Please enable JavaScript.
Coggle requires JavaScript to display documents.
Research Methods: Part 2 - Coggle Diagram
Research Methods: Part 2
Observational designs
Inter-observer reliability - many researchers carrying out the same observation their reports are then compared and score:
Total number of agreements / total number of observations x 100 .
The score that shows high inter observer reliability is any score above 80%.
Unstructured - consists of continuous recording where the researcher write down everything they see during the observation
Eval: :check: More richness and depth of detail. :no_entry: Produces qualitative data which is more difficult to record and analyse
:no_entry: Greater risk of observer bias e.g. only record 'catch the eye' behaviours
Structured - Here the researcher quantifies what they are observing using predetermined list of behaviours and sampling methods
Eval: :check: Easier as is more systematic.
:check: Quantitative data is collected which is easier to analyse and compare with other data
:check: less risk of observer bias
:no_entry: Not much depth or detail
Whilst conducting structured observations, behavioural categories can be used. This is when a target behaviour which is being observed is broken up into more precise components which are observable and measurable e.g. aggressive behaviour can be broken down to - shouting, punching, swearing etc.
Event sampling: This involves the counting of the number of times a particular behaviour is carried out by the target group or individual you are watching
Eval: :check: Good for infrequent behaviours that are likely to be missed if time sampling
:no_entry: If behaviour is very frequent there could be counting errors
:no_entry: It is difficult to judge the beginning and ending of a behaviour
Time Sampling: This is the recording of behaviour within a time frame that is pre-established before the study
Eval: :check: Reduces the number of observations that has to be made so is less time consuming
:no_entry: Unrepresentative of the observation as a whole
Self-report Techniqes
-
Interview: PPtc's give response to direct questioning from the researcher - in person/over the phone/ video call
-
Eval: :check: Can rephrase the questions that are difficult to understand :check: Can build rapport meaning the pptc may take the interview more serious :no_entry: Requires an interviewer which significantly increases the cost per pptc
Questionnaires: Pptc's give information in response to a set of questions that are sent to them - in the post/ completing a form online
Eval: :check: Dont require a trained interviewer and can be distributed easily - collecting a large amount of data - cheap and easy :no_entry: Questions the pptc's do not understand cannot be rephrased
Open questions: The questions phrased in a way that allows the pptc to answer in anyway they chose - produces qualitative data
Eval: :check: As the pptc's can shoes how they respond this can lead to more valid (true) responses
:no_entry: Qualitative data makes the data analysis between large sets of data more difficult and harder to spot patterns
Closed Questions: The question is phrased in a way that limits the pptc's response - produces quantitive data
Eval: :check: Allows easy data analysis between large sets of data - easier to spot patterns
:no_entry: Pptc';s responses limited to a fixed set it can be argued that it is less valid
-
Correlations:
In a correlational study no variables are manipulated - two co-variables are measured and compared to look for a relationship
Co-variables - the two factors/ variables that are measured / collected by the researcher the compared to each other
-
Scattergram - A graph used to the plot the measurements of two co-variables. Scattergrams visually display the relationship between the co-variables
-
Eval: :no_entry: Correlation does not show causation
:check: Correlations suggest potential cause and effect relationships which can then be tested with experimental methods
Content analysis
An indirect observational method that is used to analyse human behaviour investigating through studying human artefacts
-
To perform Content analysis:
- Decide a research question
- Select a sample - randomly/systematically from a larger quantity of all possible data
- Coding - researcher decides on categories/ coding units to be recorded e.g. occurrences of a certain word -based on the research question
- Work through the data - tally the number of times pre-determined categories are mentioned
- Data analysis on quantitative data to find patterns
-
Example: Lonely Hearts advert - 881 lonely hearts adverts were taken fro newspapers in America
Coding units for physical attractivness:
- Both genders (Athletic, cute, fit, good-looking, healthy)
- For men ( Handsome, Hunk, Muscular)
- For women (Buxom, petite, pretty)
Coding units for resources included:
- Homeowner, professional status, College educated
Research backed up the evolutionary theory that women are looking for older men with resources (so they can supply for them) and there was less focus on physical attractiveness and men were looking for younger and attractive women (cared more about the physical appearance than resources)
-
Eval: :check: Use of artefacts - not usually created for research but are taken from the real world = high external validity and findings should be generalisable
:check: Replicable
:no_entry: Researcher has to interpret data - may lead to observer bias
Thematic analysis - Researchers attempt to identify the deeper meaning of a text by reading it first and allowing themes to emerge - Themes are NOT predetermined
Eval: :check: Stop researcher imposing their own bias
:check: High external validity
:check: Easy to get a sample
:no_entry: Data not created in controlled conditions
Case Studies
Rage on data collected from an idividual, group or institution - mainly collected using interviews and observations - content analysis can be performed on written evidence = high level of detail
Qualitative data - main form of data collected due to the use of interviews - Quatitative data can be included with the use of experimental techniques, observations and content analysis
Snapshot = short term case study
Longitudinal Case Study = Long term case study - follow up pptc's after many years
Use in clinical Psychology - Unusual behaviour of individuals with brain damage - Tan (Louis Leborgne) only able to communicate using the word Tan - this helped to identify Broca's aea
Use in psychodynamic psychology: Freud used case studies to support his theory - Little Hanz for Oedipus complex
-
Eval: :check: In depth mostly qualitative investigations - range of stat collected results in realism
:check: Case studies can be used to generate hypothesis
:no_entry: Finings from one individuals case study cannot be generalised
:no_entry: Can suffer researcher bias - researchers decide what data to include
Variables
Variables in a correlational study: Co-variables are measurements that have been made to see if they are associated - assessed for a relationship
Variables in an experimental study: Independant variables are variables that the researcher manipulates. The dependant variable is what the researcher us measuring - causal relationship between the manipulation of the IV and then measurement of the DV
Crucial that the variables are operationalised - clearly stating exactly what the variable is and how it should be measured
Extraneous variables: Any variable aside from the IV that may affect the measurement of the dependant variable - may cause an error showing a causal relationship between IV and DV that isn't really there e.g. Demand characteristics, pptc variables, order effects, situational variables
-
Confounding Variables: Type of extraneous variables - variable other than the IV that changes systematically between levels of the IV - this then hides the true effects (or lack of effect) on the DV
Ethics
Practicing psychologists are expected to design and conduct research that follows a set of guidelines published by the BPS (British psychological society) and the APA (American Psychologist association)
British psychological society's code of ethics - "designed to inform and assist our members in the practical and professional application of psychology, from research and teaching to clinical practice"
'first principles' =
- Respect for autonomy, privacy and dignity of individuals, groups and communities
- Scientific integrity
- Social responsibility
- Maximising benefit and minimising harm
Ethical issues:
Informed Consent - Consent is not valid if patch are not informed of what they are agreeing to - before the experiment patch should be made aware of the aims - link to the principle that we should not deceive our pptcs
Right to withdraw - Pptc's should be told that they can withdraw from the study at any stage with no adverse consequences - includes withdrawing data collected by them
Protection from harm - The researcher is responsible for designing research that does not risk the psychological well-being, physical health, personal values and dignity of the pptcs
Confidentiality - Pptc's personal data should stay confidential and secure - data should not include their identity
-
Dealing with ethical issues:
Informed consent -
Prior general consent: Pptcs agree to a long list of potential features of a research study, not knowing which aspects will be part of the study
Retroactive: The researcher asks for consent after the pptc has taken part in the study - if the pptc does not agree their data is destroyed
Presumptive consent: Researcher asks a group similar to the sample if they would agree to take part in the research and if they agree then the researcher assumes that the pptc would agree
Deception -
Cost - benefit analysis: Assessing the potential harm to the pptcs and comparing this to the potential benefits of research to society