This report summarizes findings from a representative survey of 2,049 people aged 18 years and older in the United States who are covered by health insurance provided by their own or a family member’s current or former employer or union. 

This Public Agenda project is supported by Arnold Ventures. Cite Public Agenda when referencing these findings. 

The survey was designed by Public Agenda and fielded from March 22 to April 9, 2024 by NORC at the University of Chicago. Respondents were selected from the NORC’s probability-based AmeriSpeak panel, using sampling strata based on age, race/Hispanic ethnicity, education, and gender. Participants in this study were screened based on their insurance coverage. Respondents with insurance coverage from an employer or union qualified to complete this survey. Oversamples were collected to attain at least 200 Asian American and Pacific Islander (AAPI), Black Americans, and Latino Americans with ESI. In this nationally representative sample of people with ESI, 85 percent of respondents report having health insurance through a current or former employer of theirs or another family member; 14 percent report having health insurance through a current or former union of theirs or another family member; and 1 percent do not know whether their health insurance is through an employer or union. 

The survey was offered in English and Spanish and was administered online via the Web (2,049 completes). If a panel household had more than one active adult panel member, only one adult panel member was selected at random. Panelists were offered the cash equivalent of $3 for completing this survey. 

NORC applied cleaning rules to the survey data for quality control. In total, 151 cases were removed from the final set of completed interviews based on four cleaning rules:

  • Removing speeders (i.e., those who completed the survey in less than one-third of the median duration). 126 cases were removed for speeding. 
  • Removing respondents with high refusal rates (i.e., those who skipped or refused more than 50 percent of the eligible questions). 38 cases were removed for high refusal rates. 
  • Removing straight-liners (i.e., those who straight-lined grid item questions). 32 cases were removed for straight-lining. 
  • Removing back button users (i.e., those who use the back button to change their answers at the screening questions). One case was removed for suspicious use of the browser back button.

The final weight is developed as a product of three weights. First, AmeriSpeak Panel weights are developed for all panel members to account for their probability of selection into the sample of panel recruits, panel recruitment nonresponse adjustments, and poststratification adjustments of the recruited panel to match population benchmarks. Second, Study Specific Base Weights are developed for a study sample selected from the panel to account for their selection probabilities under the sample design. Finally, Study Specific Final Weights are adjustments of the base weights to address survey non-response through a weighting class method. Raking adjustments are then applied to the non-response adjusted weights to align the survey sample to specific population benchmarks. The final weights may be trimmed to reduce the influence of extreme weights on survey estimates.

The weighted household panel recruitment response rate for the AmeriSpeak Panel is 22.3 percent, while the weighted household panel retention rate is 78.9 percent. As for the survey sample, the completion rate is 18.2 percent, and the cumulative response rate is 3.2 percent.

The oversamples of people with ESI who identify as Asian American, Black American, and Latino American were weighted down to their proportions in the overall population in the final main study weights. The survey has a margin of error of +/- 2.82 percentage points at the 95% confidence level for all people with ESI. The margin of error for subpopulations are as follows:

  • +/- 11.81 percentage point at the 95% confidence level for Asian American respondents
  • +/- 9.68 percentage point at the 95% confidence level for Black American respondents
  • +/- 7.94 percentage point at the 95% confidence level for Latino American respondents

Sampling error is only one of many potential sources of error and there may be other unmeasured errors in this or any other survey.

To inform the development of this survey, Public Agenda conducted five online focus groups in January and February 2023 with 40 residents of Texas, Washington State, Wisconsin, North Carolina, and Pennsylvania. All participants were employed adults ages 18 years and older. All were covered by employer-based health insurance from their own employers. Findings from those focus groups were reported in a separate report, available by following this link. 

For a topline with full question wording, please follow this link. For other inquiries about the research, please email [email protected].