research methods Flashcards
What is independent groups design?
One group does condition A, one group does condition B.
Participants randomly allocated to experimental groups.
What are the advantages of independent groups design?
No order effects e.g. ppts don’t become bored/tired.
Will not guess aim - ppts only tested once, so unlikely to guess aim. So behaviour more natural.
What are the disadvantages of independent groups design?
Participant variables - The participants in the two groups are different, acts as EV. May reduce validity of study.
More participants - Need to find twice as many participants.
What is repeated measures design?
Same person in both conditions. Advantages and disadvantages are opposite of independent groups design.
What is matched pairs design?
Two groups of participants are used but they are related to each other by being paired on certain participant variables that matter to the experiment e.g. two women in their 20s.
What are the advantages of matched pairs design?
Participant variables reduced - because participants matched. This enhances validity of results.
No order effects - Participants only tested once so no fatigue effects etc
What are the disadvantages of matched pairs design?
Matching is not perfect - Matching can’t control all relevant variables. May not address all participant variables.
More participants - Need twice as many participants.
What is a pilot study?
A pilot study is an initial run through of the procedures to be used in an investigation.
What are the aims of running a pilot study?
A pilot study helps foresee any problems with the experiment, e.g. if the design works, if the ppts understand the instructions.
Problems can be tackled before running the main study.
Why does research need to be highly controlled?
To avoid the effects of extraneous variables.
What are extraneous variables?
Any variable you’re not interested in studying that could also have some effect on the dependent variable.
What is counterbalancing?
Counterbalancing is mixing up the order of the tasks, which can solve order effects in repeated measures design.
E.g. half the ppts do task with audience first and then without, other half do without first.
Any order effects would be equal across conditions and so would cancel each other out.
What is random allocation?
Random allocation (e.g. drawing names out of a hat) means everyone has an equal chance of doing either condition.
In a study, if there are more men in one group than the other, the results may be based on gender rather than IV.
Random allocation should ensure groups are not biased on key variables.
What are standardised instructions?
Standardised instructions should insure the experimenters act in the same way with all participants.
Everything should be as similar as possible for all the participants.
What is the independent variable?
The thing that is different between the two conditions.
It has been manipulated by the researcher.
What is the dependent variable?
What is being measured by the researcher.
What is a confounding variable?
If we do not eliminate or control extraneous variables, they become confounding variables - i.e. they affect the results (DV)
What is the experimental method?
The manipulation of the independent variable to measure the effect on a dependent variable.
What are situational variables?
Features of a research situation that may influence the participant’s behaviour and therefore act as EVs.
Include time of day, noise, order effects and investigator effects.
What are order effects?
The order of the conditions having an effect on the participants’ behavior e.g. the participants perform worse in the second condition because they were tired, not because of the IV.
What are investigator effects?
Any cues (other than than the IV) from an investigator that encourage certain behaviours in the participant, leading to a fulfilment of the investigator’s expectations. Such cues act as an EV. e.g. leading questions.
What are participant variables?
The differing individual characteristics that may impact how an individual responds in an experiment e.g. age, gender etc.
Controlled in repeated measures design as same participant used in both conditions.
what is a field experiment?
a field experiment takes place in the real world rather than a controlled environment.
IV is manipulated by researcher.
weakness of field experiment
less control over extraneous variable that might bias the results. makes it difficult to replicate
strength of field experiment
high ecological validity as more likely to reflect real life.
less likelihood of demand characteristics as participants don’t know they are being studied, so don’t change behaviour for experiment.
what is a lab experiment?
a lab experiment is an experiment conducted under highly controlled conditions.
participants are randomly allocated not each IV condition.
strength of lab experiments
easier to replicate as uses standardised procedure.
allows for precise control of extraneous variables and IV. allows cause and effect relationship to be established.
weakness of lab experiment
artificial environmental = low ecological validity as behaviour does not reflect real life
what is a natural experiment?
natural experiments are conducted in everyday environment of participants but experimenter has NO control of independent variable as it occurs naturally in real life.
strength of natural experiment
high ecological validity - behaviour likely to reflect real life because of its natural setting
less likelihood of demand characteristics affecting results as participants may not believe they are being studied.
can be used in situations where it is ethically unacceptable to manipulate IV e.g. researching stress
weakness of natural experiment
no control over extraneous variables that might bias the results - makes it difficult for researcher to replicate study in the same way.
what is ecological validity?
the degree to which an investigation represents real life experiences.
what are experimenter effects?
the ways that an experimenter can accidentally influence participant through their behaviour.
what are demand characteristics?
the clues in an experiment that lead the participants to think they know what the researcher is looking for (e.g. experimenter’s body language)
what are order effects?
changes in participant’s performance due to their repeating the same or similar task more than once.
includes practice effect : improved performance of task due to repetition
and fatigue effect : decrease in performance due repetition e.g. due to boredom or tiredness
what is a one-tailed hypothesis?
alternative hypothesis which specifies a direction.
directional
what is a two-tailed hypothesis?
alternative hypothesis that does not specify a direction.
non directional
what is quantitative data?
quantitative data (think quantity) is numbers-based, countable or measurable data.
tells us how much, how many or how often.
what is qualitative data?
qualitative data is interpretation-based, descriptive data.
Can help us understand why, how or what happened behind certain behaviours.
internal validity
measure of whether results obtained are solely affected by changes in the independent variable.
e.g. construct validity = does a measure measure the concept it’s supposed to (e.g. does a questionnaire actually measure IQ or something different?)
concurrent validity = is a measure in agreement with pre-existing measures that are validated to test for the same/similar concept?
internal validity can be assessed based on whether extraneous variables are controlled or eliminated ; greater the control, greater confidence a cause and effect is found.
external validity
measure of whether data can be generalised to other situations outside of research environment they were originally gathered in.
includes :
temporal validity - high when can be generalised across time.
ecological validity - whether data is generalisable to real world. e.g. lab experiments can have artificial tasks such as remembering long lists of insignificant words, which does not really happen in real life. so have low ecological validity.
external validity can be improved by replicating a study at different times and places, and obtaining similar results. for example, confidence in the generalisability (and in turn external validity) of results is increased when research is successfully replicated across cultures.
what are the ethical guidelines in psychological research?
set by BPS
consent - informed consent given?
deception - have participants been deceived? if so how could this be avoided.
withdrawal from investigation - participants should be told they have a right to withdrawal at any time, and can have their data withdrawn at later date.
anonymity & confidentiality- participants have a right to remain anonymous
protection from harm - participants have a right to be protected from physical and psychological harm.
what is meant by briefing?
giving participants information before they begin a study - opposite of debriefing
If participants are deceived, what other ethical issue does this automatically create?
Lack of informed consent
sign test
allows researcher to determine significance of their investigation/findings.
used in repeated measures design, where data is nominal.
step 1 = is difference between columns - or +?
step 2 = add up number of + and -. ignore no difference participants.
step 3 = find least occurring sign. this = s
step 4 = find n = number of participants that have difference.
step 5 = find critical value. one-tailed or two-tailed? sig level = 0.05. use n to find critical value.
step 6 = is s equal to or less than critical value? if not, not significant.
what is s in sign test?
calculated value
what would the layout be for answer in sign test?
the calculated value (s = 5) is greater than the critical value of 2. therefore the difference in depression score before and after therapy is not significant.
the null hypothesis should be accepted.
what is a random sampling?
each member of the target population has equal chance of being chosen e.g. pulling names out of a hat
advantage of random sampling
avoids researcher bias - researcher can’t choose participants they want to form the sample, which could avoid them picking participants that give preferred result
disadvantage of random sample
as randomised, may be unrepresentative sample e.g. too many women
time consuming to compile list of sample
what is systematic sampling?
every nth person picked from target population e.g. every 5th person from register.
advantages of systematic sampling
no researcher bias
can be quick if list already exists e.g. a school register
disadvantage of systematic sampling
random so could lead to unrepresentative sample e.g. too many women
what is opportunity sampling?
the researcher directly asks any members of the target sample available to take part.
advantages of opportunity sampling
quickest and inexpensive way to get sample
disadvantages of opportunity sampling
as researcher decides who to ask, there may be researcher bias.
participants chosen likely to be from sample unrepresentative of entire target population e.g. usually university students
what is volunteer sampling?
participants offer to take part in research after learning about the research e.g. on a poster or in a newspaper.
poster would usually be put where target population would see it e.g. in a pub for alcoholics
advantages of volunteer sample
advertisements can reach a large number of potential participants, especially if on newspaper or online.
easy sample to collect as participants put themselves forward, so don’t have to contact people and ask.
disadvantages of volunteer sample
volunteer bias - people who are likely to volunteer for study have different characteristics to rest of target population e.g. they are friendlier and have more free time available.
we also want people who are busy and unfriendly to generalise results to target population.
psychology sampling - western vs non- western
american university students are 4000x more likely to take part in study than random non-westerner. so are studies representative?
what is a stratified sample?
Subgroups (strata) are identified along with their proportion in the target population e.g. ethnicity, gender.
Random sampling then used to select number of participants required from within each subgroup.
e.g. 10% graduates and 90% non-graduates in target population = 1 graduate and 9 non-graduates in sample.
advantages of stratified sample
representative of target population, so results should be generalisable to target population.
as participants randomly chosen from each subgroup/strata, this avoids researcher bias.
disadvantages of stratified sample
not all subgroups accounted for, only ones that are judged to be important to study
time consuming
