Counting Care

Nicole Lee Schroeder

June 3, 2024

Illustration of the Philadelphia

A poster of “The Bettering House” which was built in 1767 between in Philadelphia. The facility was a workhouse with separate wings and floors serving as a hospital and orphanage. [Bettering House—Philadelphia Almshouse. ca. 1884, engravings, Free Library of Philadelphia: Philadelphia, PA, accessed Feb. 9, 2024,]

How we count matters. Data reflects reality only when the numbers are collected with care, analyzed critically, and narrated ethically alongside transparent methodologies. As a researcher, I’m interested in the numbers that early Americans collected, especially as they pertain to welfare. The numbers they chose to collect, the stories they spun out of these values, and the policy changes enacted because of this “factual” evidence are fascinating. In my research, I question whether reported numbers reflect reality. More importantly, I consider how welfare reformers weaponized numbers to distort reality. In my recent JER article on medical rhetoric and disability in early national Philadelphia, I look at numbers as they pertain to disabled demographics.[1] I argue that early welfare reformers set a dangerous precedent whereby disabled people were purposefully “hidden” from population counts, disregarded in welfare funding allocation, and forced into inefficient and harmful institutionalization schemes. Disabled communities today live with the ramifications of these choices, and politicians continue to manipulate data just as their predecessors did 200 years ago.

If you look at the welfare reform debates at the turn of the nineteenth century, reformers across the Atlantic world used hard data to back up qualitative claims. Reformers argued that tax rates were too high, poverty rates unprecedented, and the undeserving poor exponentially outweighed the deserving poor. In Philadelphia, a report from 1817 concluded that “the people of color; the lower classes of Irish emigrants; the intemperate and day laborers” were glutting the welfare system.[2] The implication? People who deserved welfare support (women, elderly persons, orphaned children, disabled folk) were outnumbered by “undesirable” folk who were just lazy and taking advantage of the system. This claim, reformers argued, could be backed by hard data. At different points, the report notes that 18.5 percent of the poor were African, two-thirds of the poor wasted their pensions on “ardent spirits,” and numerous men abandoned their families—an issue described as “lamentably extensive.”[3] The committee also reported, “It does not appear, from the information before the committee, that the proportion of sick is very great.”[4] Their staunch arguments led me to question, how did Americans count disabled populations in the past? Who gets included in tallies and who is disqualified? Who does the counting and how? And do their calculations hold up under scrutiny? For early American welfare reformers, disability counts didn’t add up to much of anything. Reformers argued that any evidence of health issues for those on the poor lists stemmed from terrible living conditions. How these numbers were collected, though, leaves much to be desired. Exaggerated rhetoric, methodologically unsound data collection, and false claims litter the report.

Appendix from Report of the Library Committee of the Pennsylvania Society for the Promotion of Public Economy (Philadelphia, 1817)

The Report included a proposal for a new “House of Industry” where poor people would be able to take up piecemeal tasks and artisan trades in exchange for food and medical services. The facility was notoriously crowded, and its profits were never sufficient to meet its costs. [Report of the Library Committee of the Pennsylvania Society for the Promotion of Public Economy (Philadelphia, 1817),]

The 1817 report loudly called for changes to welfare practices. Reformers wanted to end the pension system and shift everyone into workhouses. This idea that work could cure all rested on the premise that disabled people did not exist. Everyone could work, and once forced to find the glory in labor, they could be rehabilitated back into society. The unwillingness to count a rising disabled minority had huge consequences for citizens across the following decade. Reformers in the 1820s continued to call for new workhouses, capable of housing thousands. Pension records from 1829, however, demonstrate the futility of this plan. By 1829, the vast majority of pensioners in Philadelphia were elderly disabled women who had lived in the city most of their lives, whose families struggled with cyclical poverty, and whose health status severely impacted their labor capacity, but who reported working nonetheless. By definition, these women were “the deserving poor.” Why then did reformers ignore their own data?

Reformers were collecting the very data they needed to track the rise of a disabled minority in early America. Even with quantitative evidence, though, reformers refused to pay witness to this phenomenon. While the vast majority of pensioners in 1829 recorded at least one disabling health condition (if not more), reformers staunchly argued that institutionalization could reform, rehabilitate, and cure everyone. The disabled poor could be screened more effectively in institutions by medical professionals who could determine how disabled they were and how much work they were capable of performing. Shifting power from welfare officials to medical professionals, the reforms of the 1820s and 1830s outlined a dangerous practice whereby both federal and state governments ignored the growth of a disabled minority.

This sense that disabled people don’t really exist in high numbers is a mainstay of American welfare policy. Disabled people make up to 27 percent of the adult American population, yet funding for disability services lags far behind current needs.[5] Politicians refuse to acknowledge a growing disabled minority, and they continue to manipulate data to legitimize funding cuts, just as their early American predecessors did. This past fall, the U.S. Census Bureau announced a proposal to use a new set of disability questions for the American Community Survey (ACS). The current ACS uses six yes-or-no questions to determine an individual’s disability status. If someone answers “yes” to any of the six, they “count” towards disability population data.[6]

Infographic illustrating that 1 in 4 adults in the US have some type of disability

The CDC reports that 1 in 4 American adults has one or more disabilities. [Disability Impacts All of US. 2023. CDC.]

The Census Bureau proposed a change from current ACS questions to the Washington Group Short Set on Functioning (WG-SS) questions. Instead of yes-or-no questions, the WG-SS questions include four options spanning in severity (“no difficulty,” “some difficulty,” “a lot of difficulty,” and “cannot do at all.”) Under the WG-SS you only “count” as disabled if you answer you have “a lot of difficulty” with a task or “cannot do at all.” The WG-SS questions are murky and subjective. Let’s say, for example, that you struggle to hear well in noisy, crowded spaces. Under the ACS questions you qualify as disabled. But if you report just “some difficulty” with hearing under the WG-SS questions, you would NOT be considered disabled. While you might struggle to hear only some of the time, you may still benefit from cost-saving programs for hearing aids, speech-to-text technologies, or subsidized ASL (American Sign Language) courses.

The ramifications of this potential change in questions are severe. Researchers warn that this change would cut America’s disabled population by 40 percent and “reduce the prevalence of disabled people in the U.S. to 8%, down from the ACS’s current estimate of 14%.”[7] If the disabled population is cut in half, what happens to already lackluster funding programs? NDRN (National Disability Rights Network) warned that the switch to WG-SS questions might lead “to reduced funding for critical federal programs and loss of critical data demonstrating the needs of the community.”[8] The demand to ignore a growing disabled population should be viewed with scrutiny and concern. Cuts in services will lead to lower quality of life for America’s disabled population.

Interestingly enough, as activists have highlighted, the WGS doesn’t claim to effectively identify total disability data.[9] In their major documentation, when asked “Does the WG-SS identify everyone with a disability?” the group notes “The answer to this is easy. No, the WG-SS does not identify all people with disabilities, nor does it need to.”[10] Why then is the U.S. Census Bureau determined to swap out current questions for something methodologically unsound? Disabled activists and allies protested these changes by filing comments with the federal government, reporting extensively on the issue, and mobilizing to meet with representatives. Just last week, the Census Bureau reported that 12,000 comments were filed in protest.[11] Due to the sheer volume of these critiques, the Census Bureau has decided to hold off any changes for now. Despite this temporary victory, disabled communities recommend vigilance. Unfortunately, disability-rights activists are right to be skeptical about the future actions of the Census Bureau. Undercounting minoritized populations to legitimize funding cuts is intentional, and government officials have been doing it since the founding of this nation.


[1] Nicole Lee Schroeder, “An Emporium of Beggars: Medical Rhetoric, Disability, and Philadelphia’s Early Nationalist Welfare Crises,” Journal of the Early Republic 44 (Spring 2024), 57–86.

[2] Pennsylvania Society for the Promotion of the Public Economy, Report of the Library Committee of the Pennsylvania Society for the Promotion of Public Economy (Philadelphia, 1817), 12.

[3] Ibid., 17, 13.

[4] Ibid., 14.

[5] Centers for Disease Control and Prevention, Disability Impacts All of Us,” Disability and Health Promotion, updated May 15, 2023, accessed online Jan. 27, 2024, Michelle Diament, “Nation’s Disability Services at Breaking Point, Report Finds,” Disability Scoop (Mar. 2, 2023),

[6] If you’d like to read the proposal, you can see the Comment Request documentation that was released to the public at Census Bureau, “Comment Request American Community Survey” (Oct. 20, 2023),

[7] Phie Jacobs, “How many in the U.S. are disabled? Proposed census changes would greatly decrease count,” Science (Nov. 8, 2023),


[9] Lisa McCorkell (ALisaAMcCorkell), Dec. 18, 2023, Scott Landes (@scottdlandes), Dec. 12, 2023

[10] The Washington Group, “An Introduction to the Washington Group on Disability Statistics Question Sets,” Washington Group on Disability Statistics December 10, 2020, accessed online Jan. 27, 2024,

[11] Robert L. Santos, “Next Steps on the American Community Survey Disability Questions,” United States Census Bureau (Feb. 6, 2024),

Next Articles

The Pope and the Treaty Power: A Strange Incident in the North Carolina Ratification Debate
In a new companion piece to his latest JER article, Robert Smith discusses some of the unusual ways that religion came into play at North Carolina's Hillsborough Convention in 1788.
Counting Care