Observational Techniques and Designs

• An observation is a non- experimental method which involves observing and recording behaviour in either naturalistic or controlled settings: only able to investigate what they can see as observers cannot infer motive, intention, feeling or thoughts: no assumption of cause and effect
• A naturalistic observation is one which the researcher observes and records behvaviour in a natural setting- away from the lab- no manipulation/ complete absence of an independent variable.
• Naturalistic observations used when it would be inappropriate/ unfeasible to run an experiment to investigate the topic (eg: attempting to control for an experiment about how children interact with a playground naturally would invalidate the experiment)
• In a naturalistic experiment participants are/ may be unaware of the fact they are being watched, as may be going about their regular activities
• Strengths of a naturalistic observation:

  • Behaviour is natural and unforced- meaning it has high ecological validity
  • unlikely to succumb to demand characteristics
• Limitations of a naturalistic observation:

  • raises ethical concerns as an not give informed consent or the right to withdraw is not made abundantly clear
  • also may not be possible to give them a debrief
  • not replicable due to nature of method- method may be overly subjective
• A controlled observation is one in which the researcher implements a level of control, implementing replicable procedures (sometimes an IV)- procedure and phases must be carefully designed by the researcher along with predetermined behavioural categories to be measured
• In a controlled observation participants are aware that they are being observed: must be recruited for the study
• Strengths of a controlled observations:

  • replicable procedures, therefore more scientifically rigorous
  • tests are standardised- method has good reliability
  • researcher can be more confident of the cause- effect relationship
• Limitations of a controlled observation:

  • low ecological validity
  • demand characteristics may impair observation- lowering the validity of the findings
• In a covert observation participants are not aware that they are being observed, will not have been informed of this in advance, and most likely won’t be able to see the researcher observing them
• The only ethical way to conduct a covert observation is in a public context, eg: shopping mall
• Strengths of a covert observation:

  • high in ecological validity
  • low chance of demand characteristics
• Limitations of a covert observation:

  • ethical issues- deceiving particpants, cannot give informed consent, do not know about the right to withdraw
  • cannot really replicate the study- problematic in most cases
• In an overt observation particpants are aware that they are being observed and may have been informed of this in advance, may be able to see the researcher observing them
• Overt observations occur in controlled, lab conditions as the researcher is keen to test the effect of the IV on the DV
• Strengths of an overt observation:

  • ethics can be preserved as participants are aware of the fact they are being observed
  • had right to withdraw and gave informed consent
• Limitations of an overt observation:

  • Could increase demand characterises as participants are aware they are being observed
  • Damaging validity of the findings
  • May suffer from researcher bias- researcher may set up tasks that align to closely with their hypothesis- confirmation bias as striving to look for behaviours that support their hypothesise
• Participant observations is when the researcher (and possibly confederates) join the group they observing, particpants may not be aware that the researcher is an outsider
• Strengths of a participant observation:

  • able to get full involved with the group- increasing validity of their findings as now has access to real thoughts, feelings and conversations
  • able to become aware of different theories/ topics to research in the future
• Limitations of a participant observation:

  • could have a restricted view of what they wish to observe and thus missing some important behaviours- limiting the usefulness of the participant observation
  • may loose objectivity as a researcher due to their immersion in the situation- may begin to identify with those they observing (especially in long term studies)
• Non- participant observations the researcher stays separate and apart from the group they are observing, may or may not be aware they are being observed (either covert or overt)
• Strengths of a non- participant observation:

  • researcher can keep distinct objectivity, meaning less likely to become biased or subjective in their recording of behviour- validity stays in tact
  • good vantage point for the researcher as view is not restricted by being in the experiment- increasing the scope of data being able to be collected
• Limitations of a non- participant observation:

  • Researcher may lack key detail/ insight as they are removed from the ‘action’- observation may lack explanatory power
  • Could misinterpret some of the behaviours as they are apart from what they are observing- not able to ask the particpants for clarification- meaning the observations data lacks in validity
• Behavioural categories are used to record specific behaviours during one observation, categories must be designed to only record observable behaviours- no ambiguity about what is being observed
• If using behavioural categories the researchers will have previously agreed on which specific behaviours should be recorded so that all observers agree before the observation begins
• Behaviour categories have to be operationalised to make sure they are specific and cannot be confused, then further arranged int whether it is boys or girls/ general categories- able to gain further insight into the behaviour
• Despite categories of behaviour are firmly established, researcher still may be prone to researcher bias- test their reliability of their observations against other researchers recording the behaviour- consistency between results compared (inter- observer reliability)- observations conducted sepertaly to avoid conformity
• Strengths of behavioural categories:

  • Use of clearly defined, unambiguous categories enabling an objective view of what’s observed- less subjectivity
  • Higher scientific rigour
  • Use of more than one observer ensures inter- observer reliability
• Limitations of behavioural categories:

  • predetermining the categories may be limiting in terms of the types of behaviour enacted during observational session- behaviour not holy represented meaning lowers validity of findings
  • inter- observer reliability may overestimate the true agreement between observers- may not have held what behaviour was recorded to same standards as each other
• Inter- observer reliability is a way to assess if the way you have chosen to measure the DV/ co- variables is consistent across different people
• Inter- observer reliability can be assessed by comparing the results of the observers by putting both sets of results through and inferential statistical test measuring correlation to see the degree of similarity between the results
• Inter- observer reliability can be improved by: ensuring the behavioural categories are clear and objective, give training to ensure peoples expectations are similar for each category- use previously outlined assessment procedure to check the reliability of the categories
• Event sampling is when the researcher records/ tallies every time a behaviour from a specific behavioural category occurs, eg: every time litter is dropped
• Time sampling is when the researcher records all behaviours during a set time period, at a set point, eg: record observations for 15 min intervals over 2- hour period
• Strengths of Event Sampling:

  • ensures behaviours will not be overlooked or missed
• Strengths of Time Sampling:

  • allows flexibility to record any behaviour which may be relevant to research
  • offers researcher opportunity to record unexpected behaviours which may trigger new research projects
• Limitations of Event Sampling:

  • if too many of the specific behaviour occur once and are overly complex, difficult for event sampling to capture all
  • Limiting validity of method as not providing a true reflection of what occurred during observation
• Limitations of Time Sampling:

  • missing behaviours that occur outside of the allotted time period
  • limiting validity of the method as some behaviours are being over/ mis- represented in the findings