Research Methods Flashcards
What is an Aim
A statement of a study’s purpose. Stated to make it clear what the study intends to investigate
One-tailed hypothesis (directional)
When there are directions of difference or relationship. Pre-existing research. For example faster or slower
Two-tailed hypothesis (non-directional)
There is a difference but there’s no nature of difference specified, no previous research. Could state a different with no specification of where the difference lies
Null Hypothesis
A prediction that there will be no relationship between the variables in a study or that there will be no difference between the scores from the various conditions of an experiment
Correlation analysis pros and cons
Pros: Can be used for an unethical experiment. If insignificant, you can rule out a casual relationship
Cons: Correlation analysis cannot demonstrate cause and effect between variables. Extraneous variables may lead to false conclusions
Operationalisation
How we measure our variables. Allows others to see exactly how you’re going to define and measure your variables
Random allocation
This means that everyone has an equal chance of doing either condition
Counterbalancing
Half of the pptts are in condition A before condition B and vice versa. This means the first and second condition is not the same for every pptt.
Standardisation
Everything should be as similar as possible for all the pptts. For example, instructions are the same across the conditions
Extraneous variables
Anything other than the IV that could influence your results. Should be accounted for before the experiment
Confounding variables
Anything other than the IV that has influenced your results which hasn’t been accounted for before the experiment begins
Informed consent
Knowing aims and giving permission to take part, could be through a consent letter
Deception
Misleading or withholding information on purpose.
Need approval from ethics committee, full debrief
Right to Withdraw
Able to leave when desired
Make sure they are aware they can leave
Confidentiality
Details should be kept private
Could give retrospective consent
Protection from harm
No more harm than daily life
Research proposals from BPS
Presumptive consent
Ask similar group of people for consent
Prior general consent
Consent for different studies, may involve deception
Retrospective consent
Ask for consent during debrief
Independent groups
Two separate groups of pptts. One group takes part in condition A, the other condition B
Pros and Cons of Independent Groups
No order effect- only one condition
Fewer demand characteristics-pptts only knew their conditions
More pptts are needed
Individual differences as people taking part in each condition are different
Repeated Measures
There is only one group of pptts. This group takes part in both conditions
Pros and Cons of repeated measures
No individual difference as same person does both conditions
Order effect- boredom
Demand characteristics- pptts may perform to meet expectations of experiment
Matched Pairs
Two separate groups, matched into pairs for certain qualities, such as age or anxiety. One of each pair take part in condition A, the other B
Pros and Cons of Matched Pairs
No order effects
Control for individuals, can be more sure the IV changed the DV
Can be difficult to make perfect matches and is costly on money and time
Double blind
Neither pptts or researchers are aware of the aims of the investigation. Used in drug trials, eliminate investigator effects
Single blind
Pptts are not aware of the condition they are in, attempts to control the confounding effects of demand characteristics
Behavioural categories
Devising a set of component behaviours
Event sampling
Counting the number of times a certain behaviour or event occurs in a target individual or individuals
Time sampling
Recording behaviours in a given time frame, like noting what an individual is doing every 30seconds
Controlled observation
When the researcher has some measure of control over the environment, like the bobo doll experiment
Pros and Cons of controlled observation
Control of extraneous variables
Inter-observer reliability
Easy to replicate
Cannot be applied to real-life setting
May be subjective towards what the researcher wants to see
Naturalistic observation
Studying behaviour in a natural setting where everything has been left as it is normally
Pros and Cons of naturalistic observation
High external validity
Generalised to everyday life
Few demand characteristics
Replication difficult- lack of control
Uncontrolled extraneous variables
Covert observation and Overt observation
Covert: pptts not aware they are being observed
Overt: pptts are aware they are being observed
Pros and Cons of covert observation
No demand characteristics
Ethical issues as they are unaware of observation
Pros and Cons of overt observation
Less ethical issues without deception
May be demand characteristics due to known observation
Participant and Non-Participant observation
Pptt: the observer act as part of the group being watched
Non-pptt: the experimenter does not become part of the group
Pros and Cons of Participant observation
Insight, experience in situation
Increased validity
Lose objectivity, ethical issues
Difficulty in recording observations
Pros and Cons of Non-Participant Observation
More ethical
More objective
Less insight, not experiencing
Lower in validity
Structured observation
Researcher determines precisely what behaviours are to be observed and uses a standardised checklist to record the frequency with which they are observed in a specific time
Unstructured observation
The observer recalls all relevant behaviour but has no system
Pros and Cons of structured/unstructured observation
Structured: It is easier to gather relevant data because you know what you are looking for
Interesting behaviours could go unrecorded as they weren’t pre-defined as important
Just the opposite for unstructured.
Inter-rater reliability
Test should give consistent results regardless of who administers it. This can be assessed by correlating the scores that each researcher produces and comparing them
Structured interview
When the questions are decided in advance
Pros and Cons of structured interview
Easily repeated
Easier to analyse
Data collected is restricted
Interviewer bias may still occur
Semi-structured interview
Combination of both structured and unstructured techniques
Unstructured interview
The interview starts with some general aims and questions and then lets the interviews answers guide subsequent questions
Pros and Cons of unstructured interview
Detailed and in-depth info gathered
Can tailor questions
Hard to analyse answers
Requires well-trained interviewers
Primary Data
Data obtained first hand by the researcher for the purposes of the research project. Usually gathered first hand e.g. questionnaire
Secondary data
Information that pre-dates the research project and has already been gathered by others
Pilot Study
A small scale version of an investigation that takes place before the real investigation is conducted. Aim is to check that procedures, materials, measuring scales etc. work and allow the researcher to make modifications if necessary
Questionnaire
Self-report method.
Open- range of answers, unrestricted
Closed- possible answers determined by researcher, restricted
Paradigm
A set of principles, methods and techniques that define a scientific discipline
Empirical methods
Methods of an experiment that are based on data, not just theory. Best way to ensure this is to collect quantitative data with strictly controlled variables
How do theories become science
- Use an experiment to test the theory and publish the results to let other scientists see it to be published in a scientific journal
- Other scientists read it and try to repeat it by replicating the exact experiment and use the theory to make new predictions
- If all the experiments in all the world provide evidence to back it up, the theory is seen as a scientific ‘fact’
- If new evidence comes to light that conflicts the current, the theory is questioned all over again
Peer review
A process used to ensure the integrity of published scientific work. Before publication, scientific work is sent to experts in that field or peers so they can assess the quality of the work. This process helps keep scientist honest, helps validate conclusions. Sometimes mistakes are made and bad science is published- it isn’t perfect.
Popper (1979)
Argued that theories are abstract, so it’s impossible to prove them right through empirical research. Instead he claimed that a theory should be scientific or falsifiable if it can be proved wrong. Therefore every test of a theory should be an attempt to falsify it
How can psychological research impact the economy
People with untreated health disorders may need more time off work, treatments for mental health disorders came from psychological research and can help people continue a normal lifestyle, less time off work is better for the economy. Research into sleep behaviour can help shift workers who’s work disrupt sleep
Laboratory experiments
Control all relevant variables except for one key variable, which is altered to see what the effect is.
Pros: control, replication, ability to establish whether one variable actually causes change in another
Cons: artificial-not real life behaviour, pptts may respond according to what they think is being investigated, deception is often used
Field experiments
Behaviour is measured in a natural environment like a school, a key variable is altered so that it’s effect can be measured.
Pros: casual relationships, ecological validity, demand characteristics
Cons: less control as more chance of confounding variables in a natural environment
Natural experiments
Researcher looks at how an independent variable which isn’t manipulated affects a dependent variable. It isn’t manipulated because it’s an even which occurs naturally.
Pros: possible to study variable that are unethical to manipulate, pptts may not know they are in a study and act natural, ecological validity
Cons: hard to establish relationships between variable, deception often used making informed consent difficult
Quasi experiments
The researcher isn’t able to use random allocation to put pptts in different conditions usually due to the independent variable being a feature such as gender
Pros: controlled conditions, ecological validity
Cons: cannot randomly allocate pptts, hard to establish relationships between variables
Pros and Cons of self report techniques
Pros: you can collect a large amount of information quickly
Cons: bad questions, biased samples, confidentiality can be a problem, people often want to make themselves look good
Alternate hypothesis
If the data forces you to reject your null hypothesis, then you accept your alternative hypothesis instead. You can be more specific and be more precise about how they are linked, using directional hypotheses
Independent and dependent variable
IV- a variable directly manipulated by the researcher
DV- the variable you think will be affected by the changes in the IV
Validity
Internal- The study measures or examines what it claims to
External- The extent of results of the study can be generalised to others
Ecological validity
Generalisable to real life settings- generalised findings from one setting to another setting
Concurrent validity
Results from a new test can be compared to a previously well established test
Population validity
Can you reasonably generalise the findings from your sample to a larger group of people
Temporal validity
Assesses to what degree research findings remain over time
Face validity
Does the test look as though it measures what it intends to measure
Reliability
Internal- The extent to which a test is consistent within itself
External- The ability of the test to produce the same results each time it’s carried out
Inter-observer: test should give consistent results regardless of who administers it
How to prove Inter-Observer/ Inter-rater reliability
Even after you’ve defined the behaviours of interest you must ensure that the observers are putting each pptt in the right category or given the right rating
The Hawthorne effect
If people are interested in something and in the attention they are getting, then they show a more positive response, try harder at tasks. This means their results are often artificially high so the researcher’s conclusion will be invalid, the opposite would occur if uninterested.
Investigator effects
Anything the researcher does which can affect how the pptts behave. A researchers expectations may result in demand characteristics or lead to them asking leading questions.
Must do’s with data analysis from observation
There must be adequate data sampling to ensure that a representative sample has been seen. Language must be accurate and appropriate for the behaviour. Researcher bias must be avoided
Must do’s with data analysis from interviews
Context of a situation or behaviour. Researcher distinguish what is said from how they perceive it. Avoid bias in the selection of data.
Thematic analysis
Involves making summaries of data and identifying key themes and categories. For qualitative data
Pros: preserves the detail in data, new hypothesis allows new insights
Cons: hard to decide which categories to use, or whether to leave out a summary…
Content analysis
Instead of reducing detailed data into ‘numbers’ they are analysed into categories. A representative sample of qualitative data is first collected. Coding units are identified to analyse the data. The data is then analysed to see how often each coding unit occurs. Statistical analysis can then be carried out
Pros: clear summary of patterns, easy replication so + replication
Cons: can be subjective, reducing data to units removed detail
Meta analysis
Where you analyse the results from loads of different studies and come up with some general conclusions. Good way of bringing data together
Skewed Data
Positive- cluster if scores at the lower end of the data set. The mode is less than the median which is less than the mean.
Negative- more scores of data at the higher end of the data set. The mode is more than the median which is more than the mean
Quantitative data
Discrete- counted and put into categories, shown in tables and bar charts
Continuous- measured and shown in line graphs and histograms
Normal distribution
A symmetrical spread of data that forms a bell-shaped pattern. The mean, median and mode are all located at the peak
Criteria for psychology being a science
Hypothesis testing, Theory construction, Falsifiability, Paradigm Shift, Objective/empirical, Replicable- valid, reliable