Abstract
A model for developing STEM talents in elementary children through engaging curriculum and sustained teacher professional development was designed, implemented, and evaluated over the course of three federally funded Jacob K. Javits projects. Students in Grades 1 through 5 received accelerated science curriculum, supplemental engineering curriculum units, and curriculum based on biographies of scientists, inventors, engineers, and computer scientists. Teachers were prepared to deliver curriculum-specific STEM content and skills, trained to implement Blueprints for Biography, and were given information and support through summer professional learning institutes, academic-year coaching and technical assistance in talent spotting and the use of universal screening and local norms. Across the three projects, gains in student and teacher learning and engagement were documented.
Keywords
The STEM Starters model resulted in increased numbers of children from low-income households to be nominated for gifted education services.”
Introduction and Purpose
Across 15 years and three Jacob K. Javits projects, a team of researchers and skilled professional learning leaders designed and field-tested a model for developing the STEM talents of elementary students and the educators who serve them. The three projects, STEM Starters, STEM Starters+, and STEM+C2 produced gains in student and teacher learning, developed curriculum materials, implemented evidence-based and commercially available curriculum and left infrastructure in schools for sustainability and scalability. Through its published research, series of teachers’ guides, project toolkits, and online professional learning modules, the STEM Starters model is replicable in districts beyond the more than 25 schools which participated in one or more of the Javits projects awarded to the Jodie Mahony Center for Gifted Education at the University of Arkansas, Little Rock.
The purpose of this manuscript is threefold. First, the projects are described and situated in the theoretical perspectives and rationales which guide them. Second, select project deliverables are summarized. Finally, the published and the on-going research related to the three projects are referenced for readers who wish to learn more about the evidence generated through them and the lessons learned from them.
Perspectives on Talent Development in the Early School Years
The global community needs STEM talents. Early learning is critical to the development of these talents. Challenging curriculum and compelling life stories of scientists, engineers, and computer scientists animate young children to imagine STEM worlds, to identify with STEM roles, and to develop their STEM talents. The STEM Starters model is predicated on the principle that exposing young learners to challenging and engaging instructional and curricular opportunities allows academic talents to emerge in the classroom and encourages teachers to develop the knowledge and skills to spot talents and subsequently develop them. In other words, the exposure to opportunity precedes the talent spotting phase in the talent development process.
To accomplish early STEM talent development, multiple affordances need to be in place. First, in all three Javits projects described here, professional learning for classroom teachers and specialist gifted education teachers was in place for at least a year before students were served. Second, in all three Javits projects, the curriculum implementation included science or engineering-based units and children’s trade book biographies about a scientist, inventor, engineer or computer scientist linked to the STEM content in the science or engineering units. Third, in all three Javits projects, the dosage for students included two consecutive years of curriculum exposure although the STEM curriculum used in the projects was supplementary rather than comprehensive. Finally, in all three Javits projects, there was the expectation that specialized services for advanced students would be available in addition to grade-level classroom STEM experiences.
Rationale for the Focus on STEM Disciplines
One reason for a focus on the STEM disciplines for the intervention was the scarcity of STEM experiences, opportunities, and systematic curriculum implementation in the elementary classroom (Early Childhood Working Group, 2017). A decade and a half ago, science was virtually squeezed out of the elementary curriculum and engineering was a negligible presence. Yet, both science and engineering are valued domains across most global societies and adults who have knowledge and skills in these areas have traditionally been in high demand. If society values scientists and engineers, then starting early to encourage interest, persistence, and accomplishment among children is more likely to increase the talent pool (National Science Board, 2010).
A second reason for a focus on the STEM disciplines during the early school years is that these domains are lively, hands-on in orientation, and align with young children’s developmental curiosity about the world. Even as toddlers, children are nascent engineers when they build with blocks. To the dismay of many parents, pre-school children will experiment with various combinations of household substances “just to see what happens.” Children pick up rocks and stones, capture insects and frogs and observe ant farms and aquariums for hours. They ask questions about weather because they feel the wind, see clouds and lightning bolts, and get caught out in the rain or snow. The natural world is a learning laboratory for children as are our built environments. Stepping on to an escalator or going up in a Ferris wheel present children with opportunities to wonder at the imaginative ideas behind these structures, how they work, and how they improve and enrich everyone’s lives.
A third reason for a focus on the STEM disciplines is that individuals who pursue careers in STEM domains can be inspirational. Their lives are the subject of children’s biographies that expose children to the excitement, the challenges, the failures, and the successes of scientists, engineers, and inventors (Robinson, et al., 2016). STEM biographies provide a role model in a book. If a child’s neighborhood doesn’t have a resident inventor, a biography about Leonardo Da Vinci, George Washington Carver, or Hedy Lamar will introduce them to the multitude of ways people go about inventing and why they do. If a child’s neighborhood doesn’t have a resident scientist, then a biography about Jane Goodall demonstrates how early interests and patient observations lead to a life of adventure. If a child’s neighborhood doesn’t have a resident engineer, then provide the biography of George Ferris and give the famous 1893 World’s Fair wheel its fascinating backstory or introduce children to Alexander Calder, an engineer turned artist, whose mobiles demonstrate both principles of physics and aesthetics. If a child’s neighborhood doesn’t have a resident computer scientist, then children’s biographies for Ada Lovelace, Grace Hopper, and Raye Montague all bring computer programming and other computer science innovations to life for young children.
In summary, the combination of providing STEM opportunity at early grade levels, finding time in the packed elementary curriculum for engaging STEM lessons and units, and reading about STEM role models increases the likelihood that interests and talents will emerge, and teachers will be more likely to notice them in children (Robinson, 2017; Robinson, Dailey, Cotabish, & Hughes, 2014; Robinson, Dailey, Cotabish, Hughes, & Hall, 2014).
Designing, Developing and Testing a Model for Delivering Talent Development Services
The opportunities afforded to children, their educators, and researchers who work in the schools through the Jacob K. Javits Talented Students Program are significant. This small, federal program has punched above its weight since the 1990s when the first demonstration grants and a national center for research on gifted education were initially funded. The importance of sustained funding which allows for careful design of demonstration projects, real-world implementation of them in schools, and rigorous evaluations of their services cannot be overestimated. Through the Javits funding, we have developed and brought resources into schools and left those resources in place when a funding cycle ended. Across the three projects, we have worked in a variety of districts and schools from very small and isolated rural districts to metropolitan districts with thousands of students. The model has thrived in each of these settings. Along the way, we have learned what works and what does not when schools are asked to commit to implementing an innovative instructional model. In each iteration of the model across the three funded Javits projects, STEM Starters, STEM Starters+, and STEM+C2, the lessons learned have been incorporated in the next iteration of the model.
STEM Starters
The first Javits project, initially funded in FY 2008 focused on science and served students in classrooms in grades 2 through 5. The two participating districts included 70 teachers in classrooms randomly assigned to a treatment or a comparison, in this case a delayed treatment condition. The two foci of STEM Starters were professional learning for classroom teachers and gifted education specialist teachers and accelerated science curriculum delivered in both the grade-level classroom and in the elementary pull-out classroom.
An early graphic captures the simplicity of the initial STEM Starters service delivery model. The bare bones graphic was useful in communicating to the districts the compact nature of the intervention and was especially helpful to the classroom teachers and gifted education specialist teachers because they could see the scope of their commitment and the key features of the professional learning and curriculum resources they would receive see Figure 1. STEM Starters Intervention Model.
STEM Starters was the first iteration of the model. Across the three projects, districts, schools, curriculum emphasis, and approaches to identification changed, but four constants from the first project, STEM Starters which formed the basis of the model, remain. These are the delivery of services to students and teachers in both grade level and specialized gifted education classrooms, the emphasis on sustained professional learning for teachers, the implementation of existing evidence-based STEM curriculum, and the development of instructional guides, Blueprints for Biography: STEM Series, for children’s biographies.
Curriculum Units by Grade Level for STEM Starters Treatment Students.
Note. STEM = Science, Technology, Engineering, and Mathematics.
The professional learning intervention for teachers was also provided on a two-year cycle. Teachers in the treatment group received two summer institutes of a one-week duration each and the services of a professional learning peer coach with a background in high school science and K-12 gifted education preparation. During the academic year, the coach provided services in individual classrooms by modeling lessons in the early months of the curriculum implementation, then serving as a materials facilitator assisting the teacher in setting up for the hands-on science unit, and finally fading from the school, but making herself available for phone, email, and distance conferencing.
STEM Starters Teacher Professional Development Across 2 Years.
STEM Starters+
The second Jacob K. Javits project funded in FY 2014 scaled up the initial project in three ways. First, STEM Starters+ began with Grade 1 classrooms in order to work with teachers on ways to spot talents early before formal identification processes commenced, generally at the end of the academic year for Grade 2 students. Second, the curriculum implementation expanded to include engineering units as well as science units. Driven by changes in the national science standards which explicitly identified engineering design processes, the engineering units directly addressed the integration of design processes into the science standards and acknowledged the growing importance of STEM integration in the schools (Mann et al., 2011; National Research Council, 2012). Third, STEM Starters+ increased its focus on technology. A graphic of the STEM Starters+ model summarizes the scope of the scale-up see Figure 2. STEM Starters+ project design.
Curriculum Units by Grade Level for STEM Starters + Treatment Students.
Note. STEM = Science, Technology, Engineering, and Mathematics.
STEM Starters+ included scale-up and dissemination activities to further its outreach beyond the schools participating in the second Javits project. One of these dissemination initiatives was a planned activity in the scope of the original proposal; the other, an unintended outcome of interest in the project by the National Science Teachers Association-TV.
The STEM Starters+ Virtual Scale-Up Summit was conducted in the final year of the project for 50 educators from Arkansas and surrounding states to share lessons learned, project deliverables, and an implementation guide. The Virtual Scale-Up Summit included nine asynchronous modules, and a 10th live Question and Answer module delivered through the university learning management system, BlackBoard. Modules varied in length but were between 10–15 minutes in duration for a total of 3 hours for the complete online asynchronous and synchronous summit. Fifty-three participants primarily from Arkansas and North Carolina joined the Virtual Summit, received a downloadable STEM Starters+ implementation guide, a downloadable Blueprint for Biography from the Jodie Mahony Center website, and a professional learning certificate. Example asynchronous module titles included: What Does My School Implement? A Dive into the STEM Starters+ Model, Role of the Principal, the Classroom Teacher, and the Gifted Education Specialist in STEM Starters+, Effectiveness for Students and Teachers, and A Deeper Dive: An Overview of the STEM Starters+ Implementation Guide.
In addition to the planned scale-up Virtual Summit, the Mahony Center was approached to provide a professional quality video for inclusion in the Annual Conference of the National Science Teachers Association. The result was a five-minute video showcasing the Mahony Center and the STEM Starters+ project. Filmed on campus and in Terry Elementary, a STEM Starters+ school, in the Little Rock School District, the video included interviews with project staff, elementary teachers, and footage of children engaged in an experiment in damping sound from an engineering unit and another segment of a different group of children engaged in analyzing a portrait of George Washington Carver as part of their Blueprint for Biography curriculum. The video is available at https://www.youtube.com/watch?v=cLMrbHsRkC4.
STEM+C2
The third and current STEM intervention model funded in 2019 through the Jacob K. Javits Gifted and Talented Students Program added a computer science focus. STEM+C2 takes as its starting point the pathway identified by the National Science and Technology Council (NSTC) to “engage students where disciplines converge.” (pp vi). As articulated by the Council, STEM+C2 “seeks to make STEM learning more meaningful and inspiring to students by focusing on complex real-world problems and challenges that require initiative and creativity.” (pp vi). The project is guided by the premise that challenging and creative curriculum can be used to develop talents in all children and as a framework whereby educators systematically engage in the practice of talent-spotting children from underrepresented groups for subsequent gifted and talented services (Robinson et al., 2018). STEM+C2 is a systematic approach to the progression from universal screening in Grade 1 to talent development through curriculum in Grade 2 to gifted and talented programs and services in Grade 3 for students not traditionally identified and served. Moving from a “convenient clustering” of related disciplines, the project integrates science (S), technology (T), engineering (E), math (M), computer science (C) and creativity (C) (STEM+C2). The project is graphically displayed in Figure 3. STEM+C2 project design.
STEM+C2 incorporated universal screening and local norms using existing state accountability tests, thus leveraging the resources the schools already had at their disposal. The reasoning behind the choice to use existing tests was to encourage buy-in from teachers, principals, and central office personnel. Using the tests already familiar to educators in the STEM+C2 schools did not layer on additional expenses or testing tasks.
The STEM+C2 project retained the supplementary curriculum engineering units with new unit designs developed by the Museum of Science, Boston. These units, reconceptualized as Youth Engineering Solutions (YES), were incorporated into the STEM+C2 project. In addition to the science content taught through engineering and an increased infusion of computational reasoning, YES units included a graphic novella to engage students in thinking about an everyday engineering problem and seeking to solve the problem through the engineering design process.
In keeping with the STEM Starters model which included biography as a key feature, STEM+C2 developed new Blueprints for Biography: The Computer Science Series. The biographies selected for STEM+C2 represent individuals who struggled to attain their goals, whose biographies included information about their early interests and childhood experiences and whose biographies included information about the kinds of work computer scientists do. Raye Montague, an African-American computer scientist, described by the U. S. Navy as their “hidden figure,” developed computer programs to design submarines quickly enough to be built during wartime. Grace Hopper had the insight to move from binary machine language to using English words and developed the first word-based programming languages in computer science. She worked her way through school and pursued a career in computer science uncommon in the 1940s. Ada Lovelace, a 19th century whiz at mathematics and computational reasoning, anticipated how machines would eventually be programmable through performing repeated tasks sequentially and in a loop, a staple of modern computer science. A fourth Blueprint was developed to compare two biographies of the same individual, Hedy Lamar, an actress and inventor who developed frequency hopping in the early years of World War II, a key insight into the subsequent development of Wi-Fi.
Each guide in the Blueprints for Biography: Computer Science Series concluded with a science investigation or engineering design challenge as well as a computer science exploration created to build a foundation in computational thinking, a critical skill in computer science (Meadows, et al., 2024, forthcoming). In addition, the development of an instrument measuring student knowledge and enjoyment of biography as a genre was developed and used to document student outcomes (Robinson et al., 2024).
Evaluation of the project was fully integrated into the STEM+C2 design from its initial year of planning and implementation. A national advisory board composed of individuals with interests in STEM, educational policy, school leadership, and gifted education provided early input and were an invaluable source of support and insights when COVID-19 affected schools and communities in every state. Among the advisory board members were community leaders, an association chief executive officer, members of the state department of education and of higher education, an engineering education researcher and curriculum developer, central office administrators from participating districts, and the university administrator who developed online programs at the University of Arkansas at Little Rock. The professional learning components in STEM+C2 were pivoted from face-to-face institutes, coaching, and technical assistance to an online format. During the project duration, the state department of education changed accountability test vendors, but cooperated with STEM+C2 through a memorandum of understanding so that project researchers could access previous and current state test data for universal screening support for the schools and to provide continuity for the achievement outcome measures.
Curriculum Foci Across Projects Based on the STEM Starters Model
To summarize the commonalities and the differences across the three Jacob K. Javits projects that developed and field-tested the STEM Starters model, a graphic which focuses on the curriculum elements across the 15 years of design and implementation, is provided. One curricular lens is the increasing integration and expansion from science content and processes taught directly through problem-based learning (STEM Starters) to science content taught through engineering curriculum and the engineering design process (STEM Starters+) to integrated science, engineering, and computer science (STEM+C2). Figure 4 Effective Elementary STEM Curriculum Interventions: The STEM Starters Model graphically summarizes the progression. Effective Elementary STEM curriculum interventions: The STEM Starters Model.
Summary of Student and Teacher Effects
Developing and field-testing a scalable model was made possible by the value placed on the evaluation of the Jacob K. Javits projects established through the competitive proposal process. In each iteration of the model, we were able to incorporate evaluation best practices. In the first two projects, an outside evaluator was included in the project activities. In all three projects, field studies were conducted on teacher and student outcomes. The randomized designs or matched samples designs with comparison groups permitted inferences to be made about the effects of the STEM Starters model interventions. In the first project, STEM Starters, classrooms within schools were randomly assigned to a treatment or comparison group. Although there did not appear to be diffusion effects, the next two interventions, STEM Starters+ and STEM+C2, used stratified random assignment of schools rather than classrooms to the treatment or comparison condition. School-level randomization was possible because we were working with a greater number of districts in the second and third iterations of the model. Across all three projects, however, the research design was a randomized delayed treatment design. In a 5-year project with two consecutive years of intervention for each student, the participating classrooms or schools in the treatment condition were served in years 2 and 3 of the project; the comparison classrooms or schools were provided with the services in years 4 and 5 of the project. The design appealed to district leaders’ sense of equitable opportunities for their students and teachers and most likely reduced school attrition over the course of 5-year projects. In addition to the randomized field studies, qualitative and mixed methods investigations were also conducted to provide a richer reflection of project implementation.
A summary of the research and evaluation results is organized by student effects and teacher effects.
Student Effects
The student effects summarized here are largely from the STEM Starters and STEM Starters+ projects as the third project, STEM+C2, is still underway. For example, for all children in Grades 2–5 who participated in STEM Starters, we report increased science process skills, science concepts, and science content when compared with students not participating in the intervention (Cotabish et al., 2013). These results were echoed in a study of identified gifted student outcomes (Robinson, Dailey, Cotabish, & Hughes, 2014). For STEM Starters+ student results, a study of Grade 1 students reported increased achievement on an out-of-level science test, increased engineering knowledge with no evidence of gaps across all demographic groups of students, and increased engineering engagement with no gender or minority group differences, but with differences for children from low-income households (Robinson, et al., 2018). In addition, a study of STEM Starters + students in Grades 2 and 3 who were identified as gifted, reported increased engineering knowledge across all demographic groups and increased engineering engagement with females reporting slightly higher behavioral engagement than males (Robinson et al., in preparation). Finally, in a study with two replications examining the achievement of students in Grades 2–4 who were identified as gifted, the intervention not only had consistently positive effects but also those effects were stronger with each replication. This suggests that with additional iterations of implementation and refinements by the professional learning team, the teachers increase their efficacy in implementing the curriculum which increases the likelihood of improved student performance (Adelson, et al., in preparation).
Teacher Effects
Across the three Javits projects, the instrumentation for documenting teacher knowledge and skill changed, but overall, there was a sustained interest in teacher STEM content knowledge and a willingness to engage in STEM at the elementary school level. For example, STEM Starters measured teachers’ science process skills and found that the professional learning intervention increased teachers’ ability to design experiments when compared with the randomly assigned comparison group of teachers (Cotabish et al., 2011). In addition, teachers’ perceptions of their ability to teach science and their students’ ability to learn science increased from baseline to Year 1 and Year 2 of the STEM Starters intervention and persisted post-intervention as well (Dailey & Robinson, 2017). The teacher outcomes shifted in STEM Starters+ with a measure of science content rather than process and the addition of their self-efficacy for teaching engineering to children. In addition, a study of Grade 1 classroom teachers found that the professional learning intervention which included a focus on talent spotting resulted in increased nominations of children from low-income households and children who identified as an underrepresented minority for gifted and talented services (Robinson, et al., 2018). Finally, a mixed methods study of teacher response to the STEM+C2 professional learning resulted in increased knowledge of gifted identification principles and increased knowledge of biography as measured by the BIOS but not increased enjoyment which was already high. The qualitative insights concerned the participatory nature of the professional learning—teachers experienced the engineering and biography curriculum in the same ways their students did. The importance of principal support for project success was a theme especially in the context of disruptions due to COVID-19 (Meadows et al., 2025).
Lessons Learned
In addition to the research results and evaluation insights, there were valuable lessons learned about the barriers and boosts of designing, implementing, and evaluating innovative educational interventions.
First, the sustained professional learning opportunities for elementary teachers, both classroom teachers and gifted education specialist teachers, were critical to the success of the model implementation. In each iteration, teachers had access to highly skilled individuals who provided peer coaching and other kinds of technical assistance throughout the project. Three of these individuals had formal training in both science and gifted education and had previously been K-12 practitioners. Teachers and principals trusted them to be supportive, understanding of the difficulties in school scheduling, and the demands of an elementary classroom schedule.
Second, training on specific curriculum materials rather than a generalized curriculum model approach produced greater confidence in teachers and greater likelihood in fidelity of implementation in the classroom.
Third, engineering provided teachers with an academic area that allowed talents to emerge in students who might not have been high scorers in literacy or math achievement. The hands-on, design-based nature of engineering in the elementary classroom offered surprises to teachers about who was talented in ways they had not observed in other kinds of classroom lessons.
Fourth, linking science and engineering curricula with engaging biographies about eminent computer scientists, engineers, inventors, and scientists brought STEM role models into the classroom in a lively way. The Blueprints for Biography series afforded students the opportunity to not only hear a biography read to them or to read one for themselves, but also to learn the skills of interpreting portraiture and other forms of document analysis included in the curriculum guides.
Fifth, providing clear intervention descriptions and benefits to their students to teams of educators which included a principal, classroom teachers, and a gifted education specialist teacher encouraged shared responsibility and sustained commitment.
Sixth, offering small monetary incentives to teachers for data collection associated with the project increased the likelihood that pre and post-testing would be completed and shared with project staff in a timely way.
Seventh, even young children in primary grades were able to respond to electronic assessments offered through Qualtrics. This data collection strategy improved return rates for student and teacher assessments thereby increasing the confidence we have in our student and teacher outcomes. Electronic data collection was a critical tool when COVID-19 disrupted school instruction and assessment practices.
Summary of Resources and Deliverables You Can Use
The three Jacob K. Javits projects included deliverables that serve as resources to other schools who wish to implement the model as a whole or to adopt specific components which fit their needs.
First, the science and engineering curriculum components are commercially available to any district. For example, the problem-based science units developed through other Javits projects can be purchased for use. The engineering units, both EiE and YES, continue to be developed, refined, and available from the Museum of Science, Boston. Second, the Blueprints for Biography: STEM Series and Computer Science Series are available from the Jodie Mahony Center at the University of Arkansas, Little Rock. The specific trade book biographies on which they are based are available online and from bookstores across the country. The most recent Blueprints include an instrument for documenting the knowledge and engagement of children with biography designed and validated through Javits funding. The Biographical Information and Opinions Scale (BIOS) is a unique contribution to the measurement toolbox as no prior assessments for biography were available in the literature. Third, the protocols used for assisting schools to leverage their accountability tests as tools for universal screening and establishing local norms are also available through the Mahony Center. Finally, an implementation guide for STEM Starters+ and a STEM+C2 Toolkit are available digitally for schools wishing to implement the model.
Conclusions
The Jacob K. Javits funding provided a stable source of funding to design, field-test, and scale a STEM model to develop the talents of elementary students and their teachers. It allowed for a decade and a half of services to schools and contributions to the research on effective school interventions for advanced students. Through multiple field studies in districts ranging from a few hundred students to thousands, the model was flexible enough to flourish and to produce achievement and engagement results for students and increased achievement and confidence for teaching STEM in their teachers. Teacher gains in understanding the principles of talent spotting and the procedures for universal screening with local norms were also outcomes of the model in its third iteration. Overall, the STEM Starters model resulted in increased numbers of children from low-income households to be nominated for gifted education services. STEM Starters, STEM Starters+, and STEM+C2 demonstrate that designing a school-based domain-specific model of talent development and subsequently testing it empirically in the real world of schools contributes to both the research on what works in talent development and what lessons can be learned from the world of practice.
Footnotes
Acknowledgments
The author wishes to thank Kishor Kumar Kagithi Sreeramulu for his assistance in the preparation of this manuscript. The author also wishes to thank the members of the research and demonstration team across the three projects: Jill Adelson, Katherine Cash, Alicia Cotabish, Debbie Dailey, Christine Deitz, Tony Hall, Gail Hughes, Kristy Kidd, Monica Meadows, Keila M. Mugabo, and Merve Topak and the educators who implemented the STEM Starters model with fidelity.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Jacob K. Javits Program, U.S. Department of Education: S206A080026 (STEM Starters); S206A140006 (STEM Starters+); S206A190008 (STEM+C2). However, the contents of the article do not necessarily represent the policy of the Department of Education.
