Skip to content
My Glossary Project

One Sided Test

TL;DR

Section titled “TL;DR”
  • Tests a null hypothesis against a directional alternative (e.g., parameter > or < a value).
  • Used when a researcher has a specific hypothesis or prediction to confirm.
  • According to the source, they are less powerful and less flexible than two-sided tests and may have a higher probability of a type II error.

Definition

Section titled “Definition”

A one-sided test, also known as a one-tailed test or directional test, is a statistical hypothesis test in which the null hypothesis is tested against a specific alternative hypothesis that states that the population parameter is greater than, less than, or not equal to a certain value. The direction of the alternative hypothesis (greater than or less than) determines the direction of the test.

Explanation

Section titled “Explanation”

One-sided tests focus on a specific direction of departure from the null hypothesis. They are chosen when the researcher has a clear, directional research hypothesis or wants to confirm a prediction about a population parameter. The alternative hypothesis specifies whether the parameter is expected to be greater than or less than the reference value, which determines which tail of the sampling distribution is used for the test.

The source notes advantages and disadvantages:

  • Advantage: allows testing a specific research hypothesis or confirming a prediction about a population parameter when the direction is known.
  • Disadvantages: they are less powerful than two-sided tests (which test whether the parameter is different from a certain value), may have a higher probability of a type II error, and are less flexible because they only test a specified direction.

Examples

Section titled “Examples”

Drug effectiveness example

Section titled “Drug effectiveness example”

A researcher testing whether a new drug is more effective at reducing blood pressure than a placebo would set up a one-sided test with the null hypothesis that the drug and the placebo are equally effective and the alternative hypothesis that the drug is more effective than the placebo.

Height example

Section titled “Height example”

A researcher wanting to confirm that the average height of adult men in a certain population is greater than 5 feet 10 inches would use a one-sided test with the null hypothesis that the average height is less than or equal to 5 feet 10 inches and the alternative hypothesis that the average height is greater than 5 feet 10 inches.

Use cases

Section titled “Use cases”
  • When a researcher has a specific directional research hypothesis.
  • When a researcher wants to confirm a prediction about a population parameter.

Notes or pitfalls

Section titled “Notes or pitfalls”
  • According to the source, one-sided tests are less powerful than two-sided tests.
  • They may have a higher probability of a type II error (failing to reject a false null hypothesis).
  • They are less flexible than two-sided tests because they only assess a specified direction.
Section titled “Related terms”
  • One-tailed test
  • Directional test
  • Two-sided test
  • Null hypothesis
  • Alternative hypothesis
  • Type II error