Employment practices are biased or unfair when employment decisions (e.g., hiring, promotion, job transfer) are based on factors (e.g., race, gender, age, disability, religion, and physical attractiveness) that are unrelated to a person’s ability to succeed on the job. Biased employment practices are a very important issue, because selection systems control access to jobs and such other valued outcomes as opportunities for advancement and training.
Biased employment practices result in unfair discrimination in that individuals who have equal probabilities of succeeding on a job have unequal probabilities of being selected for it. Viewed in terms of selection decision errors, unfair discrimination results in a disproportionate number of individuals who are the targets of bias being classified as false negatives. These are individuals who would have succeeded on the job but were not hired.
Unfair discrimination is always a concern in cases of disparate treatment, that is, the deliberate exclusion of individuals from jobs on the basis of their membership in targeted groups (e.g., Blacks, women). In such cases, there are overt and intentional expressions of bias or prejudice against job applicants. However, unfair discrimination may also occur in instances of adverse impact. In such instances, the hiring rates for members of groups differ from one another. Adverse impact is an operational, legal, and ethical concern when group differences in hiring rates are a function of factors that are unrelated to the job. Note, however, that adverse impact is not always a sign of unfair discrimination.
The employment process in the typical organization involves the five sequential steps that are listed below. Biases can arise in one or more of them.
- Job analysis phase. A job analysis is performed to identify task requirements and the knowledge, skills, abilities, and other requirements (KSAOs) of job incumbents. Although its results need only reflect minimally acceptable KSAO levels, the results often serve as a basis for specifying the KSAOs of the ideal incumbent. This can introduce bias, because conceptions of the ideal incumbent are typically affected by the beliefs of dominant groups in organizations. As a result, subsequent phases of the selection process will be affected by the biases introduced at the job analysis phase.
- Recruitment phase. A variety of strategies are used in the recruitment phase to advertise the job opening and attract potential applicants, including listings in newspapers, trade publications, and online recruiting. Bias can be an issue when such strategies leave potential applicants unaware of job openings, precluding them from applying. One way in which this may occur is through stereotypes of ideal job incumbents influencing where organizations search for applicants. For instance, such stereotypes may lead them to recruit at private universities that are dominated by relatively affluent White students rather than universities with ethnically and socioeconomically diverse students. Bias can also be a problem if an organization’s reputation and past employment practices (e.g., a history of unfair discrimination against women and ethnic minorities) have a chilling effect on potential applicants, reducing their motivation to apply for jobs. Finally, bias can arise if recruiting strategies or materials lead potential applicants to believe that they will not fit in the organization.
- Job application phase. At the application phase, job applicants submit such materials as resumes and applications. Employers screen these materials to make initial judgments about applicant suitability. This phase is particularly susceptible to biases, because there is only a limited amount of information about applicants, and decision makers may base their suitability judgments on stereotypes about and/or affective reactions to members of various groups. This may lead to negative beliefs and feelings about applicants who are the targets of bias.
- Testing phase. In the testing phase, the applicant completes measures of predictors (e.g., cognitive ability tests, interviews, personality inventories, work samples) designed to assess his or her KSAOs. A key issue with such measures is their criterion-related validity. This is typically indexed by the degree of correlation between (a) the predictors and (b) job performance or other important criteria (e.g., absenteeism, turnover). Regrettably, criterion-related evidence does not guarantee that predictor measures lack bias. This is suggested by strong evidence of the adverse impact of many types of predictors (e.g., cognitive ability measures) against various groups. For example, research has shown that Blacks and Hispanics score lower on cognitive ability tests than Whites. Bias would be an issue if the test performance of targets of bias were an artifact of cultural biases in such tests rather than actual differences in their cognitive ability. Unfortunately, the existing evidence does not allow for well-founded conclusions about the degree to which cultural biases in tests are responsible for adverse impact. However, inferences about cultural bias are much clearer for certain types of predictors (e.g., personality). For instance, research shows that individuals from collective cultures (e.g., Asians, American Indians) have different personalities (e.g., as reflected in measures of values) than individuals from individualistic cultures (e.g., Anglo-Americans). In addition, as a result of differential exposure to stressors, Blacks have higher levels of neuroticism than Whites. Research also demonstrates that predictor measures that are based on subjective assessments of the attributes of applicants are highly subject to bias. For example, unstructured interviews are prone to such problems as first impression effects, contrast errors, and similarity errors. In addition, subjective assessments can be biased by an observer’s stereotypes (e.g., based on the applicant’s race, sex, age, disability, unattractiveness). Thus it may be better to base inferences about job suitability for many types of jobs on measures such as work samples than on measures that have high levels of subjectivity.
- Decision phase. Using predictor information, the decision maker makes judgments about the relative suitability of applicants and then selects one or more individuals to be the recipients of a job offer. Although selection procedures are typically top down, they may rely on banding. In the case of top-down selection, individuals are rank-ordered on the basis of their standing on the predictor (single or composite), and job offers are made sequentially to those with the lowest ranks. With banding, individuals who have scores on the predictor that are within given scores ranges (bands) are considered to be equivalent to one another, and job offers are made sequentially to those within the bands that have the highest scores.
In summary, biases may affect many phases of the selection process, resulting in unfair discrimination. However, it is vital that such biases be minimized or eliminated in the interest of providing fair employment opportunities to all qualified individuals.
- Dipboye, R. L. and Colella, A., eds. Forthcoming. Discrimination at Work: The Psychological and Organizational Bases. Mahwah, NJ: Lawrence Erlbaum.
- Guion, R. M. 1976. “Recruiting, Selection, and Job Placement.” Pp. 777-828 in Handbook of Industrial and Organizational Psychology, edited by M. D. Dunnette. Chicago, IL: Rand McNally.
- Society for Industrial and Organizational Psychology. 2003. Principles for the Validation of Personnel Selection Procedures. 4th ed. Bowling Green, OH: Author.
- Stone, D. L. and Colella, A. 1996. “A Model of Factors Affecting the Treatment of Disabled Individuals in Organizations.” Academy of Management Review 21:352-401.
- Stone, E. F., Stone, D. L. and Dipboye, R. L. 1992. “Stigmas in Organizations: Race, Handicaps, and Physical Unattractiveness.” Pp. 385-457 in Issues, Theory and Research in Industrial and Organizational Psychology, edited by K. Kelley. Amsterdam, Netherlands: Elsevier.
- Stone-Romero, E. F. 2005. “Personality-based Stigmas and Unfair Discrimination in Work Organizations.” Pp. 255279 in Discrimination at Work: The Psychological and Organizational Bases, edited by R. Dipboye and A. Colella. Mahwah, NJ: Lawrence Erlbaum.
- Stone-Romero, E. F. and Stone, D. L. 2005. “How Do Organizational Justice Concepts Relate to Discrimination and Prejudice?” Pp. 439-467 in Handbook of Organizational Justice, edited by J. Greenberg and J. A. Colquitt. Mahwah, NJ: Lawrence Erlbaum.