Observational design

• Unstructured observations: where a researcher records all relevant behaviour with no specified system for data collection. This produces qualitative behaviour.
• Structured observation: A researcher quantifies an observation using behavioural categories and sampling methods. This produces quantitative data.
• Behavioural categories: For structured observations, a researcher breaks up behaviour into a set of components (it is operationalised). For example a behavioural category in aggressive behaviour would be punching, hitting, verbal abuse, biting etc.
• Behavioural categories need to be objective so that the observer doesn't have to make inferences about behaviour, and must be mutually exclusive so that components don't overlap.
• Sampling procedures: Used in an unstructured observation. This is a systematic (controlled) method used to record every instance of data in as much detail as possible.
• What methods does Sampling procedures include?

  Event and Time sampling.
• Event sampling: recording the number of times a behaviour/event occurs in the target group or individual in a specific time period. This could include tallying every time a new behaviour is shown or repeated.
• Time sampling: A method of data collection in which the researcher records the behaviour of the target at set time intervals. Rather than observing the target at one time, the researcher records the target's behaviour at regular intervals.
• Types of observations are: c ontrolled, n aturalistic, c overt, o vert, p articipant and non-p articipant observation.
• Controlled observation: observing the behaviour of a target in a structured environment, wherein some variables are controlled.
• Controlled observation strength: High level of control allows observers to specifically focus on behaviours of interest.
• Controlled observation limitation: high level of control means the environment is less realistic, makes the research vulnerable to social desirability bias and demand characteristics.
• Naturalistic observation: observing behaviour of a target in an environment where it would normally occur.
• Naturalistic observation strength: Participants act spontaneously, so provides the study with realistic everyday behaviour, giving it ecological validity.
• Naturalistic observation limitation: observer may be unable to control for extraneous and confounding variables, so an unknown factor may influence behaviour resulting in low internal validity.
• Covert observation: A participant's behaviour is observed and recorded without their knowledge.
• Covert observation strength: Participants are less likely to show demand characteristics or social desirability bias, making the behaviour they show more genuine. Good internal validity.
• Covert observation limitation: This breaches many ethical guidelines as there is deception, no right of withdrawal, and no informed consent from the participant. However this can be overcome using debriefing or retrospective consent.
• Overt observation: The participant is aware that their behaviour is being observed and recorded by a researcher.
• Overt observation strengths: participants have given consent to take part in research, so it avoids ethical issues.
• Overt observation limitations: As participants are aware they are being studied, they may be less likely to behave naturally, showing demand characteristics lowering the internal validity.
• Participant observation: A method of research in which the researcher takes part in the activities of the people being studied, and becomes a part of the group they intend to study.
• Participant observation strength: allows researcher to gain a deeper understanding of the behaviour studied as they gain an 'inside' view. Also as it is covert behaviour is more natural.
• Participant behaviour limitation: Breaches ethical guidelines. Issues with gaining informed consent, debriefing ppts. Also observer bias is more likely,
• Non-participant observation: Researcher observes without taking part in the activity or event being observed.
• Non-participant observation strength: observers are likely to be more objective and less influenced by their own views, reducing observer bias.
• Non-participant observation limitation: more likely to be overt, demand characteristics shown by ppts more likely, low internal validity.