NSF Awards: 1502882
EdGE at TERC is studying how game-based learning can transform the educational landscape. In particular, we are studying the innovative teacher practice of bridging between game-based (implicit or tacit) learning and classroom (explicit) learning with a game called Zoombinis. Zoombinis is a learning game that engages players in puzzles that require increasingly complex Computational Thinking (CT). EdGE's game-based learning research attempts to broaden participation by reaching neurotypical learners and learners with cognitive differences who may struggle at traditional school activity and testing.
EdGE research is informing STEM learning and teaching through a national implementation study of over 30 classes in grades 3-8 aiming to support a foundation of Computational Thinking through Zoombinis gameplay. Participating teachers are given student logins to a research version of Zoombinis that enables data tracking and automatically delivers an online pre/post assessment for all players. These data are used in Educational Data Mining (EDM) research to measure students’ implicit Computational Thinking such as problem decomposition, pattern recognition, abstraction, and algorithm design.
Teachers in the Zoombinis research classes use bridging materials to connect game-based learning (tacit or implicit learning) to classroom content (explicit learning). These include physical materials (posters and playing cards), kinesthetic re-enactments of the game puzzles, graphical organizers, coding activities, and more that build upon the Computational Thinking skills develop in the game. The Zoombinis bridge materials also help connect Computational Thinking to science, math, and other subjects.
We are currently studying how teachers in over 60 classes in the US are using Zoombinis along with bridge materials with a diverse audience of young learners. Participating teachers are logging their activity each time they teach with Zoombinis and are participating in period interviews to understand their bridging practices. Select classrooms are also being observed. This video provides examples of bridging practices that show promise to support Computational Thinking in grades 3-8.
Barb MacEachern
Director of Outreach, EdGE at TERC
Welcome to our 2018 STEM for All Video Showcase submission! We hope you enjoy this quick snapshot into the Implementation Study of the wonderful CT game, Zoombinis. We are currently wrapping up our work with our partner teachers and their students and will soon move into the data analysis phase. We are learning so much about our work, game based engagement, and computational thinking throughout this incredible journey! We hope your interest is piqued and you engage in conversation with us over the next week. Thank for you for time and interest. Can't wait to see the comments/questions you have!
Maureen Tumenas
I love this idea! I used to do an afterschool Zoombini contest at my former school. Do you need more classes to test this? I asked my third grade teachers if they were up for this and they were so excited!. I am a tech integration specialist/STEAM lab at a K-6 school ~300 kids.
We would love to learn more about your project and get involved!
Maureen Tumenas
Barb MacEachern
Director of Outreach, EdGE at TERC
Hi Maureen. Love the Zoombinis enthusiasm! We are currently wrapping up the implementation phase of our study but send me and email and we can investigate possibilities. barb_maceachern@terc.edu
Daniel Damelin
Senior Scientist
For people trying to find a scenario where understanding what is happening naturally leads to engaging in computational thinking, Zoombinis puzzles provide a great context.
What types of classrooms are you working in an how does the context influence how teachers make that bridge? It looks like you may be working in computer science/technology classes as well as math classes.
Barb MacEachern
Director of Outreach, EdGE at TERC
Thanks for your question, Daniel! Currently, we are partnering with teachers within a wide variety of classroom settings. We have math specialist, tech integration specialists, elementary generalists who connect Zoombinis to both math and science, middle school technology specialists, middle school engineering and robotics, just to name a few. There is a base line common ground across all of our partner teachers, they integrate computational thinking into their classes or they teach computational thinking units (mostly coding). We're focusing on 4 pattern recognition, problem decomposition, algorithm design and abstraction. So, in theory, we the teachers are in a place where they can make the connections and help build a bridge or use the bridging that we provide for several of the Zoombinis puzzles.
We're still in the implementation phase of the study and will have much more information around how classroom context influences how teachers are able to bridge game play with their curriculum. Stay tuned...
Carrie Willis
Technology Director and Teacher
Game-based engagement is definitely a hot topic right now! I love the integration with Scratch.
Jodi Asbell-clarke
Director, EdGE at TERC
Thanks Carrie. Many teachers are using Scratch but don't have a CT foundation underneath the coding activity so we are excited to see how to build this bridge.
Jan Mokros
What a great extension of Zoombinis puzzles! The game is terrific in terms of learning logic and computational thinking, and you've mined it for so much more. I especially like the physical acting out of the puzzles. Scratch extends the work very well.
I know it's a research project, but are you disseminating any of the activities themselves? We'd certainly like to see that!
Erin Bardar
Bridge Materials Lead
Thanks, Jan! For right now, the classroom activities are only available to our research study teachers. We hope to share them with a wider audience after the study is over, but details of exactly how they will be shared are still undetermined at this point.
Jessica Hammer
Looks fantastic! How are you adapting your activities to serve younger (e.g. grade 3) and older (e.g. grade 8) students? Or is the teacher doing most of the adaptation to student level and abilities in real-time?
Erin Bardar
Bridge Materials Lead
Thanks, Jessica! We've developed a single set of classroom activities, which we think will (for the most part) work across the grade levels, at least as a starting point. We tried to offer a wide variety of activity types (online, up-and-moving, etc.) and align activities to standards so that teachers can choose what they find most appropriate for their students' grade and skill levels. We also encourage teachers to adapt the activities as they see fit and to share their modifications with us in their classroom logs as well as with other teachers in our study through our Google+ community. Feedback from our teachers on the materials so far has been positive for students across the entire grade/age span.
Jodi Asbell-clarke
Director, EdGE at TERC
And following up on Erin's response, in a separate CSforALL project called CodePlay, we are building a Playlist generator that will create sequences (playlists) of CT activities that are differentiated by subject area, grade level, activity type (e.g. offline, kinesthetic), and eventually by students' individual demonstrated implicit CT knowledge in environments like Zoombinis.
Mari Strand Cary
Hi! Bringing the virtual learning into the physical world is an exciting direction to take Zoombinis. Could you talk a little about the guiding principles or professional development your teachers have received thus far? How do they know what materials to use (or how or how much)? Did you have ideas going in about which materials/components would be the most impactful, engaging, or utilized?
Erin Bardar
Bridge Materials Lead
Hi Mari,
Great questions! We don't have a formal professional development component of this project, but we have several mechanisms in place for supporting our research study teachers as they implement the Zoombinis game and bridge activities. The bridge activities are organized by Zoombinis puzzle and are tagged with relevant content standards (NGSS, Common Core Math, and CSTA's K-12 Interim CS Standards), the CT skills highlighted in the activity, and the type of activity (online, hands-on, up-and-moving, etc.). Each teacher has an EdGE team member "buddy" to help them plan out which activities to use and to answer questions about content and technical issues throughout their participation. We also have a Google+ community for our teachers to share ideas and materials as well as monthly video calls with teachers and EdGE team members to discuss how things are going in the classroom and what successes and challenges they have encountered using the game and classroom activities. For the study, we have asked that teachers devote a minimum of 10 classroom hours teaching, discussing, and/or doing activities that support computational thinking, including Zoombinis gameplay and bridge activities. The teachers keep logs to document what they do in class and how much time they spend on different types of activities so that our researchers can determine how different implementation strategies impact development of students' CT skills.
Rebecca Vieyra
I'm eager to learn more about how contextualization of the classroom influences what teachers and students gain from this experience. So, that leads me to "How and what are you assessing?" It looks like from a response above that you are looking at measuring pattern recognition, problem decomposition, algorithm design and abstraction --> what tools are you using to measure growth? And, assuming some of these skills are already being taught and gained in contextualized instruction such as math and science, how do you tease out what impact the game has over CT already embedded in other content areas?
Jodi Asbell-clarke
Director, EdGE at TERC
Hi Rebecca! We are using the students' own gameplay, the patterns of behaviors they exhibit as they solve the puzzles, as implicit assessments of their CT. We then watch how that changes over time. We compare those results to pre/post online assessments we've developed and validated with Val Shute and her colleagues at Empirical Games. We are using information about the extent of game play and the extent of Zoombinis related bridging activities in class as variables to see how they impact students' learning - so we can hopefully tease out how much those variables impact their learning. I hope that clears it up for you :)
James Diamond
Research Scientist
Hi all—This is very exciting, and I love the video! Have you ever had teachers ask about the evidence for how the gameplay is an indicator of student learning? (Or, I’m asking anyway…). That might be helpful if a teacher wanted to think about asking students to replay parts of the game if a student seems to be struggling with a particular CT skill.
I’d love to talk more at some point about whether and how you’re doing something like that, as my team and I are working on a DRK-12 in which we’re trying to help teachers make connections between the game play and the targeted learning objectives.
Great job!
Jodi Asbell-clarke
Director, EdGE at TERC
Hey Jim! yes we are doing a lot of work with teachers about how they connect CT in games with classroom activity. Let's definitely talk more about the DRK12. Email me at jodi_asbell-clarke@terc.edu and we can set up a time.
Joseph Reilly
Very interesting! The near transfer of skills learned in game to other parts of the classroom is too often ignored or assumed without evidence. In regards to the detectors of implicit learning, how similar are your approaches to those you utilized with Impulse? I'm interested in stealth assessment and the ECgD framework so I was curious if you were utilizing a similar method of hand coding videos, engineering features from the log files, then testing different detector algorithms. While the bottom-up approach makes sense, re-engineering these features for each game must be time consuming. Thank you!
Jodi Asbell-clarke
Director, EdGE at TERC
Hi Joseph - while not exactly like ECgD, because we are more emergent, we end up with a similar result. We start by observing and recording the gameplay of about 70+ players at all different levels to get a good sampling of strategies used to solve the puzzles. We then distill the data logs to provide features that can be mined to automate the detection of these strategies....so it is not exactly bottom-up either. We build detectors based on the groundtruth of extensive observation and human-labelling. The labelling does take a lot of time, but then it can be replicated and scaled to much greater numbers....I hope that answers your question!
Shari Metcalf
I've been a Zoombinis fan since it first came out, and remember getting it for my nephews when they were growing up. I love that you're doing research with it now, and I think it's a great springboard for teaching about computational thinking.
I was wondering how the data mining work is going, and what you've been able to find out so far about kids game play strategies, and to what extent you're able to identify or evaluate through log files the computational thinking skills you've described.
Jodi Asbell-clarke
Director, EdGE at TERC
Hey Shari - we have labelled two of the four puzzles so far and are reaching very good reliability - that part does take a while. We expect to be done with all four of the puzzles we are studying by mid Fall. Thanks for your interest!! :)
Trevor Haney
I have always enjoyed Zoombinis and had them growing up. I think it is very important to make learning fun and exciting and I can’t imagine a better way to do it than with Zoombinis. I really like the acting out of the puzzles so the learner can have even more fun and interaction with the project. Do you believe that this style of learning can, and should be implemented at a high school level ? If it was to be introduced at the high school level what would be different if anything ?
Barb MacEachern
Director of Outreach, EdGE at TERC
Hi Trevor,
Yes, the implicate learning that takes place during game place can be made explicate at any grade level! We created 3 HS science apps a few years back for a research study we called Leveling Up and witnessed many teens making connections to science concepts as they played the game. I'm linking our research page from our website in case you'd like to read more.
Thanks for the positive Zoombinis words!
Barb
Erinne Lynch
This brings back great childhood memories! Zoombinis was my all time favorite computer game and I remember just how challenging the journey could get. As an educator, I often struggle teaching students number sense or building their computational thinking skills. With the increase in research and implementation of gaming and coding in the classroom, it's great that your team is trying to connect the two through the use of this wonderfully challenging and addicting game. It will be interesting to see how students choices and decisions throughout the game relate to their abilities in math and science. I am anxious to follow your results and will be on the lookout for an opportunity to try this with my 5th grade students! I wonder what your next steps will be? Are you hoping to make some of your resources available to the public? Thanks for sharing!
Jan Mokros
My kids grew up with Zoombinis, and my son wanted to be like Chris Hancock. It is wonderful how playful it was, and at the same time encouraged so much mathematical thinking. One piece of trivia: The name Zoombinis was not Chris and Scot Osterweil's first choice. Chris already had these little critters labeled something else, and it took a long time for him to get used to their new names!
Further posting is closed as the showcase has ended.