In this video, we'll explore hypothesis testing. Our learning objective is to recall the basic assumptions and concepts related to hypothesis testing. What is a hypothesis? Well, it's an assumption, related to a process or population. Note that, this assumption will relate to a population parameter, not a statistic. This assumption is often communicated as a statement about a particular parameter, such as the, population mean. Hypothesis testing, is a procedure, which uses sample statistics to make inferences, about a population. When we're testing for differences or relationship, we will determine if that difference or relationship is, statistically significant, or not. Statistical significance, refers to the assumption that the, observed difference or association phenomenon represents a significant departure from, what might be expected by chance and chance alone. Statistical hypotheses are generated in pairs, representing, all possible outcomes. The pair is the null hypothesis and the alternative hypothesis. These statistical hypotheses, permit researchers to establish a link, between research questions, and, the appropriate statistical test used to answer those questions. The null hypothesis, is a statement of the status quo. It always contains a statement of equality, and indicates that there is no difference, no effect, or that no relationship exists. All hypotheses statements use, parametric symbols that represent the population parameter of interest. For example, H sub zero, or H sub note, is MU equals 50. Or H sub zero, is variance one, is equivalent to variance two. The alternative research hypothesis, is also called the alternative hypothesis, the research hypothesis carries the burden of proof. In most cases, it is consistent, with what the researcher is expecting to find or support. This is the, statement that represents the decision, if the null hypothesis is rejected. When the null hypothesis is rejected, the alternative is accepted. Examples are that, the alternative hypothesis, H sub one, is that the population mean is not equal to 50, or, that variance, in group one, is not equivalent to variance in group two. We can also have a directional research hypothesis, they state not only that the null hypothesis is not true, but, there's a specific direction involved. Hypotheses can be either directional or non directional. For the directional research hypothesis, the appropriate pair of the null and alternative hypothesis would be that, the status quo is, MU is less than or equal to 50. Alternatively, MU is greater than 50. Note that by definition, the appropriately developed hypotheses represent collectively exhaustive, mutually exclusive, events. Now, for our course, we'll only be looking at, non directional, hypotheses for two groups. Some observations and cautions related to hypothesis testing first, is that, we either accept or reject the null hypothesis. We've never proven, that a difference exists. When accepting the null hypothesis, we're not saying it's true, we're just saying, in essence, that we don't have sufficient statistical evidence, to support, the null or reject the null, respectively. The development of hypotheses take place, prior to, the collection of data, to ensure integrity, not afterwards. The alternative hypothesis will lead to, a two-tailed hypothesis test, or a non directional test. The directional hypothesis leads to a one-tailed, hypothesis test.