StatisticsDescriptive Statistics

Knowledge guide

Population, Sample, Parameter, Statistic, Variable, and Data

Separate units, measurements, and summaries to identify six basic statistical terms, using a new delivery-duration example.

Six terms at three levels

A population is the full group about which a study asks a question. A sample is the subset actually measured. These describe the observational units, such as orders, households, or people.

A variable is the characteristic recorded for each unit. Data are the individual recorded values of that variable. The variable describes what can be measured; the data say what was recorded.

A parameter summarizes the population. A statistic summarizes the sample. A statistic can be calculated from sample data and used to estimate a parameter, but the two are not interchangeable.

LevelFull group or definitionMeasured subset or values
UnitsPopulationSample
MeasurementVariableData
SummaryParameterStatistic

New example: delivery durations

An analyst wants the mean delivery duration of all parcels a depot dispatched during one specified week. The population is those parcels, and a randomly selected subset is the sample. The variable is elapsed delivery time, measured in hours from dispatch to delivery. Time is quantitative and continuous even when stored with rounding.

For an illustrative five-parcel sample, suppose the recorded durations are 12, 14, 16, 18, and 20 hours. These five values are the data, and their mean is the statistic

x‾=12+14+16+18+205=16 hours.\overline{x}=\frac{12+14+16+18+20}{5}=16\text{ hours}.

The parameter is the mean duration for all parcels in the specified week. Its value is not given. The five example values are constructed for learning and are not real depot records.

Define the question before calculating

Specify the group, the time window, the measurement rule, and whether the target is a mean, proportion, or another summary. Then describe how sampled units were selected. A random sample supports population estimation, but no sampling label guarantees that missing records or inconsistent measurements disappear.

Common confusions

The sample is the set of units, not its mean. One measurement is data, not the population parameter. A sample mean is a statistic even if it happens to equal the population mean. A parameter can be unknown yet fixed for the defined population, while a statistic may change across samples.

Related question

Apply this knowledge

Use the concept guide to understand the reasoning, then return to the complete question and worked answer.

Questions on Statistics: Identify Six Terms in a Recovery-Time Study

Sources

These references support the core concepts and interpretation boundaries explained above.