Metrics of Diversity
by Marc Brenman
Social Justice Consultancy
About the author – Marc Brenman
Brenman was Exec. Dir., Wash. St. Human Rights Commission, 2004-09, Senior Policy Advisor, US Dept. of Transportation, 1995-2004. Works on race, limited English proficiency, disability, sex discrimination, LGBT rights, human rights. Co-author of “The Right to Transportation” and “Planning as if People Matter: Governing and Equity,” in press. Teaches civil and human rights, pub. policy, advocacy, governing for social justice. Consults and writes on these issues, EEO, diversity. 2010, helped Public Advocates, Inc. win first civil rights case under ARRA. for $70 million. In 2012, won 1st prize in Americans for Democratic Action Education Fund essay contest for MLK Day.
There are many definitions of diversity from a social justice perspective. We use one centered on organizations desiring to increase their employment of traditionally underrerepresented groups, or protected classes, such as race, national origin, sex, disability, LGBT status, and age. We generally limit the metrics to those in an employment context. The metrics discussed are most pertinent to the United States, but can be adapted to other countries. While metrics can cover both quantity and quality, we primarily discuss quantified ones. It is important to recognize what diversity metrics are not. They are not quotas or preferences. However, they can be used for goals and to evaluate compliance with court-ordered or voluntary settlement agreements.
Measurements are necessary to determine what needs doing, how to plan initiatives, and their success. Diversity metrics are typically collected, analyzed, evaluated, and promulgated by human resources or diversity departments. Much of the data can be obtained through employee databases. Collecting the data can present challenges, particularly if the organization has not previously done so. Federal contractors and companies required to report by the federal Equal Employment Opportunity Commission will find that they already have much of the data. However, due to self-reporting, anonymity, and fears of disclosure, there may be challenges in ensuring accurate reporting and counts of employees with disabilities, and for LGBT status.
An employer should know the demographics of its employees, and that of its service area. The latter can be found through many different sources. It is unlikely that the employer will have to create new data. For its own employees, baseline data must be established. This, and all diversity data, should be disaggregated by the categories listed and any other categories the employer wants to use, by type of job, by pay band, by division or office, and by geographical location. Data should also be disaggregated by major employment milestones, such as job applicants, hires, training opportunities, evaluations, and promotions.
Diversity in the US is increasing. This is especially true for groups like Hispanics and Asian-Americans. Efforts to improve the representation of groups should show a rate higher than natural demographic increase. If an organization only improves its diversity at the same or a lower rate than natural demographic increase, then its diversity program has failed.
Ratio measures and measures over time should be used where possible to provide means for comparison, and to show continuity of effort. Diversity metrics are sometimes considered a scorecard.
Benefits from diversity ought to be measurable. Such measures can include better relationships among diverse staff; fewer grievances and complaints; improved labor relations; reduction of noose, graffiti, and hate incidents; and more diverse hiring. Employee climate surveys can be used. Other metrics can include improvements in productivity, more innovation and creativity, job satisfaction, career development over time, retention, decrease in pay disparities, performance on exit interviews, ranking of organization as a desirable place to work, and becoming an employer of choice.
Traditional or foundational metrics include numbers related to success in workforce representation, dollars spent with women and minority businesses, and the documentation of diversity activities and training.
Diversity programs should be able to show real statistical increases in numbers and percents of recruitment, applications, hiring, retention, training, and promotion of people from traditionally discriminated and underrepresented groups. In terms of qualitative measures, employee climate surveys can include questions about perceptions of diversity, equity, welcomeness, trust, fairness, transparency, and accountability, and should show improvement over a baseline. Employee focus groups should show similar improvement. On a psychological level, social distancing scales should show decreased distance and prejudiced thinking. Measuring performance should include social equity and opportunity, quantity, quality, timeliness, cost, effectiveness, efficiency, return on investment, and customer service.
It is possible to measure the savings in recruitment costs from higher retention rates for diverse employees through building a more inclusive culture. Another measure is the reduction in absenteeism for underrrepresented affinity groups when an inclusive culture is created. The engagement ratings for employees from different affinity groups can be measured in annual engagement surveys.
Quick List of Possible Metrics