Use of hypothesis testing can be very useful during decision-making connected with statistical data. A hypothesis is a statement made about a population parameter e.g. a mean and variance of a population. Hypothesis testing is a statistical process, which gives ideas or theories and then determine whether these ideas are true or false. The conclusions in hypothesis testing never 100%, therefore all tested ideas can be only probably true or probably false.

One of the most important concepts in hypothesis testing is sampling distribution. Sampling distribution is a probability distribution of sample statistics based on all possible random samples. We have to choose randomly some amount of samples to conduct testing. The more samples size we take the better our sample curve looks normally distributed. Difference between the sample mean and population mean is a sampling error. The less this error the better result of testing. Usually we take 30 samples, which are enough to draw normally distributed curve.

Typical use scenario below will make clear the real life situation when we may use Hypothesis testing: A bottled water manufacturer states on the product label that each of bottle contains 500 ml of water. We work for the government agency that protects consumers by testing product volumes. We may agree that 500 ml on the bottle is assumed to be true or we may claim it is not true. Carrying hypothesis testing we determine our null and alternative hypothesis. The hull hypothesis is what we expect to happen before we start testing. Usually, it is a statement, that is tested and denoted with “H0”. In hypothesis testing, the null hypothesis considered as true, until we have enough proof to either reject the null hypothesis, or fail to reject the null hypothesis. The alternative hypothesis usually what we do not expect to happen during testing. It is a statement, that derives when the null hypothesis reject. The alternative hypothesis are...

(1)