Using the Extended Case Method to Expand the Scope of Policy Research: An Examination of the Educational Outcomes of a College Preparation Program for Low-Income,Racially Minoritized Students

Abstract

This article describes how the extended case method, a tool of critical qualitative inquiry rooted in ethnography, can be used to inform policy research. Using examples drawn from a yearlong ethnographic study of a college preparation program, it demonstrates the utility of the extended case method for policy research through a discussion of literature on educational policy and qualitative research methods. It then uses study findings to show how the extended case method can address challenges related to context and meaning in policy evaluation focused on causal relationships. Implications for future qualitative policy work are also offered.

Keywords

college access education policy evaluation ethnography extended case method financial aid qualitative research

While most education researchers agree that good policy is driven by empirical evidence (e.g., Cooper, Cibulka, & Fusarelli, 2008; Rippner, 2015), there is at best a tenuous connection between social science and policy promulgation (c.f. Lagemann, 2002; Stone, 2012). The government officials who actually make and interpret educational policy do so within a complex ecological system requiring ongoing negotiation and drawing on many types of information (e.g., polling data, anecdote, and personal philosophy) beyond that provided by rigorous social science (Anderson, 2015). When social science evidence is used by policy makers to shape policy, they often display a marked preference for quantitative evidence that lends the illusion of objectivity (e.g., Igo, 2007; Sellar, 2015).

This potential discontinuity between the way that policy is made and the evidence used to justify it has profound consequences for policy evaluation: Simply put, policy represents a formal intervention in educational systems but may be based on a theory of action that has never been explicitly stated (c.f. Patton, 2015; Sanderson, 2002). This ambiguity has significant consequence for social scientific inquiry since researchers interested in understanding intervention effects need to understand a policy’s theory to produce a logic model that offers preliminary hypotheses about how a particular practice or policy should shape outcomes within a given context (e.g., Cooksy, Gill, & Kelly, 2001; Fielden et al., 2007; Julian, 1997). However, good policy evaluation requires the reconstruction of a policy’s logic model by documenting the environmental factors that connect program inputs to program outcomes (Astbury & Leeuw, 2010) and then working to eliminate “known unknowns” about how the policy functions (Pawson, Wong, & Owen, 2011, p. 518). The simultaneous rigor, flexibility, and focus on social relations fostered by qualitative methods make them uniquely well-suited to exploring ambiguous theories of action (Abma, 2000, 2006; Abma & Widdershoven, 2008; Harklau & Norwood, 2005), and therefore well-suited to policy analysis.

Although the capacity of qualitative methods to respond to ambiguity and to provide rigorous policy evaluation has been reasonably well-established (e.g., Gaber & Gaber, 2018; Pasque & Lechuga, 2017; Yanow, 1999), most available qualitative approaches focus on either understanding the human consequences of policy or on the fidelity of policy implementation. This framing is consistent with the general belief that qualitative methods excel at providing explanation and description but are weaker in documenting causation or allowing for future prediction (e.g., Creswell & Creswell, 2017; Luker, 2008; Mertens, 2014). However, as Maxwell (2004) has noted, qualitative methods can be used effectively to study causation. Acknowledging this potential “recognizes the reality and importance of meaning, as well as of physical and behavioral phenomena, as having explanatory significance,” and “also recognizes the explanatory importance of the context of the phenomena studied” (Maxwell, 2004, pp. 8-9). In response to these arguments, qualitative methodologists have most frequently advanced grounded theory (e.g., Birks & Mills, 2015; Gibson & Hartman, 2013), causal case methods (e.g., Beach & Pedersen, 2016; Yin, 2017), and process tracing techniques (e.g., George & Bennett, 2005; Kay & Baker, 2015) as methodological approaches capable of addressing causation in a rigorous way.

In this article, I argue that the extended case method should also be considered as a qualitative methodology capable of exploring causal relationships for education policy analysis. The extended case method is a methodological orientation employed by critical ethnographers to refine theories (e.g., Burawoy, 1991, 1998, 2009; Eliasoph & Lichterman, 1999; Tavory & Timmermans, 2009). In the extended case method, researchers seek to test, revise, and create meso-level theories capable of explaining how human beings experience societal organizations and structures (e.g., policy, belief systems, and social movements) using sustained participant observation and deep theoretical engagement (c.f. Burawoy, 1998; Van Velsen, 1979). These attributes mean that the extended case method is uniquely well-suited to addressing Maxwell’s (2004) argument for the importance of both meaning and context in causal analysis. The emergent nature of researcher understanding in the extended case method also makes it useful for unpacking the ambiguous logic models inherent in many educational policies. To date, however, only a limited number of studies in education have used the extended case method for exploring causal relationships (e.g., Arndt & Davis, 2011; Kimball, 2016) and none of these studies have had an explicit focus on policy outcomes.

I therefore present partial findings from a yearlong ethnographic study situated in Conscious of College, the pseudonym for a college preparation program serving low-income, racially minoritized high school students in an underresourced high school. Examples from this study demonstrate the utility of the extended case method in evaluations of education policy. Focused on the way in which participants in the college preparation program understood federal legislation related to financial aid and the way in which that interpretation led them to engage in college-going behaviors, these examples show that study participants were more likely to respond to posted tuition prices than to net cost after financial aid and therefore to display suppressed college-going outcomes, which echoes prior quantitative work on the topic (e.g., Chen & DesJardins, 2010; Epple, Romano, & Sieg, 2006; Perna, 2010). However, findings from this study also deepen prior knowledge of college-going by showing that the policy logic model driving financial aid policy fails to include key factors influencing the way that participants in the college preparation program understood the costs of college-going (e.g., the timing of student definition of an institutional choice set; the resources available to some students to make sense of complex financial aid policies). These findings emerge as a result of sustained engagement with relevant social theory (Hughes, Kimball, & Koricich, in press).

Extended Case Method

The extended case method provided the research design for my investigation of the Conscious of College program. Critical ethnographers employed by or trained within the Department of Social Anthropology at the University of Manchester in 1950s and 1960s originally developed the extended case method as way to examine specific events or processes within a bounded context (c.f. Handelman, 2006; Van Velsen, 1979). Although the approach later fell out of favor, its core tenets were adapted for theory testing and refinement by later ethnographers (e.g., Burawoy, 1998; Eliasoph & Lichterman, 1999; Tavory & Timmermans, 2009). Notably, this reconfiguration of the core purpose of the extended case method led to an abductive epistemological orientation that makes it quite useful for capturing and responding to emergent findings from within the overall framework of social theory. In abduction, the simultaneous utilization of inductive reasoning (the use of specific observations to produce abstract conclusions) and deductive reasoning (the use of abstract rules to generate specific conclusions) allows for the development of theories that are well-supported by both prior scholars and empirical observation. As described by Burawoy (1998), practitioners of the extended case method begin by using existing social theory to predict what rigorously collected but localized data should show if the theories used are well-chosen and well-reasoned. If the collected data does not align with theoretical predictions, the researcher then revisits the data to reflexively examine the original prediction and determine if the flawed prediction arose from problematic theory or from bias introduced by the researcher. Assuming that the problems arose from the theory, the observed deviations then become the basis for proposed modifications to the social theories used to generate the predictions.

Notably, Burawoy’s (1991, 2009) formulation of the extended case method is intended to exist within the structures of normative social science, which remains dominated by discourses of postpositivism and pragmatism (Fay, 1996). However, the extended case method does not produce claims of “truth” or “falsity” that are “valid” in the same way that social scientists typically use these concepts (c.f. Fay, 1996; Burawoy, 2009). In the extended case method, researchers combine empirical and theoretical knowledge through the lens of personal experience to create theoretical propositions that should be tested by future empirical inquiry. In so doing, the extended case method provides an explanation of observed causal relationships but does not produce definitive causal determinations. It is therefore a useful way to sift through complex, interdependent educational learning environments, identify potential causal relationships therein between program or policy interventions and observed outcomes, and provide a theory-driven explanation for those causal relationships. That is, the extended case method serves to separate the signal (causation) from the noise (everything else) in educational environments while also providing an explanation for why the signal is found where it is. As such, it can help structure future causal inquiry while itself providing abductive evidence of causation.

As Burawoy (1991) has noted, the extended case method “seeks generalization through reconstructing existing generalizations, that is, the reconstruction of existing theory” (p. 279). Perhaps the best way to view the extended case method is as a “fact” checking mechanism for social science. It does so through four key extensions of the way that normative social science typically proceeds:

. . . the extension of observer into the lives of participants under study, the extension of observations over time and space; the extension from microprocesses to macroforces; and, finally, and most important, the extension of theory. Each extension involves a dialogue: between participant and observer, between successive events in the field, between micro and macro, and between successive reconstructions of theory. These dialogues orbit around each other, each in the gravitation field of the others. (Burawoy, 2009, p. xv)

This framing for the extended case method marks it as a distinctly reflexive approach to qualitative inquiry yet one that can produce findings that exist in dialogue with normative social science. In short, it is a highly structured method for reinserting human judgment and experience into rigorous inquiry.

Research Design for Conscious of College Case Study

In order to better understand the Conscious of College program, I used an extended case method framework to understand how participating students experienced the college-going process. I first identified a preliminary theoretical model driven by social theory that I believed should capture how Conscious of College participants made decisions about whether to apply to college, where to apply, and where they would enroll after admission. More specifically, I used the theory of dual commodification (Hughes et al., in press) to understand how student goals and behaviors would likely be influenced by institutional goals and behaviors. The theory of dual commodification, which is based heavily on social theory describing the relationship between education and consumer behavior (e.g., Baker, 2014; Cohen, 2003; Collins, 1979; Cross, 2002; Riesman, 1980), holds as follows:

Colleges provide messages (e.g., mailings to students registered to take the ACT or SAT; high school visits by admissions officers) that signal their availability to students.

Students consider their own individual commitments and relationships to family, friends, and the community as they process information from colleges.

Students can also provide messages (e.g., strong performance on tests; participation in cocurricular activities like band or athletics) that signal to colleges that they are desirable candidates for admission and merit-based financial aid.

Colleges use messaging strategies (e.g., merit-based financial aid; personal attention) to shape the college-going decisions of preferred students.

Students choose a college to attend from among those they consider to be reasonable options based on their personal commitments, relationships, and the messages provided by specific colleges.

As a represented by the prior five items, college-going is a social process that changes over time based on societal factors (e.g., educational systems; financial aid policy).

These elements of the theory of dual commodification served as theoretical predictions about how college-going would work for participants in the Conscious of College program that could be tested and refined using the extended case method.

In order to generate the data necessary to do so, I worked with a team of two research assistants to undertake a yearlong ethnographic inquiry. To collect data about the Conscious of College program, this inquiry used roughly 250 hours of participant observation as well as nine formal interviews with program participants and staff members; frequent informal interviews with both program participants and staff members; and two focus groups with program participants. Throughout the study, the research team utilized a system of fieldnotes, regular memoing, and research team meetings to produce shared understanding of the Conscious of College program. Research team members also generated verbatim transcripts of all formal interviews and focus groups—translating from Spanish to English as necessary. Fieldnotes, memos, and transcripts then formed the data for analysis.

Consistent with the extended case method’s focus on the iterative reconstruction of theory (Burawoy, 2009), the research team began with a deductive coding framework consisting of a priori codes drawn from the theory of dual commodification. However, the extended case method’s focus on emergence (Burawoy, 1998) also led us to adopt an inductive coding approach driven by our emerging understanding of the Conscious of College program. First-order inductive coding focused on identifying major ways that study data could be categorized. For example, first-order codes focused on participant identities, steps in the college-going process, and attributes of the Conscious of College program. Second-order inductive codes aggregated first-order codes into larger units of meaning or linked them with elements of the dual commodification framework. For example, second-order codes focused on participants’ views of financial aid programs, college costs, and the ranges of institutions to which they might reasonably apply or expect to attend. These second-order codes were then compared with the theory of dual commodification to understand more fully where the model accurately described participants’ experiences and where some revision was warranted.

The Conscious of College Program

Funded by U.S. Department of Education under the auspices of the TRiO Program, the Conscious of College (pseudonym) program works with approximately 60 students in Grades 9 to 12 attending New High School (pseudonym) to help them form and realize college-going aspirations. In part, due to higher dropout rates and suppressed college-going aspirations among cisgender young men and transgender adolescents (Dockery, 2012), cisgender young women participate in Conscious of College programs at roughly twice the rate of cisgender young men and transgender adolescents. Although Conscious of College is technically open to all students that meet the program’s participation criterion of serving low-income and/or first-generation students, participants almost exclusively come from racially minoritized backgrounds (primarily identifying as Black and/or Latinx). These percentages reflect the broader demographic composition of New High School, where more 90% of the roughly 1,000 enrolled students come from racially minoritized backgrounds. Notably, the percentages of students from racially minoritized backgrounds in both New High School and Conscious of College are far higher than the school-aged population of the surrounding community. However, families with higher levels of economic, social, and cultural capital use enrollment in local private high schools and statewide school choice policies allowing students to attend other public high schools or charter schools to opt out of a school district widely perceived as struggling. With decreased personal stakes in local schools, these same families have then resisted increased financial investment in the local school district. As a result, problems endemic to underfunded schools (e.g., crumbling infrastructure, low teacher morale, and limited educational offerings) also affect New High School (Oakes, 2005). In response, a school-community-university partnership developed Conscious of College to supplement existing New High School offerings related to college-going and help some of the highest achieving students attend college.

Participants in the Conscious of College program take part in two types of activities intended to support their academic development and foster their college-going aspirations. During the academic year, Conscious of College provides structured academic support and educational programs that participants must regularly attend to remain in the program. The Conscious of College space is markedly quieter and calmer than the remainder of the New High School campus. It is also administered by three professional staff members, which means that Conscious of College participants experience a lower staff-to-student ratio than other New High School students. Conscious of College professional staff also regularly initiate contact with students who they perceive to be struggling, or who have not consistently participated in required activities. In so doing, staff display an interest in understanding each student’s experience that Conscious of College participants note to be less frequently displayed by other New High School staff members.

The Conscious of College professional staff members offer regular workshops on various aspects of the college-going process (e.g., completing applications and filing for financial aid) attended by Conscious of College participants and their families. The collective intention of these programs is to help normalize the experience of college-going within the New High School environment where less than half of all students graduate from high school and a fraction of the total student enrollment end up attending college after high school. The Conscious of College also offers extensive summer programming, when program participants complete college visits and attend courses at a local college. By asking participants to take courses at a local college, the program aims to demonstrate that students participating in Conscious of College have already partially realized their college-going aspirations by the time that they graduate from high school. The stated program logic behind this activity is that students for whom the act of college-going has already been demystified will be more likely to enroll in college after high school graduation.

Reframing Theoretical and Policy Logic Models: A Case Study

To show the utility of the extended case method for policy evaluation, I now offer exemplar findings, a discussion of those findings, and a discussion of the utility of the extended case method therein. Importantly, given the methodological focus of this article, the findings presented below can demonstrate the range of data available to extended case methodologists and the conclusions to which it might lead. However, that same methodological focus makes it impossible to present these findings with sufficient depth or richness to establish the verisimilitude of the conclusions. That is, I focus only on a narrow subset of the total data available from the overall study and still present only snippets of that focused analysis. To establish the verisimilitude of findings, I triangulate them via citation to prior work that drew similar conclusions—albeit without making of the connections made herein.

This subanalysis explores how students consider financial incentives in the college-going process and how those financial incentives are structured within the overall framework of federal financial aid policy. In order to respond to these theoretical predictions, I also needed to develop an in-depth understanding of the policy logic model behind federal financial aid policy. The second half of the 20th century saw college redefined as a private rather than public good (Kimball, 2011). As part of that redefinition, federal financial aid policy shifted from efforts intended to reduce the costs of college-going for the neediest students to efforts to get all students to view college as a form of investment. In the later framing, the primary goal of federal financial aid is to reduce the perceived opportunity cost of college-going by offering readily available credit or modest reductions in cost. The logic model behind this policy framing might reasonably depicted as shown in Figure 1:

Figure 1.

Policy logic model for federal financial aid.

According to this framing, students begin the college-going process with a relatively high level of certainty that they will attend college, which leads them to an initial assessment of the opportunity cost that favors college-going. They then learn that the perceived initial cost of college will be reduced further through federal financial aid, which virtually everyone who applies for will receive (Radwin et al., 2018), and institutional aid, which is artificially inflated by high discount rates at many institutions (Baum & Ma, 2010). As a result, students will end up with a revised opportunity cost that is even more favorable toward college-going, and most people who begin this process will end up in college, which is a reasonable assumption for some student populations (Shapiro et al., 2017).

While space constraints preclude the inclusion of a full findings section, the strength of the extended case method can be seen through a briefer discussion of several key examples from the data collection process that served to alert the research team to discrepancies between the theory of dual commodification, the policy logic model for federal financial aid programs, and our observations of Conscious of College programs. We first noted the salience of financial aid to the college-going process when reviewing program artifacts as part of our initial document analysis. Not only did the program regularly communicate scholarships received by participants but it also frequently included financial aid and financial literacy workshops among their academic year and summer program offerings. For example, in the planning calendar used by Conscious of College staff and participants to structure the academic year, 4 out of the 32 listed program milestones or events directly address financial aid issues. Only steps related to submitting applications to colleges appear more frequently. Notably, additional program documents also revealed that the choice of colleges to which to apply incorporated financial considerations as well, as students were asked to consider the likelihood of receiving various forms of aid at each institution in which they expressed interest.

Interviews with participating students revealed the gaps in knowledge that these programs and practices sought to fill. For example, one sophomore noted that they knew that:

Some of them may cost more and some of them cost less. I don’t really know how scholarships work, but I think some colleges don’t offer some particular scholarships. So when I consider going to a college, I will probably consider all the stuff, like the cost, how far away it is from my house, the size of it and yeah . . .

As the interviewee stopped without finishing that thought, the research team member conducting the interview then asked, “Who do you talk to about college?” The student then responded: “No one in my family really. Pretty much in the past couple weeks, I have been talking to the tutor mentors.” The research team member then asked the student to confirm that they meant that they had just started thinking about college, which they did. For a high school student actively contemplating college applications, this late start provided a considerable disadvantage relative to other college-going students whose anticipatory socialization to college-going often begins in middle school, if not earlier (Hill & Wang, 2015). The example also reflected the reality that for many Conscious of College participants, even well-meaning parents were unsure of how to support their children’s college-going aspirations in the context of complex admissions and financial aid processes. Consistent with the theory of dual commodification’s focus on how students understand their available choices, this restricted view of the college-going landscape could artificially limit the range of options which students judge to be reasonable ones.

Likewise, interviews with Conscious of College professional staff members revealed that they considered students’ understandings of cost to be a critical factor in college-going. In response to a question about how students’ identities shaped their college-going processes, one staff person noted as follows:

Social class can be scary . . . That goes back to . . . making sure you have the funds and the finances. They might feel as though an Ivy League school is out of their reach because their parents don’t make a six-figure salary, so they can’t keep up with the tuition—not realizing that those Ivy Leagues also have opportunities and programs to embrace some people that might not be as fortunate. So they’re definitely intimidated by the cost the dollars of attending a college or university that they feel in some ways has been told to them, time after time: “It’s way out of your league.”

In response to this recognition, the Conscious of College professional staff designed its programming around the acquisition of key information about college-going as a formalized process with unwritten rules that must be learned. Recognizing that academic achievement alone would not increase the rates of college-going at New High School, the Conscious of College program sought to provide otherwise inaccessible knowledge about issues such as federal financial aid and various forms of scholarship.

In providing this socialization to the norms of college-going, even the Conscious of College professional staff admitted that they faced some pronounced limitations. For example, a second professional staff person compared program participants with students utilizing fee-based college counselors:

You can’t even begin to pretend [the advice we give is] the same as that person who costs five grand, but then people are saying that their parents are going to get that back dollar for dollar in their financial aid packages . . . so this is who my kids are up against when they’re looking at small liberal arts colleges that are semiprestigious to prestigious.

Not only do students from high-income backgrounds typically enter the college-going process with higher levels of college-going knowledge but they also have access to experiences throughout their schooling that allow them to shape and understand financial aid offers differently than Conscious of College participants (c.f. Goldrick-Rab, Kelchen, Harris, & Benson, 2016; Perna, 2010). Moreover, those same high-income students find themselves more likely to be indirectly targeted by colleges via high school visits, which are disproportionately centered in well-resourced schools (Jaquette & Salazar, 2018). As a result, students from high-income backgrounds may be more likely to receive institutional aid from colleges that are not simply engaged in extreme tuition discounting (Baum & Ma, 2010). The data from the Conscious of College program provides some preliminary evidence for this conjecture, while the Conscious of College program reports a high volume of scholarship dollars, most of those institutional aid offers come from local tuition dependent colleges, regional public universities, or community colleges.

Collectively, these examples reflect the overall pattern observed in study data related to financial aid: participants expressed a marked lack of clarity in understanding the costs of college, which influenced their college-going decisions. For some, it led them to discount potential colleges that might have been able to provide more robust financial aid offers. For others, the limited availability of good resources—for many participants, consisting only of Conscious of College staff members—meant that college-going decisions would be made with incomplete information. As a result, data supporting the subanalysis reported in this article calls into question both the way that social theory (as depicted by the theory of dual commodification) and federal financial aid policy (as depicted by the logic model presented above) understand the process of college-going. Instead, overall findings from this study echo prior quantitative studies that have shown that students have varied levels of price sensitivity (Epple et al., 2006; Perna, 2010) and that these differences can be explained in part by demographic characteristics such as race (Chen & DesJardins, 2010) and socioeconomic status (Paulsen & St. John, 2002).

By virtue of its use of the extended case method, however, this study provided key contextual information about the causal mechanisms via which participants in the Conscious of College program interpreted federal financial aid policy. It can therefore provide a roadmap for how future studies seeking to make causal attributions might be structured and can offer concrete suggestions for intervening in the way those policies are structured in order to produce better outcomes. The revised policy logic model for federal financial aid policy is shown in Figure 2.

Figure 2.

Revised policy logic model for federal financial aid.

As can be seen in Figure 2, several refinements to the initial logic model have been made to account for the lived experiences of Conscious of College participants. These lived experiences can and should inform our understanding of policy logic. Notably, these added elements to the logic model can easily be triangulated by prior data, which lends greater strength to the revised logic model. For example, prior research has shown that parents, siblings, and other family members can powerfully affect college-going (e.g., Freeman, 2012; Perez & McDonough, 2008; Rodriguez & Nuñez, 2015). Likewise, the importance of peer environments and fictive kinship networks to college-going has long been acknowledged by higher education researchers (e.g., Deil-Amen & Martinez, 2015; Hoxby & Turner, 2015; Welton & Martinez, 2014). Finally, several studies have shown the role that student understanding of financial aid can play in predicting decisions to attend or not attend college (e.g., Goldrick-Rab et al., 2016; Page & Scott-Clayton, 2016; Ziskin, Fischer, Torres, Pellicciotti, & Player-Sanders, 2014).

The unique contribution of the extended case method is not its innovation of findings, but rather its ability to link experiences, outcomes, and context in ways that support new understandings of causal relationships. That is, it tells a new story about causation by linking new and old knowledge: To that end, the revised policy logic model presented above still retains the fundamental assumption that students generate and revise their understanding of opportunity costs as part of the college-going process, but it now includes additional elements intended to represent the fact that not all students have the same level of knowledge about the costs of college-going. Specifically, a students’ understanding of financial aid is now included as an interstitial step between reduced cost arising from various forms of aid and their revised opportunity cost. This step acknowledges that, while financial aid is available to virtually all students, each individual student must understand how it works in order to benefit from it.

Since the vast majority of federal financial aid comes in the form of loans (Radwin et al., 2018), interpreting aid packages requires prior understanding of credit. Few high school students are well-positioned to generate this understanding independently—relying instead on information available from family members, school personnel, or their peers. Notably, however, many people from low-income and racially minoritized backgrounds have good reason to be skeptical of credit practices—coming from communities that have been targeted by redlining tactics that prevent the acquisition of home loans and predatory lending practices that result in usurious rates (Calder, 2001; Donohue, 2006)—and may not be well-positioned to provide advice about student loans. Likewise, many low-income and racially minoritized students attend underresourced schools where they have only limited access to college-going information (Oakes, 2005). As a result, differences in socialization to college-going norms may condition the way that students experience the financial aid process and likely also mean that preconceived, potentially negative assumptions about both institutional aid and federal financial aid structure the initial determination of opportunity costs. Not only does this finding have implications for the policy logic model for federal financial aid but it also suggests that any analysis of the dual commodification process must address contextual variations in an ongoing and systematic way; the inclusion of individual influences on college-going in the model’s second stage and societal influence in the model’s sixth stage must be interpreted expansively to account for complex interactions between context and college-going.

Implications for Policy Research

Despite the many potential benefits of the extended case method, it is seldom used in educational research at present—with only a handful of studies adopting it as a methodology (e.g., Arndt & Davis, 2011; Kimball, 2016). This study clearly shows, however, that the extended case method approach produces useful information about educational settings—including evidence that can inform understanding of policy. By showing that existing social theory and financial aid policy do not fully appreciate the importance of contextual influences on student outcomes, this study offers important implications of future policy research. For example, the examples from this study provided above show that understanding of financial aid may matter for some students as much as offering that financial aid. Consequently, in order for financial aid to be effective, educational programs targeting those populations least likely to respond positively to loan-based aid would be particularly helpful. Moreover, given that much of the interpretation of these programs takes place within the community context, outreach to families, peers, and schools would also likely improve the effectiveness of aid programs.

Examples from this study also provide a key indication of the importance of combining effective storytelling and rigorous social science. Researchers have long argued that qualitative methods can be utilized for policy evaluation (e.g., Gaber & Gaber, 2018; Pasque & Lechuga, 2017; Yanow, 1999); however, policy makers have demonstrated a preference for statistical evidence and have been slower to recognize the potential for qualitative research to offer rigorous data about program outcomes (e.g., Porter, 1995; Sellar, 2015). However, qualitative methods offer the unique capacity to provide a human perspective on policy outcomes that would likely complicate narratives relying on either the supposed objectivity of quantification or the personal connections provided by anecdote. Juxtaposing the student narrative about familial participation in the college-going process—(“no one in my family really”)—to the narratives of students with more economic, cultural, and social capital would offer an actionable form of social science in addition to a compelling form of storytelling. In short, when done well, qualitative methods offer the same levels of rigor as the most robust quantitative designs while also resulting in a clear, compelling narrative driven by human experience. At a time where the collective expertise of the academy is being called into question by both politicians and the general public, qualitative methods can offer evidence about policy effects using stories the average person can understand utilizing methodological sophistication most people do not possess (Kimball & Friedensen, 2018).

Moreover, in comparison with most qualitative and quantitative approaches, the extended case method is uniquely well-suited to address context in a detailed way (Burawoy, 1998). As noted by Maxwell (2004), good evidence on outcomes should seek to explain not just what has occurred but also why the people involved in producing these outcomes have behaved as they did (representing their meaning making) and the environments in which these behaviors took place (representing context). While most qualitative approaches can provide good information about participant meaning making (e.g., Creswell & Creswell, 2017; Luker, 2008; Mertens, 2014), not all can systematically address context independent participant accounts. In the examples above, participants provide explanations and interpretations of empirical observations made by the researchers. That is, the extended case method helps establish both that something has occurred and also to address to why it has occurred. It can do so as a result of the deep engagement with study environments fostered by the use of ethnography as a methodological approach.

The extended case method is also one of a very limited subset of qualitative methodologies that can address causation in a rigorous way—joining only grounded theory (e.g., Birks & Mills, 2015; Gibson & Hartman, 2013), causal case methods (e.g., Beach & Pedersen, 2016; Yin, 2017), and process tracing techniques (e.g., George & Bennett, 2005; Kay & Baker, 2015) among commonly used approaches. However, for qualitative methodologists trained within antipositivist paradigms (e.g., interpretivism, constructivism, and poststructuralism), the extended case method may also offer a potential improvement over each of these approaches: it is tied directly to reflexive practices and acknowledges the central role of the researcher within the inquiry process. The centrality of the researcher can clearly be seen in the overall structure of the examples provided above: beginning with dual commodification as a theoretical model, the study design also incorporated the policy logic model of federal financial aid and resulted in proposed revisions to both. This outcome occurred because the researchers using the extended case method provide simultaneous attention to theory and empirical evidence in order to understand whether the best available theoretical predictions accurately describe the observed reality. If they do not, the extended case method produces refined theoretical propositions that can then be tested utilizing more traditional methods. Due to this emergent design property, the extended case method is uniquely well-suited to addressing ambiguous problems, which are all too common in the study of social processes and policy outcomes (e.g., Astbury & Leeuw, 2010; Pawson et al., 2011).

In addition to implications related to the use of the extended case method in policy research, this article also provides two important reminders about qualitative policy research more broadly. First, it makes clear that that there is a linkage between individual outcomes and policy outcomes. That is, for many educational policies, what makes a policy successful is the impact that it has on students. While it may be convenient or even useful to aggregate students’ experiences into collective outcomes, they will always be individuals with potentially widely varied experiences (Friedensen, McCrae, & Kimball, 2017). Analyses of policy outcomes that do not interrogate these varied experiences and the meanings students ascribe to them will only ever tell part of the story. Second, due to the process of aggregation, many policy analyses present the image of monolithic social organizations (e.g., colleges, cities, and states). However, the effect and nature of implementation of policies almost always varies from context-to-context (Maxwell, 2004). In other words, there is no singular experience of a policy’s impact but rather a wide range of potential experiences due to the interface between individuals, institutions, and policies. In the storytelling around quantitative evidence, complex data can often be reduced to a relatively singular narrative (Porter, 1995). Qualitative policy work offers an ongoing reminder that reality is rarely so simple.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Author Biography

Ezekiel Kimball is an assistant professor of Higher Education and Associate Director of the Center for Student Success Research at the University of Massachusetts Amherst. His research agenda focuses on college access and postsecondary learning outcomes for minoritized student populations. Much of his recent work focuses on the higher education experiences of students with disabilities.

References

Abma

(2000). Responding to ambiguity, responding to change the value of a responsive approach to evaluation. Evaluation and Program Planning, 23, 461-470.

Abma

(2006). The practice and politics of responsive evaluation. American Journal of Evaluation, 27, 31-43.

Abma

Widdershoven

(2008). Evaluation and/as social relation. Evaluation, 14, 209-225.

Anderson

(2015). Public policymaking (8th ed.). Stamford, CT: Cengage Learning.

Arndt

Davis

(2011). Warrior spirit: Soul wound and coping among American Indians in law enforcement. The Counseling Psychologist, 39, 527-569.

Astbury

Leeuw

(2010). Unpacking black boxes: Mechanisms and theory building in evaluation. American Journal of Evaluation, 31, 363-381.

Baker

(2014). The schooled society: The educational transformation of global culture. Stanford, CA: Stanford University Press.

Baum

(2010). Tuition discounting: Institutional aid patterns at public and private colleges and universities: Trends in Higher Education Series. New York, NY: College Board.

Beach

Pedersen

(2016). Causal case study methods: Foundations and guidelines for comparing, matching, and tracing. Ann Arbor: University of Michigan Press.

10.

Birks

Mills

(2015). Grounded theory: A practical guide. Thousand Oaks, CA: Sage.

11.

Burawoy

(1991). Ethnography unbound: Power and resistance in the modern metropolis. Berkeley: University of California Press.

12.

Burawoy

(1998). The extended case method. Sociological Theory, 16, 4-33.

13.

Burawoy

(2009). The extended case method: Four countries, four decades, four great transformations, and one theoretical tradition. Berkeley: University of California Press.

14.

Calder

(2001). Financing the American dream: A cultural history of consumer credit. Princeton, NJ: Princeton University Press.

15.

Chen

DesJardins

(2010). Investigating the impact of financial aid on student dropout risks: Racial and ethnic differences. Journal of Higher Education, 81, 179-208.

16.

Cohen

(2003). A consumers’ republic: The politics of mass consumption in postwar America. New York, NY: Knopf.

17.

Collins

(1979). The credential society. New York, NY: Academic Press.

18.

Cooksy

Gill

Kelly

(2001). The program logic model as an integrative framework for a multimethod evaluation. Evaluation and Program Planning, 24, 119-128.

19.

Cooper

B. S.

Cibulka

J. G.

Fusarelli

L. D.

(Eds.). (2008). Handbook of education politics and policy. New York, NY: Routledge.

20.

Creswell

(2017). Research design: Qualitative, quantitative, and mixed methods approaches. Thousand Oaks, CA: Sage.

21.

Cross

(2002). An all-consuming century. New York, NY: Columbia University Press.

22.

Deil-Amen

Martinez

(2015). College for all Latinos? The role of high school messages in facing college challenges. Teachers College Record, 117, 1-50.

23.

Dockery

(2012). School dropout indicators, trends, and interventions for school counselors. Journal of School Counseling, 10(12). Retrieved from http://jsc.montana.edu/articles/v10n12.pdf

24.

Donohue

(2006). Freedom from want: American liberalism and the idea of the consumer. Baltimore, MD: Johns Hopkins University Press.

25.

Eliasoph

Lichterman

(1999). “We begin with our favorite theory . . . ”: Reconstructing the extended case method. Sociological Theory, 17, 228-234.

26.

Epple

Romano

Sieg

(2006). Admission, tuition, and financial aid policies in the market for higher education. Econometrica, 74, 885-928.

27.

Fay

(1996). Contemporary philosophy of social science: A multicultural approach. Malden, MA: Blackwell.

28.

Fielden

Rusch

Masinda

Sands

Frankish

Evoy

(2007). Key considerations for logic model development in research partnerships: A Canadian case study. Evaluation and Program Planning, 30, 115-124.

29.

Freeman

(2012). African Americans and college choice: The influence of family and school. Albany: SUNY Press.

30.

Friedensen

McCrae

Kimball

(2017). Using qualitative research to document variations in student experience. In Kimball

Loya

(Eds.), Using qualitative research to promote organizational intelligence (pp. 53-64). San Francisco, CA: Jossey-Bass.

31.

Gaber

(2018). Qualitative analysis for planning and policy: Beyond the numbers. New York, NY: Routledge.

32.

George

A. L.

Bennett

(2005). Case studies and theory development in the social sciences. Cambridge: MIT.

33.

Gibson

Hartman

(2013). Rediscovering grounded theory. Thousand Oaks, CA: Sage.

34.

Goldrick-Rab

Kelchen

Harris

Benson

(2016). Reducing income inequality in educational attainment: Experimental evidence on the impact of financial aid on college completion. American Journal of Sociology, 121, 1762-1817.

35.

Handelman

(2006). The extended case: Interactional foundations and prospective dimensions. In Evens

T. M. S.

Handelman

(Eds.), The Manchester School: Practice and ethnographic praxis in anthropology (pp. 94-117). New York, NY: Berghahn Books.

36.

Harklau

Norwood

(2005). Negotiating researcher roles in ethnographic program evaluation. Anthropology & Education Quarterly, 36, 278-288.

37.

Hill

Wang

(2015). From middle school to college: Developing aspirations, promoting engagement, and indirect pathways from parenting to post high school enrollment. Developmental Psychology, 51, 224-235.

38.

Hoxby

Turner

(2015). What high-achieving low-income students know about college. American Economic Review, 105, 514-517.

39.

Hughes

Kimball

Koricich

(in press). Multidirectional effects in college-going: A review of the literature and a conceptual model for future research. In Paulsen

(Ed.), Higher education: Handbook of theory and research. New York, NY: Springer.

40.

Igo

S. E.

(2007). The averaged American: Surveys, citizens, and the making of a mass public. Cambridge, MA: Harvard University Press.

41.

Jaquette

Salazar

(2018, April 13). Opinion: Colleges recruit at richer, whiter high schools. The New York Times. Retrieved from https://www.nytimes.com/interactive/2018/04/13/opinion/college-recruitment-rich-white.html

42.

Julian

(1997). The utilization of the logic model as a system level planning and evaluation device. Evaluation and Program Planning, 20, 251-257.

43.

Kay

Baker

(2015). What can causal process tracing offer to policy studies? A review of the literature. Policy Studies Journal, 43, 1-21.

44.

Kimball

(2011). College admission in a contested marketplace: The 20th century and a new logic for access. Journal of College Admission, 207, 20-30.

45.

Kimball

(2016). Reconciling the knowledge of scholars and practitioners: An extended case analysis of the role of theory in student affairs. Critical Questions in Education, 7, 269-288.

46.

Kimball

Friedensen

(2018). Professions and the post-truth era: A historiographic analysis of the role of expert judgment in the destabilization of objectivity. Education Policy Analysis Archives, 46. Retrieved from https://epaa.asu.edu/ojs/article/view/3357

47.

Lagemann

E. C.

(2002). An elusive science: The troubling history of education research. Chicago, IL: University of Chicago.

48.

Luker

(2008). Salsa dancing into the social sciences: Research in the age of info-glut. Cambridge, MA: Harvard University Press.

49.

Maxwell

(2004). Causal explanation, qualitative research, and scientific inquiry in education. Educational Researcher, 33(2), 3-11.

50.

Mertens

D. M.

(2014). Research and evaluation in education and psychology: Integrating diversity with quantitative, qualitative, and mixed methods. Thousand Oaks, CA: Sage.

51.

Oakes

(2005). Keeping track: How schools structure inequality (2nd ed.). New Haven, CT: Yale University Press.

52.

Page

Scott-Clayton

(2016). Improving college access in the United States: Barriers and policy responses. Economics of Education Review, 51, 4-22.

53.

Pasque

P. A.

Lechuga

V. M.

(Eds.). (2017). Qualitative inquiry in higher education organization and policy research. New York, NY: Routledge.

54.

Patton

M. Q.

(2015). Qualitative research and evaluation methods: Integrating theory and practice. Thousand Oaks, CA: Sage.

55.

Paulsen

M. B.

St. John

E. P.

(2002). Social class and college costs: Examining the financial nexus between college choice and persistence. Journal of Higher Education, 73, 189-236.

56.

Pawson

Wong

Owen

(2011). Known knowns, known unknowns, unknown unknowns: The predicament of evidence-based policy. American Journal of Evaluation, 32, 518-546.

57.

Perez

P. A.

McDonough

P. M.

(2008). Understanding Latina and Latino college choice: A social capital and chain migration analysis. Journal of Hispanic Higher Education, 7, 249-265.

58.

Perna

L. W.

(2010). Toward a more complete understanding of the role of financial aid in promoting college enrollment: The importance of context. In Paulsen

(Ed.), Higher education: Handbook of theory and research (pp. 129-179). New York, NY: Springer.

59.

Porter

T. M.

(1995). Trust in numbers: The pursuit of objectivity in science and public life. Princeton, NJ: Princeton University Press.

60.

Radwin

Conzelmann

J. G.

Nunnery

Lacy

T. A.

Lew

. . . Siegel

(2018). 2015–16 National Postsecondary Student Aid Study (NPSAS:16): Student financial aid estimates for 2015–16 (NCES 2018-466). Washington, DC: National Center for Education Statistics.

61.

Riesman

D. R.

(1980). On higher education: The academic enterprise in an era of rising student consumerism (2nd ed.). New Brunswick, NJ: Transaction.

62.

Rippner

J. A.

(2015). The American education policy landscape. New York, NY: Routledge.

63.

Rodriguez

Nuñez

(2015). Constructing college “choice” for Latino students: The organizational culture of an urban catholic high school. In Ceja

Perez

(Eds.), Higher education access and choice for Latino students (pp. 38-49). New York, NY: Routledge.

64.

Sanderson

(2002). Evaluation, policy learning and evidence-based policy making. Public Administration, 80(1), 1-22.

65.

Sellar

(2015). A feel for numbers: Affect, data and education policy. Critical Studies in Education, 56, 131-146.

66.

Shapiro

Dundar

Huie

Wakhungu

Yuan

Nathan

Hwang

Y. A.

(2017). Completing college: A national view of student attainment rates by race and ethnicity—Fall 2010 cohort (Signature Report No. 12b). Herndon, VA: National Student Clearinghouse Research Center.

67.

Stone

D. A.

(2012). Policy paradox: The art of political decision making (3rd ed.). New York, NY: W. W. Norton.

68.

Tavory

Timmermans

(2009). Two cases of ethnography: Grounded theory and the extended case method. Ethnography, 10, 243-263.

69.

Van Velsen

. (1979). The extended case method and situational analysis. In Epstein

A. L.

(Ed.), The craft of social anthropology (pp. 129-149). Oxford, England: Pergamon Press.

70.

Welton

A. D.

Martinez

M. A.

(2014). Coloring the college pathway: A more culturally responsive approach to college readiness and access for students of color in secondary schools. Urban Review, 46, 197-223.

71.

Yanow

(1999). Conducting interpretive policy analysis (Vol. 47). Thousand Oaks, CA: Sage.

72.

Yin

R. K.

(2017). Case study research and applications: Design and methods. Thousand Oaks, CA: Sage.

73.

Ziskin

Fischer

M. A.

Torres

Pellicciotti

Player-Sanders

(2014). Working students’ perceptions of paying for college: Understanding the connections between financial aid and work. Review of Higher Education, 37, 429-467.