Observational techniques

• What are observations?:

  • The experimental method is not always the most suitable way to research certain types of behaviour.
  • Sometimes it is more appropriate to use non experimental methods such as observations.
  • Observational techniques allow us to see what people do without having to ask them. They involve the researcher watching and listening.
• Types of observation include:

  • Naturalistic and controlled observation
  • Overt and covert observation
  • Participant and non-participant
• Sometimes observations are used in experiments as a way of measuring the DV. They are therefore not deemed a ‘research method’ but instead a technique used alongside other research methods.
• Naturalistic:

  • Behaviour studied in natural situation – researcher does not interfere in any way.
  • e.g. observing children playing in the school playground

  Controlled:

  • Occurs when the researchers control some variables
  • Can occur in laboratory or natural situations
  • e.g. watching parent/child interaction through a one way mirror
• Overt:

  • Participants are aware they are being observed
  • Overt observation can take place in naturalistic and controlled observation

  Covert:

  • Participants are unaware that they are being observed.
  • They may be informed afterwards but have no knowledge until the observation is over.
• Participant:

  • In a participant observation the observer is also a participant in the activity being studied either with or without the group’s knowledge.
  • e.g. joining a cult to observe the behaviour from the inside

  Non-participant:

  • The observer observes from a distance and does not interact with the p’s
• Observation strengths:

  • Capture what people actually do – with a low risk of demand characteristics
  • Can therefore give special insight into behaviour
• Observation limitations:

  • Observer bias – interpreters observation of a situation may be influenced by their expectations
  • Cannot demonstrate causal relationships – although observational techniques can be used in experiments and aid in detecting cause and effect.
  • Provides information about what people do but not about what they think or feel
• Reliability of observations:

  • Observations should be consistent
  • If two people observed the same activity we need to see if they agree on the behaviours seen – we have to assess inter-observer reliability – above 0.80 is considered good.
  • To calculate the inter-observer reliability: Total agreements ÷ total observations
• In order to produce high inter-observer reliability, we can do the following things:

  • Clearly operationalise behavioural categories
  • Training observers in observational techniques
  • Pilot the study to review the behaviour checklist and behavioural categories
  • This helps to reduce observer bias (interpreting behaviours from one's own perspective) and in turn, will increase the validity of the results.
• Observations - External Validity:

  • Strength - High ecological validity- involve more natural behaviours.
  • Limit - Population validity may be a problem, if for example, children are only observed in middle class homes, because we cannot generalise such findings to children from all classes.
• Observations Internal validity

  Limits:

  • Observations will not be valid if the behaviour checklist is flawed. For example, some observations may belong in more than one category, or some behaviour may not be codeable.
  • The validity of observations is also affected by observer bias – what someone observes is influenced by their expectations. This reduces the objectivity of observations. (This can be solved by using a double blind procedure where the person completing the research is unaware of the hypothesis).
• Observations reliability

  Limits:

  • Observations can be inconsistent and unreliable if steps aren’t taken to reduce observer bias
• Inter-Observer Reliability:

  • Observations should be consistent, which means ideally two observers should produce the same record. (This is also known as inter-observer reliability).
  • A general rule is that if there is more than 80% agreement on the observations, the data have inter-observer reliability. (Total Agreements ÷ Total Observations).

  Improving Reliability:

  • Observers should be trained in the use of a coding system/behaviour checklist.
  • They should practice using it and discuss their observations. The investigator can then check the reliability of their observations.
• Ethical Issues: issues relating to informed consent. Invasion of privacy, in which case participant confidentiality should be respected. The use of one-way mirrors often involves deception.
• Dealing with Ethical Issues:

  • Ethics committees can be used to approve observational designs, and researchers should consult ethical guidelines.
  • One particular BPS guideline concerns observational research – this type of research is acceptable where those observed would be expected to be observed by strangers.
  • In addition researchers should be aware that it is not acceptable to intrude upon the privacy of individuals who, even in a normally public space, may believe they are not observed.