Rational for this module
Every day, each of us make hundreds of decisions, some small, some very large. Students, as part of their academic and social growth, need to be taught structures and methodologies that will enable them to improve their decision-making capabilities.
To make good decisions, we need to have the necessary data/facts and analyze them for the best choice. This is what good science does. First, we observe phenomenon and then utilize hypothesis testing on what we perceive is reality. Thus, we have the null (H 0 ) hypothesis of no difference or equality and the alternate hypothesis (H 1 ) our view of reality, i.e. H 0 : : 1 = : 2 and H 1 : : : 1 ≠ : 2 or you can use , or > . (To make this a bit more understandable, you might want to use an example in words. For instance, : 1 = average salaries of males, and : 2 = the average salaries of females. Our null hypothesis states that the average salaries of males equal the average salary of females. Our alternative hypothesis states that the average sales of males and females are not equal.) An appropriate set of statistics is utilized so we may determine whether the null hypothesis can be rejected. We do not "accept" the alternate hypothesis because we never know all the variables and facts. If, however, we can reject the null hypotheses, we, by default, accept that the alternate hypothesis is true.
While regression analysis does not necessarily prove causality, it is a pivotal statistical methodology for looking at two or more independent variables simultaneously and showing their relationship (strength, direction and interplay) to our phenomena of interest, the dependent variable. Regression Analysis is more logic than math. It is about understanding relationships and not just a rote set of algorithms with a magical answer that tells you what to do at the end of the arithmetic More often, the first level of analysis requires more layers before a truly insightful analysis can be ascertained.
Comments: