How the Iowa caucuses work, part 1: The basics

Expanded and revised from a series published at Bleeding Heartland during the 2008 election cycle

Even in Iowa, many people are confused about what will happen at 7 pm on February 1, when hundreds of thousands of people join their neighbors at a public gathering to express a preference for a presidential candidate. This post will cover the basics of the process and how the results will be reported, which is quite different for Iowa Republicans and Democrats. Over the next several days, this series will examine other elements of the caucus system:

Part 2 will explore barriers to participation in the caucuses and why it can be challenging to turn out even highly politically engaged Iowans.

Part 3 will focus on caucus math, which can create different ways to win an Iowa Democratic precinct.

Part 4 will discuss ways a good precinct captain or other volunteers can help a presidential candidate.

Part 5 will explain why the Iowa caucuses are, in Mark Blumenthal’s words, a “pollster’s nightmare.”

Part 6 will lay out the leading arguments for and against the Iowa caucus system, as opposed to a primary.

Iowa Republican and Democratic caucuses have some common features. Interested voters must gather at specific locations by 7 pm on Monday, February 1. You cannot caucus by absentee ballot or authorize someone to caucus on your behalf. You must be at the location your party has designated for your precinct by 7 pm this Monday evening. In larger cities, most precincts will have their own caucus locations (often in a school or church), but in smaller towns and rural areas, one building may be a caucus location for several precincts, using different rooms. Democrats have set up more caucus sites this year than Republicans have. Democrats from all 1,681 Iowa precincts will caucus in about 1,100 locations, while Republicans will meet in only about 900 locations. Having to drive longer distances to a caucus site can affect turnout, especially if snowy weather hits on February 1.

The Iowa Democratic Party changed its rules last year to allow for a few satellite caucuses to accommodate people “who can’t participate in their regular precinct caucus due to a hardship, such as limitations of mobility, distance or time.” Also for the first time, Democrats living abroad or performing military service outside Iowa may participate in a “tele-caucus,” which will happen online. Registration for those new options has closed, and according to the Iowa Democratic Party, about 90 people have signed up for the tele-caucus and more than 200 for the satellite caucuses. Those numbers represent a small fraction of more than 100,000 Democrats expected to attend their precinct caucuses in person.

Students from Iowa who attend college out of state are allowed to caucus in their home precinct (often the parent’s home remains their permanent voting address). Out-of-state students at Iowa colleges and universities are allowed to caucus in the precinct where they reside. Iowa natives attending in-state colleges or universities are allowed to caucus either in their home precinct or in the precinct where they live when school is in session.

Eligible voters can register to vote or change their party affiliation to participate in either party’s caucus on Monday night. The Iowa GOP is requiring voters to bring photo identification to sign in. CORRECTION: The Republican Party is suggesting but not requiring that voters show ID at the caucus site. Democrats need not show ID or proof of address in the precinct, but they must sign a form stating under penalty of law that they reside in that precinct. Please, no conspiracy theories: large groups of people fraudulently pretending to live in area would be spotted easily at these neighborhood gatherings.

Once everyone has signed in and the temporary chair has called the caucus to order, Republicans and Democrats handle things differently.


Although caucuses are not elections, Iowa Republicans will be able to “vote” for their preferred presidential candidate on Monday night. After a permanent chair and secretary have been elected for the precinct, the chair will give supporters of each candidate an opportunity to make their pitch to the gathering. Well-organized campaigns will have lined up advocates to speak on their behalf in hundreds of precincts, but speakers at the caucuses do not need to be officially designated by the campaigns.

Then blank paper ballots will be passed out to Republican caucus-goers, who write down the name of their preferred candidate, fold the ballot, and return it to the front of the room. Ballots are counted immediately, and representatives of candidates are allowed to supervise the count. The results are announced and the precinct chair is supposed to record the numbers immediately on a form, as well as report them to the Republican Party of Iowa using an app designed for that task.

It’s often said that a candidate needs broad appeal across the state to win the Iowa caucuses. That axiom doesn’t necessarily hold true for Republicans. Deep pockets of support in areas with very high turnout can deliver victory, because a ballot in any precinct counts the same as a ballot in any other precinct.

Reporting the number of votes for each presidential candidate sounds a little like a primary, run by the state party rather than by government officials. Why can Iowa Republicans report raw numbers of caucus-goers supporting each candidate when New Hampshire law stipulates that the Granite state must hold the nation’s first primary? Because the straw poll that happens at Republican precinct caucuses is not linked to the election of county convention delegates.

Many GOP caucus-goers leave after the ballots for the candidates have been counted. Then the precinct chair moves on to other agenda items: requesting nominations for two precinct representatives to serve on the county central committee; electing those precinct committee persons; requesting nominations for delegates to the county convention; electing county convention delegates; requesting nominations for alternates to the county convention; electing those alternates; and considering party platform resolutions.

Typically, only hard-core activists stay to choose convention delegates or debate platform resolutions. For that reason, the candidate who won the most “votes” in a precinct could end up with few or no supporters elected as county convention delegates or alternates. This aspect of Iowa Republican caucuses explains how supporters of Ron Paul were able to seize control of the Iowa GOP’s apparatus in 2012, even though Paul finished third behind Rick Santorum and Mitt Romney on caucus night.


Democrats do not cast ballots at precinct caucuses. The Iowa Democratic Party will not release the number of Iowans who showed up on February 1 intending to support Hillary Clinton, Martin O’Malley, or Bernie Sanders. On the Democratic side, the Iowa caucuses are entirely a battle for delegates.

Precinct caucuses elect county convention delegates. The county delegate totals are converted into estimated “state delegate equivalents” using a mathematical formula. The winner of the Iowa caucuses is the candidate who wins a plurality of state delegate equivalents.

After caucus-goers have signed in and elected a permanent precinct chair and secretary, Democrats express their preference on the presidential race by moving to an area of the room designated for their preferred candidate.

You want your candidate to win as many county delegates as possible, and to do that, you need your candidate to be viable in as many precincts as possible. County convention delegates in each precinct are allocated according to a complicated formula. The most important part of Democratic caucus math is the viability threshold, which candidates must cross to win at least one county delegate from a given precinct.

Most of the 1,681 precincts will elect four or more county delegates, so a candidate will need the support of at least 15 percent of the people in the room to be viable. (In precincts that assign just one county delegate, the viability threshold is 50 percent. In precincts that assign two county delegates, the threshold is 25 percent. In precincts assigning three delegates, candidates will need at least one-sixth of the voters in the room to be viable.)

After Democratic caucus-goers have gone to their candidate’s area in the room, precinct captains count the supporters and report the number to the precinct chair, who announces the totals and which candidates are viable. Democrats then have an opportunity to “realign.” Supporters of candidates who fell below the viability threshold can choose a different candidate, or they can try to persuade people from one of the other groups to come over and help them be viable, or they can try to form an “uncommitted” group. A later post in this series will discuss realignment scenarios in more detail; sometimes small shifts in how many people are standing in your corner change how county delegates are allocated.

It’s also important to note that the number of county delegates for each precinct is fixed ahead of time, based on how many residents cast ballots for the top of the Democratic ticket in the last two general elections. By way of example, my precinct in the Des Moines suburb of Windsor Heights has six Polk County convention delegates to assign in the caucus. That number will not change, whether 50 people or 500 people show up on caucus night. A neighboring precinct that tends to produce more Democratic votes in general elections will have eight delegates to assign. On Monday night, even if 300 people turn out for my precinct caucus and only 100 show up in the other precinct, my precinct will assign six county convention delegates, and the neighboring precinct will assign eight.

The viability threshold and the fixed number of county delegates for each precinct are major reasons why the Iowa Democratic caucus results do not necessarily reflect the raw number of supporters who turned out for each candidate.

For the same reasons, pockets of heavy support in Iowa are less valuable for a Democratic candidate than support spread evenly across the state. This aspect of the caucus system informed my prediction that Clinton will outperform her polling numbers on Monday night. My hunch is that Sanders supporters will tend to be more clustered into a smaller number of precincts, where there is an upper limit to the county convention delegates that can be won.

As mentioned above, county delegates won by each Democratic candidate at precinct caucuses will be converted to “state delegate equivalents” when the Iowa Democratic Party reports the results on Monday night. Let’s look more closely at those state delegates.

During the 2004 Iowa caucuses, 3,000 Democratic state delegates were up for grabs. Party officials reduced that number to 2,500 for the 2008 caucuses. Last year, the State Central Committee voted to assign 1,401 state convention delegates for 2016. They are divided among the 99 counties based on voting for the top of the Democratic ticket in the last two general elections–that is, based on how many ballots in each county were cast for President Barack Obama in 2012 and Democratic gubernatorial nominee Jack Hatch in 2014. CORRECTION: In addition to the 1,401 state delegates that will come from the counties, the satellite caucuses will assign three state delegates and the tele-caucus two, bringing the total to 1,406.

At the end of this post I’ve enclosed a table showing how many delegates to the Democratic state convention have been assigned to each county.

Here are the state delegate totals for the twenty largest counties. I put the main population center of each county in parentheses. These counties will account for 950 state convention delegates, more than two-thirds of the only votes that matter as far as the Democratic Iowa caucus results are concerned (the 1,406 state convention delegates).

County – Delegates
Polk (Des Moines and most of its suburbs) – 228
Linn (Cedar Rapids) – 121
Johnson (Iowa City) – 92
Scott (Davenport and Bettendorf, the Iowa side of the Quad Cities) – 82
Black Hawk (Waterloo/Cedar Falls) – 69
Dubuque (Dubuque) – 48
Story (Ames) – 46
Woodbury (Sioux City) – 36
Pottawattamie (Council Bluffs) – 31
Dallas (far western suburbs of Des Moines) – 29
Clinton (Clinton) – 24
Cerro Gordo (Mason City) – 22
Warren (Indianola) – 22
Des Moines (Burlington) – 20
Jasper (Newton) – 18
Marshall (Marshalltown) – 18
Muscatine (Muscatine) – 18
Lee (Keokuk) – 17
Webster (Fort Dodge) – 16
Wapello (Ottumwa) – 15

Note: Although Scott County has a larger population than Johnson County, Johnson County receives more delegates to the Democratic state convention, based on how the county’s residents voted in 2012 and 2014 (it was the only county Hatch carried in the governor’s race). By the same token, Story, Woodbury, and Pottawattamie counties contain almost the same number of registered voters, but Story County receives more delegates because Democratic candidates perform better in this county, dominated by a college town.

Doing well in the twenty largest counties is important–no candidate who bombs in the main Democratic strongholds will win the Iowa caucuses. However, it’s worth noting that nearly one-third of the state delegate equivalents will be calculated based on county delegate totals coming from the 79 smaller-population counties.

Turnout levels may vary widely from county to county, but delegate totals are fixed in advance. In 2004, 122,193 Iowans came to their precinct caucuses to elect county delegates, who in turn selected 3,000 state delegates. That worked out to almost 41 voters per state delegate.

But in “the People’s Republic of Johnson County,” where Iowa City and the University of Iowa are located, turnout was very high; there were 79 voters per state delegate assigned by Johnson County in 2004. Other counties with a large student population also had relatively high numbers of caucus-goers per state delegate.

Meanwhile, there were more than 30 counties in which it took fewer than 30 voters to elect one state delegate in 2004. At the low end, it only took about 22 caucus-goers to elect a state delegate in Fremont County in western Iowa.

Howard Dean’s perfect storm failed in Iowa for many reasons, but the relatively high caucus-goer to state delegate ratio in Dean’s strongest counties was certainly a factor. The same pattern could reduce the percentage of state delegate equivalents awarded to Sanders on Monday.

That’s all for now. Check back tomorrow for a review of the main barriers to participation in the Iowa caucuses.

UPDATE: Unfortunately, the Iowa Democratic Party did not release overall turnout figures by county for the 2008 or 2012 caucuses. But I would bet that the trend continued of more voters per state delegate in counties containing large cities or major universities.

SECOND UPDATE/CORRECTION: This page includes Democratic Iowa caucus attendance by county in 2008. I calculated how many voters per state delegate attended the caucuses in the eleven counties that combined to allocate about half of the 2,500 state convention delegates the Iowa Democratic Party assigned that year.

In Johnson County, there were 134 caucus-goers for every state convention delegate (137 state delegates, 18,362 caucus attendees).

In Polk County, 44,098 voters attended Democratic precinct caucuses that accounted for 357 state convention delegates (just under 124 caucus-goers per delegate).

Story County had 121 caucus-goers for every state convention delegate (76 state delegates, 9,227 caucus attendees).

Linn County had 90 caucus-goers per state delegate equivalent (202 state delegates, 18,139 caucus attendees).

Scott County had 88 caucus-goers per state delegate equivalent (142 state delegates, 12,552 caucus attendees).

Woodbury County had just under 86 caucus-goers per state delegate equivalent (68 state delegates, 5,836 caucus attendees).

Pottawattamie County had just under 86 caucus-goers per state delegate equivalent (55 state delegates, 4,723 caucus attendees).

Dubuque County had just under 85 caucus-goers per state delegate equivalent (90 state delegates, 7627 caucus attendees).

Black Hawk County had about 80 caucus-goers per state delegate equivalent (117 state delegates, 9382 caucus attendees).

Cerro Gordo County had about 74 caucus-goers per state delegate equivalent (46 state delegates, 3416 caucus attendees).

Clinton County had just under 70 caucus-goers per state delegate equivalent (46 state delegates, 3203 caucus attendees).

THIRD UPDATE: John Deeth did the calculations for all 99 counties for the 2008 caucuses. Click through for the table. Excerpt:

An analysis of 2008 caucus attendance shows that, on average statewide, it took just under 96 people to elect a state delegate equivalent, up from 41 caucus goers per to elect a state delegate equivalent in 2004.

Some counties are full of go-to-meeting activists while others have more rank and file voters.

The easiest place to elect a delegate was Osceola County, where it took 59.67 attendees per state delegate equivalent. The bottom ten is filled with small, rural, GOP leaning counties, with the curious exception of Clinton.

And once again, the hardest places to elect state delegates were campus counties. In Winneshiek, home to Decorah’s Luther College, it took 138 people to elect a state delegate equivalent, 2.3 times as many as in Osceola. Johnson County, the hardest place to elect a delegate in 2004, is narrowly behind Winneshiek at 134 bodies per state delegate equivalent.

Iowa Democratic Party state convention delegates to be awarded, by county

County State delegates
Statewide total 1401
Adair 3
Adams 2
Allamakee 6
Appanoose 5
Audubon 3
Benton 12
Black Hawk 69
Boone 13
Bremer 12
Buchanan 10
Buena Vista 6
Butler 6
Calhoun 4
Carroll 8
Cass 5
Cedar 8
Cerro Gordo 22
Cherokee 4
Chickasaw 6
Clarke 4
Clay 6
Clayton 8
Clinton 24
Crawford 5
Dallas 29
Davis 3
Decatur 3
Delaware 8
Des Moines 20
Dickinson 7
Dubuque 48
Emmet 3
Fayette 10
Floyd 8
Franklin 4
Fremont 3
Greene 4
Grundy 4
Guthrie 4
Hamilton 6
Hancock 4
Hardin 7
Harrison 5
Henry 7
Howard 4
Humboldt 3
Ida 2
Iowa 7
Jackson 9
Jasper 18
Jefferson 9
Johnson 92
Jones 9
Keokuk 4
Kossuth 6
Lee 17
Linn 121
Louisa 4
Lucas 3
Lyon 2
Madison 6
Mahaska 7
Marion 13
Marshall 18
Mills 5
Mitchell 4
Monona 3
Montgomery 3
Muscatine 18
O’Brien 3
Osceola 1
Page 4
Palo Alto 4
Plymouth 7
Pocahontas 3
Polk 228
Pottawattamie 31
Poweshiek 9
Ringgold 2
Sac 3
Scott 82
Shelby 4
Sioux 4
Story 46
Tama 9
Taylor 2
Union 5
Van Buren 2
Wapello 15
Warren 22
Washington 9
Wayne 2
Webster 16
Winnebago 4
Winneshiek 11
Woodbury 36
Worth 4
Wright 5

Video produced by the Iowa Democratic Party to demystify the caucus process:

Maintenance Notice - As of November 14, 2023 we are still seeing issues with replying to comments...Thanks for your patience, this will be restored.

  • Three Tickets

    The Des Moines Register produced a great podcast called Three Tickets on the history and culture of the Iowa caucuses. Its available at the Register website and on iTunes.

  • Democracy can be messy

    First I want to state this anecdote is in now way intended to be accusatory. I don’t feel any ill was intended. But, as with anything, to err is human.

    My precinct, Pleasant Hill 2, had a temporary chair from Ankeny lead the caucus. He repeatedly informed us we had 9 delegates. A new chair was named, and ran the meeting with the temporary chair as secretary.

    We divided the room and undecided and O’Malley were not viable. Of the 175, 67 where Sanders, and about 4-5 where each uncommitted and O’Malley. Knowing we had to get just 2 of these to get a 5-4 delegate split I, as a Bernie supporter, didn’t work the room too hard. When we divided again, Hillary was at 100 and Bernie at 75. To our knowledge this was a 5-4 split. When the secretary (formerly the temp. chair) sat down with the newly appointed chair to send in the results, they discovered the caucus only had 8 delegates. At this point I redid the math knowing it would be a 5-3 split. You can do the %, but all Bernie needed in PH2 was 2 more people to make the delegates 4-4. It was also, given the totals probably the fair result. However, if we had known the math, would we have been able to work a bit more and gain the extra two people needed to preserver a Sanders delegate?

    Unfortunately what was most difficult was having to call up the delegates for Bernie and ask one of them to be an alternate. It was very disappointing.

    The point to this caucus anecdote is not to call sour grapes, but merely to point out Democracy in these situations can be messy, and certainly deserves our participation. Perhaps more attendance, or more advocacy would have changed the result.