Fichas
Chapter 9: Questionnaire design
Fichas del conjunto (31)
- 12 key constructs of good design:Reliability
Validity - 23 types of validityConstruct - what the question is measuring
Content - how suitable the question is for measuring that (more subjective)
Criterion - compared to established measure - 3Two types of reliabilityTest-retest
Split-halves - 47 examples of design issues:
- 55 main benefits of good designData quality,
Interviewer's task,
Respondent's experience,
Analysts task,
Perception of research - 6Data quality: 3Valid and reliable data
Minimise non-response
Minimise error - 7Interviewer's task: 2Make task straightforward as possible
Minimise questioning and recording errors - 8Respondent's experience: 3Getting and maintaining interest and willingness to participate
Making it an enjoyable experience
Making it as easy as possible - 9Analysts task: 1Make DP/analysis accurate and efficient
- 10Perception of research: 3Raise profile of research
Enhance professionalism and credibility of research
Increase goodwill of general public to research - 11Aim of questionnaire design processto convert research objectives into meaningful questions and assemble these into and effective and workable questionnaire
- 127 stages of design processClarify what you need questions to measure
Wording questions
Decide types of question and response formats
Put questions in effective layout and logical order
Design layout
Test draft
Revise and agree final version - 13Questionnaire design is based on: 7What is research problem
Types of data and constructs to test
Nature of population
Method of data collection
How will responses be captured
Other constraints (time/budget)
How responses will be analysed - 14Question content: 2Use standard questions
Clarify the meaning of concepts, definitions and indicators - 15Benefits of standard questions (3)+ Makes preparation easier
+ Tried and tested
+ Allows comparison with other tests - 16To agree definition for the concept (3):Use formal/informal qualitative research
Ask members of target groups
Check definitions used by others - 17Conceptualisation= moving from abstract concept to concrete indicators
- 18Indicators: 3May need to think about dimensions or aspects of concept
E.g. for sexism are you interested in the gender stereotyping aspect
Can specify further e.g. only in certain situation such as at home or at work - 19Question wording (3 areas)Aims, basics, sensitive topics
- 20Aims for each question: 8Valid
Relevant
Is acceptable to respondent
Understand clearly
Interpreted as intended
Consistently interpreted
Elicits an accurate and meaningful response
Meaning of response is clear and unambiguous - 21Basics of Q wording:Simple wording
Precise terms and examples/definitions where appropriate
Avoid negative phrasing
Avoid double-barrelled questions
Keep short
Specify reference period
Avoid adverbs e.g. regularly/frequently as can mean different things to different people
Use never, almost never, several times a week
Avoid hypothetical questions about future behaviour
Avoid assumptions
Avoid leading questions
Use neutral wording (e.g. not "welfare for the poor")
Ensure question is complete - interviewer should not need to use own words to explain it - 22Sensitive topics - why importantIf not handled properly can lead to question/survey refusal
- 23examples of sensitive topics (4 of 7)money, voting, religion, sexual activity, criminal behaviour, alcohol, drugs.
- 24Ways of handling sensitive topics: 3Respondent enters response themselves
Reads out code on showcard
Online/telephone interviewing - 25Cases of social desirability bias (5)Being a good citizen, Well informed/ cultured, Using vote, Occupation/income, Illness
- 26To avoid soc des biasMake it just as easy to give low prestige answer as high prestige answer
- 272 types of question structureopen ,closed
- 28Open questions 3++ Respondents feel more at ease and more in control
+ Helps build rapport
+ Gives wide range of responses - 29Open qs 3-- Require more from respondent, interviewer and DP
- More time consuming and therefore costly
- Have to code - 30Open question - when to useInclude early in questionnaire or at start of sections
- 31Closed questions 4+, 2-+ Quick
+ Easy to administer
+ Easy to process
+ Cheap
- Difficult to formulate
- Some detail is lost
Información
- Descripción
- Fichas sobre Chapter 9: Questionnaire design, creado por corin.mulliss el 28/04/2014.
- Creado por
- corin.mulliss
- Visitas
- 10
- Valoración