self report Flashcards
3 question designs
likert scales - require the respondent to say how much they agree with a statement
rating scale - using numerical scale to rate or rank something
fixed option - multiple choice with predetermined options
features of a good questionnaire
- not too long or arduous
- no ambiguous or double barrelled questions
- not have emotive questions
- avoid using leading questions
- not have too much jargon
- not have double negatives
- be simple and easy to understand
strengths of questionnaires
- collect a large amount of data very quickly
- highly reliable, can be replicated with no error, given to large number of participants
- produces data that is easy to analyse
limitations of questionnaires
- social desirability, socially acceptable answers decrease validity, responses may not be truthful due to need to seem normal or nice
- response bias (aquiescence), lowers validity, people tend to get bored and just click ‘yes’ etc for everything
- low reliability of ambiguous questions involving rating scales, we do not all have the same idea of what specific values or numbers mean
types of interviews
structured -
- pre-set questions asked in a predetermined order
unstructured -
- seems more like a conversation, more open questions and each answer determines next question
- participants encouraged to elaborate on answers
- useful for qualitative data, based on meaning
semi-structured -
- combines best of both forms
- some pre-set questions but some free flowing, room for elaboration and to ask follow up questions
features of a good interview
- have an interview schedule so the interviewer always knows what they should be covering
- be standardised for ever participant to reduce interviewer bias
- involve note taking
- not be too long or arduous to complete
- have no ambiguous or double barrelled questions
- not have emotive questions
- avoid using leading questions
- not have too much jargon
- not have double negatives
- be simple and easy to understand
strengths of interviews
structured - easy to replicate
semi/unstructured - generate more qualitative data due to development of new questions
both - generate lots of data, both verbal and non verbal, generate more honest answers if researcher builds rapport
limitations of interviews
structured - restricted to pre-set questions so can’t elaborate
semi/unstructured - hard to analyse due to different questions between interviewees
both - social desirability means data isn’t always accurate, expensive and time consuming, hard to draw conclusions due to generation of qualitative data
pilot studies
- a small scale ‘dry run’ of the investigation
- would be conducted with a small group of participants to determine questions are clear, ethical and understandable
- any questions that don’t meet this criteria would be re-written, perhaps as a result of suggestions from participants