ABSTRACT
A good descriptive analysis shows the important data in a form such that they will be clearly understood and their meaning grasped by the reader. This chapter deals with situations in which researcher wish to investigate the relationship between two given variables for which the data are already in hand. Often researcher's data measure two (or more) characteristics for each observation, and they wish to determine the existence of an association between the two characteristics. Associations can be of many varieties : positive (direct) or negative (inverse); and straight-line (linear) or curvilinear or oscillating. The chapter considers how to characterize the form of simple associations when it is reasonable to assume that, if there is a relationship, one of the variables causes the other. This is called regression analysis. The best-fitting regression list that minimizes the “sum of squares” can be found by a series of trial-and-error approximations, or commonly by mathematical formula with calculations performed on the computer.
