Methodology

We conducted an online survey in English and Spanish with a nationally distributed sample of 5,012 U.S. adults between December 17, 2024, and January 24, 2025. The survey oversampled individuals from low-income households (88 percent had annual household incomes under $65,000) who either filed taxes inconsistently or not at all in the past three years.

All participants were between 18 and 64 years old, and responses were collected from all 50 U.S. states, Washington, DC, and the U.S. Virgin Islands.

We applied quota sampling based on age, household income, and tax filing behavior to ensure adequate representation of lower-income and inconsistent or non-filing households.

The final sample included:

• 63.7 percent who had not filed taxes at all in the past three years
• 7.6 percent who had filed inconsistently (some years but not others)
• 28.7 percent who had filed taxes every year

The sample included a racially and ethnically diverse group (with 7 percent of the respondents selecting multiple identities): 65 percent white, 25 percent Black, and 11 percent Hispanic/Latino, with smaller representation from Native American (3 percent), Asian/Pacific Islander (3 percent), and other (0.9 percent) backgrounds.

We implemented quality control measures through our survey partner, Qualtrics, including screening for duplicate or bot responses, detecting straight-lining behavior, randomizing response order where applicable, and removing responses completed in significantly less time than the average, which could indicate low-quality or inattentive participation.

Respondents were asked about:

• their households’ tax filing and tax credit claiming behaviors and experiences
• employment and family structures
• barriers to claiming tax credits
• awareness of tax credits
• trusted sources of information for tax and other benefits information
• participation in other government support programs
• comfort with data sharing and integrated service delivery models

Unless otherwise noted, all figures referenced in this article represent statistically significant findings from the survey and are calculated based on item-level response rates, not total sample size.

In parallel with the national survey, we conducted a qualitative study with 25 low-income residents across Illinois between September 2024 and February 2025. This qualitative research aimed to surface emotional drivers, motivations, and lived experiences that shape how people approach tax filing.

Participants were primarily recruited through New America Chicago’s partnership with the Coalition on Family Issues (COFI), with additional outreach through community organizations, local Craigslist postings, and families in New America Chicago’s CivicSpace cohort.

The study included:

• Group discussions (two sessions): Hosted during regular COFI parent meetings, these sessions explored perceptions of a potential partnership between the state Department of Human Services (DHS) and the state Department of Revenue to improve access to tax credits. Participants raised concerns about access, prompting a deeper exploration of how past interactions with public agencies shape attitudes toward tax filing.
• In-depth interviews (in English and Spanish): One-on-one, semi-structured interviews with nine participants provided insight into how personal and family experiences with state benefits and tax systems influence filing behavior and credit awareness. Interviews explored familiarity with credits, sources of confusion or misinformation, and how previous public system experiences shape filing decisions. Participants represented a range of family structures and filing patterns across Illinois.
• In-person co-design workshop: In partnership with state representatives, a design session with 10 residents focused on co-creating messaging strategies and identifying optimal communication timing for a new tax service. The session emphasized collaborative problem-solving and centering participant voice.
• Messaging survey (in English and Spanish): Participants who had filed taxes at least once in the past three years were first assigned a behavioral archetype based on filing motivations and barriers. They were then shown a scenario involving a text message from DHS about a new tax service. From a set of 11 messages (both generic and archetype-tailored), they were asked to select up to three that resonated most.

Insights from this study informed the development and refinement of behavioral archetypes that represent distinct mindsets and patterns around tax behavior. These emerging archetypes are helping to inform our work on messaging, potential policy friction points, and approaches to more personalized support for low-income families.