Exploring Influences of Policy Collisions on Transfer Student Access: Perspectives From Street-Level Bureaucrats

Articulation Agreements

Although articulation agreements between institutions have existed since the early 1900s (O’Meara et al., 2007), state involvement has grown substantially during the past 30 years to streamline transfer of coursework processes through the implementation of statewide agreements (Kisker et al., 2011; LaSota & Zumeta, 2016; Montague, 2012; O’Meara et al., 2007; Zinser & Hanssen, 2006). Research on articulation policy has focused primarily on their design, implementation, intended outcomes, and efficacy (Montague, 2012; O’Meara et al., 2007). The majority of articulation policies and policy elements aim not to increase prevalence of transfer but to preserve credits for students who do transfer (Roksa & Keith, 2008). Articulation policies range from tightly coupled 2 + 2 arrangements (system-wide), which “specify all lower division courses for nearly all majors so that course credits are applied to programs of study consistently across the system” (p. 338), to loosely coupled arrangements between institutions for specific programs (institution-driven Hodara et al., 2017). Commonly included policy elements within articulation agreements include (a) common general education package, (b) common lower-division pre-major and early-major pathways, (c) credit applicability, (d) students granted junior status upon transfer, (e) guaranteed and/or priority university admission, (f) associate/bachelor’s degree credit limits, and (g) acceptance policy for upper-division courses (Kisker et al., 2011).

Research on the efficacy of articulation policies has found mixed results, with some studies indicating transfer students perform on par with first-time-in-college (FTIC) students in terms of grade point average (GPA; Garcia Falconetti, 2009), have better graduation outcomes if they transfer through an articulation agreement (Kopko & Crosta, 2016), and are more likely to transfer vertically if they live in a state with a blanket articulation policy across all public institutions (LaSota & Zumeta, 2016). Other research has shown that articulation agreements had no effect on increasing the prevalence of transfer (Roksa & Keith, 2008) or successful completion of bachelor’s degrees after transfer (Garcia Falconetti, 2009; Huffman, 2012). There is some evidence that system-wide policies provide the clearest transfer pathways to credit equivalency and timely degree progress, whereas institution-level arrangements are cumbersome, lack clarity, and are less effective at supporting transfer (Hodara et al., 2017). Considered collectively, the effectiveness of articulation policies seems to depend substantially on state, institutional, and policy contexts.

There are a number of challenges working against creating effective articulation agreements (Dougherty et al., 2017; Hodara et al., 2017; Montague, 2012; Wang, 2020). For example, students’ uncertainty of major and limited advising capacity, particularly at the community college, negatively influenced policy outcomes (Hodara et al., 2017). This finding suggests that a lack of on-the-ground resources may inhibit the effectiveness of articulation policy, regardless of its purpose and design. The multipurpose nature of community college courses (e.g., transfer prerequisite, core requirement for a terminal degree, certificate course) can pose challenges for constructing and assessing transferability (Cohen et al., 2014; Montague, 2012). Broad-stroke articulation policies can also fall short because of tensions or prejudices between 2-year and 4-year faculty (Bailey et al., 2015; Bailey & Morest, 2006; Dougherty et al., 2017; Montague, 2012; Wang, 2020). Articulation policy may fall short of bolstering transfer student outcomes because it may become derailed by other institutional actors, structures, and policies that exert influence during implementation. The nature of this derailment needs to be better understood.

Accordingly, we notice a few notable gaps in articulation policy literature. First, we know little about how articulation policies are interpreted and implemented. Second, without a focused street-level perspective within a specific policy implementation context, we do not know how 4-year institutions balance implementation of articulation policy alongside other institutional interests that may act as a conflict, such as EM. The addition of a focused approach within a single context is needed to complement prior large-scale studies of articulation agreements and to provide unique, more nuanced understanding of policy implementation.

EM

An emerging finding in our exploration of articulation agreements is their interaction with EM policies. The emergence of a competitive academic marketplace had led to institutions adopting strategic EM practices and policies to maintain financial stability and academic rank. Snowden (2010) finds that EM practices and policies have become institutionalized and professionalized within 4-year institutional structures and are powerful influencers on organizational decision making and spending priorities. The main goal of EM is to develop a unique identity and brand through student enrollment and marketing (Hossler & Bontrager, 2014), and EM practices exist across many parts of an institution, including recruitment, admissions, financial aid, selective enrollment, marketing, and branding (Natale & Doran, 2012). Humphrey (2006) found that EM is not just critical for maintaining financial stability, but it also bolsters institutions’ ability to grow endowments, enhance academic reputation, and increase in academic rank.

The most prevalent EM efforts discussed in the literature include tuition discounting, spending on recruitment and marketing, and selective enrollment (Hossler & Bontrager, 2014). Tuition discounting involves a shift to merit-based aid practices whereby institutions offer attractive aid packages to the students with the most competitive admission application package (Hossler, 2004; Natale & Doran, 2012). Institutions may also increase spending on recruitment of highly qualified students, marketing, and branding (Natale & Doran, 2012). The impacts of increased spending on EM have negative implications for equitable access for underserved students and have been explored by several scholars (Hossler & Bontrager, 2014; Natale & Doran, 2012). Finally, institutions can increase prestige and academic social rank through increased selectivity of enrollment, both into the institution and into specific academic programs (Frank, 2004; Hossler, 2004; Natale & Doran, 2012). Some elite universities have low acceptance rates institution-wide, whereas other 4-year institutions instead only have certain academic programs or colleges that are highly ranked or wield prestige in the professional marketplace for their graduates. In this latter case, colleges within a university may use EM policies to restrict enrollment flow into specific degree programs. This scenario is the case in our study, as the College of Engineering (COE) utilizes an EM policy to restrict enrollments into degree-granting engineering majors based on academic performance.

Although EM supports institutions’ efforts to enhance their brand and academic identity, other scholars have highlighted the subsequent detriment to college access (Hossler, 2004). Tuition discounting, for example, involves a shift away from need-based aid packaging to merit-based aid funding, thereby supporting students who already have the means to afford college (Hossler, 2004). Investments in recruiting high-scoring students siphon money from efforts to increase diversity of enrollments on campus (Natale & Doran, 2012). And increasing selectivity of academic programs, much like systems to distribute merit-based aid, typically rely on students’ performance on standardized tests, course performance, and completion of certain courses to determine eligibility. Underserved, low-income, and first-generation students are more likely to underperform on these entry metrics compared with high- and middle-income White students, thereby perpetuating inequities in the postsecondary education system (Carnevale & Strohl, 2013). This prior research has focused on best practices for marketing, recruiting, and increasing yield of transfer students. There is a dearth of literature considering the influence of EM on college access through transfer pathways; prior research has focused on best practices for marketing, recruiting, and increasing yield of transfer students (e.g., Hope, 2016, 2018; Sutton, 2018). We know very little about how community college transfer students are affected by EM, a notable contribution of this study.

In thinking across seemingly contrasting policies, an institution will be forced to decide how to balance investments between EM practices and its efforts for supporting equitable access through policies like articulation agreements. Because street-level bureaucrats serve on the front lines as policy implementers, they play a critical role in representing the institution to students in how it chooses to balance competing needs and policy interests. In cases where policies intersect or overlap, as is the case in our study with articulation and EM policies, frontline faculty and advisors are forced to make on-the-ground decisions about how to apply each policy.

Policy Implementation and Street-Level Bureaucracy

Individuals involved in the first three phases of policymaking (i.e., agenda setting, policy formulation, and policy adoption) rarely are involved directly in implementation, a job often carried out by administrators on the ground (Dunn, 2015, p. 43). We consider this fourth phase of the policymaking process in our study, particularly with respect to articulation policies coming into conflict with EM policies at the implementation phase. To describe the role that frontline administrators have in carrying out policies and procedures, Lipsky (1969) coined the term “street-level bureaucrat,” which is the theoretical lens that grounds our analysis. Prottas (1978) further explored the unique positioning of street-level bureaucrats as boundary actors between an organization and its clients, suggesting that they accrue power and control as the only member with simultaneous access to information of the organization and client. Hudson (1989) suggested that street-level bureaucracy has been underutilized as a theory to explain policy implementation and how street-level bureaucrats make decisions during implementation.

In one of a few examples, Loyens and Maesschalck (2010) linked the theory of street-level bureaucracy with ethical decision making. They found that street-level bureaucrats often face role strain during policy implementation, where they must balance their role as employee of the organization and also act in support of the client. Despite developing routines for decision making based on client characteristics, Loyens and Maesschalck (2010) found that street-level bureaucrats were often left with inescapable dilemmas of competing needs that were impossible to reconcile and, as a result, were coerced to act with discretion.

In examining the behaviors of street-level bureaucrats, several authors have found that the implemented policy often deviates from prescribed or intended policy, a phenomenon coined as street-level divergence (Brodkin, 2003; Lipsky, 1980; Maynard-Moody & Musheno, 2000). For example, caseworkers in Chicago welfare offices distorted implementation of a federal caseworker policy by adjusting the content of casework meetings. Brodkin (2003) describes how doing so allowed caseworkers to meet federal quotas to continue receiving funding. Although perverse external accountability demands are one cause for street-level divergence, it can also result from policy ambiguity, changes in organizational practice, or decisions about implementation being complex and impacting multiple stakeholders (Brodkin, 2003; Gofen, 2013; Hill, 2005; Lipsky, 1980). The complexity of meeting divergent stakeholder demands plays a significant role in the emergence of street-level divergence during policy implementation.

Gofen (2013) expanded prior work on street-level divergence by exploring how it occurs and in what forms. He found that, in examining policy implementation within the contexts of teachers, nurses, and social workers, divergence could be categorized in three ways: (a) transparent versus concealed divergence, (b) self-serving versus other-serving divergence, and (c) individual versus collective divergence. In the first category, street-level bureaucrats either chose to explicitly label their divergence from stated policy or to forgo disclosing their adjustments. In the second category, street-level bureaucrats either adjusted policy in their self-interest to make it easier to operate in support of competing stakeholder needs or adjusted policy to best suit the needs of their clients. Finally, Gofen (2013) found that divergence either could occur on the individual level with certain street-level bureaucrats shifting policy at implementation or could occur across a collection of colleagues working to enact adjustments together. These findings provide a framework to examine how street-level divergence occurs within organizations and how that divergence influences policy implementation.

In our study, we extend prior research on policy implementation by examining how street-level bureaucrats manage implementation of two converging policies. Prior research points to the complexities street-level bureaucrats encounter in trying to meet the needs of multiple competing stakeholders. However, there is little research on the influence of managing multiple policies that are dissonant in purpose but impact the same clients, a notable contribution of this study. Another contribution of this study is that, to our knowledge, street-level bureaucracy and divergence have not been examined within the postsecondary education context. We seek to understand how street-level bureaucracy can be used to explore actions of academic faculty and advisors who implement policies on the ground for postsecondary institutions.

Data Collection and Participants

We engaged faculty and advisors in semi-structured interviews to explore their experiences managing implementation of colliding policies in engineering. We complement interview data by analyzing documents that detail the articulation agreement and EM policies and analyzing university course-taking data of transfer students.

Data Analysis and Quality

Data analysis was conducted in three phases and was largely inductive. First, we used thematic analysis (Corbin et al., 2014) of transcribed interviews to explore how faculty and advisors implement colliding policies and how that management process influences the transfer of coursework process. The first round of thematic analysis involved descriptive coding (Saldaña, 2015) by a member of the research team using NVivo coding software. To validate the quality of the initial codes generated through thematic analysis, a second research team member selected a random subset of transcribed interviews and generated descriptive codes, which were then used to refine the initial codes generated by the first research team member. The research team then engaged in peer auditing, looking across themes to identify codes relevant to the research questions addressed in this study.

Three descriptive codes were most salient and were selected for a secondary round of focused coding: Articulation Agreements, EM Policy, and general discussion of Other Policy. Seeking a more nuanced understanding of these codes, the research team engaged in a second cycle of coding which included a blend of focused and evaluation coding (Saldaña, 2015). The combination of focused and evaluation coding enabled us to capture a more detailed understanding of participants’ on-the-ground experiences with each policy, including their perceptions of the effectiveness and impacts of the policies on the transfer process. The theoretical lens of street-level bureaucracy played a role in this secondary coding process, which remained inductive in nature, but helped us to narrow our focus on understanding how these policy actors engaged in policy implementation for both policies. Table 1 provides a summary of frequency counts of the occurrence of the focused codes with a count of sources (i.e., participants) for each code. Finally, the research team consolidated findings from descriptive, focused, and evaluative codes and identified themes related to each policy and their intersection.

OPEN IN VIEWER

Table 1

Frequency Counts of Focused Code Mentions and Sources

Descriptive code	Focused code	No. of sources	No. of mentions
Articulation Agreement	AlignmentofCourses_−	21	153
	Partnerships	17	92
	Perception_+	20	82
	AlignmentofCourses_+	20	79
	RigorofCourses_−	18	68
	GeneralInformation	18	52
	CurriculumCommitteeProcess	16	47
	Perception_−	15	47
	ASDegree_TechnicalCourseOverload_−	14	34
	StudentParticipationLevels	9	30
	RigorofCourses_+	10	24
	NonVCCSTransfer	4	13
	ConfusingLegalese	2	6
	OnlineCourses	2	6
	PoorTimingforAdmissions	2	4
	Remedial_Nontransferable	1	3
	NoAgreementAvailable_OutofState	2	2
Enrollment Management Policy	Advising	16	124
	ImplicationsforStudents_−	19	109
	Process	10	71
	Justification	10	42
	Requirements	11	39
	ImplicationsforStudents_+	13	34
Other Policy	ClemencyRules_−	4	9
	ExceedCreditLimit	3	8
	FallAdmissionsOnly_−	1	3
	CourseWithdrawals	1	2
	DeansOfficeImplementation	1	2

Note. A frequency count of focused evaluation codes including the number of times a code was mentioned by participants (no. of mentions) and the number of participants who mentioned the code (no. of sources). The +/− denotes participants’ evaluation associated with the code. For example, Alignment of Courses − was used when participants felt courses were poorly aligned, whereas Alignment of Courses + was used when participants perceived courses to be aligned well.

To illustrate the analytic process from initial codes to identifying themes, we provide the following example. In the initial phase of coding, we parceled out participants’ discussion related to implementation of the articulation agreement and of the EM policy. Then, knowing that street-level bureaucrats grapple with balancing the needs of the organization and its clients, we used focused and evaluation coding and found that participants described a series of negative implications for students that resulted from the implementation of the EM policy (e.g., coded Implications for Students −). Next, during the theme generation process, we discovered that many of those negative implications also linked to focused and evaluation codes from the implementation of the articulation agreement such as students becoming overloaded with technical courses in their first semester post-transfer (e.g., ASDegree_TechnicalCourseOverload). This largely inductive process, informed by street-level bureaucracy, led us to the discovery of policy collision, the prominent finding of this study.

To triangulate descriptions from the participant interviews, we generated a protocol to conduct a document analysis of the articulation agreement and EM policy. The protocol involved a systematic and iterative review of each policy examining the following elements: (a) purpose, (b) requirements, and (c) benefits and/or intended outcomes. Finally, we conducted a descriptive analysis of entering transfer students in engineering including student demographics. We also examined the types of courses in which transfer students enrolled during their first two semesters post-transfer. All courses taken by a transfer student in their first two semesters post-transfer were collated into three groups: (a) elective courses, (b) foundational preengineering non-major courses, and (c) discipline in-major courses. We then compared percentages of enrollment by transfer students in each group during their first two semesters post-transfer, which helped us to triangulate the impacts of colliding policies on transfer students’ course enrollments. We also examined transfer students’ performance (i.e., GPA) in courses during their first semester post-transfer to understand the scope of students impacted by the collision of policies.

Managing Policies as They Collide

Our analysis of interviews with street-level bureaucrats identified multiple themes. First, faculty and advisors modify parts of policies so that they work in better alignment during implementation. Next, we discuss rationalization processes that emerge as advisors, faculty, and administrators grapple with dissonance caused by the policy collision. Finally, we detail how the collision of policies negates some of the benefits included in the design of the articulation policy.

Impacts of Colliding Policies on Students

To address our second research question, we sought to understand how colliding policies influenced the transfer of coursework process. An important theme that emerged in conversations with 14 of the 26 (54%) study participants was the unintended consequences of technical course overload for transfer students in their first semester after transfer. By transferring to the university through the articulation agreement, one of the primary benefits for students is satisfying the general education course requirement at the university. When examining the articulation agreement in isolation, this arrangement seems to be a desirable and enticing benefit for transfer students and guarantees the transfer of a portion of credits required for completion of a bachelor’s degree. However, faculty and advisors discussed how this benefit can actually end up harming students, particularly in their first semester after transfer, as students’ schedules are overloaded with technical, upper-level major courses. An advisor details how an engineering program’s curriculum (i.e., check sheet) is designed to help students balance difficult technical courses with nonprogram elective courses each semester and the implications for transfer students who lose access to the balanced curriculum by design:

I worry a little bit about the burnout . . . the reason we have the check sheet the way it is is because it’s very balanced . . . you have the [Gen Ed Required Courses] littered throughout. . . . You have a balance of [program] classes and non-[program] classes. . . . But if you’re coming in and you’ve eliminated everything but the [program] classes that first year when you just start, it can be just kind of a killer. Just a motivational killer sort of thing.

Designed as an incentive for community college students to transfer via an articulation policy, students who have completed the general education course requirement are left with “nothing to take except for really, really hard classes that they’re just not quite ready for.” Table 2 shows the distribution of general education courses and compares courses allotted in the first 2 years of a program versus the latter 2 years; in aggregate, program plans are designed to have more general education courses in the latter 2 years as a way to balance the academic course load.

OPEN IN VIEWER

Table 2

Distribution of General Education Courses in Academic Plans by Discipline

Engineering degree	Gen. Ed. electives (first 2 years)	Gen. Ed. electives (last 2 years)
Aerospace	7	6
Biological Systems	7	9
Chemical	6	10
Civil	6	7
Computer	4	12
Computer Science	9	9
Construction Management	4	3
Electrical	6	10
Engineering Science and Mechanics	6	7
Industrial Systems	6	10
Materials Science	4	9
Mechanical	4	12
Mining	7	9
Ocean	7	6
Average	5.93	8.50

Note. A comparison, across engineering disciplines, of the placement of general education courses in an academic plan. We compare the number of prescribed general education courses sequenced during students’ first 2 years (i.e., freshman and sophomore year) with those recommended during students’ latter 2 years (i.e., junior and senior year).

On average across disciplines, academic plans include three credits more of general education in the latter 2 years than the first 2 years. This design helps students balance their workload in the junior and senior years as they contend with challenging upper-level major courses as well as a senior design project. Thus, although the intent of the articulation policy general education requirement waiver is to benefit transfer students, it conflicts with how classes are sequenced in academic plans across engineering disciplines for FTIC students.

The check sheets do not account for courses that transfer students actually take after transferring to the university. Consequently, we examined the classes in which transfer students actually enrolled their first semester after transfer and compared them with courses taken by FTIC students in their fifth semester, the semester most closely aligned with vertical transfer students who transfer via the articulation agreement. Transfer students enrolled in more major courses and fewer elective courses than FTIC students (Table 3).

OPEN IN VIEWER

Table 3

Comparison of First Semester Transfer and First Year FTIC Enrollments by Course Type

Course subject type	% of first semester transfer enrollments	% of first year FTIC enrollments
All electives	11.30	15.10
All STEM electives	2.60	2.20
All major	86.10	82.70
Total course enrollments	10,939	49,889

Note. A comparison of the types of courses transfer students enrolled in during their first semester post-transfer with courses FTIC students enrolled in during their first year. This time frame represents the courses students take in attempt to meet the requirements of the EM policy. FTIC = first-time-in-college; STEM = science, technology, engineering, and math; EM = enrollment management.

Further complicating the issue for transfer students is that in their first semester, along with enrolling in a higher number of difficult technical courses, on average, transfer students also had to meet the GPA (3.0) and credit completion (12) requirements of the EM policy to gain entry to a degree-granting major. Only 63% of transfer students in our data set met the minimum GPA requirement in their first semester for entry into their first choice major, with a median GPA of 3.19. An advisor reflects on frustration expressed by transfer students who feel the classes they are taking in their first semester after transfer are harder when compared with FTIC students who must meet the requirement of the EM policy after their first year of courses at the university:

[Transfer students] say things like they don’t get any “frou-frou” classes, or whatever, which I get. That’s absolutely a challenge that I see. And we try and acknowledge that without adding fuel to their fire. We try and acknowledge that at orientation by emphasizing the importance of having a balanced schedule. So you don’t need to take 18 credits of junior level whatever engineering major. Let’s balance out other degree requirements, that might help.

This expressed concern of transfer students may be warranted, as the average weighted GPA of FTIC students in their first year courses is 3.12, which includes some general education courses. However, when examining the academic performance of FTIC students in the same set of courses that transfer students take during their first semester, transfer students (3.01 average) actually outperform FTIC students (2.97 average) in those same courses. Thus, institutional data support the perception that the EM policy implicates FTIC and transfer students in different ways.

Knowing students must complete 12 credits at a 3.0 GPA to enter into a degree-granting program, another advisor modified their recommended courses for students, particularly in the first semester after transfer, to help students successfully transition to the university. The advisor describes that although students could take a set of courses to advance more quickly toward degree completion after transfer and not enroll in general education courses, it may not be a best practice for their success:

Sometimes, we also have to be careful about the number of discipline-specific courses that we’re recommending to transfers because we also don’t want to overwhelm them even though they can do something; it’s a matter of should they be doing something. But I also think we have to be careful about the types of courses that we’re putting our transfer students in right out of the gate. I think that we do need to take those things into consideration as advisors in our college and not throw them in all 4000 level classes in a specific discipline and allow them that semester of transition.

Another advisor shares a contrasting perspective and suggests that students advised to take courses to meet the requirements of the EM policy “get penalized for that because they’re not being advised to take classes that count towards their progression. It’s going to take them longer and longer to graduate.”

Although there is variation in how faculty and advisors choose to manage students’ technical course loads in their first semester, the impact of colliding policies is evident. One advisor expresses pessimism about the design of the articulation agreement for transfer students and expresses concern over how its current design hinders students’ ability to be successful in coursework after they transfer:

That’s where it comes into a question of the associate’s degree is great to have on their record, and I know that the community colleges want it for graduation statistics, but it actually in a lot of times hurts the students if they come in with a full associate’s, because there’s just not anything here for them to take, whereas if you only do one year there, we can actually insert some classes into your first year here.

It seems that, particularly within this engineering context, the potential consequences of granting students’ fulfillment of general education coursework as part of an articulation agreement in combination with EM policy requirements may place students in challenging situations.

Implications for Street-Level Bureaucrats

An important implication from this study for advisors, administrators, and faculty, serving as street-level bureaucrats, is how they reconcile and rationalize implementation of disparate policies that collide for transfer students. Organizational pressures can lead street-level bureaucrats to adopt reductive and pathological moral dispositions, including indifference, enforcement, or caregiving when acting with discretion in policy implementation (Zacka, 2017). We found that participants generated varying justifications for the existence of policies, such as rigor and equal application of policies, that are then communicated to students. We suggest that it is important for street-level bureaucrats to understand the roots of why they generate these justifications. For example, the reliance on rigor to justify the tension between these policies may be emblematic of a larger power imbalance between the university and its partner community colleges, the latter of whom are supposed to be partners in adoption of the articulation agreement, but may not actually be considered equals by the former. Transfer partnerships have little chance for success if 4-year institutions and the institutional actors enacting them do not trust the education provided by community colleges.

Advisors, faculty, and administrators also should consider how these justifications for policy collision may be interpreted when they are communicated to marginalized students. Street-level bureaucrats may be the first institutional agents—practitioners at 4-year institutions in a position of authority (Dowd et al., 2013)—to engage with transfer students and, as a result, have significant influence on those students’ first experiences post-transfer. Institutional agents play a critical role in affirming transfer students’ transfer aspirations and collegiate identity, particularly for marginalized students (Dowd et al., 2013). Relying on academic rigor to justify the existence of colliding policies that negatively impact transfer students may fall short in affirming those students’ aspirations and collegiate identities and instead perpetuate students’ experiences of transfer stigma and self-doubt as they transition to the university.

Another finding pertinent to institutional actors serving as street-level bureaucrats is street-level divergence or the adjustment of policies to fit different contexts or needs. As street-level bureaucrats at universities grapple with colliding policies that affect students differently, how should they decide when to adjust policies to better fit the needs of a subgroup of students? In our case, faculty and advisors chose to engage in concealed divergence (Gofen, 2013), informing only transfer students about their exemption from upper-level coursework registration restrictions of the EM policy. Simultaneously, universal application of the policy to all students emerged as a justification for including transfer students in the policy. As universities balance EM practices and articulation agreements with community colleges, they will often face these kinds of predicaments when choosing who will be affected by policies and how exemptions or exceptions will be made and communicated to stakeholders. Street-level bureaucrats should be mindful of how, why, and in what instances they engage in street-level divergence and understand how those decisions impact students.

Implications for Transfer Students and Policymakers

Among the most prominent implications of this study is that transfer students are caught between a policy that expands access and another that constrains their progression. Policy collisions may influence subpopulations differently, especially when a policy designed for a different target population negatively influences a particularly vulnerable subpopulation. Concerted efforts to improve access through community college pathways may be largely undermined by other policy efforts like managing enrollment through selective admission policies. Transfer students who rely on the articulation agreement for transfer may also perceive the EM policy as another unnecessary barrier as they progress toward their degree and may be dissuaded from persisting. Policymakers who seek to establish equity-minded policies (Bensimon et al., 2016; Wang, 2020) and who are authentically interested in improving vertical transfer should consider ways to design articulation and EM policies that work together instead of relying upon street-level bureaucrats to make difficult decisions about which parts of policy to prioritize over others when there is conflict. This consideration is particularly important in the context of vertical transfer pathways, where students are already marginalized by the higher education system structure (Cohen et al., 2014).

University policymakers should also consider the explicit versus implicit nature of their policies. In our case, because the purposes of the EM policy were not stated explicitly, policy implementers were left to generate their own justifications for why transfer students were subject to the EM policy, particularly as students experienced dissonance with the tenants of the articulation agreement. Having multiple justifications for policies across street-level bureaucrats can create ambiguity among transfer students as they work with peers to transition to their new institutional environment.

Another implication of unwritten policies is that, for transfer students and the community college faculty and advisors supporting them, the planning process for coursework begins well before transfer. We found that community college faculty and staff were aware of the EM policy rule and communicated its existence to students in advance of transfer. However, because the policy, and particularly the exemption, was not explicit and public knowledge, students and community college faculty may not be aware of workarounds to coursework issues that emerge from colliding policies. Thus, students with lower amounts of transfer student capital, or the knowledge, skills, and abilities needed to navigate the transfer process (Laanan, 2004; Laanan et al., 2010; Santos Laanan, 2007), may be at a severe disadvantage. If a student is considering transferring and does not know to ask someone at the university about possible workarounds to the policy, or if that student does not have community college faculty advisors with insider knowledge or connections to university faculty and staff, that student might make a different decision about the transfer pathway. This scenario illustrates an important unintended consequence of having exceptions to policies that are not stated explicitly and publicly that yield inequities between students.

Implications for Research

Prior scholarship on articulation agreements has focused too narrowly on the contents of those policies without considering broader contexts. Findings from our study suggest that we should not examine policy solutions in isolation but from the perspective of their convergence with other policies. Understanding how other university policies, such as the EM policy in this study, collide with and influence the efficacy of an articulation agreement is often not considered. Policy interactions may help explain why articulation agreements may not be effective at increasing transfer or uncover how a component of the policy intended to benefit the student may actually have the opposite effect. This finding is particularly relevant for competitive disciplines that have additional policies in place “behind the scenes” to control access. Although examining policy collisions is applicable across contexts, it is particularly important in the context of this study as universities grapple with national movements to remain competitive in a global academic marketplace while simultaneously diversifying and making campuses more inclusive. In addition, there is a tremendous need for scholarship that examines how EM policies and practices at 4-year institutions impact transfer students. There has been little research examining this issue, and findings from our study point to its importance in bolstering the success of transfer pathways.

Findings from this study also have important implications for future research and policy analysis that examines policy implementation. For example, future studies examining the efficacy of articulation agreements may need to be more granular and involve practitioners who are acting as street-level bureaucrats. In our case study, we only discovered the policy collision by engaging advisors, administrators, and faculty who implement both the articulation policy and the college’s EM policy—previous work at the institutional or state level may not capture this type of interaction that articulation policies may have with other institutional policies. In addition, there is a general lack of focus on the implementation stage of policy creation and adoption in higher education, and this study highlights the value of close examination of this stage of the policy process. Future research on policy implementation can uncover organizational conditions and actors that may influence how a policy is realized during implementation. Finally, future research should incorporate students’ perspectives to understand the perceived impact of policies on students’ navigation through vertical transfer. Although we believe the faculty and advisors in our study have a deep understanding of the system through their close interactions with students, bringing the direct student perspective into this inquiry is an important next step.

Finally, as this is a case study, we recognize the need to bound the case within the institutional and disciplinary contexts of this study. The prevalence of policy collisions, particularly those where articulation policies designed to increase access intersect with policies constructed to preserve institutional and program prestige through selective enrollments, may be most salient in institutional contexts similar to the university in this study. Effectively aligning articulation policy with other university and college policies at a lower transfer-in institution may be more challenging than at medium or high transfer-in institutions. In addition, the engineering disciplinary context of this study, with its highly sequential curriculum and competitive admissions, seems more susceptible to policy collision compared with less sequential disciplines such as humanities or social sciences or in disciplines with less competitive admission standards. In addition, the collision of articulation and EM policies as described in this study is limited only to institutions that do not allow direct entry into disciplines. Institutions that admit students directly into degree-granting majors will not encounter this specific type of policy collision. Further research is needed to understand the prevalence of EM admission policies for transfer students that exist nationally. Doing so will be a valuable direction for future research to understand the extent of policy collision between EM policies and articulation agreements.