methodology Flashcards
what is self report data?
research methods where pp’s are asked to report information about themselves without researcher interference eg. their own attitudes/beliefs
strength of self report data
- individual answering for themselves so the data is direct and not likely to be affected by subjectivity of research = less bias
weaknesses of self report data x2
- social desirability: pp may answer how they think they should answer which won’t uncover their own meanings
- people may be in different moods on one day compared to another = data may not be true for all situations
describe a survey as a research method
- questionnaires and or interviews to find out what people think of an issue
- 2 types of questions (open and closed)
- 3 types interview: structured, un-structured, and semi-structured
- a survey gathers information by asking questions of a large number of people usually written in questions or face to face interviews
A01 on questionnaires (description)
- questions usually closed + have likert scale questions too.
- for ethical and practical reasons, questions short so pp don’t give up half way through
- pilot study to family and friends to ensure required info will be gathered
- researcher must consider pp design e.g. independent, repeated, matched
what is external reliability
the consistency of a measure or finding over time
what is internal reliability
the consistency of a measure within itself - can be a problem for questionnaires as often many diff Q’s are used to measure the same trait of attitude = use split-half method (Q’s split into half and findings compared from both halves during analysis)
strengths of questionnaires x2
- reliable: bias reduced from researcher as little variation in how pp are asked for information = no interviewer effect
- reliable: well controlled procedures with same set, same order questions so study is replicable to other groups
weakness of questionnaires x2
- no flexibility, if fixed Q’s asked, valuable data may be missed as pp’s can’t expand on their answers
-social desirability: pp’s may respond with what they are expected to say
What are interviews?
involve set questions such as questionnaires, but face to face situation allows the opportunity to expand, or clarify the questions
why may interviews be chosen instead of questionnaires
- ask follow up Q’s
- respondent may be reassurance
- when access is difficult e..g child with mental health problems
structured interview?
follow a set format and may be extra instructions for using the questionnaire such as where and how to expand on asnwers
unstructured interview
not a set format and questions arise from pp’s answers
semi-structured interview
set questions and some of which can be explored further
strengths of interview x2
- questions can be explained and explored further = in depth and detailed data (both qualitative and quantitative data enrichment)
- interviews gain in-depth data likely to be valid, interviewee’s talk in own words and constrained by questions as they are with a questionnaire.
weaknesses of interview x2
- open ended questions = difficult + time consuming to analyse and may be subject to researcher bias
- social desirability
reasons why survey may be a good research method to use
- possible to generalise
- open and closed q’s = more flexibility than lab exp.
- private allows ppl to be honest = higher eco validity
- more ethical than lab
- quantitative put into graphs and charts when summarising data
- repeated for reliability
what is subjectivity
when the analysis of the results includes input from the researcher
what is objectivity
when there is no bias affecting results, including no bias from researcher’s opinions
strengths of closed questions x2
-generate standard replies therefore numbers can be generated and analysed easily = put in graph to compare data spread
- questions are same for all respondents, if meaning is same for all pp’s then the questionnaire is more reliable.
weaknesses of closed questions x2
- fore a choice of answer as pp’s must choose from a set of answers when the pp might not agree with any of the choices: may not include pp’s choice answer. Therefore the respondent may not say what they want to say producing inaccurate and invalid answers.
- difficult to compare as answers could mean different things to pp’s e.g. ‘unsure’ could mean ‘don’t know’ or ‘sometimes yes/no’ but would be scored the same = results invalid
strengths of open questions x2
- pp’s free to answer as they wish = generates more in depth and richer data.
- pp’s can interpret questions as they wish so more valid data as respondents can talk more about what they really think.
weaknesses of open questions x2
- difficult to analyse because the answers are more likely to be detailed and also different from another= selecting themes can be subjective/biased + % can’t be calculated
- pp’s do not always answer in full as open ended questions are more time consuming to complete, they often to fail to answer all questions.
types of data?
quantitative
qualitative
sampling techniques
-random
-stratified
-volunteer
-opportunity
strengths of quantitative data x2
- data can be summarised in graphs and tables, so are easier to analyse and more likely to be drawn from controlled lab sit.
- well controlled procedures and operationalised variables such as well structured questions so can be replicated to test reliability
weaknesses of quantitative data x2
- specified response are required so tend not to be valid
- demand characteristics (may guess aim of the questionnaire
strength of mean
takes all the values from the raw score into account
weakness of mean
can be skewed by anomalies
strength of median
not affected by extreme scores and useful when the scores are ordered data
weakness of median
does not take account of the values of all the scores and can be misleading if used in small sets of data
strength of mode
not affected by extreme scores and is useful as shows where the majority of scores lie
weakness of mode
tells us nothing about the other scores e.g. there may be more than one mode in the set of data
what is measures of dispersion
how far the data is spread away from the mean, median, mode
weakness of range
effect by extreme scores and does not tell us distribution around the mean
strengths of qualitative data x2
- deals with the ‘why’ rather than ‘what’ e.g. why pp continued shocking to 450V
- conducted in more natural circumstances and tend to produce more ecologically valid data as they are real life situations
weaknesses of qualitative data x2
- difficult to analyse because data can be so different it’s hard to summarise
- qualitative data is harder to replicate due to lack of control in methods and so lacks replicability
analogy for thematic analysis
fat cats slowly roll down walls
name each stage of thematic analysis
familiarisation with data
coding
searching for themes
reviewing themes
defining and naming themes
writing up
familiarisation of data?
reading data to understand the content
coding
generating labels that identify features of data
searching for themes
examine the labels and data to identify patterns of meaning
reviewing themes
checking potential themes against the data to see if they explain the data and answer the research question
defining and naming themes
a detailed analysis of each theme and creating an informative name for each one
writing up
combining together info gained from analysis
strengths of thematic analysis x2
- reduces large amounts of data into manageable summary without losing validity
- encourages researcher to derive themes from the data rather than impose pre-selected themes which is likely to achieve better validity
weaknesses of thematic analysis x2
- choosing themes is subjective
- time consuming
strength of random sampling x2
- low bias therefore representative of target pop.
- can be checked mathematically for bias so more scientific + credible
weaknesses of random sampling x2
- cannot be certain sample is representative of all groups = bias
- hard to ensure everyone in target pop. is available
strength of stratified sampling
all relevant groups have at least some representation so conclusions can be drawn
weakness of stratified sampling
difficult to know how many of each group is needed to be representative ie findings may not be generalisable
strength of volunteer sampling
pp’s more likely to cooperate and are interested so may be less likely to give biased info or go against researchers instructions (demand characteristics)
weakness of volunteer sampling x2
only certain types of people may volunteer so bias - motivation might make them behave differently
- may take time to get enough volunteers
strengths opportunity sampling x2
- allows large number of pp’s to be recruited quickly + not time consuming
- researcher has more control over who is chosen and so be able to get the sample quickly and efficiently because access is not a problem
weaknesses opportunity sampling x2
- those who are picked and willing to take part (self-selecting) = rules out anyone not available/unwilling to take part therefore bias.
- may not get representatives from all groups and so bias is more likely than other methods e.g. researchers more likely to choose pp of their own age, who look friendlier etc..
interviewer effects?
- asked them leading questions
- or may only focus on aspects of pp’s behaviour which fit their expectations
experimenter bias
expectations may also influence how they take measurements and analyse their data, resulting in errors that can lead to e.g. accepting false hypothesis.
what are extraneous variables? + Examples
variables that you can’t control that may affect the results from experiment
e.g. participant variables (age, gender, experience, mood)
situational variables (temp, noise, light)
strength of lab experiment
care controls = replicable, which means they can be tested for reliability
weakness of lab experiment
lack ecological and task validity as not measuring ‘real’ behaviour
experimenter effects?
researcher’s tone of voice, what they’re wearing, gender… etc.. (can affect pp’s answers/actions)
strength of field experiments
ecological validity + lack of demand characteristics as pp’s don’t know they’re in an experiment
weakness of field experiments
extraneous variables + harder to control variables and replicate same experiment
strength of repeated measures
pp’s do all conditions so pp’s variables (features of pp’s that might affect results) are controlled (cancel out)
weakness of repeated measures and matched pairs
order effects = bias and drawing wrong conclusions e.g. practice effect (may do second condition better from having practice and fatigue effects
strength of independent groups
no order effects as people do diff conditions
weakness of independent groups
need more pp’s = takes longer and more costly
