Measurement

Chapter 8
Cooper and Schindler

Measurement
Consist of assigning numbers to empirical events in
compliance with a set of rules
The definition implies that measurement is a three-part process
Selecting observable empirical events
Using numbers or symbols to represent aspects of the events
Applying a mapping rule to connect the observation to the symbol

Example
Studying people who attend an auto show where all of the years new
models are on display
Gender
Styling characteristics

Characteristics of Measurement
Gender

Styling Characteristics

AB C D E
Desirability
of styling to
show attendees
A-E

Gender
Empirical
Observations of show

attendees
A-E

Assign
5 if Very Desirable
4 if Desirable
3 if neither
1
2 if undesirable
1 if Very Undesirable

Mapping Rule Assign

M if male
F if female

M
Symbol

AB C D E

(1 through 5)

2 3 4 5

What Is Measured? (I)

Concepts
Objects
Include the things of ordinary experience, such as tables, people, books
and automobiles
Also include things that are not as concrete, such as genes, attitudes,
neutrons and peer-group pressures

Properties

Are the characteristics of the objects

Physical properties
Psychological properties
Social properties

Researchers measure indicants of the properties of objects

What Is Measured? (II)

Age, Years of experience, Number of calls per week

It is not easy to measure properties

Motivation to succeed, ability to stand stress, problemsolving ability, and persuasiveness
There is often disagreement about how to operationalize the
indicants

Not only is it a challenge to measure such constructs,

but a studys quality depends on what measures are
selected or constructed, and how they fit the
circumstances

Scale Classifications
Employ the real numbers systems
The most accepted basis for scaling has
three characteristics
Number are ordered (Order)
Differences between numbers are ordered
(Distance)
The number series has a unique origin indicated
by the number zero (Origin)

Measurement Scales
Nominal
No order, or origin
Determination of equality

Ordinal
Order but no distance or unique origin
Determination of greater or lesser values

Interval
Both order and distance but no unique origin
Determination of equality of intervals or differences

Ratio
Order, distance, and unique origin
Determination of equality of ratios

Nominal Scales
Partition a set into categories that are mutually exclusive and
collectively exhaustive
Counting is the only arithmetic operation
Only labels and have no quantitative value

No order or distance relationship and have no arithmetic origin

No general used measure of dispersion
Several tests for statistical significance may be utilized
Chi-square test
For measures of association, phi, lambda, or other measure may be
appropriate

Ordinal Scales
Include the characteristics of the nominal scale plus an indicator
of order
Ordinal scales are possible if the transitivity postulate is fulfilled.
An extension of the ordinal concept occurs when more than one
property is of interest
Add and average ranks is technically incorrect
Use a multidimensional scale

Have another difficulty when combining the rankings of several

respondents
Convert the ordinal scale into an interval scale
Thurstones Law of Comparative Judgment

Ordinal Scales
Examples of ordinal scales include opinion and preference scales
Paired -comparison techniques

Ordinal scales have only a rank meaning

Statistical measures
Central tendency
median

Dispersion
Percentile or quartile

Correlation
Rank-order methods

Statistical significance
Nonparametric methods

Interval Scales
Has the powers of nominal and ordinal plus one additional strength
Incorporates the concept of equality of interval

Calendar time is interval scales

Zero time and zero degree(Centigrade and Fahrenheit) are arbitrary origin

Many attitude scales are presumed to be interval

Thurstones differential scale was an early effort to develop such a scale

Statistical measures
Central tendency (Arithmetic mean)
Dispersion (Standard deviation)
others (Product moment correlation, t-tests, and F-tests)

Ratio Scales

Incorporate all of the powers of the previous ones plus the provision for
absolute zero or origin
Represent the actual amounts of a variable
Examples are weight, height, distance, and area
In behavioral sciences, few situations satisfy the requirement of the
ratio scale(Psychophysics offering some exceptions)
In business research, we find ratio scale in many areas (money values,
population counts, distances)
Statistical measures
All statistical mentioned up to this point
Multiplication and division
Geometric mean, coefficients of variation

Sources of Measurement Differences

The respondent as an error source

Situation factors
The measurer as an error source
Instrument as an error source

 Validity

Sound Measurement

Content validity
Criterion-related validity (Concurrent validity, Predictive validity)
Construct validity

Reliability
Stability (Test-retest)
Equivalence (Parallel forms)
Internal consistency (Split-half, KR-20, Cronbachs alpha)

Practicality
Economy
Convenience
Interpretability

Criteria for Evaluating a

Measurement Tool
Validity
Refer to the extent to which a test measures what we
actually wish to measure

Reliability
Has to do with the accuracy and precision of a
measurement procedure

Practicality
Is concerned with a wide range of factors of economy,
convenience, and interpretability

Validity
Internal and external
Research Instrument internal validity
Measure what it is purported to measure
Does the instrument really measure what its designer claims it does?

Three major forms

Content validity
Criterion-related validity
Concurrent validity
Predictive validity

Construct validity

Content Validity
The extent to which it provides adequate
coverage of the topic under study
Determination of content validity is
judgmental and can be approached in
several ways
Through a careful definition of the topic
Use a panel of persons to judge

Criteria-Related Validity
reflects the success of measures used for prediction or
estimation
Predict an outcome
Estimate the existence of a current behavior or condition

Predictive and concurrent validity differ in time perspective

An opinion questionnaire that correctly forecasts the outcome of a
union election has predictive validity
An observational methods that correctly categorizes families by
current income class has concurrent validity

Any criteria measure must be judged in terms of four

qualities: relevance, freedom from bias, reliability, availability

Construct Validity
One may wish to measure or infer the presence of abstract
characteristics for which no empirical validation seems possible
Attitude scales
Aptitude tests
Personality tests

Example
Measuring the effects of ceremony on organizational culture
Ceremony was operationally defined would have to correspond to an
empirically grounded theory

Convergent validity
Discriminant validity

Reliability
A measure is reliable to the degree that it
supplies consistent results
Reliability is a contributor to validity and is a
necessary but not sufficient condition for
validity
Reliability is concerned with estimates of the
degree to which a measurement is free of
random or unstable error

Stability
A measure is said to be stable if you can
secure consistent results with repeated
measurements of the same person with the
same instrument
Test-retest

Equivalence
Considers how much error may be introduced by
different investigators (in observation) or different
samples of items being studied (in questioning or
scales)
Equivalence is concerned with variations at one
point in time among observers and samples of items
Interrater reliability may be used to correlate the
observations or scores of the judges and render an
index of how consistent their ratings are

Internal Consistency
Use only one administration of an instrument or
test to assess consistency or homogeneity among
the items
Split-half techniques
Spearman-Brown correction formula

The test splitting may influence the internal

consistency coefficient
Kuder-Richardson Formula 20
Cronbachs Coefficient Alpha