Questionnaires and quantitative data analysis Flashcards
Explain overview of questionnaires/surveys
• Applicable to multiple study designs
• Common tool to collect data
• Enables standardised data collection ↓bias
• Open and closed questions
• Large or small samples
• Self or researcher administered
• Mail or postal questionnaire most common form
(plus online)
• Easy to enter data, and analyse (pre‐code)
What are pre-coded questionnaires?
The respondent is asked to choose one or more responses from a series of choices pre-determined by the researcher such as sex category
What are some key things to consider when designing the research question?
- Has someone asked it before?
- Long questionnaires = ↓response rates & respondent fatigue
- Use simple language, avoid jargon
- Pilot your questions/length
- Ensure relevance
- Do not cramp the presentation
- Vary question format to avoid boredom
- Easy to follow design
- Keep questions and answers together
- Clear instructions about how to respond
- Include respondent id on every page
What is a Likert scale and what needs to be considered?
A Likert scale allows the respondent to indicate the extent to which they agree with a certain statement. • Balanced response scale • 3, 5 or 7 options • Equal number positive and negative options • Ordinal data item
What are the advantages of self-completion questionnaires?
- Cheaper
- Quicker, Familiar to most people
- Absence of Interviewer effect
- Convenience
- Go at your own pace
What are the disadvantages of self-completion questionnaires?
• Cannot prompt • Difficult to ask many questions • Do not who answers/completes • Not appropriate for some respondents – low literacy • Respondents may not take research seriously (e.g religion in national censuses) • Lower response rate – huge problem in quantitative research • Greater risk of missing data
What steps can be taken to improve response rates?
• Good covering letter • Follow up non‐responders • Provide instructions and an attractive layout • Limit the number of open ended questions • Provide incentives (e.g. a gift voucher)
Explain what a covering letter should look like?
- Friendly but short, tailored
- Describe why the study is being done
- Mention incentives (if any) e.g. copy of results
- Utilise contacts
- Use deadlines
- Describe confidentiality/ anonymity policy
- Name and corporate phone #
What are some potential issues with the questionnaire design?
- Leading questions
- Multiple questions
- Confusing categories
- Appropriate options
Explain the issue of leading questions
- Never should be right or wrong answers
* Do not influence participants in any way
Explain the issue of multiple questions
- Don’t string questions together. Keep each question limited to one fact.
- Example: What is worrying you? Is it that your employment opportunities are limited? Is it other areas?
Explain issue of confusing questions
You should be able to clearly define the concepts in each question, and ensure that your understanding of those concepts is the same as the person completing the question
e..g not ‘do you regularly participate in sport?’
Explain issue of appropriate responses
Options listed must relate or be able to answer question
e.g. not ‘what is your favourite sport?’ and allow 3 options
How is quantitative data organised and analysed? and different types
Process of statistical analysis:
• Descriptive Statistics
• Inferential Statistics
What must data analysis always have?
A purpose such as describing, comparing, examining similarities or differences
