Statistics key terms

A number, usually written as $r$, between −1 and +1 that measures the strength and direction of the linear correlation between two variables.

The change in a value expressed as a percentage of its original value: $\frac{\text{new value}-\text{original value}}{\text{original value}}\times100$.

A composite bar chart in which each complete bar represents 100%, so segment sizes show proportions rather than original frequencies.

A unit used to describe the difference between two percentages. A rise from 4% to 7% is an increase of 3 percentage points.

A boundary value in ordered data corresponding to a specified percentage of the observations. For example, the 90th percentile marks the boundary below which approximately 90% of observations lie.

A pattern that repeats at regular intervals, such as the same ordering of roles within successive teams on a list.

A change in apparent size or shape caused by the viewing angle or apparent depth of a representation.

An estimate of population size calculated using $N = \frac{Mn}{m}$, where $$ is the number marked and released, is the second sample size and is the number marked in the second sample.

A representation of frequencies using pictures or symbols, with a key stating the value of one full symbol.

A circle divided into sectors whose angles represent the proportions in different categories.

A small trial survey used to identify and correct problems before the main survey.

An inactive treatment designed to resemble the active treatment, allowing participants’ experiences to be kept similar.

All the individuals in the group being studied, such as all the fish in a particular lake.

A numerical characteristic of a whole population, such as its mean or the proportion of its members with a particular characteristic.

A diagram showing the distribution of a population across age groups, usually with male and female populations represented on opposite sides.

A relationship in which larger values of one variable tend to occur with larger values of the other.

A distribution with a tail towards higher values; values above the median have a greater spread than those below it.

An estimate of an unknown or future result based on a pattern in data.

Preparation of data before calculations or other analysis, including data cleaning.

A small-scale trial of an experiment used to identify problems with its design or procedure before the full experiment.

Data collected by, or specifically for, the person or organisation that will use it.

An action planned or taken before a problem occurs, intended to prevent it or reduce its effects.

A number from 0 to 1 describing how likely an event is. It can be used as the expected proportion of outcomes belonging to that event.

The fraction of a group with a particular characteristic: $\text{proportion}=\frac{\text{number with the characteristic}}{\text{total number in the group}}$.

Non-numerical observations describing qualities or labels, such as eye colour or make of car.

The process of checking that products meet required standards, including using regular samples to monitor a manufacturing process.

Numerical observations or measurements representing quantities, such as the number of pets or a person's height.

A set of questions used to collect data from respondents.

The required number of members to be selected from a particular group in quota sampling.

A sampling method in which specified numbers of members are selected from particular groups, without requiring random selection within each group.