Introduction

Much applied research begins with a research hypothesis that states that there is a relationship between two variables or a difference between two parameters, often means. (In later chapters we consider research involving more than two variables.) One typical form of the research hypothesis is that there is a nonzero correlation between the two variables in the population. Often one variable is a categorical independent variable involving group membership (called a grouping variable), such as male-female or Treatment a versus Treatment b, and the other variable is a continuous dependent variable, such as score on an attitude scale or on a test of mental health or achievement. In this case of a grouping variable there are two customary forms of research hypotheses. The hypothesis might again be correlational, positing a nonzero point-biserial correlation between group membership and the dependent variable, as is discussed in chapter 4. More often in this case of a grouping variable the research hypothesis posits that there is a difference between means in the two populations.