The Network in Action

The DREME Network: A Multidisciplinary Collaboration to Grow the Field of Early Math Teaching and Learning

Until recently early childhood educators gave scarce attention to math. Literacy activities have been significantly more prevalent in preschool classrooms as well as at home. Similarly, early math researchers have been substantially outnumbered by early literacy researchers.

In 2007, Greg Duncan and his colleagues^[1] inspired a significant increase in attention to early math with their finding that mathematics skills at kindergarten entry were strong predictors of later academic achievement. This finding, along with consistent evidence that a substantial learning gap in math associated with family income exists before children enter school,^[2] indicated the need to support children’s math learning in the early years using developmentally appropriate and effective instructional approaches. Reinforcing interest in math was evidence that U.S. students’ math skills are consistently low compared to most other industrialized countries,^[3] and more recent evidence that learning loss related to COVID was greatest for math.^[4] The benefits of early opportunities to learn math were shown in studies finding that math engagement in home and preschool environments promoted preschoolers’ math learning.

Heising-Simons Foundation leaders decided to address the inattention to math in both research and practice. They used a variety of strategies, including funding community-based programs, curriculum development, district and school interventions, and policy advocacy. The most significant initiative, funded for a decade, involved a multidisciplinary group of researchers engaged in applied research related to early math teaching and learning. The goal was to rapidly and significantly expand the field of early math research. They reasoned that reaching this goal was not only essential for improving children’s opportunities to learn math, but would require more than individual investigators’ dispersed efforts.

The Foundation took a risk in committing long-term funding and giving considerable discretion to a group of researchers who had not previously collaborated. The Development and Research in Early Math Education (DREME) Network was not the first multidisciplinary network of researchers created to address a practical problem. The MacArthur Foundation, for example, has supported many such networks designed to address fundamental social issues in ways that contributed to significant improvements in policy and practice.^[5] They also reduced the risk by selecting Members in the key scholarly domains who were highly respected and well known to be effective collaborators and to be committed to the goals of the Network. Creating a multidisciplinary network to support and expand research is nevertheless uncommon.

In this Brief, we describe how that DREME Network was created and evolved, and what we learned about the ingredients of an effective multidisciplinary research network.

In the Beginning

In 2011, the Foundation funded a one-day meeting at U.C. Berkeley for a small group of researchers who studied math teaching and learning from preschool through high school. Sessions focused on a broad set of research efforts related to the development of children’s math skills, curriculum, pedagogy, and professional development. Few attendees had collaborated with each other. Most worked in the relatively siloed way researchers typically work. The meeting made clear that a great deal was known about how children learn math that was not influencing teaching practices broadly, but also that there was much more to learn. It also revealed researchers’ substantial enthusiasm for the opportunity to collaborate and to pool what was being learned across diverse but related topics using different research strategies. This meeting played an important role in the Foundation’s belief that multidisciplinary, collaborative research was possible and necessary to notably influence practice.

The Formation of the DREME Network

The Foundation selected a person to lead the Network after many conversations with people in the field. They sought a person who was a scientific leader in the field, who had participated in a similar network, and had exceptional academic leadership credentials. Further conversations with people in the field led to recommendations of Network Members. The core group began by listing the areas of expertise they believed were important to be represented. People who did research on early math teaching and learning, assessment, and teacher preparation and support were considered central to the endeavor. In addition, the group determined that it was important to include expertise related to 1) children with special needs, 2) children’s culture and language, 3) school and district policy, 4) general early childhood education policy; 4) program evaluation, and 5) learning in family and community contexts.

At the time the Network began, the research fields of early childhood education (ECE) and subject-matter teaching were relatively disconnected. People who studied ECE tended to examine education more globally, stressing the many dimensions of child development that needed attention–especially social-emotional and language/literacy, but also approaches to learning, gross and fine-motor skills, and health. Only a handful of researchers concentrated on teaching and learning math in children before they entered elementary school. Two prominent ECE math researchers were selected, and a third, who studied elementary math teacher education, was invited to expand her focus to include younger children. Appropriate people who addressed other domains (children with special needs, family and community learning, and culture and language) while also focused on young children and math were not found. Thus, with input from the field, a few people who did work in these three domains were asked to move from a more general focus or a focus on literacy to math as the central focus of their research. In brief, to expand the field it was necessary to ask many of the researchers with relevant expertise to shift their attention to math or to younger children. The opportunity to work with prominent colleagues on a set of issues they viewed as important was all that was needed to engage them. All of the Network members selected were also particularly keen to engage in work that would promote greater equity in children’s opportunities to learn math, a priority shared by the Foundation.

Ultimately the network was comprised of 12 members at 12 universities. Two of these members were added after several years of work to provide kinds of expertise that were needed as the work progressed.

Prior to this first meeting, the Network director sent network members a general summary of the mission and asked them to identify specific topics related to the mission that they thought were important to include in the work. The topics were grouped and organized into broader subjects, and discussed at the first meeting, where many of the members were meeting each other for the first time.

Three broad topics emerged:

classroom teaching and learning
families and community contexts for learning
coherence in learning opportunities from preschool through the early elementary grades

Network members formed project teams to further develop each of these broad areas. Most members participated in more than one project team.

Fine Tuning

Projects within these broad topics were mapped out, with each member of the network taking on a well-defined set of responsibilities. Each project team had a leader who helped keep the work on track. Fine-tuning was challenging because so many decisions had to be made by members who had differing perspectives. But the process benefited from the different backgrounds, methodological strengths, experiences in the field, and points of view. Below is a brief summary of the plan, as it evolved, for each area.

Classroom Teaching and Learning

The group decided on two somewhat different but related lines of work. The first group focused on teacher preparation and support, targeting teacher educators. Since teaching math to young children was not a significant focus of early childhood education at the time, many people who trained and supported teachers of young children lacked experience and expertise in early math teaching. This group of Network members decided to develop materials and resources for people who prepared and supported teachers of young children. Materials were developed for them to use in courses as well as in-service work. They included brief explanations of the math understanding young children need to develop and how young children develop math understanding. They also developed activities that teacher educators could use in a course or professional development program, as well as text with embedded video clips illustrating children’s thinking, handouts, and articles.

The other group targeted practicing teachers. They noted evidence suggesting that math learning activities have the added benefit of promoting the development of executive functions, which are important in both cognitive and social development. They accordingly created math activities that teachers of young children could use in their classrooms which are specifically designed to maximize executive functions as well as math learning. This same group developed strategies for supporting young children’s math learning throughout the day, including during transition times, outdoor play, snack time, and other routines.

Families and Community Contexts

The network members realized that materials available online for parents and caregivers were more likely to reach more educated and affluent families than the families living in poverty they most wanted to reach. Consequently, they saw their primary audience as people who educated and supported parents and childcare providers in low-income communities. They developed materials, many in both English and Spanish, that explained the math young children can learn, the development of children’s math thinking, and strategies for supporting their math skill development and reducing math anxiety through daily routines, games, songs, and reading picture books. The work began with a focus on families in home contexts, but increasingly involved other contexts in the community that link to families, such as libraries, playgrounds, and pediatric offices.

P-3 Coherence

Network members and the Heising-Simon Foundation were committed to informing efforts to create greater coherence in children’s math learning opportunities from preschool through the early grades, in addition to promoting more effective instruction. Evidence showing, for example, that kindergarten repeated much of what children learned in preschool suggested that schools were not maximizing children’s opportunities to grow their math skills.^[6]

Network members recognized that teaching at the classroom level is substantially affected by school district policies. The vision for math teaching, the curriculum and assessments used, the professional development provided, and other district policies have a profound effect on what and how teachers teach. All such policies can either support or hinder coherence and quality in math instruction from preschool through the early grades. The research thus examined the ways district policies and practices affected children’s math learning opportunities and math learning in an in-depth, longitudinal study of two districts that were endeavoring to promote more effective and coherent early math instruction. The study, which followed children from preschool through second grade, included interviews with district and school administrators and teachers, classroom observations, and assessments of children’s math skills and attitudes. The goal was to derive lessons about effective (and ineffective) district strategies. The project also produced tools, such as a math assessment instrument and a classroom observation measure, that could be used widely in research on early math.

Underlying Principles

Guiding all of the work was a set of principles to which all members subscribed. To some degree, members were selected because their previous work aligned with these principles.

Engage policy makers and practitioners meaningfully and respectfully at all levels of the work–from developing ideas and plans to disseminating findings and resources;
Integrate research into product development through an ongoing iterative process in which the value (usability and usefulness) to the user audience is assessed throughout the development process;
Work collaboratively with organizations in the community to identify needs, disseminate products, and acquire feedback on their usability and usefulness;
Prioritize research, development, and dissemination that benefits vulnerable populations, including children with special needs, multi-language learners, children of color, and children living in poverty.

Process

The Network Members invited graduate students and post-doctoral fellows (referred to as Network Affiliates) to participate in the work, thus expanding the number of researchers focused on early math who learned to do research according to the principles listed above. Over the 10 years the Network was funded, multiple relatively new researchers participated in DREME work and then moved on to other positions in various settings (e.g. universities, colleges, direct service, community organizations) where they support equity and excellence in early math learning and teach others, including new researchers in the field of early math, how to engage in ways that support this goal.

Network members, along with some affiliates, met four times a year (except during COVID) for a day and a half. They typically met at Network Member’s universities, taking advantage of university space and resources and facilitating the participation of Network Affiliates at that University.

Network meetings were organized to provide whole-group time as well as time for project teams to work. Occasionally researchers, practitioners, and policy makers in the geographical area of the meeting were invited to talk about their work that overlapped with the work of the DREME Network. Guests enriched the thinking of network members and the conversation sometimes led to new collaborations.

In addition to these whole Network face-to-face meetings, project teams met regularly on Zoom to move the work forward. Project members also met with community or other organization leaders they collaborated with. Most of the work was done within Members’ universities, often in collaboration with teams of Network Members at other universities.

From the beginning, Network members, affiliates, and partners published work and made presentations at conferences and other convenings. The conference presentations allowed members to disseminate the work they were doing to academic and broader audiences, to learn from and establish connections with researchers and practitioners working in the field, and in some cases to develop new collaborations.

Quality Control

Research was woven through all the Network’s work to develop materials and resources for practitioners. For example, a study was conducted to examine how small group activities developed by Network members were actually used in classrooms to support student learning. The study provided information for revisions in the activities. A great deal of qualitative work was done to inform tweaks to products and instruments. There were also many rigorous quantitative studies that measured fidelity, reliability, validity, and sometimes impact. For example, one study used a random assignment design to assess the effects of different kinds of picture books (one focused on counting and the other embedding counting in a story) on children’s number knowledge. Although outside experts and practitioners were often involved in evaluations, Network Members themselves were responsible for designing and interpreting research. An alternative that could be beneficial is for the funder to support people outside the Network to evaluate some of the work, especially the use of materials in the field.

In a few cases, practical tools that were initially developed for research were adapted to be useful to practitioners. For example, a classroom observation measure that was created for the study of P-2 instructional coherence was subsequently revised to be a practical tool that coaches could use to help teachers improve their practice.

Each team had a process for reviewing and approving materials. Usually at least one, and often two other members of the team reviewed material, which was then revised and sometimes re-reviewed. Because the target audience included non-researchers, depending on the particular audience, parents, parent educators, early childhood educators, teachers, and administrators were also involved in the review process. In addition, the resources were tried out in authentic contexts to make sure instructions were clear and the activity worked as intended.

Lessons Learned

Getting the right mix of people involved is critically important. Because this network was focused on improving educational practice all of the members had a track record of working with and doing work that was relevant to educational practitioners. An alternative is to include as Network members people who are in the roles of the target audience, e.g., practitioners. This is a reasonable strategy to keep practice front and center. In the case of the DREME Network, the researchers were all involved in practice, and practitioners’ voices were also introduced through the many community and program collaborators who worked with the Network members.

Below is a list of other qualities of Members that facilitated the work of the DREME Network.

Distinction in their field: There can be a mix, but a sufficient number of participants who are well respected in the field gives the Network early credibility and can be used to attract funding and collaborators.
Commitment: Participants made a long-term commitment to the work of the Network, which meant abandoning some of the individual work they had planned and moving in a slightly different direction (such as studying early math instead of early literacy). The Network work was a priority.
A desire to learn: The purpose of a multidisciplinary team is to ensure that the work is informed by theory, research, and tools from different disciplines and fields of work. Network members wanted to grow in their own understanding of problems and solutions, and to use methods that were new to them.
Flexibility: Participants were flexible; they understood that every decision involved some negotiation with collaborators, which meant they would not always get their way. They also had to navigate roadblocks (e.g., Covid, district leadership changing their plan) – to be able to move to plan B (or C) without becoming discouraged.
Humility: Everything written or created was reviewed and critiqued by others. Participants had to be comfortable with and responsive to critical feedback.
A good mentor: Interdisciplinary networks offer extremely good opportunities for new researchers to develop their interests, values, and skills — interacting with a broader set of seasoned researchers than they would ordinarily have at their own university. Network members saw mentoring a new generation of early math researchers as a core part of their work.
A good sense of humor: This is hard work, and participants are in it for the long haul. They need to be able to have fun and use humor occasionally to diffuse conflicts.
Committed leadership: A leader who is committed to the benefits of interdisciplinary work and to research-practitioner collaborations, and who is able to hold people accountable without making them feel managed.

Other lessons learned include:

Being proactive in identifying potential collaborations and opportunities to share the work is necessary for the work to have any impact. The DREME Network had a staff member who oversaw the development of the website and used social media to push out information about the work as it evolved. She also monitored conferences of organizations that might be interested in the work, and supported Members in developing proposals and presentations. Members were also responsive to invitations to make presentations, write blogs, and the like.

Collaborative work among researchers necessarily brings to the fore issues of responsibility and authorship, as well as processes for review. The project teams varied in how formalized their expectations or rules were. What was clear was the need for a shared understanding. This is especially important where students or more junior colleagues are involved. They are early in their careers and authorship is critically important to their advancement. It is also important when a group of researchers are collecting and sharing data. Clarity about what any one researcher can do (e.g., presentations, publications) with shared data is needed to avoid conflict among Members.

The Foundation’s long-term commitment with ongoing accountability was critically important. The Network was required to submit an initial proposal, and then to submit a biennial report on progress made toward the goals described in the previous proposal and on the plan for the next two years. Teams found supplementary funding for some projects from other sources. But the expectation that Heising Simons Foundation funding was likely to continue for 8-10 years allowed the Network to take on ambitious projects, to take the time to assess thoroughly the resources created, to establish enduring partnerships with practitioners and other organizations, and to train a large cohort of students and postdocs who are continuing to deepen and expand the field of early math. Ongoing accountability ensured that Members continued to focus on the work of the Network and it guaranteed regular reflection and conversation among Network Members about what had been accomplished and what were productive next steps, taking into consideration new developments in the field that could inform new directions. The flexibility of the Foundation allowed the Network to make appropriate changes in the work plan that took advantage of new collaborations and opportunities in the field.

^[1] Duncan, g.J., Dowsett, C.J., Claessens, A., Magnuson, K., Huston, A.C., Klebanov, P., Pagani, L.S., Feinstein, L., Engel, M., Brooks-gunn, J., Sexton, H., Duckworth, K., & Japel, C. (2007). School readiness and later achievement. Developmental Psychology, I(6),1428-1446.

^[2] National Academies of Sciences, Engineering, and Medicine. (2019). Monitoring Educational Equity. Washington, DC: The National Academies Press. https://doi.org/10.17226/25389.

National Academies of Sciences, Engineering, and Medicine. 2019. Monitoring Educational Equity. Washington, DC: The National Academies Press. https://doi.org/10.17226/25389.

^[3] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/53f23881-en.

^[4]Kuhfeld, Megan, James Soland, and Karyn Lewis. (2022). Test Score Patterns across Three COVID-19-impacted School Years. (EdWorkingPaper: 22-521). Retrieved from Annenberg Institute at Brown University: https://doi.org/10.26300/ga82-6v47

^[5] https://www.macfound.org/programs/pastwork/research-networks/

Rose, R. (2000) FINDING ANSWERS TO BIG QUESTIONS: OVERCOMING DISCIPLINARY BOUNDARIES THROUGH RESEARCH NETWORKS A Guide To Conceiving, Organizing, Implementing, and Monitoring Interdisciplinary Research Networks. https://www.macfound.org/media/files/rosenetworkmonograph.pdf

^[6] Mimi Engel, Amy Claessens and Maida A. Finch (2013). Mathematics Instructional Content and Student Knowledge in Kindergarten. Educational Evaluation and Policy Analysis, Educational Evaluation and Policy Analysis, 35(2), 157-178.