An exploratory analysis of Early Adopters in education innovations

Abstract

Purpose

The purpose of this study is to gain a better understanding of the conditions and motivations that influence teachers to adopt innovations.

Design/methodology/approach

Using Diffusion of Innovation theory (Rogers, 2003) and Self-Determination theory (Ryan and Deci, 2017), data from two surveys (n = 568; n = 108) and qualitative follow-up interviews of Early Adopter teachers (n = 16) were triangulated to discern relationships among their identification as Early Adopters, satisfaction of their basic psychological needs (BPN) and their implementation of an educational innovation.

Findings

Early Adopters had a positive and statistically significant relationship with the implementation of the innovation. Satisfaction of teachers’ BPN had the largest impact on innovation.

Research limitations/implications

The findings are preliminary and based on a small sample size of teachers. Reliability of the measure of BPN was not as high as the standard, but it did have the largest impact on implementation. Additional studies should explore the connections among Early Adopter teacher motivation, leadership and the satisfaction of their BPN.

Practical implications

School leaders should leverage the influence of Early Adopters to support innovation, and they should provide additional time, training and resources to supports teachers’ BPN.

Originality/value

This study examines how to identify and support Early Adopter teachers as enablers of change within schools. We know of no other studies that have used both Diffusion of Innovation theory and Self-Determination theory to understand the motivations of Early Adopter teachers.

Citation

Basileo, L.D. and Lyons, M.E. (2024), "An exploratory analysis of Early Adopters in education innovations", Quality Education for All, Vol. 1 No. 1, pp. 158-179. https://doi.org/10.1108/QEA-10-2023-0009

Purpose

Schools must innovate. For society to benefit from the rapidly evolving digital economy, students need to become proficient creators, independent and critical thinkers, strong communicators and cooperative team members. Teachers must continuously improve and expand their professional capacity to create an engaging, rigorous learning environment where students can acquire those essential skills (van Lieshout et al., 2023).

As noted by the Organization for Economic Co-operation and Development (OECD), innovation is often overlooked in education, with little data to show its impact (Schleicher, 2023). It is also difficult, often requiring systemic, second-order change (Cuban, 1988; Genlott et al., 2019; Marzano et al., 1995; Watzlawick et al., 2011). It brings with it tensions and risks that can appear to be at odds with governance of educational responsibilities (van Lieshout et al., 2023). Some innovations do not work as intended, and educators do not always know how to adapt innovations to the context of their schools. Measuring and reporting on innovation, its variety, and outcomes, is therefore essential to building knowledge of educators as they seek to improve the quality of teaching and learning in their schools (van Lieshout et al., 2023).

The purpose of this study is to contribute to the body of knowledge about educational innovation by gaining a better understanding of the conditions and motivations that influence teacher decisions to adopt innovations. This study will first provide a background regarding the diffusion of innovation in education. We will then review the study design and methodology. Finally, we will review the study findings, conclusions and practical implications for implementing innovations in education.

Diffusion of innovation in education

In this section we will outline why innovation is important in education, including how innovations are adopted and diffused; the adopter categories; and the literature regarding successful adoption of innovations by teachers. Then, we will review the background on teachers’ motivation to innovate, and we will summarize the literature regarding the supportive conditions for innovation in education. Finally, we will discuss the specific innovation for this study.

Change through innovation is challenging because it involves a paradigm shift (Marzano et al., 1995), in which the innovation is accepted or rejected as congruent with the framework of existing mental models (first order change), or the innovation changes those mental models (second order change). The notion of first- and second-order change, first defined in 1974 by Watzlawick et al. (2011), has been adopted in studies of school innovation and reform. First-order change may be seen as doing the same things we have always done, but doing them in new ways, while second-order change involves doing completely new things (Genlott et al., 2019) and transforming the way education is organized and provided (Cuban, 1988).

Within educational research literature, the terms “innovation,” “change” and “reform” are frequently used interchangeably. In their review of 30 years of studies, Ellis et al. (2023) found subtle differences in meaning, with reform more often associated with external, policy-driven change while innovation referred to individual or collective invention. They noted that few studies offered any conceptual definition or theorization of innovation, relying instead on assumptions that readers shared an implicit understanding of its meaning and effects. Using categorizations of innovation first suggested by Sternberg et al. (2003), they noted that most educational innovations were either incrementations that moved current practice forward in its current direction, redefinitions of practice, or redirections that moved practice in a new direction.

These categorizations are apparent in recent studies. Innovation in education has been defined as a change in practice and re-culturing of teachers’ beliefs and habits (Fullan, 2016), adaptation to change (Çakıroğlu et al., 2022), as trying something new in instructional practice (Dale et al., 2021) and the use of technology (Akman and Koçoğlu, 2016; Celik et al., 2014). Emergencies, such as the COVID-19 pandemic, can engender and accelerate the adoption of innovations (Çakıroğlu et al., 2022; Joshi et al., 2023. Technological change (Armstrong, 2019; Çakıroğlu et al., 2022), top-down policy initiatives (Stephenson et al., 2018; De Vocht and Laherto, 2017), new pedagogical approaches (Fitzgerald et al., 2019) and new teaching resources (Tristani et al., 2020) can catalyze innovation. Often, teachers themselves are the source and impetus behind educational innovations (Stephenson et al., 2018). Indeed, teachers are the key enablers of innovation (Dale et al., 2021).

The innovativeness of teachers is essential to improving school effectiveness and the overall agility of educational organizations (Balkar, 2015; Kozikoğlu and Küçük, 2020; Suparno et al., 2022; Swindle et al., 2022). Teachers’ willingness to adopt innovations may have an influence on student learning as well. Recent studies indicate that students whose teachers are more innovative achieve statistically significant improvements in their digital literacy scores compared to students whose teachers are less innovative (Cirus and Simonova, 2021).

Innovation is adopted and diffused

The Diffusion of Innovations theory (Rogers, 2003) explains how social change occurs and how innovation is transmitted over time within social systems. This theory identifies four main elements of diffusion:

1. an innovation;

2. communication or transmission;

3. the time during which the innovation is transmitted; and

4. the social system.

Diffusion of Innovation theory states that the rate of adoption of innovations depends upon the perceived characteristics of the innovation and its:

• Relative advantage: the degree to which the innovation improves upon current ideas or practices.

• Compatibility: the degree to which the innovation is consistent with values, past experiences, and current needs.

• Complexity: the degree to which the innovation appears difficult to understand or do.

• Trialability: the extent to which one can experiment with the innovation before committing to it.

• Observability: the degree to which the effects of the innovation are visible.

The adoption of innovations passes through predictable stages. First, people must gain knowledge of the innovation and how it works. Second, they develop either positive or negative views of the innovation which is influenced by interpersonal communication with peers. Third, people engage in activities that lead them to a decision either to adopt or reject the innovation. If they adopt it, they put the innovation to use. Finally, people seek confirmation that their decision to adopt the innovation was appropriate and justified.

Diffusion is the “process by which an innovation is communicated through certain channels over time among members of a social system” (Rogers, 2003, p. 5). For diffusion to occur, the innovation must first be implemented successfully. The degree of implementation is sometimes in the eye of the beholder, however. Those adopting the innovation may find value in aspects that were not the focus of the designers of innovation, leading to differing interpretations of the extent to which the innovation was implemented (Towndrow et al., 2010). For the purpose of this study, we defined successful implementation as teachers’ adoption of the innovation within the contexts of their own classrooms. But diffusion is not the same as implementation. For diffusion to happen, the adopter of the innovation must actively share it and influence others to adopt it.

Adopter categories.

Within each social system, people display different degrees of openness to adopt innovation. Diffusion of innovation theory (Rogers, 2003) posits categories of adopters based on their relative inventiveness and on the time it takes for them to successfully adopt an innovation. These categories are Innovators, Early Adopters, Early Majority, Late Majority and Laggards. Each classification is discussed next.

Innovators are highly attuned to information from mass media and have social networks outside the local social system. They have a high tolerance for uncertainty and setbacks. They do not depend upon the evaluation of innovation by their peers as they are the first to adopt an innovation. Early Adopters are more integrated within the local social system. They are opinion leaders toward whom others look for advice and guidance. Change agents most often seek out Early Adopters because of their strong influence on speeding adoption. The Early Adopter reduces uncertainty for others when they adopt an innovation, and they enjoy a high level of esteem among their peers. Early Majority adopt innovations just before the rest of their peers. They interact frequently with others but do not usually act as opinion leaders. They take their time to make decisions, being neither the first nor the last to adopt an innovation. Late Majority adopt innovations just after the average member of their social system. They wait until system norms have accepted the innovation before they adopt it and are influenced by peer pressure. Laggards are the last members of their social system to adopt an innovation. They do not interact with many peers, preferring to work in isolation. They are traditionalists and prefer doing what they have always done. They must be convinced that an innovation will not fail before they adopt it.

Innovators and Early Adopters are essential to the diffusion of an innovation, although they typically comprise only 16% of the population within a social system, with Early and Late Majority Adopters comprising 68%, and Laggards comprising 16% (Rogers, 2003). Figure 1 depicts this distribution.

Teachers as successful adopters of innovation

Teachers who are Innovators and Early Adopters will use unproven ideas to solve problems. Dale et al. (2021) found that they are more capable of balancing the risk of innovation with its potential to improve learning and are adept at transferring their learning about the innovation from one academic discipline to another. They also found that Innovators and Early Adopters are confident in their abilities to learn innovations on their own, are willing to devote time and effort to explore and evaluate the innovation and want to support their colleagues in applying the innovation to their teaching.

Some researchers have suggested new categorizations of teacher adopters based on attitude. De Vocht and Laherto (2017) redesigned a survey instrument to identify adopter categories at the outset of adoption based upon teachers’ interests and concerns. Similarly, Swindle et al. (2022) also suggested four categories based on the adopters’ attitude toward the innovation, the degree of fidelity with which they implement the innovation and their influence on other teachers. Coaching of teachers based on these profiles has resulted in positive shifts in the quality of implementation of innovations.

Recent research has examined the rationale teachers use to either adopt or reject an innovation (Frei-Landau et al., 2022; Joshi et al., 2023). Teachers predominantly evaluate innovations depending upon their perceptions of their usefulness and ease of use. But these perceptions also depend upon teachers’ willingness to learn the innovation independently or with assistance. Some teachers are unwilling to adopt an innovation even after they have learned about it and are aware of its potential benefits. Teachers with the highest potential as adopters are those who are motivated to try innovations, tolerant of uncertainty and self-efficacious (Dale et al., 2021; Yamamoto, 2019).

Teachers’ motivation to innovate

Teacher motivation is critical to the successful adoption of innovation. Innovation requires significant time, effort and commitment of teachers to learn and apply it to their classrooms, with a strong sense of support for their professional autonomy and competence (Gorozidis and Papaioannou, 2014). Studies of self-determination theory (SDT) (Ryan and Deci, 2017) applied to education provide many insights about the nature of motivation and how it affects teachers’ perceptions of their abilities.

SDT states that all human beings have basic psychological needs (BPN) for autonomy, competence and relatedness. Fulfillment of BPN results in greater psychological health, vitality, well-being and quality of life; thwarting of BPN leads to poor psychological health, low vitality and ill-being. For teachers, the fulfillment of their psychological needs for autonomy, competence and relatedness results in higher levels of autonomous motivation, improves their professional performance and reduces psychological stress and exhaustion (Brandisauskiene et al., 2023; Freed et al., 2021; Reeve and Cheon, 2021). Basic need satisfaction also explains both teacher job satisfaction and teacher commitment to professional learning (Jansen in de Wal et al., 2020).

The fulfillment of teachers’ BPN is also strongly associated with teachers’ motivation to engage in innovation. The sense of self-determination arising from the fulfilment of these needs contributes to sustained effort and adoption of innovation (Schellenbach-Zell and Gräsel, 2010). Therefore, the school environment needs to be supportive of teacher self-determination.

Leadership and supportive conditions for innovation

Organizational context has an influence over the way teachers respond to innovation (Swindle et al., 2022). An organizational climate that provides support and sets high expectations for teacher performance has a positive and significant effect on the innovative behavior of teachers. It empowers teachers to learn and experiment with new teaching methods (Balkar, 2015). Conditions that support teachers’ basic needs for autonomy, competence and relatedness encourage teachers’ level of innovativeness (Gorozidis and Papaioannou, 2014). Such environments are more likely to grant autonomy for teachers to make individual and collective decisions about the adoption of innovations (Akman and Koçoğlu, 2016; Celik et al., 2014).

Frei-Landau et al. (2022) observed that, as teachers progress through the phases of adoption of innovation, they often express their need for autonomy in their choices of tools, choices of peer collaborators and their desire to try things independently. They also express their need to affirm their competence by requesting modeling with feedback, opportunities for hands-on application and opportunities to adapt the innovation for their students. Relatedness may be an essential factor in teachers’ decisions to adopt innovation, as many studies reveal teachers’ strong preference for innovation in collaboration and mutual support of peers (Armstrong, 2019; Dale et al., 2021; Fitzgerald et al., 2019; Stephenson et al., 2018). Even Early Adopters are often reluctant to implement an innovation without peer support (Fitzgerald et al., 2019).

Teachers who are Innovators and Early Adopters see leadership support as the primary enabler of innovation in their schools (Dale et al., 2021). Educational leaders are key factors in creating a supportive school environment for innovation. They should be innovative and forward-thinking to create a culture of innovation in schools (Suparno et al., 2022). They must fully understand and be involved in the innovation to encourage its adoption (Tristani et al., 2020). When introducing an innovation, leaders need to make explicit how an innovation fills the gap between attaining the school’s goals and the school’s current state (Robinson, 2011). They must also understand the time it takes for an innovation to succeed and ensure teachers have the proper training. Successful leaders of innovation provide clear and specific information about the innovation and its purpose, ensure that sufficient resources and materials are available for successful implementation, provide guidance and feedback to teachers, involve all stakeholders and communicate progress throughout adoption of the innovation (Stephenson et al., 2018).

Leaders of innovation need to develop high levels of relational trust while ensuring that improved student learning remains the primary focus (Robinson, 2011). To scale educational innovations, it is essential that leaders create structures through distributed leadership that empower teachers to take risks and change their instructional beliefs and practices (Robinson, 2008). These structures must support development of the:

[…] expertise needed to ensure that the work of improvement is learned and coordinated in ways that steer educators away from practices that are ineffective and toward those that are more effective (Robinson et al., 2017, p. 35).

Distributed leadership then becomes a shared responsibility of principals and teachers to engage in continuous learning, modeling, development of expertise through inquiry and practice and reciprocal accountability (Elmore, 2004).

The innovation

In this study, we examined the perceptions of Early Adopters about their experiences with an innovation – a student-centered model of instruction (MOI). Schools use MOIs as frameworks for the way teachers plan and deliver lessons. There are multiple MOIs with each having its own syntax of conditions and expectations for learning, social structures and principles of interactions between teacher and students and among students (Joyce et al., 2009). Each model is best suited for certain learning outcomes. The focus of most of these models has traditionally been on teachers and teacher actions in the classroom – a teacher-centered model. Teacher-centered MOIs create a classroom environment that emphasizes the teacher’s voice and reinforces student passivity and compliance (Adiningrum et al., 2021; Altun, 2023). The student-centered MOI was designed to foster student agency through structures that develop students’ skills of self-regulation, critical thinking, creativity and cooperation (Dada et al., 2023). The student-centered MOI was an innovation because its focus is on the role of students and strategies students use as learners (Adiningrum et al., 2021).

The MOI introduced complexities through fundamental changes to teachers’ understanding about their instructional role. It required teachers to design instructional units that built student knowledge and cognitive skills within academic disciplines as they developed students’ intrapersonal skills, such as self-regulation and persistence, and interpersonal skills, such as collaborating and planning work. The intent of the model was to create conditions supportive of deeper learning, where students could become increasingly skilled at applying learning to new situations and novel problems (National Research Council, 2012; Pellegrino, 2015). This required teachers to develop new pedagogies linked to 21st century skills (Fullan, 2016) and acquire a new repertoire of instructional practices (Joyce and Calhoun, 2011). Some teachers, such as the Early Adopters, already had a well-aligned repertoire which allowed them to adapt the MOI to their classrooms, while others had repertoires that were not as close to the model and needed more guidance and support.

The implementation of the MOI began with a series of professional development training and coaching sessions for teachers and school leaders that started over the summer and continued periodically over the course of the year. Each school was assigned an external “school leader coach” and a “teacher coach” (also termed “field faculty”) to help teachers implement what they learned in professional development. Teachers exercised increasing professional autonomy as they became more practiced in implementing the model. Because controlling teaching practices are prevalent and culturally valued in the USA (Reeve, 2009), principals are often not well prepared to support teacher autonomy. Therefore, implementation included professional development and coaching to assist principals through the second-order change process. Both teachers and leaders looked for observable differences in their students, including students assuming increased responsibility for their own learning within student teams, using resources in higher-order cognitive tasks, and supporting each other in developing their skills as learners. Altogether, the implementation of student-centered MOI required time, practice and autonomy support for teachers.

In summary, innovation has the potential to improve the quality of teaching and student learning outcomes. Successful adoption of innovation in education is difficult work, requiring a high level of innovativeness in both teachers and their leaders as well as a supportive culture of innovation in the school. For this study, we focus our attention on Early Adopter teachers, their motivations to adopt an innovative student-centered pedagogy and how this innovation influenced their instructional practice. Furthermore, while these characteristics are key to the successful adoption of an education innovation, there have been very few, if any, which empirically investigate these relationships and Early Adopter beliefs. As such, our main research questions are as follows.

RQ1.

Can we accurately identify Early Adopters?

RQ2.

What beliefs do Early Adopters express about innovation?

RQ3.

Do Early Adopters change their practice more than others after the implementation of an innovation?

RQ4.

Does teacher motivation matter in terms of implementation of an innovation?

RQ5.

How can educators ensure a more successful implementation of an innovation?

Design and methodology

In this section we review the data collected and the methods. Our research questions aim to understand if we accurately identified Early Adopters, why some teachers adopt the innovation and change their practice and why others do not. As noted, we are using Rogers (2003) theoretical underpinnings to begin to understand how teachers impact education innovations by changing their instructional practice.

Study samples

We utilized two cross-sectional survey samples to answer the research questions. The first was a secondary data set collected from a primary source called the School Culture Survey. It included all districts and teachers that had been trained on the student-centered MOI innovation and that opted into the Spring administration of the survey (n = 568). This sample allowed us to compare Early Adopters to non-Early Adopters and answer the RQ1 and RQ3. The second set of samples provided us contextual insight into Early Adopters’ belief, motivation and how to achieve more successful implementation of an innovation allowing us to answer the RQ2, RQ4 and RQ5. Each sample is described in detail next.

The study sample from a School Culture Survey was collected from a cross-sectional survey administered over 48 days beginning in March and closing in May. The survey was administered across 1,050 teachers located in 18 schools in eight districts in the USA. We utilized survey methods from Dillman et al. (2014) which includes five contacts to elicit high response rates with participants. We calculated an overall weighted response rate of 73% (n = 729) across the schools. The lowest response rate was 30% and the highest was 100%. While 729 participants completed the survey, only 568 teachers had an instructional classroom and were included in our final study sample. The others were administrators and non-instructional teachers or coaches and they were excluded because questions were targeted towards instructional strategies used in the classroom. Those identified as Early Adopters by the field faculty were linked back to the School Culture Survey. Early Adopters represented 11% of all participants in the original sample frame, which is very close to the 13.5% reported by Rogers (2003).

The second survey was primary and collected specifically for this study. Using the Rogers (2003) definition of an Early Adopter, external field faculty identified 253 Early Adopters of the MOI innovation. The survey was administered to them over 40 days beginning in February and closing in March. Of the 253 Early Adopters, 108 completed the Early Adopter Survey (43% response rate). We surmise that this response rate is lower because teachers took it in addition to the School Culture Survey. The Early Adopter survey also included face-to-face interviews with those who agreed to participate in a follow-up video call (n = 16). Of the 108 completers, 29 agreed to participate in an in-depth interview and 16 participants completed it (55%), with 4 of the 29 canceling it due to time constraints.

Methods

Due to the nested nature of teachers within schools, Hierarchal Linear Modeling (HLM) was used to adjust for clustering of teachers within buildings (Raudenbush and Bryk, 2002). If we do not adjust for clustering, statistical significance could be inflated. We used SPSS version 29 with the Advanced Statistics module using the mixed-methods command to calculate the estimates. We chose HLM as opposed to other methods because data were collected in clusters of schools, and data were cross-sectional, so we could not infer causality. The Conditional R-Squared is used as it reflects the proportion of variance explained by both fixed and random effects. Intraclass correlation (ICC) coefficient is another measure of the proportion of total variance attributable to the group level. It is calculated as the ratio of the between-group variance to the total variance. It is a useful measure for assessing the impact of group-level predictors.

We utilized Patton’s (2014) guidance to analyze, code and categorize the qualitative data. Both the survey and the interviews asked the same the open-ended questions listed below:

Q1.

Why did you decide to adopt the student-centered MOI in your classroom?

Q2.

Since you adopted the student-centered MOI, how has your professional practice changed?

Q3.

Since you adopted the student-centered MOI, how has your relationship with students changed?

Q4.

Have you had any challenges in implementation? If so, what?

Q5.

What advice would you give to other colleagues?

Q6.

Do you have any suggestions that you feel could improve the training or coaching you have received?

Several patterns of responses and themes were apparent across questions regardless of the specific question to which participants responded. For example, challenges were reported throughout Q1 through Q6 instead of only being reported in Q4. After identifying common themes from the literature review, we analyzed all responses and coded them as positive or negative sentiments. If a respondent brought up the same theme in multiple questions, it was coded once per respondent. The following themes were coded: the five characteristics of innovation, teacher motivation and supportive conditions for implementation. The coding for each theme is described next.

The five characteristics of innovation – relative advantage, compatibility, complexity, trialability and observability (Rogers, 2003) – were coded to determine the degree to which these characteristics influenced teachers to adopt the innovation. More specifically, for relative advantage, a positively coded response was one in which the teacher identified ways in which the student-centered MOI was an improvement over their current practice, while a negatively coded response indicated that it was not an improvement. For compatibility, a positive response referred to ways in which the MOI was consistent with the teacher’s current values or beliefs about instructional best practices, and a negative response indicated that it was not consistent. Positive sentiments about complexity identified the MOI as easy to implement and negative responses indicated that it was difficult. For trialability, positive responses were those in which the teacher described attempting a small implementation of the MOI before scaling it more widely; negative sentiments stated that teachers could not try out the MOI before fully implementing it. For observability, positive responses indicated that teachers saw changes in their students’ learning behaviors and in their own professional practice; negative responses were those in which teachers remarked that there were no changes.

The second theme included coding references to the degree to which teachers were autonomously motivated or compelled to adopt the student-centered MOI. For teacher motivation, positive sentiments were those in which the teacher expressed autonomous motivation, indicating that their participation was voluntary (“I was eager to learn new skills”, “I volunteered”), while negative responses implied or stated that it was imposed (“It was mandatory”, “I was told I had to”).

Finally, the last theme referred to conditions for successful adoption. Successful adoption of the model included three characteristics: supportive leadership, time and ample training. Comments indicating a supportive leadership style were coded as positive if participants described ways that leadership was helpful or supportive in implementing the innovation, while a negative response described leadership that was a hindrance or not supportive. For time and training, respondents were counted when they mentioned needing more time or training for practice. No positive or negative sentiments were coded with needing time or additional training.

Measures.

This section describes the outcome measure, independent and control variables. The dependent variable in the HLM analysis consisted of several questions which measured the effect of the innovation on students. Items were standardized and combined into an index. The items asked participants to assess whether: students took on more responsibility because of academic teaming, students learned at a faster rate in a team than individually, the learning tasks challenged students to use higher-order thinking, and students persisted through difficult tasks. The reliability standard aims to set principles for maximum allowable measurement error. Cronbach’s alpha assesses internal consistency and can capture measurement errors that result from poor question wording. Cronbach’s alpha of 0.6 is considered the minimum reliability threshold in education sciences (United States Department of Education, 2022). To this end, the minimum reliability thresholds are highly debated and often contextual (Dunn et al., 2014; Lance et al., 2006). The Cronbach’s alpha for the outcome measure was acceptable at 0.71.

Three independent variables were included in the model. First, a dichotomous measure indicating whether the participant was an Early Adopter was included as a predictor. Second, a question asked, “How much has your practice changed this year since implementing the student-centered model of instruction?” The response categories were on a five-point Likert scale (1 = not at all to 5 = a great deal). The item was standardized for comparison purposes. We used the practice change variable as a control to isolate teachers’ perceptions of whether their practice changed as some teachers feel that their practice has changed simply because they participated in the training; however, we know that this innovation calls for second-order change. As such, we wanted to see observable differences of the innovation’s impact on students. Thus, we focused on observable student differences as the outcome measure, isolating the impact of being an Early Adopter.

For the third independent variable, we investigated whether satisfaction of teachers’ BPN influenced being an Early Adopter of the student-centered MOI. This portion of the study was exploratory in nature, as we did not find any studies which have utilized both self-determination theory and diffusion of innovation theory. It should be noted that we did not use the standard Basic Psychological Need Satisfaction and Frustration Scale (BPNSNF) (Chen et al., 2015a; Chen et al., 2015b) survey items to measure the BPN because this was a secondary data set. Instead, we utilized items that our field faculty felt addressed teacher autonomy, relatedness and competence in practice.

The items included these statements: “I exercise my own professional judgement in planning and delivering instruction” (autonomy); “I am close to my colleagues, and we are mutually supportive” and “Participation in my PLC results in a high level of teacher morale” (relatedness); and “I feel that I make a positive difference in the learning growth of my students” (competence). Items were standardized and included in an index. The Cronbach’s alpha was low (0.57) for this index. Due to the exploratory nature of the analysis, we still tested the assumption that satisfaction of teachers’ BPN influences their implementation of the innovation but note that this measure is not optimal for future studies.

Finally, there were no teacher demographic characteristics collected during the administration of the School Culture Survey. Consequently, we did not include teacher characteristics to ensure participant confidentiality so teachers could respond without fear of being identified. In the administration of our prior surveys, teachers have been skeptical of confidentiality when asked to report identifying information. This is a limitation of the study.

Findings

In this section, we will review the findings for each research question. Then, we will discuss concluding remarks and offer practical implications.

Can we accurately identify Early Adopters?

Table 1 reports the descriptive statistics below. About 21% of the Culture Survey sample were identified as Early Adopters. Recall that Early Adopters initially represented 11% of the Culture Survey sample frame. Thus, while Early Adopters represented 11% of the original sample, they also appeared more likely to participate in the survey. Mean scores are zero because indices were standardized.

Table 2 shows the correlation coefficients between the variables. Early Adopters have a small and statistically significant relationship with the student-centered MOI. Taken together, we believe that we have accurately identified Early Adopters. Additionally, other predictor variables had small to moderate and statistically significant correlations with the MOI. Surprisingly, the greatest correlation is between the BPN and the MOI at 0.456 and it is positive and statistically significant.

What beliefs do Early Adopters express about innovation?

There were 108 participants, including the 16 teachers who participated in the follow-up interviews. Table 3 below shows the frequencies of the Early Adopters whose responses included the characteristics of innovation.

Table 3 shows positive references consisted of 89% of the 55 responses about relative advantage of implementing the innovation, 85% about compatibility, 22% about complexity (ease of use), 97% about trialability and 79% about observability of the impact of the innovation. Overall, 78% of the teachers negatively referred to the complexity (difficulty) of implementing the innovation. This may reflect the challenges of second-order change and a sign that associated complexity may hinder adoption. Increasing the ease of use of the innovation may be necessary to prevent Early Adopters from abandoning it.

More specifically, the largest number of responses (109) involved the observability of the MOI, with most teachers remarking that they did see improvements in their students’ learning and classroom behaviors in responses such as these:

I've seen that my students' learning has changed. They can acquire new skills and learn how to work as a team.

There has been a noticeable climate change within our building. Our behaviors have reduced within our building as a whole. Students are more invested in their learning thus reducing disruptive behaviors. Everyone feels included and heard.

Relative advantage, compatibility and complexity had roughly the same number of responses, with most teachers responding positively about relative advantage and compatibility. For example, these responses address relative advantage and compatibility, respectively:

I decided to adopt the student-centered model of instruction in my classroom because it facilitates a student-centered learning environment. It promotes collaborative learning, and it encourages students to take ownership of their own learning. It allows students to learn from each other while practicing reflection and higher-order thinking skills.

The student-centered model of instruction aligns with my core beliefs and philosophy about teaching and learning. I've always known the importance of this instruction for myself and my students. Now we have tools and processes to develop the necessary skills to collaborate.

Nevertheless, most teachers commented negatively on the complexity of the innovation because they found it difficult to implement, for example:

Yes, implementation is overwhelming at first. It takes LOTS of training, months, before there is a natural flow among the majority of the class to work together and listen to peers.

My class is quite small, eight students, so there are times that they are opposed to teaming. I balance the day as best I can to appease all students.

The fewest responses involved the trialability of the innovation, but most of those comments were positive as teachers frequently attempted small trials of the innovative model before scaling it to a wider application in their classrooms, for instance:

Try it! Look at your practice, really look at how you are using time, space, and sounds. If you hear your voice more than your students', reflect and reframe your practice. Give yourself time and realize the students can do more than you might think.

Do Early Adopters change their practice more than others after the implementation of an innovation?

Next, a series of HLM models were run to assess whether Early Adopters were more likely to implement the innovation. Figure 2 shows the results of the first model. We found that Early Adopters were positively and significantly more likely to implement the innovation regardless of their perceptions of whether their practice changed. The Conditional Pseudo R-squared on the model was 16%. The Conditional ICC for the model was 0.05. The ICC measures the degree of dependence among observations within schools. When the intercorrelation is close to zero there is little clustering at schools. Thus, while this ICC is small, Heck et al. (2022) noted that 0.05 is often considered as a basic cutoff of evidence of substantial clustering. Nevertheless, even trivial amounts of clustering (ICC < 0.05) may still have effects on inferences when performing single-level regression (Pituch and Stevens, 2015). This suggests the need for multilevel modeling in studies particularly when survey data are gathered in clusters.

Figure 3 includes BPN as a control variable in the model to isolate the impact of the Early Adopters on the MOI. Both coefficients (Early Adopter and Change in Practice) decline in strength and the BPN becomes the largest predictor of implementation of the innovation. While not the strongest predictor in the model, it is still important to note that being identified as an Early Adopter remained statistically significant on implementation of the innovation. This model is a better fit, as the Conditional Pseudo R-squared increased to 34%. The ICC remained relatively unchanged.

In summary, Early Adopters had an impact on the implementation of the innovation regardless of their perception on the change in practice and their BPN. Furthermore, BPN was the largest predictor in the model. As such, in order get a better understanding of why, we further investigated the open-ended responses in the Early Adopters Survey for more insight particularly concerning what motivates them to implement.

Does teacher motivation matter in terms of implementation of an innovation?

The second largest number of responses (n = 89) was seen in the Early Adopters Survey for referencing motivation in terms of implementing the innovation. Table 4 shows the distribution of positive and negative sentiments.

For teacher motivation, positive responses were those in which the teacher expressed their autonomous motivation, indicating that their participation was voluntary, such as:

I wanted the opportunity for my students to experience something that was good for their future. After reading the information I found out that it was maybe something that could help me with my lessons and their development.

Negative responses implied or stated that the innovation was imposed:

It was not my decision to adopt the model of instruction in my classroom. It was by district mandate.

Considering these teachers were identified as Early Adopters, we were surprised to see that the positive and negative responses concerning autonomous motivation were nearly even, with most responses being negative. However, this may be the case simply because these were district or school wide implementations. School-wide implementations may hinder autonomy if leadership insists upon teacher compliance with implementation of the innovation. Furthermore, this finding illustrates the need to better understand how to infuse short-cycle inquiry into the intervention process to address these concerns as they arise for the successful implementation of innovations.

How can educators ensure a more successful implementation of an innovation?

Three major themes emerged as conditions for successful implementation: leadership, time and more training. For leadership, there were few responses (12), and they were predominantly negative. A positive response indicated that leadership was helpful or supportive in implementing the innovative model, for example:

I have encountered many challenges with implementation, but my administration in my building has been very supportive in adjusting things to overcome these challenges. For example, having time to plan for teaming activities and unpacking standards is a challenge. We have rearranged our Specials schedule to accommodate for additional planning time to meet with grade-level teams.

Negative responses described how leadership was a hindrance or not supportive:

Yes, I have had challenges in implementation because I have not received very much support from leaders in my school. I have received minimal lesson plan feedback as well as observational feedback although I've made consistent changes using the practices. The conclusion I've drawn is that I have more of a personal investment than the leaders at my school.

About 55% of the responses indicated that they needed more time to practice, and 54% of responses referred to the need for more training. Comments associated with time primarily involved concerns about implementation requiring more time, for example:

I wish there was more time to plan ahead for this. I feel like it was just thrown on us at the very beginning of the year, like a day or two days before students were in our classes. It has felt like we are flying by the seat of our pants at times and I believe that is because of the way we started this.

Suggestions about training covered a wide range of topics including the provision of more opportunities for applied practice:

It would have been beneficial to have more hands-on guidance and training. If an instructor was able to plan, come into the classroom, and help with a particular lesson, the teacher would be able to receive instant feedback.

For more modeling in the classroom:

One suggestion that I can give is that one of the coaches could model a lesson in a class of K-2 students and another in 3-6.

And for adaptations of the model for different disciplines, younger students, students with disabilities and English learners:

The program needs to be adapted for students that are in preschool and have special needs and are bilingual. Materials in Spanish would be very helpful. Examples (videos) need to be of preschool students and classrooms not of upper grades, as it does not relate to teaching preschool.

Overall, we found that teachers often expressed the need for autonomy in how they adopted the MOI, for modeling, for more hands-on practice, and for adaptations of the MOI for different populations of students. These themes are consistent with previous research that has identified essential elements for innovation to take root: schoolwide solidarity and commitment, ownership by the practitioners implementing the innovation, provision of adequate resources, continual opportunities for professional development, and collaboration (Fullan, 2016; Joyce, 1982).

In conclusion, we found support that the external field faculty did correctly identify Early Adopters, that the beliefs expressed by Early Adopters were much like Rogers (2003) identified in terms of the characteristics of innovation, and that regardless of the control variables included in the model, being identified as an Early Adopter had a positive and statistically significant relationship with the innovation. In sum, it appears that the external faculty have accurately pinpointed Early Adopters, and they are more likely to implement innovations. While the reliability of the BPN was not as high as the standard, it did have the largest impact on the model. In the next section, we will provide concluding thoughts and policy implications.

Conclusion and practical implications

We were able to identify Early Adopters by looking for characteristics as defined by Rogers (2003) and learned that they were statistically significantly more likely to implement the student-centered MOI than non-Early Adopters. Additionally, we found that our measure of the BPN was the largest predictor of implementation. Moreover, Early Adopters’ responses referred to all the characteristics of innovation as outlined by Rogers (2003). They believed the MOI as an innovation offered relative advantage (usefulness). Many found the MOI to be consistent with their experience and beliefs about teaching; however, they also commented on the complexity of the innovation. They valued the trialability of the MOI, with many trying out small implementations before scaling it more widely in their classrooms. Finally, they believed that implementation of the MOI led to visible and positive changes in their students’ learning behaviors.

Comments by Early Adopters were also consistent with previous studies of the influence of leaders on innovation (Dale et al., 2021; Stephenson et al., 2018; Suparno et al., 2022; Tristani et al., 2020). They stated their belief that leaders should be knowledgeable about innovation, providing guidance and resources while ensuring communication of its purposes. It is clear from their remarks that leaders can either be a help or a hindrance to implementation of innovations.

The role of school leaders in implementing this innovation was to introduce the MOI and to support implementation in their schools. To lead this innovation effectively, they needed to be fully knowledgeable about the MOI, its purpose and rationale, the changes in classroom instructional practice that it entailed and how to organize and coordinate schoolwide efforts to support these changes (Robinson, 2011; Robinson et al., 2017). Most importantly, leaders needed to know how the MOI would impact student learning and clearly convey goals to improve learning outcomes resulting from its implementation (Robinson, 2008). In some schools, principals distribute leadership to teacher teams to implement the model. In these instances, Early Adopters assumed leadership roles by sharing their practices with other teachers and providing peer feedback. This helped to spread implementation of the MOI to more classrooms as Early Adopters encouraged and supported colleagues in trying the model out.

The capacity of school leaders to support implementation and create a conducive professional climate is critical to successful implementation (Joyce and Calhoun, 2011). Consequently, leaders should participate fully in ongoing professional development and engage in the innovation alongside teachers (Joyce and Calhoun, 2015). Nevertheless, it was clear from teachers’ comments that many leaders were not well-informed about the MOI and were therefore unable to explain the rationale for the model, offer useful feedback or provide support. Instead, they depended upon instructional coaches and teachers to provide guidance. When confronted with evidence that implementation was inadequate or missing from classrooms, these leaders deflected blame on teachers as being resistant to change. This led to resentment and a sense of disempowerment from teachers who felt that their professional expertise and theoretical understanding of effective pedagogy had been discounted (Robinson, 2011).

We also found that teachers’ perception of support for their BPN – autonomy, competence and relatedness – was predictive of their implementation of the student-centered MOI. While the reliability of the BPN was not as high as the standard, it did have the largest impact on the model. The addition of the qualitative findings provided a more nuanced view of the data, however, as they were slightly more negative with teachers implementing the innovation out of compliance rather than being autonomously motivated.

Overall, the study contributes to the body of knowledge about the diffusion of innovation in the field of K-12 education by examining the importance of identifying Early Adopter teachers and supporting them as enablers of change. To the best of our knowledge, there are no studies that have used both self-determination theory and diffusion of innovation theory to understand the motivations of teachers to be Early Adopters.

As an exploratory, mixed-method study, our findings are preliminary and based on a small sample size of teachers. We attempted to isolate Early Adopters and assess whether they were more likely to implement an educational innovation. We tested the assumption that satisfaction of teachers’ BPN would influence their decisions whether to implement the innovation, but because this was a secondary data set, we did not use established scales to measure BPN and the reliability of the measure was low. Furthermore, we were unable to control for teacher characteristics or any other contexts that could impact being an Early Adopter and implementing the student-centered MOI. Another methodological limitation addresses the cross-sectional design of this study which limits our inferences on causality. While this study represents the first step of addressing the interrelations of the variables, longitudinal designs would help determine any direct impacts of Early Adopters.

As such, our findings are limited in nature and more investigation is warranted into isolating the impacts on Early Adopters on innovations. Future studies should also assess the role of motivation in implementation. Additional control variables should be used in models including but not limited to years of experience, gender and age as they could influence the adoption of an innovation. Furthermore, the role of the leader may be even more detrimental to adoption than what has previously been found in the literature. Much more investigation is warranted.

With these limitations stated, the study is significant and has important implications for educators seeking to innovate within their schools. They include:

• Understand and clearly communicate the purpose and benefits of the innovation to teachers.

• Identify Early Adopters early in the implementation and ensure they are supported.

• Leverage Early Adopters to diffuse innovation throughout the school.

• Support teachers’ autonomous motivation for implementation.

• Ensure teachers have time to learn and practice the innovation.

• Ensure training and modeling for all grades, content areas and special populations of students.

Because of the key role that Early Adopter teachers play in diffusing innovation, we recommend that school leaders identify them early and leverage their influence to help scale innovations more widely within the school. School leaders should build Early Adopters’ sense of autonomy, competence and relatedness by ensuring that they have additional time, training and resources along with supportive coaching, modeling and feedback to improve their instructional practice. MOIs should not be so rigidly implemented as to exclude or create barriers to learning or implementing but adapted to the needs and the context of the school (Joyce et al., 2009). Therefore, leaders should provide time and resources to support adaptation of the MOI to meet the needs of all students, including early childhood students, students with disabilities and students for whom the language of instruction is not their home language.

This model represented a departure from common, teacher-centered practice. It represented an opportunity to become skilled at new pedagogies and develop and refine existing skills. As identified by Joyce and Calhoun (2016), and noted by Early Adopters participating in this study, becoming a skilled practitioner in the MOI was dependent upon opportunities to:

• study and understand the rationale behind the model;

• observe the model in action;

• plan for practice; and

• receive and provide peer coaching.

As designers of the innovation and working in partnership with the educators implementing it, this study afforded us the opportunity to better understand the conditions under which teachers would and could act as Early Adopters. We saw the potential of Early Adopters as leaders of collaborative inquiry about their profession and how to improve their own schools (Joyce et al., 1999) and the circumstances that support them in this role. The mixture of qualitative and quantitative methods also revealed to us aspects of the MOI, such as the complexity of the model and how the model was introduced to schools by their leaders, that potentially worked against successful implementation and diffusion. It is clear from this study that forced implementation of innovation and a lack of leadership preparation can stifle the motivation of Early Adopters who would otherwise be the strongest proponents and accelerators of change. This knowledge is certainly helpful to researchers and practitioners as we engage together in improvement science (Bryk et al., 2015), learning and applying the lessons of experience to better implement innovations in ways that enhance learning for all students.

Figures

Figure 1.

Distribution of adopter categories

Figure 2.

Early Adopters and the innovation

Figure 3.

Early Adopters on Innovation and BPN

Table 1.

School culture survey descriptive statistics

Statistics	Early Adopter	Practice change	MOI	BPN
N	568	527	534	535
Mean	0.21	0.00	0.00	0.00
Median	0.00	−0.32	0.15	−0.07
Standard deviation	0.41	1.00	0.73	0.67
Minimum	0.00	−2.21	−2.92	−3.08
Maximum	1.00	1.58	1.68	1.45

Source: Authors’ own

Table 2.

Correlation coefficients

Statistics	Early Adopter	Practice change	BPN	MOI
Early Adopter
r	1	0.194*	0.078†	0.208*
N	568	527	535	534
Practice change
r	0.194*	1	0.083†	0.297*
N	527	527	527	527
BPN
r	0.078†	0.083†	1	0.456*
N	535	527	535	534
MOI
r	0.208*	0.297*	0.456*	1
N	534	527	534	534

Notes:

*Correlation is significant at the 0.05 level (two-tailed);

^†Correlation is approaching statistical significance at the 0.10 level (2-tailed)

Source: Authors’ own

Table 3.

Early Adopter sentiments of the characteristics of innovation

Responses	Relative advantage	Compatibility	Complexity	Trialability	Observability
Positive	89%	85%	22%	97%	79%
Negative	11%	15%	78%	3%	21%
N	55	52	55	30	109

Source: Authors’ own

Table 4.

Early Adopter sentiments of teacher motivation

Responses	Motivation
Positive	47%
Negative	53%
N	89

Source: Author’s own