Statistics key terms

The distance from the centre of a circle to its circumference.

A method for sensitive questions in which a private random event determines whether a respondent gives a prescribed answer or answers honestly, allowing a proportion to be estimated without identifying individual admissions.

A sample selected so that every member of the population has an equal chance of selection.

A numerical quantity whose value depends on the outcome of a random experiment. In a binomial model, $X$ counts successes.

Differences in results arising through chance, even when the probabilities remain unchanged.

The difference between the largest and smallest values in a data set.

The position of a value when a set of values is placed in order.

The difference between the two ranks for one pair of observations: $d=\text{rank of }x-\text{rank of }y$.

The change in a quantity divided by the time taken. Its units describe both the quantity and the time interval, such as people per year.

Data in its originally collected form, before cleaning or further processing.

An original data value before it has been converted or adjusted.

A mistake made when entering or copying a data value, such as recording a height of 175 cm as 1750 cm.

An existing source of information, such as a published report, book, database or website.

A statistically calculated line of best fit, expressed here as $y=a+bx$, used to estimate y from x.

The proportion of trials in which an event occurs, calculated as its frequency divided by the total number of trials. It can be used to estimate the event's probability.

The ratio of the probability of the same event in two groups: $\text{relative risk of A compared with B}=\frac{P(\text{event in A})}{P(\text{event in B})}$.

The extent to which data is suitable for the question being investigated, including the population, variables and time period.

The extent to which repeated measurements under the same or similar conditions yield similar results.

Repeating a statistical study using the same method under similar circumstances to check whether similar results and conclusions are obtained.

A sample that reflects the relevant characteristics of the population.

A person who answers questions in a survey or interview.

The outcome measured to investigate how it varies with the explanatory variable. Also called the dependent variable.

A weighted index measuring changes in the prices of everyday goods and services, including mortgage interest payments.

Replacing a numerical value with a nearby value at a chosen level of precision, such as the nearest second.

A subset of a population selected for a statistical investigation.

The sum of the observed values divided by the number of observations in the sample.

The number of members included in a sample, often written as $n$.

The complete set of possible outcomes of a random experiment.

The collection of data from a subset of a population rather than from every member.

A tendency for a selection method to favour some parts of a population over others, which can make conclusions misleading.