research methods Flashcards
what is an interview?
asking questions in real time and analysing answers after
- information gained directly from participants
define ‘researcher effect’
the influence of the interviewer directly on the respondent
- eg being rude
define ‘researcher bias’
interpretation of response by the interviewer
define ‘response bias’
the tendency of a participant to respond in a particular way
describe the characteristics of an interview
- mostly open questions, generating qualitative data
- smaller data sets lower generalisability
- relevant questions are used and recorded/transcribed which increases reliability
describe a structured interview
- standardised questions asked in same order
- often used in job candidate screening
what are strengths of a structured interview?
- high response rate
- quick
- comparable
- high reliability
what are weaknesses of a structured interview?
- less detail
- can lack validity
what is a semi structured interview?
guide of topics but phrasing and timing is up to the interviewer
used in clinical interviews
what is a strength of a semi structured interview?
- detailed
- flexible
- generates rich, easy to analyse data
what are weaknesses of semi structured interviews?
- could have irrelevant data
- lack reliability
what is an unstructured interview?
everyday conversation without pre determined questions
what are strengths of unstructured interviews?
- deeper discussion
- questions can be adapted and improvised to improve relevancy
what are weaknesses of unstructured interviews?
- low reliability
- time consuming
- less objective
what is a questionnaire?
a self report technique which investigates beliefs and opinions through pre determined questions where respondents record their own answers
what is an open question?
no fixed response
generates qualitative data
what is a closed question?
fixed response
generates quantitative data
what is a strength and weakness of qual data?
- more detailed evaluation
- hard to find averages
what is a strength and weakness of quant data?
- easy to analyse
- extraneous variables not considered
what is a ranked scale?
- respondents assess views on 1-5 scale
- objective
- quantitative data
- easy to draw conclusions
what is a likert scale?
- respondent rates how much they agree
- quantitative data
- allows for degrees of opinion or no opinion
what are characteristics of a good questionnaire?
- no leading / confusing questions
- open and closed questions
- avoid lists that are long or asking 2 diff things in one question = reduces fatigue
- start with interesting questions
- check / test to see if it works
what are strengths of questionnaires?
- large sample
- flexible, quick and efficient
- cost effective
- anonymous = more valid response and reduces social desirability
- standardised procedures = reliable
what are weaknesses of questionnaires
- risk of social desirability to appear favourable
- poor response rate = unrepresentative
- leading qs / researcher bias could flaw results
- written at 11 year old level = age limited
strength of quantitative data
- quick and easy
- large sample
- high generalisability
- objective and numerical
- can be easily replicated
weaknesses of quantitative data
- gathered in questionnaires could be subjective so low validity
- numerical data lacks detail which is unrepresentative of humans complex behaviour
- reductionist - narrows down to one aspect
what is thematic analysis?
A method that analyses qualitative data by finding patterns and themes within a data set.
why is thematic analysis used?
- It allows for flexibility in what the researcher wants to find with specification on any theory
- Can be used to analyse transcripts, secondary data, media, etc
- rich detailed data
Define ‘data corpus’.
A large collection of qualitative data texts being analysed for research.
Define ‘data set’.
A collection of related sets of qualitative data that is composed of separate elements but can be analysed as a whole.
Give an example of a data set in social psychology.
All the answers to an open question in a single questionnaire about obedience.
List the 6 stages of thematic analysis.
1) Familiarisation with the data
2) Coding
3) Searching for themes
4) Reviewing themes
5) Defining and naming themes
6) Producing the report
describe ‘familiarisation with data’
- Read through the data corpus
- If it is audio data, transcribe it
- Note any initial analytical observations
describe ‘coding
- Initial codes of labels using words or short phrases to identify important features of the data
- It can be done manually or with a software program
- Highlighting or post-it notes are a good way to indicate the origin of codes
- Code as many potential themes as possible
- All the data identified under the same code should be collated
describe ‘searching for themes’
- Sort all the codes into broader patterns of meaning of themes
- Mind maps and tables are a good way to sort the codes
- Some codes may form main themes or sub-themes, or even get discarded
describe ‘reviewing themes’
- Refining themes by combining or splitting or discarding on a mind map
- The themes should have a relationship and form a coherent pattern, if it doesn’t then the issue may be with the theme itself or the arrangement of data
- The themes should reflect the data corpus as a whole and the aim of the research
describe ‘defining themes’
- Each theme should be have a concise name and definition to immediately identify the ‘essence’ of each theme
- The researcher should conduct a detailed analysis on each theme (e.g. how the theme fits with the data as a whole)
- An overall narrative of the data will be formed with a final thematic map
describe ‘finalising report’
- Final analysis and writing the report
- The audience must be considered to allow for coherent and appropriate language (e.g. writing for a scientific journal or a newspaper)
- Evidence for each theme should be provided
3 strengths of thematic analysis.
- High inter-rater reliability - two researchers can both code the same data corpus to reach an agreement and remove subjectivity
- High test-retest reliability - due to the standardised procedure of finding codes and collating them into themes it can be easily replicated for other data sets
- High validity - qualitative data is rich and detailed that can provide insight into the respondent
3 weaknesses of thematic analysis.
- Low generalisability - it can get time consuming to replicate this procedure on every open question in a questionnaire and so the sample size may be low to account for this
- Low validity - the different codes and themes may not fully reflect what the data was trying to get across and so is open to misinterpretation by the researcher
- Low validity - the data corpus is qualitative and so open to interpretation when identifying codes and themes therefore has an element of researcher bias
