The Complex Process of School Change

School change is actualized through the unique local intersection of practices, histories, knowledge, experiences, identities, and social relations (Reinholz et al., 2019; Sannino & Nocon, 2008). Although innovations are usually introduced at the school level, they are implemented by individual educators with unique perspectives (Hall & Hord, 2015). Adopting or resisting an innovation is a dynamic interaction between the individual and the larger group and can be described as a co-constructive process (Datnow et al., 1998). How an educator makes meaning of school change occurs through the lens of their particular and existing knowledge, practices, norms, attitudes, identities, social relations, and needs (Datnow et al., 1998). Due to these individual lenses, each educator may perceive differences in goals and processes of an innovation and the affordances and constraints that govern the outcomes. Although teachers and students in schools experience change as individuals, STEM as a reform tool must also be understood at the system level, where change can be leveraged (Reinholz et al., 2021).

Due to the multiplicity of perspectives and agendas, schools are sites of struggle where internal and external factors influence school change processes positively, negatively, and unpredictably (Carlone et al., 2010). Chiu et al. (2015) found histories, knowledge, experiences, identities, and social relations were not individually associated with successful supports for whole-school STEM, but that successful STEM innovations were associated with multiples of these factors. School climate is one of the strongest predictors of teacher self-efficacy (Aldridge & Fraser, 2016), a crucial factor in adopting an innovation. Aldridge and Fraser (2016) specifically identified approachable administrators and affiliation with other teachers as centrally important to a positive school climate for teachers.

Implementing the Ambiguity of STEM and PBL

STEM is increasingly accepted as an effective approach to developing higher order thinking skills (Agussuryani et al., 2022), creativity (Wan et al., 2023), and other 21st Century skills. It is also becoming an organizing principle for curriculum, school, and district change initiatives. While not universally defined, STEM is often associated with interdisciplinary and PBL curriculum, authentic contextualization, and a focus on future workforce needs (Holmlund et al., 2018; Martín‐Páez et al., 2019; Thibaut et al., 2018). The banner of “STEM” is also easily disassembled and coopted to justify existing or preferred practice (Siegel & Giamellaro, 2020) without necessarily integrating disciplines or using focused curricular approaches, such as PBL (Larkin & Lowrie, 2023). STEM is thus associated with multiple, pedagogically difficult approaches that can require major shifts for individual teachers and school systems. For our purposes here, we remain agnostic regarding a specific definition of STEM, highlighting instead the participants’ multiple definitions of the construct, bound by the common label. Indeed, the murkiness of a STEM definition is central to our findings. Please see Siegel and Giamellaro (2022) for analysis of how the district in this study co-constructed STEM for themselves.

Teachers who incorporate new STEM lessons often struggle with the competing time and policy mandates of teaching the existing content while incorporating the STEM activities (Hutner et al., 2022). While specific teacher supports such as coaching (Giamellaro & Siegel, 2023), and professional development efforts that are extended (O’Dwyer et al., 2023) or use a teachers-as-students approach (Brown & Bogiages, 2019), have been shown to improve STEM implementation, these efforts exist in the context of multi-level systems that can threaten those efforts. STEM is currently implemented in K-12 schools across various levels, ranging from single activities to whole-school adoption. Even at the whole-school level, there is wide variation in the actual focus on STEM and, therefore, students’ experiences with STEM (Vaval et al., 2019). To take on the change process of becoming a STEM School requires an awareness of what STEM should look like in a reformed school accompanied by multi-level and multi-factorial supports if the initiative is to be successful (Chiu et al., 2015; Gardner & Tillotson, 2019).

Project-Based Learning is an equally fraught label as it is applied to a wide variety of curricular and pedagogical approaches that is too broad to define clearly. Within the school district described in this case study, PBL indicates curricula designed by the educators themselves, implemented over an extended period, interdisciplinary, and solving a real-world problem, often with experts from germane fields. As with STEM, the participants each held unique visions of PBL as the district co-created their curricular vision, and participants often conflated STEM and PBL (Siegel & Giamellaro, 2022). Our stance in this study is to acknowledge these variations as a critical and realistic aspect of any school change process. We tracked a K-12 school district during the first year and a half of its conversion from a traditional to a STEM-focused district. We asked What do teachers and administrators identify as affordances and constraints to implementing district-wide STEM during the first years of a STEM conversion initiative?

Given the ambiguous nature of both STEM and PBL, and the divergent ways that practitioners have implemented them, it is clear that no single model of practice can be implemented with high fidelity and then experimentally tested against a control. Factors such as Teachers’ affective responses to STEM (Brown & Bogiages, 2019; Burton et al., 2022) and collaborative teacher supports (Kelley et al., 2020) have been shown to affect STEM implementation, but complex reforms are always dependent on many inputs, converging and clashing in lived systems (Reinholz et al., 2019). There is a need to understand how educators take up these complex reforms and then experience perceived affordances that allow the reform to move forward and constraints that teachers perceive to slow the change process. In one recent study that examined affordances and constraints of STEM implementation at the highest levels of the system, national policy, Mäkelä et al. (2023) found teacher supports, curricular flexibility, and long-term vision to be affordances while limited resources and collaboration challenges acted as constraints. The present study examined affordances and constraints at the school district level, which are likely to be experienced by other educators struggling to implement complex reforms, including STEM and PBL, but likely other multi-faceted curriculum reforms as well. In so doing, this study provides guidance to school systems intending to implement similar large-scale reforms or to researchers intending to design more controlled systems that utilize affordances and minimize constraints.

Methods

We examined how teachers perceived affordances and constraints to district-wide STEM during the early implementation phase (first 1.5 years) of a school district’s change process. Using a case study approach (Yin, 2009), we analyzed participant perspectives to identify (1) categories of affordances that allowed for successes in implementation, and (2) categories of constraints that limited implementation efforts. Pattern analysis (Miles et al., 2020) of these categories was used to infer mechanisms that afforded or constrained the STEM initiative overall.

Context

Administrators from Crawford (pseudonym), a rural school district in the Northwest United States, decided that STEM PBL was to be incorporated into every grade level, across all disciplines. We have described elsewhere how this district came to define STEM, how that intertwined with PBL, and the implications of those definitions on the implementation of the innovation (Siegel & Giamellaro, 2022). Crawford has a population of 1,500 residents, a median household income of $48,000, 12% of persons in poverty, and 79% of residents completing high school or higher (United States Census Bureau, 2018). A university partner (Authors) provided support, including grant funding to hire a STEM coach and to provide teacher professional development (PD). The specific role and outcomes of the STEM coach are described in previous work (Giamellaro & Siegel, 2018). The coach’s primary role was to support teachers’ individual and collective needs as they invested in changes to their practice to implement district-wide STEM. Individualized supports included helping the teachers to find grade-level and Next Generation Science Standards (NGSS)-aligned curriculum, modeling inquiry-based instruction with students, coaching teachers on instruction, connecting teachers to external scientists and other experts, and connecting teachers within the district. Network and qualitative analyses showed connections to external experts were powerful aspects of the coach’s role (Giamellaro & Siegel, 2023).

Collective supports for teachers included PD sessions on STEM, approaches to PBL, inquiry teaching, formative assessment, curriculum development, contextualized learning, the engineering design process, and NGSS. Eight full-day sessions were required for all faculty. The university led additional, optional on-site graduate courses. About 30% of the school faculty enrolled in these two graduate courses. About 60% of faculty also participated in two week-long summer institutes designed for teachers to learn about creating STEM and PBL curricula and work collectively with peers to design such curricula for the upcoming school year. The PD opportunities were extensive, regular, and varied in delivery mode. More complete descriptions are beyond the focus of this paper and can be found in Authors (Giamellaro & Siegel, 2018; Giamellaro & Siegel, 2023; Siegel & Giamellaro, 2022). At the district level, administrators and faculty worked to redesign the district bell schedule to accommodate STEM PBL, implemented STEM-focused professional learning communities, created venues for showcasing student STEM accomplishments, expressed an acceptance that standardized test scores might dip before they rise, and otherwise prioritized STEM PBL across the district.

The authors agreed to study the implementation through design-based research, in which collected data is iteratively used to inform and improve the initiative’s implementation (Severance et al., 2016). The researchers collected data on student and teacher outcomes, educator perceptions of the innovation and their preparedness, professional networks, and other supporting data. The researchers and leadership team met regularly to examine these data and decide how to support the unfolding STEM PBL initiative. The third author focused entirely on data collection and analysis to manage inherent bias and did not design or implement any initiative aspects.

Participants & Data Sources

Most of the educators associated with the district (Table 1) were participants in this study, including teachers (n = 35) and administrators (n = 3). Only two educators in the district opted not to participate in the study, though all educators participated in the STEM PBL initiative and some related professional development. Nine teachers, one administrator, and the STEM coach agreed to keep a written journal to chronicle their experience over the first 1.5 years. Participant journalists were representative of the district faculty and varying degrees of participation in the optional STEM PBL professional development. Journalists were prompted to record any combination of (1) what they taught, (2) student reactions, (3) reflections on their changing practice, and (4) thoughts on the unfolding STEM initiative. Journal writers were asked to write at least weekly throughout the year and were given a modest stipend.

Table 1 Descriptive Characteristics of Participants
Full size table

We also conducted open conversational interviews (Brenner, 2006; Fontana & Frey, 2000) with all participants with the goals of (1) triangulating others’ perspectives with those recorded by the journalists and (2) “understanding informants on their own terms and how they make meaning of their own lives” (Brenner, 2006, p. 357). The interviewer (third author) let the informants speak about any aspect of the initiative or their professional practice at the forefront of their mind, at times pressing for more detail (Esterberg, 2002). The gist of all interviews was recorded in field notes, with verbatim passages when required to capture participant perspectives (Brenner, 2006). The interviewer was present in the school district and conducted informal interviews almost every day. Thus, all participants were interviewed regularly throughout the year.

Analysis

Analysis proceeded as semantic excerpting, axial coding, then inferential analysis (Miles et al., 2020). All journal entries (n = 600) and interview notes (n = 491) were entered into a qualitative data analysis software platform (www.dedoose.org). Two coders (first and second authors) first examined the entries and collaboratively excerpted text in which participants identified factors that advanced the STEM PBL initiative (affordances) or those that hindered implementation of the initiative (constraints) and negotiated rules for these codes (Table 2). The two coders calibrated the coding of affordances and constraints to STEM until independent coding resulted in an acceptable agreement (Kappa = 0.77). The two coders then independently coded and regularly discussed 993 excerpts for the affordance and constraint factors. These excerpts were subsequently used as the unit of analysis. Participating educators’ excerpts were evenly distributed across grade levels, positions, journals, and interviews.

Table 2 Code Definitions
Full size table

During the first-pass coding (affordance/constraint excerpting), coders attached memos to excerpts to identify factors participants associated with affordances and constraints. Six groups of factors were collaboratively identified as regularly indicated by participants: (1) external, (2) curriculum, (3) teacher, (4) administration, (5) district, and (6) fixed factors. Additionally, participants often discussed students but as different than the other factors. Further examination determined that participants often referenced students’ motivation or attendance as an outcome of the initiative but not necessarily an influence on the initiative. Student findings were therefore identified as elements of the initiative, as they were associated and meaningful to participants, yet they did not directly impact the initiative.

A second pass through the dataset followed the same coding process as the first, applying these seven identified factors and elements as codes (Table 2). The two coders first calibrated these factor codes as before and then independently applied the codes to the excerpts. These first levels of axial coding (Miles, et al., 2020) resulted in excerpts that were coded as affordances, constraints, or both, along with multiple factors and the students element. Code co-occurrence patterns indicated focus points for further analysis.

Following Miles et al. (2020), inferential analysis proceeded using a display matrix of affordance/constraint by the seven factors. To focus on the most common patterns we first collected groups of excerpts with high code co-occurrences (e.g. teachers x affordance) and interrogated the collection through repeated readings guided by the following analytical questions (Miles et al., 2020):

  1. 1.

    What causes a factor to be associated with an affordance, a constraint, or both?

  2. 2.

    Which factors tend to co-occur within an excerpt and why?

  3. 3.

    Do participants identify some factors as more influential on affordances or constraints than others?

Patterns in the answers to these analytical questions were identified and negotiated, describing how individuals see affordances and constraints differently within the same factor and how interpersonal relationships mediate the perceptions of affordances and constraints. Answers to these questions that were consistent across multiple participants, triangulated across multiple events and sources, and not contradicted by answers generated from other excerpts are reported below (Miles et al., 2020).

Results

To answer the research question, what do teachers and administrators identify as affordances and constraints to implementing district-wide STEM during the first years of a STEM conversion initiative, we first identified 513 excerpts in which participants identified only affordances to district-wide STEM, 369 excerpts indicating only constraints, and 111 excerpts identified as both. Table 3 shows the code co-occurrence frequencies resulting from the second round of coding while Table 4 lists examples of coded excerpts. Code co-occurrence frequencies initiated focus points for inferential analysis to suggest why participants associate these factors with affordances or constraints. Each collection of excerpts associated with each factor was interrogated by the first two authors and substantive results are presented in the following sections.

Table 3 Co-occurrence frequency of second round Codes as Affordance or Constraint
Full size table
Table 4 Example Excerpts Illustrating Affordances and Constraints to STEM
Full size table

External Factors

Both teachers and administrators identified factors external to the district as affordances to implementing STEM PBL more often than as constraints (284:145; Table 3). Participants generally described these external factors as acting on curriculum and related resources as a mechanism for affording or, less often, constraining STEM implementation. These external factors did not impact students directly but changed the curriculum they interacted with.

Participants described the university STEM coach most consistently as an affordance. While some participants expressed that they did not see the coach as an affordance, none named her a constraint to implementing STEM. Participants varied in the aspects of the STEM coach that they most valued. Most valued the connectivity to external experts and other resources that she provided and the connectivity within the school district. A smaller number valued the mentorship, modeling, and instructional coaching she offered, though these teachers noted this role as being particularly impactful. We describe the actions and outcomes of the STEM coach in greater detail elsewhere (Giamellaro & Siegel, 2023).

Teachers and administrators described almost all other external factors to function as an affordance in some cases and as a constraint in other cases. Local professionals and companies, governmental organizations, and other partners were seen as affordances for developing or implementing authentic STEM projects. Experts provided STEM-specific content knowledge, venues for projects, contextualization of academic knowledge, curriculum planning support, ideas on local connections to the concepts of study, and a sense of partnership for the teachers. However, forming these partnerships, particularly without the STEM coach, was seen as a constraint. While the rural context facilitated some STEM partners (e.g., farmers, park rangers, a fertilizer company, etc.), teachers also described a lack of available external partners for a range of STEM topics due to the district's rural location (e.g., computer science, physical sciences, mathematics). Participants described parents as partners at times, notably when they acted as outside experts, but were also described as needy or as creating home lives that were detrimental to learning and particularly for project-based, intensive learning.

Teachers felt that external PD by the university partner, conferences, and other sources was generally a worthwhile affordance. Again, most of these opportunities were optional and variably utilized by different teachers. In addition to the local PD and graduate coursework, the university supported teachers to travel to other STEM schools for site visits, national conferences, and participation in other STEM-focused PD opportunities that teachers independently sought out. While most teachers reported that these opportunities were an affordance in helping them to visualize or enact STEM PBL, some teachers occasionally reported back that they were a disappointment because they highlighted practices or structures that “Crawford could never do.”

State mandates were often discussed as constraints, particularly regarding standards and required standardized testing. Elementary teachers and non-STEM secondary teachers reported that they felt conflicted between the district requirement to implement STEM PBL and state requirements to teach their discipline-specific standards. Several teachers also described an ethical commitment to teach true to their fields of study. Even teachers who accepted that science and Common Core standards could be more broadly integrated recognized that it would take a considerable amount of work to do so. It would run the risk of abandoning a lesson they knew worked well for their discipline standards. The state had adopted the NGSS a few years before this STEM initiative, though none of the teachers in the district had re-aligned to them. Following extensive PD and individualized coach supports, elementary and secondary science teachers did so as part of this initiative. Within the teachers’ journals and interviews, the NGSS were regularly referred to as a structure that helped teachers understand and plan for STEM lessons.

Participants regularly described state-level testing and student preparation for the tests as a significant constraint to implementing STEM PBL across the district. Even teachers who strongly believed in PBL worried that the approach would not prepare students well enough for the year-end tests. Administrators regularly communicated that they understood that there would be a transition period and would not hold teachers accountable for test results. Most teachers were wary of this assurance, wondering if it was true and what this would mean for the district’s reputation with the state department of education and with parents. Several teachers described feeling safer to do what they knew worked well for test results, even if that meant limited or no implementation of STEM or PBL.

Curriculum Factors

Curriculum factors often interacted with external, fixed, and district factors and student elements as the mechanism mediating whether those factors acted as affordances or constraints to the innovation (Table 3). In this way, curricula were also the primary mechanism that teachers described as most directly impacting students’ day-to-day STEM experience. Curriculum factors were also more often described as affordances than constraints (170:85, Table 3). When discussed independent of other factors, curricula were perceived as available or not, time-consuming or not, or cognitively accessible or not, and thus either an affordance or a constraint accordingly.

Teacher Factors

Teacher factors were perceived more often as affordances than constraints (243:184, Table 3). Teachers and administrators primarily described disposition and pedagogical skill as assigned to either specific individuals or generalized groups of the “good teachers” or the “bad teachers”. As would be expected, these were highly subjective assessments, and most teachers would be cast into either group by various other educators. In most cases, when teachers were discussed as affordances, and particularly constraints, to the STEM PBL initiative, the idea of good versus bad or specific dispositions were not explicitly attributed to any particular teachers, and we did not press participants to do so.

Dispositions identified by participants included “adaptable” versus “rigid”, “passionate” versus “checked-out”, “eager” versus “resistant”, “open” versus “cynical”, “collaborative” versus “territorial”, “independent” versus “needy”, “committed” versus “disinterested”, and “confident” versus “timid”. In most cases, participants described these dispositional qualities as inherent to individuals rather than specific responses to the STEM PBL initiative. Skills seen as affordances or constraints were similarly associated with teaching generally rather than STEM specifically. These included pedagogy, leadership, and technological savvy.

Despite teachers being the primary implementers of the innovation, or perhaps because of it, participants more often described teacher factors in isolation than the other factors. In other words, participants were more likely to associate teachers with affording or constraining some aspect of the STEM innovation, independent of curriculum, district, or other factors. Teachers were individually credited with taking on some aspect of STEM PBL and thereby advancing the initiative, or categorically rejecting the initiative, sending ripples of constraint throughout their immediate area of influence.

Administration and District Factors

We consider the administration and district factors together in this section. Although we as researchers coded them separately, deeper analysis of the excerpts suggested that participants closely associated these factors, for example associating district structures, policies, hiring, and spending as directly a function of administrative decisions. Further, participants may have been reticent to speak directly about specific administrators and therefore referred to the district more generally. Combined, administrative and district factors were coded as affordances in 120 excerpts and as constraints in 230 (Table 3). Although the administration launched the STEM initiative within the district, participants were three times more likely to identify administrative factors as constraints than affordances to implementing STEM (89:27, Table 3).

Affordances included planning and vision for STEM, connections to and marshaling of resources, oversight of the big picture for the innovation, enthusiasm for and movement towards the innovation, and overall setting the ethos for change. These affordances also had their corollary constraints: discordant visions within the administration, unfair or unjustified distribution of resources, ideas without training or support, and rushing decisions related to the innovation.

The most consistent perceptions of administrative constraints to the innovation involved aspects of the work climate not specific to STEM. Teachers, and even administrators, frequently reported a generally toxic climate set by the administration that included a lack of recognition for teachers, favoritism unrelated to job performance, power consolidation, a lack of transparency on decisions, and being unreceptive to feedback. While the excerpts largely represented the teachers rather than administrator views, it is worth noting that the administrators’ excerpts also identified most of these administrative constraints. Participants commonly indicated that administrative constraints also led to constraints within all other factors, except students, while they did not do so with administrative affordances.

Fixed Factors

Fixed factors such as time, schedule, state funding, etc., were described as both affordances and constraints (42:57, Table 3), but minimally and not as pivotal to the success or failure of the intervention. The mandate of state standardized testing requirements was the most commonly cited constraint to implementing STEM, as student success on the tests was seen as threatened by devoting time to STEM rather than test-specific content. While fixed factors were deemed only minimally impactful to the overall success of the implementation by study participants, they are included here as an acknowledgement of the possible strain mandated policies and factors beyond the immediate control of teachers, schools, and districts place on innovations such as this initiative.

Students

When participants mentioned students, they almost always discussed them in conjunction with the factors influencing the success of STEM PBL implementation, such as external (home lives of students) or curriculum. Despite being one of the original justifications for the initiative, participants did not ascribe success or failure of the STEM innovation as directly mediated by students. As a result, students were not considered to be impacting the success of the STEM innovation.

Interactions Across Factors

To make sense of the process of implementing STEM PBL within this school district, we first grouped affordances and constraints identified by educators into like factors and identified patterns in how these factors were perceived to influence the rollout of the initiative. This approach tells part of the story but does not account for the nuance of how these factors evolved, interacted, and influenced individuals in different ways. The identified factors interacted with, attenuated, and exacerbated each other as both affordances and constraints.

Social interactions were perhaps the most important mediating phenomenon within this multi-factorial system in change. For participants, the success or failure of the innovation seemed primarily a function of the intentions and dispositions of other people in the school district and partners outside the district. This perception seemed to eclipse funding, curriculum, structures, students, and even other individuals’ actions. Given the entire body of data from across participants, we can see that these social interactions were complex and pervasive. Participants readily attributed their own ability to enact STEM as a function of their interactions with others. However, the perception of who afforded or constrained this ability was not at all consistent across the district.

No one person was consistently a lever in either direction of enactment. It is noteworthy that interactions with the STEM coach, whose entire job was focused on STEM implementation, were generally perceived as affordances. Interactions with the administration, the drivers of the initiative, but also managers of every other facet of the district, were most often perceived as constraints. Participants described administration-associated conditions as workplace climate, resource allocation, and inconsistent messaging. While these seem unrelated to implementation, participants described them as problematic because the STEM initiative required teachers to acquire support, collaborations, and resources outside of their established professional routines. Interactions with the STEM Coach were focused on one ultimate goal: STEM PBL implementation. Interactions with administrators were multifaceted and focused on STEM but also budgets, work duties, assessment, and many other considerations that are often associated with professional tension in a school system. While this administrative wearing of many hats may have been exacerbated within the small, rural district, there may be lessons to be learned about the division of responsibility within a systemic initiative.

Another critical pattern that became clear across participants and factors is that the district-wide STEM PBL initiative amplified the district’s existing positive and negative conditions. Teachers and administrators pursuing STEM PBL activated individual skill sets, dispositions, and social interactions in new ways that repositioned them to be particularly helpful or hurtful to implementing the innovation. The educators in the district did not have fixed qualities that would allow for a formulaic prediction of how they would contribute to an innovation. Rather, participants were changing their practices and professional identities in ways that were sometimes but not always perceived as productive. The systemic changes seemed to either exacerbate perceptions of individualized affordance and constraint or bring them to the surface in a way that was not apparent before the innovation. The expectation of “doing STEM” disallowed teachers to maintain structures that may have insulated them from some of the perceived frustrations that were smoldering before the initiative was introduced, such as frustration with administration. Conversely, the STEM mandate gave other teachers or even the same teachers license to work with outside experts and otherwise innovate their curriculum and pedagogies in ways that they may not have felt allowed to in the past.

Discussion

School reform is always difficult, and the complexities of STEM and PBL are no exception as teachers navigate how to implement new ideas into the momentum of existing systems. If “STEM” is to become a cohesive construct rather than a label applied to existing but unaligned educational practice (Larkin & Lowrie, 2023; Siegel & Giamellaro, 2022), particularly if implemented at the system level, an intentional approach to implementation is required (Mäkelä et al., 2023). This is difficult as schools are known to be sites of struggle where conflicting agendas, norms, and expectations collide (Carlone et al., 2010). In the case reported here, implementing an innovation seemed to catalyze a sense of struggle as participants differentially worked toward implementing the changes, all with different understandings of what would help or hinder. As in previous work, the structure of social interactions was a determining driver of the change process (Lesseig et al., 2019; Mäkelä et al., 2023; Reinholz et al., 2019; Sannino & Nocon, 2008). In some ways, these patterns are not uncommon in a school setting (Aldridge & Fraser, 2016; Hall & Hord, 2015). However, the STEM innovation itself also seemed to either exacerbate these generalized perceptions of individualized affordance and constraint or bring them to the surface in a way that was not apparent before the innovation. In pursuing STEM PBL, participants activated individual skill sets, dispositions, and social interactions in new ways that repositioned them to be particularly helpful or hurtful to implementing the innovation.

The findings suggest that not all factors were equally impactful. As in Mäkelä et al. (2023), resources and collaboration in the form of partnerships, curriculum, and time (e.g.) were at times challenging for these participants, though these factors were also affordances for some participants in this study, suggesting that these factors are not inherently constraints to STEM implementation. Similarly, while Mäkelä et al. (2023) found flexibility and administrator involvement to be affordances, in this study, we found those factors to have variable influence. Long term vision was an affordance in both studies. Still, it is not clear if the variable impact of these factors would respond to changes in the system or if they have predictable associations with outcomes. There is a need to sort this out in future empirical research.

Systemic Change

The participants in this study indicated affordances and constraints to implementing STEM PBL, many of which acted at the system level. While district-wide change can be daunting and messy at this scale, the factors identified here provide a starting point to consider which aspects of the system need attending. These data suggest that curriculum is a good place to start, in that participants often described curricula as the mechanism through which other affordances or constraints acted. Our analysis of the data also suggested external factors, such as outside experts and partners, were generally an affordance that may have been particularly important to the change initiative’s STEM focus. Teachers were asked to implement new pedagogical approaches, add new and unfamiliar content knowledge and contextualize the knowledge in real-world projects. These external experts and resources were overwhelmingly seen as an affordance, if not a requirement, for successful STEM implementation.

Another approach to address the identified factors’ systemic interactions is to note where the entwined elements reach a functional nexus in the system (Reinholz et al., 2019). As the focus of schooling in general, students’ experience, learning, and development would be a natural nexus to examine. Given that the initiative was launched primarily as a mechanism to engage students and to redesign school as a place where student thinking, engagement, and motivation could be supported more robustly, it was surprising how rarely participants spoke about affordances and constraints in relation to the students themselves. Data presented elsewhere suggest teachers may not have had enough time or support with this STEM initiative to progress to a stage at which they were more focused on student outcomes than on logistics, management, and personal professional impact (Siegel & Giamellaro, 2022).

Teachers are the agents who consistently make small and large decisions about any innovation (Hall & Hord, 2015). As such, they also represent a lens to examine interactions at a nexus. Teachers directly impact STEM PBL implementation by deciding, for example, to enact STEM in their classrooms. Although all of the identified factors were interacting, they came to a head as teachers made daily and long-term decisions. Teacher decisions to implement STEM, or to what degree, can be analyzed to determine the suite of factors preceding these decisions. In this study, we found that teachers indicated multiple factors that either afforded or constrained their implementation in most cases. A decision to move forward on developing a STEM project may have happened as a function of the STEM coach connecting to an outside expert, the administration granting a planning day, student interest in the idea, and personal motivation to do school differently. A lack of any one of those factors could derail the whole endeavor. Conversely, a teacher deciding not to enact a STEM project due to time constraints, content knowledge, or a vague sense of mission from the administration could be affected by minor changes to one or all of those factors.

This change initiative placed significant responsibility on teachers, regardless of their motivation to participate. There is both value and risk to analyzing teacher-level affordances and constraints. Outcomes of change initiatives may be attributed solely to teacher disposition or motivation if the multi-factorial nature of these affordances and constraints are not recognized. In this case study, successes and failures were credited to teachers and were often intertwined with their perceived dispositions. This practice has much peril, as an effective teacher in one area could potentially be marked as resistant to change in a case when they are using their professional judgement to do what they believe is best for their students. However, celebrating successes could also inspire future change. This finding corroborates previous work that has found teachers’ affective dispositions to be a significant predictor of the effectiveness of STEM implementation at the system level (Brown & Bogiages, 2019; Burton et al., 2022).

Administrators represent a third nexus where the identified factors are implemented as complex interactions. The administration had an outsized role in determining how factors would operate within the district and in driving a local vision of what STEM and PBL mean. The data presented here represent teachers’ emic perspectives as they worked through a challenging change process. It is natural for leadership to become a focus of frustrations one is experiencing with change (Hall & Hord, 2015). However, school change is implemented at the individual level (Hall & Hord, 2015). Participants in this case study seemed to detect an adversarial tone from the administration they believed constrained the innovation more often than administrative vision afforded STEM implementation. As in previous work, the structure of social interactions was a determining driver of the change process (Sannino & Nocon, 2008). In such a school change process, administrators must create a supportive and positive climate to be successful (Aldridge & Fraser, 2016). The participants in this study identified this as a gap and a threat to the innovation. Although the administration had mandated and was driving the innovation, inadvertently, they may have been a significant constraint to its success, a constraint noted in other systems as well (Reinholz et al., 2021).

Individual Change

It was clear from this case study that change was evolving systemically but experienced individually. An affordance for one participant could be a constraint for another. Individual perspectives on affordances and constraints varied widely, and participants experienced different factors in different ways. This variability may be partially explained as a function of differing motivations of early adopters, resistors, and people who needed to be supported well enough to adopt (Hall & Hord, 2015). The data suggest that the reality was more complex. While we did not try to characterize educators’ motivations, they all reported both affordances and constraints. Certainly, some participants were more invested in the change process, and even this evolved over time. Any individual’s perceptions of which factor tended to serve as an affordance and which tended toward constraint were also influenced by the grade and subject they taught, their background and training, their STEM interest and STEM identity, their professional beliefs, and their relationships with others, particularly administrators. All teachers in the district, including science and math teachers, were asked to change their practice and their perceptions of their own and new disciplines. In an integrated form, STEM may be unique in asking this of teachers when implementation is at the district level.

Although there is often tension between teachers and administrators who have both common and competing concerns, in this case, tension seemed to be the greatest threat to the initiative. The administrators were highly invested in STEM PBL and made significant systemic moves to make it happen. However, these interventions did not support STEM PBL in ways that many teachers found useful or even necessary. The systemic approach was needed, but an approach that honored individual and varied needs for support may have been required to achieve more widespread investment from the teachers tasked with implementation. As with many tensions in education, a balance needs to be found between the individual and the whole. A move toward meeting individualized needs is indicated to the extent that is possible.

Implications

It is important to consider that we are limited to participants’ perspectives in this study. Our task was to find patterns across those perspectives that might indicate trends across the district and that could potentially show up in other districts trying a similar initiative. This case study could inform future school change initiatives, particularly in small, rural contexts where individuals have an outsized impact.

In future district-wide STEM initiatives, we suggest considering how each of the seven factors are going to be addressed and making a plan to leverage those factors so they serve as affordances rather than constraints. Implementors and teaching faculty, in particular, must be brought into these conversations as they may not all experience a planned approach to leveraging a factor as an affordance. There must be an acknowledgment that a singular approach may not be practical for all. However, there is also a danger in being too scattered. In this case, participants sought a more cohesive approach and messaging from the administration, yet valued individualized support, and became frustrated with their unique constraints. District-wide STEM initiatives must find a balance between groups and individual supports, and this must be planned for and monitored throughout the change process.

Conversion to STEM and school change, in general, requires two levels of intervention. We must support affordances to STEM and minimize constraints experienced across the district or unit of change. We must also attend to those affordances and constraints that individuals are experiencing. If individual teacher needs are not met, change seems unlikely. If change is not addressed cohesively, changes seem equally unlikely.

Limitations

This study was embedded in the context of a single, rural school district implementing their own vision of STEM And PBL (Siegel & Giamellaro, 2022). The participants were enmeshed in a specific context of policy, training, culture, and school systems. The study predominantly relied on participant perspectives and provided an emic view of their lived experiences. Our analyses also added the etic lens of the researchers’ interpretations and pattern-finding with the rich text of the data. Thus, the findings cannot be generalized to other systems and should be interpreted as a starting point for inquiry or practice rather than specific guidelines.

The district’s small size limited our ability to aggregate teachers for analysis. For example, there is only one math and one science teacher in the middle school and one each in the high school, making it difficult to discern whether those teachers’ perceptions were due to their discipline or some other individual characteristic or experience. In larger aggregations (e.g., elementary and secondary), we found similar variations of patterns within groups as we did across groups. Thus, these analyses represent the whole group of participants when not otherwise indicated.

Conclusion

The promise of STEM as a needed reform and promising approach to preparing the next generation of change-makers is now widely accepted (Agussuryani et al., 2022; Martín‐Páez et al., 2019; Wan et al., 2023). Because STEM is still being defined and bounded, it can catalyze tensions when visions are ambitious but murky. In this case, district-wide implementation of an innovation, STEM PBL, introduced increased complexity as practitioners renegotiated identities and re-established roles they played with their colleagues and found their footing amongst shifting district expectations. Teachers need more clarity and guidance on STEM PBL, at both the individual and district levels than might be otherwise expected for more familiar-feeling initiatives. Additionally, STEM puts new demands on financial resources and practitioner responsibility, particularly in a small school district. In this case, and likely in others, social patterns may become chaotic and lead to additional problems within the structures and culture of a school.