Qualitative data are labels that classify objects.

Qualitative data has a finite number of labels.

Each label is called level.

Qualitative data are associated with two scales of measurement:

• Nominal: when they have no order
• Ordinal: when they can be placed in order

Nominal data are also called categorical.

Colour names (red, green, orange, blue) can be used to label objects (e.g. the red pencil and the blue pencil).

When used like this they don't have any order. These data are therefore qualitative nominal.

No rain, light rain, heavy rain, torrential rain are four types of rainfall.

These labels convey no quantitative information, but they can be placed in order (no rain is less than light rain which is less than heavy rain which is less than torrential rain).

When described like this, these rainfall data is therefore a qualitative ordinal.

Questionnaires often use a Likert scale (e.g. strongly agree, agree, neutral, disagree, strongly disagree)

The levels on a Likert scale are labels that describe a person's level of agreement with a statement. Data from a Likert scale can be ordered but contain no quantitative information, so they are an example of a qualitative ordinal data.

A number does not imply quantitative data

In a marathon, each runner is assigned an individual number. These numbers can be used to label a runner. The number has no numerical information attached to it. If the numbers are assigned at random then the numbers have no logical order to them.

These data are therefore qualitative nominal data.

Quanitative data are measurements (i.e. numbers where the number can be placed on a scale).

There are two types of quantitative data

• Continuous: the measurement scale has no gaps (e.g. height is measured in metres)
• Discrete: the measurement scale has gaps between values (e.g. counting objects only uses whole numbers, and not the fractional parts on the number scale in between whole numbers)

Rainfall can be measured in mm of rain per day. When measured like this rainfall is a quantitative, continuous variable.

Compare this to the example of rainfall as a qualitative, ordinal variable.

Measuring a person's height is an example of quantitative, continuous data.

These data are continuous because there is no reason to expect any gaps in the range of possible heights.

Colour can be measured as the wavelength of light from an object.

When measured in this way colour is quantitative, continuous data.

Compare this to the example of colour as qualitative nominal data.

The number of species of flowering plant in a meadow.

These data will be whole numbers because a species is indivisible.

So these data are quantitative, discrete.

What is the data type for each of the following?

1. The body mass of a shrew measured in grams
2. The number of plant species in a 2 m x 2 m quadrat
3. The relative abundance of plant species in a 2 m x 2 m quadrat
4. Land-use types (e.g. urban, arable, pasture)
5. The genes in the genome of an individual
6. The presence (or absence) of a parasite in a blood sample

Here are some hints to help identify the data types:

1. Weight, measured in grams, is written as a number.
2. This count will only include whole numbers.
3. The relative abundance can be any number from zero to one
4. Land-use types categorise land-use
5. Each known gene in a genome is labelled with a name
6. Presence/absence data is called binary data (it can take only two possible values).

What is the data type for each of the following?

1. The body mass of a shrew measured in grams
   is quantitative continuous
2. The number of plant species in a 2 m x 2 m quadrat
   is quantitative discrete
3. The relative abundance of plant species in a 2 m x 2 m quadrat
   is quantitative continuous
4. Land-use types (e.g. urban, arable, pasture)
   are qualitative nominal
5. The genes in the genome of an individual
   are qualitative nominal
6. The presence (or absence) of a parasite in a blood sample
   is qualitative (this case is sometimescan be given its own data type: binary data)