Donna: David, please introduce yourself and tell us about your ‘day job(s).’
Yes, it’s many jobs these days. So first and foremost, I’m a professor at Cornell University and serve as associate chair for the department of statistics and data science. I also work in the fields of applied mathematics, operations research, and computer science.
I’m the 2024 chair for the ASA Business and Economics Section. I’m a longtime member and longtime officer, and I’m excited about my continued work with this group. In addition to serving as the editor-in-chief of the journal Data Science in Science, I’m the new director of the National Institute of Statistical Sciences.
Donna: Let’s dive right in and talk about Data Science in Science. How do you see it differentiating itself from other publications? Also, what are the key features researchers can expect when submitting to Data Science in Science?
Data science has evolved a lot over the last 10 years, and different groups have latched on and taken a leadership role in that. I’ve seen statistical sciences more recently step up, as well. It’s a combination of different domains that come together to compromise data science. And a big promise of data science and AI more broadly is to make scientific advances. And that’s really what this journal is about—new science, discovery, testing, and anything in the whole workflow or pipeline that can be enabled through these new tools.
I want to say a disclaimer that we define as a group “science” very broadly. So, it touches on the social sciences, as well. I work in several domains across ecology, hydrology, agriculture, physical chemistry, and computer science. It’s a lot of fun, and here is where my passion comes in.
We aim to be a home for collaborative research that potentially spans multiple scientific domains and that’s newly enabled through data science. We hope we can recognize advances in data science, itself, whether they’re methodological or maybe new adaptations motivated or demanded by key scientific challenges. We also appreciate the importance of data visualization in scientific discovery and communicating those discoveries.
So, some of what I think of as key features are about rewarding collaboration, and in particular, recognizing the type of advances we make these days are largely driven by collaborative teams. It’s a combination of folks who identify as scientists and others who identify as data scientists.
Ron: What do you see as the opportunities in this new journal for our emerging scholars and researchers—people setting out at the start of their careers? Also, are there special issues envisioned for the future?
The special issue is the way the journal is taking some ownership of different collaborative domains, and it is an ideal path for new researchers, in particular, to get involved with Data Science in Science. So, the current open special issues include data science and modern finance. The intersection of climate and the environment is the second one, and then more recently, data science and the brain sciences. And we’ve had a lot of strong submissions in each of these over the last year.
Some of the new initiatives in the coming year include the intersection of AI and the federal government. We’re hoping to do a special issue on wastewater epidemiology.
That one’s a little bit more specific, but I think we can go a little bit deeper. High energy physics is on the short-term radar. In the intermediate term—something in space science, which is something I’m personally passionate about. And there’s a big community, especially the astrostatisticians, who could benefit from this new collaborative outlet. And then, further down the line, I’d like to see some more work in ecology and the environmental sciences. In terms of new researchers, we’re kind of an ideal fit because we see ourselves as very modern in our take on data science itself. I view myself as a data scientist from a statistical lens.
We are looking for new ways to partner and recognize and elevate new researchers, as well. So, thinking about things like ASA section awards for new student research and reaching out to those award winners and inviting them to submit to the journal. There are some other ways to partner with ongoing workshops and conferences, both with the ASA and more broadly at other data science conferences.
We also have an open call for a co-editor, and there’s a search committee putting that together now and reviewing applications, but that will stay open for the next couple of months. So do reach out. We are constantly looking for associate editors, and the ideal profile is someone who has a strong research track record of doing deep collaborations with scientists.
At the same time, I’m looking for what I think of as “scientific unicorns.” These are scientists who are also broadly recognized as data scientists, whether they’re self-taught or otherwise. They’re advancing and have full credibility as data scientists themselves. And I find those people through their papers. Oftentimes, I reach out and see where it goes, but we’re looking for a good balance and trying to reach both parts of the community. Junior folks, I believe that once you’ve started publishing your papers, it’s an appropriate time to start doing some reviewing yourself. We continue to expand our internal list of potential reviewers.
So, we do put out a lot of requests, but we always strive to get at least two independent peer reviews for all our submissions. For the associate editors, I am just looking for diversity overall. So scientific diversity and, even more broadly, geographic diversity. This is a big world we live in, and we want to reach all the communities helping us advance it.
Donna: If folks are interested in serving as reviewers, to whom should they reach out?
Connecting with me directly is ideal. Alternatively, they can connect with Data Science in Science.
Donna: How do you see artificial intelligence influencing the landscape of scholarly publishing?
Yes, these are wild times. AI demands a lot of attention, whether it’s just from the headlines or if it’s in faculty meetings. For me, this comes up in the publishing world acutely, as well. So, initially, there was a bit of a panic, to be honest. Are we going to be flooded with submissions generated by AI? That didn’t happen. And at the same time, of course, there are AI tools to screen submissions.
So, the initial take is that, as a community, we have to have policies and best practices in place. And the ASA has shown a lot of leadership in this space, giving it deep thought. I believe we don’t necessarily need to rush these, but as we roll out the policies, we need to continually review them, because this is something that is rapidly changing over time.
There are large ethical concerns, both in academia and business, when things aren’t held to the highest standards. The fallout in the publications world is that AI is not a legitimate collaborator, and the main reason is it can’t be held accountable for what’s written. And that’s something that, if you’re using AI tools, I caution you to review the official policies and make sure you’re doing it the way it is currently accepted. And always disclose what you’ve done. And sometimes, as we try to push boundaries, it’s the way to safeguard both yourself as the author and the reader, as well, so they’re not misled. And that’s ultimately what’s most important, because it’s that path that advances science.
Leave a Reply