# What are the differences between a regular hypothesis, null hypothesis, and alternative hypothesis?

##### 1 Answer

The Null and Alternative Hypotheses are statistical terms related to a specific binary test. A “regular” hypothesis may be any supposition or conjecture devised to explain observed phenomena.

#### Explanation:

The critical difference is that the common term “hypothesis” can be a very general opinion in common usage, while the specific null and alternative hypothesis have strict definitions applied to the analysis of statistical data. The “null hypothesis” is a condition that you wish to disprove on the basis of statistical analysis. The “alternative hypothesis” *must* be the opposite of the null hypothesis. If the null and alternative hypotheses are not opposites, you do NOT have a valid statistical test hypothesis pair.

For example, a regular hypothesis may be something like, “Most people like colored chairs.”. A null hypothesis might be, “More than 50% of colored chairs are blue.” The corresponding alternative hypothesis MUST be: “More than 50% of colored chairs are *not* blue.” It cannot be a different statement such as, “More than 50% of colored chairs are green.” It must be the opposite of the null hypothesis. If you want to test a different condition, you need to construct a different null/alternative hypothesis pair.