Station number, title, presenter, and abstract for each gallery walk station.
1. Workshop on Digitally-Mediated Team Learning
Ronald F. DeMara, Laurie O. Campbell, Samuel Spiegel, Richard Hartshorne
Digitally-Mediated Team Learning (DMTL) encompasses cooperative learning in a digital classroom-based synchronous setting with the instructor in a supportive role. Foci of DMTL include STEM problem-solving and design activities in real-time with an emphasis on leveraging educational technologies, learner analytics, and pedagogical methods, while empowering equitable participation. This poster overviews the methods and outcomes of the NSF Project on Synthesizing Environments for DMTL which convened researchers from 44 universities to identify 1, 3, and 5+ year research. Untapped opportunities for online instructional environments are identified to engage, orchestrate, and assess STEM design and problem-solving teams in classroom settings. These include:
- Leveraging instructional technology during group problem-solving activities which allows team members to adopt technical/leadership roles within a team to co-construct solutions to exercises through peer teaching and learning.
- Having the instructor serve in a supportive role in which technology provides the instructor with real-time analytics on team progress, which can then be used to provide proper scaffolding/formative feedback for maximum efficacy in reaching learning goals.
- Supporting equitable participation encompassing the human aspects in a community of learners across the range of interactions enabled within digital teams.
2. Supporting Collaborative Problem Solving in a Game-Based Learning Environment
This poster will present our story-centric game-based learning environment, Crystal Island: EcoJourneys, to highlight how the design is informed by problem-based learning (PBL), a constructivist instructional approach driven by an inquiry framework. We present key features of Crystal Island: EcoJourneys design: 1) the interactive story, 2) the whiteboard, and 3) the collaboration scaffolds, as well as discuss trace data of student interactions as they engage in the story and contribute to the chat and whiteboard tool. Our design of the game highlights a complex ecosystem that illuminates how aligning tools in the game and our commitment to a social-constructivist approach to learning, while challenging, was instrumental in supporting student learning.
3. The Reflect! platform: A tool for learning how to approach wicked problems
Wicked problems pose particular challenges — especially ethical challenges — because they can be framed in a number of different ways,depending on the often conflicting values, needs and interests of those who look at the problem. To approach a wicked problem, a multitude of societal stakeholders has to be taken into account. Because dealing with wicked problems often leads to confusion, frustration, and conflict, it is difficult to provide opportunities for learning how to deal with them in educational settings.
The Reflect! platform is an educational tool designed to support reflective, team-based consensus building about wicked problems (https://reflect.gatech.edu/). The platform provides work plans for projects that can last up to a semester. Reflect! has first been tested in two classes, each with an enrollment of about 170 students who worked in about 40 teams on problems such as micro-targeting in social media, privacy issues in brain-computer interfaces, and ethical concerns about the use of robotic caregivers for the elderly.
4. Cyber-Eye: Transforming Construction Engineering Learning for the Future Engineering Workforce Using Unmanned Aerial Systems Technology
Cyber-Eye aims to transform current CEM education by providing a technology-based learning environment that features aerial visualizations and information produced by small Unmanned Aerial Systems (UAS). Cyber-Eye learning technology environment offers a unique perspective. It enables the simultaneous visualization of in-situ construction resources, processes, and management of activities as they unfold over time. It allows for search and retrieval of UAV visualizations, sensing data, and text of solutions to problems. The output provides learners the opportunity to develop skills that integrate procedural and configurational knowledge, by enhancing their understanding of interdependencies, interactions, and constraints among integrated and specialized engineering systems and a construction project. This understanding occurs by facilitating the perception of spatial relations of entities (objects) and dynamic processes. It impacts students’ abilities to process complex information and develop essential knowledge related to CEM spatial-temporal configurations. In short, Cyber-Eye’s design will significantly scaffold individuals’ development of their spatial-temporal cognitive ability.
5. Graspable Math
Our demonstration and/or poster will showcase Graspable Math, a dynamic algebra technology tool designed for teachers, students, and anyone who wants to play and explore math within a fast, fluent, and conceptually grounded algebra system. This novel technology allows users to interact with algebraic structures in new ways. First, we allow users to manipulate and transform algebraic expressions with a touch of their finger or mouse, using gestures that are fluid and relatively error-free. Second, we allow learners and teachers to explore strategies and pathways through algebraic interactions and fluidly connect algebraic situations to other kinds of graphical representations.
6. Signing avatars and embodied learning in virtual reality.
Jason Lamberton, Athena Willis, Melissa Malzkuhn, & Lorna Quandt
The Signing Avatars and Immersive Learning (SAIL) project is ongoing, and as part of the project were are developing ASL signing avatars who will be placed in a virtual reality environment. Using gesture tracking sensors, we are creating a system in which users can learn ASL signs from avatars, and see the signs they are producing themselves in virtual reality. As development is underway, we have demonstrations available where users can enter the virtual reality environment and see the ASL content produced by avatar characters.
7. Mapping Fields in AR with Personal Mobile Devices to Enhance Visualization Skills for Education and Industry
Using only a smartphone, we have developed an app that permits learners to anchor magnetometer measurements to 3-D space that can be viewed through the camera. We have prepared a series of exploration protocols that help users explore background and non-idealized magnetic fields. Our preliminary work with users suggests that exploration with this technology helps students to think critically about the limitations of idealized field models and recognize the complexity of environmental fields.
8. Idea Thread Mapper
Idea Thread Mapper (ITM) is a research-based collaboration platform that supports student-driven knowledge building and creative inquiry across classroom communities. At the core of ITM is a visual collaboration and discourse environment where students develop extended threads of ideas to build deep understandings and work on progressively deeper problems. The online discourse tools integrate with Knowledge Forum, a world-renowned platform for collaborative learning and knowledge building. To sustain student-driven inquiry and dynamic collaboration, ITM further incorporates two types of support:
- Support for collective inquiry structuring: Idea thread mapping and analytics embedded in the online discourse environment help students reflect on their progress and co-organize their collective journey of inquiry; and
- Support for cross-classroom collaboration: Boundary-crossing tools help build connections between different classrooms, which share inquiry progress and engage in shared Super Talk.
With its solid design for open interoperation and adaptability, ITM works with Knowledge Forum and other tools across platforms and databases. Together these systems provide a new social technological infrastructure for transforming classrooms into interconnected creative communities, in which students develop 21st century competencies.
9. Learning Parallel Programming Concepts Through an Adaptive Game
Modern computing is increasingly handled in a parallel fashion and despite the growing body of work on how to teach parallel programming, little is understood about the learning of this subject. This project shed light on the challenge of learning parallel programming and gather initial data on ways to scaffold it in college-level courses. We developed an adaptive learning game called Parallel. Our research advances understandings of how people learn concepts associated with parallel programming, as well as investigating which activities enhance the learning process in this domain. Our game generates content tailored to specific students through a method entitled procedural content generation. This work will transform the transition from sequential programming to parallel programming in undergraduate computer science curricula and advance personalized learning. The demo and poster will showcase the adaptive game we developed as well as research results.
10. Collaboration Learning Analysis with Deep Neural Networks
In this poster, we present a new method to automatically evaluate the collaboration level of students in groups using computer vision techniques. Instead of single-time solicitation techniques such as self-reported questionnaires to evaluate the collaboration, we evaluate the entire process of the collaboration by analyzing image data captured from groups. We use object detection techniques such as Mask R-CNN method to process our data. This process is based on detecting people and other objects from pictures and video clips of the collaborative learning process, then evaluating the performance of students in the interactive learning and textbook settings using these collaborative indicators. Two collaborative indicators are related to the team’s proximity and time on task. We tested our approach in a controlled study of 33 teams while performing an anatomy muscle painting intervention. The results indicate that our approach successfully recognizes the differences of collaborations among teams of treatment and control conditions, F(1,33)= 11.42, p<0.005. This work may offer implications for automated quality prediction of collaborations among human-human interactions in group-work scenarios.
11. Multimodal Learning Analytics: Understanding and Improving Learning in Physical Spaces
Multimodal Learning Analytics (MmLA) is the logical and inevitable evolution of Learning Analytics that by capturing, processing, and fusing natural rich modalities of communication, such as speech, writing, and nonverbal interaction (e.g., movements, gestures, facial expressions, gaze, biometrics, etc.) creates a more comprehensive and holistic framework to study learning process wherever and however they happen. MmLA is not limited only to online or digital environments, but its main strength is to be able to analyze face-to-face, hands-on, unbounded, and analog learning settings such as classrooms, collaborative groups and labs. MmLA heavily uses recent advances in sensor technology, as well as Computer Vision, Natural Language Processing and Artificial Intelligence to automatically extract relevant features from multimedia recordings. This automation helps educational researchers to scale their studies and also increase the level of detail at which there are able to observe the individuals. This automation also enables the creation of real-time feedback systems to support students’ and instructors’ reflection about their learning and practices.
12. My Learning Analytics: a Student-Facing Dashboard
My team at UM has developed and implemented a student-facing dashboard that provides students with information intended to support self-regulation. It has three views: an assignment planning view, a activity tracking view, and a grade distribution view. Although performance dashboards are quickly becoming standard in Learning Management Systems, recent reviews of prior research on their use has not answered important questions about the impact of that use on students’ motivation and behaviors. We have data from 10+ classes, undergraduate & graduate-level, that includes log data and survey data which we are currently analyzing over the summer of 2019. The project recently won a Gold Award at the Learning Impact Leadership Institute from the IMS Global Learning Consortium.
13. Number Factory: An adaptive and inclusive elementary math interactive
As part of an research-practice partnership (RPP) to infuse Computational Thinking into STEM in grades 3-8, we are prototyping interactive learning experiences that can be adaptable using educational data mining. We will present NumberFactory, an interactive for place value learning (1s, 10s, 100s, and 1000s) with flexible scaffolds to support neurodiverse students. These scaffolds include graphical organizers and sensory customization. This type of differentiated tool is intended to leverage the overlap between CT practices and the unique cognitive strengths and needs of neurodiverse learners.
14. Student Explorations of Human Impact on Global Temperatures: Leveraging Student Choice to Promote Knowledge Integration
This research, as a part of the cyberlearning project Project Learning with Automated Network Support (PLANS), explores the impact of choice on sixth grade students’ learning through online climate change curriculum. In the curriculum, students investigate human activities that might contribute to rising global temperatures using computer-based models. We compared students who were able to choose which activities to investigate in the model (Choice) with those who had activities assigned (Standard). A two-way ANCOVA indicated that, when controlling for pretest scores, students in the Choice condition demonstrated slightly higher learning gains than those in the Standard condition. Further, there was a significant effect of teacher on learning gains. These results point to the need to better understand how teachers facilitate students’ self-directed learning and support students in making productive choices.
15. HoloLens Applications for Teaching Science
Hololens applications to demonstrate and teach fundamental concepts of electrodynamics were developed as part of a collaboration between the Case Western Reserve University Physics Department, Interactive Commons, and the Teaching and Learning Technologies Center. The applications were integrated into the curriculum of a course on electrodynamics course as means to gain experience and evaluate the effectiveness of augmented reality (AR) for teaching in the sciences. The enthusiastic feedback from the students has led to the offering of a course Augmented Reality (AR) for Teaching in the Sciences to further explore ideas for using AR in the classroom. Here we demonstrate some of the HoloLens applications and present findings from our experiences developing and using these applications in the classroom.
16. Exploring Geoscience Concepts in a Virtual Grand Canyon
Stephen Moysey, Victoria Sellers, Matthew Boyer, Catherine Mobley, Sabarish Babu, Geoff Musick, Bryson Rudolph, Kelly Lazar, Leah Wiitablake
Virtual reality has the potential to transform field education. VRFE Grand Canyon is an example of how basic concepts from introductory physical geology courses can be integrated into a virtual reality field experience. VRFE Grand Canyon is a fully immersive, game-like VR experience designed to run on smartphones using Google Cardboard. The app addresses geoscience content ranging from rock identification to the synthesis of geologic history from field observations by guiding students through a sequence of locales that emphasize different geologic materials and processes. For example, students use analysis of hand samples to investigate the link between rock characteristics and depositional environments for the Coconino Sandstone, whereas the sequence of Muav Limestone, Bright Angel Shale, and Tapeats Sandstone are used to introduce concepts of sea level change. A unique aspect of VRFE Grand Canyon is that we are embedding geoscience learning content within the context of a narrative story that follows the exploits of a prospector drawn to the Canyon by a real-world news story published in the New York Times in 1912. The use of storytelling in this example shows how Grand Canyon’s human and geologic history can be intertwined to more deeply engage and motivate learners.
17. Connecting Language, Learning and Critical Thinking
Rebecca J. Passonneau, Smaranda Muresan
To test whether a proposed educational intervention has an impact on students’ learning, researchers in education develop rubrics to assess students’ reading and writing skills. The rubrics are tested for reliability, and are applied to students’ written or oral discourse to compare students who receive a novel educational intervention to a control group. This is similar to an enterprise in natural language processing (NLP) where NLP researchers annotate corpora to add markup about meanings and categories that are not explicit in the text, and test the annotation reliability. The resulting corpora are used to train and test machine learning applications. In one of my current projects, we hypothesize that these kinds of NLP methods can transfer to automation of educational rubrics. This same project has led me to consider how educational applications of NLP might shed light on the question of how meaning is constructed through language use, for example by students and instructors, in their reading and writing.
18. Computational Thinking with Zoombinis: Results from a national classroom implementation study
Zoombinis is an award-winning, educational game targeting ages 8 through adult, consisting of a series of challenging logic puzzles designed to develop mathematics concepts essential for computer programming and data analysis, such as sets, logical relationships, dimensions, mappings, sorting, comparing, and algorithms. In a national classroom implementation study of the logic puzzle game Zoombinis, 1,273 students in elementary and middle school classes and teachers were provided with classroom materials designed to help bridge game-based knowledge with formal instruction. This poster focuses on the multilevel analysis of changes in students’ pre-post CT assessments in 57 classrooms, informed by types and quantity of CT instructional practices, gameplay duration, as well as student, classroom, teacher, and school demographics.
19. An Intelligent Tutoring System for Source Code Comprehension
Vasile Rus, Peter Brusilovsky
This interactive demo will show a first prototype of an intelligent tutoring system developed by the ongoing project CSEdPad, funded by the National Science Foundation. The CSEdPad project aims at developing an effective and engaging instructional intervention to improve comprehension and learning in introductory Computer Science courses at the college level, to reduce attrition rates and increase retention, and to ultimately produce more and better-trained graduates. The result will be a win-win-win situation for aspiring students, CS programs and their organizations, and the overall economy.
Understanding source code is a critical skill for both learners and software developers. Learners spend a significant amount of time looking at code examples, e.g., given by an instructor or in a textbook. Similarly, software developers spend a significant portion of their time looking at someone elseâ€™ code, e.g., code written by a teammate.
One working hypothesis of our project is that instructional strategies such as eliciting self-explanations will result in mental models that are more accurate, which in turn will positively impact learning, self-efficacy, and retention. The positive role of self-explanations is well documented for science learning but less so for learning computer programming.
20. Contextual Linkages Between Educational Platforms
A growing number of educational platforms are employed in coursework, often connected via LTI protocols within the course LMS. This presentation demonstrates a system that automatically connects disparate learning platforms based on the context of what was discussed in a class session. This system creates machine-generated transcriptions of class sessions and then uses those transcripts to identify keywords and phrases used during class.
The keywords and phrases gleaned through this process are then used to search other educational resources in the students’ learning ecosystem (e.g. LMS, textbook, open-education context), providing contextual linkages to benefit student study and to expose students to the value of resources assigned to class.
It is hypothesized that providing contextual links to germane resources based on the context of a studentâ€™s notes, for example, will incent students to take more complete notes as they will hence be rewarded with a more complete list of relevant links including, but not limited to, specific pages in course textbooks, documents in the student course LMS, relevant videos and questions appropriate for self-assessment on material covered in each class session. This system will also collect data on student usage of each linked platform that can inform institutions on value to student learning and use.
21. Empowering Learners to Conduct Playful Experiments
Students learning to do experimental research face many practical and conceptual challenges, including access to appropriate physical resources, understanding how to align study designs with hypotheses, and understanding concepts such as validity. This Cyberlearning project is developing Studycrafter for learners to easily practice creating and running research experiments into human behavior. StudyCrafter offers a platform with tools for student research designers to construct and distribute online interactive scenarios, and to record the choices that participants make within them. By varying different aspects of these scenarios, and comparing participants’ responses to them, students can explore questions about human behavior. We describe implementations of StudyCrafter in university classrooms, in which we are studying the affordances of StudyCrafter for promoting students’ engagement in, and understanding of research design; how the tools that we make available for students to create research studies (e.g., characters, environmental assets) can be designed to mitigate stereotype and bias issues that threaten any research design; and our explorations into designing tools powered by artificial intelligence to automatically analyze and guide students’ work.
22a. E-TRIALS: The Future of The ASSISTments Testbed
A poster describing both the history the ASSISTments Testbed (infrastructure that has allowed researchers to conduct randomized controlled trials in the ASSISTments online learning platform) and the future of this infrastructure as it evolves into E-TRIALS, an EdTech Research Infrastructure to Advance Learning Science, through recent funding from Schmidt Futures.
22b. Skyscraper Games
Skyscraper Games is a project from the Entrepreneurial Game Studio (EGS) at Drexel University’s ExCITE Center. The goal is to engage middle school students in game design and computer programming using our online Python coding tool, which features a model of the Cira Centre – a skyscraper in downtown Philly. The students then have the opportunity to display and play their games on the actual building. This project started in 2014 when Dr. Frank Lee played the world’s largest Tetris game on the Cira Centre. From 2015 – 2017, the EGS held summer workshops for middle school students in the Philadelphia area. In order to expand the reach and scope of our offerings, for school year 2017 – 2018 we piloted a professional development program for teachers and informal educators. Nearly 40 5th – 8th graders participated. On June 10, 2019, those students will be playing their games on the 29-story Cira Centre.
23a. CIRCL Educators
Judith Fusco, Pati Ruiz, Sarah Hampton, Angie Kalthoff, Steve Hickman, and Amar Abbott
CIRCL Educators, in collaboration with the Center for Innovative Research in Cyberlearning (CIRCL), aims to bridge cyberlearning research with classroom practice and broaden the community of people engaged in creating effective new learning experiences. We blog, host book clubs, lead webinars, and have developed a new learning sciences course for educators.
23b. nQuire – Inquiry Learning at Scale
The nQuire platform (www.nquire.org.uk) engages members of the public in interactive surveys and science investigations. It has been designed by The Open University, UK, in collaboration with the British Broadcasting Corporation (BBC), for â€˜citizen inquiryâ€™ that fuses citizen science and inquiry learning. An individual, group or organization can create an nQuire mission to investigate wellbeing, community or environment. Depending on the topic, participants may upload images (e.g., of wildlife in their gardens), collect sensor data (e.g., environmental noise levels), or respond to small or large challenges (e.g., to improve their wellbeing through creative activity). The platform is open to anyone to participate in or author an nQuire mission. It supports both scientific research (through customizable consent forms and personalized feedback) and community engagement (through missions that allow discussion and sharing of data). Collaboration with the BBC has resulted in large-scale engagement with the platform.
24a. International Community for Collaborative Content Creation (IC4)
A poster providing an overview of the IC4 project, a community of youth in after school clubs ranging from ages 9 – 19 that create and share digital media projects with participants in seven countries across five continents (Brazil, FInland, India, Iran, Kenya, Namibia, USA) to teach one another math, science and technology subjects. Using Epistemic Network Analysis (ENA), the project sees the development of community over time.
24b. Teaching social science through virtual reality and game-based learning
We discuss the development and delivery of a virtual reality (VR) archaeology course and the potential for VR in social science learning environments. We present the design of an immersive, interactive, virtual archaeology course for undergraduate students using the HTC Vive, a room-scale, virtual reality platform, and Unreal Engine 4, an open-source gaming engine. In Spring 2019, Virtual Archaeology was delivered as an undergraduate, general education course at a mid-western public university to relieve the burden of archaeological field school, which many students cannot attend for financial or logistical reasons. The course was designed to teach archaeological theory and field methods using a virtual excavation experience and game-based learning strategies.
This presentation will demonstrate the virtual learning environment used to convey archaeological concepts and technical skills and the game design elements incorporated into the curriculum, including game structure, game involvement, and game appeal. We will discuss the challenges of designing a curriculum incorporating virtual reality and game-based learning, pedagogical trade-offs involved in virtual vs. real-world experiences, and the expansion of virtual learning into other social sciences and natural sciences.
25a. iVisit: Using Virtual Humans in 360-degree Immersive Digital Sites to Practice Communication Skills within Complex Spatiotemporal Contexts
Construction site visits are a vital component of the current construction curriculum. Ideally, three fundamental elements compose a successful construction site visit: well-defined learning objectives, accessible and proper spatiotemporal occasions associated with those learning objectives, and subject-matter-experts on site with whom students can have interactive communication. However, due to the nature of real-world jobsites, several challenges exist that limit the application of this teaching approach. Spatiotemporal constraints make it difficult for instructors to find an appropriate, accessible, and safe location that is tied well with the learning objectives associated with their classes at specific times during an academic semester. Furthermore, additional challenges such as students not being able to see or hear in a crowded or noisy environment and unavailability of hosts who have proper knowledge and expertise to deliver required content associated to site visit learning objectives to students limit the effectiveness of such visits. This research aims to help overcome these barriers, by proposing a novel approach that leverages 360-degree digital sites with virtual human subject matter experts as conversational partners on site (iVisit-Communicate). In this poster, the design and development processes of iVisit-Communicate will be described in detail followed by a case study of its implementation.
25b. Case-based Analysis Platform to Support Ethical Decision-making
During the design process of a three-part MOOC on Ethical Decision-making for Global Managers (part of GeorgetownX), we realized that we could not rely on existing applications/platforms to support learners in the analysis of real-world cases whereby they recognize, test and strengthen their own process of decision making when confronted with ethical dilemmas.
We conducted a design sprint with key stakeholders and confirmed both the learning needs and the learner experience that could support both analysis and reflection of ethical decision-making. We framed our design on the reflective sense-making model, a model that we developed to specifically support ethical reasoning development, (Vovides et al., 2013). The following are some of the features we are incorporating in the prototype (currently in progress):
- individual asynchronous engagement
- peer to peer asynchronous engagement
- an expert-novice visualization of evidence used in decisions
- asynchronous instructor feedback
- group creation based on matching to enable synchronous engagement.
During Cyberlearning 2019, we will demonstrate the prototype for further refinement.
26a. Adapting Reading-Assistance and Automatic Text-Simplification Tools to Assist Self-Directed Learning by Deaf and Hard-of-Hearing Computing Workers
Prior research has observed lower levels of English literacy among many Deaf and Hard-of-Hearing (DHH) adults in the U.S. This poses a challenge for DHH workers in computing fields who must regularly teach themselves about evolving technologies. Artificial intelligence techniques can be used to create text-simplification tools for reading assistance. However, prior research on these tools has not focused on DHH users. We investigate these tools from an interdisciplinary perspective, addressing open questions in natural language processing, human computer interaction, accessibility, and education research to understand how we can adapt these tools for DHH users engaging in self-directed learning. To do so, we need to understand their needs and preferences, evaluate and prioritize different design parameters, adapt the tools to fit their literacy profile and to incorporate texts from the computer domain. We also need to understand how these technologies might benefit DHH users and how to measure their efficacy.
26b. Incorporating Cyber Into Any Classroom
NICERC (the National Integrated Cyber Education Research Center) is a DHS-funded non-profit that develops and distributes STEM, Cyber, and Computer Science curricula to teachers across the country at no charge. Through this project, we make available to teachers, content for all classrooms that help infuse cyber into all subjects. Computational thinking, robotics, programming, and humanities and liberal arts are topics that we bring into math, science, government, history, and English classrooms. Stop by to gain access to the content and learn how we can bring Professional Development to your school for free!
27a. Improving the Flipped Classroom Through Adaptive Learning
Most resistance to the flipped classroom is experienced in pre-class learning and classroom participation. The two are connected as the latter is mostly due to lack of preparation. To improve pre-class learning, we substituted the one-size-fits-all recipe of assigning videos, textbook content, and online quizzes with adaptive lessons. The lessons combine video lectures, textbook content, simulations, and assessment while using personalized paths for each student. Although the cognitive learning gains, measured via a final examination, were found to be of small effect size with adaptive lessons, the improvement in the affective learning was substantial as measured through a classroom environment survey. This reliable inventory evaluated seven psychosocial dimensions of the classroom environment. The effect sizes were found to be positive in all dimensions, with most improvement (medium size effect or larger) shown for cohesiveness, individualization, personalization, and task orientation.
27b. Toward automated dialogue systems for teaching communication and interpersonal skills
This demonstration highlights recent progress in the creation of an online, embodied agent for teaching communication and interpersonal skills. Our system is motivated by a near-term vision that students and workers will soon be able to develop their interpersonal skills through simple conversational practice with automated teaching agents. By practicing effective strategies for handling common interpersonal challenges, for example challenges involving communication, working in teams, resolving conflicts, negotiation, or dealing with bias in the workplace, trainees will be able to enhance their employability, their on-the-job effectiveness, and their career opportunities. The demonstration focuses specifically on negotiation skills training. The project places a strong emphasis on providing a realistic learning experience, reducing the development costs of conversational training systems, and providing individualized assessment and feedback.
28a. Overcoming the challenges of ambitious and complex collaborations among university, museum, and community partners
Leading multi-organizational team projects is a challenge. Individuals and organizations are diverse in their experience, cultural perspectives, personal motivations, and priorities. The poster will describe the Collaborative Project Management (CoPM) philosophy and the CoPM Playbook, specifically designed for Informal STEM Learning partnerships between researchers, practitioners, and community partners.
28b. Accent conversion with Golden Speaker Builder
Ricardo Gutierrez-Osuna & John Levis
Golden Speaker Builder (GSB) is a tool that allows language learners to generate a synthesized version of their own voice, but with a native accent. We describe the software architecture of GSB and the speech processing methods that underlie “”accent conversion”” process.
29a. Learning Analytics for Improving College Student Success
The six-year higher-education graduation rate has been around 59% for over 15 years; less than half of college graduates finish within 4 years. This has high human, economic and societal costs. The National Research Council has identified a critical need to develop innovative approaches to improve student retention, graduation, and workforce-preparedness. The objective of this project is to develop new computational methods to analyze large and diverse types of education and learning data to help (a) discover successful academic pathways for students; (b) improve pedagogy for instructors; and (c) enhance student persistence and retention for institutions. The project outcomes are designed to help students select courses that fit their needs, capabilities, and learning styles, and are likely to lead to (faster) graduation; help instructors to better meet student needs; and give advisors and institutions the analytics needed to improve retention and persistence.
The proposed research will produce new dynamical system modeling, collaborative filtering, and multi-task learning methods. Modeling the evolution of a student’s knowledge using a dynamical state-space system is a key innovation; the proposed research will develop novel collaborative system identification and collaborative Kalman filtering techniques for grade prediction. Technical innovations include supervised learning approaches for evolving datasets, such as linear and non-linear multi-task learning and collaborative multi-regression models with controlled grouping of the latent variables. These innovations will coalesce into three pilot applications: DegreePlanner for students, CourseInsights for instructors, and StudentWatch for academic advisors.
29b. The Impact of Gesture and Distributing Story between Multiple Characters on Children’s Engagement with an Animated Story Telling Application
We are interested in developing more effective software systems to tell children stories in order to improve their listening comprehension skills. This poster describes a prototype system in which cartoon characters tell Aesop’s fables. A study evaluated the impact of using complex gestures and distributing story text among multiple characters vs. only a narrator on children’s engagement.
30a. The Role of Data Visualizations in Writing for Impact Online
This poster presents findings from a design research study that examined how students engaged in data-informed writing practices in an online community. It explores the Impact Studio, a suite of visualization tools to help students communicate and collaborate online. The Impact Studio is the student-facing side of the analytics engine powering the globally networked Write4Change (W4C) community, which virtually linked adolescents committed to writing about problems of local and global concern. W4C leverages innovative collaborative authoring and curation tools in an online makerspace as students learned to use data to write for impact. This poster presents an overview of the five Impact Studio data visualization tools and reports on findings about how 23 youth writers (ages 14-19) used these data visualizations to trace the circulation and uptake of their writing. The findings suggest students used available data analytic tools to gauge the reach and impact of their work across the community in aggregate (to â€˜visualizeâ€™ their audience as a whole over time) but did not always use information from those visualizations in making on-the-ground decisions about their writing. This raises questions about how much support educators/online environments should provide about how and when to use data to inform writing decisions.
30b. Optimizing Statistical Machine Translation for Text Simplification
Most recent sentence simplification systems use basic machine translation models to learn lexical and syntactic paraphrases from a manually simplified parallel corpus. These methods are limited by the quality and quantity of manually simplified corpora, which are expensive to build. In this paper, we conduct an in-depth adaptation of statistical machine translation to perform text simplification, taking advantage of large-scale paraphrases learned from bilingual texts and a small amount of manual simplifications with multiple references. Our work is the first to design automatic metrics that are effective for tuning and evaluating simplification systems, which will facilitate iterative development for this task.
31a. The Emotive Virtual Patient
The Emotive Virtual is a synthetic entity that allows medical school students to practice communication in to explore new paradigms for social learning through augmented reality. The Emotive Virtual Patient is fueled by the EVP App, supported by the NSF US Ignite high-speed network. With the EVP App, students can participate in collaborative augmented reality, practicing communications skills through the HoloLens and receiving feedback through a professor remotely with the EVP App. Our research orients around new social learning paradigms that may emerge in an augmented reality environment. Our research also is focused on the Future of Work in that we posit that work and learning for medical practice will be over a high speed network in a telemedicine environment.
31b. Designing Full-Body and Tangible Interactions that Aid Reasoning about Correlation and Causation
This poster mainly illustrates preliminary work on our EAGER grant that investigates how theories of cognition that emphasize the relationship between thinking and physical action can be used to design full-body and tangible ways to interact with data-based museum installations. The aim is to identify elements of the interaction that can be used to prime people to think in ways that improve their understanding of causation and correlation. We are currently testing variations of the prototype installation at Discovery Place, a science museum in Charlotte, NC, during their regular hours and after hour events.
Additionally, the poster briefly showcases two additional research challenges that we observed during our initial work: making visitors aware that the system is interactive (despite their tremendous potential to bring in patrons and to facilitate embodied learning, full-body installations have no utility if visitors do not notice them); and, increasing the discoverability of hand gestures and body movements (museum visitors cannot consult user manuals before interacting with an installation and quickly leave thinking that the system is broken if the exhibit does not respond to their control actions).
32a. Learning on the Move by Making Public History
We have developed a set of tools and practices that allow youth and other city residents (or visiting tourists) to create and share “digital spatial story linesâ€œ (DSSLs). These DSSLs use personal narrative to map archival media, currently housed in the collections of our library and museum partners, onto city neighborhoods at a walking or biking scale. DSSLs are narrative index and media delivery structures that make vibrant aspects of past and present American Roots Music and Civil Rights activism available at a personal, embodied scale. The tools we are developing support both authoring, sharing, and using DSSLs as an organic form of public history in Nashville, TN, a rapidly growing Southern city that is understood in multiple ways as “the Music City.”
32b. The use of engineering model building activities to elicit computational thinking
The purpose of this study is to respond to these issues by looking for practical ways in which computational thinking can be incorporated into higher education for engineers. By doing so, this research builds learning environments that teach future engineers these critical skills needed in the workforce. The research questions are: (1) How can model-based learning environments be designed to elicit CT? and (2) What types of CT behaviors emerge when engineering students build computational models? Data was collected in an engineering capstone course focused on food and pharmaceutical process design. Design-based research (DBR) was used to develop the learning experience and collect data in the form of student artifacts from the classroom. Student reports and programming files for the Building the Model phase were coded deductively using the definition proposed by Selby and colleagues [5,6]. Five components to computational thinking were coded: generalization, algorithmic thinking, decomposition, abstraction, and evaluation. Evaluation had the most number of themes associated to any category, while generalization only had one theme within. Decomposition and generalization were seen very scarcely throughout the data set.
33. Using the Concept Warehouse for Adaptive Learning in Mechanics
Brian Self, Milo Koretsky, Stamatis Vokos, Dominic Dal Bello, Thomas Ekstedt
“Mechanics – the study of motion and of the action of forces on bodies” is a core topic in both physics and engineering that is rife with nonintuitive concepts. Scholars have grappled for decades with developing educational approaches to complex mechanics topics. Although mechanics includes core topics across both engineering and physics, many faculty do not form strong connections between disciplinary treatments of these common principles in ways that might enhance performance and understanding.
This project seeks to eliminate equity and performance gaps in mechanics courses by (a) developing a suite of adaptive web-based tools that incorporate videos that illustrate why a topic is relevant to diverse professionals in the real world and (b) developing adaptive tests and instructional tools.
The Concept Warehouse is an online repository of ConcepTests (multiple choice questions), Instructional Tools (virtual laboratories and simulations), and Concept Inventories. We are developing ConcepTests, and creating learning modules that incorporate videos to motivate the material, short instructional interactive videos, targeted ConcepTests and Instructional Tools, and adaptive computerized testing to better assess student understanding. By tracking student interactions with the Concept Warehouse over three successive courses (physics, statics, and dynamics), we will examine student conceptual growth longitudinally as they learn mechanics.
34a. Actionable Knowledge: Refining Educators’ Practice with the Science of Learning
Translating the findings of the learning sciences into actionable knowledge for practicing educators remains a challenge. At the National College of Education, National Louis University, we have a history of educator preparation which is focused upon experiential learning to develop teachers who implement reflective and evidence-based practice. To this end, we have developed a graduate four-course core sequence which is designed to enhance teacher knowledge of learner variability and the science of learning. In the tradition of design-based research, this poster presents initial findings from an investigation into the impact of introductory concepts of the science of learning on teacher practice. In a mixed methods study, the learning artifacts of graduate teachers enrolled in a course â€œIntroduction to the Learning Sciences and Technologyâ€ were examined for emerging themes and evidence of conceptual change. Focus group interviews were held with graduate candidates to elicit participant perceptions of the extent and impact of their new knowledge on the instructional practices. Study findings will inform future research directions into educator preparation.
34b. 3D printing problem- and project-based learning
3D printing is an additive manufacturing process of producing solid 3D artifacts. Creating a 3D printed object involves designing a 3D model of the object, usually using computer-aided design (CAD) software or a scan of an existing object. The 3D model is then converted into a 3D CAD file that is used for producing the object using a 3D printer. Rooted in the maker movement, 3D printing offers real educational benefits by promoting the culture of active learning, innovation, design and problem solving. It is holistically tied to STEM education connecting multiple STEM subject areas, particularly the areas of engineering and technology. However, just putting a 3D printer in place will not create these learning opportunities. Educators need technical and curriculum support to effectively utilize 3D printing technology for teaching and learning. We will share an example of a 3D printing project that was implemented with preservice elementary teachers in a science methods course.