Admissions Preference: To supporters of preferences, these are small, often subjective efforts to give some special consideration to an underrepresented group. Preferences are often described as a “tip” or a “plus factor.”  Opponents of preferences view them as large and often mechanical, and describe them as “race-norming” or adding “hundreds of points” to the scores of favored racial or social groups.

Adverse Impact: The negative effect a policy or action has on an individual or group. This often arises in the context of a claim that the policy or action has what is also described as a “disparate impact” on or “disparate treatment” of the claimant(s). For example, a standardized test might disqualify a disproportionate number of individuals of a given race. Or work requirements, like the strength tests applied to applicants for firefighter positions, would eliminate more women than men. The Supreme Court has held that this impact alone will not sustain a constitutional claim. Rather, the litigant must show that the policy or action was adopted with the intent to cause the effect that it has. This stringent requirement is relaxed somewhat when the lawsuit is predicated on a federal statute like Title VII, allowing the bare impact to make an initial prima face case, albeit still imposing the ultimate burden of persuasion on the plaintiff.

Affirmative Action: A catch-all term that describes various efforts an institution can undertake to become more inclusive.  For colleges and universities, these can range from outreach efforts to special tutoring to differential admissions standards.  The controversial aspect of affirmative actions in higher education centers on the alleged use of large preferences based on race, gender, class, or legacy status. See Debate on Size and Scope of Preferences, Affirmative Action vs. Preferential Treatment, and Affirmative Action Programs

Cascade Effect: If a very elite (“tier one”) college or graduate program uses large preferences to increase its diversity, for example by admitting Hispanics who would, on a race-blind basis, more likely be admitted to a second-tier school, then the tier-one’s program makes it more likely that the tier two school will also have to use preferences and/or use them on a larger scale than they would otherwise.  In other words, in a society where diversity is highly prized by most educational institutions, the actions of elite schools can have large external effects on the options available to other schools.

Critical Mass: While the Supreme Court has consistently held since Bakke that affirmative action policies may not use specific quotas in college admissions and hiring, it has approved policies that pursue a “critical mass” of underrepresented groups to achieve “diversity.”  To dissenting justices, “critical mass” may simply be a quota by another name.  For the majority, it means a significant presence of a minority group, such that individual members of the group do not feel like “tokens” who must act as “spokespersons” for their group.

Disparate Impact: An alternate way of framing the claim that a given policy or action has an “adverse impact” or a disproportionate effect on individuals, in particular groups like African Americans or women.

Disparate Treatment: A decision or action that singles out an individual or group for differentiated treatment. This is another way of saying that the action is “discriminatory,” that is, that a decision has been made to identify and use as a basis for treatment some individual or group characteristic, trait, skill, or the like. Discrimination itself is a value-neutral term and the law of affirmative action singles out for careful examination and potential invalidation those discriminatory actions that are deemed what the law calls “invidious,” that is, motivated by bias, stereotypes, and the like, that have no bearings on individual or group skills and qualifications.

Diversity, Goal of:  Since Justice Powell’s decisive opinion in Bakke, the key constitutional justification for race- and gender-conscious policies by universities has been the educational value of a “diverse” student body.  On the surface, that seems to mean a student body and faculty that reflect a broad array of backgrounds, experiences, and perspectives.  In practice, it often means a student body and faculty in which no racial group or gender is severely underrepresented.

Equal Employment Opportunity (and/or Equal Employment Opportunity Employer): Generally, a phrase that is used describing employers that do not discriminate, as a matter of either policy or procedure, in their employment decisions. This includes both fair and appropriate procedures for hiring and in the employment relationship itself. The phrase signals, for example, that the employer complies with all applicable federal and state statutes and regulations governing the employment process and relationship.

Ethnicity: The idea that a given subgroup of humanity has certain common national or cultural traditions such as nationality, tribal affiliation, religion, and language. While sometimes conflated with race, which focuses on phenotype characteristics such as skin color, ethnicity is a broader concept that has a particular focus on the history, culture, and traditions of a particular group.

Goals: The objectives that an institution pursues. This can include threshold matters like the desire to insure that all applicants for employment or admission are qualified. It might also mean that the college or university, for example, seeks to fashion a student body that is “diverse” and has particular objectives in mind related to that goal that help shape or influence the admissions process and decisions. The goals must be tied to appropriate institutional interests and cannot become rigid “quotas,” i.e. inflexible numerical targets pursued for their own sake.

Good Faith Efforts: Shorthand for a fair, open, and appropriate attempts to achieve institutional goals and to comply with all applicable laws, rules, and regulations, free from from bias and/or invidious discrimination. The phrase signals a desire to act in accordance with principles of fair treatment and the requirements of the law.

Levels of scrutiny – “strict”, “intermediate”, and “rational basis”:  Since the late 1930s, the Supreme Court has evolved three distinct levels of scrutiny that it applies in determining whether classifications used by government (or governmental surrogates) to award benefits or impose penalties, violate the Equal Protection Clause of the 14th Amendment. 

In brief, “suspect” classifications – those that have historically been adopted for predominantly invidious reasons, such as those based on race, ethnicity, or national origin — are subject to “strict” scrutiny, and can only survive if the government defendant has a “compelling” interest and uses the classification in a “narrowly tailored” way.  The term “compelling” has not been defined by the Court, but is generally understood to require a defensible institutional goal of the highest order.  Post-hoc justifications for the policy are not allowed; it must have been adopted for the specific reasons proffered in its defense.

“Intermediate” scrutiny is a step down; the requisite institutional interest must be “important” (once again, undefined by the Court), and the means employed need only be “substantially related” to the attainment of that interest.  Intermediate scrutiny is applied to “quasi-suspect” classifications – gender, most often – that are understood to sometimes be logical bases for distinct policies (e.g., related to pregnancy or medical treatment) but have also been used for invidious reasons.

“Rational basis” review is the lowest level of scrutiny, applied to all other classifications (e.g, applying special rules to gas stations or hair stylists).  It requires only that there be some “reason” for the classification, with courts refraining from judging whether it is good or bad, wise or unwise.  Post-hoc justifications are permitted, and need only show some legitimate rationale supporting the policy.  In the admissions context, rational-basis review would be applied to preferences for legacies or athletes.

National Origin: The division of individuals into distinct groups based on the nation they are from. This includes both the origin nation for immigrants to the United States and, for those who are already citizens the nation from which their ancestor hailed.

Protected Cases: Groups that have been identified for fair, open, and nondiscriminatory treatment. A typical formulation might include “race, ethnicity, national origin, gender, and creed.” As an historical matter, such groups are those that have been singled out for invidious discriminatory treatment in the past and have not been able to secure protection from such treatment through the political process. In an important decision, United States v. Carolene Products Company (1938), the Supreme Court described such groups as “discrete and insular” and in subsequent cases, has characterized them as “suspect” or “quasi-suspect,” i.e. worthy of special protection under the law. These distinctions also play into the decision to use different levels of judicial scrutiny when examining actions that claimants suspect are predicated on bias or stereotypes, rather than individual merit.

Race: The idea that various subgroups of humanity share certain distinctive physical traits, such as the color of their skin, that mark them as a separate group.

Race-Norming:  In the affirmative action context, race-norming is a practice of adjusting credentials to eliminate racial disparities.  For example, suppose that a school uses an entry test to judge applicants; among 100 white applicants, the mean score on the test is 60, and the range is 40 to 80; among 100 Black applicants, the mean score on the test is 55, and the range is 35 to 75.  The school could “race-norm” the scores by adding five points to each Black test result.  If the applicants were otherwise comparable, this race-norming should lead to similar admission rates for whites and Blacks.

Test Bias:  This term is used in two very different and contradictory ways.  One definition, often used by strong advocates of affirmative action, simply considers whether there are systematic differences in test performance across groups (most often, the concern is differentials across racial groups).  The other definition, generally used by psychometricians, considers whether a test inaccurately measures skills in a way that correlates with group membership.  Thus, for example, the Law School Admissions Test clearly produces results that vary widely by race; the mean Black score is about one standard deviation below the mean White score.  But the LSAT is not biased in the psychometric sense; it does not under predict the performance of Blacks in law school.