On Tuesday, November 5, 2024, Portland, Oregon will find out how the first tabulations of City Council, Mayoral, and Auditor votes work under our new ranked choice and multi-winner voting systems. For City Council elections, three winners will be selected from four brand new districts (the Mayor and City Auditor are elected under single winner RCV).
This new multi-seat method has some very interesting math and campaign implications. Votes can be transferred both from eliminated candidates and from winning candidates who exceed the number needed to win.
This summer, the Multnomah County Elections Division ran a practice election to create sample data and work through the counting procedures. This exercise was a great investment because it allowed the team at the Elections Division to show observers and researchers the detailed mechanics of running these unique elections. It also produced sample data for analysis and exploration. This past week, the team here at the Elections & Voting Information Center (EVIC) has been fine-tuning our use of these data to prepare for Election Day!
The sample election involved 25 candidates and required 25 rounds to identify a winner. In the council races, each round is a step in the tabulation where totals are taken from the first pick for each ballot. In round one, if there is no winner, the candidate with fewer votes is eliminated and those ballots have their second choice moved to the next round. If a candidate wins in a round — which means the candidate has enough votes to exceed the 25% plus one threshold to be a winner — those ballots are allocated as a fraction to second (or third) choices. Rounds are repeated until three candidates win.
This process produces a LARGE table of results that can be hard to read initially. The image below shows the official report – with each round’s result and transfer of votes. Can we visualize this in a simpler form?
To create a visualization we use a variation of the Bump Plot. A Bump Plot shows the changing rankings of things over time. These are popular for showing sports standings over time. But in this case, we are not just showing the ranking – we want to show the magnitude of vote totals. The Area Bump Plot (or Sankey Bump Plot) is a solution here.
We chose to first visualize the eventual winners in three shades of green to help the reader follow the progression of these candidates. Here you see two of the eventual winners started out in first and second place in the early rounds. The third winner started out in sixth place and even dropped a few positions over rounds until Round 16 where other candidates being eliminated added to their total.
On election night, many of us will look at these patterns to see how voters connected candidates in their rankings. In this sample data, two clear front runners never left the top two positions. Here we’d expect candidates with widespread recognition we also see how much of the vote transfers are not going to these candidates, suggesting the third candidate race is distinct in issues or personality.
In this hypothetical example, Alisha Ali starts lower in the rankings and even slips over rounds until Round 16 where the shift begins. In this round, voters that had their first pick (or potentially second, third, or even further choices) eliminated have Ali as their next choice at a sizeable margin – about 30%. In each subsequent round this pattern holds, meaning Ali was a consistent back up choice for voters who supported candidates Ouedraogo, Hornby, Dillon, Itoya, and Kato. In a real race, we would ask: what connects these candidates? Did they share platforms, networks, or other traits? We will wait to ask these questions tomorrow!
Paul Gronke (Reed College) and Paul Manson (Center for Public Service at Portland State University), for the Elections & Voting Information Center in Portland, Oregon*
Current version: November 1, 2024
As many news outlets have reported, on Monday, October 28, 2024, drop boxes in Multnomah County, Oregon and Clark County, Washington were set on fire by someone who attached an incendiary device designed to breach metal and create intense heat. Election officials and law enforcement in both counties have been quick to respond.
This essay is the Elections & Voting Information Center’s (EVIC) contribution to helping citizens understand what happened—to the best of our knowledge—and conveying information about these events to a national audience that is suddenly focused on election administration in our region.
What happened?
- On Monday, October 28, 2024 between 3:00-4:00 AM, incendiary devices designed specifically to breach metal and inject heat were placed on ballot drop boxes in Multnomah County, Oregon, and in Clark County, Oregon. Election officials and first responders were quickly notified of these events.
- Security personnel close to the Multnomah County drop box used a fire extinguisher on the outside of the drop box, and fire suppression equipment was automatically deployed inside the drop box.
- In Clark County, the Vancouver police were notified of a fire at a drop box at 4:00 AM, responded, and contacted the fire department. Fire suppression was ineffective.
- Explosive experts were called onsite in both cases to safely remove the devices (which may have led to the delay in action in Clark County).
- In Multnomah County, three ballots were damaged, but sufficient identifying information remained on the outer envelopes so that the voters could be contacted and given an opportunity to cast a replacement ballot.
- In Clark County, fire suppression for their drop box was far less effective. As of Wednesday, October 30, 488 damaged ballots were retrieved and voters identified. 6 ballots could not be identified, and there may be more ballots that were completely burned. Clark County has been in contact with or mailed replacement ballots to all 488. County and state election officials are reaching out through the media and other channels to encourage anyone who used the drop box in question to use Washington State’s ballot tracking system to determine their ballot’s status and to then contact the election office to see if they need to obtain a replacement ballot.
- Video surveillance in Multnomah County has identified a likely perpetrator. Law enforcement officials are using this video to try to find the perpetrator.
Updates
- Updates: Friday, November 1, 2024
- Clark County will pay “about $134,000” to maintain 24-hour security on all drop boxes.
- Updates: Thursday October 31, 2024
- In Multnomah, Portland Police have released a description of the suspect and have provided more details about the incendiary device.
- Correction (and amended in story): Clark County and Multnomah County obtained their drop boxes from the same vendor. It is unclear why fire suppression failed in one case and succeeded in the other.
- Updates: Wednesday, October 30 2024
- The Clark County Auditor issued a press release updating the situation with the burned ballots. For details, follow the link, but the takeaway is
- Elections staff have been able to identify 488 damaged ballots retrieved from the ballot box. As of Tuesday evening, Oct. 29, 345 of those identified voters had already contacted the Elections Office to request a replacement ballot. Elections staff will mail 143 ballots to the additional identified voters tomorrow, Thursday, Oct. 31.
- The Clark County Auditor issued a press release updating the situation with the burned ballots. For details, follow the link, but the takeaway is
- Updates: Tuesday, October 29, 2024
- Clark County has announced 24-hour surveillance of drop boxes in the County.
- Multnomah County is increasing the frequency of visits to drop boxes by security personnel and has increased security staffing at the Elections Building.
- The New York Times reports that markings on the incendiary devices in both counties had references to the war in Gaza.
The Importance of Drop Boxes
- Vote-by-mail and early voting both involve various ways to return ballots. Many assume that with the name “vote-by-mail”, the primary method of return is the United States Postal Service (USPS). However, this is not the case for Oregon and Washington. In States that provide the option to vote-by-mail, drop boxes and drop sites are the primary way ballots are returned.
The National Overview: Defining and Tracking Usage
According to the Election Assistance Commission’s (EAC’s) 2022 Election Administration and Voting Survey (EAVS) Policy Survey, 39 states allow the use of drop boxes. The Movement Advancement Project’s list of states that have drop boxes is similar to the EAC but with some differences. The National Conference of State Legislatures is another source, but this list is very different from the previous two. The reason for these differences seems to be that there is no single definition of what it means to be a state using drop boxes. Explicitly allowing or prohibiting drop boxes in state statutes seems clear, but there is a gray area of states that are silent, and drop boxes may or may not be being used. It’s also not clear if a “drop site” at a county office or voting center is being counted–this is “not being mailed” but it is also not a drop box.
Finally, the laws and regulations surrounding drop boxes are in flux. Take, for example, the Movement Advancement Project’s coding of Alaska as having “no applicable law or policy”. Drop boxes were used in the state in 2020, but only because the Alaska Division of Elections coordinated with the Municipality of Anchorage and deployed throughout the state. The Division recently announced an end to this program for the 2024 general election. This is one of many areas where comparing laws and procedures across states is hampered by the lack of a common definition used in all states.
In terms of usage, according to the 2022 EAVS, nearly 40% of mail ballots were returned at drop boxes (this figure is based only on states that tracked and reported drop box returns). The use of drop sites (drop boxes, elections offices and polling places) is growing over time, and the use of postal returns is declining.
The percentage of mail ballots nationwide that were returned via postal mail declined from 67% in 2016 to 53% in 2020, according to the MIT Election Data and Science Lab’s Survey of the Performance of American Elections, and shown in the future. The same survey reported that the three States with the longest history of vote-by-mail (Oregon, Washington, and Colorado) had 60% of ballots returned via drop box, an increase from 51% in 2016.
“Drop Sites” and Drop Boxes in Oregon and Washington
“Drop sites” is the term used by state elections officials in Oregon and Washington for designated ballot collection locations. Oregon and Washington election law establishes parameters about the placement and population threshold for the number of drop sites, to assure voters have access for returning their ballots and there is some level of equity of access across the states. Oregon election rules require each county to develop a security plan and review it periodically to assess threats to drop sites. (Security requirements may also be in place in Washington, but we do not know at the time of this writing.)
County offices and vote centers are required to be designated drop sites, along with providing other valuable voter services. But both states allow for additional drop sites, and counties have been innovative in finding locations. The list of sites includes the usual suspects – local city hall and other county or municipal buildings, schools, police and fire stations, and public libraries are commonly used. But other locations are used that may surprise voters from states that make less extensive use of drop boxes.
For example, Oregon’s list of sites includes:
- A sports complex: “outside 24-hour walk-up only drop box located on the south side of the Rose Quarter by the fountain & Rip City sign”
- A Walmart parking lot: “Outside 24-hour drop box North side of parking lot”
- Markets and convenience stores: “Simnasho – Three Warriors Market, Outside 24-hour drop box”
Washington’s list has a similarly diverse set of locations:
- Park & Ride parking lots/mass transit stops: “Shoreline Park & Ride”
- Fraternal organizations: “Lyle Remote Ballot Box. Lyle Lions Club parking lot”
- Gas stations: “In front of Twin Totems Store between Highway 101 and the gas pumps.”
In the EVIC Local Election Official (LEO) Surveys, conducted nationally since 2018, LEOs have consistently told us that they consider improving turnout to be “part of their job” (see the results from our 2022 survey below). LEOs consider security, access, and equity when administering elections, and that includes how they place drop boxes.
We do not know the decision-making process used by each auditor, clerk, and election director in Oregon and Washington, but we are not surprised to see these diverse locations. A tool such as the Center for Inclusive Democracy’s Voting Location and Outreach Tool provides a nice illustration of the many different considerations that go into site selection.
Washington state reports detailed data on dropbox usage since 2013 (we are not aware of other states that provide this kind of detailed report). The percentage of ballots returned via drop box peaked at 73.1% in November 2020 and declined to 61.1% in the 2022 November general election. It is interesting and not fully understood why voters continue to prefer to use a drop box, even in Oregon and Washington where there is paid postage on the ballot return envelopes and ballots need only be postmarked by Election Day.
Why Attack Drop Boxes? What about Blue Boxes?
An attack on drop boxes in these States is similar to an after-hours attack on a polling place where ballots are being held. This is not the first time attacks have been made on drop boxes. In 2020, an estimated 100 ballots were destroyed by an arsonist in Los Angeles, and 35 ballots were damaged later that year in Boston in another arson attack.
There have also been attacks made on “blue boxes” – the familiar boxes used by the USPS that lack many of the security protections of drop boxes. In Lane County, Oregon, two blue box keys were recently stolen, However, these keys were probably not stolen to target ballots. The USPS has been battling stolen “arrow keys” for years. The keys are a gateway to mail theft (stealing checks, identity theft, etc.). This is a low-grade but persistent criminal activity.
Precisely because ballot drop boxes are generally more secure and lead to speedier processing of ballots by local officials, anyone stealing an arrow key is almost certainly aiming for mail other than ballots. This is what led officials in Lane County, Oregon to warn voters not to deposit ballots in the blue boxes.
Other Observations
- Not all ballot drop boxes are equal. The drop boxes used in Multnomah County and Clark County are both manufactured by Vote Armor – Laserfab Inc (but are different models). We are not aware why fire suppression worked in one instance and not another.
- Ballot Tracking is your friend! A recent report, co-authored by researchers at the USC Center for Inclusive Democracy and researchers at EVIC, points out the many advantages of ballot tracking systems, which improve voter confidence in ballot integrity and may improve turnout. This episode highlights once again how valuable ballot tracking can be to allow voters to confirm that their ballot was received and accepted.
- Video surveillance of drop boxes is not legally required in Oregon or Washington. The only reason there was a video of the drop box in Multnomah is that the attack took place on the drop box next to the County office. While it may be desirable to have 24-hour video surveillance of all drop boxes, this may limit the locations, particularly in rural areas with less reliable internet access and potential challenges accessing power.
- Legal penalties for tampering with ballot materials vary across States. The NCSL lists the legal prohibitions against tampering with ballot systems, and we were surprised to find that in at least seven states, this is only a misdemeanor. Professor Rick Hasen makes a strong argument in an essay at MSNBC that any number of federal statutes can and should be used to prosecute arson and elections-related violence.
Policy Recommendations
Violence of any kind is deplorable, but it is particularly dangerous during a hard-fought and contentious election less than four years after an insurrection at the U.S. Capitol.
It is encouraging that no one will be disenfranchised in Multnomah County, but there is a real possibility that some voters in Clark County will not realize that their ballots were destroyed.
We agree with Professor Rick Hasen that legal protections for election workers and penalizing ballot tampering need to be strengthened.
For most states with substantial numbers of by-mail voters, drop boxes are an important part of the election system. States and counties need to evaluate the security of their drop boxes and make sure that they are using infrastructure that meets the highest security standards. These boxes are not prohibitively expensive – we were told that the Vote Armor model costs approximately $6,000, and other drop boxes are in the same range.
While $6,000 is not expensive compared to some election infrastructure, it’s also not insignificant and may be cost-prohibitive in many jurisdictions, particularly in rural areas or in areas with smaller budgets.
We recommend that States provide funding so that all LEOs who want to use drop boxes can afford them, and that all current drop boxes in use be evaluated and if necessary replaced to meet appropriate security standards.
Drop boxes are here to stay. We are not experts in the manufacture and deployment of drop boxes, however, we believe that at a minimum, 24-hour video surveillance and/or other constant monitoring systems should be integrated into all drop boxes used in the United States.
We recommend that drop boxes with monitoring technologies, that can keep election officials apprised at all times of how many ballots have been deposited into a box and have video feeds to record activity around the boxes, be explored and put into use in all election jurisdictions using drop boxes for their by-mail voters.
We also recommend that LEOs, vendors, and the engineering community evaluate the fire protection integrated into drop boxes and make recommendations about best practices for fire suppression
_______________
* Note: This posting will be updated as new information comes out. Please feel free to email the authors (linked in title) with any additional questions or information about this explainer, or contact us via our website at evic.reed.edu.
Links
Local Election Official and Law Enforcement Statements
“Elections Director Tim Scott’s statement on incendiary device at County ballot box”. Multnomah County Elections Division, Oregon. “UPDATED: Clark County Auditor statement regarding ballots destroyed in ballot box.” Clark County Auditor’s Office, Washington.
“‘Your ballot is safe’: Elections addresses voter questions about safety and security after ballot box arson incident”. Press Release: Multnomah County Elections Division, Oregon.
“ Ballot Box Fire Presser 102824 1080”. Press Conference on YouTube: Multnomah County Elections Division, Oregon.
“UPDATE: PPB Releases Photos of Suspect Vehicle in Ballot Box Incident (Photo)” Press release, Portland Police Bureau.
Media Stories
“2 Ballot Boxes set on fire in Portland and Vancouver.” Oregon Public Broadcasting. “FBI investigates fires at Vancouver, Portland ballot boxes that burned hundreds of ballots.” The Columbian.
“Ballot-box fires in Oregon, Washington are linked; police identify suspect’s car.” The Oregonian’s Oregonlive.
“New security measures in place after ballot box fires in Vancouver and Portland” KGW8 (NBC) News.
“Investigators Identify ‘Suspect Vehicle’ in Ballot Drop Box Fires in the Pacific Northwest”. The New York Times.
“Devices with `Free Gaza’ Messages Found at Ballot Box Fires.” The New York Times.
“What to know after ballot box fires in Washington and Oregon”. CBS News.
“Ballot boxes were set on fire in Oregon and Washington. What happens to the votes?” Editorial Essay by Rick Hasen: MSNBC.
“https://www.opb.org/article/2024/10/30/portland-police-ballot-box-bomber-metal-fabrication/” Oregon Public Radio
“Suspect in Vancouver, Portland ballot box attacks ‘intends to continue … targeted attacks,’ say Portland police” The Columbian.
We’re excited to share some new research from the Center for Inclusive Democracy (CID) at the USC Sol Price School of Public Policy and our team here at the Elections & Voting Information Center (EVIC).
In “Vote-by-Mail Ballot Tracking: A Multi-State Analysis of Voter Turnout and Rejection Rates”, our combined team found that voters in Georgia, Colorado and California who used a free vote-by-mail ballot tracking tool during the 2022 midterms reported higher confidence in the electoral process than non-tool users.
This study was led by CID with Sol Price School of Public Policy Assistant Research Professor Mindy Romero, PhD, a political sociologist and the founder and director of the CID, serving as the study’s Principal Investigator.
EVIC’s founder and director Paul Gronke, PhD, political scientist at Reed College, and Lisa A. Bryant, PhD, political scientist, department chair, and 2024-2026 Andrew Carnegie Fellow (Carnegie Corporation of New York) at California State University, Fresno focused on the instrument creation and surveying aspect of this project.
CID Research Associate Anna (Annie) Meier and EVIC Senior Program Advisor Michelle M. Shafer also contributed to this work.
Support for this project was provided by the Election Trust Initiative.
Read the University of Southern California press release about this report here, and read the report itself here.
By Paul Gronke and Paul Manson
Kyle Yoder and April Tan of the Center for Election Innovation and Research (CEIR) released a focus brief on the use and abuse of public records requests, how these requests have impacted local elections offices, and what legislative solutions have been shown to ease the burden placed on local election officials (LEOs).
The research team at the Elections & Voting Information Center (EVIC) has been tracking public records requests in our 2023 EVIC LEO Survey, part of a broader investigation into workloads, staffing, and administrative burdens. The survey provides additional evidence about the dramatic increase in workload resulting from public records requests, particularly in medium-sized and larger jurisdictions where these requests seem to be concentrated.
We also collected nuanced information about how LEOs respond to these requests as part of a deep dive into elections staffing in the State of Oregon. This research, combined with our LEO Survey results, lead us to propose a friendly amendment to the first of the four legislative proposals made by Yoder and Tan.
Their first proposal is to require: Processing Election-Related Records Requests at the State Level
The stated rationale is “a law enacted in the State of Washington in 2023, all requests for records from the statewide voter registration database or for any standard reports generated by the database must be made to and fulfilled by the secretary of state.”
Our research in the State of Oregon highlights a few potential issues with this proposal — mainly it is too limiting. Not all election-related public records requests are for data or reports from the statewide voter registration system. County-specific election records and local election administrative choices are unlikely to be integrated into the statewide database. Second, there is substantial variation in how much administrative authority resides at the local level versus administrative authority at the state (see this overview from the National Conference of State Legislatures (NCSL) for a quick overview).
It seems reasonable to suggest that requests for information from the statewide registration system be forwarded (or have to be submitted to) the state, but that we provide resources and capabilities for localities to respond to other requests.
Our amendment creates a wider umbrella that would ensure that all counties and local jurisdictions have some capacity to respond to these requests. This amendment (additions in bold) threads this needle:
Processing statewide voter registration system election records requests at the state level, standardizing how requests are processed by local jurisdictions, and providing funding to local jurisdictions to sufficiently respond to these requests.
We provide our survey results and rationale for this amendment below.
Survey Evidence on Changes in LEO Workload
Across the board, LEOs have reported to us that their workload has been increasing. Most dramatically, our 2023 survey shows how burdensome election work has become during the “peak” periods (which can extend as long as six weeks before and a week after Election Day).
The reasons for the workload increases are varied, with the largest impact on the smallest offices. These increases are proportionally higher because the typical LEO in a small office dedicates less than 10 hours/week to election work in non-peak times.
What about specific administrative tasks? We asked LEOs how much their workload has changed across different areas, “compared to four years ago”. We chose that time point specifically to encompass one midterm and one presidential cycle.
Three tasks stand out with the highest increase in reported workload over the past four years:
- Preparing for pre-Election Day voting
- Preparing for Election Day voting
- Public records requests
Our survey shows how these requests have become particularly burdensome for medium-sized and larger jurisdictions with the same pattern emerging for “citizen complaints” and “media requests.”
How Clerks in Oregon Deal with Public Records Requests
In late 2023, EVIC interviewed 34 of 36 Oregon clerks to identify more effective and efficient staffing of these offices. Our report, conducted in collaboration with the Oregon Secretary of State’s Division of Elections, highlighted how even in a relatively stable election system like Oregon, which has been fully vote-by-mail since 2000, the workload demands and staffing models have fallen out of sync with the current reality of the job.
And one of these realities is a big increase in public records requests.
Many topics came up in our conversations, but public records were notably high on the agenda of every clerk we spoke to. In fact, the term “public records” appears 228 times across the 33 interviews. What did clerks in Oregon tell us about these requests?
The frequency and quantity of requests has increased enormously. These two quotes are typical:
“We get inundated with public records requests …”
“… you know, go back five years, we would probably get two or three public records requests in an election cycle, now we get two or three a day …”
Also, in the past, requests generally came from the press or groups versed in laws and procedures of that state or locality. Now we are seeing a nationalization of these requests moving through informal social media channels, and citizen activists making requests that can be at times nonsensical. Clerks shared with us, for example, requests for information about polling place locations and poll workers, both of which don’t exist in our state.
All of our respondents said they worked hard to respond to requests, but how they did this varied tremendously. Some told us they developed a request form, charged for staff time, and had access to an attorney. A few told us they’d even added new positions to handle public records:
“… the county has established an office to take that load off of the clerk, because you know, you’re not the only ones who see public records requests …”
“… we’ve built into one of these new positions … will be doing public records requests. In the past, that’s not something we’ve ever considered.”
These clerks are well-positioned to handle the requests, even if the process is being misused.
Other clerks told a very different story, however. Out of a desire to be responsive and transparent, some clerks told us they answered most requests, on the phone or using email. It was not clear that they were documenting the content, frequency, or cost of these requests. A few clerks shared with us that their county had only limited access to an attorney who was not a county employee.
The variation in capacity and in how identical requests were treated across counties raised concerns for us. In our June 2023 report, we recommended:
- Standardize or potentially consolidate public records request processes for various types of emerging public requests.
- Create a new statewide position in the Elections Division to support counties in addressing public records requests.
- Encourage the use of intake forms for public records requests.
- Provide guidance on “boilerplate” or “cookie cutter” requests.
This doesn’t sound so different from the CEIR recommendation, and it isn’t in spirit. Our amended proposal provides more flexibility, allows for county-level processing in some cases, and perhaps most importantly, recognizes that budgets and funding at the local level are critical for the near term.
We can’t keep expecting local election offices to do it all and not fund them adequately.
Nonetheless, these differences are only a very minor shade of grey. We commend CEIR and many others for highlighting the abuse of public records requests and proposing reasonable policy solutions. To the degree that states can help standardize procedures, smooth processing, and reduce the burden on local offices, we are in 100% support.
Clip from a game show never to be aired:
Announcer: “26,824. Is that your final answer, Professor Gronke?”
Gronke: “Let me use my lifeline.”
Clock ticks …
Gronke: “16840.43. That’s my final answer!”
Announcer: “Your final answer to the question `How many election workers
there are in the United States is 16840 point 43??”
Gronke: “Ok, around 20,000. I am very confident that there are around 20,000
election workers in the United States. Or maybe a few thousand more …”
Announcer: On to our next contestant!
The scenario above may never appear on television, but the question is a real one, and one that evades a good answer because there is so little systematic information about the size and composition of the elections workforce.
This post and future posts will provide information about the elections workforce, drawing on results from the 2023 Elections & Voting Information Center (EVIC) Local Election Official (LEO) Survey. We hope this will contribute to efforts to improve the size, diversity, and professionalization in that workforce, and spur other efforts to improve our knowledge base about staff to monitor progress moving forward.
Why Does the Election Workforce Matter?
The quality of American democracy and the integrity of our election system relies in large part on the efforts of the elections workforce: local elections officials (LEOs) and their staff who administer elections in the fifty states, territories, indigenous areas, and in more than ten thousand counties, townships, municipalities, and other administrative units. A long research record has shown that voters express higher confidence in our election system when they have a good voting experience and when poll workers and other election officials are knowledgeable and professional. Election officials and their staffs also act as key intermediaries between political candidates and other groups who want access to the ballot to compete for public support.
A well-trained and professional election workforce is vital to American democracy and it is encouraging to see increasing attention to harassment and turnover among staff, as well as policy and research initiatives dedicated to the issue.
How Many Staff?
As valuable as these efforts are, they are hampered by a lack of systematic information about elections staff.
One place to start would seem to be the most basic: how many elections staff are there?
That seemingly simple question was posed to EVIC in September 2023, while the 2023 LEO Survey hit the field. And it is a question that we should be able to answer since we have been asking about the size of Full Time Employees (FTEs) in elections offices since 2019.
This survey question results in this powerful figure below, showing that more than half of the elections offices in the country have zero or one full-time staff person, and 90% have 5 or fewer staff members.
We can use these responses to produce a first estimate of the total elections staff, and putting aside some details (see the Appendix), 26,824 is the answer we provided. Once we adjust for the largest jurisdictions in the country, “around 30,000” was a defensible answer.
But then research and learning intervened.
We discovered something well-known to practitioners but not brought home to us until we were in the midst of a staffing study commissioned by the State of Oregon. Asking about FTEs in a local office that administers elections will inevitably capture FTEs that work on the property recording side. In some cases, these duties are split, and LEOs engage in “staff sharing” (explicitly moving a staff member from recording over to elections during peak periods).
We asked about this in the 2023 LEO Survey, and the results showed over a quarter of offices engaged in staff sharing. The larger the office, the more common staff sharing is going on (because the smallest offices don’t have staff to “share” – everyone does everything).
A second question added to the 2023 survey tries to elicit a more precise estimate of elections staff:
“Of the full-time staff, how many are fully dedicated to elections?”
The responses to this question are quite striking when compared to the figure above. Notably:
- 32% of all jurisdictions in the country have “no full-time staff” and consequently, “zero” full-time staff dedicated to elections.
- 18% of all jurisdictions have one full-time staff member (the LEO) and half of those are not fully dedicated to elections.
- 40% of all jurisdictions have 2 – 5 staff, and 37.5% of those tell us “none” of the staff are fully dedicated to elections full-time.
There are even a small number of larger jurisdictions (100,001-250,000) that do not have a single dedicated elections specialist. Such is the reality of balancing election administration with recording, licensing, budgets, and the other myriad duties assigned to LEOs nationwide.
This question was my “lifeline” — it allowed me to adjust the staffing estimate (described in more detail in the Appendix). This is my nearly “final answer” of 16,840.43.
Add some wiggle room to adjust for the largest jurisdictions, for an ultimate answer of “an election workforce somewhere around 20,000.”
If It Is Good Enough For Horseshoes … How To Get Better Estimates of Elections Staff
Our intuition is that there are somewhere between 20,000 and 25,000 elections staff nationwide, without considering additional complications like staff sharing, temporary workers, and poll workers. However, getting to that estimate required some fairly heroic assumptions.
We know that state and local governments face workforce challenges. We know that elected officials and election workers face unprecedented levels of threats and harassments for simply doing their jobs.
We believe that the research community can do better.
- EVIC can do better by improving the LEO Survey.
We can ask more precise questions about staffing. In our defense, since the inception of the LEO Survey in 2018, we have labored to make the survey instrument short and easy to complete, avoiding anything that necessitates a LEO looking something up offline. However, conversations with practitioners have convinced us that most LEOs will have a good sense of the size of their staff, and those good senses will be better than the categories we have used in the past. Barry Burden, PhD, of the University of Wisconsin Elections Research Center has successfully used a more specific item in surveys of Wisconsin clerks that we could adapt. - The Census Bureau and U.S .Bureau of Labor Statistics (BLS) can do better by coding election work and election workers.
Two resources to track employment and wages in local government are the Census of Local Governments and the Bureau of Labor Statistics National Occupational Employment and Wage Estimates. For reasons we don’t know, neither of these has a category or occupational code for elections. - The U.S. Election Assistance Commission (EAC) can do better by collecting information on staff sizes and other workforce information.
The EAC may consider collecting information on staff sizes and compensation levels, perhaps as an off-year effort separate from the biennial Election Administration and Voting Survey (EAVS), which is already a burdensome instrument.
Technical Appendix
The technical details of these calculations include projecting our survey estimates to the full population of LEOs, and making certain assumptions about the “multiplier” to use for the response ranges provided in our survey question.
Generalizing from the survey responses to the overall population is the easy part. Our sample is drawn proportionate to the size of the jurisdiction, to ensure a complete representation of medium-sized and larger-sized jurisdictions. Our sampling weight allows us to produce valid estimates. Sampling and weighting procedures are reported on the EVIC LEO Survey Methodology Page.
The more difficult part of the puzzle is choosing the appropriate “multiplier” for each cell in a crosstab of jurisdictions vs. elections staff. To illustrate, consider each of the cells in the table below. Across the columns, we need to choose values to use for “No full-time staff”, for any category with a range, and for “More than 50”. We use “.5” for responses of “no full-time staff”, the midpoint for other categories, and “50” for the largest category. (This calculation, by the way, resulted in our “first answer” of 26,824.)
Looking down the rows, things get more complicated, particularly in the bottom row, “none are full-time elections”. For example, 18% of our sample (160 respondents) said they had 1 full-time staff person (that generalizes to approximately 1440 jurisdictions). Half of those said “none” were fully dedicated to elections, and we use .5 of an FTE for these 78 cases.
We decided to use the .5 adjustment across the whole bottom row, thus (.5 * 1 * 8.8% * 8000) + (.5 * 3.5 * 15% x 8000) + (.5 * 8 * .7% * 8000) + (.5 * 15.5 * .1% * 8000) sums up our staff estimate for the bottom row. It is not clear that .5 is too high or too low.
The final decision is what to do with “more than 50”, fully aware that this column represents the 40 largest jurisdictions in the country, and, from personal communications, knowing that some of the very largest have 750-1000 employees.
Ultimately, we expect our estimate to be somewhat low, but we don’t think it is dramatically off the mark. We hope to collect more detailed information in 2024 to test this assumption.
By Paul Gronke, PhD | Paul Manson, PhD
As our Elections & Voting Information Center (EVIC) readers know, the administration of elections in the United States is a decentralized system with a complex set of diverse institutional arrangements that vary across states and sub-state jurisdictions.
Nearly 8,000 officials spread across 50 states and the District of Columbia hold many titles. They are selected by many different methods and have varying degrees of autonomy from their states (and their counties in the instance of villages, cities, etc., who conduct elections in Michigan, Wisconsin, and New England states). And, of course, they operate in very different political, demographic, and geographical environments.
In the face of this diversity, there are common features, predictable challenges, and a shared professional commitment that connects officials from the smallest Midwestern or New England township to densely populated urban and sprawling metropolitan suburban areas.
Since 2018, EVIC at Reed College has created and continuously evolved its often quoted and highly anticipated annual Survey of Local Election Officials (LEOs). And today, after the dust begins to settle on Election Day 2023 and election officials throughout the country shift focus to the 2024 primaries and presidential election, our team is proud to share our 2023 LEO Survey results and report.
The 2023 LEO Survey was conducted during the off-year (although there is no true “off-year” for election officials) by EVIC in collaboration with SSRS, our skilled survey administration partner and designer of the infographics used in our 2023 report.
In response to feedback from the elections community, researchers, and others working in the democracy space, we learned that a major challenge faced by almost all offices is summed up by one term: resiliency. In this turbulent time of rapid change, competitive elections, and increased public scrutiny, how have these offices adapted and performed?
Key takeaways from the 2023 LEO Survey include:
- Job satisfaction remains high. LEOs say that “integrity”, “service”, and “community” come to mind when they think about what they like about their jobs. When asked what they don’t like about their jobs, “misinformation”, “politics”, and “stress” are at the top of their minds.
- Peak elections workload forces most officials to stretch to their limits, or go beyond them. On a percentage basis, the increased workload during “peak” election season is truly extraordinary – from 50% to 535% higher hours worked during elections as compared to the rest of the year. This is dependent on the size of the jurisdiction.
- Turnover is twice as high as found in prior LEO surveys. However, loss of institutional knowledge may be tempered by lateral movement. For example, the average LEO in the largest-sized (>100,000) jurisdictional category has been in their current position for only 5 years, but has 16 years of experience in elections. We need to know not just about when LEOs depart, but we need to know about who replaces a LEO when they depart.
- Staffing and hiring continue to be a challenge for many election offices. Barriers to hiring include job classifications that have fallen out of sync with the skill sets currently needed to administer elections as well as non-competitive pay. LEOs are divided as to whether the political environment is a barrier.
- Misinformation is a concern among most LEOs. LEOs in smaller jurisdictions rely primarily on face-to-face communication to counter misinformation, while LEOs in larger jurisdictions rely on a broad suite of communications channels that includes social media, websites, email, etc.
On the 2023 LEO Survey page, you will find our report on the high-level survey findings. And for those of you who may be academic researchers or other data geeks (and we mean that in a good way!), you will find the following 2023 resources available to download and comb through, including the 2023 web-based instrument, crosstabs, and codebooks, as well as the codebooks, crosstabs, and questionnaires from our 2018-2022 studies.
The EVIC LEO Surveys also collect metrics to monitor the environment in which these public servants serve, their attitudes toward their work, and the situations and circumstances in which they find themselves. Since the 2020 election, the EVIC team has examined LEO perception of threats and harassment inside and outside of their workplace and the effects of this situation over time.
The 2023 LEO Survey was generously supported by the Massachusetts Institute of Technology Election Data and Science Lab (MEDSL) with additional survey communications support from the Election Trust Initiative to enable us to produce such high-quality infographics.The inaugural LEO Survey of 2018 through the 2022 Survey was funded by our collaborative partner on those efforts – Democracy Fund – without whose support this idea would not have become a reality. In addition to our current and past survey funders and their respective team members (way too many to name), we’d like to recognize our external team at SSRS and our current internal EVIC team: Jay Lee (Reed ‘19), Michelle Shafer, Simon Ahn (Reed ‘24) and Abby Durant (Reed ‘24), for their extremely valuable contributions to the success of this effort.
And last – but most importantly – we thank election officials all over the country in almost every state, for taking the time to thoughtfully respond to our survey and help us in sharing results.
If you are an academic researcher who would like to work with our data, a leader of a national, regional, or state association of election officials interested in having a presentation on the 2023 LEO Survey at your next meeting, or a member of the media interested in writing about the 2023 Survey, please contact EVIC Senior Program Manager Michelle Shafer via email.
Coverage by Nathan Wilk of KLCC Public Radio of the 2023 Oregon Staffing Study.
“It’s a flashing red light on the dashboard,” said Paul Manson, the Research Director with Reed College’s Elections and Voting Information Center.
“We had one jurisdiction share with us that they’re being outbid by the fast food companies,” said Manson. “More common too, we heard they’re even being outbid by other county governments.”
In case you missed our October 8, 2024 webinar “The Impact of EVIC’s Local Election Official Survey Program on Election Science Research and Election Administration” – or want to re-watch this fantastic discussion – the conversation between EVIC’s Founder and Director and Reed College Political Scientist Paul Gronke, PhD, and 2024-2026 Andrew Carnegie Fellows and Election Science researchers Lisa A. Bryant, PhD (California State University, Fresno), and Mara Suttmann-Lea, PhD, Connecticut College), is posted to our new EVIC YouTube channel at this link.
We encourage you to listen to this excellent discussion that took place on campus at Reed College on the morning of October 8th with webinar attendees from across the US and around the word listening live on Zoom. And please share it with your colleagues!
The Elections & Voting Information Center (EVIC) is a non-partisan academic research center with our research leads co-located at Portland State University and Reed College in Portland, Oregon. Led by Founder and Director Paul Gronke, PhD, and Research Director Paul Manson, PhD, EVIC searches for common sense, non-partisan solutions to identified problems in election administration backed by solid empirical evidence.
EVIC’s marquee project is the annual Local Election Official (LEO) Survey which our research team has been undertaking since 2018. The 2024 EVIC LEO Survey – generously funded by Democracy Fund and the Election Trust Initiative – is currently underway and still fielding as of the date of this post. We encourage all LEOs who have received a survey (not every LEO receives a survey every year) to fill it out either online or via paper copy and return it to us. We will be analyzing results soon and beginning to share them in early December. Stay tuned!
If you have any questions on the LEO Survey, or anything EVIC-related, please reach out to me at shaferm@reed.edu. Thank you!