(9) Observational Design

• if a researcher may simply may write down everything they see which is referred to as an unstructured observation
• Unstructured Observations :

  • researher writes down everything they see
  • produces accounts of behaviour that are rich in detail
  • appropriate when observations are small scale - involves a few participants
• Structured Observations :

  • used when theres too much going on in a single observation for the researcher to record at all
  • Simplifies the target behaviour that will become the main focus of the investigation
  • specific acts (verbal or physical) that make up the target behaviour would need to be clearly defined
  • allows researcher to quantify their observations (qualitative) using pre determined list of behaviours and sampling methods
• EVALUATION of Structured Observations

  Positives :

  • involves using behavioural categories - makes the recording of data easy and more systematic
  • data produced is likely to numerical - means analysing and comparing the behaviour observed between participants is more straightforward
• EVALUATION of Unstructured Data

  Negatives :

  • tends to produce qualitative data - more difficult to record and analyse as theres extensive notes
  • great risk of observer bias - as the objective behavioural categories that are a feature of structured observations are not present here
  • researcher may only record those behaviours that 'catch their eye' - may not be most important or useful
• EVALUATION of Unstructured Observations

  Positives :

  • data produce is rich in detail + depth
  • allows to make links - cause and effect relationships
• Structured vs unstructured observations
• the four different observations methods are :

  • structured and unstructured observations
  • behavioural categories
  • event sampling
  • time sampling
• Behavioural Categories - what it is :

  • to produce structured record of what a researcher sees (or hears), they first need to break the target behaviour up into a set of behavioural categories or behavioural checklist
  • similar idea to operationalisation
  • target behaviours to be studied should be precisely defined + made observable + measurable
  • there should be no need for inferences - two observers might interpret this differently so would not be a reliable category
  • must ensure to include all the ways in which target behaviour may occur within the behavioural checklist
• Behavioural Categories - what it must be :

  • categories can make data collection more structured and objective
  • categories should be clear and ambiguous as possible
  • must be observable, measurable and self evident
  • all forms of target behaviour should be included - there should not be a 'dustbin category' (others) where different behaviours are deposited
  • categories should be exclusive and not overlap -should not be difficult to distinguish (eg smiling and grinning)
• behavioural categories can also be known as behavioural checklist
• behavioural categories involve tally charts
• there are sampling methods for observations (when you record data)
• the different sampling methods for observations are :

  • continuous recording
  • event sampling
  • time sampling
• Continuous Recording :

  • key feature of unstructred observations
  • all instances of a target behaviour are recorded
  • complex behaviour : not practical or feasible
  • not used in structured observations - researcher must use a systematic way of sampling their observations (know how and when to collect data)
• continuous recording is a sampling method used for unstructured observations of behaviour
• event and time sampling are sampling methods used for structured observations of behaviour
• Event Sampling :

  • involves counting the number of time a particular behaviour occurs in a target individual or group

  used in structured observations
• Time Sampling :

  • involves recording behaviour within a pre established time frame (eg noting target behaviour after every 30 seconds)

  used in structured observations
• EVALUATION of Event Sampling

  Strength :

  • useful when the behaviour or event happens quite frequently
  • Behaviour could be missed if time sampling used
• EVALUATION of Event Sampling

  Negatives :

  • if specified event is too complex observer may overlook important details
  • too specific so not easily observable
• EVALUATION of Time Sampling

  Strength :

  • effective in reducing the number of observations that have to be made
  • Cost effective
• EVALUATION of Time Sampling

  Negatives :

  • instances where behaviour is sampled might be unrepresentative of the observation as a whole
  • Unnecessary behaviour sampled due to specific timeframes used
  • Don’t look at behaviour for the whole time
• it is recommended that researchers do not conduct observational studies alone
• Inter Observer Reliability :

  • researcher recommended not to conduct observational studies alone
  • single observers might miss important details or may only notice events that conform their opinion or hypothesis - introduces bias
  • observations carried with at least two researchers to be more objective and reduce bias
  • observers need to be consistent in their judgements / data recorded is same or very similar
  • observers need to be trained to establish inter observer reliability
• Observers establish Inter Observer Reliability by :

  • familiarising with the behavioural categories to be used
  • both observe the same behaviour at the same time - eg through small scale pilot study
  • comparing the data they have recorded and discuss any differences in interpretation
  • both observers should analyse the data from the study - calculating by correlating each pair of observations made and an overall figure should be produced