Measurement & Scaling

Dr. Rohit Vishal Kumar

Associate Professor
International Management Institute
Bhubaneswar – 751003 Odisha

Measurement
Measurement can be described as a way of obtaining symbols to represent the
properties of persons, objects, events or states under study - in which the symbols
have the same relevant relationship to each other as do the things represented

Number Property Under Study

1 Male
2 Female
We could have also assigned
M Male
F Female

But Not
Male
A
Female
A

1:1 CORRESPONDENCE BETWEEN THE NUMBER SYSTEM

AND PROPERTY UNDER STUDY
2

1
 Scaling
 The ability to assign numbers to objects in such a way that:
 Numbers reflect the relationship between the objects with respect to the
characteristics involved
 It allows investigators to make comparison of amount and change in the
property being measured

 Four (4) primary types of scales - Nominal, Ordinal, Interval and

Ratio

 Three (3) important characteristic of real number system are used

to devise the above scales:
Order : numbers are ordered
Distance : differences between numbers are ordered
Origin : series has a unique origin indicated by 0 (zero)

Scales - A Quick Overview

 NOMINAL SCALE
4  Least restrictive. Does not possess order, distance or origin
 Numbers assigned serve only as a label or tags for identifying objects, properties or events
 Example
East : 1 West : 2
North : 3 South : 4
 Permissible mathematical operations: percentage, frequency, mode, contingency
coefficients
 ORDINAL SCALE
 Possess order but not distance or origin
 Numbers assigned preserve the order relationship (rank) and the ability to distinguish
between elements according to a single attribute & element
 Example
Bata : 1st Sree Leathers : 2nd
Khadims: 3rd Titas : 4th
 Permissible mathematical operations: (+) median, percentile, rank correlation, sign test and
run test

2
 Scales - A Quick Overview
 INTERVAL SCALE
 Possess the characteristic of order and distance
 DOES NOT possess origin
 Numbers are assigned in such a way that they preserve both the order and distance but do
not have a unique starting point
 Example: Temperature scale
50o F is twice as warm as 25o F
10o C is not twice as warm as -3.9o C
 Permissible mathematical operations : (+) Mean, average deviation, standard deviation,
correlation, t, F
 RATIO SCALE
 Possess the characteristic of order distance and origin
 Numbers are assigned in such a way that they preserve both the order distance and origin
 Example: length (KM scale), weight (KG scale)
50 KG is twice as heavy as 25 KG
110.24 pound is twice as heavy as 55.12 pound
 Permissible mathematical operations: ALL

Scaling Techniques - Overview

Types of Scales

Based on Data Collection Based on

Techniques Stimulus

Variability Rating Methods Quantitative

Method Judgement
Scales Method
Paired Comparison Direct
Judgement
Ranking Method Fractionalization
Ordered Category Constant Sum
Sorting

3
 Variability Method Scales
1. PAIRED COMPARISON
 Respondent to choose one of the pair of stimulus that “dominates” the other
w.r.t some designated property of interest
 Example:
Compare 6 organizations on “the best place to work”
 6C2 = 15 paired comparison on the comparison grid

1 2 3 4 5 6 2 1 5 6 4 3
1 x 0 1 1 1 1 2 x 1 1 1 1 1
2 1 x 1 1 1 1 1 0 x 1 1 1 1
3 0 0 x 0 0 0 5 0 0 x 1 1 1
4 0 0 1 x 0 0 6 0 0 0 x 1 1
5 0 0 1 1 x 1 4 0 0 0 0 x 1
6 0 0 1 1 0 x 3 0 0 0 0 0 x
 Implicitly assumes (a) transitivity will be maintained (b) respondent has
experience of all the brands on the same attribute

1 of 2

Variability Method Scales

2. RANKING METHOD
 Requires respondent to order stimulus w.r.t Some designated property of study
 Example:
Rank 6 companies on “best place to work”
 Normally the respondent is asked to order K/N i.e.
Rank top 3 objects (=K) out of the 6 objects (=N)
 Implicitly assumes (a) respondent has experience on all the objects on the
same attribute (b) respondents ranking will correctly reflect his preference

3. ORDER CATEGORY SORTING

 Requires respondent to assign objects to ordered categories
 Useful when a large number of objects are to be rated

2 of 2
8

4
 Rating Scales
 One of the most popular & easily applied data collection technique
 The respondent is required to place the product / attribute under study on a
ordered set of categories and thereby assign a “degree of possessed
characteristic” to the attribute under study
 Rating scales can be (a) numerical (b) graphical (c) verbal (d) a mix of all three
 Example
10
Very Good [ ]
Somewhat Good [ ]
Neither Good nor Bad [ ]
Somewhat Bad [ ]
Very Bad [ ]
1

 It assumes (a) items are being capable of being ranked (b) respondent can
psychologically break the ranking into equal intervals (c) scale is ordinal in
nature

Quantitative Judgment Scales

1. DIRECT JUDGMENT SCALE
 An advancement on the rating method scale
 Assumes that the respondent is able to give a numerical rating with each stimulus with
respect to some designated attribute
 The scales used are assumed to be interval or ratio scales
 Is normally of two types
 Limited response category - The respondent is limited to choose between one of the given
categories
 Unlimited response category - The respondent is free to assume the magnitude of scale and
divide it as per his convenience
 Example:

Org. A
|  |
Very Bad Org. A Very Good
| | | | | | | | |
1 of 2
10

5
 Quantitative Judgment Scales
2. FRACTIONALIZATION
 The respondent is asked to give numerical estimates to the attributes under
study relative to a previously exposed attribute
 Example:
Assume that the Work Environment of A is equal to 1.00. Now rate the relative
Work Environment of the following companies with respect to A:
B : 1.50 C : 0.75 D : 2.20 E: 0.50

3. CONSTANT SUM
 The respondent is required to distribute a “number of points - usually 100” over a
set of alternatives such that the numbers distributed reflect the relative
magnitude of importance of alternatives
 Example: Environment : 22
Salary and Perks : 48
Job Security : 30
Openness : 06
TOTAL : 100
2 of 2

11

Scaling Techniques - Overview

Types of Scales

Based on Data Based on

Collection Techniques Stimulus

Subject Centric Stimulus Centric Response Centric

Approach Approach Approach
Cumulative Scales
Summated Scale
Scalogram Analysis
Q-Sort Technique
MDS
Differential Scale

Thurston Case V Semantic Differential Stapel Scale MA Modeling

12

6
 Development of Stimulus Scales
Follow rigorous development procedure
Create
Set up the various stimulus which will act as the parts of the scale
Test
Test the scale to see how the responses are distributed on each stimulus
Normalize
Normalize or Standardize the response categories. Re-Test if necessary
Validate
Check using Factor or Cluster Analysis whether the scales are predicting
correctly
Use
Use it in the actual survey

13

Subject Centric Scales

1. SUMMATED SCALE (LIKERT SCALE)
 Respondent are required to respond to each of the statement in terms of several degrees
of agreement / disagreement
 Each response is given a weight - (not disclosed to the respondent)
 Similar to direct judgment method in look and feel and is useful in judging the degree of
agreement / disagreement
 Example: To identify the outgoing type of personality
Please rate yourself on the following statements
SA A NAND D SD
1. I like playing cricket 5 4 3 2 1
2. I like going to parties 5 4 3 2 1
3. I love reading novels 5 4 3 2 1
4. Enjoy life is my motto 5 4 3 2 1
5. I enjoy working alone 5 4 3 2 1
Item 1, 2, 4 are favorable and carry (+2 +1 0 -1 -2) as weights
Item 3 & 5 are unfavorable and carry (-2 -1 0 +1 +2) as weights
Response set {A, SA, D, SA, SD} gets {+1 +2 +1 +2 +2} = + 8 [outgoing]
Response set {A, N, SA, D SA} gets {+1 +0 -2 -1-2} = - 4 [not outgoing]`

14

7
 Subject Centric Scales
2. Q SORT TECHNIQUE (STEPHENSON SCALE)
 Respondent are required to sort a set number of statements in predetermined categories
(usually 3 / 5 / 7 / 11) - with the restriction that at least ‘k’ statement should be placed in
each category
 Each category is given a weight and then these weight are used to determine the
subject’s attitude towards the attitude under study
 Normally used as a precursor to factor / cluster analysis

3. DIFFERENTIAL SCALE (THURSTON SCALE)

 A modification of the Q-Sort Technique
 It assumes that the respondent will agree with a subset of the statements - this
agreement in turn revealing the preference of the consumer
 The development of the statements for the purpose of the study is done using Delphi
Technique

Most Agreed with Neutral Least Agreed with

(Two Items) | (Three Items) | (Two Items)
(+1) | (0) | (-1)

15

Stimulus Centric Scales

1. SEMANTIC DIFFERENTIAL SCALE

 SEMANTIC : relating to the study of meaning and the change in meaning

 This scale uses “SEMANTIC” to understand the respondent’s “interpretation of
meaning”
 It allows the researcher to probe both the direction and intensity of respondents
attitudes using interval scaled data
 Mainly used in image mapping studies

 Example: understanding the corporate image of BATA

Powerful _X_ | ___ | ___ | ___ |___ | ___ | ___ Weak
Modern ___ | ___ | ___ | ___ |_X_ | ___ | ___ Old fashioned
Warm ___ | ___ | _X_ | ___ |___ | ___ | ___ Cold
Reliable ___ | ___ | ___ | ___ |___ | _X_ | ___ Unreliable
Careful ___ | ___ | ___ | _X_ |___ | ___ | ___ Careless

 Semantic differential requires extensive pre-testing before it can be put into actual
research.

16

8
 Stimulus Centric Scales
2. STAPEL SCALE
 A modification of the semantic differential scale
 Is an even numbered non-verbal rating scale used in conjunction with a single
adjective
 Measure both intensity and direction of response
 Example: how would you rate BATA Stores on “cleanliness”
cleanliness

-3 -2 -1 +1 +2 +3

3. MULTI-ATTRIBUTE MODELING
 Proposed by Martin Fishbein in 1967
 Uses mathematical model (usually linear model) to interpret a persons
attitude on a particular aspect
𝑨𝒐 = 𝒃 𝒊 × 𝒘𝒊

Where:
AO is the respondent’s overall attitude towards some object
BI is the respondents strength of belief on an attribute
wI is the weight associated with the strength of belief

17

Response Centric Scales

CUMULATIVE SCALES
 Consist of a set of items on which the respondent indicates agreement /
disagreement
 Based on the pattern of response - respondent preferences are ascertained

SCALOGRAM ANALYSIS
 Developed by Louis Guttman in 1958
 Builds on the cumulative scale and tries to develop a pattern of “pre-
determined responses” by scaling both respondent and responses

MULTIDIMENSIONAL SCALING
 An advancement over Cumulative and Scalogram Analysis.
 Tries to determine consumer preferences on more than one dimension
simultaneously
 Extremely difficult to develop administer and interpret

18

9
 Limitations of Scaling Procedure
 Most scales measure attitudes along a single dimension
 Human beings are more complex and are normally exposed to more than one stimuli -
product features, price, package design, advertising, brand name etc
 Scales fail to measure the extraneous influences
 Purchase decisions may be made because of pressure from boss etc. Under such
issues - and especially in areas on high involvement goods - scales and measurement
may fail completely
 It is difficult to develop “useable measures” from scales
 For example, question on “intention to buy” may not be indicative of market share in
the next 6 months
 There still exist a divergence between “what scales can capture” and “what market
research can deliver”

19

20
THANK YOU

20

10