Political Parties in the APSR: 1936 to 2006.

  • This is an example of “Excellent Graduate Professor”–which was an assignment that I completed for Week 3.
  • Pick a Journal and read an article about “class topic” in every decade. Write a 3-5 page essay detailing the progression of science from the 1930s to the present; i.e. 1936, 1946, 1956, 1966, 1976, 1986, 1996, 2006.

Challenge: pick a journal and describe changes on one specific topic decade by decade. What’s political science?

I just analyzed eight articles on political parties from 1936-2006. All articles were exactly a decade apart and published in American Political Science Review. I found that the 1936 and 1946 articles resembled Geertz’s “thick description,” and were weak on theory, limited of mechanisms, and pre-regression on math. In the 1956 article, theory, math and mechanisms proliferated in the political scientist’s toolkit. As decades past, I observed that the articles increasingly utilized math and mechanism orientated research to describe the topic. Indeed, by the 1970s, the descriptive part was the “findings and implications” section at the end of the essay—after the math section(s). Finally, the 1996 and 2006 articles continued the math analysis, but also integrated more analytic narrative. In the remainder of this BLOG entry, I will explicate the telling points of the articles under review in a systematic manner in order to demonstrate the trends I have formerly expressed.

Before 1950, the articles were generally descriptive. The 1936 article, Propaganda Activities of British Political Parties (Stoke), examined “methods” by which British political parties used propaganda in their campaigns, and contrasted those campaigns to the Americans. However, the “methods” amounted to a typology of publications and an explanation of who received and used the material. For example, an analysis of the campaign material found that it “Appeals to tradition, sentiment, or prejudice” (121), but there is no mathematical evidence—no regression analysis. Moreover, the typologies were descriptive too, whereas the publications were “Publications for Party Workers”, “Periodical Publications”, “Reference or Research Publications”, or “Propaganda and Campaign Publications” (122-124). Overall, Stoke found that the party in power benefits from campaign propaganda, because the in-party is seen as legitimate and can delegitimize opponents. Nice theory, low to no evidence.

The 1946 article, The New Political Parties of Germany (Neumann), most brilliantly described the details of the situation of German political parties in the immediate aftermath of WWII. It was a worthy of The New Yorker. It was brilliant journalism, yet weak on data with limited evidence. For example, after the unconditional surrender, the Germans had no access to publish in newspapers, nor access to announce over the radio (750). In politics, “German officials were in the unbelievable position of being ostensibly responsible for measures in which they had no part…gathering public criticism… being unable to explain their predicament” (750). This article revealed that America was the only “region” to continue to limit free expression, while the Russians were quick to help the Germans form political parties—a “shrewd move” (752). Indeed, a majority of this article was a description of Russian political movements—and the lack of American movements. Overall, the article described the political parties that the Russians created, how they were developing, and it implicitly called for Americans to move forward. There were no statistics, mechanisms, or serious theory.

The 1950s—enter regression, formal theory, and mechanical and behavioral mechanisms. The 1956 article, One-Party Politics and the Voter (Miller), first analyzed aggregate data and individual survey data, and then applied the data to single-party dominance effects upon relationships between three motivational variables [Party Identification, Issue Partisanship, Candidate Partisanship] and voting behavior [time of arrival at final vote decision, consideration given to voting for the other candidate, extent of ticket splitting]. The analysis found that “there is a strong suggestion that minority party campaigning is…a discouraging, uphill battle…Second…relates to the persistence of one-party politics” (715). Due to the mathematical analysis, Miller determined that minority voters [not voting for dominant party] demonstrate less unity regarding issue partisanship or candidate partisanship than their majority counterparts, and that these findings cannot be applied to party identification (717). I would think that this journal article caused a lot of researchers to test Miller’s theory and data.

As the decades approach the present, political party theory, mechanisms and math developed and occupied the focus of the entire article. The 1966 article, The Theory of Party Equilibrium (Garvey), challenged the idea that nonvoting was a result of apathy or passive consent (29). The theory began with assumptions [e.g., democracies are interested in an informed citizen, the democratic system is putatively synchronized with the interests of the parties] (30). Variables were reduced to letters and integrated into a complex mathematical formula. The theory called for testing and Garvey determined that his theory did reaffirm the theory of passive consent (38).

The 1976 article[1] mathematically determined how many voters supported democratic candidates by issue (e.g., Vietnam, amnesty, marijuana, campus unrest, busing, ideology). Further analysis enabled the authors to graph observed and expected party polarization in the 1972 election. The evidence demonstrated that “the outcome of the election was the result of ideological polarization within the democratic ranks that pitted the left-wing Democrats against those on the right” (778). The 1986 article[2] presented a party competition model that gauged political phenomena. Mechanisms to explain phenomena are abundant. This article assessed imperfect information, voter optima, equilibria for Party X and Y, cycles, voter decisions, effects of debates, and the lack of debates. You certainly need to understand statistics to be able to comprehend the formulas.

The 1996 article[3] used just as much math and mechanisms as before (if not more), but it also integrated a “big picture” narrative. For example, the article began with a detailed literature review of the topic—synthesizing 40 years of research into the first page. Then, the article carefully explained the data and methods used (data archives, and, percentage differences in 2×2 tables and chi-square) (825). The general findings were published in paragraph form and nine mathematical tables occupied much of the pages. The discussion and conclusion were minimal.

The 2006 article[4] essentially analyzed the topic of this brief essay: the study of political parties in journals over time. Reiter found that there were three major time periods: pre-behavioral period (beginning through 1949), the behavior era (1950-1965), and the “Recent Period,” which mostly engaged rational choice / historical institutionalism (1965-present) (616). There was a spike in the number of articles on political parties during the behavior revolution (1950-1965). Articles which examined political parties holistically declined over time, on the other hand, articles that studied mass and legislative behavior increased over time.

In conclusion, the early articles resembled excellent journalism and were weak on theory, mechanisms, and math. After 1950, complex theory, math and mechanisms entered the scientist’s toolkit. As decades past, I observed that the articles increasingly utilized math and mechanism orientated research to describe the topic (e.g., rational choice), and, that evidence from later periods commented upon earlier periods. Certainly, Miller’s (1956) math and analysis of One Party Politics shed light on Stoke’s 1936 description. Overall, a systematic review of eight articles from eight decades revealed a progression of scientific tools within political science.

            Moral of the story: take a statistics or formal modeling class for every political science course you complete. You’re going to need it.

[1] Miller, Miller, Raine, and Brown. 1976. A Majority Party in Disarray: Policy Polarization in the 1972 Election. American Political Science Review, Vol. 70, No. 3. pp. 753-778.

[2] Chappell Jr. and Keech. 1986. Policy Motivation and Party Differences in a Dynamic Spatial Model of Party Competition. American Political Science Review, Vol. 80, No. 3, pp. 881-899.

[3] Webber and Domhoff. 1996. Myth and Reality in Business Support for Democrats and Republicans in the 1936 Presidential Election. American Political Science Review, Vol. 90, No. 4, pp. 824-833.

[4] Reiter. 2006. The Study of Political Parties, 1906-2005: The View From the Journals. American Political Science Review, Vol. 100, No. 4, pp. 613-618.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s