2020_EAVS_Supporting_Statement_A_Revised per OMB request_Final

Supporting Statement A:

OMB Control Number: ####-####

U.S. Election Assistance Commission

2020 Election Administration and Voting Survey

A. Justification

1. Explain the circumstances that make the collection of information necessary.

The proposed information collection is comprised of two components: the Election Administration and Voting Survey (EAVS), which is a census of election officials on various election administration and voting topics; and the Policy Survey, which provides policy context for EAVS data. The information collection is necessary for several reasons. First, the Help America Vote Act (HAVA) of 2002 (52 U.S.C. § 20901), especially §241, requires the U.S. Election Assistance Commission (EAC) to study and report on election activities, practices, policies, and procedures, including methods of voter registration, methods of conducting provisional voting, poll worker recruitment and training, and such other matters as the Commission determines are appropriate. HAVA §202 further charges the EAC with serving as “a national clearinghouse and resource for the compilation of information and review of procedures with respect to the administration of Federal elections.”

Second, HAVA §802 transferred to the EAC the Federal Election Commission’s responsibility of biennially administering a survey on the impact of the National Voter Registration Act (NVRA) (52 U.S.C. § 20508 ). As detailed in the table below, the information the states are required to submit to the EAC for purposes of the NVRA report are found under Title 11 of the Code of Federal Regulations (11 CFR 8.7).

Information required	Corresponding 2020 EAVS and Policy Survey questions
The total number of registered voters statewide, including both ‘‘active’’ and ‘‘inactive’’ voters if such a distinction is made by the state, in the federal general election two years prior to the most recent federal general election	EAVS question A1a
The total number of registered voters statewide, including both ‘‘active’’ and ‘‘inactive’’ voters if such a distinction is made by the state, in the most recent federal election	EAVS question A1a
The total number of new valid registrations accepted statewide between the past two federal general elections, including all registrations that are new to the local jurisdiction and re-registrations across jurisdictional lines, but excluding all applications that are duplicates, rejected, or report only a change of name, address, or (where applicable) party preference within the local jurisdiction	EAVS question A3b
If the state distinguishes between ‘‘active’’ and ‘‘inactive’’ voters, the total number of registrants statewide that were considered ‘‘inactive’’ at the close of the most recent federal general election	EAVS question A1c
The total number of registrations statewide that were, for whatever reason, deleted from the registration list, including both ‘‘active’’ and ‘‘inactive’’ voters if such a distinction is made by the state, between the past two federal general elections	EAVS question A9a
The statewide number of registration applications received statewide (regardless of whether they were valid, rejected, duplicative, or address, name or party changes) that were received from or generated by each of the following categories: (i) All motor vehicle offices statewide; (ii) Mail; (iii) All public assistance agencies that are mandated as registration sites under the Act; (iv) All state-funded agencies primarily serving persons with disabilities; (v) All Armed Forces recruitment offices; (vi) All other agencies designated by the state; (vii) All other means, including but not limited to, in person, deputy registrars, and organized voter registration drives delivering forms directly to registrars;	EAVS questions A4a through A4l
The total number of duplicate registration applications statewide that, between the past two federal general elections were received in the appropriate election office and generated by each of the categories described in paragraphs (b)(6) (i) through (vii) of this section	EAVS questions A6a through A6l
The statewide number of confirmation notices mailed out between the past two federal general elections and the statewide number of responses received to these notices during the same period	EAVS questions A8a through A8h
Answers to a series of questions with categorical responses for the state to indicate which options or procedures the state has selected in implementing the NVRA or any significant changes to the state’s voter registration program	Policy Survey questions 1, 3, 6 through 14
Any additional information that would be helpful to the Commission for meeting the reporting requirement under 42 U.S.C. 1973gg–7(a)(3).	EAVS questions A1b, A2, A3a, A3c through A3j, A5, A7, A9b through A9j

Third, HAVA §703(a) amended §102 of the Uniformed and Overseas Citizens Absentee Voters Act (UOCAVA) (52 U.S.C. §20302(c)) by requiring that “not later than 90 days after the date of each regularly scheduled general election for Federal office, each state and unit of local government which administered the election shall (through the state, in the case of a unit of local government) submit a report to the Election Assistance Commission (established under the Help America Vote Act of 2002) on the combined number of absentee ballots transmitted to absent uniformed services voters and overseas voters for the election and the combined number of such ballots which were returned by such voters and cast in the election, and shall make such a report available to the general public.” The corresponding EAVS questions are detailed in the table below.

Required information	Corresponding 2020 EAVS questions
Combined number of absentee ballots transmitted to absent uniformed services voters and overseas voters	EAVS question B5a
Combined number of such ballots which were returned by such voters and cast in the election	EAVS questions B9a, B14a

Fourth, the EAC and the Federal Voting Assistance Program (FVAP), an agency of the Department of Defense, have worked together to combine their requirements to collect data about voting by UOCAVA citizens. Starting in 2014, the EAC added questions from FVAP’s Post-Election Survey of Local Election Officials to the Election Administration and Voting Survey (EAVS). This consolidation of surveys reduced the paperwork burden on state and local election offices and also made FVAP a primary consumer of the EAVS. As a part of this consolidation, the EAC and FVAP worked with the chief state election official of each state and developed standards for reporting the number of absentee ballots requested and received, and other data as FVAP determines appropriate and for FVAP to store the data reported, as required under the MOVE Act (Military and Overseas Voter Empowerment, enacted as part of the National Defense Authorization Act of FY 2010 (P.L. 111-84).

The primary use of these data by FVAP is to identify areas where the electoral process can be improved by providing an accurate picture of the UOCAVA absentee voting process. These data permit FVAP to evaluate the extent to which FVAP is achieving its mission and the actions it can take to improve the process. In addition, FVAP uses these data to evaluate if legislative changes have been successful in removing barriers for absentee voting and identify any remaining obstacles to voting by those populations covered by the UOCAVA.

Finally, it is important to note that other federal agencies rely on data collected through the EAVS, including the Department of Justice, Department of Homeland Security, the Census Bureau, and the U.S. Postal Service. This is discussed in further detail below under question 2.

2. Indicate how, by whom, and for what purpose the information is to be used.

These data are used by several sources. First, the EAC will use the data collected by the 2020 EAVS to meet its statutory requirements related to (1) the impact of the NVRA (52 U.S.C. § 20508) on the administration of elections for the period from the day after the November 6, 2018 Federal general elections until Election Day November 3, 2020; (2) the required HAVA information regarding the combined number of absentee ballots transmitted to absent uniformed services and overseas citizen voters for the election and the combined number of such ballots which were returned by such voters and cast in the election; and (3) information required by the Help America Vote Act (HAVA) of 2002 (52 U.S.C. § 20901), especially §241, that is used in the EAC biennial report to Congress. The link to previous reports is available here: https://www.eac.gov/research-and-data/studies-and-reports. The EAC also uses these data for various reports and guidance for state and local election officials.

Second, FVAP is a primary user of these data; it is required to submit a report to Congress reflecting a statistical analysis of uniformed services and overseas citizen participation in each federal general election, and also uses these data for policy-specific analyses. The EAC shares all relevant EAVS data with FVAP now that the EAC collects all local election official quantitative data for FVAP.

Third, additional users of these data include other federal agencies. For example, the Voting Section of the Department of Justice’s Civil Rights Division uses EAVS data to inform its monitoring and enforcement of federal voting laws, including HAVA, the NVRA, and UOCAVA. The Election Security Initiative at the Department of Homeland Security’s Cybersecurity and Infrastructure Security Agency, which leads federal efforts to protect U.S. election infrastructure following its designation as critical infrastructure in January 2017, uses EAVS data to inform its analyses of election infrastructure, including election technology cybersecurity threat modeling for different types of election jurisdictions. Other examples include the Census Bureau, which has used EAVS data on turnout and voter registration to help validate data collected through the Voting and Registration Supplement to the American Community Survey, and the U.S. Postal Service, which has used EAVS data to inform trend analyses on voting by mail.

Finally, EAVS data are also used by an array of public users, including academic and public policy researchers, in an effort to understand the conduct of American election administration.

3. Describe whether, and to what extent, the collection of information involves the use of automated, electronic, mechanical, or other technological collection techniques or other forms of information technology.

In 2016, the EAVS data collection was modified so that that states had much greater flexibility in collecting and reporting data. The data were collected primarily using an Excel template, but modifications were made so that states with more sophisticated capabilities could report their data as an export from their state election management system. States who needed to collect data directly from their local jurisdictions were provided with a simple Excel-based template, and these data could be easily combined into a single state report. Some localities were allowed to report their data on a paper form, or via a telephone interview-style data collection format. The reported data were also validated using human-assisted machine learning techniques.

In 2018, the EAC created an online data collection system to supplement the Excel-based template that accommodated state-level data exports; this online system replaced the Excel template that allowed for item-by-item data entry. This online system was used by more than a dozen states, especially those that rely heavily on local election offices to provide EAVS data. The state and local election offices that used the online system reported positive feedback and indicated that it reduced the time spent filling out the survey. The online system also increased data integrity by incorporating data validation checks throughout the online survey and by reducing the amount of manipulation that state officials needed to do with local-level submissions; the Excel template had the capability of automatically porting online survey submissions into a single Excel file.

The EAC plans to use the online data collection system again in 2020 to supplement the Excel-based template and will introduce additional capabilities that will allow local election offices to edit their data prior to certification and better track their local jurisdictions’ progress through the survey. It is expected that these additional functionalities will further reduce the response burden associated with the survey and will lead to higher-quality and more accurate data submissions.

4. Describe efforts to identify duplication.

In 2014, FVAP and the EAC combined their survey questions about UOCAVA voting to lessen the burden on states and localities associated with federal reporting of these data. By asking the questions once, in a single survey, both organizations have obtained higher-quality data and higher compliance with data reporting requirements. Under the memorandum of understanding between the two agencies, FVAP provided the EAC with the survey questions, which were added to the EAVS, and the EAC provided FVAP with all UOCAVA data after the survey had been administered.

In 2015, FVAP created a working group that reviewed all UOCAVA questions contained in the EAVS. The group identified all redundant questions and recommended changes to other questions so that the questions could be more easily understood and data reporting improved. The 2018 survey questions related to UOCAVA voting were reduced based on this effort, and this reduction in questions continues in the 2020 survey. In 2019, the EAC launched a similar working group to examine the voter registration questions in the EAVS, including efforts to streamline questions where redundancy may exist. It is anticipated that any changes recommended by this group will be implemented no earlier than the 2022 EAVS.

To further identify and mitigate against duplication of effort, EAC staff maintain regular communication with federal agencies known to conduct data collections on similar topics, such as the Census Bureau and FVAP. For example, when modifying questions regarding UOCAVA voters for the 2020 Policy Survey, a component of the EAVS data collection formerly called the Statutory Overview, EAC staff discussed potential revisions with FVAP and reviewed existing FVAP data collections to ensure that duplication was avoided. Similarly, when considering a potential new question for the 2020 EAVS on election jurisdiction employment, the EAC engaged officials at the Census Bureau who administer the Annual Employment Survey and Census of the Governments to inform question design and make sure there would be no duplicated effort.

5. If the collection of information impacts small businesses or other small entities,

describe any methods used to minimize burden.

This information collection does not have a significant impact on small businesses or other small entities. The chief election officials for the states, the District of Columbia, and the U.S. territories may have to request information from their local election jurisdictions, but much of this information is already routinely collected from the local election officials to certify election results and report voter turnout.

The EAC has made efforts to limit the information requested and burden on all participants. The information sought is limited to that information necessary to meet the requirements listed in response to Question 1.

6. Describe the consequence to Federal program or policy activities if the collection is not conducted or is conducted less frequently, as well as any technical or legal obstacles to reducing burden.

If the EAC does not collect this information it will be unable to comply with its statutory requirements under HAVA (52 U.S.C. § 20901), NVRA (52 U.S.C. § 20508), and UOCAVA (52 U.S.C. §20302(c)). This collection of information must be carried out every two years after each Federal general election as stipulated by NVRA and UOCAVA. In addition, FVAP will be unable to meet its congressional reporting requirements related to UOCAVA voting if the EAC does not collect this information.

Because the core questions in the EAVS will not change for 2020 and have not changed dramatically since 2006, state and local election offices are well-positioned to answer the questions contained in the survey. Many election offices already have developed data collection methods for the EAVS data or developed system queries to extract data from election management systems, which also reduces the burden of the EAVS moving forward.

7. Explain any special circumstances that require the collection to be conducted in a manner inconsistent with OMB guidelines.

There are no special circumstances applicable to this information collection.

8. If applicable, provide a copy and identify the date and page number of publication in the Federal Register of the agency’s notice, required by 5CFR 320.8(d), soliciting comments on the information collection prior to submission to OMB. Summarize public comments received in response to that notice and describe actions taken in response to the comments. Specifically address comments received on cost and hour burden. Describe efforts to consult with persons outside of EAC.

The EAC published a Federal Register Notice soliciting comments on the information collection for a 60-day period, beginning October 8, 2019, Vol. 84, No. 195, pgs. 53709-53710. A copy of the notice as published is provided as Attachment A. In addition to publication in the Federal Register, the EAC sought to maximize the public comments received by advertising the public comment period on its social media channels and by encouraging election officials, scholars, and other stakeholders in the elections community to review the documents and submit comments. Prior to submission to the Federal Register, the EAC consulted with the EAVS committee on its Standards Board (a federal advisory committee to the EAC) and members of a working group to solicit their input on potential modifications to the data collection.

EAC received 13 comments that were submitted by 12 individuals in 6 states. Comments were made regarding both the EAVS and the Policy Survey. Four comments were received from individuals affiliated with state election offices, four were received from academics, and seven were received from non-governmental organizations focused on elections or individual election advocates. Table 1 shows the number of persons providing comments by state.

Table 1: Number of Persons Commenting by State
State	Number of Persons Commenting
California	3
Connecticut	2
Massachusetts	1
Maryland	1
Mississippi	1
Washington, DC	2
Unknown	3
Total	13

The EAC analyzed the content of the comments for the EAVS and the Policy Survey separately; Table 2a categorizes the comments related to the EAVS and Table 2b categorizes the comments related to the Policy Survey. The total number of comments by content type may be greater than 13 (the number of persons submitting comments) because some comments contained multiple requests and some comments pertained to both the EAVS and the Policy Survey.

Table 2a: Comments on the EAVS
Content Type	Number of Comments	Percent of Total
Requests to add additional questions	6	31.6
Request for clarification of survey instructions	4	21.1
Requests relating to survey administration and survey tools	3	15.8
Requests to remove questions	2	10.5
Requests for adjustments to existing questions	2	10.5
Typos or grammatical errors	2	10.5
Totals	19	100.0

Table 2b: Comments on the Policy Survey
Content Type	Number of Comments	Percent of Total
Requests to add additional questions	4	33.3
Requests for adjustments to existing questions	4	33.3
Statements in support of a new question added by the EAC	3	25.0
Request for clarification of survey instructions	1	8.3
Total	12	100.0

The EAC’s responses to the suggestions for the EAVS are below:

• Requests to add additional questions: There were many suggestions for additional questions, including questions on cybersecurity and election security, the use of voting technology, mail ballot processing technology, poll workers who provide multilingual assistance, wait times at polling places, and UOCAVA ballots transmitted via online portal. However, the EAC is not certain that (a) all jurisdictions collect the data to be able to answer these questions, or that (b) the questions applied broadly to jurisdictions instead of only a small subset of jurisdictions. Additionally, the EAC is reticent to increase response burden by adding questions to an already lengthy survey. The EAC also wishes to conduct more rigorous question design and user testing of these potential questions. The EAC will not add these additional questions in the 2020 EAVS but will work with the EAC Standards Board and other stakeholders to determine if these items are appropriate to consider for the 2022 EAVS.

• Request for clarification of survey instructions: Several commenters raised issue with EAC’s revision of F1 to measure “successful” participation; this question has been revised to measure the number of voters who cast a ballot that was counted (in total and by mode of participation). Another recurring comment was a request to further define who should be counted as a poll worker; additional clarification has been added to the instructions for questions D5-D7. The instructions for questions B5-B8 clarify that UOCAVA ballots transmitted via online ballot delivery portals should not be counted with ballots transmitted via email. Question B27 has been clarified that all FWABs rejected for reasons other than being received late or because the voter’s regular absentee ballot was submitted and counted should be included in this question. Question E1c includes more clarification of what a partially counted provisional ballot is. The instructions for questions F5-F11 include more detail on the definition of a scanner.

• Requests relating to survey administration and survey tools: Some of the requests relating to how the EAC administers the EAVS (providing an online tool for filling out the survey, color-coding items in the data collection templates that do not match up with expected data patterns) have already been implemented beginning with the 2018 EAVS. Other suggestions (tailoring and/or pre-populating each state’s survey to reduce response burden, hard-coding not applicable values, providing respondents more detail about expected response patterns, releasing contact information for local election officials) are under consideration for implementing in the 2020 EAVS data collection templates, but do not require changes to the survey questions themselves.

• Requests to remove questions: Two commenters requested that the EAC eliminate the questions collecting data on the use of punch card (F9) and lever machines (F10) by jurisdictions, noting that no jurisdiction reported using this equipment in the 2016 and 2018 EAVS. The EAC will keep these questions for at least the 2020 EAVS until it is demonstrated across multiple election cycles that these machines are no longer in use.

• Requests for adjustments to existing questions: A number of commenters requested adjustments to the number of “other” categories included in questions throughout the survey. The EAC wishes to undertake more study on the appropriate number of categories for each question and will work with the EAC Standards Board and other stakeholders to consider making these changes in the 2022 EAVS. Another request was to reduce survey burden in the questions on voting technology (F5-F11); the EAC anticipates that pre-populating some of this information (as requested by another commenter) and providing additional instruction in the data collection templates in this section will reduce the time that respondents spend answering these questions.

• Typos or grammatical errors: The EAC has addressed all of the typographical and grammatical errors identified by the commenters.

The EAC’s responses to the suggestions for the Policy Survey are below:

• Requests to add additional questions: Commenters requested adding a number of additional questions to the Policy Survey, including questions about pre-registration for persons under 18 years of age, deadlines for certifying the 2020 Federal general election, allowing domestic civilians to receive mail ballots electronically, signature verification, auditing the accuracy of state voter registration databases, authentication of remote voters, additional questions about auditing procedures, and information about the age and provenance of states’ election management systems and voter registration databases. Because the primary purpose of the Policy Survey is to collect data that is relevant to and provides context for states’ responses to EAVS, the EAC has added questions on pre-registration (Q10 of the Policy Survey, to provide context to A3c of EAVS) and certification deadlines (Q32 of the Policy Survey, to provide context to Sections B, C, and E of EAVS) and will work with the EAC Standards Board and other stakeholders to determine if other items are appropriate to consider for the 2022 Policy Survey.

• Requests for adjustments to existing questions: In response to a comment, Q2a has been clarified that it asks whether an office that has responsibilities for election administration and voter registration at the jurisdiction level exists in every jurisdiction across the state. Several commenters requested additional, more detailed questions on state election auditing practices and procedures. This is an area of election administration that is evolving very quickly and is difficult to design questions that can be satisfactorily answered by all state-level respondents. The EAC’s primary goal in this section of the Policy Survey is to collect data that can be used to provide context to states’ EAVS data and to facilitate additional research by the EAC and other interested parties. As such, EAC will not be adding additional questions on election audits to the 2020 Policy Survey.

• Statements in support of a new question added by the EAC: One commenter noted the value to voters, election stakeholders, researchers, and third-party organizations of the EAC collecting information on the structure of state and local election offices and their contact information (Q2, Q2a, Q2b, and Q2c) and providing such information publicly. Another commenter commended the inclusion of a question on risk-limiting audits.

• Request for clarification of survey instructions: Requests were made to clarify the instructions in Q2, Q2a, Q2c, Q5, Q6b, Q6c, Q24 and Q31c. The requested clarifications were made.

The EAC published a Federal Register Notice soliciting comments on the information collection for a 30-day period, beginning February 11, 2020, Document Citation 85 FR 7758, pg. 7758. A copy of the notice is accessible at https://www.federalregister.gov/d/2020-02688. EAC received 18 comments that were submitted by 7 individuals. Comments were made regarding both the EAVS and the Policy Survey. Below in Table 3 and Table 4 are comments by instrument and EAC’s response to each comment:

Table 3: Comments on the EAVS from 30-day period
EAVS Item	Commenter’s Request and Rationale	EAC Response
B5	Reword this question to ask about “TOTAL blank/unvoted absentee ballots transmitted to UOCAVA voters”	The question language was determined by the Section B Working Group that revised this section of EAVS in 2016. There has been no evidence in the past two EAVS cycles that this question is being misunderstood.
B10	Reword this question to ask “Returned by Postal Mail”	The question language was determined by the Section B Working Group that revised this section of EAVS in 2016. There has been no evidence in the past two EAVS cycles that this question is being misunderstood.
B12	Add instruction “We are also interested in knowing whether the mode of transmission of the ballot to the voter affects the return rate.” Add a question “Return rate based on transmission mode to voter Of all ballots electronically transmitted to the voter (B7&B8) how many were returned?”	EAC does not want to have EAVS respondents calculate rates, as that would introduce data errors and these rates may not be calculated uniformly by all respondents. Many EAVS respondents also do not have access to this data.
F3	Add questions that collect data on the makes/models/versions of electronic pollbooks used by jurisdictions Electronic pollbooks are becoming more widely used, and it would be useful to know what systems are in use and where they are used.	EAC agrees that this is very useful data to have. However, there are challenges to collecting this data. In many cases the software used on the e-poll books is as important to know as the hardware in use, which would require multiple questions to address. In addition, in order to collect this data, EAC would need to use a closed-ended question; collecting this information in open-ended questions would mean the data would not be suitable for analysis. The EAC plans to study this option and work with election officials and industry stakeholders to consider whether this question could be added in the 2022 EAVS.
F5-F11	Remove sets of questions on punch card and lever machines	EAC will consider removing these questions once has been reported in two subsequent survey iterations that these types of voting equipment are not in use. No respondents reported this equipment in the 2018 EAVS; if this continues in the 2020 EAVS, the EAC will consider removing these questions from the survey beginning with the 2022 EAVS.
F5-F11	Change description of BMDs in question instructions Add a set of questions asking about use of hybrid BMDs/scanners This change would distinguish between BMDs that have separate vs integrated tabulators. (Without this addition, some systems do not fit into any category!)	EAC has concerns about whether this language is used widely enough that all respondents would be able to answer these questions.
F5-F11	In addition to collecting the makes/models of voting equipment, collect data on the version number of those makes and models. This provides an opportunity for gathering information about third party vendors and software, firmware and hardware version numbers.	EAC agrees that this is very important data. However, many localities do not know the version number of the equipment they use and therefore this data would be unreliable. EAC is exploring ways to make this information available in other formats aside from EAVS, or to begin collecting this information with the 2022 EAVS.
F5-F11	Modify two of the equipment uses in column D to add “for all voters”: In-precinct Election Day regular balloting for all voters In-person early voting for all voters This addition creates further clarity	EAC disagrees that this would provide clarity and believes this would make it more difficult for respondents to answer. In addition, the questions as currently written allow data users to determine whether an equipment type is used for all in-precinct Election Day voters or all in-person early voters. By using the F5a, F6a, F7a, F8a, and F11a variables, users can determine whether DREs w/o VVPAT, DREs w/ VVPAT, BMDs, scanners, and hand-counted paper ballots are the only type of equipment used in a jurisdiction, and from there can determine whether they are used for all voters.
F5-F11	Add a column to this question grid that asks whether voting machines are capable of internet connection	This information is available through non-EAVS sources. EAC is declining to make this change.
F5-F11	Add question on electronic ballot return (which states still allow electronic return of ballots and under what circumstances and for which voters such return is allowed) It is important to add a question about internet voting. Some states allow some form of internet voting (like email return of ballots) for some types of voters. It is important to know the quantity of internet voted ballots to access the security risk. It also is important to know the quantity of internet voted ballots, as well as paperless DRE voted ballots, because neither can be verified by the voter. Therefore, there is no audit trail for all of these votes.	This question would be better suited for the Policy Survey, not in Section F of EAVS. EAC does not believe that most states are at the point where they are able to track the number of internet voted ballots, although that may change in the future.
General	Add election security questions	EAC is considering developing a set of questions that would be administered separately from EAVS.
General	Comment about how to ensure less voter fraud	Not relevant to EAVS; no response needed
General	Comment about Georgia voting machines and requiring no-excuse absentee voting, paper balloting, risk limiting-audits	Not relevant to EAVS; no response needed

Table 4: Comments on the Policy Survey from 30-day period
Policy Survey Item	Commenter’s Request and Rationale	EAC Response
Q2	Collect names of chief local election officials Without this information, some key stakeholders, notably the US Postal Service, may be unable to direct important information about election conduct, administration, and security. Any package of materials that has to be addressed to a name, and not a generic "office" address, will not be deliverable. The names will also be valuable to any organization that needs a specific point of contact for local elections related information.	Since this is the first time this information is being gathered through EAVS, EAC will be evaluating the completeness and data quality of this before considering adding additional information. In addition, EAC is evaluating when this data could be released.
Q19	Add this question: Which domestic civilian absentee voters can receive their blank/unvoted ballots electronically? □ None □ Those in emergency situations □ Those needing a ballot replacement □ Those with disabilities □ All Especially important given the recent concern about the coronavirus and the possible expansion of vote-by-mail	EAC reworded this question and added it as Q23.
Q33	Add this followup Q33A Which methods are allowed for retabulations? (Check all that apply) □ machine rescanning of paper ballots □ human interpretation of paper ballots □ Other (please specify):	Difficult to design a question on this topic that accurately captures state procedures in a series of closed-ended responses.
Q34	Reword Q34 and Q34a so it conforms to generally-accepted terminology, add follow-up Q34b (whether human examination is used during audit) There are many potential types of post-election audits. For example cybersecurity audits, paper ballot security audits, pollbook audits, and procedural audits. Q34 specifically deals specifically with Post-Election Tabulation audits. It is important to distinguish them correctly for now, and for the future when surveys should consider those other post-election audits as well. The draft Q34 is not clear that the post election tabulation audit examines the actual ballots cast in the election.	The terminology used in these questions was modeled on NIST. EAC has revised Q34 and Q34a to use the term “post-election tabulation audit” instead of “post-election audit). EAC has not made the requested changes to the answer options in Q34a because they do not match the language used by NIST. EAC declines to add Q34b. EAC’s goal with these questions is to capture broad generalities about state policies and practices to provide avenues for further study.
Q35	Add a question on authentication of the signatures of remote voters (compared to a signature of record, or another method) Important given recent coronavirus concerns and possible expansion of mail voting	EAC needs to do further investigation before adding this question.

9. Explain any decision to provide any payment or gift to respondents, other than remuneration of contractors or grantees.

The EAC does not provide any payment or gift to respondents.

10. Describe any assurance of confidentiality provided to respondents and the basis for the assurance in statute, regulation, or agency policy.

There is no assurance of confidentiality.

11. Provide additional justification for any questions of a sensitive nature, such as sexual behavior and attitudes, religious beliefs, and other matters that are commonly considered private.

There are no questions of a sensitive nature.

12. Provide estimates of the hour burden of the collection of information. The statement should indicate the number of respondents, frequency of response, annual hour burden, and an explanation of how the burden was estimated.

The information collection has two parts: The Election Administration and Voting Survey (EAVS), and the Policy Survey. The estimated response burden is based on feedback provided in 2019 from thirty four individual states on the estimated total number of hours spent on gathering the necessary information and on entering the data into the electronic template. The thirty four states represented a mix of the number of reporting local jurisdictions (from 8 to 300), and of different record-keeping database architectures (i.e., top-down, bottom-up, and hybrid). The median response was 80 hours. To estimate conservatively and be cognizant of the fact that several states reported a considerably higher response burden, we use 100 hours per respondent in our calculations below. It should be noted that this estimated response burden is markedly lower than previous years. This reduction is likely due to multiple factors, including EAC efforts to streamline questions and improve data collection mechanisms over time, as well as state efforts to modify their systems and processes to accommodate this biennial collection, which has remained largely unchanged since the 2008 iteration of the survey.

Under the online method of completing the Policy Survey, where States select pre-determined response options, we estimate that the new burden for completing the Policy Survey is, on average, 0.6 hours. This reflects a dramatic reduction in estimated burden in previous years when the Statutory Overview was still in use. In 2018, the Statutory Overview survey was overhauled from an essay-answer survey to a multiple-choice survey.

The table below summarizes the burden estimates for the EAVS and the Policy Survey. Because this data collection occurs every two years, we have calculated and provided the annualized burden.

Collection Component	Number of Respondents	Respondent Burden	Total Burden	Annualized Burden
EAVS	56	100	5,600	2,800
Policy Survey	56	1	56	28
Total		101	5,656	2,828
Note: Decimals are rounded to the nearest whole number.

The estimated cost of the annualized cost of this burden is: $63,423.04, which is calculated by taking the annualized burden (2,828 hours) and multiplying by an hourly rate of $22.43 (GS-8/Step 5 hourly basic rate).

13. Provide an estimate for the total annual cost burden to respondents or recordkeepers resulting from the collection of information.

There are no capital or start-up costs associated with this information collection.

14. Provide estimates of annualized cost to the Federal government.

The estimated annual cost to the Federal Government is $551,500.

The information is collected biennially. For each data collection iteration, the cost includes: a) approximately $615,000 for a contractor to develop and manage a database system to house the state’s data; the contractor’s personnel cost associated with survey instrument development, database development, technical assistance to the states, data analysis and production of various reports, b) $216,000 for FVAP data processing, and report development, c) $260,000 for EAC personnel to manage the entire project (including salary and benefits); and d) $12,000 for Government Printing Office (GPO) report design and development. These figures sum to $1,103,000 for two years, bringing the annualized cost to $551,500.

15. Explain the reasons for any program changes or adjustments reported in Items 13 (or 14) of OMB Form 83-I.

The EAC requests a decreased number of burden hours in Item 13 of OMB Form 83-I . As noted in the response to item 12 of this justification, the decreased burden is due to multiple factors, including respondents’ increased familiarity with the EAVS survey, a more accurate estimate of the states' burden after several iterations of the data collections documented in prior years, streamlined questions and instructions, improvements to the data collection mechanisms, and the transition from the Statutory Overview survey to the Policy Survey.

The decreased burden hours entail decreased reporting on the part of the states and their respective jurisdictions. We document the cost of that burden in Item 14 of OMB Form 83-I.

The number of respondents has increased to 56 from the 55 respondents used in previous years, as the EAC intends to collect information for the first time from the Commonwealth of the Northern Mariana Islands through the 2020 EAVS. The Northern Mariana Islands have held federal elections since 2008, but have not been included in the survey since that time.

16. For collections whose results will be published, outline the plans for tabulation and publication.

The EAC is required by the NVRA (52 U.S.C. § 20508) no later than June 30th of each odd-numbered year to submit to Congress a report assessing the impact of the Act on the administration of elections for Federal office during the preceding two-year period and outlining major findings about the administration of Federal general elections= This report for the 2020 Federal general election will be delivered to Congress and publicly released on the EAC’s website by June 30, 2021.

The EAC will also make available to the public the information collected on the combined number of absentee ballots transmitted to uniformed and non-uniformed citizen voters and the combined number of such ballots which were returned by such voters and cast in the election as required by UOCAVA §102(c). The EAC will release its UOCAVA findings concurrently with the release of the NVRA report. All of the data collected through this project will be made publicly available via EAC’s website.

In addition to the report issued by the EAC, FVAP will also make the analysis of its data public from questions contained in Section B of the EAVS once its biennial Report to Congress is submitted in July 2021.

17. If seeking approval to not display the expiration date for OMB approval of the information collection, explain the reasons why display would be inappropriate.

Not applicable to this collection.

18. Explain each exception to the certification statement identified in Item 19 of OMB Form 83-I.

The EAC does not request an exception to the certification of this information collection.

7