Recent arguments by Dalton, Todor, and Krackhardt (1982) have highlighted the need to distinguish between turnover frequency (i.e., the number of separations) and turnover functionality (i.e., the nature of separations). Turnover functionality, which considers both turnover frequency and the performance level of leaven and stayers, is more critical to organizational effectiveness than is turnover frequency. We test whether work attitudes, widely praised as predictors of turnover frequency, are also useful predictors of turnover functionality. The results of our study, using a sample of 112 retail salespersons, indicate that (a) the traditional measure of turnover frequency overstates the detrimental effects of turnover on organizational effectiveness, in that 53% of the turnover was, in fact, functional, and (b) turnover functionality, which emphasizes the performance levels of stayers and leaven, is unrelated to work attitudes. The practical implications of these results and directions for future research are discussed.Copyright ©; 1986 American Psychological Association. This is a post-print version of an article originally published in Journal of Applied Psychology, 1986 Volume 71, Issue 4. The version of record is available through the American Psychological Association. "This article may not exactly replicate the final version published in the APA journal. It is not the copy of record." 2 During the past 40 years, a substantial amount of evidence in the field of applied psychology has refuted the notion that satisfaction leads to performance (Iaffaldano & Muchinsky, 1985;Lawler & Porter, 1967;Schwab & Cummings, 1970). As a result, applied psychologists have frequently justified the study of work attitudes by claiming that these attitudes are significantly related to turnover. Steers's (1984) statement that "Job attitudes affect organizational effectiveness to the extent they influence turnover" (p. 442) is common among textbook treatments of this issue. Indeed, a large volume of research supports this position (Mobley, Griffeth, Hand, & Meglino, 1979;Steel & Ovalle, 1984).A major assumption in the turnover-attitude literature is that turnover is an inherently bad occurrence and that turnover can be reduced by affecting attitudes toward the job. Hulin (1968), for example, was able to reduce turnover among clerical workers from 30% to 12% by increasing the level of job satisfaction. Recently however, many authors (Abelson & Baysinger, 1984;Dalton, Krackhardt, & Porter, 1981;Dalton & Todor, 1979Dalton, Todor, & Krackhardt, 1982;Mobley, 1982;Muchinsky & Tuttle, 1979;Porter & Steers, 1973;Staw, 1980) have suggested that the traditional treatment of turnover overstates the negative consequences associated with this behavior. Dalton and Tudor (1979), for example, argued that in many cases the individuals who leave an organization are poor performers, and that the separation of these individuals actually provides the organization with an opportunity to replace poor performers with more effective workers. Da...