Statistical Evidence in Discrimination Cases

Gastwirth, Joseph L.

doi:10.1111/1467-985x.00063

Cited by 17 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This early empirical work played a minor role in cases brought to court based on the Equal Pay Act of 1963 and the Civil Rights Act of 1964 but, according to Ashenfelter and Oaxaca (1987), over time became the primary form of analysis for decisions of disparate impact against population subgroups. The U.S. Supreme Court has heard many cases in which statistical evidence of disparate impact discrimination has been admitted in cases involving jury selection and employment discrimination, focusing on factors such as gender, race, and age (see, e.g., Ashenfelter & Oaxaca, 1987;Castaneda v. Partida, 1977;Gunderson, 1989;Hazelwood School District v. United States, 1977;Gastwirth, 1984Gastwirth, , 1997Kaye, 1982;Lee, 1989: Shafritz, Hyde, & Rosenbloom, 1986.…”

Section: The Concept Of Disparate Impactmentioning

confidence: 99%

School Finance Court Cases and Disparate Racial Impact

Stiefel¹,

Schwartz

Berne

et al. 2005

Education and Urban Society

View full text Add to dashboard Cite

Although analyses of state school finance systems rarely focus on the distribution of funds to students of different races, the advent of racial discrimination as an issue in school finance court cases may change that situation. In this article, we describe the background, analyses, and results of plaintiffs' testimony regarding racial discrimination in Campaign for Fiscal Equity Inc. v. State of New York. Plaintiffs employed multiple regression and public finance literature to show that New York State's school finance system had a disparate racial impact on New York City students. We review the legal basis for disparate racial impact claims, with particular emphasis on the role of quantitative statistical work, and then describe the model we developed and estimated for the court case. Finally, we discuss the defendants'rebuttal, the Court's decision, and conclude with observations about the role of analysis in judicial decision making in school finance.In the United States, each state is responsible for public elementary and secondary education services within its boundaries. Almost all states have chosen to set up local school districts to administer and deliver education, with the property tax as the primary local revenue source. These arrangements have created a nation of 50 school finance systems with substantial funding disparities across and within states. Plaintiffs in 45 states have challenged these finance systems in their state courts and, in almost one half the states, AUTHORS' NOTE: We thank Hella Bel Hadj Amor for assistance with some of the quantitative work and anonymous referees for helpful comments. Final responsibility for content lies with the authors. Stiefel, Schwartz, and Berne were consultants for the plaintiffs, and Berne testified for the plaintiffs in Campaign for Fiscal Equity Inc. v. State of New York (2001).

show abstract

Section: The Concept Of Disparate Impactmentioning

confidence: 99%

School Finance Court Cases and Disparate Racial Impact

Stiefel¹,

Schwartz

Berne

et al. 2005

Education and Urban Society

View full text Add to dashboard Cite

show abstract

“…Ignoring the grand jurors carried over for a second year, however, would allow a county to intentionally discriminate against blacks in the carry over process, while appointing the minimal number of blacks in the new grand jurors needed to "pass" the standard binomial test. This is a general problem occurring when one is concerned with twostage employment practices, as shown in Gastwirth (1997) who studied methods for assessing the fairness of both hiring and promotion practices.…”

Section: Number Of Blacks On Grand Juries Of 18 In Colleton Countymentioning

confidence: 99%

The Effect of Dependence on Confidence Intervals for a Population Proportion

Miao

Gastwirth

2004

The American Statistician

View full text Add to dashboard Cite

The binomial model is widely used in statistical applications. Usually, the success probability, p, and its associated con dence interval are estimated from a random sample. Thus, the observations are independent and identically distributed. Motivated by a legal case where some grand jurors could serve a second year, this article shows that when the observationsare dependent,even slightly, the coverage probabilities of the usual con dence intervals can deviate noticeably from their nominal level. Several modi ed con dence intervals that incorporate the dependence structure are proposed and examined. Our results show that the modi ed Wilson, Agresti-Coull, and Jeffreys con dence intervals perform well and can be recommended for general use.

show abstract

“…Thus, different statistical methods will be preferred depending on the nature of the legal complaint. Gastwirth and Greenhouse (1995) and Gastwirth (1997) pointed out that the available sample size for studying the promotion process depends on the number of those who were hired. They showed that anomalous results can occur when one analyzes the data for the two processes separately.…”

Section: Introductionmentioning

confidence: 99%