Conference Envisions New Methods in Political Scholarship

By Phyllis Jeffrey - On May 17 and 18, 2016 the UC Davis Department of Political Science, in collaboration with its counterpart at UC Merced, hosted the ninth annual Visions in Methodology conference. The event brought together researchers from across the nation with the goal of supporting women who study political methodology.

On the first morning, Alison Craig, a PhD candidate in political science at the Ohio University, presented her paper “Crafting a Broad Appeal: Audiences and Collaborators in the U.S. House of Representatives.” In her talk, Craig (who previously spent a decade working on Capitol Hill), exploded common clichés of congressional polarization and ideological deadlock. She utilized Congressional Dear Colleague letters—sent by House representatives to those with whom they seek to co-sponsor legislation—in order to trace congressional collaboration.

The Dear Colleague letters proved a more rigorous and dependable test of true working relationships than mere bill co-sponsorship, which sometimes signifies last-minute strategic action. By contrast, when members sign a Dear Colleague letter together (which is then sent out to others), robust collaboration is indicated. Craig found collaboration, including across-the-aisle partnership, to be a frequent practice with value far beyond the pragmatic goal of passing a law. With Congress facing intense scrutiny, collaboration (especially cross-party) can aid individual representatives in gaining requisite approval from voters, parties, colleagues, interest groups, and the media—useful for their careers, even if no legislation is passed.

Bipartisan collaboration is particularly sought, Craig found, by members of the minority party. For all of this strategic maneuvering via electronic communication, friendships still matter. Her network analysis suggests that the “I had a drink with x yesterday evening, and now we are working on a bill on y” conversations she recalls from her time on the Hill reveal an important human side of congressional cooperation.

More discretion

Beyond passing legislation, lawmakers also strive to ensure that executive agencies adhere to drafters’ intent in the manner of policy implementation. Sharece Thrower, an assistant professor of political science at the University of Pittsburgh, drew attention to the under-specified phenomenon of congressionally-granted executive discretion in a talk based on her paper (with Alexander Bolton, postdoctoral associate at Duke University’s Social Science Research Institute), “The Constraining Power of the Purse: Executive Discretion and Legislative Appropriations.”

Thrower noted that the question of how much discretion Congress gives to executive agencies is a key element by which the legislative branch may, while delegating authority, still seek to rein in the executive and stem “bureaucratic drift.” As such, it is of crucial interest in studies of the balance of power. 

Thrower and her co-author focused on appropriations legislation over the past 40 years, measuring discretion as a function of the amount of budgetary authority (in dollars) given to an agency by a congressional committee. Testing the measure against the “ally principle” (which holds that legislators will be more apt to provide discretion to executives in the presence of ideological closeness), they found that more discretion is granted the executive when Congress is more closely ideologically aligned with the President. 

Selection matters

On the conference’s second day, Georgia Kernell, assistant professor of communication studies at UCLA, explored what determines congruency between parties and voters. In her talk, “Party Organizations and Party-Candidate Congruence,” Kernell focused on intraparty candidate selection, seeking to uncover whether candidates’ appeal among voters is affected by the method of selection. 

She found that in parties where candidate selection occurs through more democratic means (selection by members, as in the U.K.’s Liberal Democrat Party), candidates enjoy a higher degree of congruency with party members, yet are farther from the median voter. On the other hand, when leaders are selected by party leaders through a highly centralized process (as with U.K. Labour), they lose appeal among their own core constituents, yet gain greater traction with average (and, crucially, swing) voters.

Dependence and production

Tess Wise (PhD candidate in the Department of Government at Harvard University) explored the very issue of causality and how we measure it, through a talk based on her paper “The Limits of Causal Inference.” Wise explained that what we mean (or think we mean) when we say one event or condition “causes” another can shape what we look for (or fail to) when we measure causation. This can lead us to miss some causal relationships, or to posit causation where the “causal” element is really acting to prevent other processes.

To illuminate this dilemma, Wise, following Ned Hall, distinguished between two distinct meanings of causation: dependence (which refers to scenarios in which one event necessarily depends on another) and production (describes something that directly acts to generate an event).

Wise offered tangible examples, the first of which illustrated dependence without production. Suzy likes to throw rocks at windows; Billy usually stops her. When Billy, rushing to stop his mischievous friend, trips and falls on a banana peel, the window’s subsequent breaking is dependent on the banana peel—but it is not produced by it. The peel merely prevents the process (Billy’s intervention) that would have prevented the process that caused the break.

In a second example, Billy and Suzy both throw rocks at a bottle with an eye to breaking it. Although Suzy’s rock hits a second before Billy’s, the fact that Billy’s rock would have achieved the same result means that this is a case of production without dependence. Wise drove home the relevance of both predicaments through real-world causal dilemmas, such as parsing the causes of opposition to immigration. In order to extricate ourselves from this causal swamp, Wise ended by advocating greater use of “descriptive inference”—procedures that help us isolate processes through “snapshots” of action. Wise’s talk served as an important reminder that good methodology—and theory—stays close to the action of social processes.

This event was funded by the National Science Foundation and received co-sponsorship from the UC Davis Institute for Social Sciences, the UC Davis Department of Political Science, UC Merced School of Social Sciences, Humanities, and Arts, the UC Merced Political Science Unit, the UC Merced Tony Coehlo Endowed Chair of Public Policy, the UC Merced Graduate Division, and the UC Merced Academic Personnel Office.

Find more information about Visions in Methodology 2016, including a full list of events, speakers and attendees, at the conference website.