Machine Learning - The Nature of Probability and Statistics - Measurement Scales

In this video, we are going to discuss measurement scales and there are four different

types we shall discuss one by one. So, there are four different types

The first type is our nominal level of measurement. Classifies data into mutually exclusive, that means non-overlapping and exhausting categories

in which no order or ranking can be imposed on the set of data

So as example, we can go for say political party. So, we are having the different values might be, say, democratic, republican, independent, etc

Maritial status we can go for, say, single, married, divorce, widowed, separated

and so on. So what is happening? Here you see, here these particular categories are non-overlapping

So there is no common in between. So they are non-overlapping and mutually exclusive and exhaustive

That means each and every data must be getting a value from those opportunities

options and there is no ordering or ranking can be imposed on the data next one

we are going for the ordinal level of measurement classifies data into

categories that can be ranked however precise differences between the ranks do

not exist as example you can consider this grade so that is we can have the

values like say a B C D EF so we're having six different grades are there and here

this is a good example of ordinary level of measurement. The next one is our internal level of measurement Ranks data and precise differences between units of measurement measure do exist however there is no meaningful zero so you can

consider this one that is a meaningful difference of one point between IQ

109 and IQ 110 so here you see here we are going for this is this is a good

example of this interval level of measurement so here what is happening it can

thanks data and also it calculates the precise differences between units of measure

So in this case the values are 109 and 110, but here we are calculating that it is there

having one point difference between. The next one is our ratio level of measurement

So possesses all the characteristics of interval measurement and there exists a true zero and

in addition true ratios exist when the same behavior. is measured on two different members of the population

If the things are not getting clear, so let us go for one example

For example, if one person can lift 200 pounds and the another can lift 100 pounds, then

the ratio between them will be 2 is to 1. So that is known as the ratio level of measurement

So the ratio level of measurement here we're going to calculate two different measure values

from two different members of the population and we are trying to calculate the respective

ratio of those values. So in our video we have discussed what are the different measurement skills are possible

and we have explained that one with some examples. Thanks for watching this video