2:Observational Measures Flashcards
Observational data
Collection techniques measure variable using the researchers senses or specialized equipment
- requires well-Defined OD
- utilize measurement apparatus
- coding schemes made by trained observers
- methods to cope with complexity
- not restricted to a particular type of research design or setting
Manual (real time or recorded for later analysis)
Behavioral: even coding, observer ratings
Physiological: instruments requiring observer judgment
Automated
Behavioral: lever-pressing, reaction time
Physiological: recording brain activity, record peripheral responses with automated equipment ( most automated)
Behavioral coding schemes
Behavioral category: classify observed. Behavioral events
Observer rating scale: quantify observed behavior on scale
Coding schemes must be defined very clearly to avoid confusion and ensure reliability
- keep categories as simple as possible to meet research goals
- take the time to carefully develop
Methods to help coding schemes
- make preliminary observations of behavior: create a list of behaviors of interest, become familiar with range of behaviors
- conduct a literature search to find coding schemes used in previous research studies: find previous validated coding scheme, adopt/modify.
Recording of behavior
Recording device: later review
-likely to miss observations, can review multiple times by multiple raters
Wearable recording devices: allow ongoing data collection in daily life
Subject reactivity
When subjects react to being watched Solutions: -blend in -wait it out so subjects familiarize to your response -measure the behaviors results
Subject expectations
Behavior may be affected by subject expectation about the hypothesis and or knowledge of assigned condition:
- minor deception: debrief after
- blind design: subjects do now know which condition they are in
Observer effects
When observers influence behavior of subjects to match observer expectations
-subtle cues from observer to subject:behavior changes
Observer bias
When observers see what they expect to see
-based on knowledge and measurement by observer: expectations influence interpretation/ measurement of behavior
Double blind design
Neither subjects nor observers know to which condition any subject has been assigned