Observations-research methods Flashcards
research aim
is a statement of what we are interested in investigating. E.g. to investigate if there is a gender difference in mobile phone use.
research question
is a question relating to what is being investigated (?). It is often more specific than a research aim.
What do you need to consider when carrying out an observation?
When to observe (which day of the week, what time, for how long?)
Where to observe (naturalistic or controlled setting)
What to observe (what behaviours are you going to focus on?
How to observe (overt or covert, participant observer or non-participant, event or time sampling?)
Who to observe (sample, sample size, sampling method)
Structured observations
A structured observation is where you look for very specific behaviours and keep a tally of every time you see a certain behaviour. To use a structured observation you need a coding frame. This means you can collect quantitative data.
Unstructured observations
An unstructured observation is where you observe participants without knowing what behaviours you are looking out for. Unstructured observations generally create a qualitative description of what you’ve seen.
Advantages of structured observations
Useful- key guidelines have been set out to find out specif things
Easy to compare results
Disadvantage of structured observations
Spontaneous behaviour is not counted for if it is not in the pre-test list.
We don’t know why we only know the amount of time something happens (a tally)
Subjective it’s coding is not spefic
Example of structured observations
Bandura
Unstructured interview advantages
more detail
more accature as we are recording everything
does not limit data that can be collected
high in construct validity
Unstructured interview disadvantage
Difficult to record all data
low inter rater reliability
very difficult to compare and analyse we cannot replicate
Example of unstructured interview
Stanford Prison experiment
Naturalistic observations
are when you observe people in a natural setting (like in a field experiment).
Controlled observations
are when you put people in a controlled, lab-like environment and observe their behaviour.
Strength of naturalistic observations
Natural behaviours seen (higher ecological validity)
May allow you to record unexpected behaviours
Weaknesses of naturalistic observations
Less control of extraneous variables
Harder to record behaviour (e.g. in distance, other people in the way)
Strengths of controlled observations
Less control of extraneous variables
Harder to record behaviour (e.g. in distance, other people in the way)
Weaknesses of controlled observations
Risk of demand characteristics if participants know they are being observed
Lower ecological validity
What is a coding frame
Set perimeters to know what berhaviours need to be tallied- only needed for structured interviews
behaviour categorizes
List of behaviours your recording
Ethics of observational studies
Observations are only acceptable in public situations where those observed would expect to be observed by strangers.
Informed consent for observations
A poster can be displayed beforehand explaining where and when the observation will be taking place, what behaviours will be observed, and advising people of what to do if they wish to avoid being observed.
Debrief for observations
A poster is displayed afterwards explaining what had happened which means that they can find out about the purpose of the study after and why they were not informed before and giving a contact number if anyone wants to find out more
What ethical issues are protected when the observations is done overtly
informed consent or right to withdraw or no deception as participants are aware of the observation and can avoid taking place is they want to and consent is also not need in a public seeting.
Participant observations
Participant observations are when the observer goes ‘undercover’ and pretends to be in the group of people that he/she is observing.
Non-participant observation
Non-participant observations are when the participants are observed by someone from outside the group (e.g. when CCTV is used).
Strengths of participant observation
Allows you to see behaviours which are difficult to spot
Provides greater depth as you can record reasons for behaviours
Weaknesses of participant observation
Observers may become too involved and lose objectivity
Presence of the observer may affect behaviour
Strengths of non participant observation
More ethical as researchers are not directly impacting the behaviour
Easier to record the behaviour accurately
Weaknesses of non participants observations
Less insight into the reasons for the behaviour shown
View may be obstructed (observing from a distance)
Covert observation
Covert observations involve observing people without their knowledge
Overt observation
Overt observations are carried out in a way the participants are aware they are being observed
Strength of overt observations
Avoids the ethical issues as participants are aware they are being studied
Can be open in tallying behaviours so it’s easier to record data
Weaknesses of overt observations
Participants know they are being observed so could change their behaviour
Biased sample as only those who agree to be observed are included
Strengths of covert observations
Allows natural behaviours to be observed (so more valid - no demand characteristics)
Weaknesses of covert observations
Difficult to record behaviours without raising suspicion
Ethical issues of a lack of consent, participants cannot withdraw
Event sampling
In the event sampling, the observer keeps a count of each time a particular pre-determined behaviour occurs. They record behaviours for the whole duration of the observation.
Time sampling
In time sampling, the time period to be observed in broken up into a series of time intervals, and records of behaviour are related to these time intervals.
Event sampling advantages
All behaviours are recorded (nothing is missed out)
Better for recording behaviours which are not frequent.
Disadvantageous of event sampling
Time consuming and difficult to keep concentration over a long period of time.
Difficult to record all behavior is there is a lot going on
Does not tell you when the behaviours occurred
Advantages of time sampling
Can see when behaviour occurs and if it changes over the time period
Does not require the observer to concentrate throughout the time period
Disadvantages of time sampling
Data recorded may not be full representative of what occurred
Can miss interesting data between time periods
Type and level of data of coding frame
quantitative and nominal level
Observer bias
Observer only records data that fits there aim-this can be unconscious
Observer effect
Participants behaviour differently due to being observed
How can carrying out the observation covertly effect the validity
Participants would show natural behaviours as they are unaware their behaviours is being observed (no demand characteristics) this increases internal validity
This could cause ethical issues as participants would not have informed consent.
How can carrying out the observations in a naturalistic observation effect the validity
Ecological validity is increased as behaviours are likely to reflect what participants typically do
Could mean a lack of control over exteronous variables may make it more difficult for observers to record all behaviors accurately.
How can having an independent observer effect the validity
The observers have no interest in proving a particular theory so will be more objective in recording the behaviour reduced the observer bias which improves the internal validity
They could still interpret the behavior wrong which reduces internal validity
How can maiming the operationalise the coding frame effect the reliability
It can improve the reliability as it should help all observers to be consistent with each other on exactly what will or won’t count in the actual use of the coding frame.
Reliability is reduced as it is very difficult to be precise on exactly what will or won’t count as each behaviour category.
How can training the observer in how to use the coding frame effect validity
It could improve reliability as this should help all observers to be consistent with each other in how they are applying the coding frame
It could be reduced as behaviours could occur which has not been anticipated in training therefore it could still be unclear on how to categorize them
How does running a pilot effect the reliability of the observation
It improves the reliability as the practice run should help bring out any possible issues which could occur
It could reduce the reliability as not all unclear or unanticipated behaviours will crop up in a practice run.
How can using a number of different observers effects the reliability
It improves the reliability as it enables a check on whether inter-rater reliability has been found
It can reduces the reliability as it throws up the issue of what to do if the observers are not in agreement with each other on the data recorded
How a large sample can effect the external reliability
It will help to establish of a consistent trend if behaviours has been found across participants during the observations
It can reduces reliability as it can be time consuming to find or observe enough participants.
Inter rater reliability
Consistency between two or more observers of some behavior and coding and compare-high comparison means high inter rater reliability.
