Chapter - 0 5 Design of Questionnaire

Questionnaire

• A set of questions on a topic or group of topics designed to be answered by a respondent.

Why a Questionnaire is needed?

• To standardize the process of data collection – helps in analysis

• To achieve speed & accuracy in collection & recording

• To achieve speed & accuracy in handling data within & between offices for analyses

3

Situations suitable for Questioning

• Knowledge, opinion, motivation, intension etc. are not open for observation

• Past events (like time & quantity of last purchase) can be studied only by questioning

4

Questionnaire Preparation not an easy task

• Respondents tend to answer every question –

whether they have the information or not

• Response depends on the way the question is asked

• Biasness may arise from interviewer and

respondents side.

5

Sources of Error in Questionnaires

• Memory: Respondents cant remember correctly• Motivation: To project a better self image,

respondents may be motivated to respond to all questions even incorrectly

• Communication: May not understand the question• Knowledge: Simply may not knowAttitude measurements are more prone to error

6

CLASSIFICATION OF QUESTIONNAIRE

Structured

Disguised

Non-disguised

Continue…

• The marketing researcher appoints some

interviewer to collect information on his behalf.

• The interviewer are expected to adhere to

the same order in asking questions as

contained in questionnaire.

• Respondent should be taken into confidence

& clearly told why the survey is being

undertaken.

• Respondent would realise its relevance &

give the desired information accurately.

• Majorly used for marketing research survey.

Continue..

₊ Structured questionnaire calls for straightforward and

simple approach on the part of interviewers as such.

even less qualified interviewer can be deployed in

canvassing such questionnaire.

₊ Easy to edit, tabulate and interpret the data it contains

₊ It can be conveniently pre-tested so that suitable

modifications can be made in the phraseology of

questions or in their sequence or both.

• Questionnaire of this type can be split into two

sub-types

• When Purpose of the enquiry is disclosed to

the respondents then it is known as non-

structured & non-disguised questionnaire.

• While in other cases it is known as non-

structured & disguised questionnaire.

Non-Structured Questionnaire

• In such case marketing researcher will be a loser

• To avoid such situation, he should give serious thought to the

specific information to be sought.

• It will 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.

Process of questionnaire design1

• Types of Information required

2 • Types of questions

3 • Phrasing of Questions

4 • Layout of Questionnaire

5 • Pre-testing of Questionnaire

STEP 1: Types of Information to be collected

• While attempting a design a questionnaire, the marketing researcher has to first ask himself what type of information he needs for the survey

• He should take this question seriously as it will have considerable repercussion on the usefulness of the survey

• For ,if he omits to collect the information on some relevant and vital aspect of his survey ,his research study is unlikely to be useful.

• If he collects some information on some issue which are not directly relevant to his study, he not only raise the cost but also increase time factor.

• It will also lead greater inaccuracy as the respondents will have to answer many more questions than are strictly necessary and he will, therefore, not be sufficiently careful in giving exact answers.

Types of Information-(Suggested by Crisp)

facts Quasi facts

Awareness

Opinion Attitude Future Plans

Reasons

Fact & Quasi-fact• Factual Information is frequently used in research• Ex: Do you own a car?, is intended to seek such factual

information• Quasi facts implies that the information received from a

respondent in not factually corrected though it may appear to be so.

• In continuation of above ex, If yes, when did you buy the car?• Here, the respondent may not remember the moment he

bought the car.• This lead inaccuracy• Apart from this in accuracy may arrive on account of the desire

on the part of respondent to leave a favorable impression on the interviewer. Ex: Do you eat fruit everyday?

Fact

• Regarding factual information the interviewer has to ensure

that the information sought is available with the respondent

and he is willing to share it.

• It should not be based on past else inaccuracy may arrive

• He should ensure that respondent should understand the

question correctly

• This will avoid miscommunication

Awareness

• Sometimes marketing researcher is interested to know

whether the respondent is aware of the existence of 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 certain awareness.

• Increase in penetration after first study

Opinion

• 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 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

verbalization of the attitude.

Difficulties in Opinion• The question on opinion sometimes poses some difficulties for the

researcher

• When an answer 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 opinion without knowing the subject at all.

• This a great disadvantage and make difficult for researcher to

verify.

Continue..• Another difficulty is to ascertain the intensity of an opinion,though this

can be take care by suitably phrasing the specific questions.

• Another difficulty is that opinion is 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 view point the 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.

Future Action Plans

• Occasionally, Marketing researcher wants to know what are the

immediate plans of the respondent in regard to certain thing.

• Ex: He may be asked whether he plans to buy a car during the next six

months.

• Such an information is normally a statement of intension 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 conclusion.

Reasons• Finally, the marketing researcher at times wants to know the reasons for a

particular choice or action of the respondent.

• For Ex: the respondent owing a car, may be asked why he bought that particular make.

• He is expected to give one or more reasons to support his answer.

• Here too, there is an apprehension that the respondent may not come out with genuine reasons.

• It must be emphasized again that the researcher should be clear about the nature of the information to be collected.

• The above classifications will facilitate him in phrasing the questions in right form.

STEP 2: Decide the types of Questions

Types of questions

Open-ended

DichotomousMultiple-

Choice

Open-ended Questions

• Open or free answer question gives respondent complete

freedom to decide the form, length and detail of the answer.

• They are preferred when the researcher is interested in

knowing what is uppermost in the mind of the respondentDifficulties in Open-ended Questions• At times it is difficult to note down the respondent’s answer

verbatim.(word to word)• Without mechanical aid it is very difficult and can miss out

some vital information contained in respondent’s answer.

Continue..

• 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.

• It is extremely difficult to compress lengthy answers in a

meaningful manner.

• Such answers may be good in quality but their quantification

becomes extremely difficult.

Dichotomous Questions

• It has only two answers in the form o f “Yes” or “No” Or “True”

or “False” Or ”Use” “Do not Use”.

• Ex: Do you use Tobacco in any way?

– Yes______________ No_________________

• There can’t be third answer, However in some cases there can be third

answer which may come from those who do not want to take

definite stand one way or other.

• Do you like Movies?– Yes________ No________ Neither like nor dislike________

Advantages of Dichotomous questions

• Dichotomous questions are most convenient or least

bothersome to respondents.

• Simply two possible alternatives are given to indicate

choice

• Such questions framing require least time and also response

takes less time.

• Answers to such questions are easy to edit, tabulate and

interpret.

Multiple choice questions• Respondents are offered two or more choices.

• The marketing researcher exhausts all possible choices and the

respondent has to indicate which one is applicable in his case.

• Ex: Which of these brand/brands do you use for washing clothes?

• Rin___, Det___, Wheel____, Fena____Super 777___,Excel___Any

other____

• Obviously the respondent is likely to take more time to answer such

questions as compared to dichotomous questions.

• More time required for editing, tabulation and interpretation.

Step-3: Phrasing of Questions• The way in which a question is drafted in very

important as a slightly suggestive words would elicit a very different answer from the respondent.

• Ex: 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 agree that it is of sub-standard quality.

Continue…

• However if the above is worded little bit

differently ,the answer is likely to be different.

• Do you think that this is a sub-standard product?

• This is not suggestive question and it is straight –

forward and respondents are not likely to be

prompted to say “yes” as in case of earlier form.

Factors to be considered while phrasing questions

• To ensure the appropriate phrasing questions, one should

be particular about the following factors-

• Difficult words should be avoided.

• Technical, special terms which an average respondent

may not understand should be excluded.

• Vague words such as “many”,”often”,”reasonably” should be

discarded

Continue…

• Lengthy questions should be avoided. this may lead

confusions.

• One should avoid combining two questions into one. For Ex:

Which of the following mode of transport is cheaper and more

convenient? (i) Train (ii) Bus.

• Questions lacking specificity should be avoided or modified

suitably so that they become more precise.

• Ex: Are you satisfied with your job?-is not sufficiently specific

because it does not provide the necessary frame of

reference to the respondent.

Order/Sequence of Questions

• G ive proper attention to order/sequence of questions to be

contained in questionnaire.

• Researcher has to built rapport in the beginning with the

respondent, it is thus necessary that questions asked at the

beginning are simple and helpful in establishing rapport.

• Difficult questions or those on sensitive issues should be

relegated to the end of the questionnaire.

• Questions of general type should be asked in the beginning.

• Those are specialized questions and needing some depth

information 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 specialized

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 of the sampled units.

• Is such case, some respondents would answer the specialized

questions towards the middle of the questionnaire instead of

towards its end.

How many questions to be asked

• The researcher has to decide how many questions to be asked

• To sustain the interest of the respondent questionnaire should

be short and but also at the same time able to obtain requisite

information.

• Too lengthy questionnaire would be a disadvantage and the

response to it may be quite poor.

• While deciding upon the no.of questions or the length of the

questionnaire researcher should put himself in the

respondent’s shoe and imagine how he would react to that

questionnaire.

Continue…

• He can calculate the probable time that might be

required to answer the entire questionnaire.

• He can also canvass the questionnaire amongst some

of his friends and acquaintances.

• their opinion and reaction will be very helpful to him in

finally deciding the length of the questionnaire.

• For this pre-testing can be done.

STE P-4 Layout / Print j ob of the Questionnaire

• It implies that the document should be set in such a way that it

leaves a favorable impression on the respondent.

• It should be neatly printed and the individual pages should not

have too many questions 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 should be set in bold types or

underlined.

Continue..• If it is really a lengthy questionnaire, special care should be

taken to reduce the size by providing two-column in a page

and by using finer types.

• But sometimes too fine printing may cause inconvenience to

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.

Step-5: Pre-testing the questionnaire

• It implies that questionnaire is tried out on a few

respondents and their reaction is observed.

• It helps to decide whether some changes in the

question-content or the wording of questions are

called for.

• If so,some specific modifications can be done.

• It would improve your response rate.

Continue…• Researcher can know suitability of the instructions given

to the interviewer as also capability.• Interviewer will also have a opportunity to familiarize

themselves with the problems they might face in the collection of data.

• P-re- testing may indicate whether particular sample design is feasible or some other to be taken.

• P-re- testing helps to the know the suitability of data for particular needs. For this, tabulation, preparation of dummy tables all done and examine whether such data would be appropriate and adequate for the objective survey.

I mportant consideration while designing a Questionnaire

1. Specify the information required for the problem under study

2. Select the type of questionnaire and the method of

administration

3. Determine the type and content of each question.

4. Be careful in phrasing each question

5. Decide the sequence of each question

6. Determine what form of response each question is likely to

have and provide spaces accordingly

7. Decide the number of questions/size of questionnaire.

8. Determine the layout of questionnaire

9. Pre-test the questionnaire among a small number of

respondents

10. Finalize the questionnaire and use it for the proposed survey.