Unfreeze‐Change‐Refreeze Model

In 1947, Kurt Lewin theorized the unfreeze‐change‐refreeze model, which argued that for change to occur, practitioners have to reject and replace prior systems. “Prior systems” are the current beliefs, practices, and routines we must change. Turns out, this is much more difficult than one would think. The book Unlearning: Changing Your Beliefs and Practices with UDL (Posey and Novak, 2020) notes how difficult it is to unfreeze simple “facts” we think we understand. For example, it is a simple fact that when there is no light, there is no color.

When you look at an apple, it looks red because wavelengths of light are reflected from the apple onto the cone photoreceptors in your eyes. You perceive the redness because of the reflection of that light. Without light, there is no color. Literally, nothing is reflected onto the cones of your eyes, so there is nothing to perceive. No electrical signal goes to your brain. Nothing is stimulated. However, regardless of these scientific facts, individuals who have been placed in a completely dark room, a room where absolutely no light can enter, still claim to see an apple as being red. They know that there is no light in the room and they have learned the science that explains how in an absence of light, there is no color. However, they still claim to see color. They use reasoning such as, “My eyes must not have adjusted to the dark yet so I could still see the redness,” or “There must have been some light that somehow got into the room.” They still see the apple as red even though it is impossible, simply because they believe the apple is red and beliefs and practices are hard to change. Too often in our systems, practices and procedures are followed because they were followed before. Consider the following responses:

• That’s how we have always done it.
• That’s what the contract says.
• No one has ever questioned this before.
• Why are we suddenly changing everything?

All of these responses come from individuals who are “frozen” in the current model. Sadly, the current model was not built for all students, and “freezing” there leads to the continuation of oppressive, inequitable, and exclusionary practices and systems.

Let’s discuss a specific example. There are silos between general and special education in many schools and districts. Students who receive special education services are referred to as special education students, or more derogatory terms (e.g., sped) as opposed to recognizing that all students are general education students first, and serving students with disabilities is the responsibility of all teachers.

Many practitioners are “frozen” in thinking about special education as a location or a program instead of programming. Terms like “my kids,” “your kids,” and “those kids” are evidence of these silos. This mindset is so pervasive in the system that many educators do not challenge it. Simply, people are “frozen” in the belief that serving students with disabilities is the responsibility of the special education department. This needs to change. Shifting mindset is challenging and painful because it means acknowledging that current beliefs do not serve all learners.

Change is possible, however, when we challenge mindsets and build systems that support collective teacher efficacy. What is problematic, however, is that when we can challenge and change mindsets, educators often “refreeze.” This frequently occurs when educators acknowledge, “I used to think ________, but now I think __________.” We need to take it a step further in our systems today and recognize, “In the future, I may think differently.”

To build inclusive and equitable systems, we must be flexible enough to know that our work is ongoing and cyclical. For these reasons, we both strongly advocate that the “freeze” is unnecessary. We have to be open to change and ready to adapt when evidence suggests there is a better way to do things. We advocate for an unlearning process that supports a cycle of reflect‐change‐growth (see Figure 1.1).

Concerns‐Based Adoption

To “unfreeze” mindsets that do not serve all learners, it is helpful to understand the concept of concerns‐based adoption and how people experience change. Hall, Wallace, and Dossett (1973) proposed this model to outline the stages that people go through as they experience change and the corresponding questions that define each stage (see Table 1.2).

When a change in an organization is introduced, people learn about this change and naturally want to know more. Let’s continue to explore the “frozen” mindset that students with disabilities are primarily the responsibility of the special education department. As we build awareness of MTSS in our organizations, it must be clear that we need inclusive practices and placements. All students, including those with significant support needs, deserve equitable access to grade‐level learning in general education classrooms with their peers.

A vision is necessary to build awareness, but it is not sufficient. All staff, including general and special educators, need a skill set that allows them to collaborate to meet the needs of all learners. All teachers need to build subject matter expertise, universally design instruction, incorporate best practices in social and emotional learning, design behavioral expectations that are culturally responsive, and design classrooms that are linguistically appropriate, culturally sustaining, and evidence‐informed. In addition to these skills, special educators tailor specially designed instruction aligned to students’ IEP goals. As we share information about MTSS, we have to be transparent about what it means to educators, their working relationships, and their job responsibilities.

As teachers learn more about the required changes, they ask, “How can I manage all this?” We must share how the system will support their collective efficacy at this stage. General and special educators need support to sustain healthy working relationships and co‐teaching models. Without ongoing professional learning, instructional coaching, common planning, and strong systems for educator evaluation, educators will feel anxious, overwhelmed, and potentially ineffective. If we want to support the reflect‐change‐growth model, we need to provide high‐quality professional learning experiences, high‐quality instructional materials, and ongoing support.

As educators and former administrators, we both know that people aren’t lining up begging for change. As we discussed with the unfreeze‐change‐refreeze model, change is painful. We have to incentivize change. Educators will ask, “Well, what happens if we do this?” We need to have an answer. For example, we can offer early adopters the opportunity to pilot new resources, opportunities to receive additional support in the form of a co‐teaching partner or instructional coaching support, and/or the opportunity to become a part of a distributed leadership team, and so on. We need to help both general and special educators recognize that change is worthwhile; to do that, we must be strategic. If we just put up a banner that reads, “Everyone Change Now. It’s the Right Thing to Do,” we can expect some significant resistance.

At the consequence stage, it is also critical that we discuss what will happen if educators do not implement changes. This is not meant to be a time to dish out threats and punishments, but we need to create a robust accountability system to support the change. We need administrators to have radical candor and say, “Implementing inclusive practices and serving students with disabilities is not optional. If you struggle with this change, we will provide instructional coaching cycles, targeted feedback, and the necessary resources to increase your efficacy.”

You may be reflecting on the Concerns‐Based Adoption Model and thinking, “There is no way we can put all of this into place right away,” which puts you in the Management Phase. Worry not! We will provide you with the necessary support and structure to move through this phase.

Implementation Science

Systems change requires us to address numerous elements. How do we create a strategic plan that addresses this? This is where the field of implementation science comes in. Implementation science is a term broadly used across several fields to describe how evidence‐based programs are put into action to produce outcomes (Fixsen et al., 2005).

The Every Student Succeeds Act (ESSA, 2015) emphasizes the use of evidence‐based activities, strategies, and interventions (collectively referred to as “interventions”). The term evidence‐based means an intervention that demonstrates a statistically significant effect on improving student outcomes or other relevant outcomes. Unless otherwise specified, evidence‐based means meets any of the four evidence levels described next. The criteria for identifying “evidence‐based” interventions based on each of ESSA’s four levels are as follows:

• Strong evidence from at least one well‐designed and well‐implemented experimental study
• Moderate evidence from at least one well‐designed and well‐implemented quasi‐experimental study
• Promising evidence from at least one well‐designed and well‐implemented correlational study with statistical controls for selection bias
• Demonstrating a rationale based on high‐quality research findings or positive evaluation that such activity, strategy, or intervention is likely to improve student outcomes or other relevant outcomes; and includes ongoing efforts to examine the effects of such activity, strategy, or intervention

The National Implementation Research Network (NIRN) conducted a meta‐analysis of more than 800 articles related to implementation practices to devise a model of implementation science, which can be simplified in the equation: effective interventions × effective implementation = improved outcomes (Fixsen et al., 2013). Ensuring that effective interventions are evidence‐based is necessary for improved outcomes, but not sufficient. As the equation suggests, we need effective implementation as well.

As consultants, we have worked with countless districts that want to change and know what effective interventions to put in place, but they struggle with creating a strategy for effective implementation. Implementation science builds on the Concerns‐Based Adoption Model by identifying specific elements, or drivers, necessary to build a multi‐tiered system that is flexible enough to provide all learners with what they need when they need it. There are three systems drivers, which include numerous elements that drive systems change:

• Leadership drivers focus on providing the right leadership strategies for different leadership challenges. A successful MTSS system requires leaders who are committed to creating inclusive environments where all students can be successful academically, behaviorally, socially, and emotionally. Important leadership considerations include scheduling, resource allocation, creating an inclusive culture and climate, and engaging all stakeholders in the work. Leadership drivers address the development of a vision, strategic planning, family and community engagement, staffing, and resource allocation. Implementation requires leaders who address the adaptive issues (such as identifying barriers and removing them and consensus building) paired with technical support (such as finding time and resources).
• Competency drivers are elements that help educators build the skill set necessary to support all learners. Competency drivers revolve around building educator capacity to affect positive student outcomes through thoughtful staffing models, offering high‐quality professional development models, research‐based coaching strategies, aligned educator evaluation models, and the development of fidelity assessments and feedback loops to monitor progress. When competency drivers are in place, educators feel prepared to design and deliver equitable and inclusive instruction academically, behaviorally, socially, emotionally, culturally, and linguistically. Teachers who receive substantial professional development—an average of 49 hours a year—can boost their students’ achievement by about 21 percentile points (Yoon et al., 2007). It should be noted that the quality of PD and its focus have more impact than the hours or “dosage” (Kraft, Blazar, and Hogan, 2018). Additionally, recent research notes that the ongoing training must be collaborative, extended over a prolonged period, and supplemented with instructional coaching (Smith and Robinson, 2020).
• Implementation drivers help systems create and sustain environments for effective educational services. These include safe schools, high‐quality instructional materials, a system that supports data‐based decision‐making, schedules that provide adequate time for common planning and student services, and creating a robust assessment framework that aligns with the overall tiered system of support.

We have used these drivers for years, forming the basis of many of our materials and resources. In this book, we layer these drivers into a framework that is more overtly equity‐focused and rooted in deeper learning. Figure 1.2 provides an image of this updated framework (Massachusetts Department of Elementary and Secondary Education, Commonwealth Consulting, and Novak Educational Consulting, 2022). To learn more about this work, created in collaboration with the Massachusetts Department of Elementary and Secondary Education, a Coherence Guidebook and companion self-assessment are publicly available on the department's website at www.doe.mass.edu/csdp/guidebook (Massachusetts Department of Elementary and Secondary Education, 2023a).

In this deeper learning model for MTSS, all drivers are present, but we have reorganized them into more student‐centered categories. In this framework, we look at the instructional design that students experience as well as their sense of belonging and agency. We are also more keenly aware of the need to integrate equitable practices into our systems design. Table 1.3 includes a crosswalk between the implementation of science‐based drivers and deeper learning systems components. The crosswalk includes all implementation drivers as well as additional elements to create a more robust and student‐centered approach.

Changing our systems is a multiyear process. Fixsen and colleagues (2005) argue that the changes take at least four years, which is why it is so critical to have a strategic action plan that supports us as we work toward a vision of inclusive and equitable MTSS. Building a system with the necessary drivers happens in stages, with districts moving from an exploration phase to installation, initial implementation, and finally full implementation (see Figure 1.3).

• Exploration. The goal of the exploration stage is to identify the need for change, learn about possible innovations that may drive change, develop a team to support the work as it progresses through the stages, and assess and create readiness for change.
• Installation. Once the team decides to move forward, the installation phase secures the support necessary to put innovative approaches into practice (i.e., pilot sites), develops feedback loops to streamline communication and measure effectiveness, and creates an action plan for initial implementation.
• Initial implementation. In this stage, teams gather data using feedback loops to measure the impact of implementation and develop improvement strategies based on the data. Implementation supports are refined based on evidence to ensure maximum impact.

  Equitable MTSS Practices	Leadership Driver	Competency Driver	Implementation Driver
  Vision
  Instructional Vision
  Instructional Design
  Curriculum Materials			X
  Equitable Practices
  Pedagogy		X
  Assessment		X	X
  Learning Environment
  Tiered Systems
  Tiered Supports			X
  Data‐Driven			X
  Access to Resources	X
  Systems and Structures
  Staff Development and Competency		X
  Structural Systems	X
  Continuous Improvement Cycles	X
  Human Resources		X

• Full implementation. In this stage, the implementation of innovations and solutions has scaled throughout the organization. Because of its documented impact on student outcomes, drivers are well‐integrated and routinely and effectively supported by ongoing professional learning, resources, and administrative support.

Improvement Science

As we build more inclusive and equitable systems through implementation science, we want to ensure that the drivers we put in place positively impact our work and move us closer to our vision of meeting the needs of all learners. First, let’s unpack the core components of systems improvement. There are six core principles of improvement outlined in Learning to Improve: How America’s Schools Can Get Better at Getting Better (Bryk, Gomez, Grunow, and LeMahieu, 2015; Carnegie Foundation for the Advancement of Teaching, 2019) and summarized here:

• Make the work problem‐specific and user‐centered. Districts and schools need to articulate the problems they are trying to solve (e.g., achievement gaps, disproportionality) by using data as well as a comprehensive self‐assessment process so it’s clear how an action plan will address areas of need.
• Variation in performance is the core problem to address. Each district and school works well for some students, but our current systems do not increase the outcomes of all learners. As a result, an MTSS implevmentation team must build a clear understanding of how the variation in performance can be addressed through thoughtful improvement planning. It is critical to remember that no single intervention or program will work for all learners. The implementation team must use data to determine what will work for specific students under a set of conditions.
• See the system that produces the current outcomes. It is hard to improve what you do not fully understand, so an understanding of root causes is critical when improvement planning. Understanding root causes allows a district or school to create a theory of action, or a hypothesis of how specific action steps will address problem areas.
• We cannot improve at scale what we cannot measure. Improving the outcomes of students requires a robust system of data that is consistently reviewed throughout the improvement process.
• Anchor practice improvement in disciplined inquiry. Implementation teams need to meet frequently to reflect on data “to learn fast, fail fast, and improve quickly” (Carnegie Foundation for the Advancement of Teaching, 2019). Having time scheduled to review data and refine improvement efforts is critical to increase the outcomes of all learners.
• Accelerate improvements through networked communities. District and school administrators must “embrace the wisdom of crowds” (Carnegie Foundation for the Advancement of Teaching, 2019). When implementation teams include multiple stakeholders and are committed to distributed leadership and elevating and celebrating the voices of all members, improvement is possible.

Keep these principles of improvement in mind as you learn more about improvement science. Meeting the needs of all learners will not happen overnight but when a committed team is committed to the process of improvement, we can create a system where all students have equal opportunities to succeed.

Improvement science is helpful as we navigate this journey. Improvement science is a methodology that focuses on adopting an evidence‐based approach that aims to improve practice using cycles of inquiry. A publication from the National Implementation Research Network notes the relationship between implementation science and improvement science (McColskey‐Leary and Garman‐McClaine, 2021, p. 2).

Implementation science and improvement science have both been leveraged to increase students’ academic, behavioral, and social emotional outcomes. Implementation science emphasizes integrating implementation supports (e.g., training, coaching, teams, leadership) to move evidence‐based practices into routine use. Furthermore, implementation science often starts with exploration activities, including identifying the need for a new practice or structures to support existing programs and creating readiness and buy‐in. On the other hand, improvement science typically begins with a specific problem of practice that requires addressing. Improvement science—a methodology that focuses on improving practice—efforts are undertaken by a data‐driven approach that aims to improve general practice, using cycles of inquiry systematically (e.g., plan‐do‐study‐act [PDSA] cycles)….

Both sciences (a) rely heavily on data to assess their respective outcomes of interest, (b) focus on improving systems, (c) simultaneously address policy and practice, (d) use improvement cycles, and (e) attend to practitioner‐level needs. Differences between improvement science and implementation science, however, should be highlighted. Primarily, improvement science is problem‐specific and user‐focused, while implementation science is context and practice concentrated. Both sciences focus on enhancing the use (adoption, implementation, and sustainment) of effective practices or programs to improve outcomes for students.

The Institute of Educational Sciences, a project of the US Department of Education (2017), notes, “Improvement science is a problem‐solving approach centered on continuous inquiry and learning. Change ideas are tested in rapid cycles, resulting in efficient and useful feedback to inform system improvements.” This comes full circle with our proposed model of reflect‐change‐growth. As leaders, we need to create rapid feedback loops to ensure that as we work toward our strategic vision, we get positive returns on our investments. We need to know what works quickly, so we can continue it while also recognizing what doesn’t work yet, so we can make necessary changes.

Improvement science helps districts understand how to focus on specific problems, introduce small measurable changes, measure the impact of these changes, and determine whether and when these changes create true improvements that should be spread more widely across the system. Improvement science is often defined by these six tasks (Aguilar, Nayfack, and Bush‐Mecenas, 2017):

• Identify specific problems
• Focus on key participants
• Attend to variation in performance (what works, for whom, under what set of conditions)
• Reflect upon the existing system that is designed to produce current outcomes (systems thinking)
• Measure processes and outcomes to assess the efficacy of strategies
• Utilize rapid Plan‐Do‐Study‐Act (PDSA) cycles to promote quick improvement

The PDSA cycle is a four‐step process that is useful in guiding continuous improvement to test an evidence‐based practice in your setting. During the planning stage, teams reflect on the problem of practice and evidence‐based solutions and create a theory of action. A theory of action provides a logical, organized set of ideas to guide the intended work of the educational system to achieve the desired results. Theories of action are commonly described using a series of “if…then” statements. For example, “If we [implement this framework], then [this will be the change in practice], so [these are the outcomes we anticipate].”

The Rhode Island Department of Education (n.d.) uses the template to share an exemplar of the theory of action: If we implement strategies to improve the culture and climate of our school, including restorative justice practices, then teachers and administrators will enhance their relationships with students and use more effective conflict resolution techniques, so students will feel more positively about their school experience and spend more time engaged in instruction.

The next step in PSDA is “do,” which requires teams to implement an action plan aligned to the theory of action and begin to collect data. Using the sample theory of action shown earlier, teams would determine how to measure student perception about their school experience (e.g., student surveys) and time spent engaged in instruction so the measures can be used to determine impact. During the study step, participants examine data and reflect on the effectiveness of the intervention. During this step, the Improvement Team compares the data with the predictions or hypotheses put forward in the theory of action to see whether there are signs of progress toward the aim or if the team needs to revise the plan accordingly.

The last step of the PSDA process is to act. This step integrates learning generated throughout the process, which can be used to adjust strategies or make changes to the plan. Often, multiple PDSA cycles are necessary to determine whether a change in practice results in intended outcomes.

Improvement science requires us to be vigilant and create PDSA cycles where we continually self‐assess, explore outcomes using multiple means, and report those outcomes in collaborative, transparent, and culturally and linguistically appropriate ways. In short, we can’t get “frozen” in our action plan if we aren’t making the desired impact on the students we serve.