2024ACSSOGITest_Supporting Statement Part B 20240402

SUPPORTING STATEMENT B

U.S. Department of Commerce

U.S. Census Bureau

American Community Survey Methods Panel: 2024 Sexual Orientation and Gender Identity Test

OMB Control No. 0607-0936

B. Collections of Information Employing Statistical Methods

1. Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection method to be used. Data on the number of entities (e.g., establishments, State and local government units, households, or persons) in the universe covered by the collection and in the corresponding sample are to be provided in tabular form for the universe as a whole and for each of the strata in the proposed sample. Indicate expected response rates for the collection as a whole. If the collection had been conducted previously, include the actual response rate achieved during the last collection.

For the 2024 American Community Survey (ACS) Sexual Orientation and Gender Identity (SOGI) test, the U.S. Census Bureau plans to sample about 480,000 housing unit addresses. This test will exclude addresses in Remote Alaska, Puerto Rico, and group quarters facilities. The universe of eligible housing unit addresses comes from the Census Bureau’s Master Address File (MAF). For more information about the MAF and how it is used for ACS sampling, see Chapter 3 of the ACS Design and Methodology Report (census.gov).

About 476,000 of the sampled addresses are expected to be mailable by the U.S. Postal Service. To solicit self-response, the Census Bureau sends up to five mailings to potential respondents. For addresses that were mailed survey materials but did not respond by mail, internet, or by calling our telephone questionnaire assistance line, the Census Bureau selects a subsample of all households and assigns them to the nonresponse follow-up data collection mode. Unmailable household addresses are sampled and also included in the nonresponse follow-up data collection operation.

This approach to solicit survey response is the same as is used in the production ACS. In 2022, the final weighted response rate for ACS was 84 percent. We expect this test to achieve a similar response rate.

2. Describe the procedures for the collection of information including:

• Statistical methodology for stratification and sample selection,

• Estimation procedure,

• Degree of accuracy needed for the purpose described in the justification,

• Unusual problems requiring specialized sampling procedures, and

• Any use of periodic (less frequent than annual) data collection cycles to reduce burden.

The 2024 ACS SOGI Test will consist of a national sample of roughly 480,000 housing unit addresses equally allocated across four treatments. The sample will not include housing units in Remote Alaska or Puerto Rico. Group quarters facilities are out of scope for the test.

The sampling methodology for the test is based on the ACS production sampling procedures that employs a two-phase, two-stage sample design (see Chapter 4 of the ACS Design and Methodology Report (census.gov) for more information).

Among addresses eligible to be sampled for the 2024 ACS SOGI test, a stratified systematic sampling methodology will be used to select the overall sample. The strata were defined to increase the number of sampled households with sexual and gender minorities (SGM) while ensuring adequate representation of rural households. There will be three sampling strata, defined based on the percentage of households with a same-sex couple and percentage of rural areas in the tract according to the following rules:¹

• Stratum 1: If the address is in a tract where the percentage of households with same-sex partners is greater than or equal to 1 percent, then the address is in Stratum 1.

• Stratum 2: If the address is in a tract where the percentage of households with same-sex partners is less than 1 percent, but the percentage of addresses in the tract considered rural is greater than or equal to 50 percent, then the address is in Stratum 2.

• Stratum 3: Otherwise, the address is in Stratum 3 (percentage of households featuring same-sex partners in a tract is less than 1 percent and the percentage of addresses in the tract considered rural is less than 50 percent).

Addresses within defined sampling strata will be geographically sorted and selected by systematic sampling. The strata will be sampled such that 45 percent of the households in sample are from Stratum 1, 30 percent are from Stratum 2, and 25 percent are from Stratum 3.

With limited data available at low levels of geography for the lesbian, gay, bisexual, transgender, queer, and intersex (LGBTQI+) populations, defining strata was a challenge. Since data on same-sex couples are available from the 2020 Census, we are using those data as a proxy to define the strata but recognize the limitations in doing so.²

Housing unit addresses selected for production ACS are not eligible to be included in the 2024 ACS SOGI test. Additionally, housing unit addresses selected for the test are not eligible to be selected for production ACS for a period of five years, consistent with the ACS restriction on sampling. This avoids burdening the public again with a similar request for data.

For the content follow-up (CFU) reinterview, nearly all sampled addresses from which we receive a response are eligible. Exceptions are made for responses that do not contain any names or contact information, for example. Sampling methods may be employed for CFU because of budgetary or operational reasons.

3. Describe methods to maximize response rates and to deal with issues of non-response. The accuracy and reliability of information collected must be shown to be adequate for intended uses. For collections based on sampling, a special justification must be provided for any collection that will not yield "reliable" data that can be generalized to the universe studied.

Similar to the production ACS, this test will employ a well-researched mail contact strategy to encourage self-response to the survey. For addresses that were mailed survey materials but did not respond by mail, internet, or by calling our telephone questionnaire assistance line, the Census Bureau selects a subsample of all households and assigns them to the the nonresponse follow-up data collection operation. Unmailable household addresses are sampled and also included in the nonresponse follow-up data collection operation.

To solicit self-response, the Census Bureau sends up to five mailings to potential respondents. The first two mailings are sent to all mailable addresses in the sample. The first contact is a letter (Attachment A) that provides instructions on how to complete the survey online, prominently displays the user identification number, explains that a paper questionnaire will be sent later if the respondent is unable to complete the survey online, and provides a phone number for respondents to call if they have questions.

The internet version of the questionnaire is available in English and Spanish and includes questions about the housing unit and the people living in the housing unit. The internet questionnaire has space to collect detailed information for 20 people in the household.

The second mailing, sent about a week after the first, is a letter (Attachment B) that reminds respondents to complete the survey online, thanks them if they have already done so, and informs them that a paper form will be sent at a later date if the Census Bureau does not receive their response. This letter includes clear instructions to log in, including an explicit reference to the user identification number.

Responding addresses are removed from being contacted again after the second mailing. The remaining nonresponding addresses are sent the third and fourth mailings. The third mailing is a paper questionnaire package (Attachment C). This package includes a letter, a paper questionnaire, and a business reply envelope. The letter reminds the household of the importance of the ACS and asks them to respond soon either by completing the survey online or by returning a completed paper questionnaire. The back of the letter includes frequently asked questions and provides basic information about the survey in English, Spanish, Russian, Chinese, Vietnamese and Korean, and provides a phone number to call for assistance in each language. The paper questionnaire has room to collect data on all questions for five people and for a few questions for seven more people.

Four days later, these addresses are sent a fourth mailing. This mailing is a postcard (Attachment D) that reminds respondents to mail back the questionnaire or respond on the internet. It also informs them that an interviewer may contact them if they do not complete the survey and reminds them of the importance of the ACS.

After the fourth mailing, responding addresses are again removed from being contacted again. The remaining nonresponding sample addresses are sent the fifth mailing (Attachment E). This mailing is a letter that includes a due date reminding these respondents to return their questionnaires by a specified deadline to be removed from the list of housing units that will be contacted by Census Bureau interviewers and thanks them if they have already done so. This letter includes clear instructions to log in, including an explicit reference to the user identification number and provides a telephone number to call if assistance is needed.

If a respondent starts to answer the survey online and provides an email address but does not complete the survey, an email (Attachment F) will be sent to the respondent to remind them to return to the survey to complete their online questionnaire. This email includes a link to the online survey and clear instructions to log in, including an explicit reference to the user identification number. This email is sent only once to a respondent, either prior to the third mailing or prior to the fifth mailing depending on when they started filling out the online survey.

The Census Bureau provides telephone questionnaire assistance (TQA) for respondents who need assistance with completing the paper or internet questionnaires or who have questions about the survey. Respondents may also complete the interview over the phone with a TQA interviewer. Respondents may call the ACS toll-free TQA numbers listed on various ACS mail materials, including information in and telephone numbers dedicated to the Spanish, Chinese, Vietnamese, Korean, Japanese, and Russian languages.

After the fifth mailing is sent, responding addresses are removed and the unmailable and undeliverable addresses (from the initial sample) are added to create the universe of addresses eligible for a nonresponse follow-up interview . Of this universe, a subsample is chosen to be included. A pressure seal reminder letter is sent to all mailable addresses sampled for the follow-up interview at the start of the interviewing month. This letter lets respondents know that a Census Bureau interviewer may call or visit them to complete the interview and encourages them to complete the survey online, if possible. This letter has not been finalized for the 2024 ACS SOGI Test but will be similar to the ACS production letter (see Attachment G). The final letter will be provided to OMB as a nonsubstantive change request prior to use in the field.

During the nonresponse follow-up data collection operation, Census Bureau interviewers first attempt to conduct an interview by phone, in most cases. If the interviewer is unable to complete a phone interview, they visit the address to conduct an in-person interview. The interviewers have several materials available to explain the ACS to households. The materials described below are the same as those used in the production ACS and are available in English, Spanish, and other languages.

Interviewers may provide an introductory letter and the housing unit informational brochure to respondents during the initial contact. The introductory letter provides respondents with a better understanding of the ACS, and how their information is kept confidential. The housing unit informational brochure helps respondents understand the value of ACS data, how people use ACS data to help their communities, and how to contact a regional office by telephone for additional questions. The letter and brochure in English are included in Attachment H.

The interviewers may also request that additional materials are mailed to households depending on questions raised by potential respondents. The “better understanding” letter emphasizes the confidentiality of the survey data and the benefits of the survey to motivate response. The ACS uses a “confidentiality letter” when a respondent indicates they have specific concerns about their response being kept confidential or their data being secure. The ACS sends a “refusal letter” when a respondent explicitly refuses to participate in the survey. The letter is short and to the point regarding confidentiality and the legal requirements to participate. It provides response options and emphasizes that the respondent will be contacted again soon to complete the interview. A Frequently Asked Questions brochure is given to reluctant respondents to provide answers to frequently asked questions regarding the ACS including information about the ACS, how the data are used, and protections for participant information. The letters and brochure in English are included in Attachment H.

Interviewers have three types of letters available to send to addresses that interviewers are unable to contact. The ACS sends a “no one home” letter to an address when the interviewer is unable to contact anyone at the sample address but thinks someone lives there. It emphasizes the confidentiality of the survey data and the benefits of the survey to motivate response. The respondent is told to call the interviewer or go online to respond. The ACS sends the “please call me” letter to respondents to encourage response. This letter can be used in lieu of other letters. The letter is short and straight to the point, as research with similar letters has shown to be highly successful in gaining response. The ACS uses the “final attempt” letter at the end of the data collection period. The letter is short; it expresses the uniqueness of being selected for the survey, the legal requirement for participation, and the different ways to respond. The letters in English are included in Attachment H.

If the interviewer is having access issues, three types of letters are available. The “management letter for multiunit buildings” helps interviewers gain access to multiunit buildings when they cannot directly access the sampled address. The “management letter for gated communities” helps interviewers gain access to addresses located in gated communities. A “seasonal home” letter is sent when the field representative suspects that the sampled address is only used seasonally and cannot contact anyone at the sample address. Field representatives may send this letter to someone who may be able to verify that the unit is seasonal. The letters in English are included in Attachment H.

Finally, the interviewer may also give a “thank you” bookmark to respondents who completed the interview in person or may send a “thank you” letter to respondents who completed the interview by phone. These materials thank the respondent for their participation and let them know they may be contacted for quality assurance purposes. These thank you materials in English are included in Attachment H.

Data collection instruments are available to respondents and interviewers in English (Attachment I) and Spanish (Attachment J). Wording is presented for in-person interviews, but telephone interview wording will be the same or very similar (the only differences would be related to administering the mode). Flashcards will also be used for in person interviews. The Census Bureau will provide an update to OMB via a Non-Substantive Change Request with the final materials.

The large sample size needed for this test creates operational and budgetary challenges in administering this test. Therefore, the sample will be split between two monthly panels. Each panel will follow the multimode data collection process outlined above. The self-response phase of this lasts for roughly two months for each panel. The nonresponse follow-up portion of the test , typically takes place in the third month of data collection following the self-response phase. Because of operational and budgetary considerations, the nonresponse follow-up portion of the test will be delayed to spring 2025. Additionally, the nonresponse follow-up workload may be spread across multiple months. The delay in the nonresponse follow-up administration could result in lower response rates, but we have no experience or literature to draw on to determine the impact.

As part of the test a CFU reinterview will also be conducted. The CFU reinterview provides the data to assess response reliability. CFU has historically been conducted as a telephone follow-up only. Because of the size of this test and due to decreasing telephone response rates, we will use both an internet CFU and a telephone CFU for this test. Respondents who initially respond online will be sent a notice inviting them to complete the CFU follow-up survey online. They will also receive a reminder mailing (if a response hasn’t been received) and an email reminder (if a response hasn’t been received). See Attachment K. Respondents who initially respond via the paper questionnaire will be eligible to complete the CFU by telephone. To improve response rates to the telephone CFU, we will send a prenotice letter informing them of the follow-up survey (see Attachment K). Telephone interviewers will then call the households to complete the CFU interview over the phone.

Respondents who respond during the nonresponse follow-up data collection operation will also be asked to complete the telephone CFU. Prenotice materials, like a letter, will be used as well. The exact methods of contact and materials for that operation are still to be determined but will look similar to those provided in Attachment K. The Census Bureau will provide an update to OMB via a Non-Substantive Change Request with the final materials.

4. Describe any tests of procedures or methods to be undertaken. Testing is encouraged as an effective means of refining collections of information to minimize burden and improve utility. Tests must be approved if they call for answers to identical questions from 10 or more respondents. A proposed test or set of tests may be submitted for approval separately or in combination with the main collection of information.

The Census Bureau received requests from federal agencies to add SOGI questions to the ACS. These requests identified legal uses for this information, including civil rights and equal employment enforcement, which met the strict requirements for adding survey content to the ACS.

The test will include new questions about sexual orientation and gender identity. Sexual orientation will be measured with one question. Gender identity will be measured with a two-part series: sex assigned at birth and current gender. Two wording treatments will be tested for the current gender question. Version 1 of the current gender question will allow for only one response category to be selected; Version 2 will allow for multiple response categories to be selected.

Table 1. Wording Treatments Experimental Design

Topic	Treatment 1	Treatment 2
Sex Assigned at Birth Question Wording	Version 1	Version 1
Current Gender Question Wording	Version 1	Version 2
Sexual Orientation Question Wording	Version 1	Version 1

In addition to testing question wording, for the internet instrument we will also test how the write-in for current gender and sexual orientation is displayed. (Sex assigned at birth does not have a write-in.) In Version A, the respondent will see all the response categories along with the write-in which will be displayed next to the “This person uses a different term” category. In Version B, only the response categories will be displayed. After a respondent selects “This person uses a different term” and clicks the “next” button, a new question will prompt the respondent to write-in the term.

Table 2. Write-in Display Treatments Experimental Design

Topic	Treatment A	Treatment B
Current Gender Write-In Display	Version A	Version B
Sexual Orientation Write-In Display	Version A	Version B

The wording and write-in display treatment conditions will be fully crossed to create four treatments: Treatment 1A, Treatment 1B, Treatment 2A, and Treatment 2B. The treatments are defined by their unique combination of question wording for the current gender question and write-in box display for the current gender and sexual orientation questions.

Table 3. Overall Test Treatments Experimental Design

Topic	Treatment 1A	Treatment 1B*	Treatment 2A	Treatment 2B*
Sex Assigned at Birth Question Wording	Version 1	Version 1	Version 1	Version 1
Current Gender Question Wording	Version 1	Version 1	Version 2	Version 2
Current Gender Write-In Display	Version A	Version B	Version A	Version B
Sexual Orientation Question Wording	Version 1	Version 1	Version 1	Version 1
Sexual Orientation Write-In Display	Version A	Version B	Version A	Version B

* In the paper and interviewer-administered modes, Treatment 1B is identical to Treatment 1A, and Treatment 2B is identical to Treatment 2A.

The remainder of the questionnaire will include all ACS questions. The current production wording will be used for the questions except the following:

• Relationship question: categories will be updated to remove gendered terms and same-sex/opposite-sex designations.

• Relationship check in automated instruments will be removed.

• Age will be asked before sex assigned at birth.

• Fertility question: the skip pattern for the fertility question will be modified to include people whose sex assigned at birth was female.

• Several questions will use “they” instead of “he/she.”

(See Attachment C for the paper questionnaires and Attachment I and J for wording for the other modes in English and Spanish).

In addition, in the internet instrument, open-ended questions will be asked at the end of the interview for qualitative evaluation purposes only. See Attachments I and J.

Comparing response reliability (as measured by a gross difference rate) between two wording treatments of 240,000 addresses allows us to detect at least a one percentage point difference in a five percent gross difference rate with 80 percent power and α=0.1; this calculation assumes a 50 percent unit response rate and a 33 percent CFU response rate. The minimum detectable difference for other evaluation measures will vary for each item being evaluated.

A content follow-up (CFU) reinterview will be conducted to measure response reliability. The content reinterview will re-ask a subset of questions from the original interview including the following:

• Basic person questions: age, Hispanic origin, race, relationship, sex assigned at birth, current gender.

• Housing questions: none.

• Detailed person questions: place of birth, educational attainment, health insurance coverage, disability, fertility, marital status/history, sexual orientation, and total income.

For the CFU, households will receive the same question treatment they were asked in the original interview. Answers to questions in the original interview will not be provided to the respondents in the content reinterview. The question wording for these items is identical to the wording used in the initial interview.

In addition, open-ended questions will be asked at the end of the interview for qualitative evaluation purposes only. See Attachments I and J.

One critical aspect of this test is to understand how proxy respondents answer questions compared to those who self-report. To conduct that analysis, some CFU reinterviews will be conducted with the same respondent as the original interview and some will be conducted with a different adult (18 or older) household member. Among eligible households, there will be three treatments for the assignment of CFU respondents:

• CFU Respondent Treatment 1 will consist of households with only one adult with a valid name. These cannot be included in the assignment to the same-respondent and different-respondent groups, but CFU responses should still be collected from them. In the CFU interview, they will be treated exactly the same as CFU Respondent Treatment 2.

• CFU Respondent Treatment 2 will consist of households with more than one adult (people age 18 or over), who have been assigned to have the original interview respondent respond again in CFU.

• CFU Respondent Treatment 3 will consist of households with more than one adult, who have been assigned to have a different person respond in CFU.

To assign CFU Respondent Treatments 2 and 3, households that have at least two adults will be stratified into two strata: those with a sexual or gender minority, and those without a sexual or gender minority. Within each strata, a household will be assigned to either CFU Respondent Treatment 2 or CFU Respondent Treatment 3 using a systematic sampling method with 50-50 probability.

By using a variety of metrics (item nonresponse, response distributions, response reliability, etc.) outlined in the analysis plan (Attachment M), this test will determine:

• How well the estimates for sexual orientation and gender identity compare to existing data sources.

• Which version of the current gender question wording is recommended for production implementation.

• Which version of the write-in display is recommended for production implementation.

• Whether the confirmation question should be asked for everyone or only for people where the sex assigned at birth and current gender question do not match.

• How well responses provided by proxy align with responses provided by self-report.

5. Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze the information for the agency.

The Census Bureau will collect and process these data. Within the Census Bureau, please consult the following individuals for further information on their area of expertise.

Statistical Aspects

Joan Hill Assistant Division Chief for Experiments and Evaluations

Decennial Statistical Studies Division

Phone: (301) 763-4286

Overall Data Collection

Elizabeth Poehler Assistant Division Chief for Survey Methods

American Community Survey Office

Phone: (301) 763-9305

1 Because low-level geographic data is not available for many of the SGM groups, data on same-sex partners is being used. The data for this come from the 2020 Census Demographic and Housing Characteristics File: PCT15: COUPLED HOUSEHOLDS, BY TYPE - Census Bureau Table, and H2: URBAN AND RURAL - Census Bureau Table.

2 Some people define the “Q” in the LGBTQI+ acronym to represent “questioning” instead of “queer.” The + (plus) in the acronym represents other sexual identities, including asexual, pansexual, Two-Spirit, and others.

Page | 6