The ATP ECPS Summit will broadcast live on December 3rd and 4th, 2020. All times EST unless otherwise noted.
11:00 AM - 12:00 PM | Session 1: Opening Keynote: Digital-first Approaches to Scaling Learning and Assessment
Today, there is increased demand for tools and technologies that support learners at scale. In this talk, Dr. Luis von Ahn, co-founder and CEO of Duolingo, will discuss how to use technology to design learner-centered solutions for the future of education. He’ll also highlight how AI and machine learning can help make language learning and testing both equitable and effective.
12:00 PM - 12:15 PM | Break
12:15 PM - 1:00 PM | Session 2: A Conversation with R&D Leaders: Making Learning and Testing Better with A.I.
In recent decades, artificial intelligence and machine learning algorithms have permeated many aspects of our lives. From the way we work, play, and interact with one another, there is hardly an area of daily living not transformed by A.I. In this session, research and development leaders from education and testing organizations will share how they incorporate A.I. technology to build better learning and assessment products. The panelists will review edtech trends and debunk common misconceptions, as well as discuss opportunities and challenges in the field.
Speakers: Ada Woo (Moderator), Chief of Staff and Senior Director of the Assessment Sciences Unit, Ascend Learning; Sergey Karayev, Head of AI for STEM, Turnitin; Saad Khan, Chief Innovation and Research Officer, FineTune Learning; Josine Verhagen, Senior Director of Psychometrics and Data science, Kidaptive; Nathan Thompson, CEO, Assessment Systems
1:00 PM - 1:30 PM | Break
1:45 PM - 2:30 PM | Session 3: Hot Topic Demos
Demo 1: HealthAssess - A simple, single solution to teaching health assessment and helping nursing students prepare for nursing practice.
The ATI Nursing Education HealthAssess™ product focuses on the development of assessment skills that are relevant to current nursing practice. It can be easily integrated into existing health assessment courses to facilitate teaching and learning, including examples to address various cultures and health alterations. This product offers repeated opportunities for learners to practice health assessment techniques as well as document findings in a safe, non-intimidating simulated environment. Learning modules and virtual simulation modules are delivered in a single solution to seamlessly weave the assessment content from learning to practicing. Learners review techniques to properly assess body-specific areas through tutorials, interactive learning activities, virtual demonstrations, and drill questions. Join us to talk with the product manager and psychometrician that worked on the development from the product’s inception through to final testing and hear about what it took to build a product which helps nursing students become proficient frontline nursing care professionals.
Presenters: Megan Jackson, Senior Product Manager of Clinical & Simulation, Ascend Learning; Ye Lin, Associate Psychometrician, Ascend Learning
Demo 2: Assess.ai
Learn about Assess.ai, a comprehensive ecosystem for educational assessment and professional testing, designed to improve test reliability and your team’s efficiency with modern psychometrics and AI.
Demo 3: What do we stand to gain by building equitable assessments?
Equitable access to education and assessment benefits the learners themselves and society at-large. In this hot topic demonstration, hear the stories of students for whom access to education will positively transform their lives and communities.
Speaker: Jeffrey Tousignant, Head of Marketing, Duolingo English Test
Demo 4: ATA's Newest AI-Assisted Remote Test Proctoring
In order to meet increasing demands for remote test proctoring due to the recent Covid-19 pandemic, ATA has recently upgraded its AI-assisted remote test proctoring (AI-ARTP) system, based on findings and lessons from its recent large-scale applications in China. Through vivid live-audience participation, this demonstration is to highlight its new key features, including its smart-phone and computer-camera dual monitoring, AI-identifications of a wide range of suspected security violations from improper examinee behaviors to prohibited objects, latency analyses and eye-tracking. Actual measures on security and protections of data, network and privacy will also be explicated. It is hoped that the audience will gain further concrete understanding and experiences on what constitutes a trust-worthy AI-assisted remote test proctoring system with regards to its industrial-grade reliability, accuracy and ethics.
Speakers: Alex Tong, ATA; Neil Yao, ATA; Lingxiao, ATA
Demo 5: Exam Content Crowdsourcing: Overview and Demonstration for Generating Items
pValue is an online marketplace that connects content and community experts with certification providers, who are looking to launch new exams. We are changing the business model of content generation and are modernizing the process to launch and create new exams. Our value is to add rigor, psychometric validity and efficiency to this new exam crowdsourcing model. Our demonstration will consist of walking-through our community content tools and the process for launching new crowdsourced exams.
Speaker: Kate O'Connor, Certiverse
2:30 PM - 2:45 PM | Break
2:45 PM - 3:45 PM | Session 4: Networking Roundtable Event with Hot Topic Demos
After watching our Hot Topic Demos, come join the presenters to have open discussions on their topics. Ask questions of the presenters and network with fellow attendees during this Roundtable Event. If you want to ask a question(s) of more than one presenter, no problem! We will have it set-up in a way where you can jump to multiple tables to converse with multiple presenters and attendees.
11:00 AM - 11:45 AM | Session 5: Day 2 Keynote: Adaptive Learning in Mastery-based Learning Environments
In the past decade, adaptive learning has transformed thousands of K-12, higher education and professional curricula. Dr. Ulrik Juul Christensen, Executive Chairman of Area9 Group, will explain how closed-loop learning and a specialized type of AI has been the basis for Area9's adaptive platforms used by over 30 million learners. He will explain how adaptive platforms can enable mastery-based learning at scale and summarize the impact on learning.
Speaker: Ulrik Christensen, CEO of Area9
11:45 AM - 12:00 PM | Break
12:00 PM - 12:30 PM | Session 6: Invited Short Presentations
This session will consist of the following 10-minute presentations:
Characterizing the quality of simulation-based task: an edit distance-based nonrandomness measure
Typical responses to a large family of simulation-based tasks are often in the form of sequences of actions. Different sequences are usually scored as different performance levels based on corresponding scoring rubrics. In practice, the scoring rules often need iterative improvements based on the responses from pilot tests. This iteration process is usually driven by experts' judgments rather than objective measures. In this talk, I introduce an edit distance-based nonrandomness measure to characterize the quality of simulation-based tasks and show how to apply this measure to a simulation-based task from NAEP TEL.
Persuasive Visualization of Learning Analytics to Increase Student Engagement
We evaluated the effectiveness of personalizing three social influences strategies (social comparison, social learning, and competition) in motivating students to engage in online learning activities in a context of a large 1st year biology class in our University. The system takes into consideration students’ privacy while providing them with personalized persuasive visualizations of their class assessments and offers students opportunities to either compare their performance with other students’ performance in the course, observe other students’ grades, or compete with other students as a way of motivating them to increase their engagement and improve overall performance.
The Master(y) Model for Learning
Historically, measurement and learning have evolved independently. Recently (e.g., Deonovic, et al., 2018) efforts have been made to close the gap and connect the two bodies of research together. Network models are at the center of this integration. To support a comprehensive real world data driven approach to integrated measurement, learning, and navigation of learners, models are needed that encode prior knowledge and insight from both fields of research. The ACT master(y) model introduced here is such a model, it encodes the hierarchical nature of skills, their intricate prerequisite relations, and the way in which real world tasks draw on a multitude of skills.
12:30 PM - 1:00 PM | Break
1:00 PM - 1:45 PM | Session 7: Peas in a Pod Discussions
Peas in a Pod Discussions are informal conversations with fellow summit attendees who share common interests. Pods do not include presentations. There are no projectors or slides. This is all about direct engagement and exploration of ideas.
Peas in a Pod 1: Does Psychometrics need to step aside? Considerations as learning design and assessment for achievement are combined in the Ed tech space.
As organizations and educational institutions compete to best serve the broad range of learning communities, simulation resources that are integrated with didactic instruction are becoming more prevalent as viable options in mainstream education. Educational institutions and learners ask for learning resources that are more interactive and closer to real life experiences as a means to close the gap between what is learned in their educational program and what needs to be accomplished on the job. To ensure the “educational program to employment-readiness gap” is addressed, instructional design or learning design is employed to concisely identify learning objectives that must be addressed, design learning content that can be packaged into designated platform containers, build assessment tasks integrated into the product that help stakeholders determine if learners have mastered the learning objectives, and integrate a research plan to assess the efficacy of the product. While this integrated approach is not new, does the increasing availability of the technology to support such solutions increase the likelihood that educational simulation solutions will displace assessments of achievement that historically have been built upon and evaluated using traditional psychometric characteristics and criteria. In this Peas in a Pod session attendees will be encouraged to share their perspectives on this topic as well as challenge each other to either support or refute these claims.
Speaker: Christine Mills, Director of Research and Applied Psychometrics, Ascend Learning
Peas in a Pod 2: Characterizing the quality of simulation-based task: an edit distance-based nonrandomness measure
Join Jiangang Hao, Research Director, Educational Testing Service from one of our Invited Sessions to continue the discussion and ask any questions you may have about their presentation.
Peas in a Pod 3: Persuasive Visualization of Learning Analytics to Increase Student Engagement
Join Julita Vassileva, Professor, Computer Science Department, University of Saskatchewan, Canada from one of our Invited Sessions to continue the discussion and ask any questions you may have about their presentation.
Peas in a Pod 4: The Master(y) Model for Learning
Join Gunter Maris, Senior Director of Advanced Psychometrics, ACTNext from one of our Invited Sessions to continue the discussion and ask any questions you may have about their presentation
Peas in a Pod 5: Teachers - The Critical Link
Teachers are the link between the various EdTech tools discussed during this summit and the learners. How can we support teachers and provide them with the knowledge and training needed to maximize the various EdTech tools entering the classroom?
Speaker: Tracey Hembry, Ph.D., Director, Education and Senior Psychometrician, Alpine Testing Solutions
Peas in a Pod 6: Defining “Equity” in the Testing Community
The summit theme of “Building Equitable Pathways to Achievement” is a worthwhile pursuit, but broad in nature. In this informal conversation, conference attendees will discuss and define equity and access in various testing domains and share the steps that their organizations are taking to address equity in education and assessment.
Speaker: Jennifer Dewar, Director of Strategic Engagement, Duolingo English Test
1:45 PM - 2:00 PM | Break
2:00 PM - 2:45 PM | Session 8: Equity and Technology for Educational Progress
Recently the adoption of technology in education was accelerated by the pandemic. Schools, teachers, parents and the industry have scrambled to find good ways to continue supporting the education triangle: teaching, learning and testing in virtual settings. We have learned that some aspects of the new tech-driven affordances work well. However, this new context also shed light on the gap in equity. Technology and access to perform any EdTech is not ubiquitous for everyone. In this panel we will discuss the lessons learned, discuss examples of tools that support student’s learning equitably, and potential solutions to the access gaps.
Speakers: John Kleeman(Moderator), Executive Director and Founder, Questionmark; Nori Wadaiei, Executive Vice President & Co-COO, Tokyo-based EduLab, Inc.; Ardeshir Geranpayeh, Head of Automated Assessment and Learning, Cambridge Assessment English; Deb Serri, VP Simulation, Ascend Learning; Kristen DiCerbo, Chief Learning Officer, Khan Academy
2:45 PM - 3:00 PM | Break
3:00 PM - 3:45 PM | Session 9: How Learning Sciences, AI, and Big Data Are Shaping the Future of Education
The educational challenge of the pandemic requires a fast and scalable response to teaching, learning and testing. EdTech seems to be providing that response through numerous AI and scientific breakthroughs. From adaptive learning and personalized tutoring systems to digital-first tests with automated content generation, automated scoring and home delivery - the possibilities are boundless. The unprecedented disruption of COVID-19 leaves little doubt that advances in learning sciences and AI are already shaping education. This exciting technological progress is projected to provide a wealth of opportunities to improve equity in education. In this session, the panel of experts representing learning sciences, policy makers, test publishers, and founding agencies will address the vulnerabilities in our current educational system and the role of EdTech as a possible solution to a high-quality educational experience for all learners. The overriding question for the experts is “How likely is it that the educational experience will be transformed?” The experts will engage the audience in a thought-provoking discussion on the impact of AI in education.
Speakers: Alina von Davier (Moderator), Chief of Assessment, Duolingo; Mark Schneider, Director, Institute of Education Sciences; Kumar Garg, Managing Director and Head of Partnerships, Schmidt Futures; William G. Harris, CEO, Association of Test Publishers; Bror Saxberg, VP of Learning Science, Chan Zuckerberg Initiative
* Please note: Schedule is tentative. Check back regularly for any updates.
ATP ECPS will be a two-day summit taking place on December 3rd from 11:00 AM EDT to 3:45 PM EDT and December 4th from 11:00 AM EDT to 3:45 PM EDT. The summit will include a variety of engaging sessions, including keynotes, panel discussions, and demos, throughout the summit.Learn more!
Looking for a unique way to reach the niche audience of assessment professionals, high-profile educational institutions and high-tech companies? Sponsor ATP ECPS and connect with industry leading companies.Customize your sponsorship package!