Observations

• Observation studies

  • where a researcher watches or listens to participants, and then records data.
  • several different types of observations that can be carried out:
  • Naturalistic Vs Controlled
  • Covert Vs Overt
  • Participant Vs Non-Participant
• Naturalistic Vs Controlled

  • Naturalistic Observation: Conducted in a natural setting where the target behavior typically occurs (e.g., observing behavior in a classroom). No manipulation done by researcher
  • Controlled Observation: Conducted in an artificial setting to control specific behaviors (e.g., Stanford Prison Experiment).
• Covert Vs Overt

  • Covert Observation: Participants are unaware they are being observed; only ethical if behaviors are public e.g police stake-outs
  • Overt Observation: Participants are aware of the observation and have given informed consent.
• Participant Vs Non-Participant

  • Participant Observation: Researcher joins the target group to gain an accurate understanding of behaviors (e.g., joining a cult).
  • Non-participant Observation: Researcher stays separate from the group, recording information in an objective manner
• Behavioural Categories

  • Structures- target behaviour is broken down into different types of behaviour. Each category should be observable and obvious so that it can be tallied/counted each time
  • E.g flirting behaviours could be broken down into eye contact, smiling, laughing, touching
  • Unstructured- write down everything they see
• Interobserver reliability

  • Observation studies should involve multiple researchers to avoid observer bias.
  • A single researcher may miss important details or record data that aligns with their expectations.
  • To ensure observations are reliable, it’s essential to check inter-observer reliability.
• Inter-observer reliability

  • two observers should produce the same observations
  • It is the extent to which the record sheets match one another
  • To do this:
  • the researchers create and agree on categories of behaviour
  • observers all watch the behaviours independently and then compare their data
  • Correlations can be used to do this
  • If the observations are not similar then the categories of behaviour need to be altered and rewritten
• Strengths of observations

  • It may provide a better indication of behaviour than questionnaires/interviews
  • no social desirability as it focuses on what people do rather than what they say they do, INCREASES VALIDITY
  • High in ecological validity because the researcher is observing real-life behaviour.
  • This is especially true in covert and natural observations studies, because demand characteristics are unlikely.
  • Data= higher validity
• Weakness of Observations

  • There are ethical issues involved in observation studies.
  • Often no consent is given when observing in public places and there is a lack of privacy This means some observations should not be conducted
  • Observers have expectations about what they will discover, and this can influence what they see or hear. This is called observer bias.
  • Therefore observations may lack validity