MEASUREMENT AND SCALING
Presented by :
• Measurement can be defined as a process of associating
numbers to observations obtained in a research study.
• The variables associated with a study are classified into two
a) Quantitative/ Numeric
b) Qualitative / Categorical
Incidentally, only quantitative variables can be measured
with the help of standard counting devices and qualitative
variables can only be observed , there is no standard device
or instrument to measure them.
For example, in case of human beings, there are certain
Quantitative( physical) characteristics like height,
weight etc and there are certain qualitative ( abstract)
characteristics like beauty, attitude, creativity etc.
Like human beings, a business organization has also some
Physical characteristics like employees, sales, offices
etc. Being physical in nature these are easily measurable.
However, there are certain abstract characteristics like
reputation of the employees, image of the
entity, motivation, work culture, commitment,
trust, customer’s perception, feelings of customers.
All these are extremely important because they help the
company to stay afloat and grow.
• Therefore characteristics have to be measured for their
meaningful assessment .This can be done by assigning some
numbers and forming scales.
CLASSIFICATION OR TYPES OF
• All measurement scales can be classified into the following
PROPERTIES OF SCALES
• Distinctive classification
• Equal distance
• Fixed origin
A measure that can be used to classify objects or their
characteristics into distinctive classes /categories is said
to have this property. For example: gender classifies the
individuals into distinctive groups, males and females.
The individuals may also be classified on the basis of their
Occupation, like student, salaried, businessman etc.
Similarly, the qualification of an individual could be used
to classify individuals into various categories such as
undergraduate, postgraduate, professional etc.
Similarly, we can classify a person based on marital
status like married, unmarried, widowed, divorced.
Categorical data is qualitative or descriptive data, which
can be made into numerical data if we code the various
For example if we record marital status as 1. married
2. Unmarried 3. widowed and 4. divorced.
Nominal data are numerical data for namesake only,
because they do not share any properties of the numbers
we deal in ordinary mathematics. For instance we cannot
write 4> 3 or 1<2.
ORDER (ORDINAL DATA)
A measure is said to have an order if the objects or their
characteristics can be arranged in a meaningful order. For
example, marks of a student (Quantitative data) can be
arranged in an ascending or descending order. As another
example, a consumer may asked to rank four telecom
service providers ( say A, B, C and D) on the basis of the
connectivity.( Qualitative data)
EQUAL DISTANCE (INTERVAL)
If for a measure the difference between any two
consecutive categories of a measured attribute are equal
then the measure is said to have equal distance.
For example, in temperature readings the difference
between 400 C and 500 C is same as between 600 C and 700 C.
Similarly the Time measurement also follow the same
A measurement scale is said to have a fixed origin if there
is a meaningful zero or absence of the characteristics.
Examples are income of an individual, sales of a
company, Profit of a company. etc.
Zero income signifies absence of income, Zero sales
signifies absence of sales
TYPES OF SCALES
Nominal Scale : This scale is used to divide the population
into various subgroups/categories or classes. It do not
satisfy the other three properties mentioned above.
It is termed as ‘nominal’, as though one may represent the
categories using numbers , the numbers are just for
namesake, they do not carry any value or order or
Example: If we put up a question like ‘which type of
vehicle is used for going to office ?
The answer could be bus, car, motor cycle, auto etc.
Numerical value can be assigned to classify these
categories like 1,2, 3, 4. Sometimes codes are used for
classification like STD codes for cities, codes for various
subjects in a university etc.
The data collected through a nominal scale is known as a
2. ORDINAL SCALE
A qualitative scale with order is called an ordinal scale.
This scale possesses first two of the four properties of
the scale , i.e. the properties of distinctive classification as
well as order or rank like 1st , 2nd, 3rd etc.
The ordinal scale places events in order, but there is no
attempt to make the intervals of the scale equal.
For example, if in a class of students , the highest mark is
95 , next is 85 and the next is 84, converting marks to
ranks will lead to 1,2, and 3. Incidentally, it may be noted
that the difference in the performance of the 1st ranker and
2nd ranker is not the same as the 2nd ranker and 3rd ranker.
Thus, one can only conclude that 1st ranker has performed better
than 2nd ranker and 2nd ranker better than 3rd ranker.
• The data obtained using ordinal scale is termed as ordinal
Some examples are:
Ratings of hotels, restaurants and movies. We can say 5 star
hotel is better than a 4 star hotel, but we cannot say that a 4
star hotel is twice good as a 2 star hotel.
Class of travel in a train or an aero plane.
grades of students in a class.
3. INTERVAL SCALE
A measurement scale whose successive values represent
equal value of the characteristics that is being measured,
and whose base value is not fixed, is called an interval
This is a quantitative scale of measure without a fixed
or true zero.
Some examples are temperature( Fahrenheit scale) , time,
longitude, latitude etc.
4. RATIO SCALE
Ratio scales are quantitative measures with fixed or true
Ratio scales has all four properties of scales that are
For example, a weighing scale is a ratio scale. Some other
examples are height, price, sales, revenue, profit etc. In all
these cases zero implies absence of that characteristic.