9 Appendices

Appendix A: Culture Implementation Plan

Appendix B: Maturity Model Additional Information

This appendix provides an overview of the outcomes of interest, data sources, and analysis methods used in the Maturity Model assessment.

Quantitative Metrics

• Leadership diversity: based on demographic characteristics from workforce data
• Overall office or directorate diversity: based on demographic characteristics from workforce data
• Office or directorate-level adverse impact (or disparate impact) ratios: These provide information about the extent to which one group is selected (i.e., hired, promoted) less than 80 percent of the time that members of another group are selected.
• Salary and salary increases (over the last five fiscal years)
• Performance award receipt: In particular, cash awards, time off awards, on-the-spot awards, and director’s awards.
• Promotion receipt
• Education levels of staff
• Staff-initiated organizational exit
• FEVS Diversity Index: The practice of including the many communities, identities, races, ethnicities, backgrounds, abilities, cultures, and beliefs of the American people, including underserved communities.
• FEVS Equity Index: The consistent and systematic fair, just, and impartial treatment of all individuals, including individuals who belong to underserved communities that have been denied such treatment.
• FEVS Inclusion Index: The recognition, appreciation, and use of the talents and skills of employees of all backgrounds.
• FEVS Accessibility Index: The design, construction, development, and maintenance of facilities, information and communication technology, programs, and services so that all people, including people with disabilities, can fully and independently use them.

Data Sources

• HR records and workforce data
• Federal Employee Viewpoint Survey (FEVS) results
• Applicant flow data
• Historical documents (e.g., process manuals, communications materials, and any other qualitative data available)

Data Analysis Methods

• Qualitative Analyses: Trained external raters evaluate the inclusion of different historically underrepresented groups based on various factors, which include but are not limited to sexual orientation, gender identity, race, pregnancy, disability, immigration, socioeconomic background, etc.
• Chi-square: A statistical test used to determine if there is a discrepancy between diversity benchmarks and observed NSF workforce data. This test compares the distribution of the number of expected members of a particular group based on diversity benchmarks to the distribution of observed group members within the same groups.
• Analysis of Variance: A collection of statistical tests used to analyze the differences between means on a given continuous variable (e.g., salary) based on predictors (e.g., demographic groups).
• Logistic Regression: A statistical predictive model that calculates the probability of a particular binary event (e.g., receiving a cash award or not) and the associated odds ratios (OR). This is a way of quantifying the strength of the relationship between variables, such as the event taking place and a characteristic of interest.

Appendix C: Underrepresentation and Barrier Analysis Tool Information

This appendix provides an overview of the data sources, analytical methods, and anticipated utility of the Underrepresentation and Barrier Analysis Tool.

Data Sources

Our team has reviewed and refined the mapping and tables developed for the NSF STEM occupational specialties with the Standard Occupation Classification (SOC) system as a proof-of-concept for the model. The model leverages data from the Current Population Survey (CPS) to develop analogs to compare the NSF workforce’s demographic information with society’s representative labor force. The model also leverages historical NSF workforce data to create predictive analytic capabilities that will be improved over time. This demonstrates the level of diversity that can be achieved if respective segments of society were available to the workforce (at the National, State, and occupational specialty levels).

The SOC system is a statistical standard federal agencies use to classify workers into occupational categories for collecting, calculating, or disseminating data. Detailed occupations are combined to facilitate classification to form 459 broad occupations, 98 minor groups, and 23 major groups. Detailed occupations in the SOC with similar job duties, and in some cases, skills, education, and training, are grouped.

Analytical Methods

Quantitative underrepresentation analysis requires accurate benchmarking of respective racial, gender, and ethnic population compositions. While the overall U.S. population may be used as a primary comparison group, any data should be filtered to reflect various factors that affect accessions, career development, retention, assignments, and other areas of the Talent Management LOE framework. Examples include age limitations, education, and other requirements provided in the Americans with Disabilities Act (ADA).

The approach will aggregate multiple authoritative data sources to develop benchmarking methods that allow an analysis of the Service’s expected composition compared to the broader labor force. Once processed, the aggregated data will present a DEIA common operating picture intended to provide leaders at all levels with data-based situational awareness of the DEIA environment to promote more informed personnel employment decision-making.

Anticipated Utility

To solicit sexual orientation and gender identity (SOGI) information and enhance the UR Tool, American Community Survey data will be incorporated, as it is more rigorous than the CPS. Additionally, Census Bureau data will be explored to account for geographic variation in the tracked metrics. Formalized data models are being developed, and simulated effects of potential policy actions will be integrated to project future outcomes of proposed changes in workforce management. Including policy-change modules allows for a cost-effective way to develop baseline expectations for the potential value of specific changes. By creating a simulation component, outputs based on the expected effects of changes, such as Position Description Modernization, can be provided. The expected effect’s magnitude can then be weighed against the potential cost to perform a cost-benefit analysis for leadership to consider before updating policies.

Appendix D: Culture Intelligence Model Information

This appendix provides an overview of the Denison Model process and additional information on the model itself.

The Denison Model and assessment provides organizations with an easy-to-interpret, action-based, and data-driven approach to culture and performance improvement based on sound research principles. At the center of the Model are the organization’s “Beliefs and Assumptions.” These are the deeply held aspects of an organization’s identity that are often hard to access, understand, or articulate. The Model highlights four key traits an organization should strengthen to be effective: Mission, Adaptability, Involvement, and Consistency, each represented in red, blue, green, and yellow respectively on the model. These traits measure the behaviors driven by the beliefs and assumptions that create an organization’s culture. Each of the four traits includes three indexes. For example, the Consistency trait measures the organization’s Core Values, Agreement, and Coordination & Integration. Each index is measured by four items for 48 items in the survey.

Denison’s research has demonstrated that effective organizations have strong capabilities in all four traits. They are most likely to have adaptive cultures that foster high involvement within a context that incorporates consistent and predictable processes and a shared sense of mission.

The Denison Model links organizational culture to performance metrics such as DEI, Innovation, Employee Engagement and Commitment, Trust, Growth, Operating Efficiency, and more. The Denison DEI metric is designed to explore further how effectively organizations promote diversity in their workforce and ensure that opportunities are equally available to all staff. The combination of the Denison Culture Model and the DEI performance metric allows NSF to understand how the culture impacts the sustainability of DEIA efforts.

Identifying Gaps and Taking Action

The strength of the Denison Model lies in its ability to identify gaps and potential areas for improvement within NSF’s culture. This approach starts with strategic priorities defined by leadership and an understanding of other organizational initiatives currently employed. These then enable the definition of organizational priorities and the formation of data-driven action plans aligned with these priority areas. Organizations that adopt an intentional culture awareness and development process have observed a notable increase in organizational performance and effectiveness.^[1]