Zero hypothesis refers to the proposition that there is no influence or relationship between phenomena or populations. Any observed difference is due to sampling error (random chance) or experimental error. Empty hypothesis is popular because it can be tested and found to be wrong, which means that there is a relationship between observations. It may be easier to see it as an invalid hypothesis or a hypothesis that researchers are trying to disprove. Another hypothesis, HA or H1, suggests that observations are influenced by non-random factors. In one experiment, the substitution hypothesis showed that experimental variables or independent variables had an effect on dependent variables. It is also well known that H0, the indifference hypothesis has two ways of stating invalid hypothesis. One is to present a statement as a declarative sentence, the other is to present it as a mathematical statement. For example, researchers suspect that exercise is associated with weight loss, assuming that diet remains unchanged. When a person exercises five times a week, the average time to achieve some weight loss effect is six weeks. Researchers wanted to test whether it would take longer to lose weight if the number of exercises per week was reduced to three. The first step in writing null hypotheses is to find (alternative) hypotheses. In short, you’re looking for the results you want with questions like this. In this case, suppose “I expect to lose weight for more than six weeks”. In this case, if weight loss is not achieved in more than six weeks, it must occur in a time equal to or less than six weeks. Another way to state the invalid hypothesis is to make no assumptions about the experimental results. In this case, the invalid hypothesis is that treatment or change has no effect on the experimental results. For this example, reducing the number of exercises does not affect the weight loss time: “ADHD is not related to sugar consumption” is an example of the null hypothesis. If the hypothesis is tested by statistics and found to be wrong, there may be a link between hyperactivity disorder and sugar intake. Significance test is the most common statistical test used to establish trust in null hypothesis. Another example of zero hypothesis is: “Plant growth rate is not affected by the presence of cadmium in soil.” Researchers can test this hypothesis by measuring the growth rate of plants growing in cadmium-deficient media versus the ratio of plants growing in plants. It contains different amounts of cadmium. Verifying the zero hypothesis will lay a foundation for further study of the effects of different concentrations of elements in soil. You may want to know why you test a hypothesis, only to find that it is false. Why not test another hypothesis and find it true? The short answer is that this is part of the scientific approach. In science, “prove” that something will not happen. Science uses mathematics to determine the probability that a statement is true or false. Facts have proved that it is much easier to prove a hypothesis than to prove a hypothesis. In addition, although null hypothesis can be simply stated, alternative hypothesis is likely to be incorrect. For example, if your zero hypothesis is that plant growth is not affected by sunshine time, you can state alternative hypotheses in several different ways. Some statements may be wrong. You can say that plants are damaged by more than 12 hours of sunlight, or that plants need at least 3 hours of sunlight, and so on. There are clear exceptions to these alternative assumptions, so if you test the wrong plant, you may come to the wrong conclusion. Empty hypothesis is a general statement that can be used to develop an alternative hypothesis, which may or may not be correct.