Transcript Team Problems/Solutions: Team Facilitation for Effective

Business Research
Gerald L. Blakely
INTRODUCTION
• Why do research?
• Purposes of research
• Ethics in research
• Ways of knowing
RESEARCH DESIGNS
• Correlational
• Qualitative
• Experimental
• Quasi-experimental
EXPERIMENTAL DESIGNS
• Pre-experimental designs
1. X O
2. O X O
3. X O
O
EXPERIMENTAL DESIGNS (cont.)
• True experimental designs
4. R O X O
R O
O
5. R
R
X O
O
6. R O X O
R
O
R
X O
R O
O
EXPERIMENTAL DESIGNS (cont.)
• Quasi-experimental designs
7. O O O X O O O
Time series
8. O X O Non-equivalent control group
O
O
9. O O O X O O O (X) O O O
Withdrawal
CAUSAL INFERENCE
• Temporal antecedence
• Covariation
• Plausible alternatives ruled out
– Reverse causation
– Reciprocal causation
– Spurious relationships
– Moderators
– Mediators
STUDY VALIDITY (Internal)
•
•
•
•
•
•
•
•
Selection of participants
Testing
Regression to the mean
History
Maturation
Mortality
Compensatory rivalry
Resentful demoralization
PSYCHOMETRIC PROPERTIES
OF MEASURES (RELIABILITY)
• Test re-test
• Parallel forms
• Inter-rater
• Internal consistency
PSYCHOMETRIC PROPERTIES
OF MEASURES (VALIDITY)
• Face
• Content
• Construct
• Criterion related
– predictive
– concurrent
SCALE CONSTRUCTION
• Determine clearly what you want to
measure
• Generate an item pool
– Redundancy
– Number
– Item characteristics
ITEM CHARACTERISTICS (cont.)
•
•
•
•
•
•
•
item length
reading level
double-barreled
reverse scored
leading questions
delicate/sensitive
social desirability
SCALE FORMAT
• Thurstone
• Semantic differential
• Summated rating (Likert type)
• Guttman
• Rank order
QUESTIONNAIRE
CONSTRUCTION
• White space
• ID
• Physical form
• Language
• Length
QUESTIONNAIRE
CONSTRUCTION (cont.)
• Simplicity
• Variation
• Page format
OPTIMIZING RESPONSE RATES
• No junk mail
• Return envelope
• Follow up
• Relevance
• Cover letter
SAMPLING: BASIC DEFINITIONS
• A good sample is one where every
member of the population has an equal
probability of being selected for the
sample
• Sampling error - difference between
sample and entire population
• Sampling biases - non-random errors due
to inadequate data or clerical mistakes
• Non-sampling biases - errors that occur
even if we sampled the entire population
SAMPLING: SAMPLE DESIGN
• Probability samples
– Simple random samples
– Systematic random samples
– Stratified samples
– Cluster samples
SAMPLING: SAMPLE DESIGN (cont.)
• Nonprobability samples
– Purposive
– Quota
– Chunk
– Volunteer
– Convenience
Kinds of Statistics
Statistics
Descriptive
Inferential
Estimation
Modeling
Relationships
Hypothesis
Testing
Kinds of Data