Parties, Elections & Voting — Explain political parties,...

SECTION 1

Historical Context & the Rise of Party Politics

The American constitutional framers famously distrusted factions, yet within a decade of ratification, organized political parties had become indispensable to democratic governance. James Madison warned in Federalist No. 10 that factions were a natural byproduct of liberty, and the challenge was not to eliminate them but to control their effects. Despite such warnings, the practical demands of mobilizing legislative coalitions and winning elections drove elites toward party formation almost immediately. The tension between anti-party ideology and pro-party practice has persisted throughout American history, shaping the rules under which elections are conducted and voters make their choices.

Understanding the evolution of parties, elections, and voting behavior requires tracing several critical junctures—moments when the institutional landscape shifted so dramatically that scholars refer to them as party system eras. Each era was defined by a dominant cleavage—slavery, industrialization, the New Deal welfare state, or cultural polarization—that restructured coalitions and realigned voter attachments. The timeline below captures the key transitions that produced the modern two-party system we study today.

1796

First Party System Emerges

Federalists and Democratic-Republicans compete in the first contested presidential election, establishing the precedent of organized party competition despite constitutional silence on parties.

1828

Jacksonian Democracy & Mass Parties

Andrew Jackson's campaign mobilizes a broad white-male electorate through rallies, newspapers, and patronage—ushering in the Second Party System and the modern mass party organization.

1860

Sectional Realignment

The Republican Party replaces the Whigs, realigning politics around slavery and sectionalism. Lincoln's election precipitates the Civil War, illustrating the stakes of electoral competition.

1932

New Deal Coalition

FDR assembles a coalition of labor, ethnic minorities, Southern whites, and urban voters that dominates American politics for a generation and anchors the Fifth Party System.

1968–1994

Partisan Sorting & Polarization

Civil rights legislation, the Southern strategy, and culture-war politics gradually sort ideological liberals into the Democratic Party and conservatives into the Republican Party, producing the highly polarized landscape of the contemporary era.

This historical trajectory raises the central questions this lesson addresses: Why does the United States sustain a two-party system when many democracies support multiparty competition? How do electoral rules shape party strategy? And what factors—social identity, issues, the economy, campaigns—determine how individual citizens vote? The sections that follow build from institutional foundations through individual-level behavior to equip you with a comprehensive analytical framework.

SECTION 2

Core Principles of Parties, Elections & Voting

Political scientists conceptualize parties, elections, and voting behavior through several foundational principles that structure the entire subfield. These principles link institutional design to strategic behavior and ultimately to democratic outcomes. Grasping them allows you to move beyond descriptive accounts of 'who won' and toward analytical explanations of why certain outcomes recur. The concept grid below identifies the five pillars on which the rest of the lesson rests.

Duverger's Law

Single-member, plurality (first-past-the-post) electoral districts tend to produce two-party systems through mechanical effects (third parties win no seats) and psychological effects (voters avoid 'wasting' their vote).

Median Voter Theorem

In a two-candidate race with single-peaked preferences along one dimension, both candidates converge toward the median voter's ideal point to maximize votes—a result from Anthony Downs's spatial model.

Party-in-the-Electorate

Scholars distinguish between the party as an organization, the party in government, and the party-in-the-electorate—the mass of citizens who identify with a party label and use it as a cognitive shortcut when voting.

Retrospective Voting

Voters may evaluate incumbents based on past performance—especially economic conditions—rather than detailed policy platforms. This 'reward or punish' logic underlies many forecasting models of elections.

Turnout & Participation

Rational choice theory predicts near-zero turnout (the paradox of voting), yet millions vote. Explaining turnout requires incorporating mobilization, social norms, expressive benefits, and institutional barriers like registration requirements.

✦ KEY TAKEAWAY

Think of the two-party system like a market with enormous barriers to entry. Just as two dominant firms in an oligopoly can crowd out newcomers through economies of scale and brand loyalty, Democrats and Republicans benefit from ballot-access laws, winner-take-all districts, and voter habit. Third parties are not merely unpopular; the institutional rules structurally disadvantage them—a central insight of Duverger's Law.

SECTION 3

The Electoral Ecosystem: A Visual Map

The relationship among parties, electoral institutions, and voter behavior forms a recursive loop: institutional rules shape party strategies, parties mobilize and frame choices for voters, voter decisions determine who governs, and governing outcomes feed back into institutional design and voter evaluations. The diagram below maps these connections as an ecosystem, highlighting the three analytical 'layers' political scientists use to organize research on elections.

The diagram illustrates the recursive loop of the electoral ecosystem. Electoral rules at the top constrain how parties organize and how campaigns operate. Parties and campaigns then mobilize and persuade voters, whose aggregate choices determine election outcomes—which in turn may reshape the rules themselves.

Notice that the feedback arrow from voters back to electoral rules (the amber loop on the right) is the longest and most indirect. Changing the rules of the game—abolishing the Electoral College, adopting ranked-choice voting, implementing automatic voter registration—requires overcoming the very institutional inertia those rules create. This is why scholars like Paul Pierson emphasize path dependence: early institutional choices constrain future possibilities, making the American two-party, SMSP (single-member, simple-plurality) framework remarkably durable even when large majorities express dissatisfaction with both major parties.

SECTION 4

How It Works: Spatial Models & the Logic of Elections

Although American politics is not typically analyzed with heavy mathematical formalism, several formal models provide powerful analytical leverage. The most influential is the Downsian spatial model, introduced by Anthony Downs in An Economic Theory of Democracy (1957). It treats voters and candidates as actors positioned on an ideological continuum, and derives equilibrium predictions about candidate positioning and voter choice.

VOTER UTILITY FUNCTION (PROXIMITY MODEL)

Uᵢ(C) = −|xᵢ − xC|

Where Uᵢ(C) is voter i's utility from candidate C, xᵢ is the voter's ideal point on a left–right scale, and xC is the candidate's position. Voters prefer the candidate closest to their own position, so utility decreases with ideological distance.

Under the assumptions of a unidimensional issue space, two candidates, and a distribution of voter ideal points with a single peak, the Median Voter Theorem holds: both candidates converge to the median voter's ideal point in equilibrium. The intuition is straightforward—if candidate A positions herself to the left of the median, candidate B can capture a majority by positioning just to the right of A while still being closer to the median voter. This dynamic pushes both candidates toward the center.

MEDIAN VOTER THEOREM EQUILIBRIUM

x*A = x*B = xmedian

In equilibrium, both candidate A's position (x*A) and candidate B's position (x*B) converge to xmedian, the ideal point of the median voter. This result assumes single-peaked preferences, complete information, and pure office-seeking motivation.

CALCULUS OF VOTING (RIKER & ORDESHOOK)

R = (p × B) − C + D

Where R is the net reward of voting, p is the probability that one's vote is decisive, B is the benefit differential between candidates, C is the cost of voting (time, information), and D is the 'citizen duty' or expressive benefit. Since p approaches zero in large electorates, R > 0 typically requires D > C—explaining why social norms and intrinsic motivation matter more than instrumental calculation.

🔍 Why Don't Candidates Always Converge?

The Median Voter Theorem's prediction of convergence is elegant but rests on restrictive assumptions. In practice, primary elections force candidates to appeal to ideologically extreme party bases before pivoting to the general electorate. Additionally, multidimensional issue spaces (e.g., economic vs. cultural dimensions), valence issues (competence, character), and asymmetric information about voter preferences all break the convergence result. Contemporary polarization suggests that office-seeking is only one party motivation; ideological policy-seeking matters too.

SECTION 5

Determinants of Voting Behavior

Understanding why individuals vote the way they do has been a central preoccupation of American political science since the Columbia school's pioneering studies in the 1940s. Over decades of research, three major models of vote choice have emerged, each emphasizing different causal mechanisms: the sociological model, the social-psychological (Michigan) model, and the rational choice model. These approaches are not mutually exclusive; contemporary scholarship synthesizes elements from all three. The diagram below organizes the key determinants of voting behavior into a multi-level causal funnel, moving from long-term structural factors at the top to the proximate influences that shape the final vote decision.

The funnel of causality illustrates how long-term factors like social structure filter through party identification and issue attitudes to shape candidate evaluations and the ultimate vote decision. Annotations on the right show which scholarly tradition emphasizes each level.

The Columbia school, associated with Lazarsfeld, Berelson, and Gaudet's The People's Choice (1944), emphasized that social group memberships—religion, socioeconomic status, urban vs. rural residence—powerfully predicted vote choice, often more than campaign persuasion. The Michigan model, developed in Campbell et al.'s The American Voter (1960), placed party identification at the center of the explanatory framework—a stable psychological attachment formed early in life through socialization, acting as a 'perceptual screen' through which voters interpret political information. The rational choice tradition, building on Downs, emphasized that voters weigh policy positions and incumbent performance.

Contemporary research recognizes that party identification is not perfectly stable; it can update in response to major events, elite cues, and shifting group alignments. The concept of partisan sorting—the alignment of ideology, identity, and partisanship that has intensified since the 1970s—means that party ID now correlates more tightly with issue positions, racial attitudes, religious observance, and cultural preferences than at any point in the survey era. This sorting complicates efforts to disentangle which factor is truly 'causal' in shaping vote choice.

Three classic models of voting behavior

Model	Key Variables	Core Claim	Key Work
Columbia (Sociological)	Religion, class, region, ethnicity	Social group membership determines vote; campaigns reinforce rather than convert.	The People's Choice (1944)
Michigan (Social-Psych)	Party ID, candidate image, issue positions	Party identification is the central, stable attitude that filters all political perceptions.	The American Voter (1960)
Rational Choice	Policy distance, retrospective evaluation, information costs	Voters are utility-maximizers who use spatial proximity or past performance to choose candidates.	An Economic Theory of Democracy (1957)

SECTION 6

Worked Example: Analyzing the 2016 Presidential Election

To integrate the concepts developed so far, consider a step-by-step analytical breakdown of the 2016 presidential election. This exercise demonstrates how institutional rules, party dynamics, candidate positioning, and voter behavior interact to produce an outcome that surprised most observers and pre-election forecasts.

Applying the Electoral Ecosystem to 2016

Step 1 — Identify the Electoral Rules at PlayThe U.S. presidential election uses the Electoral College, a system in which 538 electors are apportioned across states (mostly winner-take-all). A candidate needs 270 electoral votes to win. This means the popular vote and the electoral vote can diverge, which is exactly what happened in 2016.

Key structural fact: Electoral College rewards geographic distribution of support, not national popular vote totals.

Step 2 — Assess Party Positioning & the Primary ProcessDonald Trump won the Republican primary by appealing to anti-establishment sentiment, immigration restrictionism, and economic nationalism—diverging from the party elite's preferred ideological positioning. The open primary system in many states allowed crossover voters and irregular participants to influence the outcome. On the Democratic side, Hillary Clinton won a contested primary against Bernie Sanders, positioning her toward the center-left but leaving portions of the progressive base unenthusiastic.

Primary rules shaped which candidates emerged, illustrating how intra-party institutions constrain general-election choices.

Step 3 — Apply the Funnel of Causality to Voter BehaviorSocial structure: white working-class voters in the Rust Belt shifted toward Trump, driven by economic anxiety and cultural grievance. Party ID: most voters still followed their habitual partisanship—roughly 89% of Republicans voted Trump and 89% of Democrats voted Clinton, per ANES data. Candidate evaluation: both candidates had historically high unfavorability ratings, making short-term candidate image less decisive in favor of partisan loyalty. Retrospective evaluation: satisfaction with Obama's economic performance was mixed; swing voters in key states weighed stagnant manufacturing employment heavily.

Party ID dominated nationally, but marginal shifts in social-structural voting patterns (white non-college voters in WI, MI, PA) proved decisive under Electoral College rules.

Step 4 — Synthesize: Why Did the Outcome Surprise?Most forecasting models and polls predicted a Clinton victory because they either underweighted the geographic concentration of Trump's gains or overestimated Clinton's support among lower-propensity voters. The Electoral College amplified small state-level margins into a decisive 306–232 electoral vote win for Trump despite Clinton's approximately 2.9 million popular vote lead. This outcome illustrates the disconnect between the Median Voter Theorem's national logic and the geographically mediated reality of American elections.

Institutional rules (Electoral College) interacted with voter behavior (Rust Belt realignment) to produce a popular-vote/electoral-vote split—the second in 16 years.

SECTION 7

Strengths and Limitations of Major Electoral Frameworks

Each analytical framework introduced in this lesson illuminates certain aspects of parties, elections, and voting while leaving others underexplored. Responsible scholarship requires acknowledging these trade-offs. The table below compares the key frameworks on several evaluative criteria.

Comparative evaluation of major electoral frameworks

Framework	Strengths	Limitations
Duverger's Law	Parsimonious explanation for two-party dominance; empirically supported cross-nationally; clear causal mechanism linking electoral rules to party systems.	Does not explain third-party persistence at state/local levels (e.g., Farmer-Labor in MN); weaker in explaining transitions between party systems; treats voter psychology as uniform.
Median Voter Theorem	Elegant formal logic; generates testable predictions about candidate convergence; foundational to spatial modeling in political science.	Assumes unidimensionality, complete information, and two candidates; poor empirical fit in era of polarization; ignores primary electorates and activist influence.
Michigan Model (Party ID)	Strong empirical track record; party ID is the single best predictor of vote choice in most elections; captures habitual/affective dimensions of voting.	Treats party ID as exogenous when it may be endogenous to issue positions and candidate evaluations; less equipped to explain realignment and dealignment.
Retrospective Voting	Low information demands on voters; fits well with economic voting models (e.g., 'Time for a Change' models); consistent with democratic accountability logic.	Voters may misattribute responsibility (e.g., blaming presidents for global recessions); emphasizes valence at the expense of positional issues; limited in explaining non-economic elections.
Calculus of Voting	Highlights the paradox of turnout, prompting richer theories of participation; forces scholars to account for non-instrumental motivations.	The D term (duty) is often a residual that absorbs everything the model cannot explain; instrumentally, p × B ≈ 0 in large electorates, making the model empirically weak without D.

✦ KEY TAKEAWAY

Think of these frameworks as different lenses in a photographer's kit. A wide-angle lens (Duverger's Law) captures the broad institutional landscape beautifully but blurs individual details. A macro lens (the Michigan model of party ID) reveals the fine grain of individual behavior but loses the institutional backdrop. No single framework provides a complete picture; sophisticated analysis requires switching lenses strategically depending on the question at hand.

SECTION 8

Connections to Advanced Theory: Polarization, Realignment & Behavioral Political Science

The foundational models covered in this lesson serve as launching points for some of the most active research frontiers in contemporary political science. Three areas in particular extend and complicate the frameworks you have learned: affective polarization, realignment theory, and behavioral political science. Understanding where the introductory frameworks end and these advanced research programs begin is essential for navigating upper-division coursework and the scholarly literature.

From introductory frameworks to advanced research frontiers

Introductory Concept	Advanced Extension	Key Scholars / Works
Party identification as stable attitude	Affective polarization: party ID as social identity driving out-group hostility, not just policy disagreement.	Iyengar & Westwood (2015); Mason, Uncivil Agreement (2018)
Party systems as stable equilibria	Realignment / dealignment: critical elections that restructure coalitions, or secular decline in party attachments.	Key (1955); Burnham (1970); Mayhew, Electoral Realignments (2002)
Rational voter choosing on issues	Behavioral political science: cognitive biases, motivated reasoning, heuristics, and media framing shape voter perception of issues and candidates.	Tversky & Kahneman; Lodge & Taber, The Rationalizing Voter (2013)
Median Voter convergence	Asymmetric polarization: Republican Party has moved further right than Democrats have moved left (measured by DW-NOMINATE scores), challenging symmetric convergence predictions.	Hacker & Pierson (2005); Mann & Ornstein (2012); McCarty, Poole & Rosenthal

As you move into advanced coursework, keep in mind that the clean analytical separations drawn in introductory courses—between institutions and behavior, between party ID and issue positions, between rational and irrational voters—increasingly blur under empirical scrutiny. The frontier of the field lies in understanding how these factors interact dynamically. For example, elite polarization in Congress may be both a cause and consequence of mass sorting among voters, creating a feedback loop that the foundational models—designed for a less polarized era—struggle to capture on their own.

SECTION 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

Explain the difference between the mechanical effect and the psychological effect in Duverger's Law. Why do both effects reinforce the tendency toward a two-party system under single-member, plurality rules?

PROBLEM 2 — BASIC CALCULATION

Using the Riker-Ordeshook calculus of voting (R = p × B − C + D), suppose a voter estimates p = 0.00001, B = 500 (utility units), C = 10, and D = 12. Will this citizen vote? What does this exercise reveal about the role of D in explaining turnout?

PROBLEM 3 — INTERMEDIATE

Consider a one-dimensional ideological spectrum from 0 (extreme left) to 100 (extreme right) with voter ideal points uniformly distributed. Two candidates, A and B, position themselves at 40 and 60 respectively. According to the Median Voter Theorem, is this an equilibrium? If not, describe the strategic moves each candidate would make and identify the Nash equilibrium.

PROBLEM 4 — APPLIED

In the 2020 U.S. presidential election, Joe Biden won the popular vote by approximately 4.5 percentage points yet his Electoral College margin depended on roughly 44,000 votes across three states (Arizona, Georgia, Wisconsin). Using the concepts of Duverger's Law, the Electoral College's institutional design, and retrospective voting theory, construct a multi-factor explanation for why the popular vote margin and the 'tipping-point' margin can diverge so dramatically.

PROBLEM 5 — CRITICAL THINKING

Some reformers argue that adopting ranked-choice voting (RCV) and multi-member districts would break the two-party duopoly and reduce affective polarization. Critically evaluate this claim using Duverger's Law, the Michigan model of party identification, and the concept of affective polarization. Under what conditions might the reform succeed or fail to achieve its stated goals?

SUMMARY

Lesson Summary

This lesson examined how political parties emerged despite the framers' anti-faction sentiments, evolving through distinct party system eras defined by shifting cleavages. We established that Duverger's Law explains why single-member, plurality rules produce a persistent two-party system through mechanical and psychological effects. The Median Voter Theorem predicts candidate convergence toward the center in idealized two-candidate races, though real-world polarization and primary elections complicate this prediction substantially.

On the voter side, the funnel of causality integrates long-term factors like social structure and party identification with short-term influences such as issue attitudes, candidate evaluations, and retrospective economic assessments. The calculus of voting reveals that rational self-interest alone cannot explain why people vote; expressive and duty-based motivations (the D term) carry the explanatory weight. Looking ahead, affective polarization, partisan sorting, and behavioral political science represent the cutting edge of research, extending and challenging these foundational frameworks in productive ways.