Educational leaders' reports of conditions for supporting SEL implementation: The power of partnerships

Abstract

Well‐implemented social and emotional learning (SEL) programs are associated with positive student outcomes. Through CalHOPE, County Office of Education (COE) leaders are providing implementation support to districts and schools seeking to improve SEL delivery statewide. CalHOPE uses the SHIFT model to guide the regional implementation of systemic and equity‐oriented SEL. This study assesses the conceptually driven conditions among COEs for providing implementation support, including adult well‐being, workplace climate, and levers of transformation (forging partnerships within and outside of the organization, providing support through funding, training, coaching, and tools, and building capacity of leaders in the form of mindsets, knowledge, skills, and efficacy, to improve the structures and routines of SEL implementation—such as having a vision, SEL leadership team, data systems, and goals/plans). Ninety‐seven COE SEL leaders participating in CalHOPE reported on their conditions and activities. COE SEL leaders reported positive well‐being, workplace climate, and levers of transformation. Having a greater variety of partnerships was significantly associated with a stronger presence of all levers. Findings advance our understanding of the scaling up realities of SEL implementation in a statewide effort and highlight the value of partnerships in advancing SEL implementation.

Highlights

• County Offices of Education (COEs) representatives report positive well‐being and workplace climate.

• COEs report the presence of partnerships, supports, capacities, and routines for SEL implementation.

• Partnership Variety is associated with more levers of transformation to support SEL implementation.

INTRODUCTION

Social and emotional learning (SEL) is a widely adopted and promising prevention strategy for supporting the engagement, performance, and well‐being of young people and educators in schools (Durlak et al., 2022). SEL is the process through which youth and adults acquire and apply knowledge and skills to set and achieve positive goals, feel and show empathy for others, establish and maintain positive relationships, and make responsible decisions to support their emotional development and social relationships (Collaborative for Academic, Social, and Emotional Learning [CASEL], 2019, 2024). Research suggests that SEL works best when implemented well (e.g., Durlak et al., 2011). Program support (e.g., training) has been associated with SEL implementation quality more often than school (e.g., organization), teacher (e.g., attitudes), or student (e.g., demographic) characteristics (Ulla & Poom‐Valickis, 2023). Program support, for high‐quality implementation, is widely understood to be best accomplished through “systemic SEL” (Mahoney et al., 2021)—an approach to delivering SEL that engages multiple levels of the educational system to support and sustain the work. Systemic SEL involves four main elements: (1) developing foundational supports and plans; (2) enhancing adult SEL capacities; (3) fostering student SEL competencies and skill development; and (4) practicing regular, data‐driven continuous improvement. Transformative SEL (tSEL; Jagers et al., 2019) considers how SEL can promote equity by reflecting on root causes of inequity (e.g., poverty, systemic racism) in the promotion of social‐emotional competencies, and by providing culturally responsive, student‐centered education. Despite growing calls for shifting SEL towards systemic and transformative approaches, few models currently exist for integrating these approaches and applying them to practice (Shapiro et al., In Press). Doing so is theorized to require a collaborative, multi‐systemic prevention infrastructure (Shapiro et al., 2013).

PREVENTION INFRASTRUCTURE

In the broader education literature, district offices are conceived to support school improvement processes (Honig et al., 2009). Yet, their capacity to provide implementation support is often limited (Gamson, 2009), especially in smaller districts. Research studies often conclude that “implementation processes and outcomes fell short of district leaders' goals absent substantially increased capacity of people, including central office administrators” (Honig et al., 2009, p. 24).

Educational service agencies (ESAs) may be a way to expand upon district abilities to provide implementation support. ESAs (e.g., Educational Service Districts, Intermediate Units, Regional Educational Service Centers) are regional agencies who are “authorized by State law to develop, manage, and provide services or programs to local educational agencies [LEAs; e.g., districts]… recognized as an administrative agency for purposes of the provision of special education and related services provided within public elementary and secondary schools of a state (20 USC § 1401(5) (2006)).” In California, the State Constitution establishes a county superintendent of schools (elected or appointed by their county's board of education) who, with their staff, are commonly referred to as County Offices of Education (COEs). As described by the California Legislative Analyst's Office, “COEs provide direct instruction to students in juvenile court and county community schools, and provide a range of oversight and support to school districts in their county… Some of these activities, such as fiscal and academic oversight, are required by state law. Many other activities COEs conduct are optional, with the specific activities conducted by COEs varying significantly across the state.” (Petek, 2024). COEs have had their current responsibilities for fiscal oversight of districts (e.g., review and approval of school district budgets) since 1991, oversight of district plans (e.g., Local Control Accountability Plans) since 2013, and responsibilities for academic oversight since 2018, when COEs became part of a system of supports that “provide coordinated, needs‐based, and differentiated resources and supports to LEAs that lead to improved services for all students” (California Collaborative for Educational Excellence, n.d.). Thus, 58 COEs, all reporting to the California Department of Education and governed by county‐specific boards of education, are tasked with providing regional support, as needed, to over 1000 school districts with a highly diverse population of nearly six million students (California County Superintendents, 2022). Despite the role these intermediary organizations play in providing implementation support to districts for various academic, behavioral, and special education initiatives (Research for Action, 2014), their role has not been well articulated or studied in the peer‐reviewed literature, or applied to implementation support for SEL (Metzger et al., 2024).

We draw on the Interactive Systems Framework (ISF, Wandersman et al., 2008) and apply it to the education system as a conceptual framework for understanding SEL implementation (Shapiro et al., 2024). This positions COEs as part of the SEL implementation support system and districts and schools as part of the SEL delivery system. In this way, COEs can provide districts and schools with vital supports (e.g., tools, training, coaching, and feedback systems; Wandersman et al., 2012) for SEL implementation. Previous studies have examined the effect of implementation support provided by a program developer/publisher on the delivery of SEL (e.g., Li et al., 2023), while others have demonstrated that providing in‐district implementation support helps schools complete their standardized implementation strategy (Meyers et al., 2019). In this second study “all schools maintained an SEL leadership team that was intact throughout the 2 years of implementation. … At the end of these 2 years, each team had developed a vision for SEL in their school, had assessed their current resources and needs for supporting high‐quality SEL, established S.M.A.R.T. [specific, measurable, achievable, relevant, and time‐bound] goals for SEL, and completed an implementation plan that focused on integrating SEL throughout their school community.” (p. 59). The present paper seeks to better understand the potential of educational service agencies, such as COEs, to provide implementation support to districts and schools delivering SEL—which requires a conceptually‐driven assessment of the conditions and activities of COEs as intermediary organizations supporting systemic and transformative SEL implementation.

SHIFT model for SEL implementation

The SHIFT model for SEL implementation was developed through a systematic search and scoping review of the SEL implementation literature co‐interpreted with practice partners (Shapiro et al., In Press). It is intended to guide regional implementation of SEL, to “shift SEL” toward a more Systemic and Humanizing Implementation Focused on Transformation (see Figure SA). The SHIFT model aims to foster individual and systemic thriving where all people are engaged, highly performing, and well (individual thriving) due to favorable conditions for learning, teaching, and leading that catalyze, reinforce, and accelerate improvement (systemic thriving). As focused for the current inquiry, adult well‐being is theorized as necessary for adult engagement and performance in the task of providing implementation support. SHIFT further posits that the workplace climate—the persistent patterns of the setting that convey care and respect and create norms and expectations that transcend daily experiences—is related to adult well‐being and the success educators have in engaging in and performing the activities that support systemic and transformative SEL implementation (Shapiro et al., 2024). The model denotes these activities as levers of transformation. Specifically, there are four levers of transformation that are postulated as mechanisms of systemic and equitable implementation support: forge partnerships, provide support, and build capacity to improve structures and routines of SEL implementation. Levers consider how equity is conceptualized (i.e., equity thinking) and enacted (i.e., equity actions) in practice (Jones et al., Under Review). Ultimately, SHIFT explains the conditions for providing SEL implementation support through an understanding of adult well‐being, workplace climate, and the use of the levers of transformation. While the SHIFT model has been used thus far to guide SEL implementation, it may be useful for other change initiatives seeking to promote individual and systemic thriving.

Current study

Guided by the SHIFT model, we seek to understand the conditions for and activities of providing implementation support for systemic and transformative SEL, as perceived by educational leaders reflecting upon their SEL work at COEs (i.e., COE SEL leaders). We first ask (RQ1): What are COE SEL leaders' perceptions of their own well‐being, COE workplace climate, and COE use of SEL levers of transformation? This study is among the first to measure the conditions and activities needed for the provision of implementation support for systemic and transformative SEL in the public education system. Given the conceived role of COE representatives in supporting district and school‐level implementation, understanding for the first time how COE SEL leaders perceive their own well‐being within the context of their workplace climate is critical; further, their perceptions of the extent to which the levers of transformation have been enacted is requisite to future efforts to study the impact of these transformative levers on subsequent SEL implementation. Second, we ask (RQ2): To what extent are COE SEL leaders' reports of a larger variety of working partnerships associated with reports of stronger COE use of levers of transformation? The SHIFT model logically postulates that partnerships are requisite in the provision of implementation support (Shapiro et al., 2024). While many have examined the power of partnerships for implementing preventive interventions (e.g., Chilenski et al., 2018; Spoth & Greenberg, 2005), studies about the nature of COE partnerships (e.g., variety of partners) and more generally, the quantitative study of county‐level partnerships, have rarely been undertaken. This is in part due to constraints on sample size and limitations of generalizability to diverse counties. The present theoretically‐driven inquiry with representatives from a large number of diverse county offices of education has the potential to generate initial practice guidance for the process of enacting levers of transformation to improve systemic and transformative SEL implementation community‐wide.

Background literature

Well‐being

The field of community psychology has a long‐standing interest in promoting well‐being (Allen & Blackburn, 2023). The SHIFT model emphasizes the well‐being of all people involved in the education system—educational leaders included. Educational leaders with strong well‐being may have better engagement and performance in their roles, and ultimately be more successful in enacting change. Drawing from the concept of psychological wellness (Cowen, 1994), the SHIFT model considers well‐being to be more than the absence of psychopathology, but instead, a continuum that includes positive health markers. SHIFT operationalizes well‐being as having more frequent positive emotional experiences, less frequent negative emotional experiences, and greater resources for coping—based on Yosso's (2005) Community Cultural Wealth framework.

The well‐being of educators is consequential. Research documents the negative impact of stress and burnout on school leadership (e.g., principal) turnover and effectiveness (Boyce & Bowers, 2016; DeMatthews et al., 2021; Yan, 2020). Additionally, when employees experience positive well‐being, they tend to exhibit increased productivity, creativity, and commitment to their roles (Krekel et al., 2019; Oswald et al., 2015). This individual effectiveness aggregates at the organizational level, leading to improved overall performance, innovation, and adaptability. Thus, a community of educational leaders who are collectively well may be more likely to be engaged and retained, and have higher levels of performance.

Workplace climate

A positive workplace climate may also help leaders to engage and perform in their roles (Soutter, 2023) to advance SEL implementation. According to Ostroff et al. (2003), workplace climate emerges from the shared meaning that employees derive from their collective experiences. A positive workplace climate (e.g., safety, connection) can reduce turnover, absenteeism, and workplace stress, further contributing to organizational efficiency and effectiveness (O'Brennan et al., 2017; Stapleton et al., 2020).

Specifically, in the SHIFT model, a positive workplace climate includes workplace experiences of safety and connection, opportunities for voice and leadership, and culturally and linguistically responsive environments. Psychological safety (i.e., comfort expressing oneself and one's opinions, concerns, and ideas without fear of judgment, loss, or punishment; Kahn, 1990) and connection (i.e., feeling a sense of belonging with a set of common experiences and shared purpose; Burroughs & Eby, 1998) among workers are important conditions for working toward systemic changes (Tucker & Edmondson, 2003). Opportunities for voice and leadership (i.e., circumstances in which people feel included and valued, make contributions, and meaningfully shape decisions) are theorized to promote empowerment and well‐being (Trickett & Rauk, 2019) and have been associated empirically with the provision of implementation support (e.g., funding, training; Brown et al., 2015), successful cross‐sector partnering and the development of individual capacities for change‐making (Shapiro et al., 2015), and overall perceptions of community‐level change (Allen, 2005; Mizrahi & Rosenthal, 2001). Expanding upon the ideas of culturally responsive pedagogy (Ladson‐Billings, 1995, 2014), culturally responsive teaching (Gay, 2002), and culturally sustaining pedagogy (Paris, 2012), the SHIFT model defines culturally and linguistically responsive environments (CLREs) as socio‐politically conscious spaces with practices that intentionally and equitably integrate peoples' cultural identities into norms in ways that facilitate relevance for all and reverence for minoritized groups. CLREs are important to create a “more welcoming and affirming” (Vera, 2023, p. 235) school culture for all students, are linked to improved well‐being (Cholewa et al., 2014) and achievement (Howard & Terry, 2011; Kelley et al., 2015) for students of color, and are theorized as important for adults also in a culturally situated, community‐based, multilevel intervention (Trickett, 2009).

Levers of transformation

The SHIFT model posits four levers of transformation to be the mechanisms that change the educational system ecology to one that better facilitates and sustains SEL delivery and improves the social and emotional conditions for learning, teaching, and leading (Shapiro et al., 2024). The levers include: (1) enhancing partnerships (e.g., across sectors, regions, and system levels), (2) receiving and providing supports (e.g., funding, training, technical assistance), (3) building capacities (e.g., knowledge, skills), and (4) augmenting the structures and routines of SEL implementation (e.g., forming SEL leadership teams, developing a shared vision). The levers, described below, are posited to build upon each other—beginning with partnerships, leading to supports, which build capacity for the structures and routines of SEL implementation.

In the SHIFT model, partnership relies heavily on developing trusting, equity‐pursuing, and mutually beneficial collaborations that span sectors and regions, between levels and divisions of the education system, with families, and allied to students (Shapiro et al., 2024). Cross‐sector partnerships (e.g., education, behavioral health) form “organizational linkages” for collective action (Shapiro et al., 2015). A partnership may increase social capital and functional diversity in ways that open up external knowledge and resource inputs (Chesbrough, 2017; Coleman, 1988; Van Beers & Zand, 2014), create trusting instrumental ties (i.e., technical or tangible relationships) between social networks (Lin, 1999), and matter for collective improvement efforts (Finnigan & Daly, 2012). Between‐level partnerships may be easier to forge, but are often unfortunately characterized by a lack of reciprocity (Mania‐Singer, 2017). The SHIFT model posits that partnerships create the conditions for implementation support to be provided and received.

Supports include the funding, resources, training, coaching, and feedback systems that are critical to successful SEL implementation. Some have conceived a large role for COEs to play in providing these supports to districts and schools. For example, by seeking and distributing funding, COEs enable districts to find time for district and school site educators to attend trainings and obtain coaching. Access to supports is broadly theorized, and has been demonstrated, to increase capacities among educators—the mindsets, knowledge, skills, and confidence that are required to implement SEL effectively (e.g., Duane et al., 2025a). Adult mindsets, such as viewing SEL as a priority, are important for systemic SEL implementation (Huck et al., 2023). Yet, as much as they may be necessary, mindsets may also be insufficient (Gimbert et al., 2021). Knowledge (e.g., understanding what SEL is and why it is important) is also important for providing implementation support (Kendziora & Osher, 2016; Walker & Martin, 2020). Further, it is necessary for educators at school sites to have skills and confidence as SEL implementers (Schonert‐Reichl, 2017).

The collective capacities of educators become the collective conditions that enable structures and routines for SEL implementation support and delivery (Shapiro et al., 2024). These structures and routines include forming an SEL leadership team, creating a shared vision, conducting needs assessments, creating plans and goals, cultivating communication, providing instruction or programs, developing a continuum of supports, and enacting data‐based reflection and improvement (Meyers et al., 2019). In a pilot study we conducted of COE SEL leaders providing SEL implementation support, most reported that, after 17 months of support and capacity‐building, their COE embedded SEL in their internal structures and routines, including their COE‐wide programs, practices, and to a lesser extent, in their COE‐wide policies (Jones et al., Under Review). Most COEs reported having a shared vision of SEL for their COE and having formed an SEL leadership team, with fewer having built data systems for continuous improvement. Further, these leaders reported that their COE used “equity thinking” in the provision of SEL implementation support by avoiding deficit narratives and reflecting on systemic inequities, but took relatively fewer “equity actions.”

In summary, SHIFT SEL posits that trusting, equity‐pursuing, and mutually beneficial partnerships begets SEL implementation success by pooling people, sharing knowledge, collaborating on processes, and drawing on community cultural wealth. Such partnerships facilitate the provision and receipt of implementation support in ways that build capacity for the structures and routines of SEL implementation. Yet, in addition, there is little practice guidance on the nature of partnerships that make subsequent levers of transformation more likely. Previous studies have shown that broad involvement from diverse community sectors in collective efforts, but not necessarily the depth of partner involvement (Feinberg et al., 2004; Hays et al., 2000), predicts leader reports of impact (Florin et al., 2000) and program implementation success (Halgunseth et al., 2012). In fact, analysis from a methodologically rare community‐randomized trial indicated that the involvement of 13 diverse sectors in a community change initiative explained all of the systematic variance in community‐level systemic outcomes (Shapiro et al., 2015), rending the involvement of a variety of partners an important characteristic to explore in change initiatives.

In sum, this study seeks to examine the conditions for and activities of systemic SEL implementation through COE SEL leaders' perceptions of their own well‐being, the workplace climate of their organization, and the extent to which they perceive that their organization has implemented the levers of transformation to facilitate SEL implementation. Secondly, this study examines the extent to which having a variety of partnerships relates to stronger implementation of the levers of transformation. This study aims to fill critical gaps in the literature on real‐world attempts to implement systemic SEL, undergirded by transformative SEL, to better understand the role of the county in advancing SEL implementation statewide.

METHOD

Setting

CalHOPE Student Support (henceforth CalHOPE) is a statewide strategy, funded by the California Department of Health Care Services, to build multi‐systemic prevention infrastructure to support and deliver SEL in schools (Shapiro et al., 2024). The state‐level leadership team is composed primarily of lead education practice partners (Sacramento County Office of Education), experts on the translating research for the practice of SEL (Greater Good Science Center) and implementation (The Center for Implementation), an education technology company (Kelvin Education), and university researchers at a large public university (University of California, Berkeley). The university partners (a subset of whom are this paper's authors) provide technical assistance and evaluation support, and secondarily use this data for research purposes. There is a long history of systemic SEL implementation efforts in California (see Shapiro et al., 2024). CalHOPE builds on this to continue advancing systemic and transformative SEL implementation statewide.

The goal of CalHOPE is to improve student and adult outcomes in the education sector by strengthening systemic and transformative SEL implementation statewide. To accomplish this goal, CalHOPE has networked COE leaders to learn with and from each other about how to provide SEL implementation support in their respective regions (Eldeeb et al., 2025). CalHOPE's state leadership team provided support (i.e., funding, resources, technical assistance) to these COE leaders, who in turn provided support to Focal Sites (i.e., three to eight schools selected to receive focused supports to advance their SEL efforts) in their respective regions (Metzger et al., 2025). COE SEL leaders worked with district SEL leadership teams, directly with school site SEL leadership teams, and/or directly with school site personnel (Cirolia et al., In Press). In CalHOPE, COE SEL leadership teams intend to serve as “reliable conduits of information and facilitators of practice improvement” in their regions (Shapiro et al., 2024, p. 6). CalHOPE has thus made a concerted effort to, “build the capacity of the capacity‐builders” (p. 4).

Data collection

Survey data were collected 35 months after a CalHOPE kick‐off event. Data were collected across approximately 3 months in the Summer of 2023. Any COE SEL leader known to have participated in CalHOPE at any time (by nature of having their contact information on a practice roster, from attendance records, or from prior survey invitations or responses) was emailed a personalized Qualtrics survey invitation. Survey responses were used to filter the sample to educational leaders who said their “job is best described as situated within the county office, or otherwise serving students across the whole county” and affirmatively responded to the question: “Are you in a role at your COE where you support social and emotional learning?” All study procedures were approved by the Institutional Review Board at the University of California Berkeley.

Participants

A total of 117 COE SEL Leaders meeting inclusion criteria responded to the survey for continuous improvement and grant‐reporting purposes, and of these, 97 consented to the use of their survey responses for research purposes¹ (83% consent rate) and comprise the current study sample. Workers, by nature of their subordinate role in a hierarchical labor structure, could be vulnerable to undue influence to participate in research; thus, prospective participants were informed that their choice whether or not to participate would only be made available to UC Berkeley researchers (i.e., no information would be provided to their employer or any other third party as to who opted into the research study). Participants were assured responses would be aggregated statewide, and only disaggregated (by potentially identifying characteristics) if more than 10 people shared the unique combination of institutionally and/or individually identifying features. Participants came from 49 counties (84% of all counties) across California. Participants identified primarily as non‐Hispanic White (74%) and female (77%; see Supporting Information S1: Table S1 for additional information). Among those COE SEL leaders who reported having an SEL leadership team at their COE (83%), 88% of participants identified as team members. Many leaders (38%) had been in their current role for at least 6 years, characterizing their role as primarily administrative/operations support (56%) or instruction/learning support (29%). Overall, 87% spent “not a lot of time (0%–24%)” interacting with youth in their COE role (rural counties providing some exceptions). Respondents who consented to their survey responses being used for research purposes, and respondents who did not, did not differ by years of experience in current role, gender, race/ethnicity, or county rurality.

Measures

COE SEL leaders were invited to complete the Berkeley Assessment of Social Emotional Learning‐Leader Voice (BASEL‐LV; Shapiro et al., 2022). This self‐report survey of leaders was designed to assess their well‐being, workplace climate, and levers of transformation to help facilitate needs assessment, planning, and progress monitoring for support and capacity‐building efforts. The majority of BASEL‐LV survey items used response options on a 4‐point Likert scale (4: “YES!” means definitely true for you, 3: “yes” means mostly true for you; 2: “no” means mostly not true for you; 1: “NO!” means definitely not true for you). This scaling technique has been validated and used in multiple studies over the past 40 years (Social Development Research Group, 2005‐2019). Higher scores indicated a more favorable response. The survey also included five demographic questions (see Table S1). Scale scores were derived from the mean of scale items. Reliabilities for each scale are reported using Cronbach's alpha (ɑ).

Well‐being was measured through two constructs: Positive Emotional Experiences and Resources for Coping. The Positive Emotional Experiences scale (7 items; ɑ = 0.70) asked respondents to indicate positive affect over a 2‐week period (e.g., proud, strong, or determined; grateful, thankful, or appreciative). Two items (i.e., nervous, anxious, or jittery; down, sad, or hopeless) were reverse‐coded. The Resources for Coping Index (5 items) assessed coping with work‐relevant stressors (e.g., “Even when things are hard, I remember why I work in education”). One item was reverse‐coded (“After work, it takes me longer than it did in the past to unwind”).

Workplace climate was measured as three constructs: Safety and Connection (3 items; ɑ = 0.86; e.g., “There is a collaborative work culture”), Opportunities for Voice and Leadership (3 items; ɑ = 0.81; e.g., “My contributions are valued and respected”), and Experiences with Culturally and Linguistically Responsive Environments (3 items; ɑ = 0.85; e.g., “People consider the impact of race and racism on experiences in school”).

Respondents also reported on the four levers of transformation for the provision of SEL implementation support. Each lever was measured using a separate scale. A scale measuring Partnership Activities was created to indicate the types of activities that educational leaders engaged in with existing teams, community members, and students (3 items; ɑ = 0.67; e.g., “Student voices guide our county's SEL implementation”).

Supports were measured with two scales referencing activities in the past 6 months. A Supports Received scale (25 items; ɑ = 0.93) refers to the supports that COE SEL leaders received from the CalHOPE statewide team (e.g., “Over the past 6 months, our COE has received support from the CalHOPE Statewide team for SEL implementation in the form of…Tools (e.g., facilitator guides, presentation slides, media, curricula”). The Supports Provided scale (17 items; ɑ = 091) refers to the supports that educational leaders provided to districts and schools (e.g., “Over the past 6 months, our COE SEL leaders have provided district/school staff support for SEL implementation in the form of…Training/professional development”). Individual items included reports on funding, training, tools, and coaching supports for the practice of SEL, the continuous improvement of SEL implementation (e.g., receiving support for the utilization of SEL data), and culturally and linguistically responsive practices that center and affirm student identities.

Capacities (16 items; ɑ = 0.87) included SEL leader perceptions of their own mindsets (e.g., “Considering all other competing COE priorities, I believe SEL is very important for…Student performance (e.g., academic achievement, grades)”), knowledge (e.g., “I am able to answer questions from people in my region about…The practice of SEL”), skills (e.g., “I have the skills to remove obstacles to SEL implementation”), and sense of efficacy (e.g., “As a leader, I have confidence that I can support SEL implementation”).

Structures and Routines for SEL implementation (18 items; ɑ = 0.94;) included teaming (e.g., “Our SEL Leadership Team is effective at supporting SEL implementation”), visioning (e.g., “Our COE's shared vision guides SEL implementation”), conducting needs assessments (e.g., “Our COE has identified areas for SEL improvement based on a comprehensive needs assessment (e.g., surveys, focus groups, interviews)”), implementing plans (“Over the past 6 months, our COE has developed or revised a written SEL plan for implementation support in my region”), integrating SEL into instructions/programs (e.g., “Our COE integrates SEL practices into our COE's daily activities and routines”), and data‐based decision making for continuous improvement (e.g., “Over the past 6 months, our COE has reviewed data to reflect upon SEL improvement in my region”). An earlier version of the Structures and Routines scale was piloted with COE SEL leaders in the summer of 2022 (Jones et al., Under Review) and substantially revised for this use.

For RQ2, Partnership Variety was measured by asking leaders to report whether each of nine different partnership types (e.g., cross‐team, cross‐level, cross‐sector, cross‐region) were engaged in advancing their SEL work: (1) other COES, (2) departments within the COE; (3) districts; (4) CalHOPE focal schools (i.e., sites that were identified as receiving targeted CalHOPE supports; (5) schools other than focal schools²; (6) teachers and school staff; (7) students; (8) families; (9) community partners. Aligned with prior research, a composite measure indicating Partnership Variety was calculated by counting the partnership types endorsed.

Data analysis

RQ1: Well‐being, workplace climate, and levers of transformation

COE SEL leaders' reports of their well‐being, workplace climate, and COE levers of transformation were examined at the sample‐mean level, using intercept‐only mixed effects models. The intercept terms for these models estimated the sample‐level mean scores, adjusting for partial nesting of participants within the same counties (15 counties had a single respondent, and 34 counties had between 2 and 5 respondents). Although there is no uniform COE administrative structure statewide, it is expected that leaders from the same counties work in the same (or in coordinated) COE administrative units engaged in similar activities. Therefore, analytic models for workplace climate, partnerships, supports, and structures and routines controlled for county‐level clustering. Models for well‐being and capacities did not control for clustering, as these constructs carry information primarily at the individual level and thus are fully expected to vary from person to person. Supplementing this exploration of scale and index means, all individual items within each construct were examined at the sample mean level, to provide more specificity in determining areas of success and improvement targets.

RQ2: Partnerships and levers of transformation

Linear mixed‐effects models were used to estimate associations between the number of partnerships reported by COE respondents (i.e., Partnership Variety) and scores on the four levers of transformation (Supports Provided and Supports Received were modeled separately, resulting in five models). Models controlled for respondent race/ethnicity (dummy coded non‐Hispanic white), gender (dummy coded female), years of experience in current role (dummy coded 6+ years experience), the rural or urban status of the county where participants work (dummy coded rural), and the clustering of leaders within the same counties. Race/ethnicity, gender, and county rurality were included as covariates based on past work documenting disparities in educational experiences (e.g., Duncan & Magnuson, 2005; Musto, 2019; Roscigno et al., 2006) among these demographics. Years of experience was included because it can impact both perceptions of and influence on educational structures. Effect sizes were determined via partial eta squared (partial η2) estimates. For models that controlled for county clustering (e.g., all levers except Capacities), intraclass correlation (ICC) estimates were calculated to estimate the degree of between‐cluster variability. All analyses were conducted in RStudio (2020). Missingness ranged from 7.5% to 8.5% across all constructs; Full information maximum likelihood (FIML) estimation was used to handle missing data in regression models.

RESULTS

RQ1: Well‐being, workplace climate, levers of transformation

Mean‐level results for well‐being, workplace climate, and levers of transformation are summarized below (see Supporting Information S1: Table S2 for additional details). Raw means and standard deviations of individual items can be found in Supporting Information S1: Table S4.

Educational leader well‐being

The mean scale score for Positive Emotional Experiences (M = 3.23) indicated strong well‐being. Most respondents reported agreement (e.g., yes or YES!; mean score above 3.00) with all individual items, with item means ranging from 3.06 to 3.71 on positively worded items. Means of the two negatively worded items (reverse coded) were 2.51 for feeling nervous, anxious, or jittery, and 3.09 for feeling down, sad, or hopeless. The Resources for Coping Index (M = 3.28) also indicated strong well‐being. Most respondents reported agreement (e.g., yes or YES!; mean score above 3.00) with all individual items, with positively worded item means ranging from 3.10 to 3.61. The one negatively worded item (reverse coded) had a mean of 2.64.

Workplace climate

The mean scale scores for Safety and Connection (M = 3.15), Opportunities for Voice and Leadership (OFVL; M = 3.34), and Culturally and Linguistically Responsive Environments (CLRE; M = 2.89) indicated a positive climate overall within COE settings. At the item level, most respondents reported agreement (e.g., yes or YES!; mean score above 3.00) with individual items. Item means ranged from 2.96 to 3.46 on Safety and Connection, from 3.21 to 3.46 on Opportunities for Voice and Leadership, and 3.02 to 2.83 for CLRE. Notably, two CLRE items had relatively lower mean scores compared to the rest of the Climate items: “People consider how bias may shape their interpretations of behavior” (M = 2.83) and “People consider the impact of race and racism on experiences in school” (M = 2.83).

Levers of system transformation

After 35 months of CalHOPE there was a strong presence of levers of transformation within COEs. Respondents reported the highest mean scores in Capacities (M = 3.58), followed by Supports Provided (M = 3.39), Supports Received (M = 3.21), Partnership Activities (M = 2.98), and Structures and Routines of SEL implementation (M = 2.85). ICCs on scale scores ranged from 0.17 to 0.35. At the item level, most respondents reported agreement (e.g., yes or YES!; mean score above 3.00). Item means ranged from 2.62 to 3.22 for Partnership Activities. For Supports Received, item means ranged from 2.43 to 3.69 and for Supports Provided item mean scores ranged from 2.98 to 3.67. For Capacities, item means ranged from 3.12 to 3.93. For Structures and Routines, item means ranged from 2.44 to 3.46. Lower‐scoring items cross‐cut two general areas: centering equity in SEL implementation, and data‐based approaches to planning, decision‐making, and continuous improvement.

RQ2: Partnerships and levers of transformation

COE SEL leaders reported SEL partnerships with (1) school districts (90%); (2) CalHOPE focal schools (i.e., sites that were identified as receiving targeted CalHOPE supports; 87%); (3) departments within their COE (76%); (4) community partners (72%); (5) schools other than focal schools (72%); (6) teachers and school staff (71%); (7) students (39%); (8) other COEs (26%); (9) and families (18%). The median number of partnerships was 5.5 (Interquartile range: 4.0, 7.0). Having a greater number of partnership types was significantly (p ≤ .001) associated with higher mean scores on every lever (as reported on Supporting Information S1: Table S3): Partnership Activities (B = 0.13; partial η2 = 0.22, large effect; ICC = 0.36); Supports Received (B = 0.10; partial η2 = 0.19, large effect; ICC = 0.33); Supports Provided (B = 0.08; partial η2 = 0.17, large effect; ICC = 0.44); Capacities (B = 0.06; partial η2 = 0.13, medium effect); Structures and Routines (B = 0.018; partial η2 = 0.41; large effect; ICC = 0.49).

DISCUSSION

As explored through the SHIFT model, and CalHOPE efforts to build the capacity of the capacity builders in California K‐12 education over 35 months, we sought to better understand the conditions and activities of COE SEL leaders in statewide SEL implementation. COE SEL leaders largely reported positive well‐being, workplace climate, and enacting levers of transformation. These positive findings highlight the extent to which COE SEL leaders assess favorable conditions for providing SEL implementation support in their regions. Our findings lend empirical support to the notion that a variety of partnerships facilitate COE implementation support to advance SEL.

Our findings indicate that a variety of partnerships are in place at COEs, most commonly across levels (districts, schools, school staff), divisions (of the COE), and with community partners. More work is needed to collaborate across regions and with families and students. Research on community schools has long demonstrated the benefits of bringing these community connections into the schools (Heers et al., 2016; Oakes et al., 2017), yet we have not seen the community schools literature deeply inform the SEL literature. Community schools purposefully integrate partnerships with community organizations and families into the day‐to‐day functioning of schools. These partnerships aim to “close[ly] coordinat[e] with community resources, … actively involve parents in the educational and socialization processes, and [provide] extracurricular activities” (Heers et al., 2016 p. 1017). Thus our findings extend this literature to COEs and apply its lessons to SEL implementation.

Consistent with prior research showing that additional partnership types are needed for system change (e.g., Shapiro et al., 2015), here we find that additional partnership types are significantly associated with higher reported use of all levers of transformation. Partnership Variety expressed the largest influence on Structures and Routines (the lever least enacted). Partnership Variety was also strongly associated with partnership activities—likely because student voices cannot guide the work (the lowest Partnership Activity item) if students are not engaged as partners in the work (the antepenultimate Partnership Variety item).

Utilizing the SHIFT model, CalHOPE focused early efforts on promoting Partnerships, Supports, and Capacities. Finding lower mean scores on Structures and Routines is in line with the way that the effort was rolled out, where Partnerships, Supports, and Capacities were emphasized first. As the effort builds toward creating COE Structures and Routines for implementation support, we expect to see scores improve over time. Since the measurement of Structures and Routines was changed from the pilot assessment of Structures and Routines 1 year prior (Jones et al., Under Review), examining change over time is not currently possible—but the lag in Structures and Routines compared to the other levers persists at both time points. Specifically, we continue to see lower scores on items related to promoting equity and building data systems for continuous improvement. In the current analysis, we similarly see the lowest climate scores on experiences of Culturally and Linguistically Responsive Environments (CLREs), and lower item means on equity‐focused items embedded across different constructs. We do not know if these items began further behind at the start of CalHOPE, or whether equity‐oriented items continue to lag because equity‐oriented work is not prioritized, is uniquely hard, or takes longer to change. Previous research confirms how the school environment must “be immersed in the use of culturally responsive practices” (p. 7) schoolwide to support SEL intervention delivery in a culturally responsive manner (Barnes & McCallops, 2019), and that the status quo is upheld if equity‐oriented SEL is not a priority at the classroom, building, and district levels (Foster et al., 2022).

Grant et al. (2023) discuss how SEL implementation done with the engagement of both research and practice partners may be well positioned to advance data systems for continuous improvement. In the academic year following this survey (2023–2024), CalHOPE focal sites were supported with tools, training, and technical assistance to set up data systems that assessed and provided real‐time reporting of student and staff experiences of CLREs; this may increase these lagging indicators over time, and provide a unique opportunity to study “what it takes” to transform systems for equity. Alternatively, the use of data systems may increase without a parallel increase in equity actions, if the data is not used to make transformative changes in policy and practice (Duane et al., 2025b).

As argued earlier, education leader well‐being is needed for their engagement in systemic change‐making. Given the impacts of the COVID‐19 pandemic on educator burnout (McDaniel et al., 2024), we were pleased to see favorable well‐being and workplace climate scores in this study among COE SEL leaders. Although we do not have prior data points for comparison, we are optimistic about what this may mean for the education sector's recovery and progress. That being said, “nervous, anxious, or jittery” had the least favorable item level mean among emotional experiences (see Supporting Information S1: Table S4), which suggests another way in which there is work left to do.

Study limitations

This study sample (N = 97) is composed of relatively few individuals, but reflects substantial participation from 84% of all the COEs in the state of California. This sample may not generalize to regional providers of implementation support that are not COEs, not in the same state, or operating outside of CalHOPE. The survey administered to COE SEL leaders may be subject to social desirability bias, as well as mono‐method bias. Results should be interpreted with these limitations in mind, as we are unable to triangulate survey data with other sources of information at this time. Additionally, the BASEL‐LV is a measure to monitor engagement in a common process of improvement that took shape differently across diverse regions; it was designed to be inclusive of a broad range of activities, and therefore does not offer the precision of an implementation fidelity assessment. Further, the primary purpose of this survey was as a feedback loop for practice improvement; we were unable to measure constructs as comprehensively as we may have if data collection had been originated for research purposes. For example, Partnership Variety was very focused on nine most likely partnership types, and left out many other types of partnerships discussed in the literature (e.g., media, businesses, faith communities; Shapiro et al., 2015) that may be supporting implementation in CalHOPE. Lastly, this study does not enable us to make any causal claims about the impact of participation in CalHOPE or of Partnership Variety on COE use of the levers of transformation or downstream SEL implementation and outcomes.

Conclusions and future directions

Our work has documented the positive well‐being, workplace climate, and use of levers of transformation reported by COE representatives providing SEL implementation support. Having a variety of partnerships was associated with engaging levers of transformation, our theorized prevention support system mechanisms for advancing systemic and equity‐oriented SEL statewide. Future studies should assess the conditions for SEL implementation among leaders at other levels of the education system (e.g., district leaders, school site leaders) and across time. Future research should also use additional methods (e.g., document review, interviews) to triangulate knowledge claims emergent from this study. CalHOPE, and similar regional or statewide practice initiatives should be intentional about building equitable, data‐driven prevention systems, while applying the lessons from this study: A greater variety of partners may enable greater transformative potential, in its association with elevated COE partnership activities, supports, capacities, and structures and routines for providing SEL implementation support. County Office of Education representatives across California assess favorable conditions for serving as a prevention support system to help advance systemic and transformative SEL implementation statewide.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ETHICS STATEMENT

The Office for Protection of Human Subjects (OPHS) at the University of California, Berkeley approved of this study.

Supporting information

Supmat.

DATA AVAILABILITY STATEMENT

Data and materials cannot be provided to ensure the confidentiality of research participants.