What has been produced by political scholarship,
On American electoral behavior,
O’er the past 70 years?
Map for me the electoral research,
And tell me where ‘tis going,
Where they shall steer…
- I pray you have the time.
All the time in the world.
- Then I’ll answer [with a poetic mural].
- But ask me to yield,
- If you find that muddy points…
- Should become clear.
I await your word-reel.
- Voting research in 1940 began,
- At ColumbiaUniversity.
- ‘Tis called the Columbia Studies,
- Sophisticated research they pioneered.
- Lazarsfeld’s team surveyed 600 possible voters in a single community,
- In 1940 and 1948.
- ‘Twas two panel studies: many questions in successive interviewing,
- The [latter] Elmira one acquired fame.
- The design illuminating personal vote choice,
- During the campaign:
- Did voters flip-flop ‘tween candidates?
- Did voters even know their names?
- The “psychology of choice” found something striking; so try to accept:
- Voters confirmed that though religious and social class ties were led,
- With fellow acquaintances, and, at the polling place, they said:
- How were born is how we’re bred.
But I am confused?
What was the point of their review?
- ‘Twas conventional political science wisdom at the time,
- That voters were rational—that they calculated,
- For which President to chime.
- It revealed that political parties and the media,
- Were not the main attraction,
- Within the voter’s mind.
- It told that Culture was the culprit—Culture was the why!
- Thus began electoral sociology—these were new times!
- The best part was that their research explained new events,
- Like Truman’s latent Democratic resurgence,
- That loyalty was their drive—
- Electoral studies thus arrived.
Today, I hear, that Culture is undervalued,
That cultural explanations are forming,
And new analysis will be surging,
Like a tidal wave.
- May be true—that Culture’s finally found the throttle.
- But let’s move from the ‘40s to the ‘50s—bit by bit.
- Beginning with “The Michigan Model.” Sit, sit.
- Campbell and Kahn’s study was on foreign policy attitudes, [the original view],
- But the final questions of the survey explored political interests too.
- Whether they were voters, and if so, who would they vote for?
- Then something strange occurred. [Like a tame-boar],
- A Gallop poll failed to forecast Truman’s late rush,
- So it all seemed quite skewed, to us, an embarrassing fuss.
- Gallop predicted that Dewey would be President.
- Thus, political polling seemed bogus—and losing trust!
- Thus Campbell and Kahn chose to reinterview everyone;
- Whom had previously affirmed that they’d vote for someone,
- And for who’d they choose in that surprising election…
- Were there statistical defections?
- What were the voters’ reflections?
- Campbell and Kahn evaluated the demographics and attitudes—
- [the characteristic’s search].
- Of both voters and non-voters, of Democrats and Republicans.
- ‘Twas the important survey research.
- They analyzed psychological, sociological, and political factors;
- Factors that comprised the people’s vote. And their study now stands,
- As one of the longest political science endeavors, I should note,
- On American sands.
- In that book they attributed Truman’s win to Party ID.
- There simply were more Democrats than Republicans,
- In that election, you see?
Yes. And there were later surveys
Integrated into their analysis.
- Yes. National election survey data from the presidential elections
- Of ’52 and ’56 sparked national investigations.
- Designed to test theories of voting behavior, and a consequential
- Historical record gave it much purpose, Sine qua non.
- ‘Twas brilliant and enlightening—what they found,
- That Eisenhower’s landslide victory did not amount,
- To a shift in party loyalty. No. Not at all!
- ‘Twas a short-term deviation—One Democrat’s fall.
- And so millions were still loyal to the Democratic Party, and,
- Though party loyalty and social explanations are weak,
- When explaining the citizens’ choice, we begin to see,
- Why the candidates began to seek,
- The voters in-between.
Ike’s election was a “deviation?”
- ‘Twas their evidence, and the most refreshing bit,
- Was upon their political attitudes analysis.
- Party loyalties flowed from the New Deal, they said,
- A competent man [FDR] had regained control from prior incompetence.
- But in ’56 political science recorded a downplay of FDR rhetoric.
- No longer was “depression” or “prosperity” heard by the electorate.
- Rather negative personal attacks on Stevenson—
- The democratic candidate.
- Partisan loyalties are relatively stable but can change.
- There is “a persistent adherence and a resistance
- To contrary influence.”
- Millions to the party label demonstrate maintenance.
- Unless they all switch—then it’s a realignment.
- Let us speak more of the attitudes,
- Because the American Voter reiterated,
- That the Columbia Studies rang true—
- That political information, engagement, and reasoning,
- Were much less widespread than the elites knew.
- Rather, the people shifted upon new societal goals,
- ‘Tis not through ideology necessarily that they choose.
- No. They found no “coherent patterns of belief.”
- This implied that elites minimally advocate:
- The muse.
- Thus the American voter was powerfully influenced:
- By partisan loyalties established early in life,
- By short-term popularity manipulation, even if trite,
- By dim ideology and dim policy agendas—
- Voters were not enlightened—not politically bright.
And what do the critics say?
Have they smashed it apart to-date?
- They said a lot,
- Which is easy to do,
- When you examine other things,
- To explain a similar pursuit.
- But not much political science to my mind,
- More took out the wind,
- Than the historical events of those times.
- Like the civil rights and anti-war movements.
- ‘Twas ideologies engaged and policy agendas aflame!
- On the ground, in the capitals, in all towns!
- The peoples’ ideology and agenda became:
- Both verbs and nouns!
- The result was to question the premise of partisan loyalty
- As advanced by the Ivory Tower.
- Was the stable electorate just momentarily engaging frivolity?
- For the “Solid South” altered power!
- From democrats to republicans! They cowered!
- From democrats they deviated and departed! That hour!
- Revisionists then entered the American voter stage.
- They argued that the structure of political thinking had changed.
- That issues were now the dominant variable in play.
- That ideology was much higher than The Voter claimed.
The Changing American Voter, right?
And didn’t the Michigan team consent?
That the quality of mass attitudes indeed reached new heights?
- Miller et al. agreed, but demanded that ‘twas leadership
- Who brought the substance of issue politics to ‘ere sight.
- Then, suddenly, from the critics’ darkness—sobering light.
- Scholars surrounded the attitude consistency topic.
- Questioning why the issues thermometer from ’60 to ’64 jumped.
- Was it a significant change in question wording format?
- Yes! The Changing American Voter was a “wording” artifact!
- The “smarter” electorate did not earn a quickie brain-bump.
- The tit-for-tat among scholars made them look like…
They were going through another scientific slump.
- But then another National Election Study (NES) by the Michigan team,
- Reviewed political attitudes from ’72 through ’76.
- Compared it to their data from the ‘50s—
- ‘Twas real individual partisanship stability!
- And static individual preference continuity!
Then what, to you, is the revisionist worth?
- The study of voting behavior then changed in two ways:
- We started looking at elites vs. citizen interactions.
- Like examining Key’s “Echo Chamber”—
- That peeps choose by what they’re shown,
- From some recognized players known.
- We started taking serious our measurements.
- Like the pitfalls and power of our awesome,
- Survey instrumentation.
Is it all surveys?
- So I began with an introduction to the sociological school,
- Followed by the behavioralists / historical institutionalists.
- Let us explore the rational choice theorists too.
- The original architect is Anthony Downs,
- The book: An Economic Theory of Democracy,
- Is widely respected and renowned.
- For it was the first to expound,
- A hypothesis of how electoral politics found:
- That rational government and rational parties,
- That a Spatial Model could decree success; really,
- ‘Twas profound.
- This Downsian theory is outlined in:
- “The Statics and Dynamics of Party Ideologies.”
- Downstold of bimodal and multimodal theories,
- Explicitly economic rational choice inquiries.
- Imagine that voters range from the far left to the far right,
- That this is a one-dimensional range—a continuum.
- Bright voters know their accurate place and thus their party’s place,
- Upon this ideological spectrum.
- Downs then brought up a pertinent point, which still rings true today:
- That competing parties in a two-party system must appreciate,
- That the median voter shall usually appoint—shall choose,
- The next politician to fill the Representative’s shoes.
- Because the parties will “converge rapidly upon the center”
- Chiefly in places where most voters are moderate; like theU.S.
- So political scientists then began to rapidly address
- This Median Voter Theorem, for instance:
- If the proportion of voters on each side,
- Is about the same, regardless of group pride,
- Advise that they cancel each other out.
- That de facto the independents’ decide the victor:
- Which Candidate remains alive.
Left [——–>—->–MV1–<—-<——–] Right
MV1 = Median Voter Won
Did they integrate this theory into surveys?
- Yes! Issue voting changed. The NES began in ’68,
- To humbly request that individuals confess
- Along a downsian [spectrum] seven point scale:
- Their ideological place;
- Where their candidate(s) is consigned; and,
- The space that the parties occupy.
- Political scientists soon thereafter measured the data, and noticed:
- That voters’ positions and candidate consignment
- explained electoral choice.
- This created sincere focus.
- Downs had also said in the original book,
- Something that many preferred to overlook.
- That rational behavior in a massive society like the States, means:
- One person’s vote is quite meaningless, read:
- One vote will likely never change the course of an election,
- Thus ‘tis not rational to investigate candidate representation…
- In order to understand the candidate you personally select…
- Your cost is not worth your benefit.
- Indeed, this finally received more attention.
- Fiorina (1981) integrated Downs’ rational choice insight upon,
- Retrospective voting as candidate reflection.
- Typically, Fiorina said: The voter has the last 4 years in their head.
- That the variables to consider were instead:
- Voters’ thoughts about the economy, foreign policy; the President.
- Thus, a political party’s platform or policy—wasn’t necessarily—
- The cross-cutting element to predict who would succeed.
- This study of retrospective voting was hence challenging,
- To the traditional explanatory variable of Party ID.
- Indeed, retrospective voting called for the rebirth of V.O. Key,
- That voters adhere to conditional perceptions; that elections,
- Are post hoc political accountability.
So far I think,
That the early theorists believed,
That people were much more informed,
Than the later evidence showed.
And if this is true,
What did political scientists do?
Was this retrospective voting thing,
Just a way of accounting for an inattentive citizenry?
- Low information rationality studies,
- Accelerated in the ‘90s.
- Popkin advanced the concept,
- of information shortcuts.
- Lupia added that Californians choose,
- Referendums based on cues.
- Page and Shapiro added up all the votes of the “inattentive citizenry,” and,
- They note Converse’s “miracle of aggregation;” which established,
- That individual inattentiveness is nullified by the summative vote,
- That the electorate, at large, is stable, rational, and competent.
Don’t these scientists perform experiments?
- Think about Bartels info as an exhibition:
- In six presidential elections,
- Uninformed voters beat the odds,
- Since they beat the statistical stops,
- Of what they otherwise would have chosen.
- Without shortcuts for recollection’s use,
- Voters might have seemed obtuse.
- But via cues voters rose 3 percentage points— improved,
- Though they would’ve improved much more,
- Had they been experts in the voting booth.
- Fournier established a similar story,
- In that quiet country north of the States.
- He also found evidence that collective biases,
- Are not impacted by the campaigns.
- Bartels [and Achen] though noted,
- That “straightforward” retrospective voting is not full-proof,
- It may be “significantly skewed by systematic errors such as myopia”—
- Quite bloated,
- For people may not know whether ‘tis good or bad times—
- Seriously aloof,
- About misattribution, and, rationality too.
Electoral behavior must not be scientifically favored,
Clearly, “rationality,” is much more like an odor.
Changing when danger and hardship are savored…
How do you find, in voting behavior, causal orders?
- The American Voter consumes the metaphor: “funnel of causality.”
- In short: Understand your DV and watch closely your IV.
- Record the events as they converge and change, e.g.,
- Voter’s attitudes, economic events, and the military.
- Accounting for cross sections against the DV,
- Engaging a selective strategy—reviewing the history,
- Also recall the social and institutional things,
- Like electoral change and exogenous frames, which see:
- That temporal and causal priorities are utterly linked.
- The behavioral revolution started to make the former metaphor more concrete.
- Goldberg  followed the funnel when he undertook
- Recursive causal modeling.
- Exhibiting multi-equation models regarding partisan attitudes to:
- Vote choice, parental influences, Party ID, and sociology pools.
- Thence a “best fit” was nominated as the “pivotal position” and, perceive,
- That mathematics became mandatory reading.
- On the other hand, Brody and Page (1972) foretold:
- [Correlation is not causation, you know],
- Issue proximity and vote choice as inextricably linked may fold,
- For voters have psychological projections they hold,
- Or voters are persuaded [by other locals],
- Thus, go slow on the prima facie show.
- As another mold:
- Jackson(1975) championed the non-recursive causal model,
- ‘Tween party affiliations and party evaluations, and concludes:
- People’s appraisal of the parties platform, corresponding
- To their own preferences results in whom s/he will choose.
- That party affiliations only matter when,
- The voters recognize little party difference!
- Otherwise, Party ID appears mute!
- Jacksonthus shakes the “Michigan Model” ferociously, challenging:
- First, the dissimilar impact of Party ID, and,
- The statistical analysis of reciprocal possibilities.
How did they resolve this thing?
- Two articles in APSR reviewed this issue in 1979.
- Almost identical simultaneous equations to answer this puzzle would find,
- That Party ID is weaker than it seems, or, stronger than you know!
- For the impact of exactly the same IVs [the following]:
- Vote choices, candidate evaluations, issue preferences, and Party ID; deem:
- Different results occur when statistical models are tweaked!
- So the following silently occurred…
- …lamenting political scientists at a “Causal” dirge.
- Alas, upon exit from the dirge,
- Some political scientists began to experiment.
- Measuring variation about electoral behavior,
- They are finding causal leverage, in are coming closer,
- To scientifically realizing the electoral picture.
- And we started to relook at other data contextually. Like Markus in ‘88:
- Personal and collective economic circumstances calculate,
- That “better off” voters chose to reelect the incumbent’s
- Presidential candidate by 8 percent,
- As opposed to citizens whose pockets remained stagnant.
- Today, Green and Gerber are renowned,
- For tabulating field experiments from the ground,
- They peek at the significance of the get-out-the-vote.
- Do you vote?
When dodging Downs,
- Good. That’s true [inaudible laughter].
What then could I research?
Are there blind spots for me to choose?
- If you want to continue with electoral behavior,
- Write a much more significant literature review,
- Than the paltry verse that I have provided you.
- Other than that as a start…
- There is a real need for a person or two,
- To interpret all of the articles in our field.
- What are the deeper implications
- Of the analysis and relationships ‘at are continuously real?
- Say, how does sociology or psychology predict politics—
- Particularly when the voter is looking for a savior—
- In a polity where parties appease the median voter?
- Where Congress is a bunch of median voter promises loaded?!
- Oh, and culture. Like you said…
- Culture is far from dead.
- So start with the political languages,
- Perhaps the U.S.does have a dialectic…
- Start with Exceptional America by Abbott.
- Well then, go, research and design,
- And keep in mind:
- Electoral behavior is thoughtful like a child.
 SeeUniversity ofMichigan.
 Almost this entire poem is inspired from an analysis of: Bartels. 2008. The Study of Electoral Behavior.Princeton University (quote from page 22).
 Page and Jones (anemic Party ID), and, Markus and Converse (Party ID salient).