QUESTIONNAIRE

Questionnaire construction
4. Redrafting of the questionnaire
 If problems were found then the

questionnaire will need to be redrafted

 If there are major changes required it

is best to then do another pilot

 If the problems are minor, the

researcher may then proceed to

administration of the questionnaire to
the full sample
Questionnaire construction
5. Administration of the questionnaire
 After development of the questionnaire,

it is administered to the full sample of

respondents
 How would this be done? Postal,

telephone, structured interview. Mode of

administration differences
 The responses are then analyzed in

terms of the researcher’s aims and

objectives
Ethical Considerations
 These apply to all types of research
and generally an ethics application
will be required with human subjects
 In designing questionnaire questions,

respondents should not be misled

concerning the aims of the study or
how their data will be treated
(confidentiality aspects).
Ethical Considerations
 Ifthe questionnaire is anonymous,
those who do not choose to reply
should not be pestered
 However reminder postcards may be

sent out and then another follow up

questionnaire if necessary
 Hence, the chief investigator will

know who and who has not replied.

 De-identification and safe storage
Question and questionnaire
formats
 Interviewer schedule method
 Formal questionnaire not prepared to

be filled in by the respondent

 Rather, it guides the interviewer who

asks the questions

 Can probe to clarify response vs. self

report
 There are cost considerations for

using this method.

Interview Methods
 Needs expert interviewers
 Time consuming
 May introduce interviewer bias
Question and questionnaire
formats
 Telephone questionnaires
 May be more efficient to collect data via

telephone than via postal means

 Can get better response rate than by postal

methods
 Advantage of being ‘anonymous’ over face

to face interview method

 Can automate data entry
 Not everyone has a phone
 Difficult to administer long questionnaires
Question and questionnaire
formats
 Open-ended and closed-response formats
 Q1. How do you feel about the standard of the
treatment you received while you were a patient at
this hospital? (also focus group approaches)
 Q2. How would you rate the standard of the
treatment you received while you were a patient at
this hospital (circle one number)
 Excellent 1
 Good 2
 Moderately good 3
 Fair 4
 Poor 5
Question and questionnaire
formats
 Likert and forced-choice response
formats
 Forced-choice response does not

allow for middle of the road or

undecided answer.
Response Category Issues
 D.1 How satisfied are you with the effect of your treatment or
 care?
 Very satisfied …………………………………….... 1
 Satisfied ……………………………………………... 2
 Neither Satisfied nor dissatisfied …………… 3
 Dissatisfied ……………………………………….…..4
 Very dissatisfied …………………………............5
 D.2 How satisfied are you with the explanations the doctor or
 other health professional has given you about the results of
 your treatment or care?
 Very dissatisfied ……………………………..… 1
 Dissatisfied ………………………………………… 2
 Neither satisfied nor dissatisfied …………… 3
 Satisfied ……………………………………………. 4
 Very satisfied ……………………………………… 5
Question and questionnaire
formats – open ended questions
 What do you think is the main problem with
the average Pakistani diet?
 What is the major source of stress in your life

at the moment
 Respondents are free to answer according to

their own ideas

 Form small groups and note the responses

you obtained from group members. How

would you classify and analyze this data?
What are some issues with open ended
questions?
Question and questionnaire
formats – open ended questions
Advantages
 Useful for gathering new information

or a pilot study to refine Qs

 Allows adequate answers to complex

issues
 Allows the subject to ‘speak their

mind’
Question and questionnaire
formats – open ended questions
Disadvantages
 Time consuming for respondent
 Difficult to code and analyze
 ‘Irrelevant’ information may be

provided
 Requires greater literacy
 Different level of detail in answer
The wording and design of
questions
 The writing of good questions is an art and it
is a time consuming one
 Common problems are double barreled Qs,

leading or loaded Qs, ambiguity, and the

complexity of language used
 To obtain valid and reliable responses one

needs well worded questions

 Consider “ Do you smoke?” Yes/No. What are

some problems associated with this question.

Develop a better question concerning
smoking behavior.
The wording and design of
questions
 There are a number of pitfalls to
avoid when writing questions for
questionnaires
 1. Double barreled questions. Eg. Do

you like cars that are big and

powerful, or small and economical?
◦ These questions should be separated out
to be clear about what you need to know
The wording and design of
questions
Avoid
 2. Long questions – cause confusion

and fatigue
 3. Ambiguous questions. Avoid

vacuous words that may mean

different things to different people.
 Eg. ‘old people’ may be 20 to a

toddler or 80 for a 50 year old. Be

specific about such matters

4. Inappropriate level of wording
◦No jargon or acronyms e.g. DIY
◦No double negatives
◦Keep it simple and concise
◦Appropriate readability/literacy
level for sample
◦Culture, age and gender
appropriateness
The wording and design of
questions
Avoid
 5. Bias and leading questions

◦ The wording should not lead the respondent to

answer in a particular way (e.g. social desirability)
◦ Eg. ‘How often do you eat chocolate?’ This may
prompt some people to be less than truthful in
answering such a question.
◦ The response format may be biased too if it does
not accurately reflect the true behavior
◦ Eg. □ 1/month □ 1/week □ 1/day
The person may be eating it 3 times per day
The wording and design of
questions
 The possibility of an invalidly
administered questionnaire
 A survey on ‘attitudes to migration’

might be answered less than

honestly by respondents if the
interviewer is obviously of immigrant
background
 A patient satisfaction survey

interview administered by the

treating Dr may have similar
Guidelines
 Are the words simple, direct, and
familiar to all respondents? (avoid
technical jargon, consider regional or
cultural differences in word usage
etc)
 Is the Q as clear and specific as

possible? Can it be shortened with no

loss of meaning?
 Are any items double barreled?
Guidelines
 Are the Qs leading or loaded?
 Avoid colloquial terms
 Avoid emotionally charged terms

(e.g. reds, black leaders etc.)

 Is the Q applicable to all respondents

(e.g. how old is your wife; what is

your present occupation)
The Structure of
questionnaires
 1.Introductory statement
◦Purpose of the questionnaire –
benefits that will flow from it
◦Information sought
◦How the information will be used
◦Introduces researcher
◦Confidentiality/anonymity
◦Contact for queries and complaints
The Structure of
questionnaires
 2. Demographic questions
◦Age, sex, education history etc – good
to use examples from ABS
◦These are usually put first to warm-up
respondents – these are easy to answer
 3. Factual background questions

◦Eg. Height, weight, smoking behavior

etc
◦Also helps to warm up respondents
The Structure of
questionnaires
 4.
Opinion questions
◦ Questions requiring reflection
should be positioned after the
demographic and factual questions
◦Avoid conditional questions as
much as possible
 5.Closing statements and return
instructions
◦This usually has statements
regarding thanks for participation
◦Provide information on how to
return the questionnaire
◦Possibility of taking up issues with
the researchers
Summary
 Questionnaires are useful tools if
properly designed and administered
 Well constructed questionnaires can

yield valuable and often novel

information relatively inexpensively
 A questionnaire is a vehicle allowing

communication between the

researcher and the subject
A good questionnaire is the product
of testing and retesting
 Where ever possible, construct or

use a valid and reliable tool, or test

its validity and reliability