Research Methods Flashcards
What is validity?
The extent to which a study provides a true or legitimate explanation of behaviour
What are the two different types of validity?
Internal validity and external validity
What are the three different types of internal validity?
- control
- construct validity
- mundane realism
What are the three different types of external validity?
- ecological validity
- population validity
- historical validity
What is external validity affected by and why?
Affected by internal validity because you can’t generalise the results of a study that lacks internal validity because the results have a no meaning for the behaviour that they were studying
What are the four different types of psychology experiments?
- laboratory experiment
- field experiment
- natural experiment
- quasi experiment
Give three features of laboratory experiments
- highly controlled environment
- IV and DV is controlled
- no randomness
Give two features of field experiments
- natural, more everyday setting
2. manipulates IV but records what naturally happens to the DV
Give two features of quasi experiments
- control DV
2. no manipulating the IV - based on existing difference
Give three features of natural experiments
- takes advantage of pre-existing IV
- no control - opposite to laboratory
- would still have happened if researcher wasn’t there
Give two weaknesses of laboratory experiments
- lacks mundane realism
2. participants may not behave naturally, leading to lower ecological validity
Give two strengths of laboratory experiments
- well controlled = less extraneous/confounding variables = higher internal validity
- can easily be replicated
Give three weaknesses of field experiments
- more time consuming and expensive
- less control = more extraneous/confounding variables = lower internal validity
- participants not aware of being studied = major ethical issue
Give two strengths of field experiments
- increased mundane realism
2. more natural behaviour from participants as their not aware that they’re being studied
Give two strengths of natural experiments
- allow research to be done where the IV can’t be manipulated
- increased mundane realism and ecological validity
Give four weaknesses of natural experiments
- cannot demonstrate casual relationships because IV is not deliberately manipulated
- random allocation of participants to conditions is not possible
- sample might have unique characteristics meaning that the results cannot be generalised to the public = low population validity
- can only be used where conditions vary naturally
Give one strength of quasi experiments
allows comparison between different types of people
Give five weaknesses of quasi experiments
- can only be used where conditions vary naturally
- participants may be aware that they’re being studied = reduced internal validity
- DV may be a fairly artificial task = reduced mundane realism
- random allocation of participants to conditions is not possible
- cannot demonstrate casual relationships because IV is not deliberately manipulated
What is an ethical issue a conflict between
- what the researcher needs to conduct a useful and meaningful research
- the rights of the participants
Ethical issue - how to deal with informed consent
- participants asked formally to indicate their agreement to participate by signing a consent form
- an alternative would be to gain presumptive consent
- offer the right to withdraw
Ethical issue - how to deal with deception
- the need for this should be approved by an ethics committee using a cost benefit analysis
- participants should be fully debriefed and offered an opportunity to discuss any concerns
Ethical issue - how to deal with the right to withdraw
- participants should be told at the start of the study that they have the right to withdraw from the study at any point
Ethical issue - how to deal with protection from harm
- avoid any risks greater than in everyday life
- stop study if any harm is suspected
- offer participants the right to withdraw
Ethical issue - how to deal with confidentiality
- researchers should not record the names of any participants; they should use numbers or fake names
Ethical issue - how to deal with privacy
- do not study anyone without informed consent unless its in a public place and its public behaviour
Ethical issue - limitations of informed consent
- participants knowing all information about study may affect results
- even with informed consent this does not ensure participants really understand what they are involved in
- a problem with presumptive consent is that what people think they will or will not mind is different to actually experiencing it
Ethical issue - limitations of deception
- cost-benefit analyses are flawed because they involve subjective judgements and the costs/benefits are not always apparent until after the study
- debriefing can’t turn the clock back and participants may still feel embarrassed or have lowered self-esteem
Ethical issue - limitations of the right to withdraw
- participants may feel as though they shouldn’t leave because it will ruin the study or are being payed at the end of the study and therefore are more reluctant to leaving
Ethical issue - limitations of protection from harm
- harm may not be apparent at the time of the study and only is judged later with hindsight
Ethical issue - limitations of confidentiality
- sometimes possible to work out who the participants were using information that has been provided
Ethical issue - limitations of privacy
- there is no universal agreement on what constitutes as a public place
What are the four main principles of the most recent code of ethics and conduct?
- respect
- competence
- responsibility
- integrity
What are the three main types experimental designs?
- independent groups
- repeated measures
- matched pairs
Give three ways of dealing with the limitations of the repeated measures design
- researchers may use two different (equivalent) tests to reduce the practice effect
- counterbalancing
- in order to avoid participants guessing the aims of the study, a cover story may be presented about the purpose of the test
Give one way of dealing with the limitations of the independent groups design
- randomly allocate participants to conditions which (theoretically) distributes participant variables evenly
Give two ways of dealing with the limitations of the matched pairs design
- restrict the number of variables to match on to make it easier
- conduct a pilot study to consider key variables that might be important when matching participants
Give two strengths of the repeated measures design
- controls participant variables because the same participants are in all conditions
- you only need half as many participants as IG and MP
Give four weaknesses of the repeated measures design
- the order of conditions may affect performance
- because all participants do all conditions they may guess the aim of the experiment
- one of the tests may be more difficult than another
- a gap may be needed between each conditions to reduce effects such as boredom or fatigue
Give five strengths of the independent groups design
- controls some important variables
- participants unlikely to guess the aims of the experiment
- there would be no order effects because each participant only does one condition
- it doesn’t matter if one test is harder than the other because different participants take parts in different groups
- because there are different participants in each of the conditions they can be tested at the same time saving time and effort
Give one weakness of the independent groups design
there is no control of participant variables (eg. participants in group 1 may be more able than those in group 2) which acts as a confounding variable
Give three weaknesses of the matched pairs design
- very time consuming and difficult to match participants on key variables
- you need a lot more participants than a RM to end up with the same amount of data
- it is not possible to control all participant variables because you can only match on variables known to be relevant but it could be that others are more important
What are the six different observational techniques?
- naturalistic
- controlled
- covert
- overt
- participant
- non-participant
Give two features of naturalistic observations
- nothing is changed
2. increased external and ecological validity
Give two features of controlled observations
- researchers change/manipulate the situation
2. lacks internal validity = increased demand characteristics
Give two features of covert observations
- deception = unethical
2. if observer observes a dangerous situation do they go to help and jeopardise the study?
Give two features of overt observations
- able to ask for informed consent = ethical
2. may cause demand characteristics
Give two strengths of naturalistic observations
- offers a realistic picture of natural, spontaneous behaviour = high in ecological validity
- participants are unaware of being observed and therefore their behaviour is more natural
Give three weaknesses of naturalistic observations
- doesn’t provide information about what people think or feel
- lots of ethical problems if participants don’t know their being observed
- there can be little or no control of extraneous variables so something unknown to the observer may account for the behaviour observed
Give two strengths of controlled observations
- provide a means of conducting preliminary investigations into a new area of research - this could lead to new hypotheses for further investigations
- observer can focus on particular aspects of behaviour
Give one weakness of controlled observations
- environment can end up feeling unnatural and therefore the participants behaviour will become less natural = lacks validity
Give one weakness of overt observations
- the observer may ‘see’ what they want to see (observer bias) and may mean that different observers will see different things (low inter-observer reliability
Give three strengths of covert observations
- offers a realistic picture of natural, spontaneous behaviour = high in ecological validity
- what people say they do is often different to what they actually do, therefore observations may be more valid than self-report techniques
- participants are unaware of being observed and therefore their behaviour is more natural
Give two weaknesses of covert observations
- doesn’t provide information about what people think or feel
- lots of ethical problems if participants don’t know their being observed
Give two strengths of non-participant observations
- likely to be objective because the researcher does not take part in the group being observed
- what people say they do is often different to what they actually do, therefore observations may be more valid than self-report techniques
Give two weaknesses of non-participant observations
- doesn’t provide information about what people think or feel
- lots of ethical problems if participants don’t know their being observed
Give two strengths of participant observations
- may provide real insights of the behaviour from the ‘inside’ which otherwise may not have been gained
- observer can focus on particular aspects of behaviour
Give two weaknesses of participant observations
- more likely to be overt and therefore have participant awareness issues, however if it is covert then it raises ethical issues
- the observer may ‘see’ what they want to see (observer bias) and may mean that different observers will see different things (low inter-observer reliability
What are the five different types of sampling?
- volunteer
- opportunity
- random
- systematic
- stratified
Give three strengths of volunteer sampling
- gives access to a variety of participants
- creating sample requires little effort from researchers (other than creating an advert) as participants volunteer themselves
- there is less chance of the ‘screw you’ phenomenon as participants are often eager to take part
Give two weaknesses of volunteer sampling
- sample is biased because participants are more likely to be more highly motivated and/or with extra time on their hands
- volunteers are eager to please, increasing the chances of demand characteristics
Give one strength of opportunity sampling
- easiest method because you just use the first suitable participants that you can find = less time to locate sample
Give one weakness of opportunity sampling
- inevitably biased because sample is drawn from a small part of the population (eg. selecting people in town on a monday morning doesn’t include professional people or young children) = results cannot be generalised easily
Give three strengths of random sampling
- unbiased as participants are selected using an objective system
- unbiased - all members of the target population have an equal chance of selection
- sample should be fairly representative as each member of the population has an equal chance of being selected, so results should be generalisable to the target population
Give two weaknesses of random sampling
- not truly random/unbiased unless you select a number using a random method and start with this person and then select every nth person
- unbiased selection does not guarantee an unbiased sample (eg. all females could be randomly selected however this is not an equal representation of the population)
Give two weaknesses of systematic sampling
- need to have a list of all members of the population and then contact all of those selected, which may take some time (and they may not consent to participate or be unavailable)
- the process of selection can interact with a hidden periodic trait within the population (eg. if every fifth house in the street is a young person in a flat this is not a representative sample)
Give one strength of systematic sampling
- likely to be representative results of the target population unless certain characteristics are repeated for every nth person which is unlikely
Give one strength of stratified sampling
- likely to be more representative than other methods because there is a proportional and randomly selected representation of subgroups
Give two weaknesses of stratified sampling
- very time consuming to identify groups, then randomly select participants and contact them
- these samples require detailed knowledge of the population characteristics to ensure correct proportions,, which may not be available
What are the two different types of hypothesis?
- directional - enough evidence to predict what will happen
2. non-directional - not enough evidence to predict what will happen
Give two features of correlations
- not a research method but a way to analyse
2. displayed in a scattergram
Give two reasons why correlations are used
- when it is unethical to do an experiment
2. when it is impractical to manipulate variables
Give five strengths of correlations
- can be used when its unethical or impractical to manipulate variables
- there may be other, unknown variables (intervening variables) which explain why the co-variables being studied are linked
- if a correlation is found to not be significant, then you can probably rule out a causal relationship
- if a correlation is found to be significant, then further investigations can be justified
- as with experiments, the procedures can be repeated which means that the findings can be confirmed
Give two weaknesses of correlations
- they may lack internal/external validity
2. people often misinterpret them and assume that a cause and effect have been found, but this is not possible
Give the three types of descriptive statistics
- measures of central tendency
- measures of dispersion
- graphs
What are descriptive statistics used for?
when quantitative data is collected, it needs to be analysed to identify trends or to see the ‘bigger picture’ - this is done using descriptive statistics
Give three types of inferential statistics
- correlation
- graphical display
- parametric and non-parametric difference tests
What are inferential statistics used for?
Once you’ve described your data you need to infer something about it - this means interpreting its meaning in relation to your hypothesis
What are the four levels of measurement in order of most detailed data to least detailed data?
- nomenal
- ordinal
- internal
- ratio
What is nominal measurement?
categorical data (eg. what newspaper do you read?)
What is ordinal measurement?
named categories in order (eg. what grade did u get for each of your GCSE’s?)
What is internal measurement?
we can measure and compare the exact time because intervals are equal (eg. how quick is reaction time? - ruler drop test)
What is ratio measurement?
no categories and is continuous (eg. %)
Give three features of parametric tests
- data must be internal or ratio
- sample data must show normal distribution
- homogeneity of variance is essential
Give three features of non-parametric tests
- used for data that doesn’t meet parametric criteria
- often regarded as less sensitive/powerful
- unlike parametric tests, they require transformation of data
Give three examples of measures of central tendency
- mean
- median
- mode
Give two examples of measures of dispersion
- range
2. standard deviation
Give two reasons why we do statistical testing
- differences between groups are caused by chance
2. a statistical test is calculating the probability that chance has caused that difference
Why do we do the sign test?
we always need to ensure that our results don’t occur due to chance - the sign test is one way of achieving this certainty
Which experimental designs will the sign test work with?
- repeated measures
2. matched pairs design
Show how to do the sign test in six steps
- state the hypothesis
- record the data and work out the sign
- find the calculated value
- find the critical value of S
- is the result in the right direction?
- reporting the conclusion
Give three ways in which psychology and the economy are linked
- mental health costs UK approx £22 billion annually
- current focus in research is dementia
- research weighs up the benefits and costs of therapies (eg. drugs are cheap, but psychological therapy is much more effective yet expensive)
What is the availability heuristic?
The rule that the likelihood of selecting something is linked to its availability used in shops product placement
What is framing?
The way info is presented has an effect on whether it is successful or not
eg. 600 people in a study
test A saves 200 people in study
test B kills 66% of people in study
Test A is ‘framed’ is better because it uses the word ‘saves’ whereas test B uses a more negative word (‘kills’) and therefore people are more likely to choose it even though it is worse.
Give three features of ‘the foot in the door’ technique
- discovered in 1950’s
- the target is given ‘low-stakes; advice that is irrefutable and emotive
- once we agree with something once, it is difficult to change our mind, even if the info changes - coherence heuristic
Give examples of 4 cases that ‘the foot in the door’ can be applied to
- decision making in juries
- mental health treatment
- financial advice
- government programmes (eg. healthy eating campaigns)
Give two case studies which could have been stopped if peer review was considered
- MMR jab case
2. 11+ grammar school scandals
What is peer review?
When a piece of research has been carried out it will be written up as a paper and considered by someone else
