Measurement and

UNIT 8 MEASUREMENT AND SCALING Scaling Techniques

TECHNIQUES
Structure

8.0 Objectives
8.1 Introduction
8.2 Concept of Measurement
8.3 Measurement Issues in Research
8.4 Scales of Measurement
8.5 Criteria for Good Measurement
8.5.1 Reliability
8.5.2 Validity
8.5.3 Practicality
8.6 Errors in Measurements
8.7 Scaling Techniques
8.7.1 Comparative Scaling Techniques
8.7.2 Non-Comparative Scaling Techniques
8.8 Let Us Sum Up
8.9 Key Words
8.10 Some Useful Books
8.11 Answers or Hints to Check Your Progress Exercises

8.0 OBJECTIVES
After going through this Unit, you will be able to:
• state the concept of measurement and its need;
• explain the various scales of measurement;
• discuss the various scaling techniques; and
• identify the criteria for good measurement.

8.1 INTRODUCTION
Measurement is the foundation of scientific enquiry. As we know that research
begins with a ‘problem’ or topic. Thinking about a problem results in
identifying concepts that capture the phenomenon being examined. Concepts
are mental ideas representing the phenomenon. The process of
conceptualization involves defining the concepts abstractly in theoretical terms.
Operationalisation of concepts for research purpose involves moving from the
abstract to the empirical level. This underlines the need to measure the various
attributes of the people, the characteristics of objects or phenomenon. Further
issues like decent work, human wellbeing, happiness, quality of education etc.
which have several qualitative and quantitative dimensions are emerging
important issues of research in economics. Such issues are being probed
53
Research Design and through development of indicators and composite indexes which necessarily
Measurement involve measurement of such indicators. Hence, a student of economics is
expected to be well versed in the measurement scales, criteria of good
measurement, and important scaling techniques. This unit throws light on these
issues. Let us begin to discuss the concept of measurement.

8.2 CONCEPT OF MEASUREMENT

Simply speaking the process of assigning numbers to various attributes of
people, objects or concepts is known as measurement. Technically,
measurement is a process of mapping aspects of a domain to other aspects of a
range according to some rule of correspondence.Tyler(1963) defines
measurement as, “assignment of numerals according to rules”. Nunally
(1970)viewed that, “Measurement consists of rules, for assigning numbers to
objects in such a way to represent quantities of attributes”. Thus, Nunally
focuses on both rules and manner in which numbers are assigned to an object.

According to Campbell, “Measurement is defined as the assignment of

numerals to objects or events according to rules”. The fact that numerals can be
assigned under different rules leads to different kinds of scales and different
kinds of measurement.

Thus, measurement is a process to assign numbers or other symbols to

characteristics of objects according to certain rules. Assigning numbers permit
statistical analysis of the resulting data and facilitate the communication of
measurement rules and results. The rules for assigning numbers hence need to
be standardized and uniformly applicable. In the assignment process, there
must be one to one correspondence between the numbers and characteristics
being measured. In this process, variables can be divided into two basic types –
quantitative andqualitative variables.

8.3 MEASUREMENT ISSUES IN RESEARCH

A number of concerns arise in the process of measurement which need to be
addressed by the researcher. Some of the important issues include the following:
1) Whether the underlying characteristics of the concept allows ordering
(ordinal level) or categorizing (nominal level)?
2) Whether the features of the concept are discrete or continuous with fine
gradation?
The answer of these two questions will enable to determine the level of
measurement inherent to the concept. As a general rule, efforts should be
made that measurement represent the highest scale of measurement for a
concept because it allows the use of more powerful statistical techniques
for analysis.
3) Another important issue pertains to number of indicators for measurement
of a concept. Some simple concepts can be measured by one indicator
whereas abstract concepts are measured with more than one indicators.
How many indicators are appropriate to measure a concept is to be decided
by the researcher.
4) Another measurement issue concerns the source of valid and reliable
measure. Hence, considerable attention should be given to identify valid
and reliable measures for the study.
54
5) Measurement should be free from measurement errors like invalidity error Measurement and
and unreliability. Scaling Techniques

6) Measurement validity is distinct from internal validity and external validity

because these two are separate research design issues.
7) A final measurement issue concerns the proper use of available data. For
that, a research should be well aware of all available data sources and the
types of data available with the various data compilation agencies.

8.4 SCALES OF MEASUREMENT

Before discussing different levels of measurement, let us remember that there
are three postulates of measurement: (i) Equalities or identities, (ii) Rank order,
(iii) Additivity.
You were introduced the elementary concept and types of scales of
measurement in Unit 1. Even at the cost of repetition let us recall that Stanley
S.Stevens (1946) in his seminal paper, “On the theory of scales of
measurement” published in “Science” classified types of scales into four
categories: (1) Nominal, (2) Ordinal, (3) Interval, (4) Ratio.

1) Nominal Scale
A qualitative scale without order is called nominal scale. In nominal scale
numbers are used to name identity or classify persons, objects, groups,
gender, industry type. Even if we assign unique numbers to each value, the
numbers do not really mean anything. This scale neither has any specific
order nor it has any value. In case of nominal measurement, statistical
analysis is attempted in terms of counting or frequency, percentage,
proportion, mode or coefficient of contingency. Addition, subtraction,
multiplication and division are not possible under this scale.

2) Ordinal scale
A qualitative scale with order is called ordinal scale. In ordinal scale
numbers denote the rank order of the objects or the individuals. Numbers
are arranged from highest to lowest or lowest to highest. For example,
students may be ranked Ist, 2nd and 3rd in terms of their academic
achievement. The statistical operations which can be applied in ordinal
measurement are median, percentiles and rank correlation coefficient.
Ordinal scales do not provide information about the relative strength of
ranking. This scale does not convey that the distance between the different
rank values is equal. Ordinal scales are not equal interval measurement.
Further this scale does not incorporate absolute zero point.
3) Interval scale
Interval scale includes all the characteristics of the nominal and ordinal
scale of measurement. In interval scale numerically equal distance on the
scale indicate equal distances in the attributes of the object being measured.
For example, a scale represent marks of students using the attributes range
0 to 10, 10 to 20, 20 to 30, 30 to 40 and 40 to 50, and so forth. The mid
point of each range (ie. 5, 15, 25, 35 and 45 etc.) are equidistance from
each other. The data obtained from an interval scale is known as interval
data. The appropriate measures used in this scale are: arithmetic mean,
standard deviation, Karl Pearson’s coefficient of correlation and tests like t-
test and f-test. We cannot apply coefficient of variation in the interval scale.
55
Research Design and 4) Ratio Scale
Measurement
Ratio scale has all the characteristics of nominal, ordinal and interval scale.
It also possesses a conceptually meaningful zero point in which there is a
total absence of the characteristic being measured. Ratio scales are common
among physical sciences rather than among social sciences. The examples
of ratio scales are the measures of height, weight, distance and so on. All
measures of central tendencies that can be used in this scale include
geometric and harmonic means.

The properties of these four scales can be summarized in following tabular

form:

Properties Category Ranking Equal Zero Point

Scale interval
Nominal
Ordinal
Interval
Ratio

Properties Indicates Indicates Indicates Absolute

Scale Difference Direction of Amount of Zero
Difference Difference
Nominal ×
Ordinal × ×
Interval × × ×
Ratio × × × ×

You will notice in the above tables that only the ratio scale meets the criteria for
all four properties of scales of measurement. Interval and Ratio scale data are
sometimes referred to as parametric and Nominal and Ordinal data are referred to
as non-parametric.
Examples of Scales of Measurement –
Scale Example Statistics
Nominal Gender Frequency, percentage, proportion Mode,
Yes-no Coefficient of contingency, chi-square (x2)
Students roll number
Objects/Groups
Ordinal Class Rank, Median, percentile, rank, correlation,
Socio Economic status Range
Academic Achievement
Interval Student grade point, Mean, correlation, Range, Standard
Temperature, Calendar deviation, rank order variance, Karl
dates, Rating Scale pearson’s correlation, t-test, f-test etc.
Ratio Weight, Height, Salary, Mean, Median, Mode, Range, Variance,
Frequency of buying a Standard deviation, coefficient of variation,
product rank order variance, Karl pearson’s
correlation, t-test, f-test

56
 Measurement and
8.5 CRITERION FOR GOOD MEASUREMENT Scaling Techniques

There are three major criteria for evaluating measurement:

1) Reliability
2) Validity
3) Practicality

8.5.1 Reliability
Reliability refers to the degree to which the measurement or scale is consistent
or dependable. If we use same construct again and again for measurement, it
would lead to same conclusion. Reliability is consistency in drawing conclusion.

Reliability and Validity

. . ..
. . ..

Reliable and Valid Valid but not Reliable Reliable but not valid

8.5.2 Validity
Validity refers to the extent an instrument or scale tests or measures what it
intends to measure. This means validity is the extent to which differences
found with a measuring instrument reflect true differences among those being
tested. This can be done by seeking other relevant evidences. There are three
types of validity: content validity; criterion validity and construct validity.
Content validity indicates the extent to which it provides coverage of the issues
under study. Criterion validity examines how well a given measure relates to
one or more external criterion based on empirical observations. And, construct
validity explains the variation observed in tests conducted on several
individuals. Construct validity is closely related with factor analysis.

8.5.3 Practicality
From practical view point, measure should be economical, convenient and
interpretable. The measure should not be lengthy and difficult. The measure
can be done by highly specialized persons.

8.6 ERRORS IN MEASUREMENTS

Measurement need to be precise and unambiguous for validity and reliability of
any research study. Hence, measurement should be either free from errors or
have minimum error. Any measurement usually involves two types of errors:
• Measurement invalidity

• Unreliability

Measurement invalidity refers to the degree to which the measure incorrectly

captures the concept.

57
Research Design and Unreliability refers to inconsistency in what the measure produces under
Measurement repeated uses. If any measure, on average give some score for a case on
variable and gives other score, when it is used again, it is said to be unreliable.

The possible sources of error are:

a) Respondent by not expressing strong negative feelings.
b) Situation wherein interviewer and respondent are not in good rapport.
c) Measurer – errors may also creep because of incorrect coding, faulty
tabulations and statistical calculations or behavior or attitude of the
interviewer.
d) Instrument errors may also arise because of the defective measuring
instrument.

Check Your Progress 1

1) What do you mean by measurement?

………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
2) Give examples of interval scale and ratio scale.
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
3) How will you examine the validity of a measurement?
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
4) Identify three measure concerns that need to be addressed in the process of
measurement.
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………
………………………………………………………………………………

58
 Measurement and
8.7 SCALING TECHNIQUES Scaling Techniques

Scaling describes the way of generation of a continuum upon which measured

objects are located. Broadly, scaling techniques can be classified into two
categories: comparative, and non comparative scaling.

Scaling Techniques

Comparative Non-comparative
Scaling Techniques Scaling Techniques

Paired Rank Constant Continuous Itemised Category Cumulative

Comparison order Sum Rating Scale Rating Scale Scale Scale

8.7.1 Comparative Scaling Techniques

Comparative scales involve the comparison of objects directly with one
another. For example, in a study of consumer preferences for different telecom
services, a respondent may be asked to rank choice according to his
preferences. In this technique, data must be interpreted in relative terms and
have only ordinal or rank order properties. In comparative scales small
differences between stimulus objects can be detected.

Example
Rank the following factors in order of your importance in choosing mobile
services (Assign 1 to most important and 5 to least important factor. Please do
not repeat the ranks.)
Factor: Price, connectivity, call drop, Internet, Download speed.
Rank :……… ……… …… ……. ……….
There are various comparative scaling techniques. Three most commonly used
comparative scaling techniques are discussed here.
a) Paired comparison
b) Rank order
c) Constant sum

a) Paired Comparison
The paired comparison is most commonly used scaling technique. This
method is simply a binary choice. In this method respondents choose the
stimulus or items in each pair that has the greater magnitude on the choice
dimension they are instructed to use. This method is used when the study
requires to distinguish between the two objects. The data obtained in this
method is ordinal in nature.

59
Research Design and Example
Measurement
In a study of consumer preferences for the two brands of milk product i.e.
Amul and Mother Dairy, a consumer is asked to indicate which one of the
two brands he would prefer for personal use:

1) Which milk product of the following you prefer on the basis of taste.
Please tick mark (√).

Amul Mother Dairy

2) Which milk product will you prefer on the basis of packaging? Please
tick mark (√).

Amul Mother Dairy

3) Which product will you prefer on the basis of price? Please tick mark (√).

Amul Mother Dairy

This technique is useful when the researcher wants to compare two or

more than two objects. If there are more than two objects (for example
n objects) to compare, the total comparison will be;

n [n - 1]
Number of comparison = ;
2
n= number of objects.

b) Rank Order

This method is very popular among researchers and provides ordinal data.
In this method, respondents are provided various objects and asked to rank
the objects in the list. Rank order method is less time consuming. In this
method, if there are n objects, only (n-1) decisions need to be made.
Respondent can easily understand the instructions for ranking. This
technique produces ordinal data.

Example
Rank the various brands of mobile phone in order of preference. The most
preferred can be ranked 1, the next as 2 and so on. The least preferred will
have the last rank. No two brands should receive the same rank number.

60
 Measurement and
Scaling Techniques

Sl. No. Brand Rank

1 Nokia
2 Motorola
3 Samsung
4 HTC
5 Karbonn
6 Spice
7 Xolo
8 Gionee
9 LG
10 Sony

c) Constant Sum
In constant sum scaling, respondents are asked to allocate a constant sum
of units to a specific set of objects on the basis of pre-defined criterion. If
an object is not important, the respondent can allocate zeropoint and if an
object is most important may allocate maximum points out of the fixed
points. The total fixed points are 100. The total may be taken as some
other value depending on the study.

Example
Allocate preferences regarding a movie based on various predefined
criteria. Respondents were asked to rate each criteria in such a way that the
total becomes 100.

Criteria Respondent Preference

Content Quality 20
Music 20
Sound 15
Story 30
Breadth of Coverage (local, regional 15
Supplements, national & global)
Total 100

The data obtained in this method is considered an ordinal scale. The advantage
of this method is that it provides fine discrimination among objects without
consuming too much time. However, allocation over large number of objects
may create confusion for the respondent. This method cannot be used as a
response strategy with illiterate people and children.

61
Research Design and 8.7.2 Non Comparative Scaling Techniques
Measurement
Non-comparative scaling techniques involve scaling of objects independent to
some specific standard. Respondent evaluate only one object at a time. For
example, in a study of consumer preferences for different telecom services, a
consumer may be asked to rate a list of factors that he/she would consider
while choosing a particular telecom company.

Example
Please rate the following factors you think important in choosing mobile
service. Rate 1 to least important and 5 to most important (rating could be on
any scale. In this example, we have used a 5 point scale).

Factor Price Connectivity Call Download Internet

Drop Speed
Rating

In this scaling technique, data is usually in interval scale. It can be continuous,

metric/numeric also. Some of the commonly used non-comparative scaling
techniques are –

i) Continuous Rating Scale

ii) Itemised Rating Scale
iii) Category Scale
iv) Cumulative Scale

i) Continuous Rating Scale

This rating scale is also known as graphic rating scale.In continuous
rating scale, respondents indicate their rating by marking at appropriate
position on a line. The line is labeled at both ends from one extreme
criterion to the other. The line may contain points 0,10,20,…..,100.

Example
How would you rate a magazine with regard to its quality.

Quality Scale Measurement

Indicators
Content Most_________________x__________________Least
Coverage Above 80 60 40 20 0
Language Most__________x_________________________Least

Presentati Most_________________x__________________Least
on Style
A very large number of ratings are possible if the respondents are literate
to understand and accurately differentiate the objects. The data generated
from continuous rating scale can be treated as numeric and interval data.
The researcher can divide the line into as many categories as desired and
assign scores based on the categories under which the ratings fall.

62
ii) Itemized Rating Scale Measurement and
Scaling Techniques
In this scale, respondents are provided with a scale having numbers/
descriptions associated with each category. The respondents are asked to
select the best fitting category with the object.The commonly used
itemised rating scale are:

a) Likert Type Scale (Summated Scale)

b) Semantic Differential Scale
c) Stapel Scale

a) Likert Type Scale (Summated Scale)

Likert (1932) proposed a simple and straight forward method for scaling
attitudes that is most frequently used today. This scale is also known as
summated rating scale. Summated scales consist of a number of
statements which express either a favourable or unfavourable attitude
towards the given object to which the respondant is asked to respond to
each of the statements in terms of several degrees of agreement or
disagreement. Let us understand this method with an example.

Against the following statements relating to psychological well-being, please

tick mark any one of the following options shown against the statement

Sl. Statements Strongly Agree Undecided Disagree Strongly

No. Agree Disagree
5 4 3 2 1
1 I always live in present
and do not worry about
future.
2 Many people think without
purpose in life but I am not
such type of persons.
3 I generally feel sense of
perfection during the
moments I live in.
4 I am very well in
managing and discharging
most of my
responsibilities.
5 I give importance to the
innovative ideas and
experiences that challenge
my thought about me and
the world.

Depending on the wording of an individual item, an extreme answer of strongly

agree or strongly disagree will indicate the most favorable response on the
underlying attitude measured by the questionnaire. This scale is relatively easy
and quick to compute. The data in this scale is of interval scale.

Some of the important advantages of this scale are: (i) it is relatively easy to
construct as it can be performed without a panel of judges, (ii) it is considered
more reliable because respondents answer each statement included in the
63
Research Design and instrument, (iii) it can easily be used in respondent centred and stimulus
Measurement centred studies, (iv) this takes less time to construct. However, there are several
limitations of this type of scale i.e. (i) with this scale, we can simply examine
whether respondents are more or less favourable to a topic but it is difficult to
tell how much more or less they are, (ii) This does not rise more than to a
ordinal scale structure, (iii) the total score of an individual respondent has little
clear meaning since a given total score can be secured by a variety of answer
patterns.

b) Semantic Differential Scale

This scale was developed by Osgood, suit &Tannenbaum (1957). Semantic
differential scale includes a seven point scale in comparison to Likert’s five
point scale. The semantic differential scale is based on the proposition that an
object can have several implied or suggestive meanings to an expressed
opinion. The semantic differential scale can be scored on either -3 to +3 or 1
to 7. This scale can also provide interesting comparison between products
organizations and so on. The results of this scale are further analysed by the
factor analysis.

Example: A researcher has developed an item seeking your perceptions

regarding magazine. Please mark√on each line that best indicates your
perception. Please be sure to mark every scale and do not omit any scale.

India Today Magazine is

+3 +2 +1 0 -1 -2 -3
Easy to read ____ : ____ : ____:____:_√__:____:____ Hard to read
Modern ____ : ____ : ____:____:_√__:____:____ Old fashion
Rational ____ : ____ : ____:____:____:____:____ Emotional
Unreliable ____ : ____ : ____:____:____:____:____ Reliable
Worthless ____ : ____ : ____:____:____:____:____ Valuable
Unorganized ____ : ____ : ____:____:____:____:____ Organized
Orthodox ____ : ____ : ____:____:____:____:____ Liberal
Vain ____ : ____ : ____:____:____:____:____ Modest

Semantic differential scale provides a very convenient and quick way of

gathering impressions on one or more than one concept. The data generated
from this scale can be considered as numeric in some cases, and can be
summed to arrive total scores adjectives. It must define a single dimension and
each pair must be bipolar opposites labeling the extremes.

Like Likert Type Scale, this scale has also several advantages: (i) it is an
efficient and easy way to secure attitude from a large sample, (ii) the total set
of responses (both direction) gives a comprehensive picture of the meaning of
an object as well as a measure of the subject of the rating, (iii) it is a
standarised technique that can be easily repeated.

However, this technique escapes many of the problems of response distortion

found with more direct methods.

64
c) Stapel Scale Measurement and
Scaling Techniques
Stapel scales are named after John stapel who developed these scales. Stapel
scale consists of a single criterion in the centre with 10 categories numbered
from -5 to +5 without a neutral point (zero). The scale is usually presented
vertically. Negative rating indicates that the respondent inaccurately describes
the object and positive rating indicates that the respondent describes the object
accurately.

Example
Rate the Departmental store by marking (√) on the following factors. +5
indicate that the factor is most accurate for you and -5 indicate that the factor is
most inaccurate for you.

+5 -5
+4 -4
+3 -3
+2 -2
+1 -1
Services Products
-1 +1
-2 +2
-3 +3
-4 +4
-5 +5

In this scale, data generated is interval data. In this method, data can be
collected through telephonic interview. The data obtained from Stapel
scale can be analysed in the same way as semantic differential scale.

iii) Category Scale

In Category scaling method, objects are grouped into a predetermined
number of categories on the basis of their perceived strength along certain
dimension. This scale is a dichotomous scale. This method is useful for
socio demographic questions. The response is this category we get,
typically ‘Yes’ or ‘No’ type. Data in this category is either nominal or
ordinal. Under this category scale, instructions and response tasks are
quick and simple, and many options can be included.

Example
A) Single Category Scale
1) Do you own a house?

Yes No

65
Research Design and 2) Do you own a car?
Measurement

Yes No

3) Do you own a Mobile phone?

Yes No

b) Multiple Category
1) Please indicate your income by marking (√) in the income
group you fall

5000 - 10000
10000 - 15000
15000 - 20000
20000 - 25000

iv) Cumulative Scale (Guttman Scale)

Like other scales, it consists of series of statements to which a respondent
expresses his agreement or disagreement. The statements are in a form of
cumulative series i.e. in a way, an individual who replies favourably to
say item no. 3 also replies favourably to item no. 2 and 1 and so on. The
individual’s score is worked out by counting the number of points
concerning the number of statements he answers favourably. Knowing
the total score, we can estimate as to how a respondent has answered
individual statements constituting cumulative scales.

Let us understand with the illustration of an example.

Here is an example of a Guttman scale - the Bogardus Social Distance
Scale:
(Least extreme)
1) Are you willing to permit immigrants to live in your country?
2) Are you willing to permit immigrants to live in your community?
3) Are you willing to permit immigrants to live in your neighbourhood?
4) Are you willing to permit immigrants to live next door to you?
5) Would you permit your child to marry to an immigrant?
(Most extreme)

Cumulative scale (scalogram analysis) like any other scaling technique has
several advantages as well as limitations. The advantages are – It assures that
only a single dimension of attitude is measured, (ii) Researcher’s subjective
judgement is not allowed to creep in the development of scale since the scale is
determined by the replies of respondents.

66
Its main limitation is that (i) In practice perfect cumulative or unidimesional Measurement and
scale are very rarely found and approximation is used, (ii) Its developmental Scaling Techniques
procedure is cumbersome in comparison to other scaling methods.

Check Your Progress 2

1) What is the distinction between comparative measuring scale and non
comparative measuring scale?
……………………………………………………………………………….
……………………………………………………………………………….
……………………………………………………………………………….
……………………………………………………………………………….
2) What do you mean byLikert Scale? Give its two examples.

……………………………………………………………………………….
……………………………………………………………………………….
……………………………………………………………………………….
……………………………………………………………………………….
3) What are the sources of measurement errors?

……………………………………………………………………………….
……………………………………………………………………………….
……………………………………………………………………………….

8.8 LET US SUM UP

Operationalisation of concepts for research purpose need measurement of
various attributes of the people, the characteristics of objects or phenomenon.
Measurement is a process to assign numbers to the characteristics according to
certain rules. Measurement scales are of four types – nominal, ordinal, interval
and ratio. A number of concerns arise in the process of measurement.
Important among them are: the scale measurement i.e. the validity, reliability
and practicality, number of indicators for measurement etc. Scaling techniques
are broadly of two types – comparative, and non-comparative. Within
comparative category, paired, rank order and constant sum are important
techniques. Within non-comparative techniques, four sub-categories i.e.
continuous rating scale, itemized rating scale, category scale and cumulative
scale have been covered in this unit. All these scaling techniques have their
relative advantages and limitations and are used depending upon the goal and the
nature of characteristics of objects/phenomenon to be measured.

8.9 KEY WORDS

Measurement : The assignment of numerals to objects or events
according to rules.
Nominal : Numbers are used to name identity or classify
persons, objects, groups or gender.
67
Research Design and Ordinal Scale : Scale with order is called ordinal scale.
Measurement
Interval Scale : In this scale numerically equal distance on the
scale indicate equal distances in the attributes of
the object being measured.
Ratio Scale : A conceptually meaningful zero point in which
there is a total absence of the characteristics
being measured.
Validity : The extent to which an instrument or scale tests
or measures what the measurement or scale i.e.
consistent or dependent.
Reliability : The degree to which the measurement or scale
i.e. consistent or dependent.
Practicality : Measure should be economical convenient and
interpretable.
Errors in : Discrepancies between the obtained scores and
Measurement the corresponding scores.
Scaling : The way of generation of a continuum upon
which measured objects are located.
Comparative Scaling The comparison of objects directly with one
another.
Paired Comparison : A binary choice, respondents choose the objects
in each pair that has the greater magnitude on
the choice dimension.
Rank Order : Respondents are provided with various objects
and asked to rank the list of objects.
Constant Sum : Respondents are asked to allocate a constant
sum of units to a specific set of objects with
respect to predefined criterion.
Non Comparative : Scaling of object independently to some specify
Scaling standard.
Continuous Rating Respondents indicate their rating by marking at
Scale appropriate position on a line.
Itemised Rating Scale Respondents are asked to select the best fitting
category with the object.
Likert Scale : Respondents are asked to select the best fitting
category with the object.
Semantic Differential Seven point scale based on the proposition that
Scale an object can have several implied or
suggestive meanings to an expressed opinion.
Stapel Scale : A single criterion in the centre with 10
categories numbered from -5 to +5 without a
neutral point.
Category Scale : Objects are grouped in to a predetermined
number of categories on the basis of their
perceived strength along certain dimension.

68
 Measurement and
8.10 SOME USEFUL BOOKS/PAPERS Scaling Techniques

Stevens S.S. (1946): On Theory of Scales of Measurement, Science New Series,

Vol. 103, No. 2684, pp 77-680.

Kothari C.R. (2008): Research Methodology Methods and Techniques, New

Age International Publishers. Chapter 5 page 69 to 82.

Babbie Earl, (2010): The Practice of Social Research, 12th Edition,

WodsworthCengage Learning, CA, USA.

Malhotra, N.K. and S. Dash (2009): Marketing Research, An Applied

Orientation, Pearson Education, New Delhi.

Singh, A.K.: Tests Measurements and Research Method in Behavioural

Sciences, Bharti Bhawan Publishers and Distributors, Patna.

Gregory R.J. (2006): Psychological Testing History, Principles and Applications,

Pearson Education, New Delhi, India.

Michael S. Lewis-Beck (ed) (2004): The Sage Encyclopedia of Social Science

Research Methods, Vol. 2, Page no. 161 to 165, Sage Publications, New Delhi,
India.

8.11 ANSWERS OR HINTS TO CHECK YOUR

PROGRESS EXERCISES
Check Your Progress 1

1) See Section 8.2

2) See Section 8.4
3) See Section 8.5
4) See Section 8.3
Check Your Progress 2
1) See Sub-section 8.7.1 and 8.7.2
2) See Sub-section 8.7.2 under the subhead Itemised Rating Scale
3) See Section 8.6.

69