By Christian Schunn

IFL co-director

Decades of research in mathematics, English, and science education have made it clear that students make substantial progress only when they experience rich learning tasks in their classes that require the students to engage in sense making; it is very difficult for teachers to arrange for strong student learning outcomes when the curriculum for students is weak. However, we are now learning that a good, cognitively demanding curriculum for students, while necessary, is far from sufficient. For example, in a recent analysis of the outcomes of 50 different sets of science curriculum materials, only the materials that included additional supports for teachers reliably produced good learning outcomes for students; the presence of extra supports for students mattered much less. And we are also learning about why the common ways of rolling out curricula rarely obtain the student outcomes that are being sought. From these new insights, districts can become smarter about how they roll out new curricula.

Teachers are (and should be) in control of the curriculum details. Building a coherent experience for students across subjects and across years requires district-level thinking. Further, finding and creating good curriculum resources is rarely a simple matter, and so coordinating at the building or district level creates opportunities for a few experienced people to put together a stronger curriculum. However, every classroom is unique in terms of the rich set of knowledge, skills, interests, and challenges the students bring. And each teacher is unique in terms of areas of strength. That is, neither students nor teachers are cookie-cutter replicas. Student learning is not an assembly-line process like putting a standard wheel on a standard car. (Interestingly, it turns out that the traditional assembly line wasn’t great for high-quality yield car manufacturing either.) So to obtain strong outcomes for students, each teacher must customize the curriculum resources provided by the building or district, aiming towards a shared vision of the goal for students, but adjusting for where the students and the teacher currently are. Further, for both conceptual understanding and motivational reasons, acknowledging and supporting professional agency in teachers builds their commitment and performance.

From curriculum fidelity to integrity. Unfortunately, not all customizations turn out to be good for students. Many research studies have documented the ways in which curricula that initially produced strong learning outcomes for students later suffered “lethal mutations” in which critical aspects were removed by teachers during the customization process. A critical bridge task or closing task might be skipped for lack of time; student sense-making activities (the most active learning ingredient) might be lost because too much scaffolding is given to students; a foundational hands-on learning experience might be removed because instructions on how to set up the hands-on activity were lost. Some of the most famous publications about growing student outcomes by leaps and bounds in real classrooms later turned into no-effect-at-all stories as things went to scale because of all these lethal mutations.

An initial reaction to such unfortunate outcomes as curriculum materials go to scale is to design for implementation “fidelity” or make the curriculum “teacher-proof.” Very detailed scripts for teaching are provided. Policy memos mandating strict adherence are circulated. Observers are sent out into the field to make sure each teacher implements the curriculum with fidelity, which is defined as following the script exactly. When framed in this fidelity-of-implementation way, the observed student outcomes can only be worse than what was initially designed. Most obviously, teachers will resist such encroachments on their professional agency, reverting to the old curriculum resources when the observers are not present. But more importantly, teachers are prevented from making the curriculum better for their students than it was initially designed if told they must use the curriculum “as is.” It is important to understand that the base materials and detailed teaching scripts tend to aim for the lowest common dominator, provide little room for building on unique strengths in a given classroom, and have no room in the plans for the basic chaos of real classrooms such as various kinds of lost teaching days.

A more productive framing than fidelity is “integrity.” A real ship maintains integrity when it is still able to function as intended after repairs—it holds water and it can still move in the desired direction. Similarly, a curriculum maintains integrity when teacher adaptations help students achieve the desired goals of each activity or unit. Changing an example to one that is more culturally relevant to the students of a class brings improvement to the student outcomes, as long as the new example is a good example of the key conceptual point the old example served. A curriculum functions to introduce new concepts from the discipline, allow students to test and refine their understandings and skills, and encourage students to develop positive attitudes about their relationship to the discipline. Each activity in the base curriculum should serve such functions, and teacher adaptations maintain integrity if they also serve the same function as that which was adapted.

Supporting teachers with educative curricula. If integrity, rather than fidelity, is the goal, how should teachers be supported? The common finding of lethal mutations means that, on average, teachers are not well positioned to meet the integrity goal as they customize. Here is where the extra supports for teachers in the curriculum play such a critical role, explaining why it was the presence of extra teacher supports, rather than the student supports, that best predicted positive student learning outcomes. A new wave of research on curricula design has shown that curricula that do well at scale are “educative” for the teachers. That is, they are designed to educate the teachers (rather than educate only the students) so that they can make productive adaptations.

A number of educative curriculum features have been identified. One common educative strategy is to turn teaching scripts into a “worked example.” Similar to an example shown to students in which a solution is worked out with explanations, a teaching worked example has annotations (aimed at the teacher) for the purposes of the student activities and the components of the teaching script. An annotation might explain what the initial demonstration is supposed to do for students or explain what a particular question given to the class might reveal about student understanding. Another common strategy is to show the relationship of particular activities within a large conceptual learning arc: What is assumed about what the students already understood from prior activities, what will be learned this time, and where will the next step go? A third common strategy is to identity ways customization might be done within particular parts of the curriculum given typically occurring classroom variation. For example, what extra resources or supports might be given to English learners around reading a complex text? What websites might provide a range of examples that could also function in similar ways but allow for customization to student varying interests and prior learning?

Teachers need supportive professional development. Even with educative curriculum materials, some teacher professional development is still required. The curriculum materials for students that offer the strongest opportunities for new growth in students typically require large changes in teaching practices and knowledge for many if not most teachers. If very small changes would have been all that were required, teachers would have long ago made those small changes on their own. For teacher learning, similar as with student learning, the curriculum materials form a base, and then a learning community (with other teachers and/or a coach or teacher leader) plays a critical role in learning so that they can successfully enact new ways of teaching.

Too much simultaneous change means teachers don’t get enough support. The need for supportive teaching professional development to go alongside large curriculum change is obvious to most people in the educational system, but the most common outcome is that teachers receive woefully inadequate support. Why does this happen? The ironically constant feature of most districts is change itself: Every single year, many new initiatives are being rolled out at once across the system (e.g., new online systems, new socio-emotional learning methods, new evaluation methods, new assessments, and new curricula). Each initiative requires time from the relatively limited professional learning calendar of leadership and teachers. So then only a small number of hours can be devoted to any given initiative. The research literature says that approximately 20 hours is needed as a minimum, whereas teachers often experience only 4 to 8 hours at a one-day district-wide event.
A little PD is rarely better than nothing. What’s an administrator to do given the large gap between what is typically needed and what can be typically afforded? First, four observations and then some strategies. The first observation is that one of the reasons that there is so much change is because of the failures of what were actually good ideas that result from attempting too much change at once. If districts that engage in high levels of change slowed some of the changes down, more of the efforts would be successful and fewer changes would be needed down the road. The second observation is that sometimes a curriculum change with only a small amount of professional development is completely equivalent to a curriculum change with no professional development at all; the small amount of professional development typically does not translate into perceptions of quality learning experiences by teachers or improved outcomes for students. Third, sometimes a curriculum change with only a small amount of professional development leads to lower student outcomes than if the curriculum change had not occurred at all. Fourth and finally, the amount of professional development that teachers need to produce to obtain improved learning outcomes appears to be related to how large the curriculum change is for the teachers in terms of their existing knowledge and skills. For curriculum changes that are within their comfort zone, an educative curriculum and no PD appears to produce good learning outcomes; whereas for curriculum changes that are far from their comfort zone, an educative curriculum and 24 hours of PD appear to be the minimum requirement for improved learning outcomes.

With these observations in mind, administrators can enact the following strategies:

  1. Eliminate “exposure” PD events that consist of one day or less for new curriculum changes. Those appear to be a complete waste of teacher PD time when that event is the only PD planned for supporting that particular curriculum change that year.
  2. Only engage in curriculum change if then new curriculum includes educative curriculum features.
  3. Align supports for change in terms of integrity of implementation rather than fidelity of implementation.
  4. Reduce the number of curriculum changes in a given year to only those that can be supported with sufficient teacher professional development.
  5. Focus teacher professional development time on those aspects of the curriculum that are most novel, rather than reviewing all the changes.

For more information on educative curriculum:

Ball, D. L., & Cohen, D. K. (1996). Reform by the book: What is—or might be—the role of curriculum materials in teacher learning and instructional reform? Educational Researcher, 25(9), Retrieved from

Beyer, C., Davis, E. A. (2009). Using educative curriculum materials to support preservice elementary teachers’ curricular planning: A comparison between two different forms of support. Curriculum Inquiry, 39(5), Retrieved from

Davis, E. A., Palinscar, A. S., Smith, P. S., Arias, A. M., & Kademian, S. M. (2017). Educative curriculum materials: Update, impact, and implications for research and design. Educational Researcher, 46(6), 293-304.

Mihalakis, V. & Petrosky, A. (2015). Collaborative professional development to create cognitively demanding tasks in English language arts. In Supovitz, J. & Spillane, J. (Eds.), Challenging standards: Navigating conflict and building capacity in the era of the common core. New York, NY & London: Rowman & Littlefield.

Pareja Roblin, N., Schunn, C., & McKenney, S. (2018). What are critical features of science curriculum materials that impact student and teacher outcomes? Science Education, 102(2), Retrieved from