Please enable JavaScript.
Coggle requires JavaScript to display documents.
Research Methods: Part 3 - Coggle Diagram
Research Methods: Part 3
Peer - review
Before publication in a journal an authors scientific paper is assessed by people who are experts in the same scientific area as the author (peers)
The role of peer review in scientific processes - peer review is conducted ocross the sciences, it is how the quality of scientific work is assessed and ultimately influences how science is carried out by practicsing scientist
Eval: :check: The scientists knowledge that their work will be checked and assessed by feels experts is though to self-regulate the quality of their work
:check: The media often reports scientific findings - the peer-review system aids in science communication helping journalists and the public decide if scientific claims should be trusted
:no_entry: Journals may struggle to find suitable peers in new of very specialised areas
:no_entry: Professional rivalry exits in academic fields - this can result in peers rejecting papers on non academic grounds
-
Double blind peer review: The author does not know who the peers are and the peers do not know who the authors are
Eval: :no_entry: Blinding is problematic authors can be identified by writing styles and anonymous reviewers can be more likely to steal ideas
:no_entry: Publication bias towards publishing positive findings results in the file draw problem - important negative findings never being published
-
Reliability
Saying results are reliable is another way of saying the results are consistent - if the researchers replicate their study exactly, they will get similar results
Range off measurement tools too collect data from pptcs:
- Questionnaires and Interviews
- Experimental conditions
- Observations
External reliability - The extent to which a measure is consistent when repeated e.g. the results of the study are consistent with an exact replication at a different time and/or with different participants
Assessing external reliability: Test-retest - Repeat the study using the same procedures/ measuring devices at different times and test the correlation between the two versions
Inter-rater reliability - Two (or more) observers record behaviours during the same observation using the same behavioural categories; then they test the correlation between each tally of behaviour to identify if the behavioural categories are appropriately operationalised
Comparing with test of correlation - The level of correlation is tested using a test ion correlation such as Pearson's R or Spearman's Rho - a correlation of 0.8 or higher is usually accepted as a strong correlation
Internal Reliability - The extent to which different parts of a measure are consistent with itself e.g. If a 100 question IQ test is divided into two 50 question tests, the results for each set of questions with the same pptc's would be similar
Assessing internal reliability: Split half method - assess measures that test one variable with multiple questions, for example questionnaires or tests of IQ: Split the tests into two parts - pptcs complete both parts - test the strength of the correlation between the two parts of the measure - a strong correlation indicates internal reliability
Improving reliability
Observation - Improving the training given to observers can improve their accuracy in assigning a particular observed behaviour to the correct behavioural catergory
Interviews - Use structured interviews rather than unstructured interviews - the interview will then include a script the interviewer can follow ensuring each pptc has a similar experience and provides answers - can then compare responses
Questionnaires - Use closed questions to reduce the range of possible answers and if there is an established questionnaire that tests for what you need to measure, use that instead of creating a new test
-
Validity
Researchers gather data and produce findings is psychological study - we need to question if it is truthful to say that these findings accurately reflect natural behaviour - psychologists have to consider both the internal and external validity of their results
Internal Validity - Questions the cause and effect relationship between the change the researcher made to the IV and the observed change in the DV - if the change in the DV was influenced by any other factor than the IV (due to lack of control) - the findings lack internal validity
Social Desirability Bias - pptc's hide their genuine opinions/behaviours and instead act/respond in a more socially accepted way to 'look good'
Demand Characteristics - Pptc's think they have discovered the aim and behave in a way they believe will produce the results supporting the researchers theory
Investigator effects - The researchers behaviour influences the behaviour of the pptcs - researcher bias - when the researcher consciously or unconsciously influences the results - the pptc's behaviour could be influenced by the researcher using more positive body language/tone of voice - may also be bias in the way they interpret the pptc's responses
Uncontrolled extraneous variables - Lack of control - this is not using standardised procedures, identifying and eliminating extraneous varibales or not controlling pptc variables by randomly assigning pptc's to groups
External Validity - Questions ion a studies findings can be generalised beyond the study - so from the sample used to the target population and the experimental set-up to other "real world" settings and activities
Ecological Validity - The extent to which the findings of any particular study can be generalised to other environments e.g. moving a test of obedience from lab setting to a busy street, would obedient behaviour be replicated?
Mundane realism - The extent to which the tasks/materials/activities used in an experimental set up are similar to the stimuli used in real-life
Population validity - The extent to which the sample in the study is representative of the target population (gender, age,ethnicity)
Temporal validity - The extent to which the findings from a study can be generalised to other time periods - generally asked of older studies questioning if the findings on topics like social influence, attachment, relationships and gender would be the same if researchers conducted the study in modern society
-
Improving validity
Internal validity - Impoved by demonstrating a high level of control over variables
- Random allocation - controls pptc variables
- Standardised procedure - controls extraneous variables
- Counter-balancing - controls order effects
- Single and double blind trials - controls researcher bias and demand characteristics
- Peer review - controls researcher bias
External Validity - Improved by demonstrating that findings are generalisable
- Replicating findings in multiple settings improves ecological validity
- Replicating findings with diverse groups of people improves population validity
- Replicating historical studies improves temporal validity
Features of science
Scientific concepts
Empirical method:
-
The process of collecting data from direct experience - data gathered from direct observation of pptc's
- Observation
- experimentation
- self-report
- case studies
- Content analysis
Objectivity - Data should be collected and interpreted in ways that avoid bias - data is not influenced by the researchers opinions or expectations
Improving objectivity:
- Systematic data collection - carefully planned out and consistent for each pptc
- Double-blind - Researchers who don't know the research aims collect the data
- Peer-review - identifying bias research
Replicability - Scientist must record their method and produce standardised procedures so that other scientists can repeat their experiment or observation
If same results are found in the replication it increases the confidence and validity of the results
Falsifiability - Karl Popper argued that the ability to collect supporting evidence for a theory is not enough fo the theory to be scientfic - must be empirically tested meaning the theory can be tested in a way that demonstrates it is not true
Example of Black Swan: "all swans are white" - a falisifiable theory - while all previous observations of swans were white a single observation of a black swan in Australia was sufficient to falsify the theory
In psychology - Claiming "human behaviour is due to the existence of the soul which gives us free will" is not a scientific argument - the soul is an unfalsifiable concept (not observable) so it cannot be shown not to exist
Freud's ideas of the ID, Ego and Superego are unfalsifiable - they are not open to observation or empirical experiment therefore cannot be falsified
Paradigm Shift
Philosopher Thomas Khun suggest scientific fields develop in a series of scientific revolutions known as paradigm shifts
Scientists hold their own set of assumptions known as paradigms and they gather evidence to support these views - However sometimes new theories or evidence is generated that dont fit the old paradigm (initially rejected) - After sometime with further evidence to support the new paradigm the majority of the scientific community feels they can no longer support the old paradigm and move to the new paradigm - Paradigm Shift
In psychology: - Early psychologists used introspection to generate theories - Freud used case studies and Wundt used experiments, this was a paradigm shift from earlier religious and philosophical explanations for human behaviour - The movement from psychoanalytic approach to behaviourism was another paradigm shift - the next paradigm shit to cognitive neuroscience occurred around the 1970's - highly controlled experiments and new scientific devices such as fMRI and PET scans
Hypothesis testing
Stages of scientific theory construction: "bottom up" - Observation - psychologists start by observing natural behaviour in the real world
- Construct a testable hypothesis - to allow the observed behaviour to be tested
- Conduct an experiment and gain experimental data - collection of empirical data in controlled conditions. If significant using stats test researcher can claim a cause and effect relationships
- Propose a theory that explains the results
Top down method - Researchers start with unestablished theory and develop a hypotheses to test the theory
The more a theory can withstand testing its assumptions with hypothesis testing, the more confidence there should be in the validity of that scientific theory
-
Types of data
Meta-analysis - collects and combines data from previously conducted and published studies using similar research questions
Eval:
:check: Reliable data - large sample size
:no_entry: Researcher has no control over the quality of the data collected
:no_entry: Significant studies more likely to be posted while non-significant results are less likely to be published
:no_entry: choice of which studies to include could be biased
-
-
-
Secondary Data - Researchers use information previously collected by a third party - another researcher or organisation
Eval: :check: Often exists and is often already analysed - less time-consuming and less costly
:no_entry: Decreased validity as data is not collected for the specific research question
:no_entry: Decrease validity as the researcher had no role in data collection therefore cannot ensure that all variables were controlled
Descriptive Statistics
Stats that summarise:
Measures of central tendancy:
Mode - most common, two modes = bi-modal, more than two modes = multimodal
-
Median - middle number
Eval: :check: Not affected by extremes :check: easy to calculate :no_entry: does not take into account all values
-
-
Percentages:
- x/y X 100 = X%
- Percentage = y-x/x X 100