OFCCP Week in Review: August 2022 #2 | Direct Employers Association

Tuesday, August 2, 2022: In a webinar discussing its compensation data study, the NASEM Expert Panel noted flawed design, data issues, and recommendations

The Expert Group Behind the National Academies of Science, Engineering, and Medicine (NASEM) report the EEO-1 Survey Component 2 Limited Data Collection Quality Assessment for the 2017-2018 reporting years hosted a online seminar to discuss his method and conclusions in more detail. See last week’s WIR: “The National Academy of Sciences has published its study on anticipated earnings data; EEOC Commissioners Respond With Notably Different Takesfor more details on the report’s findings. Here are some highlights from the presentation.

Survey design is outdated and useless for self-assessment

The current design of the EEO-1 survey has not been revised since the mid-1960s, panelist Don Tomaskovic-Devey, a sociologist at the University of Massachusetts, noted. Employers have told them that this design is completely useless for salary self-assessments. “Employers have been very clear on this,” he said.

Job categories and salary ranges are too broad due to this outdated design. The current design “describes the US economy in 1965 better than it does today,” Tomaskovic-Devey observed, adding that it was designed largely for the pre-computing world.

Measurement problems abound

The panel cited several measurement issues of concern, including:

  • The measure of salary should be total compensation, not taxable compensation.
    • Box 1 on W-2s does not measure all offsets (Conclusion 3-1)

To note: References in parentheses refer to the written report of the Panel

  • Salary bands were too wide (Conclusion 3-2)
  • Job categories were too broad and outdated (Finding 3-3)
  • Sex and race/ethnicity measures did not meet federal best practices and standards, respectively (Finding 3-4)
  • No data was collected for some protected groups (Finding 3-5)
  • Data on legitimate causes of pay gaps, such as education and experience, were not collected (Finding 3-6)
  • Business and establishment identifiers are neither consistent nor unique (Finding 4-2

Committee recommended moving to individual compensation data collection

Moving to collecting salary data at the individual level would solve many problems, the panel said. Employers already report individual salary data to several state and local government agencies. “It’s something employers already know how to do and do well,” Tomaskovic-Devey commented.

Collecting wage data at the individual level would reduce respondent burden while improving data quality, panelists agreed. In addition, it solves several measurement problems (Recommendation 3-7). Moving to individual data will increase accuracy and reduce costs, Tomaskovic-Devey said. It also solves a lot of problems with times. Individualized data will allow for a much better adjustment of raw wage differentials and can include education and experience. “It’s a much more flexible way to work with long-term data,” he observed. The Bureau of Labor Statistics’ transition to individual wage data in the Statistics on employment and wages by profession can be instructive for the EEOC, the panel said.

Additionally, corporate stakeholders said their reporting burden would decrease if the EEOC collected individual data, in part through improvements in data collection, said group chairman William Rodgers III, who is vice -president and director of the Institute for Economic Equity of the Federal Reserve Bank. of Saint Louis. This would allow employers to tell “better stories at a lower cost,” he noted.

Data is valuable

“The United States is a bit unusual in that it doesn’t have data like this compared to other countries,” Tomaskovic-Devey pointed out. Getting this kind of data will be useful for both the EEOC and employers, said Valerie Rawlston Wilson, director of the Economic Policy Institute’s program on race, ethnicity and economics. This data is “essential to filling this hole in the information infrastructure needed to [legal] app,” she said.

EEOC should look to other agencies for examples on how to move forward

Panelists repeatedly noted their recommendation that the EEOC consult with statistical agencies such as the Census Bureau and the Bureau of Labor Statistics, which already collect data at the individual level, but do so without class demographic factors. legally protected that the EEOC would need. .

The study did not drastically address privacy issues, but panelists noted that there are a number of methods that statistical agencies such as the Bureau of Labor Statistics and the Census Bureau use to protect confidential information, and the EEOC should work with these agencies. “Other federal agencies have long since addressed this issue,” Tomaskovic-Devey said.

Judith K. Hellerstein, professor of economics at the University of Maryland, acknowledged that there may be legal barriers preventing agencies from sharing statistical data and working together to share information. She also noted that processing data at the individual level is something the EEOC has not yet done and will require more computing power and related resources.

Other ratings

A few additional observations to note:

  • Panelists repeatedly mentioned that “additional data cleaning” needs to be done.
  • “We are aware that the EEOC is conducting an internal assessment of data modernization,” Tomaskovic-Devey said. Therefore, some of the panel’s recommendations that “may seem drastic” are actually not so extreme.
  • The panel repeatedly and strongly recommended pre-testing, field testing, and stakeholder involvement.
  • They also noted that the report makes it clear that the collection is only raw data and that unadjusted data does not prove any violation of the law. These data can only provide initial indications of where discrimination might occur, but they do not constitute proof of a violation of the law.

Andrew B. Reiter