Apr 25, 2019 | Atlanta, GA
Like the majority of research in IC, it comes down to the people
Miranda Parker was early on in her time as a Ph.D. student in the School of Interactive Computing (IC) when she began her first quantitative study. She wanted to see whether they could model the variables that influence whether a school would or would not adopt computer science (CS) as a class for its students.
Prior to the study, the hypothesis was that variables like median income or enrollment numbers or the population of students who qualify for free and reduced cost lunch programs could be an indicator of whether or not computer science was implemented. Lower income levels, for example, might correlate to schools that just didn’t have the resources to deploy such programs.
Somewhat to Parker’s surprise, the short answer to that question was – no. No, a higher median income didn’t mean more computer science; no, schools with lower free and reduced lunch numbers didn’t teach computer science at a higher rate; no, higher enrollment didn’t necessarily mean more young students yearning to learn how to code.
On the surface, that first study might have felt like a failure. If the goal was to prove that income disparity equated to a disparity in who was gaining exposure to a key part of their education, then it may be fair to describe it as such. However, Parker looks back on that study as a key component of what has guided her research at Georgia Tech ever since.
It wasn’t a failure, she said. It just helped open her eyes to some realities she may not have noticed otherwise.
“Part of me wanted my first study to fail because part of me didn’t want to be able to say, ‘Oh, yes, these three things mean more computer science,’” she said. “Sure, it’s snazzy. It’s easy to put on a Facebook post. But it’s so much more complicated than that. And I’m glad that it’s more complicated than that.”
Over the years, Parker, who studies human-centered computing with a focus on computer science education, has gained a deeper understanding of what might influence a public high school in Georgia to offer computer science education. None of the above items are among them. What she said has shown some correlation is a bit more complex.
“If a school had computer science in 2016, the correlation was that it also had computer science in 2015, 2014, and 2013,” she said.
Okay, but how did it get started in 2013? That’s part of the question her research is trying to uncover.
“That’s an endless cycle,” she explained. “You had it before, now you still have it. But how did you get it to begin with?”
One thing she’s learned, which can be said for a majority of research in IC, is that it comes down to the people. Who is involved with a school and what connections do they have to a particular subject? If a connection has worked in CS in the past or may be passionate about adding that to the school, the results indicate the school is much more likely to employ that subject.
Makes sense, right?
“If a school has someone who can teach computer science and there are parents saying we need to teach computer science, then whether it’s rural or urban or high or low income, it doesn’t matter,” Parker said. “They will have computer science. But if there’s no one there to push them, it’s much less likely.”
It’s not just a person, either. Organizations like Georgia Tech’s Constellations Center for Equity in Computing and the Center for Education Integrating Science, Math, and Computing are also championing K-12 CS educational opportunities.
But, Parker said, being successful is a bit more complicated than just serving CS up to the masses in communities that are unfamiliar with these and other organizations.
“Computer science isn’t the end all, be all,” she said. “If a school is in a more agricultural-based county, that may benefit the school more than a heavy computer science program would. It’s about finding how computer science can most benefit students in different ways for different areas.”
The most encouraging thing about that research, Parker said, was that the failure of her original study showed her one important piece of information.
“You don’t need high income to have computer science,” she said. “It really can be for everyone. That’s an important piece of information to know.”
Parker is aiming to finish her Ph.D. work in the fall and will decide between pursuing a faculty position, which she is leaning toward now, or other opportunities that may present themselves down the road. Former Georgia Tech Professor Mark Guzdial, now a faculty member at the University of Michigan, is Parker’s advisor.