Patricia A. Taylor

Income Inequality in the Federal Civilian Government

An analysis of income inequality by race and sex within the federal civil service yields three findings of importance for income attainment and human capital research. First, large differences in salary between minority/sex groups remain after occupational stream and a number of employment-related variables are controlled. Second, institutionalized discrimination explains only one-half of these salary differences. Finally, within the federal civil service, the pay structures of minority and nonminority women are more similar to each other than are the pay structures of any other two groups of employees.

The Celebration of Heroes Under Communism: On Honors and the Reproduction of Inequality

Drawing on the work of Goode, Mills, and others, a theoretical synthesis of the place of formal honors in systems of inequality is presented. A multivariate analysis shows that the effects of ascribed characteristics on the receipt of awards operate indirectly, while employment position and professional achievement have different effects for different types of awards. And for the first time for a Communist society, we demonstrate in the analysis with data from Yugoslavia how membership in the Communist party, participation in the National Liberation Front of World War II, and other political status variables affect formal status recognition. All political status variables strongly affect receipt of honors, even with controls for ascribed, positional, and achievement characteristics. The results demonstrate how the elite of a society use formal honors to reproduce a system of inequality.

Confidence, concentration, and competitive performance of elite athletes: a natural experiment in Olympic gymnastics.

During the womens all-around gymnastics final at the 2000 Olympics, the vault was inadvertently set 5 cm too low for a random half of the 36 gymnasts. The error was widely viewed as undermining their confidence and adversely affecting their subsequent performance. This paper examines whether the vault problem had such a carryover effect. Both pretest scores (from preliminary rounds) and posttest scores (from the final) are available on vault, bars, beam. and floor. Manipulation checks establish that the error had experimental impact on vault performance. However, from comparing means, from analysis of covariance, from multiple regression, and from statistically adjusting the official scores, it is clear that the vault error had little if any effect on later performances or on the final standings. Elite athletes in a closed-skill sport apparently learn to concentrate so well that most can recover from a mishap and refocus successfully for the next effort.