The effects of violating the normality assumption underlying r

Using randomly generated samples from populations having varying rhos, different shapes of distributions, and varying sample sizes, the effects of violating the population normality assumption underlying r were examined. Results indicate that the normality assumption underlying r is robust (i. e., that violation of the population normality assumption does not seriously alter the interpretation of r). Violation of the population normality assumption appears, therefore, to be insufficient reason to deny r a place as a major tool for sociological analysis. ne of the most widely used statistics in the socialsciences is Pearson’s r. Frequently, however, social scientists face a situation in which their data appear not to meet the normality assumption of r. Nonnormal-looking sample data may reflect a concept that is not normally distributed in the population. It may also reflect a concept that is normally distributed in the population but for which either sampling or measurement error has occurred. The normality assumption applies to the former, not the latter. Therefore, nonnormal sample data may not violate the normality assumption at all. When a concept is not normally distributed in the popula-tion, the researcher may respond in a number of ways. He may ignore the fact that he has violated the normality assumption of r and use it as an estimate of rho, he may attempt to transfor