Scales of Measurement

Four scales of measurement are
Nominal,
Ordinal,
Interval,
and
Ratio


1. Nominal

Nominal levels of measurement are used to distinguish between features only on the basis of
qualitative information.

Nominal data does not imply quantitative differences. We can only say that A is different to B.

It is meaningless to add,subtract multiply or divide nominal data. If points on a map were coded
to represent towns, buildings, population clusters or mountains they would represent nominal
data.

Nominal Data Set

A = Russia
B = Spain
C = France
D = Poland


2. Ordinal

Ordinal scales involve differentation by class, but they also differnentiate within a class of features
on the basis of rank according to some qualitative measure.

Only rank is involved in ordinal scales. We are able to say that object A has a higher rank than
object B, but we cannot say by how much.

Examples might be differentiating between major roads and minor roads or small, medium and
large cities.

Ordinal scales inform the map reader that some features are larger or smaller, more important or
less important or younger or older.

Ordinal Data Set
A = Melbourne
B = Sydney
C = Brisbane
D = Perth


3. Interval

Also known as the cardinal level of measurement, interval scales add information about the
distance between ranks.

To employ an interval scale we must use some kind of standard unit. For example, we
differentiate between temperatures by using the standard unit of degrees celcius.

We distinguish among elevations by using the arbitrary datum of mean sea level. We cannot
multiply or divide interval scale data. For example, it would be incorrect to say 40 C = 2 * 20 C.

Interval scales have no true or absolute zero. A temperature of 0 C does not imply an absence of heat, it
is just the point at which water freezes.


4. Ratio

Ratio data is the highest measurement scale. All forms of arithmetic operations can be
meningfully applied to ratio scale data.

Distance, tonnes of wheat, number of people are examples of ratio scale measurements.

Zero has real meaning in ratio level measures. It is not possible to have less than zero people or
to travel less than 0km.

Ratio Data Operation

30 metres x 30 metres = 900 metres