observation

Question 1

Sampling Behaviour

Answer 1

Sample behaviour at particular times, in specific settings, and under particular conditions to allow generalisation of findings to describe people, settings, and conditions beyond those used in the sample – i.e. to establish external validity
To identify representative samples of behaviour, use a combination of
Time sampling
Choose time intervals for making observations either systematically (e.g. first day of the week; morning from 8 to 11) or randomly, or both
Not effective when the event of interest occurs infrequently
Event sampling is used instead – useful for observing behaviour during events that occur unpredictably, such as natural or technical disasters
May not be representative of the same behaviour at other times
Situation sampling
Can significantly increase external validity
Studying behaviour in different locations and under different circumstances and conditions
Reduce the chance that the findings is unique to specific circumstances or conditions
Subject sampling is required if the population for the situation is too large
Could select either systematically (e.g., every 10th student at the cafeteria) or randomly

Question 2

Observation Methods

Answer 2

Direct observation method: decide on the extent to which researchers will intervene in the situation they observe – to change or create the context for observation
Observation without Intervention
Aka naturalistic observation
Passive recorder of events as they occur naturally
Within moral and ethical constraints

Question 3

with intervention

Answer 3

Participant observation: observe people’s behaviour and participate actively in the situation they are observing
Undisguised: the people who are being observed know that the observer is present for the purpose of collecting information about their behaviour
Reactivity – people react to the fact they are being observed by changing their normal behaviour (more about this in slide 11)
Disguised: the people who are being observed do not know they are being observed
Structured observation: observer intervenes in order to exert some control over the events they are observing; “set-up” a situation so that events can be more easily recorded
Using a confederate to behave in a certain way in order to create a situation to observe behaviour
Can be conducted in a natural setting or a laboratory
Important consideration: consistent procedures and structure observations as similarly as possible across observations

Question 4

Indirect (Unobtrusive) observation method

Answer 4

Complement hypothesis testing based on other methods, such as direct observation, experiments, and surveys
Important advantage: non-reactive
Can be obtained by examining physical traces and archival records
Physical Traces: remnants, fragments, and products of past behaviour
Use traces: physical evidence from the use or nonuse of an item (e.g., cigarettes in ashtrays; cans in recycling bins; fingerprints on a murder weapon)
Natural-use traces
Controlled-use traces
Products: creations, constructions, or other artifacts of behaviour (e.g., fashion, television shows, music, food

Question 5

Recording Behaviour

Answer 5

Narrative records – usually comprehensive record of behaviour, but not always (e.g. field notes are not comprehensive)
Written, audio or video recordings
Classify and organise the data to test hypotheses about behaviour
Should be made during or soon after behaviour is observed; observers must be carefully trained to record behaviour according to established criteria

Question 6

Interobserver Reliability

Answer 6

The degree to which two or more independent observers agree
Could be agreeing with the same degree of error but we are more inclined to believe that observations are accurate and valid when two or more independent observers agree than when data are based on the observations of a single observer
The more independent observers agree, the more confident we become

Question 7

Influence of the Observer

Answer 7

Reactivity: presence of an observer influences the behaviour being observed; hence individuals’ behaviour may not represent their typical behaviour
Demand characteristics - Individuals may react by trying to behave in ways they think the researcher wants them to behave
External validity is threatened as behaviour is not representative of their behaviour outside the research setting

Question 8

Ways to control reactivity

Answer 8

Limit participants’ knowledge of their role in the study or about the hypotheses of the study (ethical concerns have to be considered)
Observers conceal themselves (disguised instead of undisguised)
Use indirect or unobtrusive method
Adapt participants to the presence of the observer
Habituation: observers enter into the setting on many different occasions until the participants stop reacting to their presence
Desensitisation: exposed to observers in gradual increments

Question 9

Observer Bias

Answer 9

Systematic errors that results from an observer’s expectation; occurs when researchers’ biases determine which behaviours to observe and when observers’ expectations about behaviour lead to systematic errors in identifying and recording behaviour
Expectancy effects
Controlling observer bias
Aware that it might be present
Use “Blind” observers – do not know the reasons for the observations or the goals of the study