Behavioural Observation Flashcards
Behavioral Observation def
The recording of the observable responses of individuals
-> Draws heavily on theory & research tradition of behaviourism
What are the types of responses we examine? (3)
Responses that are observable with the NAKED EYE during a relatively BRIEF period of time in a PARTICULAR environment
Behavior is different from biochemical & neurological events because:
Biochemical & neurological events are NOT observable with the naked eye
Behavior is different from personality traits because:
Personality traits are “observed” only if behavior is aggregated over relatively long periods of time and in a number of environmental contexts.
Emphasis is on ___ behavior, quantification, ___ levels of inference, and assumptions of _____________
OVERT, low, environmental causality
Condition to Perform Behavioral Observation
Develop a set of rules – CODING systems – for extracting information from the stream of behavior
The rules/coding systems for extracting information from the stream of behavior specify: __ (7)
- target behaviors
- sampling of behaviors
- dimensions of behaviors to assess
- observation settings
- observers
- data recording
- data scoring
2 Behavioral Categories
Molar and Molecular
(Behavioral category): Molar
- BROAD units of behavior; e.g., “aggresses”, “plays”
- Defined in terms of LARGER response units that have functional significance over extended periods
- Capture psychologically meaningful behaviors but require observers to make INFERENCES
(Behavioral category): Molecular
- More specific behaviors; e.g., “bites”, “smiles”
- Defined in terms of smaller, immediate response units (i.e., specific sequences of motor movements)
- Harder to interpret post-data collection, but easier to use and can be collapsed into molar categories for analysis
Operational Definition def
Clear, specific, and measurable description of each behavior to be observed that ensures consistency in observation and assessment
To guarantee adequacy, operational definitions should: (8)
(1) Be Objective and focus on directly observable components of the target behavior
(2) Be Clear, unambiguous, and easily understood
(3) Require little or no inference -> see examples in the slides
(4) Include Descriptive names,
(5) Include General definitions,
(6) Include Elaborations describing critical behavior components,
(7) Include Typical examples,
(8) Include Questionable instances to enhance clarity and applicability.
Selecting Observation Settings: What type of settings? (2)
(1) Naturalistic settings
(2) Laboratory/Analog settings
Naturalistic settings characteristics (4)
- Naturalistic behavior – no constraints on behavior
- Multiple kinds of situations
- Multiple kinds of behaviors
- Need access
Laboratory/Analog settings characteristics (4)
- Greater control over stimuli
- Increase chance of observing infrequent behavior
- Possible sample biases
- Limitations on sampling of situations and behaviors
Types of Recording techniques/Sampling procedures (4)
- Real-time observations
- Event recording
- Duration recording
- Interval recording
Sampling prodecure: Real-time observations def
Recording event FREQUENCY and DURATION as they naturally occur without interruption
[Sampling procedure] Event recording def (2)
Aka frequency recording or tally method
- When FREQUENCY of response is the dimension of interest.
- Recording each occurrence of the target behavior during an observation SESSION or within specified INTERVALS
Sampling procedure: Duration recording (2)
- When DURATION of response is the dimension of interest.
- Recording the length of time from the beginning to the end of each occurrence of the target behavior
Sampling procedure: Interval recording
Aka Time sampling.
- Dividing an observation session into brief intervals and recording the target behavior if it occurs within any part of each interval.
Sampling procedure: TYPES of Interval recording
- WHOLE-interval recording: Behavior is scored only if it occurs throughout the ENTIRE interval
- PARTIAL-interval recording: Behavior is scored if it occurs only PARTIALLY within the interval
- MOMENTARY recording: Behavior is scored if it occurs during the moment when the interval ENDS
Observer effects/errors def
Systematic errors in behavioral observations that may result from using human observers
TYPES of Observer effects/errors (3)
(1) Reactivity
(2) Observer bias
(3) Observer drift
[Observer effects/errors] Reactivity def
Change in subjects’ behavior due to the presence of an observer
[Observer effects/errors] 5 contributing factors to Reactivity
- Valence or social desirability of behavior
- Subject characteristics
- Conspicuousness of observation
- Observer attributes
- Rationale for observation
[Observer effects/errors] Observer bias def
Systematic error in assessment stemming from observers’ expectancies, prejudices, and information-processing limitations.
[Observer effects/errors] Observer bias can be reduced by ______ (2)
- Rigorous training
- Use of low-inference observation codes.
[Observer effects/errors] Observer drift def
A decrease in observer consistency and accuracy after training
[Observer effects/errors] Observer drift develops when ____ (3)
- A recording-interpretation bias has evolved
- Definitions/measurement procedures are altered to suit changes in target behavior
- Observers are bored or inattentive
[Observer effects/errors] Observer drift can be reduced by ____ (2)
- Providing continuing training
- Calibrating all observers simultaneously
2 types of Reliability of Behavioral Observation Data
- Interobserver (rater) Agreement
- Intraobserver Reliability
Reliability of Behavioral Observation Data: Interobserver (rater) Agreement def & coeff
- Consistency among data collected by multiple independent observers
- Coefficient kappa: Agreement between two (or more) observers corrected for chance
Reliability of Behavioral Observation Data: Intraobserver Reliability def & coeff
- Consistency of data when a single observer repeatedly observes the same behavior
- Test-retest reliability, equivalent-forms reliability, split-half reliability, Cronbach alpha
Types of Validity of Behavioral Observation Data (3)
- Content
- Criterion
- Construct
Content Validity of Behavioral Observation Data: Questions asked
- Has the universe of target behaviors completely and unambiguously defined?
- Have the settings and temporal factors completely defined?
- Does the coding system adequately samples from the behavioral domain of interest?
Criterion Validity of Behavioral Observation Data: Questions asked
Have criteria defined at the same level of specificity as observed behaviors?
Construct Validity of Behavioral Observation Data: Questions asked
Implications of molar vs. molecular behavioral categories for construct validity
=> molar = easier to assess overarching constructs; molecular = harder to capture the overall construct
=> molar = easier to test convergent validity (broad behaviors are more likely to correlate with the larger construct); molecular = harder to assess (may not correlate strongly with the overall construct or other actions within the construct)
