Variables

Among the first activities in statistical analysis is to count or measure: Counting/measurement theory is concerned with the connection between data and reality. A set of data is a representation (i.e., a model) of the reality based on numerical and measurable scales.

How data is presented and analysed depends on the type of data collected. Certain statistical methods are valid for certain data types only. Any characteristic being measured or observed (e.g. assessment results, choice of canteen meal) is referred to as a variable. Since data collected for statistical purposes can take on a variety or spread of values that are not pre-determined, these variables are termed random variables. E.g. When measuring assessment results, the values of the data can range from 0% to 100%. The random variable under study is Assessment Result. The individual data values are the values between zero and one hundred. They are also referred to as observations.

Data may be classified in different ways according to the nature of the random variable under study. Random variables may be classified as:

Qualitative: which yield non-numeric answers. Examples are eye colour, gender, marital status (i.e. married, divorced, single, widowed, living together) This type of information cannot be manipulated mathematically.

Quantitative: which yield numeric responses that can be mathematically manipulated. Data is measured by Quantitative random variables (e.g. Income, Prices, Assessment results).

Data of random variables can also be classified as:

Discrete Data (or discrete random variables): are data that can assume specific values only, (usually whole numbers). Discrete random variables are characterised by data that is countable. E.g. The number of employees in an organisation.

Continuous Data (or continuous random variables): are data that can assume any numerical value. Continuous random variables are characterised by data that is measurable. E.g. income, age, time to complete a task, the height of a person.