ABSTRACT
This chapter discusses the selection and application of appropriate statistical methods to answer basic associational research questions. These statistical methods are usually used in single-factor designs with many ordered levels of the independent variable. The most common correlation coefficient used to describe the relationships between two continuous variables is the Pearson product-moment correlation, which is represented by r. When one or both of the variables is not normally distributed or there are violations precluding the use of parametric statistics, the Spearman correlation or the Kendall tau (t) is usually used. The chapter examines two very commonly used statistical tests: the Pearson product-moment correlation coefficient, r, and the chi-square (x2) test for independence. It discusses common problems associated with the reporting and interpretation of correlation coefficients and the need to present effect sizes and confidence intervals. The chapter also discusses the special case of the chi-square test with one degree of freedom because of its prominent use in medical research.
