Survey Design Flashcards
Describe open questions
- Allows freedom in answer
- Allows participant to give more detail/explanation
- Difficult to analyse
- Used more in qualitative
Describe closed questions
- Participants pick best fit from series of provided answers
- Unable to give detail
- Easy to analyse
- Used more in quantitative (can be analysed with SPSS)
Describe rating scales
- Collect data on attitudes/perception
- For every statement, participants rate how much they agree/disagree
- Likert scales - scales 1-5
Describe validity
Does it tests what it intends to
Includes: face validity, content validity, concurrent, predictive, construct validity
Face validity:
- From the appearance of the questions do they measure what they intend to
Content validity:
- Do the questions on the scale cover the important characteristics of the concept being measured; to gain content validity the questions need to be grounded in literature
Concurrent validity:
- Does the scale correlate well with other measures of the same concept taken at the same time
Predictive validity:
- Does the measure accurately predict future behaviours
Construct validity:
- Is there a well-developed theoretical and conceptual understanding of the concept being measured
Describe reliability
Produces same scores on different occasions
2 types: consistency over time and internal consistency
Consistency over time:
- How much a test n produce similar scores on subsequent trials
- Can be produced by participants remembering previous answers
Internal reliability:
- Split-half reliability = looking at first half of questions and then the second; do they correlate well
- Odd-even reliability = looking at set of scores; do the odd and even questions correlate well
- Cronbach’s alpha = procedure used to measure reliability of psychometric instrument; co-effiecient will increase when correlations between test items increase (ideals is 0.7 and above)
