non-experimental method Flashcards
what is the easiest way to describe the difference between a correlation and an experiment?
correlation is an association between variables (no IV and DV)
experiment is a test of difference- how IV affects DV
how can cause be established in an experiment?
experiments involve the manipulation of one variable (IV) to see the effect on another (DV)
why CAN’T causation be established for correlations?
correlational variables (co-variables) cannot be manipulated. They can only be measured.
what do correlational studies investigate?
the likelihood of two variables being related, it can tell us the strength and direction of an association between co-variables
what is a positive correlation?
as one co-variable increases, so does the other
what is a negative correlation?
as one co-variable increases, the other decreases
what is zero correlation?
there is no relationship between the two co-variables
what are 3 strengths of correlations?
-can study relationship between variables that cannot eb manipulated (e.g. for ethical reasons)
-it is possible to correlate a large amount of data and secondary data can be used
-if two variables are correlated, we know the strength and direction of an association which is often the starting point for further research
what is the main limitation of correlations?
association does not mean causation, cause and effect cannot be established- we don’t know which co-variable causes the other to change
this brings in the third variable problem
what are the three types of observation?
controlled vs naturalistic
covert vs overt
participant vs non-participant
what is a controlled observation?
the observation happens in a carefully controlled and structured environment
what is a good example of a controlled observation?
Mary Ainsworth’s strange situation to measure attachment in children
what is a naturalistic observation?
an observation where the researcher studies behaviour in its real-life or natural setting without any intervention or control from the researcher
what is covert observation?
where the participant does not know they are being observed- it must be behaviour that is happening in public anyway because you can’t get informed consent
what is overt observation?
where the researcher is clearly visible/ not hidden and participants know the are being observed and have given informed consent
what is a participant observation?
where observer goes ‘undercover’ and joins the group that is being observed- they become part of the research
what is a non-participant observation?
where the observer remains separate from those they are studying and observes in a more objective way
give two strengths of a naturalistic observation?
high external validity
gives better insight into real behaviour
give two limitations of a naturalistic observation?
ethical issues
replication hard as many extraneous variables
give two strengths of a controlled observation?
more replicable and more control over extraneous variables
more ethical as people aware of all factors of observation
give two limitations of controlled observations?
low external validity
observer bias
give a strength of a covert observation?
avoids demand characteristics
give a limitation of a covert observation?
ethical issues of watching people without informed consent
give a strength of an overt observation?
more ethical
give a limitation of an overt relationship?
more demand characteristics
give a strength of participant observations?
allows researcher to gain insight which may improve external validity
what are two limitations of participant observations?
observer may become too involved and unable to give an objective account
ethical issues e.g. deception and invasion of privacy
what are two strengths of non-participant observations?
-avoids ethical issues
-observer can remain separate and objective
what is a limitation of all observations?
cannot demonstrate causal relationship
what makes a good observation (6)
- be clear on what behaviours you are looking for by using behavioural categories
-be clear on how you are recording observation
-be clear on how you will analyse data
-consider ethical issues and how you will address them
-has more than one observer; inter-observer reliability
-avoid interfering with the observation
what are behavioural categories?
when a target behaviour is broken down into components that are observable and measurable
what is event sampling?
counting the number of times a particular behaviour (behavioural categories) occurs
what is a strength and limitation of event sampling?
+ good for when a behaviour occurs infrequently and could be missed through time sampling
-important details may be overlooked during more complex observations
what is time sampling?
recording behaviour (behavioural categories) within a pre-determined time frame, record which behaviours can be seen every nth time
what is a strength and limitation of time sampling?
+ reduces number of observations that have to be made
- behaviour sampled may be unrepresentative of the observation as whole
what is meant by inter-observer reliability?
observations should be carried out by at least two observers, with the same behavioural categories, who independently observe the observation
it helps avoid observer bias
if results of the two observers correlate…
it suggests the observation was accurate and has inter-observer reliability
what is the 3 step process for how do ensure inter-observer reliability using same behavioural categories?
- each observer independently rates observation
- correlate findings using a scattergraph
- the observation is reliable if the level of agreement is 80% or above
what is a self-report technique?
any method where a person is asked to explain their own feelings, opinions, behaviours on a topic
what do questionnaires involve?
a set of pre-written questions asking participants about their attitudes, thoughts, feelings and behaviours
what is the difference between an open and closed question?
open- no restriction on how to answer
closed- limited set of set answers
what type of data do open questions produce?
qualitative- wide range of more detailed data but harder to analyse
what type of data do closed questions produce?
quantitative- lacks depth but easy to analyse
what are the 3 types of closed questions?
Likert scale- a statement is given and participants choose their level of agreement
rating scale- similar to Likert but participant chooses a numerical value (not agreement)
fixed choice- participant is able to choose from a range of options
what are the strengths of using open questions?
-provides full and in depth information
-allows participants to answer how they really want to
what are the strengths of using closed questions?
-easy to answer
-comparisons can be easily made between different answers
what are the limitations of using open questions?
-participants may not bother to answer at all
-difficult to compare answers with others
what is the limitation of using closed questions?
-participants may not be able to answer how they want to
what are the strengths of questionnaires in general?
cost effective- allows lots of data to be collected relatively cheaply (since can be widely distributed and doesn’t need a researcher present)
data can be easily analysed
what is the main limitation of questionnaires in general?
people may not answer honestly; response bias, acquiescence bias, lack of self-awareness, social desirability bias
what is meant by response bias as a limitation of questionnaires?
there may be a certain type of person who actually answers questionnaires so findings may not be representative of the target population
what is meant by acquiescence bias as a limitation of questionnaires?
more likely to blindly agree to statements that have been given to them
what is meant by social desirability bias as a limitation of questionnaires?
type of demand characteristic where participants want to portray themselves or be viewed in a positive way, so may not answer honestly
what is an interview? what are the 3 types?
a ‘live’ encounter where the interviewer asks a set of questions to a single participant about their and/or experiences
structured, unstructured and semi-structured
what is the difference between structured and unstructured interviews?
structured- a list of pre-set questions that make up the interview. Set order and no follow up questions.
unstructured- works a lot more like a conversation, interviewer guides the choice of topic but in a free-flowing way. Interviewee is encouraged to elaborate.
what are 2 strengths of structured interviews?
-easily to replicate due to standardized format
-can ask about everything and ensure nothing is missed out
what are 2 limitations of structured interviews?
-likely to be social desirability bias
-participants not free to answer how they want and interviewers cannot ask follow up questions to gain insight
what is a strength of unstructured interviews?
provides rich data- interviewer can ask follow up questions so can get more detailed responses and insight.
what are 2 limitations of unstructured interviews?
-bias can easily occur as interviewer can change questions and may ask them in a leading way
-difficult to make comparisons between responses as not all participants asked same questions
how is interviewer bias reduced?
-standardise all questions
-have a different (independent) researcher carry out the interview rather than the main researcher for a study.
what 4 errors should be avoided when making questions to use in questionnaires and interviews?
-overuse of jargon (unnecessarily complicated terms) that the interviewee may not understand
-use of emotive language and leading questions
-double barrelled questions
-double negatives
explain why using interviews rather than questionnaires may be a better method in some research?
what is the difference between qualitative and quantitative data?
qualitative is descriptive and non-numerical
quantitative is usually given numerically, it is data that can be counted or categorised
what research methods produce qualitative data?
interviews, questionnaires with open questions, case studies and some observations
what research methods produce quantitative data?
experiments, questionnaires with closed questions
what are the strengths and limitations of qualitative data?
+ in depth so gets views and feelings of ppts
-difficult to replicate
-difficult to make comparisons
-often based on subjective accounts, maybe retrospective and open to distortion
what are the strengths and limitations of quantitative data?
+comparisons can be made
+more objective
+trends can be identified
-not in depth
-may produce statistics that look impressive but lack real meaning
what is the difference between primary and secondary data?
primary- data collected for the purpose of the investigation by the researcher.
secondary- data that has already been collected by other researchers and has often been subject to statistical testing so we know the significance level of the data
which research methods produce primary data?
experiments, questionnaires, interviews or observations that are being used for the purpose they were created
which research methods produce secondary data?
includes research conducted by other psychologists, government records, information on websites, employee absence records etc
what is a strength and limitation of primary data?
+authentic and fit for purpose
-takes a lot of time, effort and expensive
what is a strength and limitation of secondary data?
+ inexpensive, requires minimal effort, already statistically tested
-variation in quality of data so might not meet researchers needs
what is a meta-analysis?
the process of combining data from a large number of studies that have been carried out in a particular area of study, to see if findings are consistent
it is a type of SECONDARY data
what types of data does a meta-analysis include?
quantitative and qualitative
what is a strength and limitation of meta analysis?
+ increases confidence in findings so can generalise to wider target pop
-prone to publication bias and the file drawer effect as the researcher may nit include all relevant studies in meta-analysis, which leads to bias result
what is the difference between publication bias and the file drawer effect in a meta-analysis?
publication bias- publishers are more likely to publish ground breaking pieces of information
file drawer- studies are much more likely to remain unpublished if they do not find a significant result
