DEMOCRATIC IOWA CAUCUSES

Although most states hold presidential primaries, Iowa holds caucuses. Whereas primary voters’ role in the process ends once they cast their ballot, caucus attendance is a more involved process. Iowa’s Democratic caucuses feature a two-stage balloting process. Before caucus-goers arrive at their precinct’s designated caucus site, precinct captains appointed by the campaigns take up their position in the location.Footnote ² After signing in, caucus-goers congregate around their preferred candidate’s captain. These groups form the first alignment. The precinct chair tallies the number of votes for each candidate. Those caucus-goers who supported candidates who receive at least 15% of the vote cannot change their vote, but those who did not may vote for any candidate when the second alignment is called. In a 15-minute period between alignments, campaign representatives and supporters attempt to persuade other caucus-goers to support their candidate. All candidates with at least 15% of the vote on the second alignment will receive a portion of the precinct’s county-level delegates, who ultimately determine the statewide delegate allocation.Footnote ³

The Democratic Iowa caucus serves as a useful model for a simulation for several reasons. First, its institutions are distinct from other nominations. The discussion-driven nature of caucuses provides opportunities for students to fully engage in conversations in ways that a simulated primary election does not.

Second, Iowa’s traditional first-in-the-nation status contributes to distinct features. The field of candidates in Iowa is larger during the campaign than afterward: less-serious candidates often drop out after lackluster caucus performances.Footnote ⁴ By choosing the Iowa caucuses, I ensured that students participating in the simulation would be confronted with a wider range of alternative choices and would need to give the exercise more thought than they would have if I had assigned a later contest with fewer candidates.

Third, the Iowa caucus often has a disproportionate impact on presidential nominations. For decades, political scientists have considered how the serialized nature of presidential primaries influences outcomes. Early contests typically influence the dynamics of those that follow. Sometimes candidates who exceed expectations go on to greater successes. However, these early successes tend to draw much scrutiny, which can backfire for the candidates’ campaigns (Bartels Reference Bartels1988). In 2020, Vermont Senator Bernie Sanders’s strong performances in early contests lead to widespread concern among many Democratic elites and voters who feared that Sanders would falter against Donald Trump in a general election (Berman and Harris Reference Berman and Harris2020). Focusing on the Iowa caucus allowed the simulation to proceed without an undue amount of attention on prevailing narratives that arose later.

The fourth related reason involves normative concerns surrounding Iowa’s prominence. Many political scientists have questioned the wisdom of having Iowa and New Hampshire—small states with electorates that tend to be demographically and ideologically unrepresentative of the nation—hold positions of prominence in the primary calendar. These concerns intersect with broader concerns about the representativeness of primary and caucus electorates. Voters in these contests tend to be older, wealthier, whiter, and more ideological than the electorate at large (Bartels Reference Bartels1988; Haskell Reference Haskell1996). This simulation does not aim to instill a negative view of primaries and caucuses in students; however, it should encourage them to consider how US electoral institutions elevate certain voices, potentially at the expense of others.

GOALS OF THE SIMULATION

The first goal of this simulation was to furnish students with an illustrative example of how electoral institutions affect outcomes in a democratic system. The way in which candidates are chosen and elected has a tremendous influence over the workings of government and the number of viable political parties in a system (Epstein Reference Epstein1986). Varying electoral institutions structure disparate incentives for candidates, parties, and voters, and they shape differing outcomes across contexts. For example, some states and municipalities recently have adopted forms of ranked-choice voting for primary and general elections. Many other states have abandoned caucuses in recent years, adopting primary-driven presidential nominations instead.Footnote ⁵ Students who participate in this simulation should come away with an understanding of how the Iowa caucus’s structure shapes its results. They also should consider how different institutions might catalyze different outcomes.

The second goal was for students to think critically about how different voters make decisions in elections with different institutions. Before the simulation took place, each student was assigned a unique character. They were given the character’s name, age, and a brief backstory in addition to the character’s issue preferences and their first choice for the Iowa caucus in 2020. Whereas some students likely held similar worldviews to their assigned characters, many undoubtedly held very different opinions on politics and society than their assigned character. In an era of heightened affective polarization in the United States, in which many citizens view those who disagree with them with suspicion and contempt (Iyengar, Sood, and Lelkes Reference Iyengar, Sood and Lelkes2012), it is valuable for students to consider how others’ lived experiences and backgrounds may bring them to different conclusions about events and policies. Students engaged in this simulation should seriously consider their character’s reasons for supporting their preferred candidates and weigh how that character would act if they were an actual person participating in an actual contest.

SIMULATION ASSIGNMENTS AND DEBRIEFING

This simulation was used in an intermediate-level political parties and elections course with a maximum enrollment of 45 students at a large public university in the Spring 2023 and Spring 2024 semesters. I created a series of characters for this simulation. Each was given a brief description, which detailed their issue preferences and motivations, and assigned a most-preferred candidate to support in the first alignment.Footnote ⁶ Students were randomly assigned to a character two weeks before the simulation. The distribution of characters and their preferences were assigned to approximate the preferences of the 2020 Democratic caucus’s actual attendees.

Students were required to write a brief paper about their character’s political views.Footnote ⁷ Given that the campaign websites for the 2020 candidates had long since been taken down, I provided an online database of different candidates’ positions published by Politico during the 2020 primaries (see www.politico.com/2020-election/candidates-views-on-the-issues). Students were encouraged to research other sources of information on the candidates, and they were given advice about how to access contemporary news articles. They were tasked with using these resources to rank their character’s preferences and asked to identify which, if any, candidates their character would be willing to support in addition to their first choice.Footnote ⁸ Students were required to write a defense for their character’s first choice and a justification of their ranking of the candidates, which they brought on the day of the simulation. To incentivize students to participate fully in the preparatory work, those who wrote a strong defense of their candidate and provided a detailed discussion of their character’s rankings were eligible to earn extra credit on the midterm exams.

On the day of the simulation, the class followed the institutions of the Democratic Iowa caucus as closely as the constraints of the classroom allowed. After the students signed in, they were directed to stand in their preferred candidate’s designated position, and the first alignment was held. Despite absences, the Sanders and Buttigieg supporters were locked in after the first alignment in both the Spring 2023 and Spring 2024 simulations. Lively discussions ensued in the 15 minutes between the two alignments. The students took this exercise seriously and participated thoroughly, generally adhering to their character’s preferences and motivations. A debriefing session was held after the simulation. Several students who had changed their vote spoke of their reluctance to support a candidate that their character ranked lower. Multiple students mentioned that they would not have changed their mind were it not for the structure of the caucus. After discussing their initial reactions, the class discussed the broader implications of the Iowa caucus. Students voiced their appreciation for the direct, good-faith debates that the caucuses facilitated, but they expressed concerns about the investment of time that the caucus required from caucus-goers. At this stage, they were presented with data from the actual 2020 Iowa caucus and were informed that their characters likely did not reflect a representative sample of all Iowa Democrats. The Spring 2024 students also discussed the Democratic Party’s reasons for stripping the Iowa caucuses of their first-in-the-nation status that year.

Following feedback from students in the Spring 2023 semester, the Spring 2024 Iowa caucus simulation was held in the wake of a pair of other simulated nominations. Each of the class’s three meetings during the week of the simulation was devoted to a particular exercise. The first exercise focused on the “King Caucus” of 1824, wherein the class simulated the initial style of US presidential nominations. In the King Caucus, members of Congress convened to select their party’s presidential candidate, which closely resembles leader-selection methods used in most contemporary democracies. The second exercise modeled the Democratic National Convention (DNC) of 1968 and was closely adapted from Proctor’s (Reference Proctor2022) excellent simulation. This model of nominating presidential candidates involved a wide-ranging selection of delegates. A majority of these delegates were party elites and elected officials who were free to choose whichever candidate they wanted; a minority were “pledged” delegates who were bound to support a given candidate in the primary. These additional simulations provided students with historical case studies in which outcomes were shaped by institutions.

In the Spring 2024 semester, the day of the Iowa caucus simulation also featured a simulated 2024 Republican Iowa caucus. In most regards, the Republican Iowa caucuses do not resemble their Democratic counterparts. Although the GOP’s caucuses likewise are held in person at a set time in the evening, voters cast secret ballots rather than publicly voting for a particular candidate. Conversations among attendees can and do occur in the early stages of a given precinct’s caucus, and precinct captains do give three-minute speeches for their candidate’s campaign. However, nothing in the Republican caucuses approaches the formalized discourse and opportunity for strategic vote changing present in the Democratic caucuses. Although the three-minute speeches might change certain minds, many Republican caucus-goers attend their precinct with a clear idea of who they are going to vote for that is unlikely to change. This seems to have been particularly true in the 2024 caucuses, which chiefly pitted Donald Trump against Florida’s Governor Ron DeSantis and former South Carolina Governor and UN Ambassador Nikki Haley (Godfrey Reference Godfrey2024). Multiple rounds of voting are held only in the rare instance of an exact tie. However, the Republican Iowa caucuses do resemble Democratic nominations in one key way: their delegates are proportionally allocated, whereas most Republican caucuses and primaries allocate all of their delegates to the plurality winner.

As in the simulated Democratic Iowa caucuses, students were assigned a character with a most-preferred candidate, a brief backstory, and a set of issues about which they chiefly cared. For the sake of time and to avoid giving some students extra work, the three-minute speeches were not simulated in the Spring 2024 semester. Students were instructed to prepare for the Republican caucus the same way as they would for the Iowa caucus. On the day of the simulation, the students were introduced to the institutions of the Republican Iowa caucus, which in many ways resemble a primary more closely than the Democratic Iowa caucus. The first choice of the character assigned to each student functioned as their vote. In a debriefing session, students expressed mixed feelings on the Republican caucus’s differences from the Democratic caucus. They tended to appreciate the Republican caucus’s more straightforward procedures. However, many took issue with the fact that the caucus—which demanded a greater investment of time than a typical primary—ultimately functioned as a primary and did not afford the same opportunities for structured debate and dialogue.

ASSESSING THE SIMULATION

Two questions about the exercise were included in the course evaluation in Spring 2023.Footnote ⁹ The students’ feedback informed the additions to the simulation exercise in Spring 2024: a pretest and a posttest were administered to gauge the educational utility of the Iowa caucus simulation exercise, as well as the other exercises. Given the constraints of the course, it was not feasible to administer the simulation to certain students as a test condition, making these examinations necessary to evaluate the effectiveness of the simulation. The pretest and posttest featured six questions about the Iowa Democratic and Republican caucuses. Several required straightforward, factual responses; others required more input and open-ended thought. A total of 32 students were present on the day of the Spring 2024 simulated Iowa caucus. Table 1 presents their mean and median scores on the pretest and the posttest.Footnote ¹⁰

Table 1

Student Scores on Iowa Caucus Items; Pretest and Posttest, Spring 2024

Note: The survey includes 32 students who were present for the day of the simulated Iowa caucuses.

A few students delivered remotely satisfactory answers on the pretest, but most earned full credit on five of the six posttest responses by delivering correct and detailed answers. The only exception pertained to the differences between the Republican Iowa caucuses and other Republican nominations. Most students either did not identify the key differences—that is, the time investment required of caucus-goers and the fact that the delegates from Republican Iowa caucuses are allocated proportionally—or discuss them sufficiently; accordingly, they did not receive full credit for their response. The results of the posttest suggest that the simulation bolstered students’ knowledge about electoral institutions and their effects on political outcomes.Footnote ¹¹ Most of the students delivered thoughtful, nuanced answers to the fourth question, which asked about the advantages and drawbacks of the Iowa caucuses. Many students emphasized that the structure of the caucuses placed different demands on those individuals whose candidate did not receive at least 15% of the vote on the initial ballot. This pressured them to make difficult decisions about whether to support their most-preferred candidate or another who aligned somewhat with their preferences.

Students also were provided with an open-ended prompt pertaining to their preferences across nomination types. On the pretest, 16 of the 32 students whose responses are presented in table 1 offered answers on their preferred mode of presidential nominations; on the posttest, 31 did so. Moreover, several students changed their answers between the pretest and the posttest. One student who had supported the current primary system changed their answer to the King Caucus; another changed to the convention system. Each of those students cited the drawbacks of primaries and caucuses relative to older models in which elected officials chose their own standard-bearers. Overall, students’ responses to this prompt suggested that they had seriously considered the merits and drawbacks of different institutional approaches covered in the simulations and that they had drawn their own conclusion about these systems of nominating candidates. Many addressed certain caveats of their preferred system. Some students who championed primaries emphasized their preference for proportional methods of delegate allocation. Others stressed the need to take measures that would ease voting and encourage turnout in primaries or caucuses, which tend to be notoriously low-turnout events.

Students in the Spring 2024 simulation also were given the opportunity to directly express their feedback on the exercises through a survey instrument, which was conducted with the approval of the university’s Institutional Review Board (IRB). Sixteen students in the class opted to respond to this survey:Footnote ¹² 14 indicated that they had enjoyed the simulation “a great deal” or “a lot” and 14 indicated that they had enjoyed the week of exercises more than a typical week of lectures. More important, the students’ responses suggested that they believed the simulation had been a worthwhile learning experience that helped them to absorb and retain the course material. Each respondent indicated that they would feel at least as comfortable with an exam question pertaining to materials covered in the simulation as they would with questions drawn from lectures and readings; 13 stated that they would be more comfortable with such questions.Footnote ¹³ Written responses to open-ended questions revealed that different students especially enjoyed different aspects of the simulation. Some wrote about preferring the Iowa caucuses; others favored the King Caucus or the DNC. Other students attested to their enjoyment of the research and preparatory work they had undertaken, and others appreciated the in-class debates and dialogues.

The open-ended survey responses illustrated how the exercise could be improved. Two students indicated specifically that they would have liked a more involved Iowa caucus simulation; one of those students emphasized that they would have preferred a more thorough Republican caucus. An adaptation of the Spring 2024 simulation assignments could facilitate such a development. Every student could be assigned to give a speech or perform a task in character on a given day. A sample of students could be assigned to deliver three-minute speeches for the Trump, DeSantis, Haley, and Hutchinson campaigns. Other students would be assigned to chair the Democratic campaigns in the other simulated caucus, and some would be tasked with advocating for or against a particular candidate in 1824 or 1968 or for specific policies to be included on the 1968 Democratic platform. This exercise would constitute part of their simulation grade and might provide an even more immersive, authentic experience of how caucuses and conventions operated in different eras of American political history. Given that students’ scores on the posttest improved least on questions pertaining to the Republican Iowa caucus, I believe this pivot would bolster the exercise.

A lingering caveat concerns the experiences of students who personally supported the position and candidate they were assigned to defend in the simulated Democratic caucus. One goal of the Spring 2024 survey was to gauge these differences. Seven students indicated that they believed their assigned character for the Democratic caucus happened to be a fairly close match for their personal view. Three of those seven stated that they would have enjoyed the simulation less if their assignment had been farther from their personal view. Of the nine students who indicated that they did not feel that their character was a close match for their personal view, five stated that they would have appreciated the exercise more if they had been tasked with advocating for a position and a candidate that more closely matched their preferences. Because the university’s IRB required the survey to be fully anonymized, it is impossible to ascertain whether students whose character happened to align with their views scored higher on the posttest. However, those five students also indicated that they enjoyed the simulation “a great deal” or “a lot.” It seems plausible to suggest that being tasked to defend a position they did not necessarily align with did not diminish their experience in the simulation. Moreover, when asked to provide feedback on aspects of the overarching simulation exercise that they enjoyed, five of the 16 students mentioned in general terms that they had “enjoyed being assigned a character and trying to get into their head,” as one student put it.

Ultimately, this simulation was a successful classroom exercise. The students in both semesters engaged with their character and the situation. Many directly voiced their enjoyment of the exercise, whether in the Spring 2023 course evaluation or the Spring 2024 survey. By providing the students a space to participate in a modeled version of a salient political event, this simulated Iowa caucus proved to be a highlight of my course. Even after the Democratic Party stripped the Iowa caucus of its first-in-the nation status, this simulation remains a useful illustration of how electoral institutions influence outcomes. Instructors in future semesters may use a simulated Democratic Iowa caucus in the same way as they might use a simulated national party convention or King Caucus. The differing experiences of students in Spring 2023 and Spring 2024 suggests that the simulated Iowa caucus presented in this study is more effective when paired with other simulated nominations. Instructors may use a series of exercises in a given week or across the course of a semester to illustrate the ways in which varying institutional structures generated different outcomes across time in the United States.