Methodology

The Partnership for Public Service created the Best Places to Work in the Federal Government® rankings in 2003 to provide the most comprehensive rating of employee engagement across federal agencies and their subcomponents. We use the term employee engagement to refer to the satisfaction and commitment of the workforce and the willingness of employees to put forth discretionary effort to achieve results. 

The 2020 Best Places to Work rankings include 482 federal agencies and their subcomponents: 17 large agencies, 25 midsize agencies, 29 small agencies and 411 subcomponents. We group agencies by workforce size to provide comparisons of agencies that may face similar management challenges. Large agencies are those organizations with 15,000 or more employees. Agencies with 1,000 to 14,999 employees are included in the midsize category. Small agencies are those with at least 100 but fewer than 1,000 employees. Subcomponents—the subagencies, bureaus, divisions, centers and offices within agencies—need to have at least 100 employees to be included in the rankings. The number of employees was determined by using OPM’s FedScope database at the end of fiscal 2020 unless otherwise noted.

The majority of the data used to develop these rankings was collected by the Office of Personnel Management through its Federal Employee Viewpoint Survey. The survey was administered as a census to all full-time and part-time, permanent, nonseasonal executive branch employees between September and November 2020.  It was completed by 624,800 federal workers, for a response rate of 44.3%, a 1.7-point increase from 2019. Visit OPM’s website to learn more about the FEVS methodology.

The rankings also include responses from more than 303,800 additional employees at 12 agencies which were surveyed at the same time and had a response rate of more than 50%. The Congressional Budget Office, Consumer Financial Protection Bureau, Federal Deposit and Insurance Corporation, Government Accountability Office, Millennium Challenge Corporation, Department of Homeland Security’s Office of Intelligence and Analysis, Peace Corps, Securities and Exchange Commission, Smithsonian Institution, Tennessee Valley Authority’s Office of the Inspector General and the United States Postal Service’s Office of the Inspector General  provided data from their own surveys. In addition, the rankings incorporate responses from employees at the nation’s intelligence agencies, which conducted a similar survey but did not report the number of respondents.

For the third year in a row, the Department of Veterans Affairs did not participate in the 2020 FEVS and instead administered its own internal survey. Since the VA’s survey included the three questions needed to calculate an agency’s Best Places to Work employee engagement score, the department is included in our rankings. However, like other agencies that conduct their own surveys, we have not included the VA’s data in the calculation of the 2020 government-wide score. The OPM data was weighted to represent all federal employees on demographic characteristics such as age, gender, race, supervisory status, agency size and occupation. The survey results have a margin of error of plus or minus 1.0%.

The Best Places to Work statistical model, developed with CFI Group in 2003, is based on the same methodology used in CFI Group’s American Customer Satisfaction Index. The Best Places to Work engagement score is calculated based on the percentage of positive responses to three workplace engagement questions and is weighted according to a proprietary formula developed with Hay Group (later acquired by Korn Ferry) in 2007.

The Partnership changed how it calculates the percentage of positive responses to each FEVS question in 2020. Previously, we divided the number of positive responses (e.g., the number of respondents who answered “agree” or “strongly agree”) by the total number of people who completed the survey. Beginning in 2020, the Partnership divided the number of positive responses to each question by the number of people who answered that particular question. This change resulted in smaller denominators, filtering out respondents who skip questions, and slightly larger percent-positives.

The workplace category scores are calculated by averaging the percentage of positive responses to questions across different aspects of the employee experience, including effective leadership, employee skills–mission match, pay, recognition and work–life balance. For 2020, categories were added to reflect employee views related to the COVID-19 pandemic. This includes employee perceptions of their agency’s performance during the pandemic and the support they received from their agencies and their leaders.  

The Partnership and BCG would like to thank OPM for its excellent administration of its employee survey, without which the Best Places to Work rankings would not be possible.