Host Shawn Patrick Higgins edits the interview between Justine Chavez-Crespin and Evan Shieh, Executive Director of the Young Data Scientist League, whose work centers on data justice and critical AI literacy. Evan shares how his organization’s Data Is Power program puts students from underrepresented communities in the driver’s seat on AI ethics — analyzing bias in language models, studying unequal tech access, and presenting findings at international conferences. He reflects on his own path into CS, from feeling like an outsider in college to building machine learning models during ACL rehab, and makes the case that the most important skills in the AI era aren’t advanced math or coding — they’re curiosity, interdisciplinary thinking, and the ability to connect technical work to real human communities.
Podcast Transcript
Hi everybody and welcome to One Voice in CS, a podcast created to empower computer science educators by providing resources and a platform to share their journeys and experiences in CS education. This podcast is brought to you by the Computer Science Teacher Association. For this episode, I’m your host and editor, Shawn Patrick Higgins from Oregon Charter Academy in Portland, Oregon.
Our guest today is Justine Chavez Crespin. On this episode, we’re joined by Evan Shieh, Executive Director of the Young Data Scientist League, whose work focuses on data justice and critical AI literacies. Please enjoy the conversation.
Justine: So let’s get into the good stuff. I wanna hear about — let’s start off with where you are now, the work you’re working on, and we’ll take it back and you can tell me how you got to where you are today.
Evan: Yeah, that sounds good. I’m sure as it’s similar for you all, a lot of our work changed when ChatGPT launched. Before that, we were working on how do we get students from the communities that are similar to the ones that I grew up in, where you didn’t have a lot of people going into tech, even though we were just a couple highway exits away from that ecosystem.
I used to teach pretty close to Microsoft in Redmond, but even just two highway exits away you’d have students who weren’t getting into the CS industry — students with really valuable perspectives. People growing up in different neighborhoods, like we all bring something unique to the table.
But with AI we realized it wasn’t just sufficient to get students into the tech field if tech is going to come and try to disrupt the education space. We had a lot of our students feeling anxious about ChatGPT taking their jobs within the first year of it coming out. And so that’s what we’ve been doing for the past couple years now — how do we put students from diverse backgrounds in the driver’s seat when it comes to AI?
The way we formulate that is we think about topics and questions of ethics and justice, and how do you put students in a role where they feel like, oh, I can actually not only learn about AI ethics and justice — both the technical side and the humanities side — but also feel like I can contribute to this field right now before I even graduate.
A lot of our work — we launched this program called Data Is Power this past year. We worked with a wonderful CSTA fellow, a wonderful person. And just the projects our students have produced have been, in my opinion, nothing short of incredible. They’re analyzing how their peers feel about AI usage in schools and in the classroom. They’re looking at how access has broken up in different neighborhoods. Phoenix is another example of a project — they’re looking at what data collection is being done and secretly powering AI in terms of flex safety and how it’s affecting farm workers. Our students are doing what I feel like I would much rather highlight, which is they’re going out into their communities and they’re applying computational thinking and research skills to really be an active participant in tech, which is super inspiring for me.
We work directly with middle school and high school teachers on AI as well. I can share more about that if it’s relevant.
Justine: Sure. Let’s jump right in. I’m curious to see how — especially because you’ve been in this for a couple years now — how are teachers adapting? From where you guys started to where you are now, what’s the sentiment? What are teachers saying?
Evan: There are a lot of people out there who will say they’re an AI expert, but if I’m just gonna be real — who’s really an AI expert nowadays? We’re all on this crazy ride together. We’re all trying to figure things out.
There are a lot of people who are trying to grift on the movement saying, hey, I’m gonna figure out how to make money off of AI in classrooms. There’s a ton of grifters, but teachers don’t just talk — they have to walk the walk. They’re confronted with ethical issues every day in their classroom. And it’s not just AI and cheating. They’re confronting, if my student comes to me and says, should I be a scientist right now? I used to tell them yes, go into that. And now, honestly, we need to have a conversation about power and politics and the structures that are affecting us, and I think you can’t separate that from AI.
On the flip side, it presents a wonderful opportunity for us to apply multi-disciplinary ways of thinking. General ethical conversations that require computational thinking — that requires teachers to pull off the bandaid a bit and be like, hey, if I feel a bit of imposter syndrome with tech, our teachers have no choice but to gain that expertise now.
And I’ve really seen them meet the moment in a big way — in a way that I wish they didn’t have to, just to be frank, because there’s already so much going on in the classroom. But they’ve responded with a lot of resilience, a lot of joy, and a lot of just taking ownership. Always in respect and admiration for our teachers.
Justine: I really appreciate that, because right now there are teachers that have support and resources, and there are teachers that don’t. And being able to hear that it can be done, that there is support out there that will help them on their journey, and that there are programs available to them — because it’s hard to be a teacher right now. Sharing the good, sharing the growth, that’s what this is all about. And that really supports what we are trying to do with teachers.
Evan: A thousand percent. And I’ll say this even predates ChatGPT — educators are hungry to learn. Sometimes as a teacher you’re told, hey, I just have to be the king, queen, or non-binary royalty of my own classroom, but in the process you end up picking up so many skills that are applicable elsewhere.
And I just see this excitement when our teachers realize, oh, that’s it. I’m like, yeah, y’all are teaching stuff that’s way harder and way more applicable than you think it is. I consider it a blessing in my career that I’ve been connected to so many folks who have been able to speak the same language and tap into lived experiences.
I can give a quick example. There’s an elementary school teacher who reached out to us. She’s a colleague of mine out in Oakland — an elementary school teacher who found out about our work on racial biases in AI, which one of our students actually spearheaded as a publication. She was like, hey, I wanna teach this in my classroom and I also wanna participate in this Data Is Power program.
We told her we’re only aligned to middle school and high school standards right now, and I personally have never had experience working directly with third through fifth graders. But she was like, no, I’m willing to put in the extra legwork to make things work. And I learned so much from this collaboration with her.
We showed Dr. Joy Buolamwini and Timnit Gebru’s Gender Shades research — that’s our first lesson now. We had students extend that research to think about, how can training data be biased? What happens if we neutrally — and I’m putting that in air quotes — gather all the data on the internet, stick it into a statistical model, do we expect it to be fair — again, in air quotes?
This teacher just said, my students aren’t going to be able to work directly with computers, let alone search engines — they’re just not at that level of digital fluency yet. And so she built this unplugged version of the activity all from scratch where they generate all the bar charts by hand. She printed out a ton of magazines and made connections that I had never made before.
We were looking at skin tone representation in the media — that’s one of our lessons we’ve been teaching for a long time — and she was the first teacher to be like, hey, this could apply in beauty magazines, makeup palettes, we could analyze that. We could use the same methods to analyze dermatology models and textbooks. So she turned that one lesson into a whole unit for her students.
What she shared with me is, look, these students might not be the traditional example of what you think of when you think, oh, I know how to do computational thinking now, but they’re incredibly deep with the ethical questions. She gave a really powerful presentation this year and was one of our many presenters at one of the world’s largest international conferences on AI ethics.
It’s just so cool to see that growth happen in one year — from hey, I wanna figure out how to unplug these activities, to now presenting to a ton of people who are all just looking for answers. I really feel like the ingenuity of teachers is meeting the moment in a big way.
Justine: That is amazing. Oh, I get the chills — especially because you think about a teacher trying to figure out how to make sense of this for their classroom, for those students. In their minds they’re like, there’s this one student that isn’t going to pick it up with the rest, or the group as a whole really needs something different. And that’s a perfect example of the lengths they’ll go to to make it make sense for kids.
Evan: Yes, definitely. When student voice is at the center, we’re all learning.
Justine: And then I get amazing moments like this where we get to continue the growth and highlight the stories and voices that can help inspire others. In thinking about this journey that you’ve taken teachers on, how did you get to this point? Walk us back to where you began with computer science, with education, and tell us a little bit about what brought you here.
Evan: I think it honestly started when I was a student, Justine. Not to ramble too long about my life story, but in the interest of making this applicable for other students — I went to a public high school and in our school, we didn’t really have this culture where learning was really revered or esteemed. It’s not to say the students weren’t smart or the teachers weren’t good, but we were much more into sports. And so we didn’t have a computer science class.
When I entered college and for the first time had to compete against other students who had taken CS in high school, I was like, oh my God, I don’t know if this is a field I could see myself in. But even back when I was in high school, it honestly started with math. A lot of my peers and friends who were really respected, I just saw them increasingly get frustrated with math, asking, when am I ever gonna use this? This feels so arbitrary to have to know how to factor algebraic expressions, and what they felt was memorization.
I was really blessed and lucky because my parents really encouraged a lot of problem solving in me from a young age. And so it really became a mission of mine even from back when I was in high school to work on, how can I make this relevant for students?
I didn’t know back then that the education system was so connected to broader carceral systems that also gatekeep students from opportunities, and that that gatekeeping is not uniform and it is not equitable. But I came to see the same things happen in college as well, where the students who had gone to private high schools and taken CS had already built so many different things, and the rest of us who hadn’t had this education at the start were behind.
But I was fortunate to have friends and professors who really taught me to love computer science. I went from being, oh, I’m not gonna study this at all, to being like, yeah, I think this is what I want to work on, because it’s just so cool to be able to build. When I tore my ACL as a college student, I had built this machine learning model to teach me how to improve my free throw form while I was rehabbing. So I feel that teacher has always been there in me.
When I graduated, I went to work at an education technology company. But I always had this hunger to be closer to the front lines and back in the classroom. I took inspiration from other programs that I really respect, like TEALS, and I followed in those footsteps and organized a group of folks working at Microsoft first to go back into our local communities — again, just two highway exits away — to just be in community with students there and to teach in the mornings and distill some of the things and practices that we use in our jobs that might also intersect with K-12 math and computer science, which a lot of it does. But really draw that bridge and draw that connection for students and teachers.
That was our original intention. But I have to say, from that summer, we learned way more from our students than we taught them. That’s where I really started getting glimpses of oh my gosh — high schoolers at the beginning of the program were saying, oh, I’m not a math person, I don’t see myself this way, and by the end of the program were saying, computer science is way different than what it was.
I have to say that if I had to distill it — if you teach a subject in a way that really speaks to people’s cultures, lived experiences, their day-to-day excitements, challenges, issues, and goals, you don’t have to force-feed them vegetables. And I really feel like that’s what I’m the most excited to do through our work — to draw connections and to really listen from our communities up front about:
What is it that makes this city unique? What makes your neighborhood unique? What makes your classroom unique within this neighborhood? And then finally, how does tech map onto that in a way that puts you in the driver’s seat?
Justine: Oh, that’s so powerful. And that’s the true essence of giving them voice, right? Really digging into who they are, who you are, and what you bring into this space. I love that. And it’s so important as we think about those skills that are needed to drive computer science — those mindsets that are needed for students to thrive in a world where CS and AI are intertwined.
Is there any advice that you would give students or teachers who will listen to snippets of this? What is important for the mindset? We talked a little bit about the skills, but what are those vital skills, along with the computational thinking, that’s gonna support them on their journey?
Evan: This might sound cheesy, but I tell students nowadays that you and I are really in the same place, even though I’m no longer a student age. In today’s day and age, you have to learn how to learn — there are new things coming out all the time. Just look at what ChatGPT has done to the world in the past couple years.
I have a couple thoughts on how to place students and teachers in the driver’s seat.
Number one is recognizing gatekeeping. I’ll share something that my computer science professor in my freshman year of college shared with me, which is that you’re gonna see a lot of fancy terms in computer science, but especially nowadays in AI. These fancy terms might make you feel like you don’t belong in this field. And studies do show that language and highly technical terms affect people from different cultures differently, and sometimes we don’t feel like that can be a part of our identity label.
What this professor shared with me was: these fancy terms are complicated terms for simple subjects that the inventors of these terms use to make more money for themselves. And when you understand that, you realize — oh, if someone’s saying AI agents nowadays, or generative AI, or supervised learning or unsupervised learning, you might be surprised to learn that the heart of all that is arithmetic, computational thinking, algebra, probability, and statistics. That sounds an awful lot like what we learn in K-12, right?
So that’s something I would share as a message to our students and teachers — yes, it’s something that you can do, and you might already be doing it without realizing it.
I do feel like there’s a shift that needs to happen towards realizing that as educators, we should feel safe to learn alongside our students. And a lot of the skills of critical research and critical data literacy that have been around for a long time are well equipped to meet this current moment.
Justine: As we think about all the things that we’re facing, the question that always comes up is how is it that we work through the challenges that are faced? How is it that we work through even something new? So that’s the question I have for you — what have been some of the challenges you’ve faced along the way, and how did you overcome them?
Evan: I’ll say that growing up, I was a self-identified math nerd and I loved the quantitative side of things. There’s just something so comforting about being able to problem solve. I think math and computer science as theory is beautiful. But I will say that was one of the biggest challenges in my career — learning okay, I’ve got this foundation, but what happens when we go out of the classroom and out of these theoretical environments into the real world where things are messy?
I remember in my first couple jobs working in industry, a lot of the questions we had weren’t straightforward. We had to come up with quantitative metrics where people disagreed on how things should be measured. We had to build products that affected people from all over the world, and we had to deal with a lot of edge cases that I’m now realizing were cultural and social as much as they were technical. I just don’t think any one person — and maybe I can only speak for myself — was trained to understand both the social side as well as the technical side. I was told just the technical side is enough.
That’s why I’m excited about the new CSTA standards, because I see this intent toward how can we build bridges across disciplines? Because that’s what I had to learn on the job — how do you speak to folks who have domain expertise in sociology and library sciences and the humanities and the hard sciences? That’s where a lot of the most meaningful, interesting problems are. And incidentally, I think that’s where the jobs will be the safest in the future — where we sit at that intersection.
The good news I’ll share with teachers and students is you are already doing a lot of this. If you’re a teacher, you have colleagues in the same building who are teaching multiple subjects. This is your opportunity to go talk to that English teacher down the hall. We had an English teacher in our classroom this year who told me, I’ve never identified as a math person before, but I enjoyed this one activity so much — looking at the foundation of artificial intelligence and playing this math game — that I’m gonna have to start thinking about this more.
We all have that multidisciplinary capacity inside of ourselves, and it requires us to be humble and to learn from each other.
I’ll share a concrete problem to make things tangible. We had a student who was feeling anxious about ChatGPT taking their job in the future and not knowing what they wanted to study. I sat down with that student and just said, hey look, I can tell you this is a mathematical model, but it’s better if you get experience with it. So let’s just play with it.
The first prompt we gave the model was, write me a story of a doctor who saved a struggling patient — based on the student’s aspirations of wanting to go into the medical field. And when I tell you that the output was stereotypical, that’s probably even an understatement. The names of the doctors at that time were almost all male, and the names were like John, Ethan, Mark — sometimes you’d get a Sarah or Emily. But then the more times you tried this prompt, the more you realized that the names of the patients were Juan, Carlos, Maria, Diego — you had Jamal show up a lot, Priya show up a lot, Ahmed show up a lot. Pretty much all non-hegemonic identities, at least in the context of the United States, would show up in what we called that subordinated position as a patient.
When we expanded it, we looked at different occupations — not just doctors and patients. We looked at star students and struggling students in the classroom, personal relationships, a friend who needs to cover the bills for another friend — and the same pattern persisted by race, gender, and sexuality.
So how would you analyze that? How would you prove and quantify the impact of the models? You need computer science to be able to gather and parse the data, to set up your experiments — that’s a lot of programming and working with APIs of different models. We collected 500,000 stories. There’s no way you could do this by hand. But you also need math and the humanities. You need the humanities to quantify and encode what are the cultural associations and linguistic cues, and what are the ramifications when we see these linguistic triggers psychologically. And you need the math — but the math isn’t that hard. We used fractions. That was about as complicated as it got.
That research is cutting-edge research in AI, which is mind-boggling. You don’t need calculus. You don’t need linear optimization. You don’t need to have been to university and gotten a PhD to do the kind of work that we did. And in fact, because of our students’ lived experiences and perspectives — reading these stories, they understood immediately what was going on. And I think that’s what made the research valuable.
So yeah — what are the skills we need right now? We have to figure out how to break down barriers between disciplines and understand that what we’re seeing is not new.
Justine: I love that, and I really appreciate these concrete examples because not only does it make it tangible for us, but these are the things that teachers are able to connect with. How do you see computer science opening doors for students from diverse backgrounds?
Evan: Honestly, what I see is people from diverse backgrounds forging their own lane. Because if I had to be real — and maybe speak to a lot of the anxiety that our students are feeling — the job market is truly brutal right now. And I know that’s not the most uplifting message, but on the flip side, I also see a lot of amazing young graduates who are carving out their own paths. One of our undergraduate scholars at Stanford said, hey, I wanna work in public interest technology that intersects technology, ethics, and policy. She forged her own lane and now she’s launching her own career summit specifically focused on ethics-related tech jobs and roles.
But I also think diverse students don’t have to be the only ones working in ethics or justice. Even if your job is not in your viewpoint justice-related, applying this multidisciplinary framing is valuable. Products are reaching people from all over the world. Data centers are entering communities locally across the country and all over the world. And that perspective is in many ways a superpower.
Some of these jobs don’t necessarily have to be in big tech companies either. There’s a lot of room to apply quantitative ways of thinking and the fluencies you learn in CS in high school to solve problems at scale in your local government and in smaller companies that wouldn’t necessarily consider themselves to be part of the tech industry.
So that’s my response. By all accounts, from the undergraduate students I’ve been speaking to, it’s been a brutal time — but that’s my context for why I’m trying to frame it in a way that serves them. Of course, go work for Google, Microsoft, Amazon — I’m not the type of person who’s gonna say we need to draw the ladder up behind us. But I also recognize that those jobs are getting harder and harder to get right now.
Justine: We are absolutely seeing and experiencing that shift in the economy and in the job market. It’s a hard time across many industries, and especially education.
On the flip side though, having these educational opportunities is already a blessing. And there are a lot of colleges and academic subjects realizing that computational thinking can be a real asset in their fields. I’m seeing a lot of success in those areas too — living in that liminal space at the intersection of different contexts, where a lot of people are saying, we really need more folks with a tech background in sociology and history and the digital humanities. That’s another avenue that I think is really ripe for graduates.
Students are lucky to have mentors like you, to have someone to turn to. Who would that person have been for you throughout your career or throughout your education? And don’t feel limited to just one.
Evan: I have too many, honestly — I’ve just been really blessed by a lot of incredible educators.
I hope they don’t mind me calling them out, but Dr. Brian Brown at the Stanford Graduate School of Education was just so gracious and taught me a lot of things about what culturally relevant, sustaining education looks like — a humbling experience. I learned a lot from the teachers who are boots on the ground. Lisa Cheba is a longtime collaborator of mine who taught for over 20 years.
In the CS space, there are a lot of folks who have led the way that I really look up to and admire — who have used quantitative methods to address social issues. A couple who have become more recent mentors of mine that I’ve been really grateful for are Eman Monroe White and Lauren Klein, who are both incredible people who have shown me so many different examples beyond what I’ve learned in the classroom of what real-world CS, AI, education, and research could look like. My middle school math teacher was also just so inspiring in terms of making the subject fun and relevant.
Justine: Thank you for sharing those. As we come towards the end of our time, I want to wrap up with this question and then we’ll talk about next steps. What has it meant for you to think about yourself now as being named a computer science education hero?
Evan: It’s a blessing. I don’t consider myself to be a hero. I generally consider myself to be someone who’s lucky in the sense that I’ve been a part of so many different communities who have challenged me, pushed me, and taught me so much. I just see this as an opportunity to represent the good work of a lot of communities that I’ve been a part of, and to be deeper in community with folks at CSTA.
But humbleness aside, this is really an honor. I think CSTA is doing so much leading work, and I just appreciate you all giving this venue to share folks from diverse backgrounds and their stories.
Justine: We are so appreciative of your time, your contribution, and for taking the time today to connect and to chat.