Research Methods Year 2 Flashcards
What is a case study?
An in-depth and detailed investigation, description or analysis of a single individual, group, institution or event.
They often focus on things which seem unusual, such as rare disorders. However, they can focus on typical cases.
What type of data do case studies produce?
Qualitative data.
A case history of the individual concerned may be created using interviews, observations or questionnaires.
The person in a case study may be subject to experimental testing to assess what they’re capable of, this could produce quantitative data.
How long do case studies take?
Case studies tend to be longitudinal, which means it takes place over a long period of time and may involve gathering additional data from family and friends of the individual.
What are the advantages of case studies?
- They can offer rich, detailed insights which can shed light on unusual forms of behaviour. Therefore it may be more effective than a questionnaire or experiment.
- They contribute to our understanding of typical functioning. HM gave more understanding into typical memory processing and the existence of separate stores in the STM and LTM.
- They can generate hypotheses for future studies and could lead to entire theories (Oedipus)
What are the disadvantages of case studies?
- Generalisation of findings are an issue when dealing with a small sample size.
-The information which makes it into the final report is subjective and based on what the researcher wants.
-Personal accounts from family and friends may be inaccurate, which means the evidence may be low in validity.
What is content analysis?
An observational research technique that enables indirect study of behaviour by examining communications that people produce.
For example, texts, emails, conversations, TV, film and other media.
The aim is to summarise and describe the communication in a systematic way so overall conclusions can be drawn.
What is coding?
The initial stage of content analysis where communication is studied and analysed by identifying each instance of the chosen categories.
Why does coding take place?
Some data sets are large so need to be broken down into meaningful units.
How does coding happen?
It could involve counting the number of times a particular word is mentioned to produce quantitative data.
What is thematic analysis?
An inductive and qualitative approach to analysis that involves identifying implicit or explicit ideas within the data. Themes often emerge once the data has been coded.
How does thematic analysis take place?
- Themes are identified - any idea that is recurrent. Themes are likely to be descriptive.
- Collect a new set of data to test the validity of themes and categories.
- If the new data adequately explains the themes, the researcher will write a final report that uses direct quotes from the data to illustrate the theme.
What are the advantages of content analysis?
-It is useful to get around many of the ethical issues which can arise from most psychological research.
-It is flexible as it produces qualitative and quantitative data depending on the aim of the research.
-The material which an analyst wishes to study is likely to already be in the public domain so there isn’t a need for permission to be obtained.
-High external validity
What are the disadvantages of content analysis?
- People tend to be studied indirectly so the communications they produce are usually analysed outside of the context so may be taken out of context.
-It could be subject to bias as analysts preconceptions could affect research.
-It could suffer from a lack of objectivity when more descriptive forms of thematic analysis are employed.
What is reliability?
Refers to how consistent a measuring device is. This includes psychological tests or observations which assess behaviour.
If a particular measurement is made twice it could produce the same result it is reliable.
What do psychologists tend to investigate?
Abstract concepts such as attitudes, aggression, memory and IQ.
What is test-retest reliability?
A method of assessing the reliability of a questionnaire or psychological test by assessing the same person on two separate occasions. It can also be applied to interviews.
There must be sufficient time between the different occasions to ensure the participant can’t recall their answers but not so long their attitudes change.
What does test-retest show?
This shows the extent to which the test can produce the same answers.
If the same or similar results are produced then it is reliable (strong correlation).
What is inter-observer reliability?
The extent to which there is an agreement between two or more observers involved in observations of behaviour.
It is measured by correlating the observations of two or more observers.
The observers should have similar behavioural categories which could be established in a pilot study and would watch the same events but record data separately.
What does inter-observer reliability do?
It gets rid of subjectivity, bias and unreliability.
What is a general rule of inter-observer reliability?
A general rule is that if the total number of agreements/ the total number of observations is more than +.8 the data has high inter-rater reliability.
What is inter-observer reliability known for interviews?
Inter-interviewer reliability
What is inter-observer reliability known as in content analysis?
Inter-rater reliability
How can reliability be improved in experiments?
Standardised procedures - they are the same each time
How can reliability be improved in observations?
Operationalising behavioural categories and making sure they are measurable/ self-evident.
Categories should not overlap and all possible behaviours should be covered on a checklist to prevent people making their own judgements which causes inconsistency.
If reliability is low, observers may need further training in using the behavioural categories.
How can reliability be improved in interviews?
Ensuring it’s the same researcher each. If this isn’t possible, all interviewers must be properly trained.
This is easily avoided in structured interviews where the interviewers behaviour is controlled.
How can reliability be improved in interviews?
The test-retest method. The correlation should exceed +.8
If a questionnaire has low test-retest reliability, it could require some items to be deselected.
A solution could be to use more closed questions which are less ambiguous.
What is validity?
The extent to which an observed effect is genuine. Whether it measures what it is supposed to measure and can it be generalised beyond the research setting that it is found in.
Can studies produce reliable data which isn’t valid?
Yes
What is internal validity?
Internal validity refers to whether the effects observed in an experiment are due to manipulation of the IV and not something else.
What is a major threat to internal validity?
Demand characteristics - whether participants respond to them and act in a way they believe is expected.
What is external validity?
External validity refers more to factors outside of the investigation. These include whether findings can be generalised to other settings, other populations and other eras.
What is ecological validity?
A form of external validity.
The extent to which findings from a research study can be generalised to other settings and situations.
If the task used to measure the dependent variable is not likely in everyday life, then it has low ecological validity.
What is mundane realism?
The extent to which experimental findings can be generalised everyday life.
What is temporal validity?
The extent to which findings from a research study can be generalised to other historical times and eras.
Whether they hold true over time.
How can validity be assessed?
Face and Concurrent validity
What is face validity?
A basic form of validity where a measure is scrutinised to determine whether it appears to measure what it is supposed to measure.
e.g - does a test of anxiety look like it measures anxiety?
What is concurrent validity?
The extent to which a psychological measure relates to an existing and well-established similar measure.
Give an example of concurrent validity.
A new intelligence test may be given to a group of participants and the IQ scores would be compared to their performance on a well-established test (Stanford Binet test).
Close agreement between the two sets of data would indicate that the new test has high concurrent validity. This is indicated by a correlation of +.8
Which data has higher ecological validity - Qualitative or Quantitative?
Qualitative methods are seen to have higher ecological validity than quantitative methods. This is because in depth and detailed data is better able to reflect a participant’s reality.
How can we improve the validity of qualitative research?
-Interpretive validity which is the extent to which the researchers interpretation of events matches that of their participants.
This can be shown through the coherence of the researchers narrative and inclusion of direct quotes from participants in the report
-Triangulation which is the use of a number of different sources as evidence. For example, data compiled through interviews with family, diaries and observations.
How can we improve the validity of observations?
Observational research may produce findings that have high ecological validity as there is minimal intervention by the researcher - especially in covert observations.
-Behaviour categories that aren’t too broad. If they overlap it could decrease validity.
How can we improve the validity of questionnaires?
- Incorporating a lie scale within the questions to assess the consistency of a respondents response and control against social desirability bias.
- All data being submitted is anonymous
How can we improve the validity of experiments?
-Using a control group because it means that a researcher is better able to access whether changes to the DV were due to the IV
-Standardise procedures to minimise the impact of participant reactivity and investigator effects on validity.
Use single-blind and double-blind trials.
What is a single blind trial?
When participants are not made aware of the aims of the study until they have taken part. This reduces demand characteristics.
What is a double blind trial?
When a third party conducts the investigation without knowing the purpose. This reduces demand characteristics and investigator effects which improves validity.
What are statistical tests?
Tests used to determine whether a significant difference or correlation exists. Overall, it shows whether the null hypothesis should be rejected or accepted.
What factors are used to decide if a statistical test is used?
-If the researcher is investigating a difference or correlation
-What experimental design is being used
-The level of measurement
How would you know if a difference or correlation is being studied?
This should be obvious through the wording of the hypothesis. Correlation can include investigations looking for a relationship or association.
How would you know the experimental design of a study?
This is based on whether it is an unrelated or related design
What are unrelated and related designs?
Unrelated - Independent groups
Related - Repeated measures and Matched pairs
What are levels of measurement?
Quantitative data can be classified into types or levels of measurement.
What are the 3 levels of measurement?
-Nominal
-Ordinal
-Interval
How do you know if the level of measurement is nominal?
-Data is represented in categories
-Data is discrete, which means one item can only appear in one category
How do you know if the level of measurement is ordinal?
-Data has a natural order
-Data does not have equal intervals between each unit
-Data is based on subjective opinion so lacks precision
How do you know if the level of measurement is interval?
-Data has a natural order
-Data does have equal intervals between each unit so is precise
-Data is objective
What is the appropriate measure of central tendency to be used by each level of measurement?
Nominal - Mode
Ordinal - Median
Interval - Mean
What is the appropriate measure of dispersion to be used by each level of measurement?
Nominal - N/A
Ordinal - Range
Interval - Standard Deviation
What is probability?
A measure of the likelihood that a particular event will occur where 0 indicates statistical impossibility and 1 statistical certainty.
How are statistical tests and hypothesees linked?
Statistical tests determine which hypothesis is true and whether we accept or reject the null hypothesis
