jroese@lssu.edu
GLOSSARY
As with any new topic, knowledge of the language of statistics is a prerequisite to understanding the concepts. Many of the
terms used by statisticians are commonly used words with rather imprecise definitions (e.g. population, sample, mean).
However, in a statistical context these words take on very specific meanings. The proper use if these terms is an essential
element in communicating statistical information.
Confidence interval
A range of values that, with a measurable degree of certainty, contain a population parameter.
Continuous
Interval data in which observations can assume an infinite number of values (i.e. decimal values).
Data
The set of observations of a population or sample.
Degrees of freedom
A measure of the amount of information available from a sample or samples.
Discrete
Any scale of measurement in which observations are restricted to a finite set of mutually exclusive
categories or integer values.
Homoscedastic
A term referring to the similarity of variances between two or more populations of interest.
Independent
A term used to indicate the lack of influence the selection of one observation has on the selection of
other observations.
Interval
A scale of measurement in which observations are measured on a proportional basis (e.g. number
of individuals, weight, length).
Mean
A measure of central tendency among the observations of a population or sample; the arithmetic
average of the data.
Median
A measure of central tendency among the observations of a population or sample; the middle value
in an ordered data set.
Mode
A measure of central tendency among the observations of a population or sample; the most
frequently occurring value(s) in the data.
Nominal
A scale of measurement in which observations are classified into mutually exclusive categories
which lack any ranking or proportional relationship (e.g. male and female; herbivore, carnivore,
omnivore).
Observation
The value of a variable for an individual sampling unit.
Ordinal
A scale of measurement in which observations are classified into mutually exclusive categories
which can be ranked, but which lack a proportional relationship (e.g. slow, and fast; small, medium,
large).
Parameter
A quantity that summarizes or characterizes a population.
Parametric
Refers to statistical procedures that depend on the estimation of parameters.
Population
All objects of a similar type, or some designated subset of these objects (e.g. all white-tailed deer in
Michigan).
Sample
A subset of a population, usually understood to be randomly selected.
Sampling Unit
The objects which comprise a population or sample. These are the subject upon which
measurements are made (e.g. a deer).
Standard deviation
A measure of dispersion of the observations around the population or sample mean; the square root
of the variance.
Standard error
A measure of dispersion of a group of sample parameters around a population (e.g. standard error
of the mean).
Statistic
A quantity that summarizes or characterizes a sample.
Variable
A measurable characteristic of the sampling units that may assume any one of a set of values (e.g.
the weight of each deer).
Variance
A measure of dispersion of the observations around the population or sample mean; the sum of the
squared differences between each observation and the mean.
Variation
Differences, naturally occurring or designed, between observations.