Measurement scales

TYPES OF MEASUREMENT SCALES

Levels of Measurement

 Nominal Scale Named variables

 Ordinal Scale Named + ordered variables

 Interval Scale Named + Ordered + Proportionate interval in between variables

 Ratio Scale Named +Ordered+ Proportionate interval between variables+

Can accommodate absolute zero
Levels of Measurement

 Nominal, Ordinal, Interval and Ratio are defined as the four fundamental levels of
measurement scales , used to capture data in questionnaires and surveys.
 Each scale is an incremental level of measurement , i.e., each scale fulfils the functions of the previous
scale and all survey question scales like likert, semantic, differential, dichotomous etc. are the derivations
of these four fundamental levels of variable measurement.

Nominal scale is a naming scale where variables are simply named or labelled with no specific order
Ordinal scale has all its variables in a specific order beyond just naming them
Interval scale offers labels, order as well as a specific interval between each of its variable options
Ratio scale bears all characteristics of Interval scale in addition to that it can also
accommodate the value of “Zero” on any of its variables
Nominal scale : 1st level of measurement

 Also called as categorical variable scale,

 A scale used for labelling variables into distinct classifications and does not involve a quantitative
value or order.
 It is the simplest of the four variable measurement scales.
 Calculations done on these variables can be futile as there is no numerical value of the options. As
numbers associated with variables only represent tags for categorization or division.
Example : where do you live ?


1. village 2. town 3. city numbers does not represent any value for purpose of calculation or
order, it is used as tags for categorization.
Nominal scale : 1st level of measurement

For instance : a customer survey , which brand of smartphone do you prefer ?

1. apple 2. Samsung 3. one plus 4. xiaomi .
Here only the name of the brands is significant , there is no need for any specific order of these brands . Number
assigned to the respective brands does not have any numerical value
, it just represent as labels for the brands.
When a researcher wants to find out the preferred brand among the four from all the data collected, he uses
the numbers to the respective brands as quantifying instruments and answering the final question.
Like: of total 150 data forms, in simple form
1. got 25 votes , 2. gets 40 votes 3. gets 30 votes 4. gets 55 votes
The answer to the question is 4 is the preferred one i.e., 4. Xiaomi is the preferred one .
Ordinal scale: 2nd level of measurement

 Ordinal scale is defined as a variable measurement scale used to simply depict the order of variables and the
difference between each of the variables

 Ordinal scale is generally used to depict non mathematical ideas like frequency, satisfaction,
happiness, a degree of pain etc.

 Ordinal scale maintains descriptional qualities along with an intrinsic order But is void of an origin of scale
thus the distance between variables can’t be calculated.

 Descriptional qualities indicate tagging as nominal scale with relative position of variables

 Origin of this scale is absent due to which there is no fixed start or true zero.
Ordinal scale: 2nd level of measurement

 Some common examples are status at workplace, order of product quality, order of agreement or
satisfaction.
 The scale is generally used in market research to gather and evaluate relative feedback about product
satisfaction, changing perception with product upgrades etc.
 For example consider a semantic differential scale question. how
satisfied are you with our services ?
1. Very unsatisfied 2. unsatisfied 3. neutral 4. satisfied 5. very satisfied Here the order of
variables is of prime importance and so is labelling.
As 'very unsatisfied’ will always be worse than ‘unsatisfied’
as well as
‘satisfied’ will be worse than ‘very satisfied’ but we don’t know how much.
So the order is relevant to the result and so is naming.
Ordinal scale: 2nd level of measurement

Analysing results based on the order along with naming becomes a convenient process as it can be presented in
tabular or graphical formats.

Methods such as Mann-Whitney U test and Krushal Wallis H test can be use to analyse ordinal data .

These methods are generally implemented to compare two or more ordinal groups.
Other examples of ordinal scales: socio economic status( low income, middle income, high income ),
educational level( +12 level, Graduate, Post Graduate) etc.
Interval scale: 3rd level of measurement

 Interval scale is defined as numerical scale where the order of the variables is known as well as the difference between
the variables.
 The variables which have familiar , constant and computable differences are classified using the interval scale.
Interval indicates distance between two entities.
 This scale is effective as it open doors for the statistical analysis of provided data like using mean, median, mode to
calculate the central tendency. Only drawback is it has no pre decided starting point or true zero value.
Interval scale contains all properties of ordinal scale in addition it offers a calculation of the difference between
 variables. The main characteristics of this scale is the difference between variables is meaningful.
Interval scale is a scale of measurement created by units of equal size., the difference between any two values is
calculated by subtraction. Ratio of values have no meaning because the value of zero ‘0’ is arbitrary.

Examples of interval scale: Celsius/ Fahrenheit temperature scale(50ºC- 30ºC), time scale( 4 pm-8


pm), IQ score scale(75-94,95-114), Credit score scale(550-650,651-750).
 Interval scale: 3rd level of measurement

Following questions fall under this type of scale.

a) what is your family income ? b) what is the temperature in your city.
For example: consider a time scale
When using a 12 hour clock , we compare the time of 4.00 pm in afternoon to 8.00 pm in the evening , it is possible to say that
the difference in the time is 4 hours
On comparable 24 hour format, we compare time of 16.00 in the afternoon to 20.00 in the evening, it is possible to say that
the difference is 4 hours
Now if we look at the ratios of these numbers i.e.,
8/4=2 and 20/16= 1.25 ,
the ratios are different indicating that
these ratios have no meaning.
Look in the figure considering the differences from 0.
 Ratio scale: 4th level of measurement

 Ratio scale is defined as a variable measurement scale that not only produces the order of variables but also
makes the difference between variables known along with information on the value of true zero.
Having the option of the true value zero, the difference between the two variables is the same and there is a
 specific order between the options . So varied inferential and descriptive analysis techniques can be applied to
the variables.
The ratio scale has all the properties of the previous three measurement scales. Best examples

of ratio scale are speed, distance, weight and height.

In market research , use of ratio scale can be

 It provides the most detailed information for calculation like finding central tendency using statistical techniques
such as mean, median, mode and other methods like geometric mean, the coefficient of variation or harmonic
mean etc.
Ratio scale: 4th level of measurement

 Because of existence of true value zero, the ratio scale does not have negative values. Ratio scale
 examples : Following questions fall under this type of scale
a) in what range does your daughter’s current height fall ?
1. less than 5ft 2. between 5ft- 5.5ft 3. between 5.5ft 4. over 6ft
-6ft
b) in what range does your weight (in Kg) fall ?
1. less than 50kg 2. between 50 kg- 70kg 3. between 70kg- 90kg 4. above 90kg In ratio scale ,
ratio remains same if change the measuring parameters like
If measured in kilograms , 60 kg-40kg= 20 kg, ratio, 60/40=1.5
If measured in grams, 60000gms-40000gms= 20000gms, ratio, 60000/40000=1.5 , The
ratio remains same for both measurement formats.
Ratio scale: 4th level of measurement

 Use of Ratio data and analysis,

At the fundamental level, ratio scale data is quantitative in nature due to which all quantitative
analysis techniques such as SWOT, TURF , Cross- tabulation, conjoint etc. can be used.
While SWOT and TURF will analyse ratio data in such a manner that can help create roadmaps of
how to improve products or services and cross tabulation will be useful in understanding whether new
features will be helpful to the target market or not.
 Levels of measurement in nutshell

Offers (variables) Nominal Ordinal Interval Ratio

sequence of variables established - Yes Yes Yes

Mode Yes Yes Yes Yes

Median - Yes Yes Yes

Mean - - Yes Yes

Difference between variables (evaluated) - - Yes Yes

Addition & subtraction of variables - - yes Yes

Multiplication & division of variables - - - Yes

Absolute zero - - - yes