ABSTRACT
The scientific community investigates phenomena in the world. The areas for scientific inquiry are many and have led to the creation of numerous academic disciplines, e.g., botany, biology, education, psychology, business, music, and so forth. The first step in any academic discipline that conducts scientific investigation is to ask a research question. Research questions can be expressed in many different ways. For example, “In the upcoming election, who will be elected President of the United States?” or “Which is better, margarine or butter, in lowering cholesterol?” The next important step is to design a study, then gather data and test the research question. This requires converting the research question into a statistical hypothesis. There are many different kinds of statistical hypotheses depending upon the level of measurement (nominal, ordinal, interval, or ratio) and type of research design used in the study. A statistical hypothesis is the cornerstone to testing the two possible outcomes, which are always stated in terms of population parameters, given the kind of data collected (percents, ranks, means, or correlation coefficients). The two possible outcomes of a statistical hypothesis are stated in a null (H0: no difference) and alternative (HA: difference exists) format using symbols for the population parameter. The alternative statistical hypothesis is stated to reflect the outcome expected in the research question. This involves either a directional (greater than) or nondirectional (difference exists) expression. The null hypothesis is the corresponding opposite expected outcome of less than/equal or no difference, respectively. A research question and statistical hypothesis for each type of data is listed.
