Measuring Sexual Orientation and Gender Identity on the ACS

A recent working paper from the Census Bureau presented at the Conference of European Statisticians outlined the Bureau’s current work to test sexual orientation and gender identity (SOGI) questions on the American Community Survey (ACS).

The Bureau proposed in FY 2024 to dedicate some of its increased ACS funding in its budget proposal to “Research and test question wording on SOGI [sexual orientation and gender identity] topics, specifically for proxy response, response in self-response modes, and translation.” The Census Project explained this and other details in our FY 2024 funding recommendation.

“Not having population-level data from a census is one of the major challenges in studying the characteristics of the LGBTQIA+ population,” the paper authors explained. “The current research will consist of both cognitive and field testing. Cognitive testing will include testing of questions in English and translation and testing of questions in Spanish. Field testing will include self-response using both paper and internet modes and examine question wording, response options and placement. One important area of research this testing will illuminate is the quality of proxy reporting of SOGI information in demographic surveys.”

In addition to providing background, the working paper looks at key areas of concern for the SOGI content on a survey like the ACS, and emphasizes the need to “collect these important data about SOGI populations and to better represent the diversity of the American people,” while focused on collecting “the data as accurately as possible and in a way that maintains the integrity of the ACS as a descriptor of the people and communities of the United States.”

NPR also has a useful article related to this topic, “What the 2020 census can — and can’t — tell us about LGBTQ people.” https://www.npr.org/2023/05/25/1174739030/census-lgbtq-data-same-sex-households-ask-about-sexuality

Study on Poverty Measurement Sparks Controversy

A new study from the National Academies urging the Census Bureau to update the methodology for calculating the Supplemental Poverty Measure (SPM), “An Updated Measure of Poverty: (Re)Drawing the Line,” also recommended “that the more comprehensive SPM replace the current Official Poverty Measure as the primary statistical measure of poverty the Census Bureau uses” and “expanding its use in recognition of the needs of most American families such as medical care, childcare, and housing costs.”

However, Scott Winship, Director of Poverty Studies at the American Enterprise Institute argued in a white paper that, instead of providing “the necessary information,” the Academies panel “attempted to entrench a specific type of poverty measure further into the bureaucracy of federal statistics without regard to the fundamental question of what best informs public understanding of the needs of poor Americans. The evidence suggests that key features of the SPM make it less accurate at identifying the poor than” the official poverty measure in use today.

Winship warned that the recommendation from the panel to replace the current official poverty measure with the SPM was “outside the panel’s mandate” and “reflects value judgements outside the realm of science—judgements lacking consensus among poverty measurement experts. This recommendation only fits in a ‘consensus study’ because the panel features even less ideological diversity than did its nearly invariant 1995 predecessor.”

Sen. Marco Rubio (R-FL) recently responded to the controversy by urging the Census Bureau to “disregard these recommendations and commission a new, politically balanced report.”

New Report on Where to Count the Prison Population

A new report from the National Conference of State Legislatures (NCSL) looks at “how states account for people who are incarcerated at the time redistricting occurs—also known as prisoner reallocation, prison gerrymandering reform or, as NCSL refers to it, inmate data reallocation. By whatever name it’s given, the process subtracts inmates’ data from the location where they are incarcerated and adds it to the place they called home before their imprisonment.”

Some advocacy groups want the Census Bureau to count prisoners at where they used to live, rather than their current residence (prison).

NCSL charted a significant increase in “inmate data reallocation” between the 2010 Census and 2020 Census, going from only two states having such policies, to 13 states in 2020 redistricting. The report explores the pro and con positions for these policies, and delves into the experiences of those 13 states that implemented such policies in the 2020 redistricting cycle: “California, Colorado, Connecticut, Delaware, Maryland, Montana, Nevada, New Jersey, New York, Pennsylvania, Rhode Island, Virginia and Washington. Illinois passed an inmate data reallocation law prior to the 2020 cycle but delayed its implementation to 2030. Advisory commissions in New Mexico and other states did some reallocation work, but the reallocated data sets were not used in final maps adopted by policymakers.”

There has been some state and federal legislative moves on this topic recently as well (see the April Census Project Update).

Report on Census Undercount of Asian American and Pacific Islander Communities

“While Asian American and NHPI communities were overcounted nationally” in the 2020 Census, according to a new report from Asian Americans Advancing Justice (AAJC), “some states had undercounts in both 2010 and 2020. This is a problem because, despite a reported national overcount of these communities in 2020, some Asian American and NHPI communities were still undercounted at lower levels of geography.”

The report also found disparities in the undercount of young children, which, while “common in other racial and ethnic groups,” it “also exists among Asian Americans” and “is not evenly distributed throughout the country.” AAJC discovered, by contrast, that Native Hawaiian and Pacific Islander (NHPI) young children were not undercounted, and urged further research to help determine why.

GAO No Longer Considers the Decennial Census “High-Risk” But Will Keep an Eye on 2030 Census Planning

The Government Accountability Office (GAO) has removed the decennial census from its list of high-risk federal programs. It had been on the list for a long time. However, GAO will be keeping close watch on preparations for the 2030 Census.

GAO found, “The Census Bureau slowed decades of cost growth while completing the 2020 Decennial Census during a pandemic. However, the continuing undercounts of segments of the population as reported by the Bureau signal that 2030 Census planning should be monitored for emerging risks.”

Costs “escalated with double-digit growth rates in its cost per housing unit from the 1970 through the 2010 Census, after adjusting for inflation. To achieve cost savings, the 2020 Census relied on several innovations. These innovations included re-engineering its field data collection with automated case management. Budget uncertainty caused the Bureau to scale back testing of these innovations in 2017. We are removing the Decennial Census from our High-Risk List because of progress in multiple areas. For example, the Bureau collaborated with independent outside entities for strategies dealing with quality concerns, chartered a high-level governance group, and monitored and demonstrated progress with priority recommendations. We are monitoring the 2030 Census planning—already underway—for emerging risks and challenges. If risks increase or we observe problematic challenges, we will consider moving the Decennial Census back to the High-Risk List.”

GAO plans to monitor various risks in 2030 Census planning, like: “Budgetary uncertainty”; “Inadequate testing”; “Late IT decisions”; and “Declining participation.”

The GAO report also shared a useful life-cycle time frame for the 2030 Census, below.

Should the ACS Eliminate Questions About Ancestry?

In response to a recent Federal Register notice, the Leadership Conference on Civil and Human Rights is urging “the Census Bureau to pause its research into possible elimination of the ACS ancestry question,” calling it “premature” and deserving of “additional scrutiny and transparency.”

  1. “The rationale for considering elimination of the ancestry question is not supported by any publicly available research.”
  2. “There is no publicly available information to evaluate the suggestion that data collected through the race question can meet the specific legal and programmatic needs underlying the justification for the ancestry question.”
  3. “Possible revisions to the OMB standards for collecting and presenting data on race and ethnicity could affect the way people respond to a proposed combined question, thus making any decision on whether to eliminate the ancestry question premature.”
  4. “The absence of public briefings or discussion of the possible elimination of the ancestry question is not in keeping with the Census Bureau’s new focus on transparency and stakeholder consultation.”

Read the Leadership Conference’s full April 7, 2023 comments for more details.

Reconsidering Data Collection on Native Americans

“The concept of “race” can be a problematic descriptor for Native American identity, which is a political and legal identity in addition to a racial one,” according to a new research report from the Brookings Institution. “By rethinking how the federal government collects and publishes data on Native Americans, we can begin to better assess the challenges they face and ensure the population is accurately represented when that data is used in research and policy.”

The report also explores how other countries have separated “data collection around Indigenous populations from racial identity. For example, both Canada and Australia ask about Indigenous identity in a separate census question, rather than as only one option among several races.”

It further recommends empowering “tribes to collect and manage data about their own populations and territories.”

The authors conclude that, “the current practice of measuring Native Americans using mutually exclusive, single-race data is not working well.”

Outside Testimony on Census FY 2024 Funding

Stakeholders can help to advance Census Bureau funding priorities by sharing their own written testimony with the U.S. House of Representatives and U.S. Senate Commerce, Justice, Science (CJS) Appropriations Subcommittees.

Outside witness testimony must be submitted to the House CJS Appropriations Subcommittee by April 14, 2023 and to the Senate CJS Appropriations Subcommittee by May 12.

Instructions from the subcommittees need to be followed carefully.

When drafting their own testimony, stakeholders can follow or build upon The Census Project’s FY 2024 funding recommendation for the Census Bureau.

It is recommended that the testimony explicitly state support for ensuring funding for the Census Bureau is a priority in the Fiscal Year 2024 CJS appropriations bills and to encourage the subcommittees to provide the agency with $2 billion in FY 2024.

The White House FY 2024 Budget Request for Census Bureau

The White House issued its Fiscal Year 2024 (FY 2024) budget on March 9, 2023, including $1.606 billion for the Census Bureau — a $121 million increase from the $1.485 billion appropriated in FY 2023.

The President’s FY 2024 budget requests $375,673,000 for Current Surveys and Programs (a $45 million increase from FY 2023) and $1,230,331,000 for Periodic Censuses and Programs (a $75 million increase from FY 2023).

The Census Bureau’s detailed budget submission to Congress is here.

The Census Project will issue its FY 2024 funding request shortly, recommending that the Census Bureau receive $2 billion ($394 million more than the President’s request, and a $521 million increase from FY 2023) to support 2030 Census preparations, pursue necessary technical innovations, expand programs like the Population Estimates, and enhance surveys, especially the American Community Survey (ACS) (informed by the ACS: America’s Data at Risk report).

The FY 2024 request for the Current Surveys and Programs account highlights:

  • Current Economic Statistics: $249 million in FY 2024 (an increase of $28 million from FY 2023), including:
    • A “new program designed to measure the production of advanced and emerging technologies by U.S. businesses.”
    • Expansion of “the Post-Secondary Employment Outcomes Program, which provides data on earnings and employment outcomes for college and university graduates by degree level, degree major, and post-secondary institution.”
    • A new “annual Puerto Rico Economic Survey and a monthly/quarterly economic indicator collection for Puerto Rico.”
  • Current Demographic Statistics: $127 million in FY 2024 (an increase of $18 million) to:
    • Establish and maintain “an infrastructure that supports improvements to intercensal population estimates, including improvements to the estimates base used to develop the annual population estimates.”
    • Formalize “a pilot program to re-use administrative records to improve measurement of health care characteristics and advance the nation’s understanding of population health.”
    • Research “innovative approaches to generating estimates about smaller population groups for the Current Population Survey” and “formalizing the Community Resilience Estimates program.”
  • State Children’s Health Insurance Program (SCHIP): $19 million in FY 2024 (same as FY 2023). “Mandatory appropriations are provided by the Medicare, Medicaid, and State Children’s Health Insurance Program Balanced Budget Refinement Act of 1999 … to support data collection by the Current Population Survey (CPS) on the number of low-income children who do not have health insurance coverage. Data from this enhanced survey are used in the formula to allocate funds to States under the SCHIP program.”

The FY 2024 request for the Periodic Censuses and Programs account highlights:

  • Periodic Economic Programs: $166 million in FY 2024 (a decrease of $24 million from FY 2023), including the Economic Census and the Census of Governments:
    • Data collection for the Economic Census, plus “follow-up activities to increase response, complete data collection, complete the process that captures company changes to update the master list of businesses, perform micro and macro analytical data review, and release “first look” national industry data.”
  • 2030 Census: $408.9 million, which is $160.1M over the FY 2023 funding level.
  • American Community Survey: $259.8 million, which is $9 million over the program’s FY 2023 funding level
    • “In 2024, entering the third year of its program lifecycle, the 2030 Census will approach its first major milestone, the selection of an operational design. Building on successful innovations implemented for the 2020 Census, the Census Bureau is researching ways to further enhance the program’s design.”“The American Community Survey (ACS) will continue efforts begun in 2023 to construct better question-wording on sexual orientation and gender identity topics.”
    • The ACS will also “continue to provide a testbed for innovative survey and data processing techniques that can be used across the Bureau.”
  • Geographic Support: $115 million in FY 2024 (a decrease of $1 million from FY 2023).
  • Enterprise Data Collection & Dissemination Systems: $280 million in FY 2024 (an increase of $54 million from FY 2023) supporting “major data collection, processing, and dissemination systems and associated research for the Census Bureau’s programs,” including:
    • Onboarding programs into the new dissemination system.
    • Integrating the “Enterprise Data Lake (EDL) with the Data Collection and Ingest for the Enterprise (DICE) programs. The DICE program will deploy functionality in support of several demographic and economic surveys and provide operational support for use of the DICE systems. It will also begin developing functionality to support onboarding additional surveys in subsequent years. Finally, the program will expand the use of ingest capabilities for third-party and administrative data.”
    • Implementing the Evidence Act “to increase research at the Census Bureau, support more complex, multi-agency, large dataset projects, and bring new types of researchers to the Census Bureau, including those new to research and in need of mentoring, and an initiative to improve the Census Bureau’s ability to measure the impact of the environment and natural disasters on people and economy.”
    • Advancing “software engineering and data science applications at the Census Bureau” and continuing “research on improving data collection methods.”

The President’s budget anticipates a drop in full-time employees (FTEs) at the Census Bureau from 2,926 in FY 2023 to only 2,477 in FY 2024.

Congressional Dear Colleague Letters Circulating In Support of FY 2024 Funding For Census Bureau

Dear Colleague letters in support of Census Bureau funding are circulating in the U.S. House of Representatives and U.S. Senate for Fiscal Year (FY) 2024:

  • In the House, Rep. Judy Chu (D-CA-28)’s letter urges House appropriators to “prioritize as much funding as possible for the U.S. Census Bureau.” Signatures from other House offices are due by March 29.
  • In the Senate, Sen. Brian Schatz (D-HI)’s letter urges Senate appropriators to “prioritize funding for the U.S. Census Bureau by providing the agency with $2 billion.” Signatures from other Senate offices are due by March 31.

These letters are important opportunities for members of Congress to communicate their support for funding the Census Bureau to the Chairs and Ranking Members of the House and Senate Commerce, Justice, Science (CJS) Appropriations Subcommittees. The CJS subcommittee has jurisdiction over funding for the Census Bureau.