hypothesis test


hypothesis

 [hi-poth´ĕ-sis] a supposition that appears to explain a group of phenomena and is advanced as a bases for further investigation.alternative hypothesis the hypothesis that is formulated as an opposite to the null hypothesis in a statistical test.complex hypothesis a prediction of the relationship between two or more independent variables and two or more dependent variables.directional hypothesis a statement of the specific nature (direction) of the relationship between two or more variables.Lyon hypothesis a hypothesis about development of X chromosomes in the embryo; see lyon hypothesis.Monro-Kellie hypothesis [mun-ro´ kel´e] an explanation of the maintenance of intracranial pressure: The skull is viewed as a closed container housing brain tissue, blood, and cerebrospinal fluid; a change in any of these three components will affect the other two. If the volume added to the cranial vault is equal to the volume displaced, the intracranial volume will not change.nondirectional hypothesis a statement that a relationship exists between two variables, without predicting the exact nature (direction) of the relationship.null hypothesis the hyothesis that the effect, relationship, or other manifestation of variables and data under investigation does not exist; an example would be the hypothesis that there is no difference between experimental and control groups in a clinical trial.hypothesis test the abstract procedure that is the theoretical basis of most statistical tests. A hypothesis test decides between two hypotheses, the null hypothesis (H0) that the effect under investigation does not exist and the alternative hypothesis (H1) that some specified effect does exist, based on the observed value of a test statistic whose sampling distribution is completely determined by H0. The decision is made to reject H0 and by implication to accept H1 when the test statistic falls within a given set of values called the critical region. This region is so determined that the probability of rejecting H0 when it is in fact true (a so-called Type I error, the reporting as significant results that are only the result of random variation and not a real effect), is set at a specified level (symbol α). When this level is set before the data are collected, usually at 0.05 or 0.01, it is called the significance level or α level. It is now more common to report the smallest α at which the null hypothesis can be rejected; this is called the significance probability or P value. The ability of the test to accept a true alternative (and thus to detect a real effect when it exists) is termed the power of the test. Note that no statistical test actually tests the H1.

hypothesis test

A statistical procedure to determine the probability (p) that the observed result of a test or trial could have been obtained if the NULL HYPOTHESIS were true.