2023-06064. Population Estimates Challenge Program

AGENCY:

Census Bureau, Department of Commerce.

ACTION:

Final rule.

SUMMARY:

The Bureau of the Census (Census Bureau) amends the regulations for the Population Estimates Challenge Program which will provide eligible general-purpose governmental entities (local governments) with the opportunity to file requests for the review of their population estimates for 2021 and subsequent years in forthcoming estimates series, beginning with the Vintage 2022 series that is scheduled to be published in 2023. Under this program, a governmental unit may file a challenge to its official population estimate by submitting additional data to the Census Bureau for evaluation, or by identifying a technical error in processing input data or producing the estimates. Specifically, the Census Bureau amends its regulations to update the regulation's references pertaining to the input data which are used to produce the official population estimates and revise the evidence required to support a challenge. In this final rule, the Census Bureau responds to comments received during the public comment period—closed on December 22, 2022—on the notice of proposed rulemaking posted in the Federal Register pertaining to ways in which the Population Estimates Challenge Program might be improved.

DATES:

This final rule is effective on April 24, 2023.

FOR FURTHER INFORMATION CONTACT:

Requests for additional information should be directed to Amel Toukabri, Chief, Local Government Estimates and Migration Processing Branch, Population Division, 301-763-2461 or [e-mail].

SUPPLEMENTARY INFORMATION:

Background

The Census Bureau typically releases annual population estimates, in accordance with Title 13 of the United States Code (U.S.C.). These estimates are typically based to some extent upon the most recent Decennial Census of Population and Housing and compiled from the most current administrative and survey data available for that purpose. Although not required by any statute, the Census Bureau also typically offers an opportunity for local units of general-purpose government (hereinafter collectively “governmental unit”) to challenge these official estimates through its Population Estimates Challenge Program. Under this program, a governmental unit may challenge its population estimate by submitting additional data to the Census Bureau for evaluation, or by identifying a technical error in processing input data or producing the estimates. If the additional data are accepted during the review period by the Census Bureau, resulting in an updated population estimate, the Census Bureau will provide a written notification to the governmental unit and publish the revised estimate at www.census.gov. If the additional data are not accepted for a revised estimate, the Census Bureau will notify the governmental unit. In the challenge process, the Census Bureau will only accept a challenge when the evidence provided indicates the use of incorrect data, processes, or calculations in the estimates.

In this final rule, the Census Bureau amends its regulations to: (1) update the regulation's references pertaining to the input data which are used to produce the official population estimates, and (2) revise the evidence required to support a challenge.

The Census Bureau also solicited comments from the public about ways in which the program might be improved. In particular, the Census Bureau welcomed comments about (1) the methodology used in preparing the annual Population Estimates, (2) the sources of data that the agency considers (or does not consider) in preparing the annual Population Estimates, and (3) what sorts of factual or methodological arguments the agency considers (or does not consider) in evaluating a potential challenge.

Based on the public comments received, the Census Bureau is amending its final rule to: (1) retain the flexibility to accept a physical copy of challenge materials rather than exclusively accepting digital challenge submissions; and (2) increase communication with localities by encouraging the appropriate Federal State Cooperative for Population Estimates (FSCPE) members to serve as conduits with local governments in the review of pre-release estimates, to the extent that this is possible given data confidentiality requirements for pre-release data. Furthermore, to the extent that research findings indicate that additional recommendations/changes support the development of accurate estimates and ensure equity for all general-purpose governmental units, the Census Bureau is open to expanding the scope of the Challenge Program. Nonetheless, this is contingent on the Population Estimates Program (PEP)'s programmatic capacity, future research priorities, and the outcome of such research. For the reasons explained below, at least at this time, the Census Bureau is not implementing the comments that advocated for an expansion of the Program ( e.g., the use of more flexible methodology, a wider range of data sources to support a challenge, and removal of the county control). The Census Bureau intends, however, to study more closely the issues raised in those comments, and commits to conducting future research that will allow for regular enhancement of the current methodologies, and which would improve both the accuracy of the population estimates and the Challenge Program.

Currently, the Census Bureau begins the process of preparing population estimates by updating population information from the most recent decennial census and other sources with information found in the annual administrative records of Federal and State agencies. The Federal agencies provide tax records, Medicare records, and some vital records and group quarters information. The FSCPE members, designated by their respective governors to work in cooperation with the Census Bureau's Population Estimates Program to produce population estimates, also supply vital statistics and information about group quarters like college dorms or prisons.^[1] The Census Bureau combines census base data, administrative records, and selected survey data ( e.g., data from the American Community Survey, American Housing Survey, and the Building Permit Survey) to produce current population estimates that usually begin with the last decennial census. Additionally, the Census Bureau's general-purpose governmental units' population estimates are provided to the FSCPE agencies in preliminary form for review and comment to resolve data processing issues identified during that period. For the purposes of this program, the District of Columbia is treated as a statistical equivalent of a county and, therefore, is eligible to participate.

A major priority for the Census Bureau is balancing the need to use the 2020 Census counts at the lowest level of estimates geography as the starting point in estimates production with the statutory obligation to protect the respondents' confidentiality at every stage of the data lifecycle. Since the 1990 Census, the Bureau has added “noise”—or variations from the actual count—to the collected data to ensure privacy and confidentiality. For 2020 Census data, the Census Bureau applied noise using a newer disclosure avoidance technique based on “differential privacy.” ^[2] The Census Bureau uses a housing unit method to distribute a county population to places within its legal boundaries. The components in this method include housing units estimates, average household population per housing unit, and an estimate of the population in group quarters. The estimation formula was simplified to increase the accuracy of the estimates and minimize the impact of differential privacy on the population estimates by reducing the number of components requiring privacy protection used to generate population estimates. Consequently, the occupancy rate and Persons Per Household (PPH) previously used in this method were replaced with the average household population per housing unit. The household population and the group quarters population used in the calculation of the estimate are the only two components subject to differential privacy protection compared to the prior three components—occupancy rate, PPH, and group quarters population—that would have otherwise required privacy protection. Therefore, the PPH and occupancy rate components are no longer inputs used to produce those population estimates. The distributive housing unit equation used to calculate the population estimates for governmental units is simplified to accommodate the application of the disclosure avoidance technique prior to releasing the estimates. As a result, the Census Bureau amends 15 CFR part 90 to revise: (1) the regulation's references pertaining to the input data which are used to produce the official population estimates, (2) where to file a challenge, and (3) the evidence required to support a challenge. These changes are captured in the updates to §§ 90.2, 90.7, and 90.8.

Previously, the Census Bureau published a final rule on January 9, 2020, in the Federal Register (85 FR 1100) to announce that the Census Bureau would temporarily suspend the Population Estimates Challenge Program to accommodate the taking of the 2020 Census and subsequent review and evaluation activities. Efforts to resume the program were delayed in response to the changes to the operational schedule for the 2020 Census which occurred due to the impacts of the COVID-19 pandemic. The Census Bureau followed the suspension of the Population Estimates Challenge Program with two Federal Register documents published on November 22, 2022. A document was posted in the Federal Register (87 FR 71240), entitled “Resumption of the Population Estimates Challenge Program,” which resumed the suspended program to provide eligible entities the opportunity to file requests for the review of population estimates for 2021 and subsequent years beginning with the Vintage ^[3] 2022 series that is scheduled to be published in 2023. That rule also made clear that challenges to previous estimates series (for which the 90-day limitations period set forth in 15 CFR 90.6 had long since elapsed) would not be accepted. The resumption document did not implement revisions to the program or its requirements. On November 22, 2022, the Census Bureau published another notice of proposed rulemaking in the Federal Register (87 FR 71269) for its program, entitled “Resumption of the Population Estimates Challenge Program and Proposed Changes to the Program.” In that announcement, the Census Bureau solicited comments from the public about ways in which the program might be improved.

The proposal was available for comment during a 30-day period that ended on December 22, 2022. The Census Bureau has now reviewed these comments and responded to them in this final rule.

Summary of Comments and Responses

The Census Bureau received 13 comments during the comment period pertaining to 11 separate topics. The contents of these recommendation received can be categorized according to a few main themes: (1) improving the Population Estimates Challenge process; (2) allowing the use of more flexible methodologies and a wider range of sources of data in both preparation of the annual population estimates and in challenge submissions, while increasing experts' participation such as through additional involvement from the FSCPE; and (3) investing in continuous research to improve the accuracy of the population estimates with a focus on the estimates base and the group quarters population. A summary of these comments and the detailed responses by the Census Bureau are provided below:

Comments Theme: Population Estimates Challenge Program Rulemaking

Topic 1. An Expansion of the Scope of the Challenge Program

Several commenters favored a challenge program that actively encourages participation and is open to considering a wider variety of data sources than is currently accepted. Many suggested that the program accept data sources typically available to or curated by localities, such as (but not limited to) electric utilities, address lists, public school enrollment data, and local property tax records. Yet, these commenters also noted that an expanded challenge program will stretch the capacity of PEP.

Commenters also acknowledged that revisions based on local government input or alternative methodologies do not always improve estimation accuracy, and so standards need to be maintained for accepting data in support of challenges. As an example of such criteria, it was recommended that the Census Bureau consider whether the alternative population estimate is developed by “a methodology and/or data set that appears in and is used consistently within the applied demography literature?” Then, in instances where that is not the case, “does the locality provide research on the validity of that data set and why this methodology works better, and do applied demographers (at the Census Bureau or elsewhere) agree with the findings?”

Even if a revision improves accuracy, commenters cited the potential for it to introduce other issues, such as questions of equity: challenges are issued seeking a higher population estimate, but not all jurisdictions have the resources to file a challenge and identify the necessary supporting evidence. Thus, governments with the resources to challenge could be in a position to receive greater shares of population and corresponding funding, possibly to the detriment of jurisdictions that do not challenge, perhaps due to fewer resources ( e.g., many localities lack the technology, infrastructure, and/or expertise needed to compile, analyze, and present data in a manner that meets the foregoing requirements). Commenters emphasized that an expanded challenge program needs to be mindful of such concerns.

Response 1.

The Census Bureau recognizes that potential expansion of the scope of the Population Estimates Challenge Program could be beneficial, although specific changes are contingent on PEP's future research priorities and findings. Furthermore, PEP not only maintains that methodologies considered should be consistent with or advance applied demography literature, but also acknowledges the efficacy of engaging and sharing findings with outside experts to enhance PEP's challenge decision-making process. As we define and progress in research impacting the Challenge Program, this will be a priority for PEP.

PEP is open to expanding the scope of the Challenge Program where science indicates that such changes support more accurate estimates and ensure equity for all general-purpose governmental units and the public. The Census Bureau is also considering alternative data sources, including administrative records, and methodologies for estimates production. This work is being led by the Base Evaluation and Research Team (BERT), which is tasked with researching the feasibility of taking coverage measures and/or administrative data into account in the development of the estimates base. Accordingly, the Census Bureau will explore the issues raised in the comments more closely in the future and will continue to consider possible ways to improve the Challenge Program. Until the results of these efforts suggest revisions to our approach, we foresee no changes in response to these comments at this time.

Topic 2. Challenge Process Recommendations

Comment 2.1  One commenter expressed concern that the proposed changes to the Challenge Program would require that jurisdictions file challenges solely through email and would eliminate the option of submitting a “hard copy” of challenge materials through the U.S. postal service or some similar delivery option. The commenter urged the bureau to retain the latter option on behalf of localities that still lack email access through broadband technology and are thus unable to electronically send large document files quickly and easily— particularly in remote or rural areas of the country.

Response 2.1  The Census Bureau will still maintain the ability to receive a hard copy of challenge materials through the U.S. postal service, recognizing the need to retain the option of submitting a physical copy, rather than exclusively accepting digital submissions.

Comment 2.2  Many commenters commended the Census Bureau for explicitly specifying in the proposed regulations that a phone number and email address will be provided for questions that localities may have about the Challenge Program. They also urged the Census Bureau to ensure that there is sufficient staffing to provide prompt responses through either of these modes to inquiries made by localities.

Response 2.2  The Challenge Program staff in PEP prioritizes the experience of localities and the challenge process and routinely responds to inquiries ( i.e., email, written requests, or telephone calls) in a timely manner. Additionally, as part of our commitment to continuous improvement, we will seek opportunities to further streamline the process of responding through all available modes.

Comment 2.3  Several commenters indicated that a more robust and improved population estimates challenge process would likely result in a higher volume of requests. Thus, the commenters stated that the Census Bureau must provide the necessary resources, including adequate staffing, to meet this need. In any event, the Census Bureau should not be in a position to use lack of staffing or staffing hours as a reason for limiting appeals moving forward.

Response 2.3.  The Census Bureau concurs that increases in the challenge process will lead to incremental stresses on existing capacity. Consistent with our continuous improvement activities, we are reviewing both the response process and resource flexibilities to facilitate the processing and turnaround time of a challenge while retaining the integrity of the challenge review process.

Comment 2.4  Numerous commenters also recommended a “change in nomenclature” to replace Challenge Program, which described as an “adversarial term.” To represent that successful challenges are the result of a cooperative partnership with local stakeholders to improve population estimates, a term that reflects the spirit of cooperation engendered by the new program should be used. One such suggestion was “Improvement Program.”

Response 2.4  The Census Bureau acknowledges that successful challenges should be properly viewed as the result of a cooperative partnership with local stakeholders to improve population estimates. However, the current name of the Population Estimates Challenge Program encompasses a longstanding relationship and history, and so alternatives—as well as the best means to potentially transition to a new name for the program—must be carefully considered. For example, a name change might make it difficult for local governments to easily find the necessary information if they are interested in challenging their population estimates, particularly the ones with fewer resources. We will include this possibility in future discussions with stakeholders and the public to determine the feasibility and benefit of this proposed change.

Comment 2.5  Commenters were also concerned that the 30-day comment period on the proposed regulations was too short, preferring at least 60 days for comment on an issue of this significance, arguing that a longer comment period would have enhanced the quality of feedback and helped demonstrate a more consistent approach to advancing stakeholder engagement.

Response 2.5  Although a longer comment period would have been more convenient for some, it was not possible to extend the comment period without jeopardizing PEP's ability to process and respond to comments received and subsequently update program materials accordingly in advance of the release of the upcoming Vintage 2022 county population estimates, which will be subject to challenge within 90 days of their release.

Comment 2.6  Other commenters stated that with more resources and by embedding the call for feedback on methodology and data sources into the challenge program, there would be more opportunities to raise awareness of alternative methods and data sources and implement methodology changes to improve the estimates program.

Response 2.6  The Census Bureau recognizes the importance of providing a mechanism for methodological feedback and input on data sources. So as to enable the Challenge Program staff to focus their time on processing challenges according to the program guidelines in place at that time, the Census Bureau is ensuring that this type of feedback may be shared via other means. In particular, BERT has created a dedicated email address, [e-mail], to provide stakeholders with an avenue for sharing ideas relating to alternative data sources or methodologies. This email address is currently active and will be advertised to localities as a destination for data and methodology suggestions relevant to their specific area.

Comment 2.7  Multiple commenters recommended that the FSCPE members more directly serve as conduits with local governments in the review of estimates in a pre-release format and to coordinate challenges. To more effectively have direct rather than secondary input into the production and review of the data, it was stated that FSCPE State representatives should have Special Sworn Status.

Response 2.7  The Census Bureau has already been actively consulting with the FSCPE member agencies regarding the Challenge Program. Additionally, PEP encourages the appropriate FSCPE members to serve as conduits with local governments in the review of pre-release estimates, to the extent that this is possible given data confidentiality requirements for pre-release data. A Memorandum of Agreement governs the partnership between the Census Bureau and the State agencies. The current agreements are set to expire in 2024, at which point it is anticipated that the agreements will be revised and renewed. During that renewal process, PEP will initiate discussions about the feasibility, expectations and responsibilities of the Census Bureau and the FSCPE members pertaining to annual data review.

Comment 2.8  Many of the commenters recommended that the Census Bureau keep external partners apprised of challenge requests that are occurring, decisions that have been made on challenges, and areas of concern about the challenge process on a regular basis, suggesting that this information sharing occurs through presentations to the Census Scientific Advisory Committee (CSAC). They also advised the Census Bureau to increase its communications about the challenge process to be more inclusive of all governmental units (especially small towns and cities), supporting the recent recommendation by the CSAC that the Census Bureau conduct webinars on the ability of local governments to improve statistics by partnering with the Census Bureau in the Population Estimates Challenge Program and Special Censuses.

Response 2.8  The Census Bureau concurs with this recommendation. PEP strives to make timely information readily available to its State partners in the FSCPE via the regular monthly meetings of the Steering Committee, Research and Methods Subcommittee, and Data Input Subcommittee, in addition to the twice-yearly meetings with the full membership. It has been the practice of the Challenge Program to keep the FSCPE members aware of any challenge requests in their States and to officially share the challenge outcomes. We plan to continue with this practice and further expand the outreach to other interested stakeholders, such as the CSAC via briefings and presentations. Another venue PEP is exploring to improve communication is the development of a video which would walk local governments through the process of submitting a challenge, and which would be supplemental to the Population Estimates Challenge Program Guide already made available on its website ( www.census.gov).

Topic 3. Subcounty Estimation Formula and Updates to the Persons-per-Household and Occupancy Rates

At the subcounty level, numerous commenters argued that the Census Bureau's recent decision to eliminate vacancy [occupancy] and person-per-household (PPH) rates by combining them into “average population per housing unit” is inconsistent with the literature on how to produce accurate population estimates. They recommended that the Census Bureau reverse this decision.

One commenter urged the Census Bureau to ensure that the replacement in the estimation formula of the PPH and occupancy rate components with the average household population per housing unit does not have a detrimental impact on the accuracy of estimates produced for localities with undercounted populations and inform the public of analyses that demonstrate that outcome.

Additionally, several commenters recommended that the Census Bureau should allow cities and other governmental units to challenge the April 1, 2020 population base used for the annual estimates with all relevant, reliable data, especially with respect to housing occupancy/vacancy and PPH rates. Furthermore, many commenters recommended that the Bureau should accept revisions to these components as part of the Challenge Program if a locality can provide reasonable and sufficient evidence of change.

The recommendations suggested some methods that the Census Bureau may research in order to develop post-Census Day PPH and occupancy components, such as the expanded, modeled, or indexed use of IRS filer and exemption data at the sub-county (place or minor civil division) level; the use of local street address listings or local annual town Censuses, where available; USPS data; and the use of other State or local administrative records, including school or program enrollment information; or other high-quality data sources.

Response 3.

PEP combined the occupancy and PPH components to adhere to the Census Bureau's modernized disclosure avoidance requirements which are designed to protect the confidentiality of respondents. By combining PPH and occupancy into the average population per housing unit, the number of terms in the distributive housing unit equation subject to the application of differentially private noise was minimized, subsequently minimizing the impact of the noise on the estimates and maximizing their overall accuracy.

Furthermore, whereas PEP recognizes that the “ average household population per housing unit” ratio may not be a standard demographic measure, the new formula is mathematically equivalent to the old version where the “ Occupied Housing Units total” in the first numerator and second denominator in line (2) shown below cancel each other out. Thus, the formula in use is a simplified version of the previous formula which no longer requires housing characteristics measures such as the occupancy and PPH. Replacing the PPH and occupancy rate with the “ average household population per housing unit” does not structurally change the formula; therefore, the replacement will not introduce additional error to the population estimates.

Where:

SUBCO RESPOP_t: Subcounty resident population total at time t

HU_t: Housing unit total at time t

OCC Rate_{Base year}: Occupancy Rate at base year

HHPOP_t: Household population total at time t

GQPOP_t: Group quarters population total at time t

Consequently, there is no foundation for the commenter's concern as both the previous and updated equations result in the same subcounty household population value for a specific subcounty area.

Although occupancy rates and PPH no longer factor into the calculation of subcounty population, PEP will consider conducting research on whether local PPH and occupancy data may be submitted in a subcounty challenge as an alternative to the 2020 Census household population per housing unit ratio used in the distributive household equation. Pending the research findings, this could be a potential mechanism to challenge the data in the April 1, 2020 population estimates base. Beyond that, the possibility of challenging the estimates base at the subcounty level would be contingent on identifying alternative sources of data which were of sufficient quality to serve as replacements for the other population components drawn from the Census results.

At the county level of geography and above, it is not feasible to break down the base population into challengeable components, as it is created by integrating results from the 2020 Census, 2020 Demographic Analysis estimates, and Vintage 2020 estimates. As such, challenges to the population estimates base must remain out of scope.

The Census Bureau appreciates the research ideas contributed by the commenters. To enhance the accuracy and reliability of the subsequent estimates, and to contribute to a longer-term goal of continuous improvement in the estimation processing, PEP will explore the practicability of the suggested research topics.

Topic 4. Re-Evaluate the Use of the County Control and Revisit HUBERT Research

Numerous commenters recommended that the Census Bureau re-evaluate the use of the county control (or what some commenters referred to as a “county cap”) when processing sub-county population estimates challenges. Some argued that adhering to the control creates a situation whereby governments with the resources to successfully challenge receive greater shares of population and resultant funding, to the detriment of jurisdictions that do not challenge, perhaps because they have fewer resources.

Various commenters suggested that research led by the Census Bureau's Housing Unit-Based Research Estimates Team (HUBERT) in 2007 to 2008 be revisited and updated. This research program assessed whether, for some counties, a housing unit-based method of calculating population change at the county level was more effective than the cohort-component method used by the Census Bureau. The commenters highlighted how the Census Bureau has relied on the HUBERT research for many of its current decisions about data and methods used in the production of its annual population estimates.

Pending an update of the HUBERT research, several commenters recommended that the Census Bureau utilize the findings from the original HUBERT research to:

• identify the 30% of counties that were more accurately estimated using HUM than the cohort-component approach;
• identify the characteristics of the counties that are better estimated by the HUM; and
• use that information to classify all counties and apply the method that is more accurate for each county type.

Response 4.

The Census Bureau recognizes that more current, in-depth research is needed to reevaluate the use of the county control for the incorporation of successful challenges. While no changes in response to this comment are being made at this time, PEP plans to update HUBERT research with current data and examine the impact of the county control to inform future changes to the methodology. When possible, PEP intends to research the feasibility and logistics of this change to ascertain if it is methodologically sound, including soliciting input from our FSCPE partners.

Comments Theme: Population Estimates Program

Topic 5. Investing in the Population Estimates Program and Ongoing Stakeholder Input

Numerous commenters stated that PEP has limited capacity to execute much needed, updated research. Therefore, they proposed an expansion and additional investment in PEP and in the FSCPE partnership. Many commenters urged the Census Bureau to make improving the population estimates a high priority and work to increase the resources necessary by reallocating or requesting additional resources to support a continuous year-round estimates research program throughout the decade.

Several commenters also advised to build and maintain collaborative relationships throughout the decade with State, local and Tribal governments to take in additional anonymized official datasets to improve the estimates, and to seek ongoing stakeholder input, both from government entities as well as from other organizations.

Many commenters recommended that the Department of Commerce provide a modest level of direct funding to every FSCPE representative so that every State can participate. At present, the effort is defined by a Memorandum of Agreement under which some States fund their participation, while others do not. This arrangement leads to some States participating actively in the FSCPE and others participating at extremely low levels or not at all.

Response 5.

The Census Bureau concurs with the importance of ensuring continuous research on population estimates related topics throughout the decade. In fact, planning and conducting prioritized research on an annual basis are ingrained in PEP's mission and construct a vital phase typically carried out after the conclusion and release of population estimates series, and before the new estimates production cycle starts. This yearly research cycle has been and will continue to be a priority for PEP, allowing for regular enhancement of the current methodologies used to improve the population estimates' accuracy, while approving a limited number of research topics to work on annually that are manageable by PEP and defer the remaining list of research topics to future years.

The Census Bureau acknowledges the value and significance of the partnership with FSCPE State agencies and their role in producing high-quality estimates products. The Census Bureau also concurs that it would be beneficial to engage with the FSCPE partners to explore creative and effective means to benefit from their local knowledge and suggestions, beyond what is currently done. PEP believes that the efficacy of the partnership is enhanced by robust participation across the country.

PEP continues to explore ways to enhance outreach and increase current States' participation through the FSCPE partnership to provide local data inputs that are consistent with PEP's current methodologies for use in the annual estimates production such as housing unit components of change, vital statistics records, and group quarters reporting. For example, PEP will collaborate with more active States and seek input from less active States to identify options to encourage more participation. Additionally, PEP has created an email address associated with the BERT research— [e-mail] —which could be advertised to localities as a destination for data and methodology suggestions relevant to their specific area.

Topic 6. Ways To Improve the Population Estimates

Some commenters recommended that the Census Bureau consider more flexible methodologies ( e.g., allow for probabilistic modeling in addition to demographic accounting methods) and broader use of administrative data to ensure meaningful opportunities to improve the accuracy of the estimates including appropriate improvements to the estimates base. Several commenters specifically recommended that “the Census Bureau move from using just one method to estimate the total population of States and counties (cohort component method), to using multiple methods to produce the State and county estimates more accurately.” They also suggested that “the Census Bureau allow targeted, localized methods that do not apply to the entire country if they improve local accuracy.” At the subcounty level, some commenters recommended the Census Bureau use its new research on Construction Starts  ^[4] based on artificial intelligence and satellite imagery as inputs for the number of housing units in HUM ^[5] estimates.

Many commenters recommended that the Census Bureau conduct research on the efficacy of Internal Revenue Service (IRS) return data in reflecting overall migration patterns. They also suggested “researching the use of United States Postal Service Change of Address data for permanent moves to be incorporated into the estimated migration rates.”

Response 6.

The Census Bureau's ability to implement flexible/multiple methodologies and the use of a broader array of data sources is contingent upon two major components: first, ensure the capacity to conduct research on the proposed methodology and adhere to PEP's scientific and methodological principles; and second, the outcome of the methodological research. New methods must be found to be sound (based on solid reasoning respectful of the attributes of the input data as they relate to the estimation tasks), accountable (understandable and replicable), robust (insensitive to small departures from assumptions, reasonably accurate under changing demographic conditions), parsimonious (reflecting a simpler strategy versus a more complex one whenever possible), and to produce valid results.

Given the vast range of individual geographies for which PEP produces estimates, we prefer methods and data that can be applied to entities across a geographic or multiple geographic levelsbased on the principle of parsimony and the accuracy and equity of the results. This is an important distinction because it underscores how differential methods across a geographic level are significantly labor-intensive to incorporate and require substantially longer time so as to enable PEP to research and test alternative methods and data to ensure equitable accuracy of population estimates across geographies).

PEP will research the feasibility and logistics of alternative methods, including investigating how new and current research taking place at the Census Bureau, such as the efforts underway to modernize construction indicators, can be leveraged to improve the accuracy and reliability of the estimates. Accordingly, although no changes in response to this comment are being made at this time, the Census Bureau will continue to conduct research and consider possible ways to improve the Challenge Program and population estimates.

Assessing the quality of the data is of the utmost importance in PEP's estimates production cycle. Therefore, PEP first evaluates time series of IRS filing statistics to identify any data quality issues that need to be addressed. PEP already makes use of the United States Postal Service (USPS) National Change of Address (NCOA) data as a benchmark to assess the quality of the IRS-based migration rates, and to validate permanent moves. Additionally, PEP compares migration trends between IRS and NCOA data to capture changes in domestic migration patterns across the country and particularly in disaster-hit counties.

Topic 7. The Estimation of Group Quarters Population

Some commenters specifically supported PEP's ongoing research on alternative methods and data sources as it pertains to the estimation of group quarters populations. One of the commenters echoed the recent recommendation by the CSAC that the Census Bureau should collect group quarters lists by individual facility and include capacity and attendance information from FSCPE members for the estimates base and throughout the decade.

Response 7.

The Census Bureau welcomes and concurs with the suggestion of continuous research on alternative methods and data sources for the estimation of the group quarters population. For instance, the 2020 Post-Census Group Quarters Review (PCGQR) operation—unique to this decade—was created in response to public feedback to improve the counts of specific GQs. This program improves the accuracy of the GQ population in the estimates base: if the PCGQR review process finds discrepancies in these population counts supported by sufficient documentation, approved revisions to the group quarters population are provided to PEP. These updates are incorporated into the base population for upcoming vintages of estimates, as the production schedule allows.

Additionally, the Census Bureau concurs with the recommendation to coordinate with the FSCPE regarding contents of their future Group Quarters Report data that they provide to PEP on an annual basis.

Topic 8. Re-Evaluate the “College Fix” in Estimates Production

Referencing the current methodology for the annual population estimates, two commenters encouraged the Census Bureau to re-evaluate the criteria used for “College Fix” counties. PEP's application of a “college fix” in the estimates is used to improve the estimates for counties with high shares of college-enrolled population, which would otherwise erroneously be “aged forward” within the cohort-component methodology. Instead, the “college fix” allows this population to be replaced each year by the newly incoming students, producing a more demographically reasonable age structure for that population. The commenters note that this adjustment is particularly needed because the sources used to directly capture migration in the estimates, namely IRS data, have proven less effective for capturing the migration of college-aged cohorts—“a phenomenon that is easily demonstrated by looking at Census-to-Census survival of these cohorts as versus populations estimated using IRS data.”

Response 8.

The Census Bureau concurs with the recommendation to re-evaluate the College Fix criteria and overall method to ensure the reasonableness of the resulting population estimates for affected counties. Therefore, to contribute to a longer-term goal of continuous improvement in the estimation processing, PEP will seek to include this research on the College Fix.

Comments Theme: Coverage and Improving Census Base Population

Topic 9. Inaccurate Estimates Due to Undocumented Immigration

The commenter stated that “the U.S. Census will continue to be inaccurate as long as there is uncontrolled illegal immigration due to the wide-open southern border,” referencing “thousands of `got-aways' in 2022 alone.” The comment asserts that these individuals secure alternative living situations which result in their omittance from the census count. Additionally, the commentor maintains that the U.S. Census limits data collection to “safe suburban environments” to the exclusion of “urban, violent areas,” leading to inaccurate population counts for cities.

Response 9.

The goal of the Census Bureau is a complete and accurate census. The U.S. Constitution requires the census to count every resident in the nation. During the 2020 Census, the Census Bureau went to great lengths to count everyone, including people residing in housing units, including apartments and mobile homes; people in complex living situations; people who live or stay in a group living arrangement; and those experiencing homelessness. This included counting people where they received services, outdoors, and at other locations where they are known to sleep. The Census Bureau understood that many noncitizens were fearful that participating in the census could expose them and their families to harm, so the bureau continued working with trusted voices in local communities to encourage people to participate. We also hired locally, and our staff collectively spoke dozens of languages.

The Census Bureau's Community Partnership and Engagement Program (CPEP) had 18 distinct initiatives that further enhanced focus on historically hard-to-count populations. One of those initiatives was the Foreign-Born and Immigrant Program. The CPEP specialists were placed locally on the basis of low-response score and population density. The specialists focused on local engagement and outreach, and specialized in languages specific to historically hard-to-count populations in their local community. The 2020 Census was the first census where everyone could respond online, by phone, or by mail. Census takers made in-person visits to every household that did not respond via one of these methods to make sure people who lived there were counted. These visits were made across the country, in all of the urban, suburban, and rural areas where people live. The Census Bureau also conducted a robust Integrated Communications Campaign to reach everyone living in the United States with information on how the 2020 Census was easy, safe, and important. The cornerstone of this effort was a research-based communications campaign that covered all levels of geography.

As a result, the findings from the Census Bureau's official coverage evaluations indicate that young children aged 0 to 4, the Black or African American population, the American Indian and Alaska Native population—especially on reservations—and the Hispanic population were likely undercounted in the 2020 Census. The Census Bureau takes these findings very seriously and is working to mitigate these issues in the 2030 Census. For example, in 2022, the Census Bureau formed the Young Children Working Group, which focuses on the coverage of young children and improving data on this population. In addition, the Census Bureau formed a separate working group in 2023 focused on researching ways to improve the coverage of other Historically Undercounted Populations (HUPs). The Census Bureau is actively conducting outreach to stakeholders, partners and community organizations to expand and strengthen a trusted messenger ecosystem across the nation.

Topic 10. Persistent Undercounts in 2020 Census, Misallocation of Federal Funding, and Improving the Census Base

Several of the comments were related to the issue of correcting undercounts that persisted in the 2020 Census, emphasizing that undercounts misdirect Federal and State funding. Although the April 1, 2020 population estimates base has been identified as a possible mechanism for mitigating undercounts, many commenters acknowledged that the expectation for the estimates to compensate for coverage errors in a multi-billion-dollar census to achieve a fairer distribution of funds might be unrealistic.

Various commenters suggested that the Census Bureau should research the best ways to make coverage adjustments, determine the feasibility of incorporating administrative data sources while maintaining quality standards, and build in an opportunity for feedback before final decisions around the base population are made. Furthermore, numerous commenters expressed support for a continuation of the Census Bureau's efforts to research population base enhancements, but recommended the research examine possible adjustments at a more local level ( e.g., adjusting age distributions in the estimates base by county, rather than applying the same distribution adjustment equally to every county across the country). Additionally, a few commenters advised the BERT research to make use of existing administrative records files and the Census Bureau's Frames Program to improve the accuracy of baseline data. Most commenters strongly supported the Census Bureau's creation of BERT and expressed ongoing support of their work.

Other commenters noted that the 2020 Post-Enumeration Survey (PES), one of the Census Bureau's official coverage measures, is not sufficiently robust for adjusting undercounts at the local level. They recommended improving the estimates base by adjusting State-level counts using PES State-level results and then incorporating administrative data from programs such as Medicaid; the Supplemental Nutrition Assistance Program (SNAP); the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC); etc. to distribute the increased or decreased population to the counties within each State.

A collection of commenters also suggested that the total population in the April 1, 2020 estimates base no longer be restricted by the total population from the 2020 Census counts, due to the risk of underestimating children in some locations and inflating estimates of children in other locations, potentially at the expense of other age cohorts.

Response 10.

The Census Bureau appreciates the expression of support for BERT and the team's efforts to build the most accurate estimates base possible. BERT has specifically been formed to research the feasibility of taking coverage measures such as the PES, Demographic Analysis (DA) and other administrative records into account in the development of the estimates base. Moving forward, BERT's research findings will inform decisions about what 2020 Census data or administrative sources are used in the development of the estimates base, and whether there are adjustments that can be made to the Census data used in the base which could be applied in equitable, methodologically sound, and demographically plausible ways. As such, this research entails a careful evaluation of all data sources which would potentially be used to enhance or adjust the estimates base so that data (including the coverage measures from PES and DA) are not used in ways that extend beyond their design capacity. The research also includes assessing the impact of the blended estimates base on specific populations, such as the population of children, to ensure that the way the sources are combined in the estimates base produces the likeliest distribution by demographic characteristics.

Moreover, the BERT research includes collaboration with other administrative records-based projects underway at the Census Bureau, such as the demographic frame developed by the Frames Program. These joint efforts will lend insight into the suitability of administrative data sources for researching coverage issues. This includes the potential to target specific populations or geographies, including the possibility of developing differential adjustments at subnational geographic levels.

The BERT research is a prime example of how we are striving to mitigate coverage issues, and the work being undertaken by this team is a major priority for the Census Bureau.

Theme: Other Comments

Topic 11. Availability of BERT's Research Plans and Results

Commenters recommended that the Census Bureau make publicly available:

a. A detailed research schedule for each BERT subject matter component.

b. A detailed representation of BERT's short-, medium- and long-term goals and key decision points.

c. Its evaluations of how specific decisions on population and housing base adjustments impact final statistics for States and sub-State areas.

d. Methodological reviews solicited by the Bureau from external researchers on BERT, PEP, and any potential application of privacy protection impacting PEP.

Response 11.

The Census Bureau acknowledges the importance of transparency regarding the work of BERT to the extent possible given the nature of the research. We will be seizing upon promising findings as our research progresses, and this will vary from one specific approach to the next. As such, it is neither advisable nor prudent to adhere to a strict, detailed research schedule by subject matter component. The research process will evolve as findings and insights emerge. With this in mind, BERT has distinct plans to promote transparency and disseminate information. These include regular public briefings as well as a dedicated email address, [e-mail], which provides stakeholders with a mechanism for sharing ideas relating to data sources or methodology, or to request information.

Changes From Proposed Rule

The following are changes to the Challenge Program procedures resulting from the public comments received:

(1) One commenter requested that the Census Bureau provide for some flexibility in the rule to allow submission of a physical copy of challenge materials through the U.S. postal service or some similar delivery option, rather than exclusively accepting digital submissions. The Census Bureau acknowledged the issue and agreed to implement appropriate language in § 90.7 to address the request, and specify in the “Population Estimates Challenge Review Guide” a physical address where local governments could submit challenge materials to the Census Bureau for review and evaluation.

(2) Many commenters advised to allow the FSCPE members to serve as conduits with local governments in the review of pre-release estimates. The Census Bureau encourages the designated FSCPE agency in each State to serve in that role to the extent possible given data confidentiality requirements for pre-release data and has added specific language to § 90.9 to reflect the Census Bureau's intent.

(3) The Census Bureau amended § 90.9 to address local governments' demand for increased communications about the challenge process to be more inclusive of all governmental units (especially small towns and cities).

Summary of Provisions Implemented by This Final Rule

In November of 2022, The Census Bureau resumed the Population Estimates Challenge Program to provide governmental units the opportunity to challenge population estimates for 2021 and subsequent years in forthcoming estimates series, beginning with the Vintage 2022 series that is scheduled to be published in 2023. The Census Bureau now amends its regulations to: (1) ensure that the regulatory text more accurately describes how the Population Estimates Challenge Program has always functioned and is expected to function in the future; (2) update the regulation's references pertaining to the input data which are used to produce the official population estimates; and (3) allow the designated FSCPE agencies in each State to serve as conduits with local governments in the review of pre-release estimates, to the extent possible given data confidentiality requirements for pre-release data. These changes are captured in the proposed updates to §§ 90.2, 90.7, 90.8, and 90.9. At this time, the Census Bureau is making no technical changes to its regulations except in the sections noted below:

Sections 90.2 and 90.7—to ensure that the regulatory text more accurately describes how the Population Estimates Challenge Program has always functioned and is expected to function in the future. This proposed clarification does not reflect any operational changes.

Section 90.8—to update the challengeable components of change.

Section 90.9—to allow the designated FSCPE agencies in each State to serve as conduits with local governments in the review of pre-release estimates, to the extent possible given data confidentiality requirements for pre-release data.

The Census Bureau in Section § 90.2 revises its policy which is to provide the most accurate population estimates possible given the constraints of resources and available statistical techniques. It is also the policy of the Census Bureau, to the extent feasible, to provide governmental units the opportunity to seek a review of and provide additional data for these estimates and to present evidence relating to the accuracy of the estimates.

The Census Bureau in § 90.7 updates information about where to file a challenge for the governmental units that would like to initiate a challenge process after the population estimates are posted on the Census Bureau's website ( www.census.gov). A request for a population estimates challenge must be prepared in writing by the governmental unit and filed with the Chief, Population Division, Census Bureau by sending the request via email to [e-mail] or to a physical address that the Census Bureau will specify in the updated version of the “Population Estimates Challenge Program Review Guide” to be posted in the census.gov website. The governmental unit must designate a contact person who can be reached by telephone or email during normal business hours should questions arise regarding the submitted materials. In the event that a county-level governmental unit or statistical equivalent is not an active general-purpose government, the FSCPE member agency may serve as sponsor of the challenge and the governor will serve as the highest elected official.

The Census Bureau also amends § 90.8 by revising paragraphs (a), (c), and (d) that specify the evidence required for the challenge process. The types of data that are submitted must be consistent with the criteria, standards, and regular processes the Census Bureau employs to generate the population estimate. The Census Bureau will provide additional Web-based information describing the data that are required and how the governmental unit may contact the Census Bureau.

The Census Bureau in § 90.9 adds language to allow the designated FSCPE agencies in each State to serve as conduits with local governments in the review of pre-release estimates, to the extent possible given data confidentiality requirements for pre-release data.

The sections that feature changes to the regulations are noted as “Revised” in parentheses, for the public's convenience:

The following are the current sections of part 90 procedure for challenging population estimates.

Former	Effective April 24, 2023
PART 90 PROCEDURE FOR CHALLENGING POPULATION ESTIMATES	PART 90 PROCEDURE FOR CHALLENGING POPULATION ESTIMATES
90.1 Scope and applicability	90.1 Scope and applicability.
90.2 Policy of the Census Bureau	(Revised) 90.2 Policy of the Census Bureau.
90.3 Definitions	90.3 Definitions.
90.4 General	90.4 General.
90.5 Who may file a challenge	90.5 Who may file a challenge.
90.6 When a challenge may be filed	90.6 When a challenge may be filed.
90.7 Where to file a challenge	(Revised) 90.7 Where to file a challenge.
90.8 Evidence required	(Revised) 90.8 Evidence required.
90.9 Review of challenge	(Revised) 90.9 Review of challenge.

Regulatory Flexibility Act

The Regulatory Flexibility Act (RFA), as amended by the Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), 5 U.S.C. 601 et seq., generally requires an agency to prepare a regulatory flexibility analysis of any rule subject to the notice and comment rulemaking requirements under the Administrative Procedure Act (5 U.S.C. 553) or any other statute, unless the agency certifies that the rule will not have a significant economic impact on a substantial number of small entities. Under section 605(b) of the RFA, however, if the head of an agency certifies that a rule will not have a significant impact on a substantial number of small entities, the statute does not require the agency to prepare a regulatory flexibility analysis. Pursuant to section 605(b), the Chief Counsel for Regulation, Department of Commerce, submitted a memorandum to the Chief Counsel for Advocacy, Small Business Administration, certifying that this final rule will not have a significant impact on a substantial number of small entities.

Number of Small Entities

This final rule would impact only governmental units, some of which may be considered a small entity under the RFA. The RFA defines “small entity” as a small business, small organization, or small governmental jurisdiction. Specifically, the RFA defines “small governmental jurisdiction” as the government of a city, county, town, school district, or special district with a population of less than 50,000. Using this criterion, the Census Bureau estimates that around 37,000 small governmental jurisdictions would be impacted by this rulemaking.

Economic Impact

The Census Bureau does not anticipate any economic impact as a result of this final rule. This rulemaking intends to resume the implementation of the Population Estimates Challenge Program in 2023 to provide eligible entities the opportunity to file a challenge to population estimates for 2021 and subsequent years in forthcoming estimates series, beginning with the Vintage 2022 series that is scheduled to be published in 2023. There are no direct costs imposed on governmental entities (units) that wish to initiate a challenge under the Population Estimates Challenge Program.

Executive Orders

This rulemaking has been determined to be not significant for purposes of Executive Order 12866. This final rule does not contain policies with federalism implications as that term is defined in Executive Order 13132.

Paperwork Reduction Act

This final rulemaking does not contain a collection of information subject to the requirements of the Paperwork Reduction Act (PRA), 44 U.S.C., Chapter 35.

Robert L. Santos, Director, Census Bureau, approved the publication of this notification in the Federal Register .

List of Subjects in 15 CFR Part 90

• Administrative practice and procedure
• Census data
• Population census
• Statistics

For the reasons set forth in the preamble, Census Bureau revises 15 CFR part 90 to read as follows:

PART 90—PROCEDURE FOR CHALLENGING POPULATION ESTIMATES

90.1

Scope and applicability.

90.2

Policy of the Census Bureau.

90.3

Definitions.

90.4

General.

90.5

Who may file a challenge.

90.6

When a challenge may be filed.

90.7

Where to file a challenge.

90.8

Evidence required.

90.9

Review of challenge.

Authority: 13 U.S.C. 4 and 181.

§  90.1

Scope and applicability.

Between decennial censuses, the Census Bureau annually prepares statistical estimates of the number of people residing in States and their governmental units. In general, these estimates are developed by updating the population counts produced in the most recent decennial census with demographic components of change data and/or other indicators of population change. These rules prescribe the administrative procedure available to governmental units to request a challenge to the most current of these estimates.

§ 90.2

Policy of the Census Bureau.

It is the policy of the Census Bureau to provide the most accurate population estimates possible given the constraints of resources and available statistical techniques. It is also the policy of the Census Bureau, to the extent feasible, to provide governmental units the opportunity to seek a review of and provide additional data for these estimates and to present evidence relating to the accuracy of the estimates.

§  90.3

Definitions.

As used in this part (except where the context clearly indicates otherwise) the following definitions shall apply:

(a) Census Bureau means the U.S. Census Bureau, Department of Commerce.

(b) Population Estimates Challenge means, in accordance with this part, the process a governmental unit may use to provide additional input data for the Census Bureau's population estimate and the submission of substantive documentation in support thereof.

(c) Director means Director of the Census Bureau, or an individual designated by the Director to perform under this part.

(d) Population estimate means a statistically developed calculation of the number of people living in a governmental unit to update the preceding census or earlier estimate.

(e) A governmental unit means the government of a county, municipality, township, incorporated place, or other minor civil division, which is a unit of general-purpose government below the State.

(f) A non-functioning county or statistical equivalent means a sub-State entity that does not function as an active general-purpose governmental unit. This situation exists in Connecticut, Rhode Island, for selected counties in Massachusetts, and for the Census Areas in Alaska.

(g) For the purposes of this program, an eligible governmental unit also includes the District of Columbia and non-functioning counties or statistical equivalents represented by a FSCPE member agency.

§  90.4

General.

This part provides a procedure for a governmental unit to request a challenge of a population estimate of the Census Bureau. The Census Bureau, upon receipt of the appropriate documentation, will attempt to resolve the estimate with the governmental unit.

§  90.5

Who may file a challenge.

A request for a challenge of a population estimate generated by the Census Bureau may be filed only by the chief executive officer or highest elected official of a governmental unit. In those instances where the FSCPE member agency represents a non-functioning county or statistical equivalent, the governor will serve as the chief executive officer or highest elected official.

§  90.6

When a challenge may be filed.

(a) A request for a challenge to a population estimate may be filed any time up to 90 days after the release of the estimate by the Census Bureau. Publication by the Census Bureau on its website ( www.census.gov) shall constitute release. Documentation requesting a challenge of any estimate may also be filed any time up to 90 days after the date the Census Bureau, on its own initiative, revises that estimate.

(b) If, however, a governmental unit has a sufficiently meritorious reason for not filing in a timely manner, the Census Bureau has the discretion to accept the late request.

§ 90.7

Where to file a challenge.

A request for a population estimate challenge must be prepared in writing by the governmental unit and filed with the Chief, Population Division, Census Bureau by sending the request via email to [e-mail] or to a physical address that the Census Bureau will specify in the updated “Population Estimates Challenge Program Review Guide” to be posted in the census.gov website. The governmental unit must designate a contact person who can be reached by telephone or email during normal business hours should questions arise with regard to the submitted materials.

§  90.8

Evidence required.

(a) The governmental unit shall provide whatever evidence it has relevant to the request at the time of filing. The Census Bureau may request further evidence when necessary. The evidence submitted must be consistent with the criteria, standards, and regular processes the Census Bureau employs to generate the population estimate. Currently, the Census Bureau challenge process cannot accept estimates developed from methods different from those used by the Census Bureau. The Census Bureau will only accept a challenge when the evidence provided indicates the use of incorrect data, processes, or calculations in the estimates.

(b) For counties and statistical equivalents, the Census Bureau uses a cohort-component of change method to produce population estimates. Each year, the components of change are updated. These components include births, deaths, migration, and change in the group quarters population. The Census Bureau will consider a challenge based on additional information on one or more of the components of change or about the group quarters population in a locality.

(c) For minor civil divisions and incorporated places, the Census Bureau uses a housing unit method to distribute a county population to places within its legal boundaries. The components in this method include housing units estimates, average household population per housing unit, and an estimate of the population in group quarters. The estimation formula was simplified to increase the accuracy of the estimates following the application of differential privacy as per the Census Bureau's new disclosure avoidance framework. As a result, the persons per household (PPH) and occupancy rate components were replaced with the average household population per housing unit. Additionally, the Census Bureau will consider a challenge based on data related to changes in an area's housing stock, such as data on demolitions, condemned units, uninhabitable units, building permits, or mobile home placements or other housing inventory-based data deemed comparable by the Census Bureau. The Census Bureau will also consider a challenge based on additional information about the group quarters population in a locality.

(d) The Census Bureau will also provide a guide on its website as a reference for governmental units to use in developing their data as evidence to support a challenge to the population estimate. In addition, a governmental unit may address any additional questions by contacting the Census Bureau at 301-763-2461 or by sending emails to [e-mail] or by delivering mail to a physical address that the Census Bureau will specify in the updated version of the “Population Estimates Challenge Program Review Guide” to be posted in the census.gov website.

§  90.9

Review of challenge.

The Chief, Population Division, Census Bureau, or the Chief's designee shall review the evidence provided with the request for the population estimate challenge, shall work with the governmental unit to verify the data provided by the governmental unit, and evaluate the data to resolve the issues raised by the governmental unit. Furthermore, the designated FSCPE agencies are encouraged to serve as conduits with local governments in the review of pre-release estimates, to the extent that this is possible given data confidentiality requirements for pre-release data. Thereafter, the Census Bureau shall respond in writing with a decision to accept or deny the challenge. In the event that the Census Bureau finds that the population estimate should be updated, it will also post the revised estimate on the Census Bureau's website ( www.census.gov).

Dated: March 20, 2023.

Shannon Wink,

Program Analyst, Policy Coordination Office, U.S. Census Bureau.

Footnotes

1.   https://www.census.gov/​programs-surveys/​popest/​about/​fscpe.html.

2.  For more information about the differential privacy technique, visit Understanding Differential Privacy ( census.gov).

3.  Annually, PEP revises and updates the entire time series of estimates from April 1, 2020 to July 1 of the year for which the estimates are published which is referred to as the vintage year. The term “vintage” is used to denote the entire time series created with a consistent population starting point and methodology.

4.  Econ Current Surveys Update Construction Re-engineering ( census.gov).

5.  HUM: Distributive housing unit-based methodology, which is used by PEP to produce subcounty population estimates.