Observation Flashcards
Participant
- The observer acts as part of the group being watched
Participant Evaluation
+ Good vantage point of the behaviours
+ Detailed/qualitative data collected
- Presence of observer may affect behaviour e.g. cause demand characteristics
- Observer bias due to the subjective/biased viewpoint
Non-Participant
- The observer doesn’t become part of the group being observed
Non-Participant Evaluation
+ The observer viewpoint remains objective
- Lack of detailed/qualitative detail due to not much insight
Naturalistic
- Observations occur in a natural setting
Naturalistic Evaluation
+ Ecologically valid as natural behaviour is observed
- Extraneous variables may affect the DV due to lack of control
Controlled
- Variables are controlled and manipulated by an experimenter
Controlled Evaluation
+ Only the IV affects the DV due to the good controls
- Lacks ecological validity due to the artificial setting
Structured
- The observer determines the behaviours being observed
Structured Evaluation
+ This is objective as a good coding frame is used
+ Quantitative data is easy to gather and present
- Misses out on other key observations
Unstructured
- The observer records everything that happens
Unstructured Evaluation
+ Qualitative/detailed data is collected
- Difficult to summarise/present data
- Too much to be observed
Covert
- Participants are unaware of being observed
Covert Evaluation
+ Natural behaviour is observed
- Unethical as consent has not been gained from the participant
Overt
- Participants are aware of being observed
Overt Evaluation
+ Ethical as consent has been gained from participants
- Unnatural behaviour e.g. social desirability bias
Event Sampling
- Behaviour is recorded each time it occurs
- Observation categories/coding frames are used which produce quantitative data that’s easy to compare, summarise, analyse and present
- However it gives a restricted view of what’s happening
- E.g. if too many things happen at once it’ll be hard to record
Time Point Sampling
- The observer records what the participant is doing at fixed intervals
- E.g. every five seconds over twenty minutes
- The issue with this is that behaviours may be missed during the time of non-observation
Time Event Sampling
- When behaviour is recorded when it occurs during a fixed time set for observation
- E.g. the first 10 minutes of every hour
- Some behaviours may be missed during the time of non-observation
Reliability In Observations
- Difficult to replicate observations due to confounding variables
- Consistency can be checked in observations using inter-rater-reliability
- Reliability is improved by using a good coding frame
Validity In Observations
- Overt observations and observer bias result in a low validity
- Validity is checked by asking participants through self report
- Validity is improved by using wider categories or single blind techniques
Inter-rater-reliability In Observations
- Observers agree beforehand on what will be observed
- A pilot study is carried out to ensure that it works
- Each observer then observes the same thing independently
- At the end of the observation the results are compared
- The greater the level of similarity, the greater the inter-rater-reliability
- This means that the coding frame/scheme is consistent and works the same way as it produces the same results when used by different people
