Professional Documents
Culture Documents
Behavioral: Nonbehavioral:
Observational Studies
Behavioral:
Non-verbal (expressions & body movements) Linguistic (words of annoyance, surprise; conversation recordings) Extralinguistic (includes vocal pitch, loudness; interaction-interruption; pronunciation, accent; temporal-rate of speaking) Spatial (how persons relate physically to others)
Observational Studies
Nonbehavioral:
Record analysis (written, photographed, videographed) Physical condition analysis (safety features, store upkeeps, merchandise availability, cleanliness) Physical process or activity analysis (time/motion studies, traffic flows, paper-workflow in offices)
Observational Studies
Observer
Participant Relationship:
Direct / Indirect observation Participant is aware / unaware Observer s role Unstructured / structured Natural setting / laboratory
Observer Training
Experience level
Data Collection
Event sampling (what to observe) Time sampling (when to observe) One or two observers
How to observe
Unobtrusive Measures
Physical traces
Erosion (measures of wear) Accretion (measures of deposit) Wear & tear on book pages of library books Frequency of replacement of floor tiles in museums
Examples:
Experimentation
Are studies involving intervention by the researcher beyond that required for measurement There is at least one dependent and one independent variable
Experimentation
Advantages:
Ability to manipulate independent variable Extraneous can be controlled Convenient and cost effective Replication: experiments producing similar results
Experimentation
Disadvantages:
Artificiality of the laboratory Generalization based on experiments is open to question There could cost over-runs There are limits to types of manipulation and controls that are ethical
Experimental Designs
Laboratory Experiments:
Ability to control certain variables but limitation in respect of realism in research setting Respondents are likely to be influenced, because they are aware of the expected study Benefit is the reality of the setting Provides a high level of external validity
Field experiments
Experimental Designs
Definitions:
R = Random assignment of subjects O = Observation or measurement of dependent variable. If there are more than one observation then subscripts O1, O2 etc will be used X = the exposure of a test group to an experimental treatment (independent variable)
Experimental Designs
Quasi- or Semi-Experiments
Pre-Experimental Designs
X O1 X
One-Group Pretest
Posttest Design
O2
Pretest
Completely Randomized Design Randomized Block Design Latin Square Design Factorial Design
Pretest Design
Experimental group R X O1 Control group R O2 O1 is composed of treatment effect and extraneous factors O2 is composed of extraneous factors only Treatment effect = O1 O2 This design is less expensive & less time consuming
High price
Store#5 Store#1 Store#8
Medium price
Store#3 Store#4 Store#2
Factorial Design
0.5% A E I M
1% B F J N
1.5% C G K O
2% D H L P
2% 5% 7% 10%