Elections in Theory and Practice in the United States

Elections in Theory and Practice in the United States was a lecture given in PS 240 on September 15th, 2008.

Possible Exam Questions

• Explain the differences in voter decision making as outlined by the Michigan Model and Rational Choice Model. Include a discussion of potential problems within each model.

Lecture Material

Columbia School

Based on The People's Choice 1944 (rare), then in 1948, and the 1960 edition is what you can find on Amazon

This study focused on narrow groups, on the theory that sociological influences (group membership, family, friends, other societal backing) give a voter their political preferences. Group membership is the sole determiner of voting behavior.

A major flaw of the study is that it focused on too small of a sample; targeted towns, individual communities, counties, etc. rather than population centers, states, etc. where a sociological influence would not have as great of an influence due to the wide variety of opinions that exist.

Literary Digest

LD has been correct in their guess of who the next president would be through poles the magazine sent out from 1916-1932. In the 1936 which pitted Roosevelt against Landon they got ahead of themselves and their poll was the largest it had ever been conducted--10 million people(1/5 of registered voters). They were horribly off in their prediction that year, largely because of the people they sent cards to, the supporters of the candidate who would lose were more likely to return them. Literary Digest predicted the election to split 57/43 in Landon's favor when it actually went 38/62 to Roosevelt. The reason why the prediction was so far off was due to Literary Digest taking too large of a sample and, perhaps more importantly, those who had more free time were more likely to send back the postcards. Thus those with higher socioeconomic status were far more probable to respond to the survey and tended to support Landon.

The result of the 1936 Literary Digest inaccuracy was a dramatic shift in how voting polls are conducted. Sample size, wording, and questions started to get examined and analyzed in order to ensure a fair, accurate poll.

Michigan Model of Voting

Based on The American Voter (1960), which examined the 1948, 1952, and 1956 elections

This model assumes that partisan identification is entirely the result of the social influences of group membership and family influence. Partisan identification in turn determines policy attitudes of voters, their attitudes to group benefits, and their attitudes towards candidates. In short, socialization as a group member (i.e. gender, class, religion, age) combined with family influences determines partisan identification which determines an individual's vote. Their findings concluded that people cast their ballot primarily due to partisan identification. partisan identification-longterm stable psychological affiliation towards a party. one of the authors of the american voter did a study following its publishing where he followed certain people over multiple years and learned that while party support remained constant, policy support swung rapidly.

Critique of the model: It can be adjusted to work no matter the outcome.

Rational Choice Model

Based on The Changing American Voter (1976) and Anthony Downs' An Economic Theory of Democracy (1957)

The Changing American Voter argues that individuals make informed, consistent voting choices based on a cost/benefit analysis. Downs showed that purely rationally, there is no reason to vote, since an individual's chance of affecting an election outcome is near nil, so the cost of voting should far outweigh the benefit. In order to explain voting, Downs introduced the D-term, which stands for civic duty. The D-term is a tautological problem--it is so vague that it cannot be proven wrong. A different assessment of cost/benefit can be made to fit results.

Secret Ballot

The lecture also noted the necessity of making distinctions between U.S. elections before the 1930s when the secret ballot had fully spread in usage and the elections prior to then, which are not comparable to modern elections due to the absence of the secret ballot. The secret ballot had been introduced in America in 1896. Prior to this year, ballots would be picked up from party chairpersons; the ballots color-coordinated by party. This makes any sort of voting studies useless because job security and social standing (among other things) could rely on how one would vote. The term "split-ticket voting" came out of this period because in order to vote for persons of two separate parties, two different ballots would be needed to be picked up, torn in half, and turning in together. The first major voting studies came after the 1930s.

psephology-scientific study of voting.