t-test

t-test

UK[ˈtiː test]US[ˈtiː test]
n

A statistical test used to determine if there is a significant difference between the means of two groups.

Morpheme Breakdown

t
test
t

statistic

test

procedure

Etymology

The term "t-test" is a modern statistical compound. The "t" refers to Student's t-distribution, a probability distribution introduced in 1908 by William Sealy Gosset. Gosset's employer, the Guinness brewery, prohibited employees from publishing, so he used the pen name "Student," leading to the distribution's name. The word "test" has a longer history, originating from Latin testum (an earthen pot), which was used in assaying metals, thus evolving to mean any critical examination or trial. Combined, "t-test" literally means "the test based on the t-statistic," a specific procedure for comparing group means within the framework of Student's distribution.

Analysis

Structure: t (statistic) + test (procedure) t (English, named after the t-distribution developed by William Sealy Gosset, who published under the pseudonym "Student"); test (Latin, from testum meaning "earthen pot," later "trial," functioning here as a noun meaning a procedure for establishing quality or performance).

Examples

The researcher used an independent samples t-test to compare the average scores of the control and experimental groups.

A paired t-test is appropriate when measuring the same subjects before and after an intervention.

The p-value from the t-test indicated that the difference between the two means was statistically significant.