Thanks for checking us out Dalila. Our sample size isn't quite big enough for me to make sweeping statements about gender differences but our recent doctoral graduate featured several girls in her dissertation that indicated definitive changes in STEM interest. Overall I do think there's considerably more interest in Minecraft among boys as compared to girls but that a lot of this has to do with the context in which it's played and what's learned. Since the game is a flexible platform a lot of the popular derivatives are violence and competition oriented but there's also a large share social and community oriented play. Our server emphasizes exploration, wonder and building in response to environmental concerns and also makes use of NPC guides that show women in important STEM roles. We hope this will help to create for a more inclusive and effective experience.

Right now we change the server based on behaviors we can observe through data records and feedback. We've set up a framework of data collection methods and user modeling in preparation to create AI agents that will more actively guide players. For instance, a player who shows interest in Mars will encounter altered NPC dialogue and path trajectory suggestions based on this interest. More broadly, we've been adding less abstract worlds, moving from variations on what Earth might look like to direct space application settings like asteroid mining or investigating microscopic alien life. We're hoping these contexts will bring out more avenues for personalized learning.

Thank you so much for sharing this exciting project! Chad - your challenge is a very common teacher challenge and it is helpful to see that you have tried different formats. I am curious if you have any data about participants in the camps and if there are any interesting trends in their demographics. It's exciting to see you have a very high rate of return and I am wondering who is participating. When I taught middle schools students, I often found it challenging to engage students in activities for sustained periods of time (either CS or Spanish). Certainly some of my students would remain engaged throughout, but those students were in the minority. So, I guess my question is, are there certain populations of students you see participating more than others? If so, have you thought about how you might engage specific sub-groups that might not be participating as much yet?

Pati - we have quite a bit of data about camp participants, including surveys, interviews and demographics - and some secondary information about IEP's and school performance due to broader preexisting partnerships. The high rate of retention and engagement is as a result of ongoing partnerships with organizations that have long-standing community ties and reputation - Urbana Neighborhood Connections Center, the Champaign-Urbana Community Fab Lab and, this upcoming summer, 4-H and University of Illinois Extension.

They're not explicit 4 hour blocks of play time - but instead broken up into different kinds of game play activities, recess, snacks, talks and discussion and even off days where we might do something related, like water bottle rockets. They are typically embedded in a larger camp experience so students are doing different things in the morning and are together for a portion of the overall summer.

Some students definitely participate more than others, like most informal learning programs. I'd defer to Matt, our graduate student, who can speak specifically to data, but my impression is that major factors that impact participation include experience playing games, interest in previous programs and issues that impact all education experiences like engaging teachers or role models, ability to attend camps every day and if kids get to collaborate with their friends.

As for suggestions about how to keep kids engaged for longer - a lot of this has to do with age and choice. The older kids are and the more they can control what they get to do the longer they stay engaged, in my experience. Hope that answers your question - feel free to drop a line if you have more specific questions.

Hi Pati, I have been involved mostly with the camps at the Urbana Neighborhood Connections Center. As mentioned, we have seen several participants return to our camps each year. In fact, the first week-long camp of 2020 had 3 students participating for the third time in 3 years, and 4 students participating for their second time. We added an interview question to ask students why they return to the camp each year. We got one response saying being with friends, a couple responses saying they liked playing Minecraft, and another couple saying they thought the camp was fun and they like learning about Astronomy and there is always something new to do in the camp. We do try to add new components to the camp each year, such as the Lunar Crater map last year. The addition of new content helps continue interest. One thing we will need to be cognizant of is how to differentiate the experience for our returners and our new students. 

In terms of demographics, here is a simple breakdown of how participants at UNCC have identified in terms of gender and race the past three years: 

2018 - In total we had 9 males and 13 females. Of those that completed surveys, 14 identified as Black and 2 as Biracial. The rest are unknown due to incomplete surveys.

2019 - In total we had 13 males and 8 females. There were 18 that identified as Black, 1 as White, 1 as Biracial, and 1 unknown. 

2020 - In total we had 8 females and 6 males. There were 12 participants that identified as Black and 2 as Biracial. 

As you can see, 2020 was a hard year and our numbers were a little lower, but there has been a pretty consistent interest from females to participate in our camps, and we hope that continues to grow. 

Let me know if you have any further questions!

We've had a few approaches to this so far. The pilot was something called the Minecraft Interest Engine, which looked at the kinds of biomes explored and blocks interacted with that might relate to areas of interest.

Our new approach looks more at which NPC's and areas they choose to interact with, the elements they include in their habitat builds as well as the kinds of questions they ask or observations they make. @Matt here can speak more to the qualitative findings from camps but anecdotally I'd say there's just a lot of work and opportunity for just introducing the kids to what all STEM can include and why it matters. The way many encounter STEM in school is in the form of things like math equations and memorization, not exploration, imagination, collaboration and application.

Thanks Susan!  This is at the heart of our project really, and we are very eager to take advantage of the fine-grained data that we have available from our server. We are looking at how kids explore our maps, what (and how many) observations they make, and how far they get with quests. But we also look at other kinds of data, including some basic interest surveys, interviews, and field notes taken by researchers. Looking at these different data sources, we try to piece it all together into a profile of each learner and understand which aspects point to evidence of interesting having been triggered. 

One of our surprises was that while our camp has always focused on Astronomy (what-if questions about Earth and exoplanets), interest emerges in many different STEM areas. It is as if Minecraft is just a vehicle for their pre-existing interests to be pursued. So as informal educators and researchers, we have embraced this by not promoting specific kinds of projects, but rather allowing them to pursue camp projects as long as they can justify it to us. For example, building is a core part of the Minecraft experience, so many students pursue engineering related goals of designing habitats. Some choose to emphasize the challenge of agriculture on these worlds or simply making sure resources are available for survival. It isn't always perfect, kids will revert back into their usual Minecraft habits (building dream homes), so we ask targeted questions and do what we can to keep them thinking about the big ideas. Ultimately, if they come in and we use Astronomy as the cover story, but they leave thinking a lot about civil engineering or biology, we are very happy about that!

Right now our back-end does some rudimentary analysis on their activities and attempts to link it to STEM areas (producing a profile of STEM relevance based on play). But as of now, this needs to be far more linked to their actual play rather than just the parts of the game they interact with. We are exploring machine learning to classify their activities and hoping to strengthen potential claims about interest. 

Hi Susan, Chad mostly covered what I was going to say, but I will reemphasize the capacity of Minecraft to both interact with preexisting interest and trigger new interest in STEM. Players can develop their STEM interest regardless of previous experience playing Minecraft, too. In one of our 5-day, virtual summer camps in 2020 we had a participant who stated she previously liked science but her interest had dissipated. She had never played Minecraft, but she ended up with the highest overall score on in-game self-explanation prompts, and she mentioned in her interview how much she enjoyed thinking about 'what-if' questions. The Minecraft experience revived a pre-existing interest in science through gaining new knowledge. Another participant in our 2020 camp was a highly motivated performer (5.0 GPA in school!), and he set a record across all years for the number of in-game observations made. 

Whether learners want to gain new knowledge, explore new areas, build new habitats, or accomplish task set before them, there are many ways to play that can build on or trigger interest playing on our Minecraft server. 

Hi Mark, thanks for checking out our work. Minecraft is an adaptable environment that can accommodate a wide range of academic domains. Minecraft does allow players to explore and build underwater. Player can see a variety of underwater life, as well as explore colorful coral reefs. If you want to more closely simulate the experience of doing research underwater then you might want to customize the experience. We use a variety of Plugins, which computer code packages allowing us to alter environments. For example, we have a plug in that allows players to take measurements of environmental conditions on the moon, such as pressure, temperature, radiation, oxygen levels, and air speed. We can set these values by region so they change as the players explore the map. We also altered jumping physics so that each time a player jumps on the moon they go higher than if the player jumps on an Earth map. Even without making these backend modifications, it is still feasible the teacher could create a one-to-one marine bio research station for students to explore. I have seen a model of the International Space Station in Minecraft, and we have created our own space station to complement our hypothetical worlds. In short, there are a lot of possibilities for recreating environments in Minecraft. The level of fidelity to realistic Earth conditions is possible and depends a lot on finding appropriate packages and incorporating them into your own server. 

Thanks, definitely looking forward to creating and exploring this environment!

Thanks for your interest and comment, Suzanne. In my lab I mostly emphasize learning in informal contexts with funding from AISL (Advancing Informal Science Learning) so our research questions focus on the design of the learning experiences, interactions with the world and agents, and what interest triggering looks like in games, summer camps, afterschool, etc.

All of that said, there is so much potential for leveraging informal learning designs in support of formal learning (See NSF/CADRE's statement on it, and "crossover learning" from the 2015 Innovating Pedagogies report) that we have definitely kept our eye on the potential. Our summer camps are fairly structured and include supporting educational materials that would likely be of value to K-12 science teachers (although we have not explicitly linked them to standards beyond some discussion of the relationships to NGSS). Notably, we working with the Minecraft: Education Edition team at Microsoft to create an Astronomy lesson for their repository which will leverage what-if questions to a degree. 

I hope this is helpful and if you have any advice for us to help support teachers, we are all ears! 

Great question, Candice. I led the team from NOVA/PBS that produced the Exoplanet Lab part of the WHIMC project, and I can tell you that we were very intentional in casting the science experts featured in our game – who are also going to live virtually in our Minecraft world – because we know how powerful such role models can be for young folks. By featuring great scientists from diverse backgrounds who clearly love their work, we're hoping to send the message to students that they can do science too, no matter their background. You can check out this page that we made so that students and teachers can get to know them better.

Just wanted to say kudos on this future and providing the kids with role models they can identify with!

Thanks, Dalila!

Interestingly we've seen less traffic than we might normally. I think it has to do with the reality that lots of kids are on screens all day and trying not to fall behind in school and our normal venues for bringing participants in through class activities and teacher networks have been less possible. We're anticipating more participants come this fall when things return to normal more so.

Hi Satabdi, thanks for your positive feedback and questions!  We certainly try to do bridging-style activities whenever possible in our camps. For example, we do go "lids down" frequently (when in person) and have discussions about the the science ideas that underlie the simulated worlds. Our curriculum has instructional content designed for before and after campers enter the worlds, with the first emphasizing features of Earth and the second emphasizing the differences they may have experienced and why. We use the observations they make as a focus point of the discussions. 

We also have done group quiz games about the concepts they are encountering while exploring which was really interesting. They sort of acted like they didn't like it at the time, but then in the interviews (which were private), they raved about the group quiz telling us to do more of them. To us, this suggested some interesting about the social context we were in!

Regarding differences between players, we definitely have some guesses about how they might cluster. There are some clear differences based on prior Minecraft experience - it is partly confidence, but also desire to show what they can do and apply their knowledge. We are thinking about how best to characterize their behaviors on the server still, but have run some initial analyses of the quality of observations (e.g., how much of the maps are they visiting, how many observations do they make). They are consistent across maps and days in both, but some choose to go see a ton of stuff and make fewer observations, while others make more observations but in concentrated areas.  

thanks again!

Hi Stacey - I'm glad your son has found it to be a great outlet.

Re: setting up a similar server - we have all of the world files (https://whimcproject.web.illinois.edu/worlds/#i...) and plugins (https://whimcproject.web.illinois.edu/downloads) available on our website but you'd need a Minecraft server of your own to host it. It also takes some familiarity with Spigot (the platform) and time to configure. Our server is setup to receive random joiners from PBS Nova and elsewhere and we're working on making the self-driven independent learner experience stronger. Most kids will get more out of it in a workshop format or playing with collaborators though.

Re: FISKE / Boulder - I'll reach out via email :)