Statistics key terms

Dividing a population into non-overlapping groups according to one or more characteristics before sampling.

A sampling method in which the population is divided into strata and a random sample is selected from each, with sample numbers proportional to the stratum sizes.

One of the non-overlapping groups into which a population is divided before taking a stratified sample.

A relationship in which the points on a scatter diagram lie close to a straight line.

The outcome category being counted in a probability model. It need not be a desirable outcome.

A number that summarises a feature of a data set, such as its average or spread.

The total obtained by squaring each rank difference and then adding the results, written $\sum d^2$.

A distribution whose shape is a mirror image on either side of its centre.

A sampling method that selects members at regular intervals from an ordered population list, usually after a random starting position.

The thinner, extended part of a distribution towards unusually low or unusually high values.

A table recording observations as marks, usually arranged in groups of five to make counting easier.

The people who will use a representation. Their statistical knowledge and information needs affect which format is appropriate.

The expected value of the statistic being monitored when a process is working correctly.

The group in an experiment that receives the treatment being investigated.

A probability calculated from a mathematical model or assumptions, such as equally likely faces on a fair die.

Observations of a variable recorded at successive times.

The whole number of observations or trials being considered, including those that do and do not have the characteristic of interest.

A transformation that adds the same fixed amount to every data value, including subtracting an amount.

A branching diagram showing possible outcomes of successive stages of a random experiment.

A straight line drawn through the middle of a time series or its moving-average points to represent the general trend.

One repetition of an experiment, such as one toss of a coin.

An axis that displays only part of the numerical range, for example a vertical axis beginning above zero.

A table that classifies each item by two characteristics, using rows for one characteristic and columns for the other.

Data presented as individual values rather than combined into broader categories or class intervals.

The items belonging to either set or both sets.

The value marking the upper quarter of an ordered dataset, with about three quarters of the values below it, written as $\text{UQ}$.

The extent to which a test or measurement measures what it was intended to measure.

A characteristic that can take different values, such as a person's weekly listening time or whether they are a student or teacher.

The square of the standard deviation, written $\sigma^2$. Its units are the square of the original measurement units.

A diagram using circles inside a rectangle to show membership of sets, including their overlaps and outcomes outside the sets.