observational design

• researchers should not conduct observational studies alone. single observers could miss important detail and only notice events that confirm their hypothesis. this introduces bias into the research process. to make recording data more objective and unbiased, observerations should be carried about by two researchers. the observers should be consistent in their judgments. therefore, observers must be trained to establish inter observer reliability
• inter observer reliability: 1. observers should familiarise themselves with the behvaioural categories used 2. they should observe behaviour at the same time which could be during a pilot study 3. observers should then compare the data they have recorded and discuss any differences in their interpretations 4. finally observers should analyse the data from the study. inter observer reliability is calculated by correlating each pair of observations made and an overall figure is produced.
• during the observation design, the researcher must choose how they will record the data. the researcher may write down everything they see known as unstructured observation which produces a rich detailed account of behaviours and involve few participants. however, there may be too many behaviours to record in a single observation. in the scenario, a structured observation is used which allows the researcher to quantify the target behaviours into a pre determined list of behaviours and sampling methods which will be the main focus
• structured vs unstructured: structured observations involve the use of behavioural categories which make the recording of data easier and more systematic. this because the data produced is ussually quantitative meaning that analysing and comparing the observed behaviour is easier. in contrast, unstructured observations produce qualitative data which is more difficult to analyse and record. unstructred observations benefit from more richness and detail in the data collected. however, they may be risk of observer bias as behavioural categories are not used.
• in order to produce a structured interview, the researcher must break the target behaviours into behavioural categories. target behaviours should be precisely defined and observable and measureable. for example the behaviour 'affection' could be broken down into hugging, smiling etc.
• the use of behavioural categories can make data collection more structured and objective but its important that the categories are very clear and unambigous. they should not reqiure further interpretation. also, researchers should ensure that all possible forms of the target behaviour are included in the checklist. categories should be exclusive and not overlap for example smiling and grinning would be too similar
• continuous recording of behaviour is a kay feature of unstructured observations in which all target behaviour is recorded. however, for complex behaviour this method is impractical. so instead event or time sampling is used
• event sampling involves counting the number of times a particular behaviour occurs in a target individual. whereas time sampling involves recording behaviour within a pre established time frame
• event sampling is useful when the target behaviour happens infrequently and could be missed if time sampling was used. however, if the specified evemt is too complex, the observer may overlook important details if using event sampling. time sampling is effective in reducing the number of observations that have to be made, however, those instances when behaviour is samples might be unrepresentative of the observation as a whole