observational designs Flashcards
observer bias
Observer bias happens when a researcher’s expectations, opinions, or prejudices influence what they perceive or record in a study.
behavioral categories
A system of behavioural categories is a list of examples of behaviour that is likely to occur during an observation
These examples should be observable, objectively defined, operationalised and unambiguous
sampling methods (methods to record behavior)
event sampling and time sampling
event sampling
this involves counting the number of times a particular behavior is carried out by the target group or individuals researchers are watching
event sampling strength
its good for infrequent behaviours that are likely to be missed if time sampling was used
event sampling weakness
if complex behaviours is being observed important details of the behaviour may be overlooked by the observer
time sampling
involves recording behaviour within a pre-established time frame
how is time sampling carried out
e.g the researcher will set a regular time interval of 30 seconds where they record the behaviour for 30 secs and continue to do this until the observation is complete
-the researchers will then watch the recordings and make note of all the behaviours that align with the behavioural category
when is time sampling most suitable
when there are multiple behaviours that the researcher wants to observe
and when these behaviours happen very frequently
time sampling strength
-reduces the number of observations that need to be made=recordings can be watched back repeatedly
-most appropriate for recording ongoing behaviours that are likely to occur in a social setting e.g conversations,playing together
time sampling weakness
-the small amount of data collected within a time frame may be be unrepresentative of the observation as a whole
unstructured observations
consists of continuous recording, where the researchers write down everything they see during the observation
strength of unstructured observation
-more richness and depth of detail=qualitative data produced
weakness of unstructured observations
-produces qualitative data which is more difficult to record and analyse
-greater risk of observer bias e.g only record eye catching behaviour
structured observations
where the researcher quantifies what they are observing using a predetermined list of behaviours and sampling methods
strength of structured observation
-easier to conducted as it is more systematic
-qualitative data is collected which is easier to analyse and compare with other data
-there is less risk of observer bias
weakness of structed observation
-not much depth of detail
-difficult to achieve high inter observer reliability as filling the predetermined list in is subjective
inter-observer reliability
measures the degree of agreement between different people observing the same thing
what happens if researchers conduct observations alone
single observers may miss important details or may only notice events that confirms their opinions or hypothesis =introduces bias into the research
how to increase inter-observer reliability
-observations should be carried out by two researchers
-these observers must familiarize themselves with the behavioural categories being used
-the researchers could conduct a pilot study and observe the same behaviours at the same time
-the observers should measure inter observer reliability by using a correlation
Why should behaviour categories be used in an observation?
Using a set of behaviour categories should enable the recordings of X behaviour made by the x(two students) to be consistent. This establishes into observer reliability.
= the observations can be carried out reliably and there will be less subjective interpretations of what is X and what is not
