An Examination of the Bidirectional Relationship Between Federal Civil Rights Guidance and Research

As the primary agency responsible for enforcing federal civil rights laws in the educational context, the U.S. Department of Education Office for Civil Rights (OCR) issues policy guidance to help ensure that school districts and universities understand their legal obligations. These policy guidance documents have direct connections to topics studied by educational researchers (e.g., disproportionality in special education, race conscious admissions policies in higher education, transgender student inclusion, etc.). However, we do not have an empirical understanding of how this guidance is used by the research community. Nor do we have a strong grasp on the extent to which this guidance is explicitly informed by research. It is important to acquire an empirical understanding of the bidirectional relationship between research and educational policy in the context of civil rights enforcement in order to determine areas of strength and those in need of improvement. Specifically, this study poses the following research questions: 1) How often and in what ways do scholars explicitly use OCR policy guidance to inform their research? What are the characteristics of this research? 2) Conversely, how often and in what ways does OCR explicitly use research to inform policy guidance? What are the characteristics of research cited in OCR guidance?


Conceptualizing the Bidirectional Relationship
Similar to researchers who have identified ways to promote research use by practitioners (e.g., Schneider, 2015, noting the importance of "visibility," "acceptability," "feasibility," and "transportability"), researchers have sought to conceptualize the ways with which policymakers rely upon research. According to the literature, policymakers rely upon research for different purposes, including "instrumental," "political," "conceptual," "imposed," or through the research process, itself (Tseng, 2012, p. 7). Weiss (1979) identified six models of research use in policy. The "knowledge-driven model" views the relationship between research and policy as a "linear view that research findings may be communicated to impel action" (Nutley, Walter, & Davies, 2007, p. 285).
The "problem-solving model" is a "second linear model that begins with the end users of research and the problems they face, before tracking back in search of useful findings" (p. 285). The interactive model is "modelled as a set of non-linear less predictable interactions between researchers and users, with research impact happening through complex social processes of 'sustained activity'" (p. 285). The political model views research as a tool to fuel "adversarial systems" (p. 285). The tactical model relies upon research for strategic reasons, to justify actions in the face of pressure or to support inaction. The enlightenment model is much more nuanced, viewing impact as more of a progression (p. 285). In recognition of these various models, this study acknowledges the complexities of the intersection of education research and policy, particularly civil rights guidance.
As not only knowledge producers but also consumers, constituents, stakeholders, and experts in the field of education, researchers are uniquely situated to engage in the broader policy discourse. While there may not be a designated body of research that examines "policy use in research" per se, we can point to various words or phrases, such as policy-oriented research or policy-informed research, or references to "scholars as policy actors" (Welner, 2012, p.7). Moreover, as reviewers, we might ask authors to consider the broader policy context in which their work is situated (Lewis, Burke, & Decker, forthcoming) which may have direct ties to the significance, implications, or timeliness of a particular study. Policy contextualization and knowledge of the law, in particular, can help researchers better understand the institutional factors that may inform research questions, design, analysis, and implications.
While less is known about policy use in research, at least from an empirical perspective, according to Levin (2004), "efforts to strengthen the impact and value of research in education need to pay attention to all three contexts -production, use, and the linkages between them" (p. 13). Similarly, within the context of law, scholars have called for a "synergistic relationship" (Superfine, Goldman, & Richard, 2019, p. 544) between law and social science research and researchers have acknowledged the "wide gaps in understanding between the legal and research communities" (Marin, Yun, Garces, & Horn, 2019, p. 2). Grounded in this literature, a bidirectional framing of the relationship between research and policy within the civil rights context allows us to investigate the inherent interdisciplinarity of education policy work, from both the "research use" side of the relationship and the "policy use" side of the relationship.

Literature Review
This study is situated at the intersection of two areas of inquiry: (1) educational inequities that implicate civil rights and (2) the relationship between research and education policy. Together, these two broad areas of inquiry form the foundation for a study on the relationship between research and OCR guidance.

Educational Inequality
In regards to the first area of research, consistent with OCR's enforcement activities, scholars have sought to identify and address inequities related to important issues such as student discipline (e.g., Gregory, Skiba, Noguera, 2010), diversity in higher education and k-12 schools (e.g., Garces & Jayakumar, 2014), the impact of policies and practices related to transgender students (e.g., Lewis & Eckes, 2019), the use of seclusion and restraint (e.g., Gagnon, Mattingly, & Connelly, 2017), sexual assault (e.g, Harper, Maskaly, Kirkner, & Lorenz, 2017), bullying and harassment (e.g., Meyer, 2015), meeting the needs of English Learners (e.g., Jiménez-Castellanos & García, 2017), resource allocation (e.g., Malen, Dayhoff, & Croninger, 2017), and special education identification (e.g., Sullivan & Osher, 2019). This list of topics is by no means exhaustive, nor is the accompanying list of references. On the contrary, each topic is representative of individual bodies of research that touch on the nuances of vast inequities that exist in education research, policy, and practice. For present purposes, we merely aim to demonstrate that a large and burgeoning area of research exists that is germane to OCR's role in enforcing federal civil rights in the educational context.

The Relationship between Research and Policy
Although it seems there is an ever-present appeal for evidence-based practice, the relationship between research and policy is not always clear (Tseng, 2012). Policymakers and researchers are often thought of as two separate communities (Perna, 2016). Scientific research is utilized by policymakers in a number of different ways, most often in an indirect conceptual way, making it difficult to see the connection (Huston, 2008;Tseng, 2012). Nonetheless, researchers have been keenly interested in the relationship between research and public policy for decades (Weiss, 1978).
Research use in policy. According to Tseng (2012), researchers are concerned about research quality and dissemination. Increasing the impact of research requires a concerted effort; findings must be communicated in a manner that is comprehensible to a broader audience that may or may not have expertise in education or research. In traditional scholarly publication venues, researchers' reach is limited by constraints such as accessibility, including the time associated with the publication process and the fact that highly revered research among scholars is often hidden behind paywalls. On the demand side, it is important to consider who utilizes research (such as policymakers, intermediate organizations, etc.), how research is defined (by researchers, policymakers, etc.), how it is obtained, and how it is decoded and translated by its consumers. (Tseng, 2012). Relationships are key to the research-policy connection. The role of relationships is also significant because research might find its way to policymakers through indirect means. For example, organizations may act as "intermediaries" between researchers and policymakers (Lubienski, Scott, & DeBray, 2011).
The research community must "embrace, or at least accept, the unique challenges and inescapably political nature of the knowledge enterprise, especially when engaging with nonresearcher audiences" (Pfleger, Wilson, Welner, & Bibilos, 2018, p. 3). As evidenced by the number of policy guidance documents that have been rescinded since the Trump Administration took office, civil rights enforcement is similarly entangled with politics.
Research use in law. With roots in important cases such as Brown v. Board of Education (1954), connections between social science and law, are not new. Existing research explores the role of research in law, in particular (e.g., Lewis & Bray, 2019;Horn et al., 2018). Less directly, researchers can change public discourse or public opinion related to issues at the heart of legal cases (Welner, 2012). More directly, researchers can influence education law through expert testimony (e.g., Welner & Kupermintz, 2004), amicus briefs (e.g., Horn et al., 2018), and state or federal legislative hearings (e.g., Perna, Orosz, & Kent, 2019). For example, in a recent study, Perna, Orosz, & Kent, (2019) utilized critical discourse analysis to review congressional hearings related to the Higher Education Act. Within the context of amicus briefs, researchers have identified strengths and limitations of the existing relationships between research and case law. For example, educational research examines the role of research in cases related to diversity in higher education (e.g., Horn et al., 2018) and k-12 schools (e.g., Frankenberg & Garces, 2007). Moreover, Lewis and Bray (2019) found that research played a limited role in a recent Supreme Court decision that examined the quality of education to which students with disabilities are entitled under federal special education law. Recognizing disciplinary and professional "divides", Marin, Yun, Garces, and Horn (2019) surveyed "knowledge producers" to get their perspective on the use of research in law. When considering extra-legal resources, respondents indicated that peer-reviewed research should at least moderately influence jurisprudence. Methodologically, survey respondents tended to prefer "methodologically-derived" findings, favoring quantitative methods over qualitative methods (p. 12). These studies provide the foundation for the "research use" side of the relationship between peerreviewed research and OCR guidance.
Policy use in research. While educational research discusses the relationship between research and policy, most of the literature focuses on research-informed policymaking or engaging in research that will inform policy; less is known about policy use in educational research or policyinformed research. Nonetheless, policy informs research in significant ways. For example, in 2017, Educational Administration Quarterly released a special issue on the Every Student Succeeds Act (Fernandez, LeChasseur & Weiner, 2017). Similarly, following affirmative action bans in California, Florida, Texas, and Washington, Garces (2013b) studied the direct impact of these policy decisions and the related implications for diversity in higher education. Zatta and Pullin (2004) discussed the implications of state and federal policy related decisions regarding alternative assessments for students with disabilities (Zatta & Pullin, 2004). As another example most directly related to the present study, Koss, Wilgus and Williamsen (2014), discussed the ways in which restorative justice "could enhance institutional responsiveness and provide options that in some cases may better achieve the underlying goals of Title IX, the DCL, and the field of student conduct management" (p. 243). As these examples illustrate, research is essential to understanding the implications of policymaking. Within the context of civil rights enforcement, policy-informed research has the potential to identify the ways in which policy choices may constrain or enhance the everyday work of educational leaders. Moreover, policy use in educational research may serve as proxy for researchers' policy literacy or the ways in which they see connections between their research and the broader policy context in which their work is situated. A systematic examination of policy use in research seems to be limited, although some researchers have identified the need for research to be informed by practice (e.g., Yohalem & Tseng, 2015).
The literature reviewed in this section is intended to be illustrative. The body of research implicated by this study is vast, thereby supporting the significance of a study situated at the intersection of two important large areas of research inquiry. While research exists regarding equity and diversity and the relationship between research and education policy, we are not aware of any research that examines the relationship between research and OCR policy guidance, in particular.

Policy Context
In contrast to policy more broadly, which is primarily discretionary, OCR operates in accordance with mandates set forth in federal law. Preventing, identifying, and remedying discrimination should be the primary catalyst for OCR action. Given the large body of research germane to civil rights in education, research can play a meaningful, supportive role in the issuance of civil rights guidance .
OCR is responsible for enforcement of numerous federal civil rights laws in the educational context, including the Americans with Disabilities Act, Section 504 of the Rehabilitation Act, the Civil Rights Act of 1964, and Title IX of the Education Amendments of 1972. The inherent vagueness of these laws requires policy interpretation by the enforcing body (Worthington, 2017). Under the Administrative Procedure Act, OCR has the authority to issue "interpretive rules" and "general statements of policy" (5 U.S.C. §553(b)(3)(A)) that result in policy guidance.
The U.S. Office of Management and Budget (OMB) publishes a document outlining best practices for federal agencies distributing policy guidance (U.S. Department of Education, 2019). OCR issues two types of policy guidance applicable to this study. The first type of guidance is a "significant guidance document." This type of guidance covers possible conflicts with actions of other agencies, the impact of modifications to financial programs, and novel legal issues resulting from legal mandates. The second, simply referred to as "guidance," is an "agency statement of general applicability and future effect," that provides the agency's policy or interpretation on a statute, regulation, or technical issue. Between January 20, 2009, and January 19, 2017, OCR released forty guidance packages. Each guidance package contained one or more documents (e.g. dear colleague letters (DCLs), questions and answers (Q&As), and frequently asked question (FAQs)), totaling seventy-seven documents. Illustrative examples of covered topics include: addressing and preventing sexual violence, inclusive policies for transgender students, the use of seclusion and restraint, racial discrimination in student discipline, supporting English Learners, civil rights in juvenile justice facilities, bullying and harassment, and the civil rights obligations of charter schools.
Under the Obama administration, OCR revamped the structure of guidance. By avoiding overly technical language and through the use of examples, OCR sought to issue guidance of practical use to the public U.S. Department of Education, 2012). At the same time, OCR placed a stronger emphasis on research, as evidenced by its release of the most comprehensive set of Civil Rights Data Collection (CRDC) survey data in the history of its collection. According to OCR, the 2009-10 survey covered 85 percent of the nation's K-12 students, while the 2011-12, 2013-14, and 2015-16 surveys were universal collections covering nearly every public school and district in the nation. In the past eight years, OCR has revamped and expanded the CRDC to be more widely accessible to the public (instead of primarily a resource within government)…(U.S. Department of Education, 2016a, p. 2).
Therefore, under the Obama administration, OCR engaged in unprecedented efforts in support of both research and policy clarification. In a recent essay, Lewis, Garces, and Frankenberg (2019) stated that OCR's policy guidance under the Obama administration was responsive to or informed by "(1) developments in the law, (2) findings from OCR's other activities (i.e., complaint investigations, Civil Rights Data Collection, compliance reviews, and technical assistance), and (3) findings from researchers or external experts" (p. 52). Although the authors did not conduct an in-depth analysis of research use in OCR guidance, the article provided examples to illustrate OCR's use of research or external experts, thereby serving as an important inspiration for the research questions posed in the present study.

Research Design and Methods
As part of a larger project on the relationship between research and OCR's enforcement activities, this study involved a thorough review of (1) all guidance issued under the Obama administration from January 20, 2009, to January 19, 2017, and (2) all peer-reviewed research mentioning OCR guidance during the same time period. This timeframe covers the duration of the Obama administration. Based on the "comprehensive" and "practical approach" of the Obama administration (Lewis et al., 2019, p. 52), combined with changes under the Trump administration, this timeframe makes for a clean dataset for political and substantive reasons.
For the purpose of this study, we operationalized the definition of research as follows: articles published in peer-reviewed scholarly journals with relevance to education. This definition allowed us to maintain a uniform basis for comparison of the bidirectional relationship. For our primary analysis, we selected to exclude law review articles given differences in purpose, intended audience, methods, etc. Moreover, since law review articles rely heavily on legal analysis, they are more likely to make connections to OCR's work, thereby making them a less interesting publication venue to analyze in the present study. The decision to exclude law review articles was also informed by a preliminary review of the guidance documents, which revealed a limited emphasis on law review articles. We did, however, included peer-reviewed journals that address legal issues.
While we recognize our definition of research may present some limitations, the decision to define research in this manner stems from the need to use a definition that permits a feasible examination of the bidirectional relationship between research and policy. Although we could have included other forms of research such as government reports, policy briefs, and non-government research reports within our definition of research, it would be impossible to retrieve a similarly comprehensive set of resources when looking for sources that cite OCR guidance. This uniform definition allows for practical and meaningful comparisons.
First, on the "research use in policy" side of the bidirectional relationship, we reviewed all guidance documents produced by the U.S. Department of Education Office for Civil Rights between January 20, 2009, and January 19, 2017. The guidance documents were found on the Department of Education website. Seventy-seven documents were found for this time period. Before focusing on the documents citing research, we reviewed and coded all documents for types of resources cited (research, legal references, etc.). This provided us with an overview of the types of sources that OCR relies upon to issue guidance. Out of all 77 documents, only six documents contained references to peer-reviewed research published in scholarly journals.
The six documents were coded to capture (1) topic of the guidance document, (2) the civil rights law from which OCR drew its authority to issue the guidance (e.g., Title IX), (3) dimension(s) of identity addressed in guidance (race, sex, disability, etc.) and (4) how the research was used. We also coded for the types of sources in each individual policy guidance document citing research in order to allow for comparison between research citations and other sources of authority within individual documents. For each individual study cited, we coded for characteristics such as methods, publication venue, discipline, and whether or not the publication venue is open access.
Second, on the "policy use in research" side of the study, we reviewed the ProQuest Multiple database (inclusive of 46 scholarly databases) and the SAGE database with the search terms "Office for Civil Rights" or "Office of Civil Rights" 2 AND "guidance" OR "dear colleague" in any part of the article from January 2009 to December 2017. The search yielded 719 peer-reviewed articles published in scholarly journals. We reviewed each abstract and references to determine if the article pertained to information in OCR guidance documents. Articles which did not utilize OCR produced guidance were excluded. Also excluded were book reviews, commentaries, or journal introductions and articles with generic references to OCR. After applying the exclusion criteria, 90 articles remained. We coded each article to capture (1) the topic of the article, (2) which OCR guidance document was referenced, (3) dimension(s) of identity addressed in the article (e.g., race, sex, disability, etc.), (4) how the OCR guidance was utilized, (5) methods used, (6) publication venue, (7) discipline, and (8) whether or not the journal was open access. Figure 1 summarizes the methods for this study.

Findings
Our findings revealed  interesting trends in the connections between peer-reviewed research and OCR guidance. This section will present findings related to each side of the bidirectional relationship, followed by a discussion of any overlap or disconnect between the two sides.

OCR Guidance Citing Research
Before delving into the guidance citing research, it is helpful to provide an overview of all references cited in OCR guidance issued under the Obama administration. Overall, OCR relied upon or cited a range of types of resources. Table 1 provides a summary of references cited in all OCR guidance issued under the Obama administration.

40
Legal references comprised the largest group of reference types, with 1,030 references. The second most common type of reference was policy guidance documents issued by OCR or the Department of Education more broadly (125). The third most frequent reference type was government-based research or reports, with 108 occurrences. Following this category was peerreviewed research, with 69 references. The three lowest frequencies were non-government reports/research (40), books/unpublished reports (5), and other types of references, including websites, guidance from other agencies, manuals, etc. (40). While peer-reviewed research accounted for about 4.9% of the total references cited in the guidance, when we combine the categories that together create a broad category of research (peer-reviewed research, government reports, nongovernment reports, and books/unpublished research), the percentage increases to 15.6% of the overall references. Overall, between 2009 and 2017, OCR released six documents referencing research, as defined by this study. Each document and the overarching guidance package it originated from are summarized below. The remainder of the findings section will focus on the policy guidance documents that explicitly referenced peer-reviewed research. Within individual documents, research played varying roles.
Extensive use of research. Some documents relied heavily on research, including citations to various studies within footnotes or endnotes and recurrent use of research throughout a document. On October 1, 2014, OCR issued guidance addressing resource equity in schools and its connection to civil rights protections (U.S. Department of Education, 2014f). Related to the Obama administration's "Excellent Educators for All" initiative, this guidance provided school leaders with information on the legal requirements for educational resources under Title VI of the Civil Rights Act of 1964, and touched on important topics such as academic programs, technology, and safe schools for all students regardless of race, color, national origin or zip code. The "Dear colleague letter: Guidance to ensure all students have equal access to educational resources" (U.S. Department of Education, 2014a) included in the guidance package utilized fifteen years of OCR civil rights data from every public school in the U.S. This DCL directly referenced peer-reviewed research more than fifty times, constituting the most in depth use of all OCR guidance reviewed (e.g., student achievement [Dobbie & Fryer, 2011]; teacher quality [Boyd, et. al, 2008]; teacher quality [Clotfelter, Ladd & Vigdor, 2010]; student achievement [Adamson & Darling-Hammond, 2012]; racial funding disparities [Baker & Green, 2005]).
Research was used extensively to support primary claims regarding the relationship between racial disparities and school discipline. More specifically, this letter utilized research to: (1) Show how the effect of school quality is different from the effect of poverty of a neighborhood and the provision of wrap-around services; (2) Show schools offer differentiated programming that supports career readiness; (3) Support the value of art and music programs; (4) Show advanced courses correlate with higher achievement regardless of entering status; (5) Correlate extracurricular activities and student achievement; (6) Present evidence demonstrating that schools that serve students of color often have less advanced and gifted program offerings; and (7) Show school districts with more students of color have facilities in disrepair, less resources for technology, and less qualified staff.
Although less so than the resource allocation document, another document provided multiple references to research. OCR's guidance package on racial discrimination in special education, released on December 12, 2016, addressed all schools, including charter schools, and reminded them of their obligations under Title VI of the Civil Rights Act of 1964 (U.S. Department of Education, 2016i). Specifically, the DCL included in the guidance package zeroed in on the right to a free appropriate public education (FAPE) (U.S. Department of Education, 2016d). This letter utilized research in an extensive manner, as both support and bases for propositions, with multiple references in the document to several pieces of research. Specifically, research was used to accompany discussion regarding implicit bias and "complex factors" that may contribute to the referral and identification processes (Coutinho et.al, 2002;Flores et. al, 2010;Hibel et. al, 2010;Morgan et. al, 2013Morgan et. al, , 2015Okonofua & Eberhardt, 2015).
Moderate use of research. Unlike the DCLs on equitable resources and special education identification, the DCL on discrimination in school discipline (U.S. Department of Education & U.S. Department of Justice, 2014) utilized research in a moderate manner. This letter explained that federal law prohibits discrimination in discipline in public school districts under Title IV of the Civil Rights Act of 1964. Unlawful discrimination based on race occurs in two ways: "first, if a student is subjected to different treatment based on the student's race, and second, if a policy is neutral on its face -meaning that the policy itself does not mention race -and is administered in an evenhanded manner but has a disparate impact, i.e., a disproportionate and unjustified effect on students of a particular race" (p. 7). 3 The guidance document was intended to help schools "avoid and redress" racial discrimination when disciplining students. A set of recommendations was included in the guidance package (U.S. Department of Education, 2014e) to illustrate ways to be compliant with federal laws.
Research cited in this letter was used in a moderate fashion, to provide support and introduce some concepts, but there were only singular references to each piece of research. The cited research provided additional analysis of discipline data from the CRDC (see e.g., Gregory & Thompson, 2010;Rocque, 2010;Skiba et. al, 2011;Skiba et. al, 2002). This letter and guidance package were withdrawn by the Department of Education in December, 2018.
Limited use of research. Three guidance documents made limited use of research, citing research in one or two instances. On December 15, 2015, OCR released a guidance package including a DCL addressing voluntary youth service organizations (U.S. Department of Education, 2015c). In contrast to the DCLs listed in the previous sections, the DCL on youth service organizations was brief and did not provide illustrations for implementation (U.S. Department of Education, 2015a). It presented responses to questions from school districts on single-sex programming provided by outside organizations. This guidance discussed Title IX responsibilities regarding single-sex programming, ways to determine how an organization qualifies as a "voluntary youth service organization", and how to comply with federal laws. In this guidance, research is utilized in a limited manner. Specifically, research provided support for one sentence about the connection between parent/community involvement and student/school performance (Kirby & DiPaola, 2011).
Consistent with the Obama administration's initiatives in support of equal access for women in education, on June 15, 2016, OCR also issued guidance related to equity in career and technical education (U.S. Department of Education, 2016g). Under the Perkins Act, schools must be sure that all students, regardless of gender, have access to high-quality secondary and postsecondary career and technical programs. The DCL included in the guidance package (U.S. Department of Education, 2016b) addressed the continued gender disparities in career and technical education and the legal obligations of schools. Specifically, the letter provided requirements for participation in nontraditional fields, and examples of issues that may arise for schools. This letter utilized research in a limited manner to provide insight into ambient bias that may affect gender equity in education (Cheryan et al., 2014), stating in a footnote that "scientific studies reveal the unexpected importance of a classroom's symbolic features, such as objects and wall décor, in influencing student learning and achievement in that environment. Symbols inform students whether they are valued learners and belong within the classroom, with far-reaching consequences for students' educational choices and achievement" (U.S. Department of Education, 2016b, p. 7).
Reflecting a similarly limited role of research, on July, 26, 2016, OCR released a letter and integrated substantial resource guide that was responsive to complaints of discrimination against students with ADHD and provided information on the obligations of schools under Section 504 (U.S. Department of Education, 2016e). The DCL broadly covered the legal obligations of schools to students with ADHD under Section 504. Specifically, the guidance informed schools that they must evaluate students that need special education or services, differentiated services based on the individual needs versus stereotypes about disabilities, and reiterated that schools must provide due process in ADHD cases. This document utilized research in a limited manner to provide information on evaluating students with ADHD (Arcia, et. al, 2000), and racial disparities in diagnosis of ADHD (Morgan et. al, 2013).

Characteristics of research cited in guidance.
In terms of the characteristics of research cited in guidance, there was a methodological trend that favored quantitative research (Table 2). Specifically, quantitative research comprised 68% (47/69) of all peer-reviewed research cited in OCR guidance. The next most frequently used methodological approach were meta-analyses or literature reviews, which accounted for 10% (7/69) of the peer-reviewed research cited in OCR guidance. Other types of articles were less frequent: qualitative research (4/69), policy analysis (1/69), mixed methods (2/69), and intervention studies (5/69). The 69 peer-reviewed articles represented a wide range of publication venues. A number of disciplines were represented, including economics and finance, education, law, medicine, and sociology. Only three peer-reviewed articles were published in a venue that is publicly accessible, one of which is an educational research journal (Education Policy Analysis Archives [EPAA]). Four articles were published in the American Educational Research Association's journals: Educational Researcher (1), American Educational Research Journal (1), Education Evaluation and Policy Analysis (1), and Review of Educational Research (1). Other high-profile education journals were represented in the list of publication venues, including Peabody Journal of Education (1), Teachers College Record (2), American Journal of Education (2), and Sociology of Education (1). The most cited journals were the Economic of Education Review (4) and Social Education (4). All eight of the references to these two journals were in the same guidance document, the 2014 on equal access to educational resources. There were also some more content specific journals in the sample such as Journal of Aesthetic Education (1), Journal of Labor Economics (2) and Journal of Educational Administration (2). A summary of the most cited publication venues is provided in Table 3.

OCR guidance not citing research.
A particularly noteworthy finding was that most guidance documents did not cite peer-reviewed research. The topics of these documents cut across the field of education. Several documents provided clarification on recent court decisions regarding diversity in education. For example, OCR discussed the use of race conscious admissions policies as a means to achieve diversity in higher education and utilizing race as a factor to achieve diversity in k-12 schools (U.S. Department of Education, 2013b, 2014b, 2016j). Other topics include those focused on balancing the physical health and safety of students with civil rights in schools (U.S. Department of Education, 2013cEducation, , 2014e, 2015bEducation, , 2016c. Notably, there were a number of guidance documents regarding students with disabilities that did not reference research (U.S. Department of Education, 2013aEducation, , 2014cEducation, , 2014dEducation, , 2016c. Moreover, topics that received much media attention, i.e. sexual violence and transgender students, also did not reference research. Overall, while there were some guidance documents that cited research, most guidance documents did not make explicit connections to peer-reviewed research. Summary across guidance documents. Table 4 provides an overview of the sources cited across the six guidance documents. In only one of the guidance documents (resource allocation) was research the most frequent type of resource cited. This particular document relied upon resources more often than the other guidance documents. Overall, laws tended to comprise a large portion of the resources across the six guidance documents.  It is worth noting that some guidance documents included a caveat regarding the use of research. For example, in the DCL on racial discrimination in special education, a footnote indicates that "the cited publications throughout this guidance are provided for informational purposes only. Their citation is not intended to suggest endorsement by the Department or the Federal government of the authors' conclusions" (U.S. Department of Education, 2016d, p. 11) Providing a bit more detail, the DCL on resource allocation specifically states: This letter cites to leading scholarship in the field of education in the endnotes to demonstrate the importance of the resources discussed to the quality of education that students receive and to document the disparities in access to these resources across the Nation. These citations, however, are intended to illustrate the problems we face rather than to provide an exhaustive account of the state of the research. OCR weighed information gleaned from research alongside the experience of decades of OCR enforcement of civil rights protections in our schools to determine which resources to prioritize in this letter. As with all investigations, OCR retains discretion to investigate complaints of discrimination in access to resources not discussed in this letter and will consider the fact-specific contexts of all complaints in evaluating allegations of discrimination, including evidence that in a particular school or district, the relationship between resources, the quality of education, and student outcomes may not follow typical patterns (U.S. Department of Education, 2014d, p. 2).
These statements reinforce OCR's primary role as an enforcement agency while also highlighting the ways in which research may provide meaningful support and context to this role.

Research Citing OCR Guidance
Ninety peer-reviewed research articles were found to reference OCR guidance. The articles addressed a myriad of significant topics, including but not limited to: sexual assault/sexual harassment/sexual violence (26 articles); bullying (9 articles); LGBTQ students (8 articles); participation in athletics for students with disabilities (8 articles); the use of technology to meet the needs of students with disabilities (7 articles); school diversity and segregation (6 articles); student discipline (5 articles); and special education eligibility and identification (4 articles). Other topics, such as single sex education or access to advanced courses for English Learners, appeared less frequently.
For the purpose of this study, research was coded under a dimension of identity, as dictated in federal civil rights laws (race, ethnicity, national origin, sex, or disability) if it was a main focus the article or if it provided any discussion of OCR's work related that dimension of identity. As such, depending on the research topic, some articles cut across more than one civil rights law or dimension of identity and are represented more than once in the figures presented in this section.

Race, ethnicity, and national origin.
Thirty-four of the 90 articles that cited OCR guidance included information on topics related to race, ethnicity, and/or national origin. Figure 2 reveals the range of topic areas referencing race, ethnicity, or national origin. Articles in this category did not cluster around any one topic area. The majority of articles in this section relied on OCR guidance as the reference to the legal framework for purposes of explaining compliance with federal policy, or simply as the status of the law. Most of these references were found in either the introduction to the article or as part of the literature review.
Some articles referenced data presented in OCR guidance. Information from the 2014 DCL on the nondiscriminatory administration of school discipline was used mainly to illustrate the discrepancy in expulsion and suspensions among black and white students. In one article, it was used specifically to show discipline disparities experienced by black girls (Annamma, Anyon, Joseph, Farrar, Greer, Downing, & Simmons, 2019). Barret and Katsiyannis (2015) utilized the 2014 DCL to underscore their discussion of recidivism among black and white students.
Lastly, the use of OCR guidance took a more critical perspective in a few articles. For example, Williams and McDermott (2014) looked at guidance in the context of the political climate and examined whether the timing and delivery of OCR's 2011 guidance on race-conscious policy was tied to the administration's reluctance to address race directly. In their reference to OCR's guidance on race, McDermott, Frankenberg and Diem (2014) commented that the Obama administration took some time to issue clarifying guidance after a related Supreme Court decision, which may have exacerbated existing resistance to integration policies.  Figure 2. Topics of Articles Addressing Race/Ethnicity/National Origin Sex. More than half (57) of the total research articles that cited OCR guidance documents (90) were categorized under sex. Not surprisingly, many involved a direct response to the 2011 DCL requirements for sexual misconduct policies under Title IX (U.S. Department of Education, 2011). Another significant guidance document, Questions and Answers on Title IX and Sexual Violence (U.S. Department of Education, 2014h), was issued because OCR continued to receive requests for clarification under Title IX. This guidance addressed fourteen areas of concern to assist educational institutions in meeting their obligations to students and the federal regulations. The articles citing these documents covered an array of topics, including reporting policies, adjudication, best practices, prevention, due process concerns, sexual assault nurse examiners. Again, guidance was most often referenced as the legal framework within the introduction or background of an article. What differs about the articles in this section is that many were practice-oriented and provided information on how to comply with the guidance. OCR's guidance was referenced as "the" legal framework for policies and practices related to sexual misconduct in education. For example, Koss, Wilgus, and Williams (2014) discussed the use of restorative justice approaches to improve compliance with the 2011 DCL.

Race/ Ethnicity/ National Origin
It is worth noting that some articles took a critical posture in their analysis of OCR's approach to Title IX, often focusing the lack of clarity in the then-new regulations and the resulting confusion among administrators. For example, Cantalupo (2014) noted "OCR's own guidance and an April 2011 'Dear Colleague Letter' (DCL) regarding sexual violence never explain how one would go about initiating an investigation or where one might file a complaint, even while referring to OCR investigations" (p. 231). Others focused on whether the 2011 guidance on sexual assault went far enough to protect students (see e.g. Eisenberg, Lust, Mathiason & Porta, 2017;Harper, Maskaly, Kirkner, & Lorenz, 2017). Articles under the umbrella of sex discrimination also addressed concerns related to inclusive policies and practices and general climate issues impacting LGBTQ students. To illustrate, Palmer and Greyak (2017) studied victimization among LGBTQ students and interactions with school discipline and the criminal justice system.  Figure 3. Topics of Articles Addressing Sex Disability. Forty-four articles were categorized under disability in this study. Discussion of legal obligations and the civil right of students as defined by OCR guidance were prominent article topics. Whereas many articles in the race, ethnicity, national origin, and sex categories had a critical stance, the majority of articles in this category addressed OCR guidance in a practical fashion, focusing on compliance with federal law. As with the previous categories, most references to OCR guidance were found in the background portions of the articles, setting out the status of the legal landscape on disability in education. For example, Yell, Katsiyannis, and Houchins (2016) provided an overview of court cases, administrative decisions, and guidance related to bullying and harassment of students with disabilities.

Sex
Articles outlining compliance under this section also were more specific in nature than other categories; much research was aimed at precisely how school administrators could meet obligations as set out by OCR guidance. Edyburn (2013) and Barron (2011) explored OCR guidance on the use of emerging technologies and accessibility in the field of special education. DeMartini (2016) and Yell, Losinski, and Katsiyannis, (2014) discussed the obligations of school districts with respect to the participation of student with disabilities in extracurricular activities. Other topics included the legal requirements for schools under specific guidance. As an example, Zirkel, Granthom, and Lovato (2012) explored a 2012 DCL on Section 504 obligations in the context of school nurses. They found that OCR's interpretation of Section 504 obligations differed from that under the Americans with Disability Act Amendments Act. Intersectionality. Forty-three articles covered more than one dimension of identity. The intersectional nature of the subject matter addressed in these articles could not be attributed to race, ethnicity, national origin, sex, or disability, separately. This is not surprising, given that multiply marginalized students are subject to the same interlocking systems of power, discrimination, and oppression that affect U.S. society in general (Crenshaw, 1991). For example, a number of articles that covered topics in both the race and sex category involved bullying in schools and its relation to sexual harassment, with reference to DCL on peer harassment issued in October, 2010 (U.S. Department of Education, 2010). Another example of intersectionality is best illustrated in Marsh and Noguera's 2017 study on the effects of labeling on black male students.

The use of guidance in research.
Research referenced OCR guidance in a variety of ways, representing a spectrum of policy use, from passing references to serving as the focal points for articles. Some research used data cited in OCR guidance documents to emphasize the significance of a particular issue. To illustrate this method of policy use, one article on the topic of racial disparities in student discipline stated: First, the process starts with self-awareness as teachers recognize their own cultural lens and biases. This is an important place to start because, according to a recent ''Dear Colleague'' Letter issued by the U.S. Department of Education's Office of Civil Rights and U.S. Department Justice's Civil Rights Division (2014), Black students without disabilities were three times as likely as their White peers without disabilities to be expelled or suspended from school… (Marsh & Noguera, 2017, p. 472) In contrast to this specific use of policy guidance, there were also instances when OCR guidance was used to support a general idea about the rights of students. One article turned to OCR guidance to support the assertion that "the United States is a nation that guarantees the educational rights of all children regardless of their backgrounds, including national origin (U.S. Department of Education Office for Civil Rights, 2011)" (Kanno & Kangas, 2014, p. 849). Other articles used OCR guidance as a guiding framework for understanding or analyzing a specific issue. A few articles even asked study participants or individuals analyzing case studies to answer questions directly related to OCR guidance. To illustrate, one case study asked participants to review the DCL on harassment and bullying and determine whether the scenario presented for analysis conforms with the requirements of the letter (Martin & Beese, 2016). The issuance of a specific guidance document also served as the impetus for some of the articles that were reviewed for this study (e.g., Block, 2012). In these instances, the articles often served as important opportunities to present the appropriate legal framework and its implications for policy and practice. Reflecting a more limited use, some studies used guidance merely as a reference, without directly referring to the guidance in the content of the article.
Characteristics of research citing guidance. Similar to research cited in guidance, research citing guidance represented a range of methodological approaches, disciplines, and publication venues. Unlike research cited in guidance, research citing guidance favored qualitative research methods (28). The second most common category was conceptual pieces, with 19. Quantitative research methods were employed in 12 of the articles. Close in frequency, with 11, were legal or policy analyses. The remaining designs were: literature review/meta-analyses (8), mixed methods (6), best practices (5), and intervention studies (1).
Represented disciplines were similarly diverse, with the most common being education journals. The next closest was psychology or behavioral health, with 11 articles. In terms of publication venue, the most common was the Journal of Interpersonal Violence (6), followed by the Journal of Cases in Educational Leadership (4), the Journal of Physical Education, Recreation and Dance (4), Trauma, Violence and Abuse (4). Some additional general and special interest journals, including policyspecific journals were represented more than once (e.g. American Educational Research Journal, Education Policy, Journal of Disability Policy Studies). Seven of the journals were open access journals. Law (peer-reviewed) 3 Other 8  Summary of research citing OCR guidance. Research citing guidance covered a myriad of topics that cut across race, ethnicity, national origin, sex, and disability. As evidenced by the examples presented in the previous section, researchers employed diverse methods of policy use in the articles reviewed for this study. Moreover, research was published in venues representing the cross-disciplinary work of OCR.

Overall Summary of Bidirectional Relationship between Research and OCR Guidance
Connections between research and policy appeared in both research and OCR policy guidance. On the "research use in policy" side of the relationship, guidance citing research touched on broad topics such as students with disabilities, gender equity, resource allocation, youth service organizations, and student discipline. Overall, there were six guidance documents (out of 77) that cited peer-reviewed research. Explicit, policy-connected research addressed issues such as students with disabilities (charter schools, bullying, ADHD, Section 504, extracurricular activities, the Americans with Disabilities Act, technology), Title IX and sexual violence, transgender students, English Learners, immigration, student discipline, and harassment. Seventeen out of the 40guidance packages or their included documents were cited in the literature reviewed for this study. Notably, only two guidance documents both cited research and were cited by researchers. The Appendix summarizes the use of research in OCR guidance and the use of OCR guidance in research.
As for the characteristics of research, research cited in guidance tended to favor quantitative methods while research citing guidance tended to favor qualitative research. Education journal articles were the most common discipline represented in both directions. Interestingly, when considering the most common publication venues, there was not a lot of overlap between the research cited in guidance and the research citing guidance.

Discussion
Returning to the idea of impact, introduced at the beginning of this article, existing literature examines various policy-related avenues that educational researchers may utilize to influence law and policy. These options include serving as an expert witness (e.g., Welner & Kupermintz, 2004), actively engaging in policy discourse (Pfleger et al., 2018), building relationships with intermediary organizations (Debray, Scott, Lubienski, & Jabbar, 2014), participating in the amicus brief process in court cases (Garces, 2013a;Lewis & Bray, 2019), and indirectly shaping public opinion (Welner, 2012). However, to our knowledge, this is the first study to examine the relationship between research and civil rights guidance, particularly from a bidirectional perspective. Research demonstrates that "any given policy decision is the result of many forces, with research sometimes included-but power, communications and politics often determine the research that makes it this far" (Pfleger et al., 2018, p. 16). As researchers seek to bridge the gap between research and policy, specifically in the legal context, OCR guidance is an empirically underexplored dimension of the gap.

Potential Disciplinary Barriers and Professional Norms
As we analyze our findings, the six models of research use in policy, as identified by Weiss (1979), are worth revisiting. To summarize, the models were as follows: the "knowledge-driven model," "the problem-solving model," the political model, the tactical model, and the enlightenment model (Nutley et al., 2007). In the present study, we cannot say with certainty why certain research was cited in some instances but not others. However, our findings reinforce the complex interactions between research and policy, as identified in the framework for this study.
While research was cited in OCR policy guidance, the use varied across documents and only six out of 77 cited peer-reviewed research. It is worth noting that this number would have been greater if we had used definition of research that extends beyond peer-reviewed research, including government-inititated research and non-government reports or policy briefs, which tend to be more easily accessible to the public. As demonstrated in the literature, there is a threshold issue with access to peer-reviewed publications (Daly & Finnigan, 2014). Although some publication venues, such as EPAA, provide the public with free access, many publications remain behind a paywall. Related to publication access is the process, itself, which tends to be long and inconsistent with the needs of policymakers (Yohalem & Tseng, 2015). Some journals have begun to take this into consideration, but efforts to improve the timeliness of the publication process are still a work in progress.
From a disciplinary standpoint, OCR, as opposed to other divisions within the U.S. Department of Education, represents a legal perspective, which may come with different disciplinary norms and perspectives. As such, the use of social science research could be considered somewhat nontraditional to those who are guided by their legal training. An overall look at all resources cited in the OCR guidance reviewed for this study hints that this may be the case. Legal references such as statutes, regulations, and case law appeared approximately 15 times more often than peer-reviewed research. This number is even greater if you combine legal citations with policy guidance documents. Depending on the disciplinary background of individuals crafting guidance, there is a possibility that OCR policymakers, like judges (Morgan & Pullin, 2010;Welner & Kupermintz, 2004), have a difficult time sifting through relevant educational research because it is not their primary area of expertise. An interactive model of research use (Nutley et al., 2007;Weiss, 1979) would be ideal to address some of these potential disciplinary barriers.
Of course, the use of research may also depend on the purpose or impetus for the issuance of a particular guidance document. For example, if the purpose of the guidance is to clarify or reiterate a legal standard based on recent developments in case law, perhaps policymakers see less of a need to identify relevant research. Related to the purpose could be a concern that the guidance may be more likely to be perceived as being guided primarily by research rather than being guided by the law, thereby making it more susceptible to being challenged in court or rescinded by a future administration. In this regard, the use (or nonuse) of research may be viewed as tactical (Nutley et al., 2007). For example, when the 2016 transgender guidance was rescinded, OCR argued that the guidance did not "contain extensive legal analysis or explain how the position is consistent with the express language of Title IX, nor did they undergo any formal public process" (U.S. Department of Education and Department of Justice, 2017, p. 1). However, in addition to legal resources, social science research has a rich history in law; within the educational context, its use in OCR policy guidance is merely an extension of an already robust and growing use of research in law through other means, such as amicus briefs, expert testimony, etc. (e.g., Garces, 2013a;Lewis & Bray, 2019;Superfine, Goldman, & Richard, 2019).
Both research and law are complex and in a constant state of evolution. As an illustration, in the last few years, multiple OCR guidance documents have been rescinded, including some mentioned in this article. Policy-informed research and research-informed policymaking requires ongoing maintenance. Keeping track of this activity and its relationship to your research as a "policy actor" (Welner, 2012, p. 7) should not be oversimplified.

The Perception of Research: Purpose, Function, and Methodological Approaches
It is important to acknowledge that research is not used in a vacuum and there is a political dimension to the process 2014). As Tseng (2012) stated, "rather than viewing politics as a nuisance to be set aside, it behooves us to increase our understanding of how the political and policy process works and how it influences research acquisition, interpretation, and use" (p. 8). Bound by a presidential administration, specifically the Obama administration, this study's design acknowledges the political nature of policymaking, including the issuance of OCR policy guidance. As stated previously, the Obama administration engaged in novel efforts to integrate research into OCR's work. In OCR's 2009-2016 report titled "Achieving Simple Justice," OCR described their work under the Obama administration as follows: One of the Obama Administration's highest priorities has been to deliver that simple justice by safeguarding students' access, and reducing discriminatory barriers, to educational opportunity. Over the past eight years, the U.S. Department of Education's Office for Civil Rights (OCR) has significantly contributed to this priority through vigorously enforcing federal civil rights laws, collecting comprehensive data on equity and opportunity gaps in schools, issuing timely civil rights policy guidance, and enhancing the public's awareness of civil rights and of OCR's work (U.S. Department of Education, 2016a).
As evidenced by this study, the research-policy gap is intertwined with other forces and sources of debate. According to the literature, there is a "general perception that educational research is low quality" (Feuer, Towne, & Shavelson, 2002, p. 5). Of the research-policy connections that did exist in this study, from a methodological perspective, policymakers relied more heavily on quantitative research methods, while researchers citing guidance tended to favor qualitative research. This inconsistency and lack of consensus is reflected in the scholarship as well. According to Feuer, Towne, and Shavelson (2002), Educational researchers themselves are often their own harshest critics (e.g., Kaestle, 1993). They are often joined by a chorus of social and physical scientists, engineers, and business leaders who lament weak or absent theory, accumulations of anecdote masquerading as evidence, studies with little obvious policy relevance, seemingly endless disputes over the desired outcomes of schooling, low levels of replicability, large error margins, opaqueness of data and sources, unwillingness or inability to agree on a common set of metrics, and the inevitable intrusion of ideology at the ground level (p. 5).
Within law in particular, authors similarly argue that there are "wide gaps in understanding between the legal and research communities about the rigor and value of social science in legal decisionmaking" (Marin, Yun, Garces, & Horn, 2019 p. 2). On the "policy use" side, our findings demonstrate similar issues related to the functional value of research. While some researchers are considering the broader policy context in which their work is situated, many are not. Although we cannot definitively state why this occurred, we theorize that some researchers might consider their work to lack policy implications due to methodological considerations such as generalizability (e.g., Feuer, Towne, & Shavelson, 2002). However, publication venues appear to be interested in research with connections to policy. In fact, some education journals' mission statements explicitly mention implications for policy as a desirable, if not required for publication. For example, Peabody Journal of Education: Issues of Leadership, Policy, and Organizations, Education Policy Analysis Archives, Educational Evaluation and Policy Analysis, and Journal of Education Policy, the American Journal of Education appear to have an interest in policy-related research. Not surprisingly, some of these journals appeared in both directions of our study. Making explicit connections to policy guidance has the potential to improve upon existing efforts to promote the "knowledge-driven model" where "research findings can be communicated to impel action (Nutley et al., 2007, p. 285).
As for publication venue, the most commonly occurring publication venues (cited in guidance and citing guidance) did not have much overlap. Although a source of much debate within the research community, there are metrics that we often turn to in order to judge publication quality and impact, including but not limited to impact factor. The present study presents important implications for this broader discourse. For example, Social Education, the flagship journal for the National Council for the Social Studies was the most cited publication venue in OCR guidance. Although this journal is not indexed, nor does it have an assigned impact factor, this study provides evidence of this journal's impact on policy. As such, this study raises considerations for the ways in which we assess impact in the field of educational research.

Limitations
This study has a few limitations worth noting. First, our definition of research is confined to peer-reviewed research. While this presents some limitations, this definition not only captures the primary forum where researchers publish their work, it also allows for an examination of the bidirectional relationship between research and policy guidance. Consequently, this definition was critical to bound the study. Here, we focus specifically on explicit references to research and policy guidance. To capture the more nuanced connections between research and OCR policy guidance, future research could explore existing networks and policy-based relationships between authors of guidance and researchers. This research design would align more closely with the enlightenment and interactive models of research use (Nutley et al., 2007;Weiss, 1979). On the consumer side, future research could involve interviews or surveys with policymakers who participated in the creation of policy guidance. On the production side, future research could aim to better understand how cited authors disseminate and distribute their research findings. Research might also seek to undercover researchers' understanding of OCR, policy guidance, and the role that research should play in policy guidance. The present study serves as a foundation for these follow-up studies.
Another consideration is the inclusion criteria related to the year of publication. Given publication timelines, our research design may not have captured all research citing guidance, particularly later guidance. However, there is almost a year in between the last guidance issued in 2017 and research publication parameters (December 31, 2017). As a result, our design for this study provided enough time for researchers to at least add reference to even the last OCR guidance documents of the Obama administration. Research will likely continue to cite OCR guidance issued under the Obama administration, including the time period outside the scope of the present study, which may be explored in future research examining the longevity and impact of Obama era guidance.

Implications for Research, Policy, and Practice
As an initial exploration into the relationship between research and OCR policy guidance, the present study reveals some implications for researchers, policymakers, intermediary organizations, and practitioners. As indicated earlier in this article, the Obama administration engaged in unprecedented efforts to create user-friendly guidance, including through the inclusion of research . Nonetheless, even within the guidance documents citing research, we identified areas of improvement. Although researchers may have played an implicit role in the issuance of these guidance documents, explicit references to research allow for consumers to consult research to inform the implementation of the requirements set forth in the guidance. While we recognize that there are many ways that researchers may influence policy guidance, our findings nonetheless revealed some important missed opportunities.
Under the Trump administration, OCR has rescinded multiple guidance documents, spanning issues such as race-conscious admissions policies, inclusive policies for transgender students, student discipline, and sexual assault. These topics, as well as other topics under the purview of OCR, are the subject of a large body of research. As OCR continues to identify future actions, it is important for researchers to be aware of policy activity and play an active role in informing future actions. This study demonstrates that in recent history, OCR has explicitly turned to research to support its policy guidance and as researchers, we can work to build upon these explicit connections.
The social science research community is already actively engaged in efforts to influence policy. Within the context of law in particular, researchers are mobilizing around cases of national significance related to race conscious admissions policies in higher education and diversity in k-12 schools (Brief of American Social Science Researchers, 2013; Brief of 531 social scientists and scholars, 2018; Brief of the American Educational Research Association et al., 2013). Moreover, in issuing guidance, OCR seeks input from multiple stakeholder groups (Testimony of Catherine Lhamon, 2014; U.S. Department of Education, 2016f, 2016k, 2018. Researchers should continue to be involved and increase their involvement in these discussions. Furthermore, it is critical for researchers to continue to engage in research with an eye toward implications for OCR policy guidance. Similarly, the research community should study the implications of policy guidance so that OCR can better understand the potential impact of a decision to rescind guidance. Relationships are key to bridging the gap between research and policy. According to Pfleger, Wilson, Welner, & Bibilos (2018), "a key element of such public scholarship calls upon researchers to join with others to more directly communicate research with policy-makers and practitioners, fostering democratic public problem solving in ways that resonate with broad concerns" (p. 17). Tseng (2012) argues that the communication of research requires translation, an act that may be done by researchers, intermediaries, or through a joint effort in which intermediaries serve as "relationship brokers, bringing researchers and decision-makers together to focus on core problems of practice or policy" (p. 12). Similarly, Marin, Yun, Garces & Horn (2019) argue that "identifying and networking those knowledge producers who are already inclined to work with legal practitioners and matching them with one another-that is, connecting communities who already share a more common understanding-could meet the goal of narrowing the gap between producer and practitioner communities" (p. 19). These ideas will be important to consider within the realm of research and OCR policy guidance as well.
The present study revealed important areas of improvement for policymakers. "Research use is contingent, interactive, and iterative. It involves people individually and collectively engaging with research over time, bringing their own and their organization's goals, motivations, routines, and political contexts with them" (Tseng & Nutley,p. 165). The data reviewed for this study showed opportunities for policymakers to increase the breadth and depth of explicit connections to research, including diversifying topics, publication venues, and methodological approaches.
In the context of civil rights enforcement, policy guidance serves as a communication tool with direct connections to local policies and practices. Practitioners can look to policy guidance not only to understand civil rights obligations but also to identify research that may be useful to the implementation of policies and practices that identify to identify, address, and prevent civil rights violations.

Conclusion
Understanding and strengthening the relationship between research and policy is important for researchers, policymakers, and practitioners. There are opportunities for growth on both the policy side and the research side of the relationship. Educational researchers could place greater emphasis on the policy context in which their work is situated and make explicit connections to this context. Policymakers could build upon their explicit use of high quality, cutting edge research. To bridge the current divide, it is important to "build relationships and trust, shore up capacity, create conditions for evidence integration, and develop partnerships" (Tseng & Nutley, 2014, p. 168). In the context of the U.S. Department of Education Office for Civil Rights, a synergistic relationship between research-informed policy guidance and policy-informed research promotes equal access and opportunities for historically marginalized students.