QUESTIONNAIRE DESIGN
QUESTIONNAIREDESIGN
QUESTIONNAIRE A questionnaire is formal set of questions
or statements designed to gather information from respondents that will accomplish the goals of the research project.
Questionnaires measure people’s attitudes, behavior and feelings toward just about everything.
ATTRIBUTES OF AN EFFECTIVE QUESTIONNAIRE
It should be user friendly. The questionnaire should look professional. It should be valid. It should be attractive and motivational in
nature. The questionnaire should encourage
respondents to answer honestly and accurately.
QUESTIONNAIRE CONSTRUCTION PHASES
PRE- CONSTRUCTION
QUESTIONNAIRE CONSTRUCTION
POST-CONSTRUCTION
Determine questionnaire objectivesTarget respondent characteristics
Resource constraintsDecide on data-collection method
Question formatNumber of response choices per question
Questionnaire wordingOrganization and layout of questionnaire
Obtain client approval
Pre-testRevise questionnaire based on pre-test resultsAdminister questionnaire and tabulate results
PRE-CONSTRUCTION Determine questionnaire objectives.What needs to be achieved from a data gathering effort? Target respondents characteristics.What characteristics of our target respondents affect
questionnaire designs? Resource constraints.What are the limits of our company resources? Decide on data collection method.How should information be obtained?How should the research efforts be?
QUESTIONNAIRE CONSTRUCTION
Question Content The utility of data . Effectiveness in producing data. Participants ability to answer accurately. Respondents willingness to answer
accurately. Effect of external events
QUESTION FORMAT Open-Ended Questions Close-Ended Questions Dichotomous Questions Multiple-Choice Questions Ranking Questions Checklist Questions
Open end questions No predefined response choices Also called infinite response or unsaturated
type questions Acts as a memory prompts Useful when narrative answers required
and when researcher is uncertain Freedom of response
Closed end questions Questions, which restrict the interviewee’s answers
to pre-defined response options. The uniformity of the questions makes them easier to
record and analyze results quantitatively. They give respondents a finite set of specified
responses to choose from. Motivates respondents to complete the questionnaire.
Dichotomous questions Offers only two response choices Breaking category into two parts Understanding respondents demographic
composition or behaviour inclination Not time consuming
Multiple-Choice Questions Closed end questions that give
respondents several choices. The three issues that must be considered
are: Options should be exhaustive Options should be worded carefully Response options should be distinct from one
another
Contd.. Example:
How would you rate the food quality at the restaurant you last visited?
-Excellent
-Very good
-Good
-Fair
-Poor
Ranking Questions The respondent ranks the response options listed
on a continuum basis in order of preference. Example:
Factors that influence your buying from a particular supermarket. Rank in order of preferenceConveniently locatedHelpful sales staffAvailability of all requirementsRegular discount offered
Checklist questions The participant has the freedom to choose
one or more of the response options available.
Different from multiple choice questions.
Question Wording Factors to be looked into while framing
questionnaires:-Shared Vocabulary-Unsupported Assumptions-Frame of Reference-Biased Wording-Adequate Alternatives-Double barrelled questions-Positively & negatively worded questions-Generalizations & Estimates
Shared Vocabulary Interactive language to be kept simple &
understandable Highly technical language should be avoided as
much as possible Words used should not be ambiguous or vague.
Example: Vague: How satisfied are you with Sterling Resorts?Better:How satisfied are you with the hospitality at Sterling Resorts?
Unsupported Assumptions Assumptions should be explicitly stated not
implied.
Unsupported assumptions lead to exaggerated estimates.
Frame of Reference A single word can have several
connotations under different situations. The frame of desirability should be made
clear. Example:capacity may be connotated
differently by an industrialist & an educator.
Biased Wording Biased Wording should be avoided. Awareness of desirable response leads
to shift of focus from actual response.
Example:Biased: Do you think TV has negative effect on children?
Unbiased: What are your views on the effect of TV on children?
Adequate Alternative Every question should have ample
number of alternatives. They should be explicit rather than being
implicit.Example: Do you prefer eating in restaurants?Which do you prefer eating-eating in restaurants or eating at home.
Double- barrelled Questions A single question that asks for two
responses. Interpretation of the responses not
effective.
Example: Do you believe most Japanese automobiles are comfortable and worth their price?
Positively/Negatively worded questions
Respondents are often guided by the directions of the questions.
Responses are different when surveys are either exclusively positive or negative.
Combination of both is desirable.
Generalizations & Estimates Proper structuring of questions to avoid
generalizations. Answers which require calculations should be
avoided.Example:How many times have sales promotion influenced you to switch brands over the last one year?Better: How many times in the last month have sales promotion influenced you to switch brands?
Questionnaire Sequence Important to elicit required information from the
respondents. The sequencing steps are:
-Lead-in questions
-Qualifying questions
-Warm-up questions
-Specific questions
-Demographic questions
POST CONSTRUCTION
Pre-testing Trial run of the questionnaire using a small
sample from the target population. Helps to refine the instrument and identify
errors. Two stages involved: Personal interviews
-Protocol analysis
-Debriefing Administer the survey
Revise the questionnaire At this stage the flaws and ambiguity
identified in the pre-test are eliminated. It also tests if right sample selection
procedures are adopted. Preliminary indication of what to expect
from the full test.
Administer Questionnaire & Tabulate Results
The interviewers should :
-code the information
-tabulate the data
-conduct appropriate statistical tests
-locate the relationships and differences among variables
-draw conclusions
By:
Akhil AgarwalDeeksha Khanna
Dilip PujariGitanjali.S.Raheja