This Research in Service to Practice project is funded by the Advancing Informal STEM Learning (AISL) program, which seeks to advance new approaches to, and evidence-based understanding of, the design and development of STEM learning in informal environments. This includes providing multiple pathways for broadening access to and engagement in STEM learning experiences, advancing innovative research on and assessment of STEM learning in informal environments, and developing understandings of deeper learning by participants.
The project will research the educational impact of social robots in informal learning environments, with applications to how social robots can improve participation and engagement of middle-school girls in out-of-school computer science programs in under-resourced rural and urban areas. The use of robots to improve STEM outcomes has focused on having learners program robots as tools to accomplish tasks (e.g., play soccer). An alternate approach views robots as social actors that can respond intelligently to users. By designing a programmable robot with social characteristics, the project aims to create a culturally-responsive curriculum for Latina, African American, and Native American girls who have been excluded by approaches that separate technical skill and social interaction. The knowledge produced by this project related to the use and benefits of social programmable robots has the potential to impact the many after-school and weekend programs that attempt to engage learners in STEM ideas using programmable robot curricula.
The project robot, named Cozmo, will be programmed using a visual programming language and will convey emotion with facial expressions, sounds, and movements. Middle school girls will engage in programming activities, collaborative reflection, and interact with college women mentors trained to facilitate the course. The project will investigate whether the socially expressive Cozmo improves computer science outcomes such as attitudes, self-efficacy, and knowledge among the middle school female participants differently than the non-social version. The project will also investigate whether adding rapport-building dialogue to Cozmo enhances these outcomes (e.g., when a learner succeeds in getting Cozmo to move, Cozmo can celebrate, saying "I can move! You're amazing!"). These questions will be examined research conducted with participants in multi-session after-school courses facilitated by Girl Scout troops in Arizona. The project will disseminate project research and resources widely by sharing research findings in educational and learning science journals; creating a website with open source code for programming social robots; and making project curriculum and related guidelines available to Girl Scouts and other educational programs.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
Aligning for Impact: Computer Science Pathways Across Contexts [CS-PAC] is an NSF INCLUDES Design and Development Launch Pilot. It broadens participation of students who are underrepresented in computer science by using the convening and policy-making power of the Georgia State Department of Education to coalesce school district leaders to implement K-12 computer science education. The project provides a national model for how to work toward systemic change. With the State Department of Education's coordination, several school districts will collaboratively seek improvements in their own student participation rates. The coordination of data reporting and analysis, resources, communications, and policy promote more equitable participation in computer science education. Research emerging from this project informs other states about how to collaboratively shape computer science education policy and policy implementation.
Using a Collective Impact approach to systemic change, the project creates sustainable institutional change at the community, state, and national levels. Qualitative and quantitative data provide descriptions about how to utilize alignment strategies within Collective Impact in three different contexts: rural, suburban, and urban. Outcomes utilize a regression discontinuity analysis to justify successful implementation as well as qualitative analysis of implementation efforts that were deemed most effective by all stakeholders. The project outputs directly affect over 88,000 students across five districts and indirectly affect over 1.7 million in Georgia alone. The culminating project goal is the development of a coherent framework for aligning K-12 computer science education pathways.
This is an Early-concept Grant for Exploratory Research supporting research in Smart and Connected Communities. The research supported by the award is collaborative with research at DePaul University. The researchers are studying the use of technologies to enable communities to connect youth and youth organizations to effectively support diverse learning pathways for all students. These communities, the youth, the youth organizations, formal and informal education organizations, and civic organizations form a learning ecology. The DePaul University researchers will design and implement a smart community infrastructure in the City of Chicago to track real-time student participation in community STEM activities and to develop mobile applications for both students and adults. The smart community infrastructure will bring together information from a variety of sources that affect students' participation in community activities. These include geographic information (e.g., where the student lives, where the activities take place, the student transportation options, the school the student attends), student related information (e.g., the education and experience background of the student, the economic status of the student, students' schedules), and activity information (e.g., location of activity, requirements for participation). The University of Colorado researchers will take the lead on analyzing these data in terms of a community learning ecologies framework and will explore computational approaches (i.e., recommender systems, visualizations of learning opportunities) to improve youth exploration and uptake of interests and programs. These smart technologies are then used to reduce the friction in the learning connection infrastructure (called L3 for informal, formal, and virtual learning) to enable the student to access opportunities for participation in STEM activities that are most feasible and most appropriate for the student. Such a flexible computational approach is needed to support the necessary diversity of potential recommendations: new interests for youth to explore; specific programs based on interests, friends' activities, or geographic accessibility; or programs needed to "level-up" (develop deeper skills) and complete skills to enhance youths' learning portfolios. Although this information was always available, it was never integrated so it could be used to serve the community of both learners and the providers and to provide measurable student learning and participation outcomes. The learning ecologies theoretical framework and supporting computational methods are a contribution to the state of the art in studying afterschool learning opportunities. While the concept of learning ecologies is not new, to date, no one has offered such a systematic and theoretically-grounded portfolio of measures for characterizing the health and resilience of STEM learning ecologies at multiple scales. The theoretical frameworks and concepts draw together multiple research and application domains: computer science, sociology of education, complexity science, and urban planning. The L3 Connects infrastructure itself represents an unprecedented opportunities for conducting "living lab" experiments to improve stakeholder experience of linking providers to a single network and linking youth to more expanded and varied opportunities. The University of Colorado team will employ three methods: mapping, modeling, and linking youth to STEM learning opportunities in school and out of school settings in a large urban city (Chicago). The recommender system will be embedded into youth and parent facing mobile apps, enabling the team to characterize the degree to which content-based, collaborative filtering, or constraint based recommendations influence youth actions. The project will result in two measurable outcomes of importance to key L3 stakeholder groups: a 10% increase in the number of providers (programs that are part of the infrastructure) in target neighborhoods and a 20% increase in the number of youth participating in programs.
While the term 'failure' brings to mind negative associations, there is a current focus on failure as a driver of innovation and development in many professional fields. It is also emerging from prior research that for STEM professionals and educators, failure plays an important role in designing and making to increase learning, persistence and other noncognitive skills such as self-efficacy and independence. By investigating how youth and educators attend to moments of failure, how they interpret what this means, and how they respond, we will be better able to understand the dynamics of each part of the experience. The research team will be working with youth from urban, suburban and rural settings, students from Title I schools or who qualify for free/reduced-price lunches, those from racial and ethnic minority groups, as well as students who are learning English as a second language. These youth are from groups traditionally underrepresented in STEM and in making, and research indicates they are more likely to experience negative outcomes when they experience failure.
The intellectual merit of this project centers on establishing a baseline understanding of how failure in making is triggered and experienced by youth, what role educators play in the process, and what can be done to increase persistence and learning, rather than failure being an end-state. The research team will investigate these issues through the use of qualitative and quantitative research methods. In particular, the team will design and evaluate the effectiveness of interventions on increasing the abilities of youth and educators in noticing and responding to failures and increasing positive (e.g., resilience) outcomes. Research sites are selected because they will allow collection of data on youth from a wide range of backgrounds. The research team will also work to test and revise their hypothesized model of the influence of factors on persistence through failures in making. This project is a part of NSF's Maker Dear Colleague Letter (DCL) portfolio (NSF 15-086), a collaborative investment of Directorates for Computer & Information Science & Engineering (CISE), Education and Human Resources (EHR) and Engineering (ENG).