What is Hypothesis Testing? Definition and Explanation

Hypothesis Testing is the technique of making an assumption regarding a population parameter. This type of assumption can be true or not. Therefore, the process includes formal operations by the statisticians for accepting or rejecting the hypotheses.hypothesis testing, research, researcher, mathematics

Understanding Statistical Hypotheses

In hypothesis testing, one can find out if a hypothesis is true or not by examining the whole population. This is usually not possible so the researchers mostly study a random sample taken from a particular population. A hypotheses is rejected in the case where there is an inconsistency between the sample data and statistical hypothesis.

Kinds of Statistical Hypotheses

Two basic kinds of statistical hypotheses exist:

Null Hypothesis

Denoted by H0, the null hypothesis is mostly the hypothesis which sample examinations result mainly from chance.

Alternative Hypothesis

Denoted by Ha or H1, the alternative hypothesis is the one which sample examinations are affected by a non-random cause.

Example of Hypothesis Testing

An example of hypothesis testing can be the examination of whether a coin is balanced or fair. In this case the null hypothesis can be that half of the time when the coin flipped it resulted in Tail while the other half resulted in Head. On the other hand, the alternative hypothesis can be that the total number of tails and heads will be distinct from one another. These hypotheses can be expressed symbolically in the following ways:

H0: P = 0.5
Ha: P ≠ 0.5

If for example, the coin was flipped for 50 times and it resulted in 10 tails and 40 heads. With such an outcome, the null hypothesis will be rejected. Therefore, we will state, that on the basis of the evidence, the coin was possible not balanced and fair.

Four Steps of Hypothesis Testing

Following are the four steps which are followed in hypothesis testing:

Stating the Hypotheses

This step includes the stating of alternative and null hypotheses in a way that both of them are mutually exclusive.

Develop Analysis Plan

This step involves the way in which the sample data can be used in order to assess the null hypothesis. The assessment sometimes emphasizes on one test statistic.

Assess Sample Data

Here one has to find the test statistic value which is explained in the analysis plan.

Evaluate the Results

The decision rule mentioned in analysis plan is applied in this stage to reject or accept the null hypothesis.