Observational Design

• What different observational designs can be put into place in an observation?

  • Structured
  • Unstructured
  • Behavioural categories
  • Sampling methods
  • Inter-observer reliability
• In structured observational designs, target behaviours are simplified and operationalised. This allows the researcher to quantify their observations using a pre-determined list of behaviours and sampling methods.

  strengths:

  • easier to record data
  • data is more likely to be quantitative (numerical) —> easier to analyse
• In unstructured observational designs, the researcher writes down everything they see. This kind of observational design is appropriate for small-scale observations e.g. observing couple interactions.

  strengths:

  • data is rich in detail

  limitations:

  • produces qualitative data —> harder to record and analyse
  • observer bias
• In behavioural categories (a.k.a behaviour checklists), target behaviours are precisely defined (operationalised) e.g. affection can be broken down into hugging, smiling, holding hands etc. Each behaviour must be observable, so a vague category wouldn’t be reliable as it’s open to interpretation. There should be no ’dustbin category’, all possible behaviours must have their own category.

  strengths:

  • objective
• Sampling methods

  In unstructured observations, there is a continuous recording of data. In structured observations, researchers must record data systematically using either event sampling or time sampling.

  • Event sampling = counting the number of times a particular behaviour (event) occurs in a target group/individual.
  • Time sampling = recording behaviour within a pre-determined time frame using a behavioural checklist e.g. recording what the individual is doing every 30 seconds.
• Sampling Methods

  Event sampling:

  strength - useful when behaviour is infrequent and could be missed in time sampling

  limitation - observer may miss important details

  Time sampling:

  strength - reduces the number of observations that must be made

  limitation - data may not be representative
• Inter-observer reliability

  To make data recording more objective and unbiased, observations should be carried out by at least two researchers. Observations should be consistent.

  How it works:

  • observers should familiarise themselves with the behavioural categories
  • they then observe the same behaviour at the same time as part of a small-scale pilot study
  • observers compare the data they recorded and discuss any differences
  • observers analyse the data form the study
  • IOR is calculated by correlating each pair of observations made and an overall figure is produced