Collecting Primary Data: Questionnaires

1
Collecting Primary Data Questionnaires
2
Objectives

• After this session you will be able to
• Plan and design valid and reliable
  questionnaires.
• Describe the processes involved in collecting
  primary data, including piloting.
• Demonstrate the skills for writing appropriate
  individual questions and designing
  questionnaires.
• Write appropriate documentation to accompany
  questionnaires.

3
Types of questionnaire
Questionnaire
Self-administered
Interview-administered
Postal questionnaire
Delivery collection questionnaire
Telephone interview
Focus group
Structured interview
Onlinequestionnaire
4
Why questionnaires?

• Low costs (time and money).
• Quick inflow of data.
• Can reach many people.
• Quick coding of closed questions.
• Reliability of data.

5
Questionnaires disadvantages

• Low response rates (threat to reliability).
• No opportunity for probing questions.
• No opportunity for respondent to seek clarity
  (threat to validity).
• Socially desirable responses (SDRs) (threat to
  validity).

6
Face validity

• Do the questions make sense?
• Are all concepts in the questions understandable?
• Are the questions interesting and motivating to
  answer?
• Are the instructions for completing the
  questionnaire clear?
• Is the questionnaire of the right length?

7
Achieving face validity

• Get your questionnaire reviewed by 15 of your
  peers.
• Get them to write down the problems they find on
  the questionnaire.
• Talk to them.
• Note which elements need improving and comment on
  how you did this in your Methodology.

8
Design construct validity

• Construct validity an instrument measures the
  constructs it was intended to measure.

Theory
Concept
Constructs
Question sets
Marketing
Relation-ship Marketing
- Loyalty- Internal customers- Supply chain
Question sets
- Loyalty- Internal customers- Supply chain
9
Measuring attitudes reliability

• Reliability the consistency between two measures
  sufficient numbers of questions for each
  construct.

• Loyalty
• To me, brand loyalty is important.
• I tend to buy the same brand all the time.
• If I like a brand I always buy it in preference
  to others.
• Brand loyalty is unimportant in my buying
  decisions.

10
Reliability of responses

• Loyalty
• To me, brand loyalty is important.
• I tend to buy the same brand all the time.
• If I like a brand I always buy it in preference
  to others.
• Brand loyalty is unimportant in my buying
  decisions.

Note these questions will be randomized
scattered throughout the questionnaire
11
Measurement of reliability

• Cronback alpha at least 0.7.
• If below 0.7 you need to eliminate or make
  changes to low reliability questions.

12
Questionnaire design closed questions

• Seek a predefined response Yes/No True/False,
  etc.
• Easy to code and analyse.
• Make group comparisons easy.
• Low cost, but
• May restrict richness of responses.

13
Questionnaire design open questions

• How, why, what ?
• Good for eliciting rich responses, but
• Harder to analyse.
• Example You have indicated a loyalty to Krispy
  Crisps please explain why -------------------
  -----------------------

14
Questions to avoid (threats to validity)

• Leading Why do you like Krispy Crisps?
• Double Do you like Krispy Crisps or Walkers
  Crisps? YES/NO
• Assumptive How often do you eat Krispy Crisps?
• Hypothetical Suppose you were asked to eat 20
  packets of Krispy Crisps?
• Knowledge based Recalling the recent Krispy
  Crisp advertising campaign

15
Good (valid) questions are

• Simple and easily understood.
• As short as possible.
• Unambiguous.
• Not likely to offend.
• Provided with categories such as Dont know and
  Not applicable.

16
Whats wrong here?

• Whats your view on crisps?
• State your age
• Without effective best practice crisp making
  cannot prosper.
• How many packets of crisps have you stolen?
• COMPLETE AND RETURN THIS FORM

Under 20
20 - 25
Over 30
25 - 30
17
Response categories (1)
List question What do you think is the most
important influence on the success of
theorganization in the next two years?Please
?as many responses as you think accurate.
Changes in government policy affecting the legal regulation of the market ?
The entry of new competitors to the market ?
The impact of the companys current reorganization strategy ?
Foreign exchange rates ?
18
Response categories (2)
Category question How often in an average week do
you use our e-banking facilities? Please ? one
response.
Never ?
Once ?
23 times ?
45 times ?
6 times or more ?
19
Response categories (3)
Ranking question Please indicate in the boxes
provided which features you believe arethe most
important when visiting our superstore(1
indicating the most important, 2 the next most
important, etc.) .Please leave blank those
features that have no importance at all.
Ease of car parking ?
Low prices ?
Friendly staff ?
Store loyalty card ?
Variety of goods ?
20
General Instructions

• Title of survey
• This survey is being undertaken by your name as
  part of name of course at the University of
  Surrey to state purpose.
• It will take approximately x minutes to complete.
• All information provided is completely
  confidential.
• Thank you for providing your comments.
• Please return toby

21
Specific instructions go to
12 Do you eat Krispy Crisps?
?
No
Go to 18
Yes
13 How many packets per week?
22
Pilot the questionnaire

• Through piloting you may decide to
• Change the wording of individual questions.
• Add/eliminate some questions.
• Amend the instructions.
• Decide not to use the questionnaire!

23
Summary

• Designing individual questions involves a
  rigorous process of analysis to avoid ambiguity,
  leading questions, double questions and simply
  misleading questions.
• Questions must be clearly linked to the purpose
  of the research
• Questionnaires should start with questions that
  are easy to answer, interesting and transparently
  linked to the purpose of the research.
• Questionnaire layout and the use of typography
  can make a questionnaire easier to complete and
  more appealing to respondents, enhancing the
  response rate.
• Clear, well set out instructions on completing
  the questionnaire can also boost the response
  rate.
• Web and email questionnaires offer a new and
  potentially powerful tool, but also require
  additional design skills.
• All questionnaires, whether paper-based, email or
  Web-based, require thorough piloting which will
  include evaluation of accompanying documentation,
  instructions, individual questions, types of
  question, question sequencing, the use of scales
  and skip instructions basically, everything!