficult for the researcher to verify. Another difficulty is to ascertain the
intensity of an opinion, though this can be taken care of by suitably phrasing the specific
question. We shall revert to this a little later. Another difficulty is that as opinions are many-
sided, the same respondent will give different answers based on different aspects such as social,
legal, moral, economic, etc. It will not be evident to the researcher which viewpoint the
respondent has taken. Attitudes are very important as they give an indication of the past, or the
likely future behaviour of the respondents. Studies on consumer behaviour are possible only
when information on attitude is adequately available. Chapter 9 is devoted to Measurement and
Scaling while chapter 10 is on Attitude Scales.
Occasionally, the marketing researcher wants to know what are the immediate plans of the
respondent in regard to a certain thing. For example, he may be asked whether he plans to buy a
car during the next six months. Such an information is normally a statement of intention of the
respondent. One does not know whether it will be implemented or not. Any information
collected under this category has to be used with great caution otherwise one is likely to arrive at
wrong conclusions.
Finally, the marketing researcher at times wants to know the reason for a particular choice or
action of the respondent. For example, the respondent owning a car, may be asked why he
bought that particular make. He is expected to give one or more reasons in support of his choice.
Here too, there is an apprehension that the respondent may not come out with genuine reasons.
It must be emphasised again that the marketing researcher should be clear about the nature of
the information to be collected. The above classification of the requisite information will
facilitate him in phrasing the questions in the right form.

types of Questions
The second important aspect in the designing of a questionnaire is to decide which types of
ques- tion are to be used. Questions can be classified in various ways. One way of classification
is as follows:
∑ Open-ended questions
∑ Dichotomous questions
∑ Multiple-choice questions.
5
Green, Paul E. and Donald S. Tull, Research for Marketing Decisions, Englewood Cliffis N.J., Prentice-Hall,
Inc., 1970, p. 118.
 An open-ended or simply ‘open’ or ‘free answer’ question gives the respondent complete
freedom to decide the form, length and detail of the answer. Open questions are preferred when
the researcher is interested in knowing what is uppermost in the mind of the respondent.
However, open questions pose certain problems. At the time of the actual interview, it becomes
difficult for the interviewer to note down the respondent’s answer verbatim. If the interviewer
has to take down the answer all by himself without any mechanical aid, he is quite likely to miss
some vital information contained in the respondent’s answer. Further, if several interviewers are
conducting interviews and each one is recording the answers to opinion questions according to
his understanding, and in his own way, then there is likely to be an element of bias in the
recorded answers. Another difficulty in respect of open questions is that it is extremely difficult
to compress lengthy answers in a meaningful man- ner. Such answers may be good qualitatively
but their quantification becomes extremely difficult, if not impossible. The dichotomous
question has only two answers in the form ‘yes’ or ‘no’, ‘true’ or ‘false’, ‘use’ or ‘do not use’,
etc. An example of a dichotomous question is:
Do you use tobacco in any way?
Yes No

There cannot be a third answer. However, in some cases, there may be a third answer which
may come from those who do not want to take a definite stand one way or the other. For
example, take the following question:
Do you like to watch movies?
Yes No

Neither like nor disike

The third alternative may be included so as to provide for those respondents who do not have
a positive preference or aversion to movies.
It may be pointed out that dichotomous questions are most convenient or least bothersome to
respondents, who have simply to indicate their choice from the two possible answers. As such,
these questions require the minimum possible time of the respondents. Also, answers to such
questions are easy to edit, tabulate and interpret.
In the case of multiple-choice questions, the respondent is offered two or more choices. The
marketing researcher exhausts all the possible choices and the respondent has to indicate which
one is applicable in his case. For example, the following is a multiple-choice question:
Which of the following brand/brands do you use for washing clothes? Rin , Det
, 501 Blur Bar , Super 777 Bar , Wheel , Bonus
, Swastic , Any other (please specify ).
Obviously, the respondent is likely to take more time to answer a multiple-choice question as
compared to a dichotomous one. Also, more time is required in the editing, tabulation and inter-
pretation of data.

Phrasing of the Questions

The next issue in the preparation of a questionnaire is how to phrase the questions. The way in
which a question is drafted is very important as a slightly suggestive wording would elicit a very
different answer from the respondent. Consider, for example, the following question:
Don’t you think that this is a sub-standard product?
 A question of this type would prompt respondents to answer in the affirmative. Many of them,
who do not have a definite opinion about the product, are likely to agree that it is of sub-standard
quality. However, if the above question is worded a little differently, the answer is likely to be
dif- ferent. Suppose this question is put as follows:
Do you think that this is a sub-standard product?
A question of this type is not a suggestive question. It is a straight forward question and
respon- dents are not likely to be prompted to say ‘yes’ as was the case in the earlier question.
It will be interesting to know that a little difference in wording makes a good deal of
difference in the answers. Payne 6 cites an example in this regard. He gives the following three
questions and the related affirmative responses:

Quesion Percentage of respondents saying ‘yes’

1 Do you think anything should be done to make it easier
82
for people to pay doctor’s or hospital bills?
2. Do you think anything could be done to make it easier for
77
people to pay doctor’s or hospital bills?
3. Do you think anything might be done to make it easier for
63
people to pay doctor’s or hospital bills?

These questions were put to three matched samples of respondents—one question to each
type. The only difference was in the use of the words ‘should’, ‘could’, and ‘might’. This change
in one word changed the percentage of affirmative responses by as much as 19 per cent, which
clearly proves that the phrasing of questions has to be done with great care so that there is no
room for any ambiguity.
In order to ensure the appropriate phrasing questions, one should be particular about the fol-
lowing factors:
1. Difficult words should be avoided as far as possible. Likewise, technical or special terms
which an average respondent may not understand, should be excluded.
2. Vague words such as ‘many’, ‘often’, ‘generally’, ‘on the whole’ and ‘reasonably’, should
not be used.
3. Lengthy questions should be avoided. Too much verbosity makes questions lengthy, and
is likely to confuse the respondent.
4. One should avoid combining two questions into one. For example, the question—Which
of the following modes of transport is cheaper and more convenient? (i) Train (ii) Bus
It is quite likely that according to the respondent both the attributes, namely cheapness
and convenience may not be applicable to either of the two modes of transport. One may
be cheaper while the other may be more convenient.
5. Questions lacking specificity should be avoided or modified suitably so that they become
more precise. For example, the question—Are you satisfied with your job?—is not
sufficiently specific because it does not provide the necessary frame of reference to the
respondent. One may be satisfied, say, from the viewpoint of emoluments but one may
not be satisfied with the type or nature of one’s work. These are two distinct aspects
which perhaps cannot be taken care of by one question.
6
Panyne, Stanley L., The Art of Asking Questions, Princeton, N.J., Princeton University Press, 1951, pp. 8–9.
Order of Questions
Another aspect that should receive the attention of the researcher is the sequence or order of
ques- tions to be contained in a questionnaire. Since, in the beginning, the researcher has to
establish some rapport with the respondent, it is necessary that questions asked at the beginning
are simple and thereby helpful in establishing the rapport. Difficult questions or those on
sensitive issues should be relegated to the end of the questionnaire. Further, questions of a
general type should be asked in the beginning while those which are specialised, needing some
in-depth information from the respondents, should be left to the end.
However, care should be taken to sustain the interest of the respondent until the last so that he
is able to answer specialised questions in a normal manner without fatigue and indifference. If
the questionnaire is very lengthy, two or three sets of the same can be printed where the order of
questions can be changed by a scheme of rotation for the sampled units. In such a case, some re-
spondents would answer the specialised questions towards the middle of the questionnaire
instead of towards its end.

how Many Questions to be asked?

The researcher has also to decide how many questions are to be asked. We may add that the
number of questions is not so important as the actual length of the questionnaire. We have just
mentioned above that the researcher has to sustain the interest of the respondent until the last
moment so that the interview can be completed successfully and the requisite information
obtained. Too lengthy a questionnaire would obviously be a disadvantage and the response to it
may be quite poor.
While deciding on the number of questions or the length of the questionnaire, the researcher
should put himself in the respondent’s shoes and imagine how he would react to that
questionnaire. He can calculate the probable time that might be required by the respondent in
answering the entire questionnaire. He can also canvass the questionnaire amongst some of his
friends and acquain- tances. Their opinion and reaction will be very helpful to him in finally
deciding how lengthy the questionnaire should be. This is, in a way, pre-testing of the
questionnaire which will be discussed towards the close of this chapter.

Layout of the Questionnaire

Finally, the researcher or someone on his behalf has to decide about the layout of the
questionnaire. This implies that the document should be set in such a way that it leaves a
favourable impression on the respondent. It should be neatly printed and the individual pages
should not have too many ques- tions so as to appear crowded. Proper spacing between the
questions and within a question should be provided for. The more important wordings to which
the researcher would like to draw the attention of the respondent, should be set in bold types or
underlined. If it is really a lengthy questionnaire, special care should be taken to reduce its size
by providing two columns in a page and by using finer types. But, this can be done up to a
certain point for too fine a print may cause inconvenience to the respondent. The questionnaire
should have ‘easy looks’ which means that it should be short and printed on superior quality
paper so that writing with pen or pencil is smooth.
Mail Questionnaire
So far the discussion was confined to the designing of questionnaires to be filled in by personal
interviews. In fact, the type of questionnaire to be designed depends on the type of survey.
Broadly, there are three types of survey namely, personal, mail and telephone. As far as the
telephone survey is concerned, it is not commonly used in India. As such, the personal interview
and mail survey are the only two methods. Since a mail survey needs a questionnaire which
should have some additional characteristics, it is necessary to look into this aspect in some
detail. First, we should know the advantages and limitations of a mail survey.

advantages7
1. It is much easier to approach a large number of respondents spread all over the country
through post.
2. A mail questionnaire will not have any distribution bias as it will not show any particular
preference or dislike for a certain neighbourhood, household or individual. This,
however, is not the case with the personal interview as it is likely to be affected by the
personal prefer- ences or dislikes of the individual interviewer.
3. Likewise, a mail questionnaire is free from any interviewer’s bias and errors which may
undermine the reliability and validity of the results emerging from the survey.
4. Where the question asked is such that it cannot be answered immediately and needs some
thinking on the part of the respondent, it is the mail questionnaire which will be most
appro- priate. A respondent can think over the question leisurely at home before giving
his answer. The quality of answers is, therefore, likely to be superior to that obtained in
the personal interview.
5. Since a large number of respondents can be approached all over the country through mail
in a short period, a mail questionnaire saves a lot of time in collecting the desired
informa- tion.
6. There is a good deal of saving in the total cost of a mail survey as cost of travelling,
boarding and lodging of interviewers is not to be incurred. This enables the organisers of
the survey to complete the investigation within a limited budget.
7. In case of mail questionnaires, there is no difficulty in having central supervision and
con- trol over the survey operations over a large region. This, however, is not the case
when a large number of interviewers are appointed to cover different territories. As they
have to be instructed, supervised and checked, these tasks cannot be managed from one
centre alone.
8. Mail questionnaires also avoid the bias arising from any inhibitions in answering
questions. In particular, when questions are of a personal nature, the respondents may
hesitate to answer them in the presence of an interviewer. This type of inhibition will not
be there if the mail survey is undertaken.
9. Finally, mail questionnaires will not have the problem of non-contacts in the strict sense
as might be the case in personal interviews when the interviewer finds that the
respondent, being away from home, is not available.
Limitations
The mail questionnaire suffers from some major limitations which are mentioned below:
1. It is not suitable when questions are difficult and complicated. In such a case, the help of
interviewers is required to offer some introductory explanation to the respondent. Further,
in all such surveys where the main object is to get the respondent talking or to undertake
deep probing, mail questionnaires are completely unsuitable.
 2. When the researcher is interested in the spontaneous answers of the respondent or his
own answers uninfluenced by others who may influence his thinking, the mail
questionnaire is inappropriate.
3. In case of mail questionnaires, it is not possible to verify whether the respondent himself
or herself has filled in the questionnaire. If a questionnaire is addressed to a housewife
concern- ing the expenditure on durable items in the family, it is she who is supposed to
answer it. However, she may ask her husband or someone else in the family to fill in the
questionnaire on her behalf. It should be evident that such answers may not be correct.
Further, they may not reflect the opinion of the particular respondent whose opinion was
sought.
4. The researcher has to accept the answers as they are provided in the mail questionnaire.
In case there is any inconsistency or ambiguity in the answers, it will be difficult for the
researcher to make use of such a questionnaire. He cannot further probe into the same to
get some additional information or to remove the inconsistency or ambiguity.
5. The respondent, in case of a mail questionnaire, may go through his answers after he has
filled in the entire questionnaire and may make certain modification in his original
answers as a result of which these answers cannot be regarded as independent.
6. A mail questionnaire does not allow the researcher to supplement the information by his
personal observations. That will be possible only when the questionnaire is canvassed by
him personally.
7. Finally, a mail questionnaire normally has a relatively poor response compared to a
question- naire canvassed personally. In the latter case, even if the respondent is initially
reluctant to answer the questionnaire, the interviewer can explain the purpose of the
survey and point out its relevance to the respondent who may then agree to answer the
questionnaire.

additional Considerations for the Preparation of a Mail Questionnaire

As the interviewer is just not available for any explanation or clarification that a respondent is
likely to need, it becomes necessary to prepare a mail questionnaire with greater care and more
thought. It would, therefore, be worthwhile to look into additional factors that can significantly
improve the quality of a mail questionnaire.
1. Mail questionnaires should be shorter than the questionnaires to be used personally
otherwise the response rate would be affected adversely.
2. The wording should be extremely simple so as to avoid any misunderstanding.
3. In case a lengthy mail questionnaire has to be used, it may be desirable to send an
advance letter, seeking the cooperation of the respondent. This may be followed by the
questionnaire. Such a practice will prepare the respondent mentally to receive a lengthy
questionnaire and a reasonably high response rate can be maintained.
 4. Wherever necessary, suitable explanations should be provided so that respondents under-
stand the questions in the proper perspective and in the manner in which the researcher
intended.
5. A covering letter must invariably be enclosed with the mail questionnaire. The covering
letter should explain the purpose and importance of the survey and solicit the cooperation
of the respondent. It should be emphasised in a covering letter that the information to be
given by the respondent will be kept strictly confidential and that his identity will not be
disclosed.
6. A mail questionnaire should also be accompanied by a pre-addressed and stamped
envelope to facilitate the respondent to return the same as soon as it is filled in without
incurring any expenditure himself.

Pre-testing the Questionnaire

Once the questionnaire is ready, it should be pre-tested. Pre-testing of the questionnaire implies
that it is tried out on a few respondents and their reaction to the questionnaire is observed. It
helps the researcher decide whether any changes in the question-content or the wording of
questions are called for. If so, specific changes that are desirable can also be ascertained and
incorporated in the questionnaire. This would improve it and if it is a mail questionnaire, it
would perhaps increase the response rate as well.
The other benefit of pre-testing the questionnaire is that the researcher can know the
suitability of the instructions given to the interviewers as also their capability. In case certain
changes are required, the same can be introduced. Interviewers will also have an opportunity to
familiarise themselves with the problems they might face in the collection of data. This apart,
pre-testing may indicate whether a particular sample design is feasible or some other sample
design, which may be more appropriate, should be selected. Sometimes, pre-testing of a
questionnaire is undertaken to find out the suitability of data for particular needs. For this
purpose, the researcher may have to tabulate the data collected in the pilot survey (or pre-testing)
and prepare dummy tables. With the help of these tables, one can examine whether such data
would be appropriate and adequate for the objectives of the survey. In the light of this
investigation, the questionnaire can be revised to elicit additional information.
The foregoing discussion may be summarized as below:

important Considerations while Designing a Questionnaire

1. Specify the information required for the problem under study.
2. Select the type of questionnaire and the method for its administration.
3. Determine the type and content of each question.
4. Be careful in phrasing each question.
5. Decide the sequence of questions.
6. Determine what form of response each question is likely to have and provide space accordingly.
7. Decide the number of questions/size of the questionnaire.
8. Determine the layout of the questionnaire.
9. Pre-test the questionnaire among a small number of respondents.
10. Finalise the questionnaire and use it for the proposed survey.
 In the appendix to this chapter, a specimen questionnaire is given. It relates to the passenger survey
conducted by the Quick Airlines Corporation.
The questionnaire was designed specifically to collect data from passengers who availed them-
selves of the services of the Quick Airlines Corporation during the two weeks when the survey was
on. It will be seen that the first four questions give the passenger identification data. The next four
questions relate to the flight details including the date and number of the flight. Subsequent ques- tions
aim at collecting information on the views of passengers on the quality of service provided by the
Quick Airlines Corporation. Some of the questions are dichotomous, resulting in only two responses,
yes or no, while others provide five options on a scale such as excellent (5) to very poor
(1) in regard to a particular service. There are only three open-ended questions in the entire ques-
tionnaire. The last question invites suggestions from the passengers for improvement of the quality of
service rendered by the Corporation. It will also be noticed that this questionnaire is extremely simple
and it would not take much time of the respondents in providing the information sought.

1. What is a questionnaire?
2. What are the different types of questionnaires?
3. Distinguish between structured-non-disguised and non-structured-disguised questionnaires.
What are their relative advantages and limitations?
4. What are the characteristics of a good questionnaire?
5. What factors would you take into consideration while designing a questionnaire?
6. Explain the following types of questions, giving a suitable example in each case:
(i) Open-ended (ii) Dichotomous
(iii) Multiple-choice (iv) Opinion-oriented
(v) Leading
7. What are the advantages and limitations of (i) open-ended questions, (ii) dichotomous ques-
tions, and (iii) multiple-choice questions?
8. How is the sequence of questions relevant in a questionnaire?
9. What are the advantages and limitations of a mail questionnaire?
10. What additional considerations should be borne in mind while designing a mail question-
naire?
11. What is a pre-test? How is it helpful?
12. Design a three-page questionnaire on a hypothetical field survey of your choice.