Quantitative research in public management session 3

Session 3 – Part 1
Measurement

School of Business
I N T E R N AT I O N A L U N I V E R S I T Y - V I E T N A M N AT I O N A L U N I V E R S I T Y H C M C


Measurement and Scale

•Common properties being measured :

• Demographic/Economic/Social: age, gender, income, edu.
• Psychological / Life style
• Attitude / belief / value / preferences
• Awareness / knowledge
• Intention: purchase, not purchase, etc.
• Motive: need, want, motivation, etc.
• Behavior: what, how much, who, when, where, how, etc.

Tri D. Le

2


Generals

•Measuring = assigning numbers to properties/ attributes of objects/ events according to given rules.
•Numbers = symbols to represent the property in the empirical system.
•The nature of the relationship existing in the empirical system determines types of numerical manipulations.
•BR involves human attitude, behavior, etc.  the measurement becomes more complicated.
•Less advanced measurement scales are used in BR

Tri D. Le

3


Researcher’s “two worlds”
Learning

Learning

positive impact

motivation

performance

Theoretical World

Empirical World

Exam grade

Eager to attend class
Read/Write

linear correlation

Understand contents
Knowledge acquired

Active participation

Tri D. Le

4


Tri D. Le

5

Operat
e

Defnit

Variabl

onal

t

ct

Concep

Constru

on

Review of Terms


Measurement

Select
measurable phenomena

Develop a set of
mapping rules

Apply the mapping rule
to each phenomenon

Tri D. Le

6


Characteristics of Measurement

Tri D. Le

7


Types of Scales

Nominal

Nominal

Ordinal
Ordinal

Interval
Interval

Rato
Rato

Tri D. Le

8


Levels of Measurement

Classifcaton
Classifcaton

Nominal
Nominal

Ordinal
Ordinal

interval
interval

Rato
Rato

Tri D. Le

9


Nominal Scales

Mutually
Mutually Exclusive
Exclusive

Collectively
Collectively Exhaustive
Exhaustive Categories
Categories

Classification
Classification Only
Only

Tri D. Le

10


Levels of Measurement

Classifcaton
Classifcaton

Nominal
Nominal

Classifcaton
Classifcaton

Ordinal
Ordinal

Order
Order

interval
interval

Rato
Rato

Tri D. Le

11


Ordinal Scales

Nominal
Nominal Scale

Scale Characteristics
Characteristics

Order
Order

Implies
Implies greater
greater than
than or
or less
less than
than

Tri D. Le

12


Levels of Measurement

Nominal
Nominal

Ordinal
Ordinal

interval
interval

Classifcaton
Classifcaton

Classifcaton
Classifcaton
Order
Order
Classifcaton
Classifcaton

Distance
Distance

Order
Order

Rato
Rato

Tri D. Le

13


Interval Scales

Ordinal
Ordinal Scale
Scale Characteristics
Characteristics

Equality
Equality of
of interval
interval

Equality
Equality of
of distance
distance between
between numbers
numbers

Tri D. Le

14


Levels of Measurement

Classifcaton
Classifcaton

Nominal
Nominal

Classifcaton
Classifcaton

Ordinal

Ordinal

Order
Order
Classifcaton
Classifcaton

interval
interval

Distance
Distance

Order
Order

Rato
Rato

Tri D. Le

Classifcaton
Classifcaton

Distance
Distance

Order
Order

Natural
NaturalOrigin
Origin

15


Ratio Scales

Interval
Interval Scale
Scale Characteristics
Characteristics

Absolute
Absolute Zero
Zero

Tri D. Le

16


Examples of Data Scales

Tri D. Le

17

Summary of scales

Tri D. Le

18


Sources of Error

Respondent

Situaton

Measurer

Instrument

Tri D. Le

19


Measurement errors

•Systematic error occurs in all observations; caused by method bias (tools) or social desirability
responses (respondents)

•Random error occurs randomly in a number of observations.

Om = Observed measurement

Se = Systematic error

Ts = True score
Re = Random error

Om = Ts + Se + Re

Tri D. Le

20


Validity and Reliability

 Validity: The extent to which the measure avoids both systematic and random errors.
 Validity: “truthfulness”
 Are we measuring what we intend to measure ?

 Reliability: The extent to which the measurement avoids random error.
 Reliability: consistency, stability, accuracy

Tri D. Le

21


Understanding Validity and Reliability

Tri D. Le

22


Evaluating Measurement Tools

Validity

Criteria
Criteria

Reliability
Reliability

Practicality

Tri D. Le

23


Validity Determinants

Content

Criterion

Construct

Tri D. Le

24


Summary of Validity Estimates

Tri D. Le

25