observational design

• What are the factors that must be present for a credible observation?

  • Behavioural categories
  • Sampling procedures
  • Inter-observer reliability
• What is the definition of behavioural categories?

  dividing target behaviours into subsets through coding systems

  • must be clearly defined, observable, and measured

  e.g, laughing/on phone/talking - must be relevant to specific observation
• What are the types of sampling procedures?

  • Event sampling - counting the amount of time a behavior is presented by an individual/group
  • Time sampling - counting behaviour present in set time frame (e.g, every 30 secs)
• What is the definition of inter-observer reliability?

  the extent to which observers record similar behaviour when assessing/observing behaviour

  • should be 80% agreement
• How to achieve inter-observer reliability?

  • multiple observers
  • pre-decide behavioural categories
  • compare behaviour + calculate correlation (should be 0.7)
  • pre-decide sampling procedure: time sampling must be at the same time
  • observe the same individual/group
• Evaluate event sampling

  Strength - infrequent behaviours won't be missed = representative

  Weakness - complex/infrequent behaviours may be overlooked
• Evaluate time-sampling

  Strength - less observations = less time consuming

  Weakness - behaviour sampled may be unrepresentative