Observations and techniques

• What is an observation?


  A technique where researchers will watch and record behaviour
• Why are they used?

  • to study behaviour in more natural settings so the results are more realistic
  • to study behaviours which would be unethical to manipulate or cannot be directly manipulated by researcher e.g. observing aggression in sports or at a bar after drinking
• Types of observations

  • participant or non-participant : observer becoming actively involved in situation or not
  • overt or covert : aware they're being observed or not
  • controlled or naturalistic : watching behaviour in a controlled or natural setting
• Participant vs non-participant

  Strength of participant/weakness of non-participant:

  • the researcher can experience the situation so can give insight - researcher can lose valuable insight if not involved

  Weakness of participant/strength of non-participant:

  • researcher is involved so results are likely to be subjective - if researcher isn't involved they can remain more objective
• Overt vs covert

  Strength of overt/weakness of covert:

  • PPs know they're being observed so observation is more ethical - if PPs are unaware there are ethical issues like lack of consent

  Weakness of overt/strength of covert:

  • a PP knowing they're observed may lead them to act a certain way (Hawthorne effect, 'Screw you' effect, social desirability) - if PPs are unaware of observation there is less likely to be demand characteristics
• Controlled vs naturalistic

  Strength of controlled/weakness of naturalistic:

  • setting is controlled so extraneous variables are less likely - if environment is natural it is difficult to replicate because of extraneous varibales

  Weakness of controlled/strength of naturalistic:

  • as setting is controlled behaviours are less likely to be realistic - if observation is in a natural settings the behaviours are more likely to be natural too
• Inter-observer reliability

  The extent to which two observers agree when assessing or observing behaviours - there should be at least 80% agreement
• How to achieve inter-observer reliability

  • have multiple observers
  • create behavioural categories beforehand (should be defined, observable and measurable)
  • observe same behaviour at same time
  • compare observations and calculate correlation between them
  • could also record scene using a camera
• Sampling methods

  Event sampling:

  • counting number of times a particular behaviour (event) occurs in a target group

  Time sampling:

  • recording behaviour within a pre-established time frame e.g. every 30 seconds
• Event vs time sampling

  Strength of event/weakness of time:

  • behaviours may occur frequently which won't be missed - instances when a behaviour is sampled may be unrepresentative of observation as a whole

  Weakness of event/strength of time:

  • if event is quite complex some behaviours may be overlooked - less observations have to be made therefore less time consuming