Business Research for Decision Making Sixth Edition by Duane Davis Chapter 7 Foundations of...
-
Upload
clare-burns -
Category
Documents
-
view
213 -
download
0
Transcript of Business Research for Decision Making Sixth Edition by Duane Davis Chapter 7 Foundations of...
Business Research for Decision MakingSixth Edition
by
Duane Davis
Chapter 7
Foundations of Measurement
PowerPoint Slides for the Instructor’s Resource Manual for
Foundations of Measurement
concerned with the assignment of numbers to empirical rules according to a set of rules
goal is to translate characteristics and properties of empirical events into a form that can be analyzed.
1. Observable empirical events2. Use of numbers to represent these
events3. A set of mapping rules
Measurement:
NecessaryComponents:
Example of the Relationships Among the Three Components of MeasurementFigure 7.1
The Measurement ProcessFigure 7.2
Levels of Measurement
NominalOrdinalIntervalRatio
Nominal
Lowest level of measurement Numbers assigned to categories but
the numbers have no meaning1. 1= male; 2= female
2. 1= sophomore; 2 = junior; 3= senior Loss of information if other scales
converted to nominal
Ordinal
Rank order on some characteristic or property
Intelligence, aptitude, and personality test scores are generally ordinal E.g. not amount of intelligence, but rank order
of intelligence
No absolute zeros and the intervals are not equal.
Ordinal Example
What is your age?1 = under 182 = 18-223 = 22-304 = 30-405 = over 40
Ordinal Example
Overall, I am satisfied with myself 1) strongly disagree 2) disagree 3) neutral 4) agree 5) strongly agree
Interval
One unit on the scale represents the same magnitude on the trait or characteristic being measured across the whole range of the scale.
For example, if anxiety were measured on an interval scale, then a difference between a score of 10 and a score of 11 would represent the same difference in anxiety as would a difference between a score of 50 and a score of 51.
Rare to non-existent in behavioral research
Ratio
Have a true zero, so distance between points has meaning
Incorporate the properties of all other scales
What is your age?
Three Basic Types of Validity in Measurement
Table 7.2
A B
a b
Construct
MeasureConcurrent validity
Predictive validity
Reliability
Refers to the consistency and stability of a measurement scale.
Generally use a coefficient of reliability to assess the reliability of a measure.
Can range from 0 to 1.0, with 1.0 perfect reliability.
Guidelines for Minimum Reliabilities: 0.7 for exploratory research 0.8 for basic research 0.9 or better in applied settings where
important decisions will be made from scales
Methods for Reliability Assessment
1. Test-Retest Method
2. Alternative Forms
3. Internal Consistency
- Split-half
- Cronbach-Alpha/KR-20
Illustrations of Possible Reliability/Validity Situations in Measurement
Figure 7.3
Improving reliability
Write items that are unambiguousAdd more items of equal kind and quality
General rule of thumb is at least 3-4 items for any scale
Make the instructions clear and standard
Managerial Considerations(Why get involved?)
Bad measurement, no value to managers
Scaling sets limits on analysisGlobal and international research
compounds problemsMeasurement issues are early in
the business research process – compounds problems
Managerial Strategies for Ensuring Good Measurement
Require clearly defined constitutive concepts
Require operational definitionsRequire multi-item scales when possibleRequire validity, reliability and pretest
assessments when possibleRequire analytical foresight
Key Managerial Questions Pertaining to the Measurement Process
Exhibit 7.2