A qualitative narrative policy framework? Examining the policy narratives of US campaign finance regulatory reform

Abstract

In 2010, the narrative policy framework was introduced as a positivist, quantitative, and structuralist approach to the study of policy narratives. Deviating from this central tenet of the narrative policy framework, in this article we show that the framework is quite compatible with qualitative methods—and the various epistemologies associated with them. To demonstrate compatibility between qualitative methods and the Narrative Policy Framework, we apply classic qualitative criteria to an illustrative case examining policy narratives in US campaign finance reform. Drawing on elite interviews, we illuminate competing policy narratives rooted in distinct democratic values that exhibit variation in how victims and harm are defined, how blame is attributed to villains, what policy solutions are put forth, and policy narrative communication strategies. Our incorporation of qualitative methods within the narrative policy framework is critical for the framework's overall development as it provides opportunities for more detailed description, inductive forms of inquiry, and grounded theory development in policy areas where sample sizes, access, and salience may limit quantitative approaches.

Keywords

Campaign finance narrative policy framework qualitative NPF regulation qualitative methods

Introduction

In complex policy debates, narratives are strategically deployed by competing stakeholders to influence policy. In this article, we apply the narrative policy framework (NPF) to qualitatively study these policy narratives. The NPF provides theoretical means to disaggregate the component parts of competing policy narratives, examine how they vary, and identify patterns. However, to date, the NPF has primarily drawn upon quantitative methods such as surveys, statistical content analysis, and experiments to study policy narratives, with only a small number of qualitative studies (Pierce et al., 2014). In this article, we propose to build upon this small number of qualitative studies by providing a guide on how to conduct Qualitative NPF research. However, to formally situate qualitative methods within the NPF requires a conceptual re-calibration of the framework as qualitative research often has different intended ends, epistemological assumptions, and methodological criteria than quantitative research.

Given the NPF's postpositive origins (Jones and McBeth, 2010), and the rich history therein of qualitative narrative analyses (cf. Riessman, 1993), we posit that the framework's theoretical scaffolding is well-suited to incorporate qualitative methodologies—including their potentially divergent epistemologies and goals—without compromising the original aspirations of the NPF. In our view, the solution is quite simple. First, we set aside debates about quantitative versus qualitative and propose that the NPF's central assumptions, theory, and structure remain intact. Next, we argue that different qualitative standards for evaluating research can be inserted in lieu of quantitative standards when conducting Qualitative NPF research. Using this “plug-and-play” approach, we argue that the framework could be improved through more frequent use of induction, grounded theory, and thick description. Such research benefits the NPF by allowing for the examination of policy phenomena that is less amenable to quantitative approaches, while also potentially expanding its ability to speak to normative policy concerns. Additionally, showing that the NPF is consistent with qualitative research embracing alternate standards and epistemologies exhibits a theoretical platform that is not just multi-method, but also epistemologically adaptive. If readers accept our contention and subsequent analyses, then the NPF has the potential to become a genuine bridge between rival methodological approaches (i.e., positive and post-positive) within the discipline of public policy that have heretofore billed themselves as mutually exclusive (Jones and McBeth, 2010).

To illustrate Qualitative NPF, we examine elite policy narratives in contentious debates about campaign finance reform taking place in the United States. Drawing on elite interviews, we show how campaign finance regulation stakeholders engage in a narrative battle rooted in two competing democratic values: equality and individual expression. While some believe that money amplifies democracy by representing political interest and commitment, others view money in politics as a threat to the very fabric of democracy, arguing that it corrodes political equality by giving a select few a louder voice in the political process. We apply the NPF's traditional deductive policy narrative elements to illustrate that these competing narratives substantively vary in how they define victims and harms, attribute blame for the harms to villains, and promote different champions and policy solutions. However, in using inductive methods common to qualitative approaches we also show how the strategies used to disseminate narratives to the general public are quite different between the competing groups.

The article is organized in the following manner: first, we discuss the role of narratives in regulatory reform, and then introduce the NPF as a means to simplify and examine regulatory policy debates. In doing so we posit where and how qualitative methods might be best situated within the framework. We then provide a guide on how to conduct Qualitative NPF research through our analysis of elite stakeholder interviews on campaign finance reform, providing detailed information on how to implement NPF in an interview setting, as well as how to analyze qualitative data using the NPF. Next, we assess our qualitative application of the NPF using Lincoln and Guba's (1985) classic trustworthiness criteria—a criteria quite different from the science-based approaches typically found in NPF research. We conclude by evaluating our Qualitative NPF application within the context of the existing understanding of the framework, contending that there is significant advantage to be had for policy scholarship by having an approach to narrative broad enough to embrace diverging epistemologies and methods.

Narratives and regulatory reform

The regulation literature has a longstanding tradition of using narrative methodologies. For example, narratives have been examined in terms of their constitutive authority to produce social meaning and allocate power. Ewick and Silbey (1995) explain: “as socially organized phenomena, narratives are implicated in both the production of social meanings and the power relations expressed by sustaining those meanings…” (Ewick and Silbey, 1995: 200; see also Ewick and Silbey, 2003; Riessman, 1993). Narratives have also been used to describe regulatory environments. In the area of health and safety regulation, Almond (2009: 366–368) finds that media-reported stories in the United Kingdom increasingly focus on tales of regulatory unreasonableness that “transcend factual validity,” are “defiantly anti-regulation” and offer support for the hegemonic narrative of “free-market individualism”.

Haltom and McCann (2004) identified a similar trend in the United States. Examining media narratives stemming from business lobbyists, the authors observed that the purpose of the narratives was to protect hegemonic corporate interests from civil liability by framing certain tort litigation as frivolous. In these examples, narratives were used as a means to critique and explain hegemonic social narratives of individual responsibility and market-based approaches to regulation (Gray, 2006, 2009). The NPF, which we now turn, systematizes the examination of narrative and offers an opportunity to disaggregate narrative sub-components to better understand their role in shaping regulation.

The narrative policy framework

Explicitly building upon primarily interpretive, qualitative, and inductive forms of narrative analysis in the study of public policy (e.g., Fischer and Forrester, 1993; Hajer, 1993; Roe, 1994; Stone, 2012) the NPF was introduced in 2010. Billed as an alternative to prior approaches to narrative in public policy, the NPF was described as “a quantitative, structuralist, and positivist approach to the study of policy narratives” (Jones and McBeth, 2010: 330). While both approaches to narrative in public policy focus attention on how political actors construct meaning through stories they tell, the means by which they approach narrative have been viewed as distinct, perhaps even contradictory (Smith and Larrimer, 2013: 233–234). Recent scholarship by Jones and Radaelli (2015), however, argues that the NPF is quite compatible with its interpretive, inductive and qualitative roots. This article aims to demonstrate said compatibility.¹

As a theoretically deductive approach to the study of narrative in public policy, the NPF begins by providing a generalizable definition of policy narratives (Table 1), defining policy narratives as consisting of a setting, characters (hero, villain, and victim), a plot, and a moral of the story (i.e., policy solution) (Jones et al., 2014; McBeth et al., 2014). The setting consists of policy-consequential facts (e.g., legally binding codes). Characters are the relevant actors in a policy narrative, consisting of those that are harmed (victims), those that perpetuate the harm (villains), and those that will correct the situation (heroes). Connecting the setting to the characters and the characters to one another is a plot that spells out the causal arrangements of the policy problem. For example, an NPF researcher might invoke the classic work of Stone (2012) to model plot by describing a villain's actions within a policy narrative as intentionally or inadvertently harming the victim. Plots affect how blame is assigned to the villain, what actions are needed from the hero and what moral is to be gleaned from the story. The moral of a policy narrative is typically a policy solution.

Table 1.

Narrative policy framework concepts

Concept	Definition
Policy narratives	Consist of four core elements—settings, characters, plots, and morals of the story (policy solutions).
Policy narrative element: The Setting	Consists of “legal and constitutional parameters, geography, economic conditions, and other factors regularly deemed relevant by policy actors involved or associated with a public policy” (McBeth et al., 2014: 228).
Policy narrative element: Characters	Three categories of characters: “victims that are harmed by the problem, villains that intentionally or unintentionally cause the harm and heroes that provide or promise relief from the harm” (McBeth et al., 2014: 228).
Policy narrative element: Moral of the Story	The policy solution promoted by a policy narrative.
Policy narrative element: The Plot	Plots link policy narrative elements by establishing relationships between characters, their policy settings, and the moral of the story (McBeth et al., 2014).

NPF research is divided into three levels of analysis (McBeth et al., 2014: 230–246). The macro-level focuses on institutional and cultural policy narratives; the meso-level on group and coalitional policy narratives; and, the micro-level is concerned with the influence of policy narratives on individuals. In practice, NPF level demarcations have led to a micro-level focus on public opinion relying heavily on quantitative surveys and experimentation (e.g., Jones, 2013; Jones and Song, 2014), a meso-level focus relying on sizeable enough populations of documents to perform quantitative content analyses to describe subsystem policy narratives (e.g., McBeth et al., 2005; Shanahan et al., 2013), and a small analytical and qualitative literature at the macro-level (e.g., Ney, 2014). The de facto foci and methods produced by the NPF's levels of analyses illuminates a space for—perhaps even a need for—qualitative research within the framework.

Recent assessments of the role of public opinion in public policy indicate that public opinion is probably most influential in high salience policy areas (e.g., Burstein, 2014). Given that most of what we might call public policy is low salience, NPF theorists have concluded that at least at the micro-level where the focus has been on public opinion “the NPF may be expending considerable energy on explaining a small slice of variation…” in public policy (McBeth et al., 2014: 247). A different kind of problem has emerged at the meso-level where quantitative content analysis is the dominant method used to examine policy narratives. Theses analyses, by definition, have been performed where data are sufficient to gather enough sample to perform statistical analyses. Taken in tandem, the aforementioned problems at the micro and meso-levels draw forth interesting questions. What if a policy issue is of low salience and data to content analyze insufficient for statistical analyses (e.g., an obscure rider buried in a low-visibility congressional committee)? Or, what about those situations where there are only a few stakeholders whose opinions matter (e.g., a judicial ruling)?

At present, and due to its quantitative orientation, the NPF is methodologically ill-equipped to address the small “n,” low salience, but potentially important policy areas. However, given recent qualitative NPF scholarship (e.g., Crow and Berggren, 2014; O'Bryan et al., 2014) and arguments made in favor of integrating NPF with its inductive, interpretive and qualitative roots (Jones and Radaelli, 2015), we do not believe that the NPF is theoretically ill-equipped to do so. On the contrary, we believe that formally integrating qualitative methods into the NPF's existing theoretical apparatus will go far in providing the NPF means to address the preceding questions.

Qualitative NPF

Traditionally, specifications of the NPF have regularly expressed need for studies of policy narratives to exhibit scientific (i.e., quantitative) characteristics (Jones and McBeth, 2010). Later specifications have followed suit, expanding the list of terms to include familiar scientific concepts such as transparency, replication, and hypotheses testing (McBeth et al., 2014; Shanahan et al., 2013). Rather than situate ourselves on either side of an unfruitful divide (quantitative vs. qualitative methods), we submit that there is value and a need for both quantitative applications of the NPF and Qualitative NPF.

To integrate qualitative methods with the NPF, we propose a simple two-step operation. First, leave the traditional NPF framework intact—this includes assumptions, the model of the individual, policy narrative elements, and levels of analyses. Second, remove the quantitative standards for evaluating research and insert commonly recognized qualitative standards (e.g., Creswell and Miller, 2000). In other words, one may extract the quantitative criteria and insert one of several qualitative criteria (discussed below) when drawing on the NPF. While the Qualitative NPF inclusion of different epistemologies may leave the NPF with the sticky problem of reconciling findings put forth by different epistemologies, this is not a problem unfamiliar to social science. Indeed, many have heaped praise on this mixed-method “problem,” arguing that it serves the social sciences quite well (Creswell, 2014; Creswell et al., 2003; Johnson and Onwuegbuzie, 2004).

Qualitative NPF is inclusive of epistemological positions ranging from science-based (objective epistemology) to constructivist (interpretive epistemology) to more critical perspectives. Epistemology is therefore not a barrier to conducting Qualitative NPF research. However, all Qualitative NPF researchers, regardless of epistemological position, should aim for transparency in their research design and attempt to meet the standards of methodological rigor that are associated with different epistemological positions.

Those conducting Qualitative NPF research who adopt an objective epistemology will find the transition from traditional quantitative NPF research to Qualitative NPF a rather smooth one given that all that is needed is a willingness to change research methods, from quantitative to qualitative. Qualitative NPF conducted from an objective epistemology assumes that there is a close relationship between quantitative and qualitative analysis. For those researchers adopting this epistemological position we recommend consulting Brady and Collier (2004) who argue that quantitative and qualitative methods share many of the same standards around methodological rigour despite the diverse set of tools between the methodological approaches (but also see King et al., 1994).

In contrast to Qualitative NPF research rooted in objective epistemology, some qualitative researchers may instead wish to adopt a more constructivist or interpretive epistemological approach, as we do here. According to Creswell and Miller (2000),

constructivists believe in pluralistic, interpretive, open-ended, and contextualized (e.g., sensitive to place and situation) perspectives toward reality. The validity procedures reflected in this thinking present criteria with labels distinct from quantitative approaches, such as trustworthiness (i.e., credibility, transferability, dependability, and confirmability)… (126).

Lincoln and Guba's (1985) Naturalistic Inquiry represents a classic work that exemplifies an interpretive epistemological approach. However, the interpretive epistemological perspective has often been perceived as less welcomed in NPF research. This article seeks to change this perception. Thus, in our study of campaign finance regulatory reform we have opted to evaluate the robustness of our application of Qualitative NPF research by using the trustworthiness criteria proposed by Lincoln and Guba's (1985). Lincoln and Guba (1985) suggest that researchers outline the ‘trustworthiness’ of their qualitative research by examining the study's credibility, transferability, dependability, and confirmability. In Table 2, we detail these classic criteria. We return to the trustworthiness of our research in the concluding sections of this article.

Table 2.

Lincoln and Guba's Naturalistic Qualitative Standards (1985)

Credibility: The study should be credible to those from which the information was gathered and to critical readers.

Transferability: The transfer of findings from one context to another is a goal and should be assessed.

Dependability: This criterion refers to consistency in research processes.

Confirmability: Results should be supported by the members of the community they are derived from and other external sources whenever possible.

A guide on how to conduct Qualitative NPF research

In what follows, we offer a guide on how to conduct Qualitative NPF research. The first step in Qualitative NPF is to identify and describe the policy issue. In our case, the policy area is United States' campaign finance reform. Second, decisions must be made on research design and method. In our study, we conducted semi-structured interviews. Third, after collecting the data, decisions must be made on how to analyze the data. Here, we discuss both our codes and the coding process. And, finally, how one presents the policy narrative elements (settings, characters, plots, and morals of the story) is a matter of writing and presentation style. We now illustrate these four steps in greater detail.

The Qualitative NPF policy issue: US campaign finance reform

Data from the 2012 US election shows that less than 1% of the American electorate gives more than $200 in elections, and the majority of campaign funds come from these donors (OpenSecrets.Org, 2013). Politicians in their quest for reelection spend the majority of their time raising money, which means spending their time on the 1% of the electorate that donate funds. As one member of Congress observes:

Fundraising, it overwhelms everything. There are fundraising breakfasts, lunches and dinners in this town. It's just constant. I think it just permeates our constant behavior, always thinking about what is my fundraising goal for this month? Have I reached it? How is my opponent doing? Have I raised enough money? … In an average day in Congress, we finish our voting at some time in the afternoon and before people even take a breath they are out of the building over to the party committee offices and getting on the phones to do what's called ‘call time’. You get on the phone and you start calling people and asking them for money … (Confidential Interview with United States Member of Congress).

While this special relationship between politicians and the donating class of the American public is an empirical fact, elites often hold competing visions of what this relationship means (e.g., Black, 2000; Prosser, 2010). Moreover, while elite interpretations vary, the general public is relatively uninformed about campaign finance laws and regulations (Grant and Rudolph, 2004; Heerwig and Shaw, 2014) and—given the low salience of the issue—the public's role in campaign finance reform policy is also predictably limited (Burstein, 2014). Thus, to the extent campaign finance regulation does become part of the political discourse it would not likely be driven by citizen involvement. Rather, the inertia of the debate is more likely to be directed by a small group of elites who place priority on the issue.

Scholarship on the normative implications of campaign finance on democracy permeate the literature and the predominantly elite discourse, producing two distinct camps. Representative of the first camp, Lessig (2011) argues that the institution of Congress is corrupt because members are beholden to money, when their proper dependence should be constituents. Like others in this camp (e.g., Ackerman and Ayers, 2004), Lessig's solution is to, in one form or another, regulate the money. On the other side of the debate, authors such as Samples (2006) view money in politics as a creative countervailing force to incumbent advantages that establishes meaningful linkages between members of Congress and their constituents. Samples and others (e.g., Smith, 2003), see money as a force for democracy and “the regulator as an obstacle” (Gray and Silbey, 2014: 125) to good governance. This reading of the normative positions on campaign finance yields a discourse structure that is congruent with Grant and Rudolph (2004), who find that campaign finance is socially constructed in such a way that produces conflict between two values: expression (i.e., money as free speech and ought not to be regulated) and equality (i.e., money disrupts free participation and ought to be regulated).

In what follows we examine how elites interpret this issue, as it is their actions that likely drive policy. More specifically, we examine elite interpretations of campaign finance reform as manifest in the stories they tell.

Qualitative NPF data collection: Interviews

Our data consists of 29 interviews with key stakeholders in the United States campaign finance arena. The interviews were conducted between 2 January 2011 and 18 May 2011. To populate our list of stakeholders we first conducted an extensive review of the campaign finance literature identifying individuals and organizations that were repeatedly referenced. This process led to the identification of various elite actors, including political columnists, members of Congress, lobbyists, academics, and members of organizations involved in campaign finance debates. We examined websites of identified organizations to obtain relevant contacts and merged these lists with names identified in the literature. Our initial list consisted of 130 potential interviewees, but through a snowball sampling technique we were able to expand our list to 141.

Stakeholders were then divided into five categories predetermined by our initial review of the literature.² The first category, titled the System Change Group, consisted of 43 individuals concerned about the inequalities produced by the current campaign finance system. These individuals tended to publicly espouse positions regulating campaign spending and contributions and also tended to show support for publicly financing campaigns. The second category of individuals, the Status Quo Group, consisted of 16 stakeholders identified to either support the current system of campaign financing and/or support reforms that would remove contribution limits. In addition, two practically oriented categories were also included: 18 researchers from academic institutions (Researchers Group) and 54 individuals who had been or were currently involved in the enforcement and implementation of campaign finance policies and regulations (Practitioners Group). Finally, there were 9 individuals classified within the Unsure Group that did not clearly fit in any of the previous groups.

We first emailed the stakeholders with a request to conduct an interview. If the initial attempt at contact failed, a follow-up email was sent two weeks later. Stakeholders who were interviewed were asked to provide additional names of potential interviewees; new names were added to the list and the same protocol for establishing an interview was followed. In total, 141 stakeholders were contacted; 29 were interviewed (23 male; 6 female). We interviewed 8 stakeholders from the System Change Group (18.6% response rate), 6 stakeholders from the Status Quo Group (37.5% response rate), 4 Researchers (22.2% response rate), 7 practitioners (13% response rate), and 4 individuals from the Unsure Group (44.4% response rate). Our overall response rate was 20.6%.

All but one interview was conducted by phone. On average, interviews took approximately thirty-five minutes to complete. Of the 29 interviewed, seven were professors currently working at universities. Nine were affiliated with interest groups involved in the campaign finance debate. Seven interview subjects were affiliated with prominent conservative or progressive think tanks. Six were political operatives consisting of one member of Congress, one individual that ran for Congress but lost, a nationally recognized member of the ‘Tea Party' (an informal name for a group united by a shared suspicion of interventionist government), a nationally renowned columnist, a high ranking member of one of the two national political party committees, and a high ranking member of a state-level third-party organization.

Interviews were semi-structured with follow-up questions pursuing emergent themes unique to specific interviews. In Appendix A, we provide an example of the interview instrument. Given that our interview subjects consisted of elites, additional pre-interview time was required for some interview settings (e.g., Dexter, 2006: 31–72; Gray, 2013a). Our extensive review of the campaign finance literature allowed us to be familiar with each interviewee prior to the interview. However, additional time prior to each interview was still needed to re-familiarize ourselves with specific interviewees. This preparation contrasts with traditional interviewing practices where one learns about an individual's perspective and the social meanings they attach to a particular topic through the process of the interview itself (e.g., King and Horrocks, 2010; Silbey, 1994).

Given that many of our subjects had a public position on campaign finance we made sure that during the interview we tried to illicit and understand the particular policy narratives employed by each during the interview. In other words, even though we knew their position beforehand (i.e., especially those with clearly espoused public positions), we are confident that the interviews still provided accounts of the social meanings interviewees attached to the subject of campaign finance, as well as the narrative elements they used when giving their responses. The assurances of confidentiality and anonymity also encouraged interviewees to be sincere when answering questions.

Qualitative NPF data analysis

Interviews were recorded, transcribed verbatim, and then both inductively and deductively analyzed by the research team with the assistance of the qualitative software program ATLAS.ti. The deductive aspect of our coding included the traditional NPF policy narrative codes: setting, plots, characters, and policy solutions (i.e., the moral of the story). Following the principles of grounded theory (Charmaz, 2006; Glaser and Strauss, 1967), we also inductively coded for emerging patterns, themes, and concepts. For instance, we observed variations in the accounts given about the perceived motivations of opponent's policy positions. We subsequently added additional codes that described different motivational accounts of villains in policy debates (such as simply being unaware, or being uncaring, and even evildoers). Grounded Theory induction also facilitated our addition of a policy narrative “Strategy” code that was used to identify the intentional use of narrative elements to persuade or manipulate.

The value of grounded theory in qualitative data analysis is that it allows codes to be developed as one reads and analyzes the interview transcript texts rather than limiting the process of coding to only pre-determined concepts and patterns.

Prior to coding Qualitative NPF researchers should be intimately familiar with coding protocols and, if not, they should first consult a qualitative text on the subject, such as Saldana's (2013) “The Coding Manual for Qualitative Researchers.”

Qualitative NPF findings: Elite campaign finance narratives

The setting: Conflicting democratic values

The policy narrative setting is the physical, ideational, and discursive space in which regulation takes place (see Levi-Faur, 2011; Gray and Silbey, 2014, 2011; Parker and Nielson, 2011 for corollaries in the regulation field). This space begins with key actor's understandings of the specific legalities governing the transactions of campaign finance. All interviewees demonstrated expert level understanding of the basic legal structure and regulations of campaign finance (donation limits, reporting requirements, funding procurement, etc.). However, what these regulations meant to the interviewees varied. Evidenced in our data, interviewees were categorized in one of two camps representing either the value of expression or the value of equality, creating divergent understandings of the same legal and empirical campaign finance setting. Proponents of expression held that regulation of campaign finance violates an individual's ability to express their beliefs through money. For instance, those who advocate individual expression are represented by the following quotes:

The dissemination of speech… requires money. Therefore, restrict the money, you restrict the speech.

Citizens should be able to contribute to any candidates or any organization, whether it's a political party or somebody like Pro-choice America, or The National Rifle Association, or some other entity, or the republican party, the democratic party, they should be able to contribute as much as they want to without any limits.

The most toxic reforms to the system [are the] ones that limit the ability of citizens, candidates, PAC's, and parties to effectively communicate their message by imposing either limits or burdens that inhibit them… They're anti-free speech. They're anti-First Amendment…

In contrast, proponents of equality tended to argue that money in politics is corrosive to society.

The fundamental problem with the other direction people go, [i.e. wanting to] just get rid of all regulation of politics [and arguing that] this is going to inhibit speech… is that it exacerbates the speech problems we have currently in the system, which is, your speech is not measured by your creativity or the thoughtfulness of your ideas. It's measured by the size of your bank account… Well below one percent is providing almost all of the money funding campaigns, and that fraction is not at all representative… most of those giving money are… either employed by, or otherwise connected to organized lobby groups… are overwhelmingly male, overwhelmingly white, [and] upper income…. If you can vote once but you can give $10,000, you obviously have more influence than somebody who has voted once and can only give $10, and if you can vote once and give $10,000,000, especially if you can do it anonymously or secretly, you know, you may be able to turn the election. So, I'm not a big fan of letting rich people decide who writes their tax laws and regulates their pollution.

As shown in Table 3, our coding revealed that depending on the democratic value premise that one begins from (equality or expression), a general narrative plot derived from each value begins to develop with generally consistent characters (victims, villains, and heroes) that culminates in value specific policy solutions.

Table 3.

The regulatory narrative of equality versus expression

Democratic value	Victim	Villain and plot	Hero	Moral of the story (policy solution)
Expression (money as free speech and ought not to be regulated)	Individuals	Public funding and regulation	Advocates of the Status Quo	Status Quo
		Motivation of villain	Advocates of the Status Quo	Status Quo
		1. Unaware: inadvertent harm	Advocates of Increased Expression (i.e., More Money allowed in Campaign Finance)	No limits or increased limits on money in campaign finance
		2. Uncaring: (mechanical harm)		No limits or increased limits on money in campaign finance
Equality (money disrupts free participation and ought to be regulated)	Society	Money in Politics	Advocates of system changes that place limits on campaign finance	Public funding
		Motivation of villain		Constitutional amendment
		1. Unaware: (inadvertent harm)		Constitutional amendment
		2. Uncaring: (mechanical harm)		Lower limits
		3. Evil doers: (intentional harm)		Matching funds

The cast of characters: Victims, villains, and heroes

The battle between the democratic values of equality and expression resulted in distinctly different policy narrative character configurations. For instance, those who valued expression defined the victims of regulated campaign finance as individuals who are unable to donate at their preferred level, on the basis that their rights to free speech are undermined. In contrast, those who valued equality claimed that deregulation of campaign finance results in collective victimization, as only wealthy individuals are able to have their views expressed and adopted.

Competing narrative portrayals play an important role in shaping regulatory approaches and preferences among policymakers. For instance, if one believes that the victims of campaign finance regulation are individual members of society unable to fully express themselves through their use of money (i.e., the expression camp) then the “threat” (i.e., the villains) is actors advocating for restrictions on campaign finance. The identification of the “hero” in the expression camp becomes equated with those individuals or groups fighting to advance or maintain “the victim's” right to expression in a democratic society. Overall, the moral of the story for the expression camp is that the public is best served through the adoption of policy solutions that remove restrictions on the use of money in political campaigns.

In contrast, those who interpret the setting through the democratic value of equality insist that the victim of unlimited campaign finance is society as a whole. The villain is any individual or group that denounces regulation or, conversely, advocates for more unregulated money in the political system. The “heroes” in this case are those fighting to protect the equality within a democratic society. Policy solutions that increase equality for all members of society are advanced, which in practical terms means limiting campaign finance donations and/or public financing of campaigns. Thus, the moral of the story for the equality camp is that the public is best served through regulations that allow greater participation in the political system by all members of society regardless of financial capacity.

Interpreting the plot: The villain's motivation

Both the expression and equality camps characterized the other as a villain causing great harm, but there is variation in the way that they interpreted motive, with each camp providing different conceptions of harm with a corresponding plot and causal arrangement. To interpret these causal arrangements, we rely on the work of Stone (2012), who characterizes causal arrangements as intentional, mechanical, accidental, or inadvertent (but also see Crow and Berggren, 2014). Mechanical harm is designated by a system operating as designed, but causing harm in the process. Accidental harm is caused by random (surprising) acts. Inadvertent harm is the product of well-intentioned (or at least not malevolent) actions mistakably causing harm. And finally, intentional harm is the product of nefarious actors actively seeking to do damage (see Stone, 2012: 213–214).

Unaware villains (inadvertent harm)

Both sides in this debate generally believed that the other side is genuine, but nonetheless were villains because they inadvertently cause harm. For instance, the expression camp regarded the equality camp as sincere but naïve and lacking an adequate understanding of how politics works in everyday practice. The equality camp also believed in the genuine nature of many expressionists but regarded them as misguided and often ideologically driven (Table 4).

Table 4.

Attributing motives to the inadvertent harm caused by competing stakeholders

Expression camp on equality reformers	Equality camp on expression reformers
You know, most people are pretty upfront and straightforward and honest … But, they're wrong … they're not looking at the reality of how society works	I think there is a class of reformers or people on the other side that genuinely believe that their version will make the system better …
Well intentioned people who sort of have a particular world view that you know, sees a connection that I frankly don't see … their answer is, “Well, we'll just get private money out of the system and just have government pay for it.” I mean I think in a lot of cases it's sort of a naïve perspective on politics	The opponents of campaign finance reform … these are people who I respect, whose conclusions I don't share, but for at least those core advocates, it is not a matter of being a mouthpiece for special interests, but it is something they believe … I think there is a genuine commitment to this idea under the first amendment that no restrictions ought to be placed on spending of money
The genuinely sincere ones … these people have problems with is the fact that politics is a transactional business … you do something for me, I'll do something for you	As to the kind of good governance reformers, I think they're genuine but misguided
I know too many people who disagree with me vehemently and are well intentioned, good, smart people … [they] truly believe that if you took private money out of politics the results you would get out of your representative government would be more akin to what the people would want … a better representative government than we have right now. And I just think, I don't think that's true … I think they're wrong	The ideology of money is speech. In other words, people with money should not be restricted in any way from using that money to register their points of view within the political process … So it's certainly ideological people I think whose motives are quite pure

Uncaring villains (mechanical harm)

While both sides of the debate on campaign finance reform argued that that other side was genuine but wrong, a different picture emerged when the general political system became the focal point. Briefly converging, a meaningful minority from both the expression side and equality side suggested that campaign finance reform is hampered by the tendency of incumbents to support policy that maximizes their chances of reelection.

Reforms are all about protecting incumbents generally. And then if you can game the system, then specifically that … It's not that incumbents are bad people. It's just they're human … They want to make it easier to stay in office … And so they'll use the law to do it (Expression Stakeholder).

Most of it is gaming … In the long run I think both sides learn how to play the game by whatever rules are there … So any time the rules change it will benefit one side or the other in the short term. In the long term they all learn to adapt and govern (Equality Stakeholder).

I think the bad guy is basically a broken system…it is our general inertia for a lot of reasons to not want to change (Equality Stakeholder).

Among those who believed that reforms are hampered by self-interested incumbents, villains were not regarded as evil, but rather as normal people who were behaving according to the incentives dictated by the campaign finance system. By locating the cause of harm in the political system within which political actors are embedded, this interpretation fits with the current stream of research on institutional corruption that focuses on lobbying and Congress (Lessig, 2011; Gray, 2013b, 2015; Jorgenson et al., 2013).

Evil villains (intentional harm)

Interestingly, not a single expression advocate believed that equality advocates are evil. Those on the equality side of the fence, however, did occasionally regard the motivations of their opponents as intentionally creating harm for purely selfish reasons. One interviewee noted that,

The politicians who pass reform are relentlessly self-interested, and it is only very rare moments when they actually think about the public interest and regulating democracy (Equality Camp).

Reoccurring villains described in this way were “generalized” members of large financial institutions, who equality interviewees argued had a clear sense of economic self-interest and a disinterest in the wellbeing of society.

Do I think that the Chamber of Commerce is well-intentioned behind their efforts to repeal all caps [on campaign finance]? No, I do not think they are well-intentioned. I think they view it as a way to seek even greater advantage than they have now.

… Groups like the Chamber of Commerce [in a US state] have invested considerable resources in opposing [public funding]. I think in that case, there is a basic self-interest question less a matter of policy principle.

… [It is] the narrow self-interest concerns of prominent economic actors … under the current system, the winners are people with lots of money, and who now can use that money to advance their interest in the political process, whether that's getting a fat contract from a government official making sure no regulations are passed that inhibit a company's ability to do business … I think the money behind the muscle tends to be economic interests that benefit from a big money-driven political system.

In these examples resentment is targeted primarily toward the financial institution, where its members are depicted as aggregated, unitary constructs, rather than toward specific individuals.

Narrative strategies and the moral of the story

Our inductive coding process allowed us to identify policy narrative communication strategies used by both the expression and equality camps.

When people are doing this framing, there's often a little bit of dishonesty or wordplay, but generally, people are, I think being honest. They're just, again, framing the issue in a way, how you look at something, what frame you see it through determines what you see (Expression Advocate).

Both sides also showed awareness of the lack of understanding and concern about the campaign finance system among ordinary citizens, which seemingly influence their communication strategies. According to one expression advocate,

When you have complexity you make it very difficult for people who aren't terribly sophisticated politically in terms of understanding the nuances of the law and what is and is not allowed.

While both expression and equality advocates wish to increase the awareness of citizens, they do so with different policy goals in mind. For expression advocates this generally means deregulation of campaign finance coupled with, in some instances, disclosure policies (i.e., revealing where all monies come from) and an expectation of individual responsibility on the part of voters to decide for themselves if money is having a negative influence in politics.

It seems to me if you had no limits on the amounts that could be given, which is to say if candidates who are not rich were treated like those who are rich, [and candidates] have the ability to spend unlimited sums of their own money, and couple that with full disclosure, the public could make up its own mind about whether receiving large sums of money was unseemly, corrupting (Expression Advocate).

According to this expression advocate, disclosure plus individual responsibility is the most appropriate policy. This individual continues,

Let voters decide. Voters must be presumed competent to decide or we wouldn't have elections in the first place.

Later in the interview, the interviewee disclosed that he fully supported the deregulation of campaign finance. When asked about his earlier insistence on a disclosure policy he stated,

I'm actually not that fired up about full disclosure but it helps sell the program … People can decide … People can make up their minds.

Yet, a few minutes earlier in the interview, the interviewee stated,

The public doesn't think about it [politics]… most Americans are not that interested in politics.

Disclosure, then, for some appears to be used as a strategic tool coupled with deregulation or the status quo to help sell (de)regulatory programs, as there is at least some understanding that exposing where the money is coming from will fall upon an ambivalent public.

The expressionists take note of public ambivalence about campaign finance and make efforts to incorporate it into their narrative strategy. When one expression advocate was asked what narratives worked well for his side when he attempted to convince the public of their position he acknowledged the importance of knowing the audience:

The public isn't terribly interested. And yet, it's also something that people have an opinion on in the sense that they feel they can understand the issue … They're not terribly interested, but that means what they learn about it, what they think about it, comes from the headline, you know, the brief comment they catch on the radio, or like say just glancing at the headline page of the newspaper, a joke on Leno, something like that …That makes framing particularly important. Framing's always important in politics, but this is an issue where I think it might be even more important than in many, because people will have opinions, and yet not be interested in delving deep into the matter. I have actually been very conscious of that…

For expression advocates the catchy headline and the provocative phrase are key starting points (see Table 5 for examples) that help citizens adopt a narrative that fits with the democratic value of individual expression, and, in turn, their preferred policy solution.

Table 5.

Narrative strategies used by expression advocates

Welfare for politicians	“Food stamps for politicians. That is one of the arguments or phrases that resonate well with the public.’; ‘Sometimes it's not so much the language, but just demonizing things… we'll call it, you know, ‘bailouts for politicians”
Public financing	‘‘On public financing, we're usually on the offense. Public financing has always been a loser for the other side. Why? Because voters don't pay a lot of attention to this, and we're thinking sort of middle class whites as much as anything, not the upper middle class … the Sarah Palin, blue collar Republican … they're deeply suspicious … I want them to think, when they hear, ‘Campaign finance reform, you know, I think somebody's getting screwed by this, and I think it's probably not the guy who's trying to convince me that I should support it. I think it's probably me’ … So definitely there's a sort of a battle of rhetoric and an effort to you know, to kind of frame the issue and gain the offensive.”
Clean elections are dangerous	“I mean, I don't think we're going to become, if we adopt more of these clean election systems or something, we're not going to become a totalitarian government, certainly not overnight or anything, but… I do think it's dangerous. I mean, look at little ways already in which the government tries to control.”
The poor need the rich	“Where do political parties, for example, that have historically represented the poor, and you know, the underserved and so on, where do they get their money from? They get it from large contributions … the way in which outsiders, the dispossessed, and so on traditionally are able to gain political power, is through the efforts of individuals who will spend large sums of money.”

In contrast to expression advocates, those promoting the democratic value of equality tended to focus on developing narratives that educate the public.

You can take any issue you please, I think, and make a compelling argument … in the public interest as opposed to a narrower set of special interests… [However] those solutions are repeatedly either off the table, or watered down, compromised … Because the money that these narrow interests wield … ultimately it leads into a kind of corruption, conflict of interest argument, which we find in repeated polling, the public really accepts. It's a complicated argument, and trying to be intellectually honest is important.

Additionally, while supporters of equality contend that educating the public is necessary, our data also showed a desire by these stakeholders to attach campaign finance to a policy surrogate (Table 6). McBeth et al. (2007: 91) describe the policy surrogate as the attempt to “strategically entangle policy issues in larger, emotionally charged debates … to gain competitive advantage.” While there has been a desire on the part of the equality camp to hitch campaign finance to more salient policy surrogates, a general theme in the interviews was that this had not been done effectively. Table 6 lists coded surrogate strategies.

Table 6.

Policy surrogate narrative strategies used by equality advocates

The environment	“The place where we begin is not with campaign finance reform, but with the issues that people care about … the issues that we meet every day … So for environmentalists, why is it that we have been unable to put in place any kind of policy to deal with climate change in this country, to change our energy policy?”
Money and power	“Campaign funds in and of itself doesn't sound inherently interesting, but when you dig a little bit deeper and you talk to people … they get the connection between power and money; between the pursuit of money in order to obtain or maintain power. And when you phrase it in those terms it actually becomes quite an interesting and compelling and relatively sexy issue …. When you connect it to policy, people begin to really think more deeply about it in a much less superficial way when you connect, you know, huge oil and gas interests to contributions to parties to, and the failure of Congress to even address issues like global warming and other things, just as like one very small example.”
Appeal to trust in history	“The laws in 24 states were passed almost 100 years ago because of the Carnegies of the world, the Rockefellers, the Teapot Dome Scandals, and in Montana in particular the copper barons who ran the state. And that was a political environment that had reached a level of corruption that the people acted, and they did so in a, for a particular reason. For us to eliminate that completely [via Citizens United] says that, “Okay, we're going to trust that the political environment is different now than it was then.”

Elite campaign finance policy narratives: Discussion and summary of findings

Our Qualitative NPF analysis dissected elite policy narratives related to campaign finance reform. We examined variation in narrative elements including characters (heroes, victims, and villains), plot, and strategy for two different campaign finance camps, one rooted in expression and the other in equality. Our analysis revealed that expression and equality stories about campaign finance are quite different.

In the expression camp, the threat was perceived to come from well-meaning but naïve pro-regulation groups that want to impose restrictions on how campaigns are financed. For the expressionist, such regulations inadvertently undermine democracy because they limit an individual's ability to express their desire to influence politics, making every potential donor a victim. Moreover, campaign contributions are not only a reflection of desire, but also the intensity of that desire that can be gauged by the amount of money donated. For this group the preferred policy solutions involve tolerating the status quo, or increasing or doing away with contribution limits. Strategy wise, the expressionists focused on creating catchy sound bites that would resonate with disengaged citizens. The heroes of their story advocated this position.

The equality camp used narratives to portray campaign finance quite differently from the expression camp. The equality camp attempted to create policy narratives that educate citizens on the consequences of money in politics. They perceived the victim as society and the threat coming from the destruction of political equality brought about by the undue influence of money in politics. The villain in their story tended to be portrayed in one of two ways. First and like the expressionist camp, the villain was conveyed as unaware and simply a product of an incentivized environment that naturally generates self-interested behavior. The second portrayal saw the villain as malevolent, intentionally doing harm to the public to serve their self-interest. In this case, the villain was usually more abstract, either a corporation or a business representing organization such as the Chamber of Commerce. Policy solutions by this camp tended to favor anything that would restrict contributions, and frequently advocated for publicly funding campaigns. There was also a genuine want in this camp to find one or more policy surrogates that would make the issue more salient.

Despite the different stories, at least one point of convergence was identified. Both groups seemed willing to agree that whatever system was in place or likely to be put in place, it would favor incumbents. Thus, both groups thought that any reform in any direction would have to overcome the opportunistic nature of members of Congress and their likely support for only those solutions that favor their reelection. Should a compromise position ever emerge between the expressionists and equality camps, it is likely to involve capitalizing on both groups cynical interpretations of incumbent politicians.

Evaluating the robustness of our Qualitative NPF research

To assess the robustness of our Qualitative NPF research we rely on the four trustworthy criteria of credibility, transferability, dependability, and confirmability established by Lincoln and Guba's Naturalistic Inquiry (1985) as exogenous standards for assessing qualitative research (Table 2). In operationalizing these four criterions we lean on the oft-cited work of Shenton (2004) which identifies specific practices observed to help studies meet these criteria.

Our transparency in detailing our research design, procedures for data collection, and analyses help our study meet the criterion of dependability. While our results are likely not replicable in a scientific sense, they are repeatable (e.g., one could use the same instrument and the same data collection procedures on a similar population), which satisfies the dependability criteria (Shenton, 2004: 71). According the Shenton (2004: 68), such transparency also makes it more likely that subsequent studies can engage ours and potentially transfer our findings to different contexts, allowing comparisons with similar NPF studies in different policy areas. Beyond comparison, however, our findings are also transferable as they could be tapped to develop testable hypotheses for the purposes of determining their generalizability. For example, future NPF researchers could explore specific hypotheses testing the extent to which strategies of manipulation or persuasion permeate other high or low salience policy areas. Transparency also lends to the credibility of our findings. Our detailed description of data and method illuminate that we are carefully accounting for the specific context of campaign finance in the United States by selecting an appropriate interview pool that hones in on vested organizations and key stakeholders whose opinions on campaign finance are likely to be the opinions that matter. Or, stated a little differently, we believe we have the right set of interviews to come to terms with elite positions about campaign finance in a very broad—albeit tentative—sense, and that being transparent about the population that let us generate our conclusions will lead the reader to a similar conclusion. In Lincoln and Guba's trustworthiness criteria, this is the definition of credibility. Finally, transparency makes it easier for the reader to determine if the findings are “the result of the experiences and ideas of the informants” (Shenton, 2004: 72), thus satisfying the criterion of confirmability. We believe our findings are the result of informants ideas and experience due to the open ended semi-structured nature of the interview and while we did use deductive coding schemes (which would arguably force interviewees into preconceived and less “free” categories) we also employed inductive coding free of such a priori constructs. In any case, it is our transparency that makes it possible for the reader to make that determination for themselves. Beyond transparency, however, we note additional features of our research that lend to its trustworthiness.

Prior examination of the literature and a detailed understanding of our interviewees also contributed to our studies trustworthiness. Such understanding lends itself to the credibility of the study, whereby one might assess how closely our findings align with the academically accepted ‘reality’. Aside from providing what we believe to be a full (but not exhaustive) depiction of the campaign finance setting, our study also demonstrates the criterion of credibility by validating extant findings related to campaign finance, particularly as our findings about elite policy narratives corroborate values of expression and equality found in the general public (e.g., Grant and Rudolph, 2004). Also lending to the credibility of our study is the use of an established theoretical framework in the NPF.

The NPF has a history of producing valid and reliable findings dating back a decade or more (see McBeth et al., 2014) and is acknowledged by the field of public policy as a legitimate framework for empirically studying policy narratives (e.g., Cairney & Heikkila, 2014; Weible, 2014). In this small, simple, and yet meaningful way, the NPF helps lend to the credibility of our qualitative study by simply being the NPF in so much that it allows us to draw on a previously legitimized framework, as Shenton (2004: 73) suggests researchers do. Beyond credibility, however, the NPF also adds something to our study that helps in its transference from one context to the next.

Regarding the trustworthiness criteria of transferability, our concern is with at least the potential for our study's transference to other contexts, as it allows for a reciprocal conversation between qualitative and quantitative researchers within the larger NPF. Without this conversation findings within each methodological camp move along separate tracks, building knowledge independently. The NPF facilitates this conversation by providing a common language—a theoretical bridge—for qualitative and quantitative studies, allowing theoretical objects (e.g., characters, plots) to be compared, evaluated, and critiqued via different methodologies. In other words, findings from qualitative studies such as ours can morph from being transferable, to testable hypotheses that inform, challenge, or modify the theories invoked by quantitative scholars to build their NPF models.

Moving forward with Qualitative NPF

The NPF's defining purpose is to empirically understand the role of policy narratives in shaping public policy. To date that purpose has been met by applying primarily science-based quantitative methodologies in high salience policy areas. These types of studies—primarily content analyses, surveys, and experiments—always require sufficient data (“N”) for statistical analyses. In both cases sample size, at least in part, inadvertently guide the NPF's policy focus. But what if a policy area lacks both of those attributes? What if the NPF needed to study the policy narratives of a select few stakeholders due to their importance as agenda setters or decision-makers? Because of the scientific orientation of the framework, we believe it is currently methodologically ill-equipped to address these questions. However, we have also argued that the NPF is theoretically robust and with a slight recalibration the framework can be modified to incorporate qualitative methods embracing alternate epistemologies to address policy areas where limitations on access and available samples restrict the framework's applicability.

The study presented here to illustrate a Qualitative NPF relied upon 29 elite semi-structured interviews. We employed the standard NPF policy narrative theoretical constructs, but through the use of grounded theory guided induction, we were also able add to those constructs, providing a more detailed explanation of the campaign finance regulatory policy narrative arena. Importantly, such exposition illustrates several facets of expression and equality policy narratives that if left to the devices of “normal” science-based NPF would have been missed: notably, elite strategies related to communicating with the general public and a more nuanced conception of character motivations within the competing policy narratives. Thus, if we extrapolate from our findings, we suspect that by incorporating qualitative methods with alternate epistemologies into the NPF that it will facilitate theory development via the additional lenses provided by induction and descriptive analysis—and the conversation that occurs between different methodologies as a result. The potential importance of this conversation extends beyond the NPF.

At least since the early 1990's the field of public policy has been divided between those that approach policy scholarship from the interpretivist perspective and those that approach public policy via attempting to emulate the natural sciences. This division has created silos of scholarship advancing and moving quite independently of one another. If the NPF is truly epistemologically adaptive—which we believe we have shown it is—then the framework presents a theoretical bridge of common language and concepts for researchers from both camps to broach a better understanding of meaning-making in public policy. While some may be averse to a turn by the NPF to qualitative methodologies with alternate epistemologies, it is worth remembering that the NPF was born of a merger of interpretivist theory and scientific method. Given the interpretivist's alpha to science's omega in the genesis of the framework, it seems self-consuming to exclude the former for the latter.

Footnotes

Acknowledgments

We would like to thank the Edmond J. Safra Center for Ethics at Harvard University and the Cultural Cognition Project at Yale University for providing the administrative, logistical, and financial support necessary to collect the data used for this research. We would also like to thank the anonymous peer reviewers.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Appendix A

References

Ackerman

Ayers

(2004) Voting with Dollars: A New Paradigm for Campaign Finance, New Haven: Yale University Press.

Almond

(2009) The dangers of hanging baskets: ‘Regulatory myths’ and media representations of health and safety regulation. Journal of Law and Society 36(3): 352–375.

Black

(2000) Proceduralizing regulation: Part I. Oxford Journal of Legal Studies 20(4): 597–614.

Brady

Collier

(2004) Rethinking Social Inquiry: Diverse Tools, Shared Standards, 2nd ed.. Rowman and Littlefield.

Burstein

(2014) American Public Opinion, Advocacy, and Policy in Congress: What the Public Wants and What the Public Gets, New York: Cambridge University Press.

Cairney

Heikkila

(2014) Comparing theoretical approaches. In: Weible

Sabatier

(eds) Theories of the Policy Process, 3rd ed. Boulder: Westview Press.

Charmaz

(2006) Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis, London: Sage.

Creswell

(2014) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, Thousand Oaks, CA: Sage.

Creswell

Clark

VLP

Gutmann

(2003) Advanced mixed methods research designs. In: Tashakkori

Eddlie

(eds) The Handbook of Mixed Methods in Social & Behavioral Research, Thousand Oaks: CA: Sage Publications, pp. 209–240.

10.

Creswell

Miller

(2000) Determining validity in qualitative inquiry. Theory into Practice 39(3): 124–130.

11.

Crow

Bergrren

(2014) Using the narrative policy framework to understand stakeholder strategy and effectiveness: A multi-case Aanalysis. In: Jones

Shanahan

McBeth

(eds) The Science of Stories: Applications of the Narrative Policy Framework in Public Policy Analysis, New York: Palgrave Macmillan.

12.

Dexter

(2006) Elite and Specialized Interviewing, European Consortium for Political Research (ECPR).

13.

Ewick

Silbey

(1995) Subversive stories and hegemonic tales: Toward a sociology of narrative. Law & Society Review 29(2): 198–226.

14.

Ewick

Silbey

(2003) Narrating social structure: Stories of resistance to law. American Journal of Sociology 108(6): 1328–1372.

15.

Fischer

Forester

(1993) The Argumentative Turn in Policy Analysis, Durham, NC: Duke University Press.

16.

Glaser

Strauss

(1967) The Discovery of Grounded Theory: Strategies for Qualitative Research, Chicago: Aldine.

17.

Grant

Rudolph

(2004) Expression vs. Equality: The Politics of Campaign Finance Reform, Columbus: Ohio State University Press.

18.

Gray

(2006) The regulation of corporate violations: Punishment, compliance, and the blurring of responsibility. The British Journal of Criminology 46(5): 875–892.

19.

Gray

(2009) The responsibilization strategy of health and safety: Neo-liberalism and the reconfiguration of individual responsibility for risk. British Journal of Criminology 49: 326–342.

20.

Gray

(2013a) Insider accounts of institutional corruption: Examining the social organization of unethical behaviour. British Journal of Criminology 53(4): 533–551.

21.

Gray

(2013b) The ethics of pharmaceutical research funding: A social organization approach. Journal of Law, Medicine & Ethics 41(3): 629–634.

22.

Gray

(2015) A sociological perspective on institutional corruption. In: Ferguson

(ed.) Global Corruption: Law, Theory and Practice, United Nations Office of Drugs and Crime, pp. 71–74.

23.

Gray

Silbey

(2011) The other side of the compliance relationship. In: Parker

Nielson

(eds) Explaining Compliance: Understanding and Explaining Organizational Responses to Regulation, Edward Elgar Publishing, pp. 123–138.

24.

Gray

Silbey

(2014) Governing inside the organization: Interpreting regulation and compliance. American Journal of Sociology 120(1): 96–145.

25.

Hajer

(1993) Discourse coalitions and the institutionalization of practice: The case of acid rain in Britain. In: Fischer

Forester

(eds) The Argumentative Turn, Durham, NC: Duke University Press, pp. 43–76.

26.

Haltom

McCann

(2004) Distorting the Law: Politics, Media, and the Litigation Crisis, University of Chicago Press.

27.

Heerwig

Shaw

(2014) Through a glass, darkly: The rhetoric and reality of campaign finance disclosure. Georgetown Law Journal 102: 1443–1450.

28.

Johnson

Onwuegbuzie

(2004) Mixed methods research: A research paradigm whose time has come. Educational Researcher 33(7): 14–26.

29.

Jones

(2013) Cultural characters and climate change: How heroes shape our perceptions of climate science. Social Science Quarterly 95(1): 1–39.

30.

Jones

McBeth MK (2010) A narrative policy framework: Clear enough to be wrong? Policy Studies Journal 38(2): 329–353.

31.

Jones

Radaelli

(2015) The narrative policy framework: Child or monster? Critical Policy Studies. Epub ahead of print 18 June 2015. DOI: 10.1080/19460171.2015.1053959.

32.

Jones

Shanahan

McBeth MK (2014) Introducing the narrative policy framework. In: Jones

Shanahan

McBeth

(eds) The Science of Stories: Applications of the Narrative Policy Framework in Public Policy Analysis, New York: Palgrave Macmillan, pp. 1–26.

33.

Jones

Song

(2014) Making sense of climate change: How story frames shape cognition. Political Psychology 35(4): 447–595.

34.

Jorgensen P, Ferguson T and Chen J (2013) Understanding Who Bought (and Continues to Buy) Your Politician. Paper presented at the Edmond J. Safra Center for Ethics at Harvard University. 13 February.

35.

King

Horrocks

(2010) Interviews in Qualitative Research, London: Sage Publication Ltd.

36.

King

Keohane

Verba

(1994) Designing Social Inquiry: Scientific Inference in Qualitative Research, Princeton University Press.

37.

Lessig

(2011) Republic, Lost, Boston: Twelve.

38.

Levi-Faur

(2011) Handbook on the Politics of Regulation, Edward Elgar Publishing Limited.

39.

Lincoln

Guba

(1985) Naturalistic Inquiry, London: Sage Publications.

40.

McBeth

Jones

Shanahan

(2014) The narrative policy framework. In: Weible

Sabatier

(eds) Theories of the Policy Process, 3rd ed. Boulder: Westview Press, pp. 225–266.

41.

McBeth

Shanahan

Arnell

(2007) The intersection of narrative policy analysis and policy change theory. Policy Studies Journal 35: 87–108.

42.

McBeth

Shanahan

Jones

(2005) The science of storytelling: Measuring policy beliefs in greater yellowstone. Society & Natural Resources 18: 413–429.

43.

Ney

(2014) The governance of social innovation: Connecting meso and macro levels of analysis. In: Jones

Shanahan

McBeth

(eds) The Science of Stories: Applications of the Narrative Policy Framework in Public Policy Analysis, New York: Palgrave Macmillan, pp. 207–234.

44.

O'Bryan

Dunlop

Radaelli

(2014) Narrating the ‘Arab Spring’: Where Narrative Meets Heuristics in Legislative Hearings. In: Jones

Shanahan

McBeth

(eds) The Science of Stories: Applications of the Narrative Policy Framework in Public Policy Analysis, New York: Palgrave Macmillan, pp. 107–130.

45.

OpenSecrets (2013) Donor demographics: 2012 overview. OpenSecrets.Org. Available at: http://www.opensecrets.org/overview/donordemographics.php (accessed 27 May 2013).

46.

Parker

Nielson

(2011) Explaining Compliance: Business Responses to Regulation, Edward Elgar Publishing Limited.

47.

Pierce

Smith-Walter

Petrson

(2014) Research design and the narrative policy framework. In: Jones

Shanahan

McBeth

(eds) The Science of Stories: Applications of the Narrative Policy Framework in Public Policy Analysis, New York: Palgrave Macmillan, pp. 27–44.

48.

Prosser

(2010) The Regulatory Enterprise: Government, Regulation, and Legitimacy, Oxford: Oxford University Press.

49.

Riessman

(1993) Narrative Analysis. Qualitative Research Methods Series 30, London: Sage Publications, Inc.

50.

Roe

(1994) Narrative Policy Analysis: Theory and Practice, Durham, CT: Duke University Press.

51.

Saldana

(2013) The Coding Manual for Qualitative Researchers, 2nd ed. Thousand Oaks, CA: Sage Publications Ltd.

52.

Samples

(2006) The Fallacy of Campaign Finance Reform, Chicago: University of Chicago Press.

53.

Shanahan

Jones

McBeth

(2013) An angel on the wind: How heroic policy narratives shape policy realities. Policy Studies Journal 41(3): 453–483.

54.

Shenton

(2004) Strategies for ensuring trustworthiness in qualitative research projects. Education for Information 22: 63–75.

55.

Silbey

(1994) Patrick David: To bring out the best… to undo a little pain. In: Kolb

(ed.) When Talk Works: Profiles of Mediators, San Francisco: Jossey-Bass Publishers, pp. 61–104.

56.

Smith

(2003) Unfree Speech: The Folly of Campaign Finance Reform, Princeton: Princeton University Press.

57.

Smith

Larimer

(2013) The Public Policy Theory Primer, 2nd ed. Boulder, CO: Westview Press.

58.

Stone

(2012) Policy Paradox: The Art of Political Decision Making, 3rd ed. New York: W.W. Norton & Company.

59.

Weible

(2014) Moving forward and climbing upwards. In: Weible

Sabatier

(eds) Theories of the Policy Process, 3rd ed. Boulder: Westview Press.