observation презентация

observation

Problems with survey research Respondents interpret different meanings from questions Omission of key terms when reading questions Reliance on people’s memories of their behaviour Social desirability effect Threatening questions - invalid answers Interviewer bias Gap between stated and actual behaviour Capacity to answer questions

What is structured observation? A method of systematically observing people’s behaviour By grouping behaviour into categories Direct observation, not reports or inference, make it an alternative to survey methods Aggregates and compares behaviour of everyone in the sample Uses a cross-sectional research design

Types of observation research Structured/systematic observation Participant observation (qualitative) Non-participant observation Unstructured observation (inductive) Simple/contrived observation - a form of participant observation where the situation is left as-is or is changed (Webb et al.1966)

Observation without Intervention Naturalistic Observation: Observation in natural (“real-world”) settings without an attempt to intervene or change the situation. Goals: Describe behavior as it normally occurs, examine relationships among naturally occurring variables. Naturalistic observation helps to establish the external validity of laboratory findings. Naturalistic observation is used when ethical and moral considerations prevent experimental manipulation. Role of emotions (emotional ties with observed peope|events)

Observation with Intervention Most psychological and marketing research involves observation with intervention. Researchers typically choose from three methods of observation with intervention in natural settings: Participant observation (‘observing participation’} Structured observation Field experiment

Participant Observation Observer is an active participant in the natural setting he or she observes. undisguised: people in the setting know they are being observed disguised: people don’t know they are being observed Disguised participant observation helps control for reactivity, one of the main problems associated with observation.

Participant Observation Reactivity occurs when people change their usual behavior because they know they are being observed. Participant observers may sometimes lose their objectivity by becoming too involved in the situation. Participant observers may influence the behavior of people they are observing.

Structured Observation Researcher sets up (structures) a specific situation in order to observe people’s behavior. Examples: clinical and developmental observations of behavior, psychological testing Structured observation is useful when behavior is difficult to observe as it naturally occurs. Researchers often use confederates to create the structured situation. Problems occur when observers don’t follow the same procedures across observations or observers, and when important variables are not controlled.

Field Experiment Researcher manipulates an independent variable in a natural setting and observes behavior (dependent variable). There must be two or more conditions to compare (independent variable). Researchers often use confederates to create these different conditions. Researchers attempt to have the most control in field experiments.

Lab Observation

Recording Behavior Researchers can obtain a comprehensive record of people’s behavior (e.g., video tapes) or select specific behaviors to record. The method for recording behavior determines how the results of the study are eventually measured, summarized, analyzed, and reported. Qualitative Quantitative

Qualitative Records Researchers use narrative records when they want a complete (comprehensive) reproduction of people’s behavior. Examples: videotapes, audiotapes, field notes Narrative records should be made during or soon after behavior is observed. Observers must be carefully trained. Advantage: Can review the record often to observe behavior. Disadvantage: Costly, time-consuming.

Quantitative Records Selected Behaviors: Researchers choose the behavior they want to observe and ignore other behaviors. Researchers need to decide how they will measure the behavior. Examples: frequency of behavior, duration Recording techniques include checklists and electronic recording and tracking.

The observation schedule Must have a clear focus and be easy to use Categories must be inclusive (cover all options) and mutually exclusive It must be easy to operate Specifies categories of behaviour to be observed and how to allocate behaviour to a category Clear guidelines needed so that observers can distinguish between behaviour categories A pilot study is useful to iron out any problems

Flanders (1970) Interaction Analysis Categories (FIAC)

Strategies for observing behaviour Record incidents, events and interventions Observe for short periods of time, repeatedly Observe for long periods of time, continuously e.g. Jenkins et al (1975): ‘the observation hour’ Time sampling Record whatever is happening every x minutes e.g. ORACLE study (Galton et al, 1980)

Sampling in structured observation Sampling people - Random sample of individuals to observe Sampling time periods - Observe same individual(s) at different, randomly selected times Non-probability samples - Cannot use a probability sample if there is no sampling frame, thus reduced external validity Limited external validity, in any case

Forms of sampling Ad libitum* sampling Record whatever is happening at the time Focal sampling (most common form) Observe a specific individual for a set period of time Scan sampling Record behaviour of everyone in the group at regular time intervals Behaviour sampling Observe whole group to see who was involved in a particular behaviour * Latin for "according to a performer's wishes"

Issues of reliability and validity Reliability Inter-observer consistency (measured by Cohen’s kappa) Intra-observer consistency (lapses of attention) Measurement validity Does schedule measure underlying concept? Implementation of schedule Presence of observer - reactive effect

What is the reactive effect? Webb et al. (1966) wrote about the ‘reactive measurement effect’, in which the research subject knows s/he is participating in research, thus invalidating the data: • The guinea pig effect (awareness of being tested) • Role selection (participants adopt a particular kind of role) • Measurement as a change agent (the very fact of a researcher being present) • Response sets (mostly relates to survey research when the respondent replies to a set of questions in a consistent but clearly inappropriate manner). Reactive effects are likely to occur in any research in which participants know they are the focus of investigation.

Field stimulations Salancik (1979) used ‘field stimulation’ to describe a form of observation in which the researcher directly intervenes in a setting and observes the results (a form of contrived, unobtrusive observation) Quantification of the outcomes is the desired goal

Criticisms of structured observation Imposes an irrelevant framework on the social setting Neglects the meanings and intentions behind behaviour (motives are inferred but not investigated) Neglects the social context of behaviour Generates fragmented data - difficult to see the wider picture

On the other hand... It is more accurate than interviews and questionnaires see what people really do, not what they say they do It is a useful accompaniment to other methods study behaviour, attitudes and social context

Home reading A. Bryman Social Research Methods 4th edition. Chapter 12.