Research Methodology

(Scales of Measurement)

Dr Arvinder Kaur
 Three fresh MBAs joined a consulting company. The first assignment given to them was
to design and conduct a study to compare the perception of the patrons of Domino’s
Pizza with Pizza Hut. As the first step, they conducted an exploratory research by
informally talking to the management of both the pizza joints. They also conducted three
focus groups so as to gain insight into what the consumers are actually looking at while
buying pizza. The output of the unstructured interviews and focus groups resulted in
identifying various information needs that could be used in designing the relevant
questionnaire. Some of the relevant information was on gender, age, income, frequency
and occasion of eating pizza, ranking of the attributes that are sought while choosing
pizza joints, and comparative perceptions of Domino’s and Pizza Hut. This information
was to be employed in designing the questionnaire. One question that came into the
minds of the three MBAs was how to measure the attitude and analyse the information
thus obtained from the survey. For this, it was necessary to assign numbers or symbols
to the characteristics of the objects. Assignment of numbers permits a statistical analysis
of the data. The numbers assigned and the subsequent analysis could be different,
depending upon the type of question asked. On one hand, there can be questions used
to measure different psychological aspects such as attitude, perception, image and
preference of people with the help of a certain pre-defined set of stimuli. On the other
hand, there can be questions on gender, marital status, ranking preference for different
flavours, income and age.
 Measurement and Scaling
Measurement means assigning numbers or other symbols to
characteristics of objects according to certain prespecified
rules.
 We measure not the object but some characteristic of it.
Thus, we do not measure consumers, only their
perceptions, attitudes, preferences or other relevant
characteristics.
 . There are two reasons for which numbers are usually assigned.
First of all, numbers permit statistical analysis of the resulting
data and secondly, they facilitate the communication of
measurement results. 11/30/2024
Measurement and Scaling

Scaling involves creating a continuum upon which measured objects

are located.
 Suppose you want to measure the satisfaction level towards Jet-
Airways Airlines and a scale of 1 to 11 is used for the said purpose.
This scale indicates the degree of dissatisfaction, with 1 = extremely
dissatisfied and 11 = extremely satisfied. Measurement is the actual
assignment of a number from 1 to 11 to each respondent whereas
the scaling is the process of placing the respondent on a continuum
with respect to their satisfaction towards Jet Airways.

11/30/2024
Measurement Scales

The most widely used classification of measurement

scales are
(a) Nominal scale,
(b) Ordinal scale,
(c) Interval scale, and
(d) Ratio scale

11/30/2024
 Nominal Scale
 Numbers are assigned for the purpose of identification of the objects. Any object
which is assigned a higher number is in no way superior to the one which is
assigned a lower number.
 In the nominal scale there is a strict one-to-one correspondence between the
numbers and the objects.
 Each number is assigned to only one object and each object has only one number
assigned to it.

11/30/2024
Examples of nominal scale:
 What is your religion?
(a) Hinduism
(b) Sikhism
(c) Christianity
(d) Islam
(e) Any other, (please specify)
A Hindu may be assigned a number 1, a Sikh may be assigned a
number 2, a Christian may be assigned a number 3 and so on.
 Examples of nominal scale:
 Are you married?
(a) Yes
(b) No
 If a person is married, he or she may be assigned a number 101 and an unmarried
person may be assigned a number 102.
 In which of the following departments do you work?
(a) Marketing
(b) HR
(c) Information Technology
(d) Operations
(e) Finance and Accounting
(f ) Any other, (please specify)
 Here also, a person working for the marketing department may be assigned a number 1,
the one working for HR may be assigned a number 2 and so on.
Example

Nominal scale measurements are used for

 Identifying food habits (vegetarian or non-vegetarian),
 Gender (male/female),
 Caste,
 Respondents,
 Brands, Attributes, Stores, the players of a hockey team and
so on.
 Arithmetic Operation for Nominal Scale
 The assigned numbers cannot be added, subtracted, multiplied
or divided. The only arithmetic operations that can be carried
out are the count of each category.
 Therefore, a frequency distribution table can be prepared for the
nominal scale variables and mode of the distribution can be
worked out.
 One can also use chi-square test using nominal scale variables.
 Ex: Use the chi-square test of independence when you
have two nominal variables, each with two or more
possible values. You want to know whether the
proportions for one variable are different among values
of the other variable
 Ordinal scale

 An ordinal scale measurement tells whether an

object has more or less of characteristics than some
other objects.
 However, it cannot answer how much more or how
much less. An ordinal scale tells us the relative
positions of the objects and not the difference
between the magnitudes of the objects.
Example of Ordinal scale

 Suppose Shashi scores the highest marks in

marketing and is ranked no. 1;
 Mohan scores the second highest marks and is
ranked no. 2; and
 Krishna scores third highest marks and is ranked no.
3.
Example of Ordinal scale
 Another example of the ordinal scale could be the CAT score
given in percentile form.
 Suppose a candidate’s score is 95 percentile in the CAT
exam. What it means is that 95 per cent of the candidates that
appeared in the CAT examination have a score below this
candidate, whereas only 5 per cent have scored more than
him.
 The actual score is how much less or more cannot be known
from this statement.
 Examples of Ordinal scale
 Rank the following attributes while choosing a restaurant for dinner.
The most important attribute may be ranked one, the next important
may be assigned a rank of 2 and so on.
Arithmetic Operation for Ordinal Scale

 In the ordinal scale, the assigned ranks cannot be added,

multiplied, subtracted or divided. One can compute median,
percentiles and quartiles of the distribution. The other major
statistical analysis which can be carried out is the rank order
correlation coefficient, sign test.
 The ordinal scale data can be converted into nominal scale
data but not the other way round.
 Interval Scale

 In the interval scale, it is assumed that the respondent is able to answer

the questions on a continuum scale.

 In the interval scale the difference of the score on the scale has
meaningful interpretation.

 It is assumed that the respondent is able to answer the questions on a

continuum scale.
Interval Scale

 The mathematical form of the data on the interval scale may be written
as
Y = a + bX where a ≠ 0

 The interval scale data has an arbitrary origin (non-zero origin). The
most common example of the interval scale data is the relationship
between Celsius and Farenheit temperature. It is known that:
Examples of Interval scale
 Arithmetic Operation for Interval Scale

 The numbers on this scale can be added, subtracted,

multiplied or divided.
 One can compute arithmetic mean, standard deviation,
correlation coefficient and conduct a t-test, Z-test,
regression analysis and factor analysis.
 As the interval scale data can be converted into the
ordinal and the nominal scale data, therefore all the
techniques applicable for the ordinal and the nominal
scale data can also be used for interval scale data.
 Ratio scale

 The ratio scale measurement can be converted into interval, ordinal and
nominal scale. But the other way round is not possible. The mathematical
form of the ratio scale data is given by Y = bX.
 In this case, there is a natural zero (origin), whereas in the interval scale we
had an arbitrary zero. Examples of the ratio scale data are weight, distance
travelled, income and sales of a company, to mention a few.
 Consider the following examples for ratio scale measurements:
1. How many chemist shops are there in your locality?
2. How many students are there in the MBA programme at IIFT?
3. How much distance do you need to travel from your residence to reach the
railway station?
 Examples

 Consider the following examples for ratio scale measurements:

1. How many chemist shops are there in your locality?
2. How many students are there in the MBA programme at IIFT?
3. How much distance do you need to travel from your residence to
reach the railway station?
 Mathematical Operations for Ratio
scale
 All the mathematical operations can be carried
out using the ratio scale data.
 In addition to the statistical analysis mentioned in
the interval, the ordinal and the nominal scale
data, one can compute coefficient of variation,
geometric mean and harmonic mean using the
ratio scale measurement.
Types of scale, characteristics, examples, permissible statistical techniques
Practice Questions
 1. Indicate the type of measurement scale you would use for each of
the following characteristics. Why did you choose the scale you did?
Develop the appropriate question for each characteristic and the
scale chosen.
(a) Colour of a dishwasher
(b) Age of a TV
(c) Occupation
(d) Brand loyalty
(e) Readership of a newspaper
(f) Intention to purchase a TV
 Practice Questions
 Indicate the type of scale (nominal, ordinal, interval or ratio) that is being
used in each of the following questions:
(a) How large is the market size for shampoos?
(b) In which of the following functional areas of management do you wish to specialize in the second
year?
(i) Marketing (ii) Finance (iii) HR (iv) IT
(c) State the order of your preference for the following colours.
(i) Grey (ii) White (iii) Blue (iv) Green (v) Black
(d) Was the research methods course difficult to understand?
Yes_________ No___________
(e) In which month were you born?
(f) How do you rate the quality of food at the Golden Dragon restaurant?
1 = Very poor, 2 = Poor, 3 = Neither good nor poor, 4 = Good, 5 = Very good
TYPES OF SCALING TECHNIQUES
 Comparative Scales

 In a comparative scale, it is assumed that a respondent

makes use of a standard frame of reference before
answering the question.

 A question like ‘How do you rate Barista in comparison to

Cafe Coffee Day on quality of beverages?’ is an example
of the comparative rating scale.

 For example, respondents may be asked whether they

prefer Chinese in comparison to Indian food.
Consider the following set of questions generally used to compare various attributes of Domino’s
Pizza and Pizza Hut.
 Paired Comparison Scales
 Here a respondent is presented with two objects and is asked to select one according to
whatever criterion he or she wants to use. The resulting data from this scale is ordinal in
nature.

 As an example, suppose a parent wants to offer one of the four items to a child—chocolate,
burger, ice cream and pizza. The child is offered to choose one out of the two from the six
possible pairs, i.e., chocolate or burger, chocolate or ice cream, chocolate or pizza,
burger or ice cream, burger or pizza and ice cream or pizza.

 In general, if there are n items, the number of paired comparison would be (n(n – 1)/2).

 Paired comparison technique is useful when the number of items is limited because it
requires a direct comparison and overt choice. In case the number of items to be compared
is large (say 10), it would result in 45 paired comparisons
 Paired Comparison Scales
 Let us assume that there are five brands—A, B, C, D and E—and a paired
comparison with two brands at a time is presented to the respondent with the
option to choose one of them. As there are five brands, it will result in 10 paired
comparisons. Suppose this is administered to a sample of 250 respondents with the
results as presented in Table

 The above table may be interpreted by assuming that the cell entry in the matrix
represents the proportion of respondents who believe that ‘the column brand is
preferred over the row brand’. For example: In brand A versus brand B comparison
it can be said that 60 per cent of the respondents prefer brand B to brand A.
Similarly, 30 per cent of the respondents prefer brand C to brand A and so on.
 Find Ordinal Scale from the above data
 To develop the ordinal scale from the given paired comparison data in the
above table, we can convert the entries in the table to 0 – 1 scores. This is to
show whether the column brand dominates the row brand and vice versa. If the
proportion is greater than 0.5 in the above table, a number of ‘1’ is assigned to
that cell, which means that the column brand is preferred over the row brand.
Whenever the proportion is less than 0.5 in above table, a number of ‘0’ is
assigned to that cell, which means column brand does not dominate the row
brand.
Find Ordinal Scale

 To get the ordinal relationship among the brands, we total the

columns. Here the ordinal scale of brands is D > B > A > C >
E. This means brand D is the most preferred brand, followed
by B, A, C and E.
 Find Interval Scale
 In order to obtain the interval scale data from the paired comparison data, the
entries in the table can be analysed by using a technique called Thurston’s law
of comparative judgement, which converts the ordinal judgements into the
interval data. Here the proportions are assumed as probabilities and using the
assumption of normality, Z-scores can be computed. Z-value has symmetric
distribution with a mean of ‘0’ and variance of ‘1’. If the proportion is less than
0.5, the corresponding Z-value has a negative sign and for the proportion that
is greater than 0.5, the Z-score takes a positive value. The Z-scores for the
paired comparison data is given in Table
 Find Interval Scale

 The entries in Table 7.4 show the distance between two brands.
Assuming that the scores can be added, the total distance is computed.
The average distance is computed by dividing the total score by the
number of brands. This way one obtains the absolute position of each
brand. Now the highest negative values among all the column is added
to each entry corresponding to the average value so that by change of
origin, interval scale values can be obtained. This is shown in the last
row and the values are of interval scale, indicating the difference
between brands. Brand D is the most preferred brand and E is the least
preferred brand and the distance between the two is 0.696. The distance
between brand C and E equals 0.381.
Practice Questions
 Suppose 100 consumers were asked to indicate their preference for five brands of
car tyres, namely Coca Cola, Limca, Pepsi, Thumps up and Sprite. Figures below
indicate the proportion of times the brand mentioned in the column was preferred
over the brand in the row. Compute the distance between the brands and comment
on the results.
BRAND BRAND

Coca cola Limca Pepsi Thumbs up Sprite

Coca Cola 0.50 0.80 0.59 0.52 0.77

Limca 0.20 0.50 0.60 0.46 0.56
Pepsi 0.41 0.40 0.50 0.61 0.60
Thumbs up 0.48 054 0.39 0.50 0.67
Sprite 0.23 0.44 0.40 0.33 0.50
 Practice Questions
 The table below presents a paired comparison data. It states the observed proportion by
stating that brand i (column of the table) is preferred to brand j (row of the table). Use the
data to prepare an ordinal and an interval scale.
Practice Questions
1. Develop a Likert scale to measure the perception of bank customers towards the
concept of Internet banking.
2. Develop a semantic differential scale to measure the image of two coffee joints—
Cafe Coffee Day and Barista.
3. Design a 5-item Likert scale to measure the opinion of the general public for what measures
should be taken to ensure the safety of women in the Indian cities.
4. Assume that a manufacturer of a line of packaged meat products wanted to evaluate consumer
attitudes towards the brand. A panel of 500 regular consumers of the brand responded to a
questionnaire that was sent to them and that included two attitude scales. The questionnaire
produced the following results:
• The average score for the sample on a 25-item Likert scale (five-point) was 105.
• The average score for the sample on a 20-item semantic differential scale (seven-point) was 106.
The vice president has asked you to indicate whether these customers have a favourable or
unfavourable attitude towards the brand. What would you tell him? Please be specific.
 Rank order scaling

 In the rank order scaling, respondents are presented with several

objects simultaneously and asked to order or rank them
according to some criterion.
 Rank the following soft drinks in order of your preference, the
most preferred soft drink should be ranked one, the second most
preferred should be ranked two and so on.
 Constant sum rating scaling
 In constant sum rating scale, the respondents are asked to allocate a total of 100
points between various objects and brands.
 The respondent distributes the points to the various objects in the order of his
preference.
 Allocate a total of 100 points among the various schools into which you would like to
admit your child. The more the points you allocate to a school, more preferred it is to
be considered. The points should be allocated in such a way that the sum total of the
points allocated to various schools adds up to 100.
 Q-sort technique
 The Q-sort technique was developed to discriminate among a large
number of objects quickly. This technique makes use of the rank order
procedure in which objects are sorted into different piles based on their
similarity with respect to certain criterion.
 Suppose there are 100 statements and an individual is asked to pile
them into five groups, in such a way, that the strongly agreed
statements could be put in one pile, agreed statements could be put in
another pile, neutral statements form the third pile, disagreed
statements come in the fourth pile and strongly disagreed statements
form the fifth pile, and so on. The data generated in this way would be
ordinal in nature. The distribution of the number of statement in each
pile should be such that the resulting data may follow a normal
distribution. The number of piles need not be restricted to 5. It could be
as large as 10 or more as the large number increases the reliability or
precision of the results.
 Non-comparative Scales

 In the non-comparative scales, the respondents do not

make use of any frame of reference before answering
the questions. The resulting data is generally
assumed to be interval or ratio scale.
 The respondent may be asked to evaluate the quality
of food in a restaurant on a five point scale (1 = very
poor, 2 = poor and 5 = very good).
 Likert scale

 This is a multiple item agree–disagree five-point scale. The

respondents are given a certain number of items (statements) on
which they are asked to express their degree of
agreement/disagreement.
 This is also called a summated scale because the scores on individual
items can be added together to produce a total score for the
respondent.
 An assumption of the Likert scale is that each of the items
(statements) measures some aspect of a single common factor,
otherwise the scores on the items cannot legitimately be summed up.
 In a typical research study, there are generally 25 to 30 items on a
Likert scale.
Likert scale statements to measure the image of the
company
 Semantic differential scale
 This scale is widely used to compare the images of competing brands,
companies or services.

 Here the respondent is required to rate each attitude or object on a

number of five-or seven-point rating scales.

 This scale is bounded at each end by bipolar adjectives or phrases.

 The difference between Likert and Semantic differential scale is that in

Likert scale, a number of statements (items) are presented to the
respondents to express their degree of agreement/disagreement.
However, in the semantic differential scale, bipolar adjectives or
phrases are used.
Select bipolar adjectives/phrases of semantic differential scale
 Graphic rating scale

 This is a continuous scale, also called graphic rating Scale. In the

graphic rating scale the respondent is asked to tick his preference on a
graph. Consider for example the following question:
 • Please put a tick mark (√) on the following line to indicate your
preference for fast food.

 To measure the preference of an individual towards fast food one has to

measure the distance from the extreme left to the position where a tick
mark has been put. Higher the distance, higher would be the individual
preference for fast food.
 Graphic rating scale

 Another way of presenting the graphic rating scale is through smiling

face scale. The following example would illustrate the same.
 • Please indicate how much do you like fast food by pointing to the face
that best shows your attitude and taste. If you do not prefer it at all, you
would point to face one. In case you prefer it the most, you would point
to face seven.
 Itemized rating scale

 In the itemized rating scale, the respondents are

provided with a scale that has a number of brief
descriptions associated with each of the response
categories. The response categories are ordered in
terms of the scale position and the respondents are
supposed to select the specified category that
describes in the best possible way an object is rated.
Itemized rating scales are widely used in survey
research.
 Stapel scale

 The Stapel scale is used to measure the direction and intensity of an

attitude. At times, it may be difficult to use semantic differential scales
because of the problem in creating bipolar adjectives.
 The Stapel scale overcomes this problem by using only single adjectives.
This scale generally has 10 categories involving numbering –5 to +5
without a neutral point and is usually presented in a vertical form.
 Stapel scale

 They will choose a value of +5 if the restaurant very

accurately describes the attribute and –5 if it does not
describe at all correctly the word in question. Suppose
a respondent has chosen his options as indicated by *.
This shows that the respondent slightly prefers the
quality of food and is of the opinion that the quality of
service is totally useless.
Thank You