A Different Kinds of data

A.1 Learning Objectives

The properties of your data will dictate how it will be analysed.

  • How you understand a variable dictates what you can do with it.
  • Consider how you will handle variables before doing statistics and data analysis.

A.2 Variable Classes

Independent Variables are64 fixed parameters studied by the experimenter, e.g. genotypes, developmental stages, cell lines, growth conditions, etc.

Dependent Variables are65 what you have measured, and change according to the independent variables, e.g. expression levels, blood pressure, intensity, presence/absence of a substance, etc.

A.3 Data Classes

Categorical variables are66 qualitative, with countable set of possible values that differ in kind, e.g. location, genotype, time interval, etc. There are two scales of categorical variables: nominal and ordinal. They are defined according to the properties in the following table:

Table A.1: Types of categorical scales.
Scale Ordered Example
Nominal ✗ Location, organ
Ordinal ✓ Dose, copy number

When categorical variables dictate how other variables, either categorical or continuous, are to be grouped, they are referred to as factors and their groups as levels.

Continuous Variables are67 quantitative and do not naturally fall into discrete categories, e.g. time, weight, and expression level, although they can be coerced into an ordinal variable.

  1. In plotting, the dependent variable is usually on the Y-axis, the independent variable on the X-axis.↩

  2. i.e. Dependent variables change according to the state of the independent variable.↩

  3. You may also see this referred to as discrete data, which refers to distinct, non-overlapping groups. Other times they are called qualitative, and as we’ll see on page ??, they are also called factors.↩

  4. You may also see this referred to as quantitative data, but we find the phrase continuous more meaningful and less confusing.↩