Embedding Math Narratives in Professional Learning: How to Shift Beliefs and Nurture New Instructional Practices Among Math Teachers
Insights about professional development and coaching programs from Math Narrative Project field tests
This article features field tests supported by the Math Narrative Project and the Gates Foundation. These field tests were led by: Bank Street, ConnectED, Impact Florida, OpenStax, MathTrust, Partnership for Los Angeles Schools, PERTS, TeachFX, Teaching Lab, and WestEd.
In this article from the Math Narrative Project, you will learn:
- How organizations that provide professional learning and coaching programs for math teachers integrated messaging recommendations from the Math Narrative Project; and
- Insights that organizations identified about how to shift beliefs and nurture new instructional behaviors among math teachers.
Negative beliefs and emotions experienced by students learning math — like believing some people are just bad at math or feeling stressed or confused when learning math — make it difficult for them to persist in learning higher-level math. Emerging evidence from the Math Narrative Project shows that it is possible to use narrative interventions to change beliefs and behaviors in ways that support students to engage in cognitively demanding work, to persist through productive struggle, and to succeed in learning.
The Math Narrative Project, supported by the Gates Foundation, is advancing evidence-based messaging and narrative change strategies to improve math instruction and outcomes for students in grades 6-10 with a focus on Black and Hispanic students and students from lower-income households.
The Math Narrative Project uses narrative as a tool to improve students’ math learning so that more students feel motivated and supported to learn more math. Narratives are systems of stories that shape our attitudes and behaviors and help us make meaning of the world around us. Those stories are shaped by the messages that we hear regularly as well as our lived experiences. The Math Narrative Project has explored how narratives influence students, teachers, and parents, and how we can amplify positive narrative messages about math learning that empower students and challenge and replace the narratives that hold them back.
Delivering narrative messages to students alone is insufficient to shift students’ beliefs and behaviors if classroom conditions, including adult practices and the beliefs and emotions they communicate to students, contradict or fail to reinforce positive messages aimed at students. The Math Narrative Project developed nine narrative messaging recommendations, which help teachers understand their role in shaping students’ emotions and beliefs about math learning and help support teachers in making shifts in their teaching practices. To test the effectiveness of these evidence-based messaging recommendations and validate them in a real-world context with current math teachers, the Math Narrative Project convened a community of practice made up of leaders in advancing math learning and teaching. In 2025, a sub-cohort of the community of practice, experts in professional learning and coaching for math educators, led field tests to explore the efficacy of embedding messaging recommendations from the Math Narrative Project into their professional learning tools and services.
Based on these field tests, here are five insights on how narrative interventions can shift beliefs, shape positive emotions, and nurture new instructional behaviors among math teachers that are critical to student success.
1. Emotions: Supporting math teachers to navigate the deeply emotional nature of teaching math decreased defensiveness, nurtured empathy, and fostered the adoption of positive teaching practices.
Research shows that teaching math is a deeply emotional experience for educators, as well as for students. Field tests confirmed what earlier phases of research from the Math Narrative Project showed — that teachers experience both positive and negative emotions about teaching math. Professional learning programs can support teachers to navigate these emotions — leveraging their positive emotions while helping them recognize and learn how to manage their negative emotions. Evidence from field tests showed that it was especially important to communicate in a way that allowed teachers to maintain a sense of themselves as good teachers and that acknowledged the many challenges they face.
When it comes to helping teachers manage their own emotions, field tests demonstrate the importance of messaging that helps teachers feel supported to further strengthen their practices — while avoiding making them feel criticized. In field tests focused on professional learning, educators expressed a range of emotions about teaching math. Field tests demonstrated the importance of supporting teachers to manage their negative emotions about teaching advanced math.
In their field test, ConnectED embedded positive math narratives into their reciprocal coaching program. According to ConnectED, teacher defensiveness can often be a barrier in their professional development and coaching work — especially when teachers are asked to reflect on how their own assumptions and practices may negatively impact students. Their field test showed that teachers were far more open to self-reflection when informed by video-based teacher and student voices and while processing these messages with a trusted colleague. Their field test demonstrated how their reciprocal coaching program facilitates significant shifts in teachers’ approaches to building relationships with students. Teachers realized, for instance, that building connections with some students requires time outside of instruction. They even brainstormed ways of breaking through students’ “protective shells” and ways to encourage students to ask more questions.
Baseline research from the field tests indicates that teachers’ negative emotions like guilt and discomfort were especially pronounced related to race, ethnicity, and culture. MathTrust’s field test focused on shifting educators’ negative perceptions of Black students by embedding videos of Black students in their professional learning series. The Partnership for Los Angeles Schools embedded student video interview content into professional learning sessions for math teachers. Both field tests had similar findings — teachers initially expressed strong emotions like discomfort about the negative classroom experiences of Black and emerging bilingual students.
There is good news from the field tests: Professional learning sessions that embedded narrative strategies from the Math Narrative Project were able to shift these negative emotions — lowering defensiveness and nurturing greater empathy and compassion among teachers.
For instance, PERTS and WestEd incorporated Math Narrative Project messaging into animated videos that teachers encountered on PERTS’ Elevate professional learning platform as part of continuous improvement cycles focused on establishing positive learning conditions. Their field test results found that teachers felt emotionally “supported and seen” when professional learning materials authentically acknowledged their “limited time and heavy workload.” Acknowledging early in a professional development program the real-world limitations that math teachers face helps teachers to manage the very human instinct to respond with defensiveness.
Students interviewed by MathTrust also spotlighted both students’ positive and negative classroom experiences — effectively elevating students’ own voices to draw attention to teacher behaviors that make it easier as well as harder to learn. The combination of both types of stories created a sense of urgency for teachers to build better relationships with their students. In this field test, initial emotions of guilt and discomfort increasingly shifted to joy, affirmation, and renewed purpose as teachers made changes and witnessed increased student trust and engagement. In pre- and post-intervention surveys, MathTrust asked teachers about their comfort level with adapting their instructional practices to meet the needs of Black students. Before MathTrust’s intervention, only 12.5% of teachers reported that they were very comfortable making these changes. Following the MathTrust intervention, that number rose quite dramatically to 62.5% of teachers.
2. Messengers: Students and teachers were effective messengers — helping to reduce defensiveness among teachers and nurture greater compassion for their students.
Students and teachers as messengers are a powerful component of successful professional learning programs. Teachers reported that hearing directly from students and other teachers was more effective for them than top-down institutional messages or directives.
Hearing directly from students about their lived experiences — both positive and negative — in learning math created an “emotional spark,” that began to nurture empathy and compassion among educators for their students.
For instance, Malachi, a Black male fifth grader interviewed by MathTrust, shared what it’s like to feel stereotyped in the classroom. When he is struggling with a difficult math problem, he said he sometimes expresses negative emotions like anger or frustration — leading his teachers to believe he is going to cause trouble. “Just because I’m mad doesn’t mean that I’m going to do something wrong,” he explained.
Malachi shared that his ideal math teacher would teach “advanced math,” would “help with the specific questions I need help with, but who does not automatically assume that I’m going to do something wrong.”
Messenger-focused narrative interventions like Malachi’s video had a profoundly different impact on educators than reviewing statistics or assessment charts, according to MathTrust. It connected inequity to real students with names and faces, which engaged teachers on both a rational and emotional level.
Bank Street embedded student video messages in a professional learning session. Bank Street observed that hearing students describe exactly how their teachers’ actions shaped their learning experience made abstract recommendations feel “real”. This opened teachers’ minds and prompted them to take tangible actions, such as implementing new classroom community guidelines.
For instance, a 7th-grade student shared that she held back from speaking in class after making a mistake and hearing her peers’ negative reactions, which highlighted the need to normalize errors. As a positive counter example, another 7th-grade student described how her teacher’s explicit community guidelines—which included reminders to not criticize classmates and to treat mistakes as part of learning—made her feel comfortable sharing her mathematical thinking. Bank Street’s field test demonstrated the importance of math teachers normalizing mistakes as a part of learning math and establishing classroom discussion norms in order to protect student thinking during discourse.
Using a podcast to deliver messaging to teachers, Impact Florida observed that exposing teachers to student voices prompted “increased empathy and awareness” about how deeply negative math classroom experiences impact students’ ability to learn math. For instance, Michele Knowles, a high school math teacher, conducted a survey of her 9th grade students. She realized that her habit of saying, “you should already know this,” or “we learned this in the fourth grade,” was hurting their self-esteem. She changed her practice when reviewing content; rather than telling her students they should remember previous math concepts, she now asks the class to “remind me of our steps,” allowing them to demonstrate their knowledge without feeling judged if they forgot.
Hearing from students changed Michele’s beliefs and classroom practices. Hearing from Michele — and other math teachers — also helped shape teachers’ attitudes and behaviors. Teachers as messengers, especially those who modeled their own willingness to make small, yet important, adjustments to better meet the needs of their students, reduced defensiveness among teachers by demonstrating shared struggle and mutual vulnerability.
“The podcast challenged the way I introduce previously learned skills,” one teacher reflected. “I need to choose my wording carefully so as to not discourage the students before the lesson has even begun.” Another shared, “It caused me to be more introspective on how I speak to students.”
To balance the challenges, the podcast also highlighted positive classroom dynamics. The later episodes focused on concrete, authentic relationship-building practices and illustrated “what becomes possible when relationships are centered in math instruction,” showcasing the positive outcomes of supportive, relationship-driven environments.
Many of the field test partners — Bank Street, ConnectED, MathTrust, Partnership for LA Schools, PERTS, WestEd, and TeachFX — utilized short videos featuring real students and/or teachers. Some important characteristics for successful videos: Most were short (2 to 3 minutes) — edited down from longer interviews conducted with a structured interview guide designed to elicit students’ experiences and emotions learning math. Many were shot over Zoom. Together, the real messengers along with the casual, low-production quality of the videos made the messengers come across as more authentic and relatable.
It’s also important to note that coaches were effective messengers when they used their deep content knowledge as educators and proximity to classroom practice to connect narrative messages to actual instructional moves.
3. Reflection: Programs that created intentional time for self-reflection among educators shifted beliefs and behaviors and avoided triggering potential defensiveness.
Reflection is a crucial precursor to shifting deeply held beliefs and changing behaviors. Several field tests created structured opportunities for teachers to reflect on the messaging they were encountering. The reflection process creates a low-risk environment for teachers to examine their own beliefs, behaviors, and practices and the impact on their students. This step can mitigate any potential defensiveness to adopt new instructional practices. It also strengthens beliefs that all students can learn higher-level math with the right support.
As noted above, when ConnectED asked teachers directly about their beliefs and behaviors that might negatively impact students, teachers often responded with defensiveness. This is a typical human characteristic — people have a deep psychological need to see themselves as capable, good people and well-intentioned. Structured reflection — reorienting existing teacher practices around reflection — proved to be effective at reducing defensiveness. This facilitated openness to change.
ConnectED created opportunities for teachers in their professional learning program to work with a coaching partner to engage in realistic self-appraisal and productive self-reflection. In post-session surveys following their professional learning sessions, ConnectED noted changes in how teachers interpreted student behaviors as well as changes in their actions in the classroom. Rather than assuming disengaged students simply “didn’t care,” teachers began changing their daily routines to better understand and respond to student behavior. For instance, one elementary teacher began checking in on students’ morning realities—such as whether they had eaten breakfast or slept well—and used that information to adjust the length and physical format of the day’s lesson.
“When kids shut down/give up, the assumption is that they don’t care about learning math or being a good student. Kids DO CARE and WANT to do well but when they feel that they are not able to learn math or do well, they shut down/give up,” shared a math teacher following a professional learning session led by ConnectED. “And it will take a lot of poking, prodding, encouraging and working one-on-one to build up their confidence.”
Utilizing reflection also amplified the effectiveness of narrative messages. When teachers in the TeachFX field test watched videos of fellow educators sharing their authentic classroom challenges, it nurtured mutual vulnerability and acted as a catalyst for organic peer-to-peer sharing. As one math teacher noted, after reflecting on the automated feedback from the TeachFX app, “I realized that I have not been prioritizing building relationships as much as I thought. It is easier to have conversations with extroverted students, but I need to make more of an effort with introverted students who may not talk as much.”
To amplify the impact of a five-part podcast series featuring real Florida math teachers, Impact Florida utilized community events and breakout groups led by facilitators to provide teachers with structured time to process their emotional responses to the narrative messaging they encountered in the podcast. Discussing their reflections together allowed teachers to connect the stories described in the podcast to their own classrooms, build on each other’s contributions, hear how peers were grappling with similar challenges, and engage in collaborative problem-solving. Student survey data from Impact Florida showed increases in students feeling comfortable sharing their thoughts and opinions in class, as well as an increase in students agreeing that their teacher responds to student suggestions to make the class better.
4. Integration: The integration of narrative strategies into professional learning tools, products, instructional practices, and coaching structures made it easier for teachers to adopt and implement these strategies in their day-to-day work.
Field tests found success in embedding narrative recommendations directly into the tools and methods they were already using to connect with their audience. Organizations learned that integrating math narrative messaging directly into existing, trusted spaces — such as ongoing professional learning sessions and coaching structures — facilitates adoption of positive math narratives. By integrating narratives into established workflows, it transforms abstract concepts into the necessary infrastructure that enables rigorous instruction.
Integration bridges the gap between abstract beliefs and concrete instructional action—how teachers, for example, launch tasks, respond to mistakes, facilitate productive struggle, and structure math discourse so students feel safe sharing their thinking. By giving teachers tools and instructional practices they can use that align with their new, more positive beliefs and emotions about teaching and learning math, organizations helped educators translate mindset shifts into everyday classroom practice. This guidance included how to respond to a student expressing frustration when math learning becomes difficult.
TeachFX paired narrative professional learning with an automated feedback app, allowing teachers to record their instruction and immediately see how effectively they were integrating narrative strategies to prioritize relationship-building with students in their classroom within their existing time constraints.
Teaching Lab embedded narratives into existing pedagogical routines like “Lesson Internalization” and “Structured Discourse” protocols. This ensured that messages about normalizing struggle and affirming mistakes were directly tied to how teachers planned their questions and facilitated discussions.
OpenStax built math narrative recommendations directly into EduStax, their AI-driven lesson planning tool. They used prompt engineering of their AI agent to ensure that narrative recommendations like celebrating mistakes and encouraging productive struggle were organically integrated into lesson plans. This approach avoided requiring teachers to learn about narrative interventions separately and rewrite lesson plans themselves for the sole purpose of integrating more positive math narrative messaging.
5. Timing & Repetition: Sequencing and repetition of narrative messages increased their effectiveness and impact.
Teachers’ readiness for different messages evolves. Field tests demonstrated the importance of sequencing messages to match the rhythm of the school year. A single exposure to narrative messaging is rarely enough to produce the type of durable change needed to have positive impacts in the classroom.
Impact Florida found that teachers’ readiness for different messages shifts over time. They suggested beginning with the narrative recommendation to prioritize relationship-building at the beginning of the school year when many teachers are focused on establishing classroom culture. Later in the school year, once relationships were more established, teachers appeared more eager to grapple with more complex messages related to productive struggle, normalizing mistakes, student capability, and relevance. Based on their experience, they recommend designing messaging arcs that intentionally progress with the school year — sequenced and revisited over time rather than addressed in isolation.
Across the field tests, organizations learned that while a one-time exposure can create an initial emotional spark or prompt immediate reflection, shifting deeply held beliefs and translating them into lasting instructional behavior requires consistent, repeated exposure over an extended period.
MathTrust found that repeatedly playing the exact same student interview videos at different points along the journey of their four-month professional learning series deepened the impact. As teachers’ mindsets began to shift, they noticed new details in the videos and engaged in increasingly nuanced and courageous conversations.
Because finding time for lengthy professional development is difficult, TeachFX recognized the need to weave smaller doses of messaging into frequent, brief interactions. Embedding a carefully worded sentence or two of messaging into daily app workflows, lesson reports, or follow-up communications creates a sustained context that continuously reinforces the narrative.
Evidence from field tests demonstrated the importance of sequencing narrative messages across the school year and revisiting them through recurring coaching, professional learning communities, tools, and follow-up communications.
Conclusion
Narrative interventions directed exclusively at students alone are insufficient if classroom conditions and messages communicated by adults contradict those messages. Shifting teacher beliefs and behaviors is essential to creating an environment where students can persist through productive struggle and succeed in cognitively demanding math learning.
Field tests from the Math Narrative Project’s community of practice have demonstrated that this shift is possible when professional learning programs intentionally embed narrative strategies into their design and execution.
The community of practice has gathered compelling evidence to show that professional learning programs can: lower teacher defensiveness to increase uptake of positive instructional strategies; nurture compassion for students’ emotional experiences learning math; and fundamentally transform the classroom conditions to support and motivate all math learners.