Observations in psychology: types, strengths and weaknesses

Observations

1. observations are where researchers observe behaviour in real life or in a lab without manipulating an IV. the method is useful when it is not possible to manipulate the variables or it can be the starting point for further research.

2. Types of Observation

   • Participant Observation
     • where researchers join the group being studies and observe them
       • very high in EV if the participant does not get caught
         • very detailed and accurate
           • first hand information
       • difficult to record data
         • almost impossible to replicate
           • ethical problems with using people without their knowledge and deceiving them
   • Non-Participant Observation
     • where the researchers observe from a distance and remain external to the group being studied
       • high EV
         • lower chance of DC
           • can be used to validate findings from other research
             • can be used to find out things like animal behaviour which may not be able to be studied in a lab
       • can't really tell cause and effect in variables that are observed and not manipulated
         • lack of control over the conditions and not manipulated
   • Structured Observation
     • where a specific coding or rating system for recording data is used
       • more control = easier to replicate or do again
         • easier to check for inter-rater reliability
       • lower EV than other two methods
         • participant reactivity where participants play up because they know they are being watched - like DC
     • Time Sampling/ Event Sampling
   • Unstructured Observation
     • you can write freely about what you are observing
     • behaviour might be missed
       • impossible to check for inter-rater reliability
         • high chance of observer bias
   • each of these types of observations can either be OVERT (in the OPEN) or COVERT (under COVER)

3. Observer Bias

   • problem with observation studies
     • it is where the observer only records data or may interpret behaviour so that it fits in with their aim or hypothesis
       • this would reduce the validity of the results
         • the problem can be reduced by having more that one observer.
           • however this may lead to a further problem of inter-rate reliability

4. Inter-rater Reliability

   • this is the extent to which different researcher observing the same behaviour at the same time and record the same results
     • same results = IRR is high
       • different results = IRR low
     • IRR can be increased in a observation study by ensuring that the categories of behaviour are clearly understood and the the observers are trained in the recording shhet
       • researchers could also carry out a PILOT STUDY to identify any problems before carrying out the actual study

5. Recording Observational Data

   • Video Recording
     • Audio Recording
       • Hand-written notes
   • Sampling Systems
     • Time Sampling
       • where behaviour is recorded for a specific period of time e.g. 1 hour, or at specific time intervals e.g. every 30 seconds
         • lots of control
           • easier to replicate
         • might miss behaviours if the happen outside the time intervals
     • Even Sampling
       • where record are made of specific behaviours each time they occur
         • harder to control
           • harder to replicate
         • less likely to miss behaviours as you write it down when as and when it happens