AQA GCSE Statistics key terms

Data that already exists and is reused in an investigation.

A region of a circle bounded by two radii and the arc between them; a slice of a pie chart.

A systematic tendency for a selection method to favour some population members over others.

The possibility that a topic or question makes participants uncomfortable or reluctant to answer honestly.

A collection of items defined by a particular characteristic or rule.

Shapes with the same shape but possibly different sizes, whose corresponding lengths are all in the same ratio.

A sample selected by chance so that every population member has an equal probability of selection and every possible sample of the required size is equally likely.

A model of a real-world process used to generate data and investigate possible outcomes.

The asymmetry of a data distribution: values extend further on one side than the other.

A measure of the correlation between the rank orders of two variables, commonly written as $r_s$.

A table of cells arranged in rows and columns, which can store data, perform calculations and generate charts.

A measure of how spread out data values are around their mean.

A value calculated by subtracting the data set's mean from a raw value and dividing by its standard deviation. It gives the signed number of standard deviations the value lies from the mean and has no units.

A structured process for a statistical investigation: initial planning, data collection, data processing and presentation, interpretation of results, and evaluation and review.

A display that splits each numerical observation into a stem and a leaf, grouping and ordering the data while retaining the individual values.

Dividing a population into distinct groups before selecting a sample.

One of the groups into which a population is divided during stratification. The plural is strata.

A relationship in which the points closely follow the overall trend. AQA classifies coefficients of $0.6$ or above, or $-0.6$ or below, as strong.

The outcome being counted in each trial. It need not be a desirable result; every other outcome is classified as failure.

A value that summarises an aspect of a dataset, such as its mean, median or interquartile range.

The total obtained by squaring the difference between the two ranks for each pair, then adding these squares: $\sum d^2$.

Deliberately withholding a statistical value from publication, for example to prevent confidential information being revealed.

A distribution whose shape is balanced about its centre, with matching patterns on either side.

A property of a distribution whose two sides are mirror images about its centre.

Selecting members of a population at regular intervals, usually from a sampling frame after a random starting point.

An arrangement of data in labelled rows and columns.

The thinner, extended end of a distribution, where relatively few observations lie.