Non-Experimental Methods Flashcards
What is an observation?
when a researcher does not manipulate a variable to see what effect this has on another variable, instead, they simply record what happens
What is a naturalistic observation?
an observational study conducted in the environment where behaviour would normally occur e.g. work, school, public streets
What are the strengths of using a naturalistic observation?
- Participants in natural environment, likely they are doing what they would normally be doing - high ecological validity
- Often participants do now know they are being observed so behaviour unaffected by demand characteristics or evaluation apprehension
- Used in situations where other techniques unethical/impractical (e.g. studying children who do not respond well to experiments)
- Depending on the type of observation, detail may be richer and more detailed than in many experiments
What are limitations of using naturalistic observations?
- Since no variables are manipulated, there is no control over extraneous variables, cause and effect relationships cannot be found
- May be possible observer bias influences results
- Replication can be difficult - hard to check reliability
- Ethical concerns due to privacy
- Also, confidentiality
How can observer bias be reduced?
Using inter-observer reliability
What is a controlled observation?
an observation study where the researcher controls some variables - often takes place in a lab setting
What are strengths of using a controlled observation?
- Less risk of unwanted extraneous variables influencing behaviour than with naturalistic observations which increases researcher’s ability to interpret the findings
- Easier to replicate due to element of control to check findings for reliability
- Participants would have agreed to come in to be observed, less of an issue with consent
What are the limitations of using a controlled observation?
- Participants will know they’re being observed - may affect behaviour due to demand characteristics or evaluation apprehension, may reduce the internal validity of the research
- Observer bias, may influence the results to fit their theory
- Being in a controlled environment may affect how participants behave, especially if that environment is artificial which may reduce the ecological validity
What is a covert observation?
When the participant is not aware they’re being observed
What are strengths of using a covert observation?
- Removes the problem of participants’ knowledge that they’re being observed - increases the internal validity
What are limitations of using a covert observation?
- Ethical issues - cannot give consent
What is an overt observation?
participants are informed that they are being observed
What are strengths of using overt observation?
- Removes problems with consent
What are the limitations of using overt observations?
- As they know they’re being observed, may change their behaviour which will obviously affect the internal validity of the research
What are participant observations?
an observation study where the researcher actually joins in with the group or takes part in the situation they are observing
What are the strengths of participant observations?
- some groups hard to research without joining them e.g. religious cults, and participant observations offer a way to study these kinds of groups
- the direct involvement of the researchers in the group being studied means they can develop a deep understanding of the motives and behaviour of those being observed
- additional source of information
What are the limitations of participant observations?
- objective information becomes more difficult as de to subjectivity, there may be bias as they are part of the group and make the group appear more positively
- may be difficult to record observations at same time as taking part, may have to record these later which may be unreliable
- ethical issue - participants may disclose information to researcher who is part of the group that they would not disclose to someone they thought was separate to that group
What is a non-participant observation?
when the researcher does no become part of the group being observed
What are the strengths of a non-participant observation?
- as researcher is not part of the group, presence in less likely to affect behaviour of the group members, particularly if it is covert - internal validity higher
- observer is able to be more objective if they are not part of the group - they are not biased to portray the group in a positive light
- observations can be recorded at the time, reducing concerns about unreliability of the observer’s memory affecting the data recorded
- no ethical issue of betrayal of trust
What are limitations of using a non-participants observation?
- as the observer is not part of the group, they lack access to the objective experience of group membership
- not being in the group makes it harder for the researcher to gain a deep understanding of the meaning of the behaviour being observed
- difficult to make observations over a long period of time without being part of the group
What is an unstructured observation?
when the researcher simply records all of the relevant behaviour
What are strengths of unstructured observations?
- allows for rich, detailed descriptions of the behaviour produced
What are limitations of using unstructured observations?
- may be too much to record
- open to observer bias as they only record behaviours that are interesting to them personally
What are behavioural categories?
key behaviours, or, collections of behaviour, that the researcher conducting the observation will pay attention to and record - involve coding, or rating behaviour, rather than trying to describe it
What makes good behavioural categories?
- observer should not have to make interpretations or inferences about the behaviour, simply record explicit actions
- all relevant behaviours are covered
- no behaviour ends up being recorded in more than one category (they are mutually exclusive) do not overlap and are unambiguous - operationalised
What are strengths of using behavioural categories?
- operationalised categories are more objective - less chance of bias
- easier to check for reliability, inter-observer reliability means observations between two sets of observations are reliable (+0.80 correlation) if this is not found, then we know the observations are unreliable. Observers should be trained in how to use the categories, to help improve the reliability
What are limitations of using behavioural categories?
- operationalising behaviour in this way may lose some of the richness of the information and we may not know the context behind it
- if a relevant behaviour occurs, but there is no category to record it in, then that behaviour goes unrecorded (pilot studies are often used to try and address this issue, but there is always the possibility that something new may occur with no category)
What is event sampling?
a target behaviour is identified and the observer records it every time it occurs (kept like a tally)
What are strengths of using event sampling?
- produces quantitative data, easier to analyse and check for reliability
- reduced the possibility of the observer only focusing on one thing, and ignoring other behaviours
What are the disadvantages of using event sampling?
- a possible problem is unanticipated behaviours that are relevant but are not recorded, and so the observation may not reflect what actually occurred
- takes no account of when the behaviour occurred, or the order in which they occurred which may be important
What is time sampling?
a way of sampling the behaviour that is being observed by recording what happens in a series of fixed time intervals (e.g. every 30 seconds)
What are strengths of time sampling?
- avoids the problem of the observing only recording behaviours that they find interesting
What are the limitations of time sampling?
- relevant behaviour may occur between the time intervals and so the observation may not reflect what actually occurred
What are self-report techniques?
asking participants to report on their own thoughts/feelings (questionnaires and interviews)
What are questionnaires?
predetermined set of written questions that participants fill in themselves, are always structured, can be completed with or without the presence of the researcher
What are the strengths of using questionnaires?
- quick and easy to administer, cost-effective way of collecting large amounts of data as they can be completed in the absence of the researcher can be very large samples of participants (especially when conducted via the internet)
- standardised questions, easy to replicate and check for reliability
- anonymous, reduce chances of social desirability bias
- if completed out of presence of researcher, removes the problem of experimenter bias
What are the limitations of using questionnaires?
- difficult to avoid preparing questions that participants will not misunderstand, or understand differently
- participant samples may be biased - e.g. if people have a lot of time on their hands
- respondents may not be completely truthful
- researcher is unable to ask any followup questions
How should questionnaires be constructed?
- consider the order, should start with questions that put participants at ease and then continue onto more important ones
- some items to disguise the aim of the research and reduce the chance of demand characteristics
- avoid leading questions
- designed to reduce the possibility of an acquiescent response set
How do you check the consistency of a questionnaire?
- test-retest method: questionnaire given to the same sample twice, with a suitable interval in-between - if responses are the same on both occasions, then the questionnaire is reliable
- split-half technique: items on questionnaire randomly split into two groups, scores from two groups of items and compared and if the test is reliable, both halves of the questionnaire should give similar scores
How do you check the validity of a questionnaire?
- face validity: do the items look as if they’re measure what it sets out to measure (subjective and not a true test)
- give sample whose attitudes about the topic the questionnaire is exploring are already known - if questionnaire is reliable, results of the questionnaire will reflect the already known views
What are closed questions?
questions where there are fixed choices of responses e.g. yes/no determined by the researcher - they generate quantitative data
What is a likert scale?
a number from 1-5 is allocated to each of the range of statements e.g. strongly agree/ agree/ neither agree or disagree/ disagree/ strongly disagree (quantitative data)
What are the strengths of using closed questions?
- easy to analyse responses and use the data to check for correlations or check hypotheses, also to check for reliability - also saves researchers time
- as questions are easy to answer, easier for participants to complete the questionnaire which means it is likely the researchers will obtain a larger sample of data
What are the limitations of using closed questions?
- data may lose richness and detail
- respondents cannot elaborate on their answers
- set of responses may not include the answer the respondent actually wants them to give, forcing them to choose an answer that is not really correct for them
What are open questions?
questions that allow the respondent to give whatever answer they like, rather than choosing from a set of fixed choices - free response method produces qualitative data
What are the strengths of using open questions?
- rich, detailed data is obtained, participants can elaborate as they wish, explaining and qualifying their answers
- respondents are not forced to choose an inappropriate response from the limited options and better represents their actually views
What are the limitations of using open questions?
- the data is harder to analyse, so it is harder to use the data to check for correlations or test for hypotheses, or check for reliability
- participants may not want to write detailed answers and as a consequence have fewer participants completing them
What is an interview?
participants asked questions by the researcher, rather than having the questions in a written form - can be structured or unstructured
What is a structured interview?
interview with a predetermined set of questions asked in a fixed order
What is an unstructured interview?
may involve some questions that are formulated in advance, but order and what other questions asked, depend heavily upon responses by the participant
What is a semi-structured interview?
all participants asked similar, though not necessarily identical questions, but the order of these and precise questions asked will depend on answers given
What are the strengths of using interviews?
- structured: use standardised questions, easy to replicate the research and check for reliability, possible to compare responses from different participants more easily as they have been asked the same questions
- good way of dealing with complex and sensitive issues, can respond to how the participant is, moving away from difficult topics if it is clear they are upset
- allow researcher to explore any areas of interest that arise in more depth
- the variety of interview types makes them flexible, allowing for both quantitive and qualitative data to be collected
- any misunderstandings on part of the participants can be correct by the researcher
What are the limitations of using interviews?
- respondents may not be truthful, social desirability bias may occur
- interviewer effects may also bias responses
- interviewers need to be properly trained
- ethical issues may arise when discussing sensitive subjects - participants may become distressed or admit more than they wish - hard for a participant to say that they wish to stop
- respondents may be unable to put their words into true feelings about a topic - leading to a source of bias
- detailed responses make it hard to analyse data and hard to check for correlations, test hypotheses or check for reliability
- researcher has to be present for ever interview, far more time consuming that questionnaires and small samples tend to be obtained
What else should be considered when designing interviews alongside whether it is structured, unstructured, or semi-structured?
- Who is the most appropriate interviewer?
- The order of questions?
- The way the questions are phrased - leading questions should be avoided and careful phrasing on sensitive topics
What is a correlation?
a quantitative method where the values of two variables (co-variables) are calculated from each participant and the correlation coefficient is calculated to see if a relationship exists between them
What is a positive correlation?
as one variable increases, so does the other
What is a negative correlation?
as one variable increases, the other decreases
What is a correlation coefficient?
the degree to which two variables are related, a number between -1 and +1, 0 indicates no correlation
What are the strengths of using correlations?
- enabled predictions to be made
- correlation coefficient allows us to know not only whether a relationship exists, but what type, and the strength of it
- ethical way of conducting research as there is no manipulation of variables
- allows research to be conducted where manipulation of a variable is impractical/impossible
- since large amounts of data on a number of variables can usually be collected easily, it is possible to check whether or not associations exist between a large number of variables
What are the limitations of using correlations?
- cause and effect relationships cannot be found
- may be another variable involved
- correlational studies only work for linear relationships
What is the difference between correlations and experiments?
experiments involve the manipulation of the IV and measuring the effect on the DV, while controlling the other variables, however, in correlations, no variables are manipulated, only measured