The National Science Foundation (NSF) supports political science research by faculty, as well as by undergraduate and graduate students. In fiscal year 2011, the Political Science Program spent about $1,000,000 on new research, training and workshop projects, and dissertation grants.Footnote ¹ In 2009, this important source of research funding was threatened when senator Tom Coburn (R-OK) filed an amendment to eliminate political science funding from the NSF budget. Coburn argued that political science “really is not science at all” and that “theories on political behavior are best left to CNN, pollsters, pundits, historians, candidates, political parties, and the voters, rather than being funded out of taxpayers' wallets.”Footnote ² The Coburn Amendment, as it came to be known, was defeated in the Senate with little debate. Studies analyzing that amendment found that while party was the largest determinant of senators' votes, the number of highly rated political science PhD programs in each state, the proportion of each state's population with an advanced degree, and whether the senator had majored in political science as an undergraduate significantly influenced the votes as well (Uscinski and Klofstad Reference Uscinski and Klofstad2010).

Following the defeat of the Coburn Amendment, it seemed political science NSF funding was safe. Thus, when representative Jeff Flake (R-AZ) sought to diminish NSF funding in May 2012, the American Political Science Association (APSA) was seemingly caught off-guard, and political scientists were only told of Flake's efforts as they unfolded. Unlike the Coburn Amendment, there was little build-up to Flake's proposals because they were not subject to a drawn-out committee process.

Representative Flake first introduced an amendment to reduce overall NSF funding by $1.25 billion, returning it to “pre-stimulus” 2008 funding levels. The vote took place May 9, 2012; the amendment was defeated 121-291 with 19 not voting. The 179 Democrats who voted opposed the amendment; the 233 Republicans had a near even split, 121-112. News of the amendment's defeat quickly spread through the blogosphere, eliciting a hesitant sigh of relief from political scientists. As Greg Koger of TheMonkeyCage.org commented, “the good news is that even a Tea-Party tinged House in the midst of a huge budget shortfall realizes that the NSF plays a vital role in our society. On the other hand, it is not clear what the margin would be for an amendment targeted at political science.”Footnote ³ Political scientists would soon find out.

In response to his initial defeat, Flake proposed a second amendment (hereafter referred to as the “Flake Amendment”) that would leave NSF funding intact, but bar the use of funds for political science research (this amendment borrowed language from the 2009 Coburn Amendment). When Flake began his remarks on the proposed amendment by mentioning debt and spending, he pointed out that the measure did not actually reduce spending, as his previous failed amendment would have. He therefore argued through most of his remaining remarks that political science research wasted “taxpayer dollars on a meritless program.”Footnote ⁴

Ironically, Flake, who holds a master's degree in political science from Brigham Young University, pointed out that political science research “has its benefits. The work of political scientists advances the knowledge and understanding of citizenship and government, politics, and this shouldn't be minimized.” But, he went on to minimize it anyway, claiming that many current NSF grants would not pass the “laugh-test,” and that while “these studies might satisfy the curiosities of a few academics,” he “seriously doubts society will benefit from them.”Footnote ⁵ The irony in Flake's reasoning was carried further a few days later in a post on his Facebook page:

In response to Flake's floor discussion, representative Chaka Fattah (D-PA) argued that although the NSF budget was substantial, ignorance is worse: “It is important that we understand the political dynamics, radicalization of populations around the world, how political parties operate in the former Soviet Union, all of the other issues that are being studied.” Fattah holds an undergraduate degree in political science, and like Flake, a master's degree as well. He noted it is both easy and politically expedient for Flake to call out certain NSF funded projects:

With discussion only from Flake and Fattah, the vote to eliminate political science funding took place just before midnight on May 9, 2012, and passed 218-208 with 5 not voting. Most of the support came from Republicans: 213 Republicans and 5 Democrats. The 181 noes came mostly from Democrats, but 27 Republicans voted no as well. With passage secured in the House, the spending bill with the Flake Amendment moved to the Senate. The Flake amendment did not become law. However, in March 2013 Senator Tom Coburn sponsored an amendment to eliminate political science funding; but a compromise was reached to keep political science funding but limit it to projects “promoting national security or the economic interests of the United States.”

Little media attention was given to Flake's efforts prior to reaching the House floor. But, after the amendment to eliminate political science funding was passed, the value of funding political science research elicited an uncharacteristic amount of discussion in the popular media. Charles Lane (Reference Lane2012) and Ezra Klein (Reference Klein2012) of the Washington Post both addressed the topic, the latter supporting political science funding, the former decrying it. The New York Times Sunday Review gave space for discussing political science funding as well (Stevens Reference Stevens2012). Much of this coverage appears to have been driven by the efforts of political scientists to justify political science research and funding in the blogosphere. In particular, TheMonkeyCage.org highlighted past NSF-funded research and extolled its value.

Even with the slim chance that the Flake Amendment will ever become law, its passage greatly concerned political scientists, and academics writ large (e.g., “A Different Agenda” 2012). This is the second time in three years NSF political science funding has been targeted by Congress. Although both Coburn and Flake justified their efforts due to concerns over government spending, neither of these amendments would have cut overall spending; they would only keep funding from going to political science researchers. And, while other social science disciplines continue to fly under the radar, political science is continually targeted.

The Flake Amendment did not address issues salient with the wider public or the media, nor did career or party fortunes appear to rest on the outcome. We have little reason to expect that either party would expend much effort in cajoling members to vote one way or the other (e.g., Bawn and Koger Reference Bawn and Koger2008). Given this, the vote on the Flake Amendment provides an ideal case to determine what factors, other than party, influence legislators to support or not support political science NSF funding. The analysis that follows demonstrates the impact that representatives' relevant individual, constituency, and institutional characteristics have on their vote.

RESULTS

We present the results of logistic regression analysis of votes on the Flake Amendment in table 1. Positive coefficients indicate an increased likelihood of voting in favor of the Flake Amendment (cutting political science funding). Given that logistic regression coefficients are not readily interpretable, for significant coefficients we present the substantive influence we estimate each variable to have had on the likelihood of voting for the Flake Amendment.Footnote ¹⁴ Each estimate was made while holding all other factors in the model at their means. Dichotomous variables are interpreted by comparing the likelihood of voting for the amendment between the two categories (e.g., Democrats versus Republicans), while nondichotomous variables are interpreted by comparing the minimum and maximum value of the variable. The model performs well; it correctly predicts 93.19% of the Representatives' votes, and provides a pseudo R ² of .67.

Table 1

Regression Analysis of “Yes” Vote to Eliminate NSF Political Science Funding

* p ≤ .10

** p ≤ .05

*** p ≤ .01 (robust standard errors in parentheses)

Model Type: Logistic Regression

Notes : Standard errors were clustered by state; robust standard errors in parentheses. The model correctly predicts the vote of 93% of Representatives.

Individual Characteristics

In section A of table 1, we see that representatives who majored in political science as undergraduates were more likely to support funding political science research by voting against the amendment. Substantively, representatives who majored in political science were 16 percentage points less likely to vote for the amendment. This suggests the affinity students have for the subjects they studied during their college years carries on later into life. For example, representative Robert Aderholt (AL-4), a Republican, majored in political science as an undergraduate and voted to keep political science funding. However, not all political science majors supported political science funding: representative John Barrow (GA-12), a University of Georgia political science undergraduate, was one of five Democrats who voted to end NSF funding for political science.

The model shows that holding a master's degree in political science or public administration was not a significant predictor of the vote. As previously mentioned, Flake holds a master's degree in political science. In addition, holding a law degree was not a significant predictor of the vote. However, those holding doctoral degrees in political science were significantly more likely to support political science funding. Members who hold a PhD in political science are 31 percentage points less likely to have voted to end political science NSF funding than those who do not. All representatives holding a political science PhD excepting Republican Tim Huelskamp (KA-1), voted to keep political science funding.

Constituency Characteristics

Moving to section B of table 1, we find that political science-specific district characteristics are highly related to representatives' voting behavior. The number of top-50 political science PhD programs in one's district increases the likelihood of voting against the Flake Amendment. Having the maximum number of top-50 political science PhD programs in one's district decreased the likelihood of a yes vote by 36 percentage points. These results suggest that representatives may be unwillingly to take potential funding from prominent political science departments within their districts. The number of political science NSF grants awarded in each district in 2011 also correlates with votes on the Flake Amendment. Representatives from districts with the highest number of grants received were 34 percentage points less likely to support the amendment compared to representatives from districts with no grants. The results also show that House members who represent districts with higher percentages of residents with advanced degrees and higher incomes were not more likely to vote for or against the Flake Amendment.

Institutional Characteristics

In this age of party polarization (Poole and Rosenthal Reference Poole and Rosenthal2007), one would expect party to be an overwhelming predictor of voting behavior in general. In contrast, however, given that political science funding is neither salient to the wider public nor tied to party reputation, one might expect party to play a lesser role in the Flake Amendment vote. Our results show that party is the most potent predictor of the vote. Democrats were 87 percentage points less likely to support cutting political science funding than Republicans. Section C of table 1 suggests that while party affiliation was an important determining factor, tenure was not.