Observations are the procedure of watching and then consequently recording and documenting the behaviour of the human or animal participants. This usually has 2 standard ways; A naturalistic observation is conducted in the participants’ normal atmosphere without any intereference from the researcher (who are observing) them in their usual physical and natural enviornment. A controlled observation is done mostly in an artificial situation such as a laboratory but it is also possible in a participants; natural envionment but with strict controls.
If one consideres the whole spectrum of possible behaviours it is a possibility that observations may be non-focused – if this lack of strict controls continues then it is deemed an unstructured observation.
A structured observation however is designed to concentrate on a specific set and range of behaviours, record them and then proceed to categorize them. This also helps for the testing and verification of the study’s reliability via a technique called inter-rated reliability (the consensus of 2 or more experimenters to verify the validity of the study by judging on the degree of agreeement in their respective research results via the same, common methods.)
Observations are often also conducted in social settings, either participant or non participant. A participant observer is a part of the observed situation/setting. Whereas, a non-participant observer does not become a part of the observed situation that is being studied which is done by observing through a one-way-glass or by staying reserved from the group of participants.
Bandura et al. tested the extent a ‘role model’ has on a child’s spectrum of agression and agression being socially learned. He observed the children through a two-way mirror along with instructions of participant observer which included their sex, step-by-step model behaviour and verbal prompts of agression and a 5-point scale of agfression on which the observers recorded children making it a structured observation, one with strict controls.
- Event sampling: When the experimenter is looking for as particular behavior and a tally chart or documenting record is kept of every instance of that behavior.
- Time Sampling: When the experimenter/observes notes down the behavior at designated, certain times. (eg, a 10-second interval)
Observers often variate their stance as:
• Overt observers: The observers are openly watching and documenting the participant behaviour, with the participant knowing that they are in a study being observed on different scales. Their behaviour, attire is indicative of them being observers.
• Covert observers: The observers are NOT openly watching and documentating the participant behaviour, without the participant’s consent as they cannot be aware they are being observed (if vice versa that would make the particpants exhibit demand characteristics).
- Does not raise ethical concerns over the issues of deception, informed consent or even privacy.
- It is practical and thus can be conducted over an extended period of time.
- The researcher can make notes and record details without having to rely on memory as they don’t have to worry about blowing their cover.
- It is natural and researchers can ask a number of questions (structured observation)
- A very high risk of demand characteristics which lowers validity.
- High risk of incurring socially desirable responses from the participants.
- The results may not always be representative which questions the credibility of the research.
- Allows for techniques like ‘time sampling’ and ‘event sampling’ which ensure a structure to the observation
- Increases validity – less or no exposure to aim and thus demand characteristics
- Reduced effects of social desirability.
- They can be better controlled and moniotored as the researcher can dig deeper and assess more.
- There is ahigh rate of inter-rate reliability as two researchers may even be simultaneously observing
- Raises ethical issues such as informed consent, deception and privacy of the particpant.
- Patients may feel distress at the violation of their privacy which again raises ethical concerns.
- The legality of this is often questioned.
- There is laso the risk of identity being unveiled which would discredit the whole operation.
- Hard to sustain or conduct over a long period of time.
Advantages of Observations:
- The observed behavior is natural, authentic as they are unaware – this increases ecological validity.
- The data collected is often quantitative though structured controls which has clearly defined categories is on terms with being objective and statistically comprehendible via analysis.
- Chances gathering extremely rich data is very high if the observation is unstructured.
- If participants are unaware, risk of inducing demand characteristics is improbable which increases validity.
Disadvantages of Observations:
- The participant cannot explain or elaborate for the cause of them behaving a certain way as it is a subjective approach (which when asked might expose the aim of the research).
- Observations may not be reliable due to natural and logistical issues such as view obstruction, missing out on details, relying on memory and etc.
- Naturalistic observations make it hard for controls to be established and this in turn, makes it harder to control confounding variables – making it difficult to formulate a cause-and-effect relationship.
- Difficulty in replication.
Various ethical issues arise – deception, lack of informed consent as people are being observed without their permission.