#112 Data Journalism in the Age of COVID-19

Hi, I'm Maggie, VP of Curriculum at datacamp, and I know firsthand how overwhelming it can be to keep up with the rapid changes in data and AI. But staying ahead doesn't have to be hard. That's why our curriculum team has curated the best learning content to help you and your team stay ahead, covering everything from data and AI literacy for busy executives to cutting edge generative AI for developers, and everything in between. Whether you're upskilling yourself or your team, we make learning practical, engaging and flexible. It just works. Power your data and AI transformation journey. On datacamp.com. You're listening to Data Framed, a podcast by Data Camp. In this show, you'll hear all the latest trends and insights in data science. Whether you're just getting started in your data career or you're a data leader looking to scale data driven decisions in your organization, join us for in depth discussions with data and analytics leaders at the forefront of the data revolution. So let's dive right in.

Hello everyone, this is Adele, Data science educator and evangelist at datacamp. One thing we covered a couple of months ago during Data Literacy Month was how data journalists curate data stories for the wider public. You know, over the past two years, data journalism has definitely hit the mainstream, especially given one of the biggest stories of the decade, COVID 19. And there's no better person to talk with about the state of data journalism today than Betsy Lady Zette. Betsy Laitizetz is a science, health and data journalist focused on COVID 19. She runs the COVID 19 Data Dispatch, a publication that provides news resources and original reporting on pandemic data. She's also a journalism Fellow at documenting COVID 19, a public records and investigative project housed at Muckrock and the Brown Institute for Media Innovation. Her work has appeared in Science News538, MIT Tech Review, the COVID Tracking Project, and other outlets. Throughout the episode, we chat about the state of data journalism today, the skills needed to break into data journalism, how the media and public institutions succeeded and failed in reporting on COVID 19 and what the future of data journalism looks like and more. Now onto today's episode. Betsy, it's great to have you on the show. I'm excited to speak to you about the state of data journalism today, best practices delivering data stories. Your work covering COVID 19 and more. But before, can you give us a bit of a background about yourself and your work?

Yeah, so I'm a data journalist who is on the science and health beat, mostly writing about COVID at the moment. I have a hybrid job. So I write the COVID Data Dispatch, which is a blog and newsletter about COVID data, mostly focused on the US I work part time at Mukrock, which is a public records data and investigative nonprofit. So I'm mostly again doing Covid and public health related stories and then I freelance for outlets like Science News, FiveThirtyEight.

So I'd love to set the stage for today's chat by actually discussing the state of data journalism as an industry as a whole. The past two years have definitely been a boon for data journalism with the coverage of COVID 19 election seasons and more. Given this context, how have you seen the landscape of data journalism evolve over the past few years and maybe expanding it slightly? How have you seen the appetite of different audiences you serve change as the field has evolved?

Yeah, I think the pandemic really created a gigantic appetite for data journalism. I mean, I think back to early 2020, when everybody in the world was just starting to learn about the scale of crisis that Covid would become, and we started to see these gigantic dashboards. I think John Hopkins was one of the first, and then a lot of news outlets like the New York Times, Washington Post, so forth kind of created their own dashboards to provide case data and testing data and hospitalizations and basically any metrics that they could get their hands on. And I think people really had an appetite for that. We saw folks who wanted all the data points all the time. And at that time, I volunteered for the COVID tracking project, which also was a major source for these kinds of metrics and the COVID tracking project. Every day after our team of volunteers updated the data, there would be a Twitter post sort of sharing the day's numbers, and immediately it would get hundreds of likes and retweets and so forth and people commenting on what the day's numbers were. So I think that kind of real time interest was really unique. But as the pandemic has gone on, I've seen my colleagues who work with data trying to be more constructive in looking for what audiences actually need on a day to day basis, rather than just throwing a gigantic dashboard at you with every possible metric. We're thinking more about what are the questions that readers have and how can we answer those questions with data? How can we provide audiences with local data or local information about their communities, which I think is what people really find useful and find actionable these days.

That's really great. And adjacent to the rising interest in understanding how Covid is spreading or Covid data, have you seen also the Rising interest generally in data journalism, covering all sorts of topics from health data to like election data. Have you seen the appetite also evolve on that front?

I think so. I mean, I am still relatively new to this field. I graduated college about three and a half years ago, so most of my career kind of has been the pandemic. But I do think when I look at like journalism job postings and stuff, it seems like all so many newsrooms want to have data people on staff now, although that's still more of a niche within journalism. But I think there's definitely an interest and I know like my colleagues who are in the kind of science and health beat, a lot of those folks are interested in gaining more data expertise.

And I really want to discuss the ins and outs of having covered COVID 19. But before I want to focus on the skill set of the data journalist today, starting off maybe with a skill set, I'd love to know from your perspective, what are the different skill sets needed to break into data journalism today and how does it differ from traditional roles such as data scientists or data analysts?

Yeah, I think it can differ a lot based on what you want to do. People often see data journalism as a kind of a niche or assume that all data journalists have the same skill set. But I think there are really like sub niches within that. Right? So there are people who really focus on coding. Maybe they're like a whiz in Python and they can do really complex analysis. There are other folks who more specialize in visualization, like building those dashboards or making really unique visualizations in like D3 or JavaScript. And then there are folks like, I would say I fall into this third category focusing more on exploring, explaining data, writing about data, or maybe using data to kind of power investigative reporting. I think those are other kinds of categories. So depending on which area you know, one is interested in, you can sort of tailor the kinds of projects that you take on or the kind of skills you focus on to that. And as for how data journalism differs from other data roles, I think in my view the focus is really communicating data to the public. Like if you were working as a data scientist, say in a corporate role, you would probably be focused more on communicating within your company because that's what the needs are and what your role is. But as a data journalist, I'm really thinking about is my work going to be accessible to a really broad audience and are people with kind of a limited data background or a limited science background going to understand what I'm doing? I Find that data journalists think about data as like a source, right? We interview sources, maybe we treat documents as sources, we treat like scientific papers as sources and the data. A data set can be a source that needs to be interviewed or needs to be. You need to ask it questions and then you need to come up with some kind of insight that you're going to share with your audience.

That's really great. And you mentioned here one thing is the distinction between different roles is that communication happens to the wider public as opposed to within an internal setting. What are the nuances associated with delivering data stories or communicating data to a wider public that you know may be lost on folks who work in internal data roles?

I think really the consideration of making your work accessible is so key to me both when I'm dealing with data as a more complicated source and then also dealing with topics like Covid that are themselves more complicated. Like you really want to think about what are the really big takeaways that you want your readers to get from something. And in that case maybe it's okay if you're not like delivering all of the super complex or niche or most interesting pieces of information, but really giving people a takeaway that's going to be useful in their day to day lives or is going to answer like a burning question that they have. Although I think people can handle complexity and maybe we can talk more about that later.

Definitely. And given this conversation on skillset is just how niche certain data journalism roles can be. And I think for listeners who may be interested in breaking into such a career path, the path is definitely not charted. As you think about different data roles such as data scientists, data analysts, while still not established and nascent in a lot of different ways, the blueprint to becoming a data scientist and data analyst is relatively established. You think about learning, whether in university or online courses, doing Kaggle, then portfolio projects, getting your name out there. Similarly, what does a blueprint look like for data journalism roles today?

I think it can be similar in terms of taking courses. Although I will say, at least in my experience, many of the data journalists I know did not necessarily go to school for it. That's certainly the case for me. I double majored in English and biology in college. I took one course that involved some R, but that was the extent of it. I was definitely not like a computer science person and I know a lot of folks who are in similar positions where maybe they start off in a general reporting role and then become more interested in data, or maybe they take advantage of journalism courses or One association that folks can get involved with is the IRE Investigative reporters and Editors. They are investigative and data focused. So they offer a lot of training courses, boot camps, which I have done a couple of times. Those are really great. There's also societies like the Data Visualization Society, other kind of communities that you can find to help identify courses or even just get feedback from other kind of data journalists on projects around work.

And you mentioned here projects. I think portfolio projects is a huge aspect of breaking into data science and just in any data role really. Walk us through what are the different types of portfolio projects you can find in a data journalism career path versus other traditional data roles.

Yeah, so I can talk about a few of my own projects. Although I mentioned I focus on like explanatory and investigative stuff. I have done some of those other kinds of projects that are more analysis focused, they're more visualization focused. So I mentioned I work for Muckrock, which is a kind of public records investigative outlet. And so a lot of the projects I've done there are using data as a tool to interrogate a broader question. And then I've also done projects for like Science News is a science specific news outlet in the US where I've done some like visualization based stories for them. One recent example of a piece I think is coming out soon is I produced a map of clinics offering long Covid treatment in the US So that story was really like trying to compile a list from a couple of different sources and then just making a giant interactive map for people. So that's one type of story. You can also do things that are more like building a novel analysis. I did a story over the summer for Gothamist wnyc, which is a local outlet in New York, basically looking at how PCR testing access had declined in the city in early 2022. So that was kind of me taking a public website and analyzing it and sort of providing novel data set that had not existed before. And then of course there are also giant dashboards or trackers that folks can work on. So like the COVID dashboards or an election tracker, things that can kind of be like a bigger project that gets updated over time.

I love these types of projects and you know what definitely jumps out at me from listing all of these projects is definitely there needs to be some form of novel analysis there. It needs to be a data set that is publicly available that people need to dig into. The public has burning questions around. It needs to be some form data visualization focused. Would you agree these are three main principles for a solid data journalism portfolio project.

Yeah, definitely, I would agree. Focusing on public data or sort of questions of public interest, because I think that's how you're going to get readers.

And I love how you focus here, especially on your own relevant experiences. And I think this segues well to discussing general best practices for delivering data stories as a data journalist. A key component of being a successful data journalist is delivering data stories. And that requires balancing the sophistication of a data story, but also the accessibility, which is something that you mentioned earlier when discussing best practices for delivering stories for a wider public. So as someone who is working on the front line of really complex topics such as COVID 19, maybe walk us through the key principles that you've learned in delivering stories the wider public.

Yeah, so definitely there is that balance between wanting to have a simple accessible takeaway, but then also not dumbing it down. Because I find in writing about COVID that readers really can handle complex topics. Like if I want to go in depth on say, how hospitalization data works in the United States, I can give an in depth explanation and people are going to read it and engage with it. But also some folks are going to stop at the headline or are going to stop at the first few lines of the story. So it's thinking about how you structure your work using those classic journalistic principles of an engaging lead, a clear nut graph that all applies for a data story as well. And then for those readers who are interested in the complexities or are interested in knowing how you got to the conclusions that you did, sharing your methodology, sharing your source information, like acknowledging the caveats of the data or the caveats of the analysis, all of that is super important.

Yeah, I completely agree. And maybe like, as an advice for aspiring data journalists, would you advise them to err on the side of complexity or accessibility when making a lot of these different decisions?

I don't think I can say one thing. I think it depends on the project. I think that you can always like in a data story, as in really any kind of writing, like you have to get another person to read it and give you their feedback. Like, this is why editors are great. They can tell you if you you are being too confusing or if you're failing to provide the general takeaway. So that can be kind of a helpful way to figure out which direction to go in.

Definitely rely on the editors. Now another key component here of delivering data stories is also the data visualization side of things. Right. Walk us through some of the Visual best practices that you've learned across the years of delivering high quality data stories.

Yeah, so this is definitely one area where I consider myself less of an expert compared to many other data journalists. But. But I do a lot with those simpler tools like Flourish and data wrapper and stuff I always think about is trying to keep it as simple as it can be. You don't want to, as I said, you don't want to give people all the data points at once. You want to make sure they know what they're looking at. Thinking about colors, what's the sort of the mood or the emotion that you're trying to create? Like sometimes with COVID map it can be appropriate to have like really eye catching reds because you want somebody to think oh, it's bad in this area, like oh, this is not a good situation in the red zone. But other times, like if I'm making a map of vaccine data for example, I would probably use like cooler colors or something that evokes like the places that are more, are more vaccinated. Maybe that's a positive for those communities. And then also thinking about like clear titles, clear labels, clear annotations, making sure that your source is in there and is linked. If it's like an online visualization so that people can go look up the source data if they want to, making sure that it has a timestamp. If you're working with a data set that is updated frequently so readers know maybe what they're looking at might no longer be the most recent data, that kind of thing. That's all important.

Yeah, that's really great. And you mentioned here a lot of the times, like making sure that the sources, the methodology is always mentioned. Ensuring that readers have ability to go deep dive into that particular aspect of the data data story. What do you think is a great checklist, maybe for ensuring that the audience has all the necessary knowledge, but also a great checklist from a data quality perspective to ensure that the methodology is sound. As a data journalist.

Yeah. So I think you want to kind of answer the basic questions of like what is the source data? And then to any extent you can talk about it, where is the source data coming from? Is it from like a government agency? Is it from a scientific paper? Is it from like a survey? And then what gives the data authority? Like if it's coming from government or something, then that's a given. But if it's coming from a scientific paper then maybe you want to know like oh, these are researchers from XYZ institution and they have expertise in this topic, and then you want to talk about, like, what did you do with the data? Did you do an analysis, or are you just presenting what exists in the data set? Did you select a specific column or a specific field for some reason to present? And can you maybe give that reason as to why that field seems most important or why it might be most relevant to your story? When is the data from, what's the timestamp, who might be represented by the data, and is anything missing? Are there any kind of major caveats that you need to provide? Yeah, I feel it's really like, journalists also often talk about answering all those who, what, when, where, why questions. And I think you can think about a similar set of questions with your methodology. And if the reader wants to do their own analysis, what's the information that they would need to either replicate what you did or do something similar, maybe for their community or answering kind of a related question.

Okay, that's awesome. You really put a lot of emphasis on making sure that it connects a lot to journalistic best practices as well, and connecting it to traditional journalism, which I find great for audiences. Now, I think this a lot connects as well to your ability to shape the story of a particular data story. Right. So as a data journalist, what are the additional nuances that you need to take into considerations when shaping the story and the narrative that the data is telling you?

Yeah, one thing I find really important is to let the data shape the story, not the other way around. Like, I've run into this before where you come up with an idea, or maybe you come up with an argument, and then you say, well, let's go go out and see if we can find data that supports this argument, as opposed to finding the data and seeing what argument comes out of it, or like seeing where your evidence leads you. People can have the same problem in reporting, like talking to sources, where I might go into a story about COVID or something and say, well, I have this argument I want to make, and now I want to find experts who are going to give me evidence that will support my argument. It can be hard to not fall into that. So you always have to give yourself room for finding something you don't expect and adjusting your story accordingly. Another thing I find really important is explaining and leading into uncertainty. I think this is particularly true for Covid. But you find this in many other data sets as well, where, if you're looking at, say, results of an election poll, you don't want to just write the story as though these Data are really definitive and definitely reflect the entire country. Right. You're probably dealing with a sample, and the sample might not be as representative as you want it to be. So you have to explain that or you have to talk about what's not being included or what's not being represented in the data.

Okay, that's really great, especially on the last point. I think 2016 proved a lot that polling can be misleading as something to look at. Focusing on that particular aspect that you mentioned here of making sure the data tells the story or shapes the narrative and not vice versa. Can you walk us through maybe an example of a story that you were working on where looking at the data made you update the initial hypothesis that you've had? And what was that process like?

I think one example that comes to mind is just covering the pandemic right now in the United States. We've been in this moment for the last kind of two months or so, I would say, where everybody is anticipating a fall surge just because both in winter of 2020 and winter of 2021, we had a big surge in Covid cases. And experts have tied that to colder weather. Like, people are gathering indoors more. Soon we're going to have the holidays, which is going to mean travel and all of that stuff. But we're not really seeing huge spikes in numbers yet right now. As of the end of October, when we're recording this, and even in sources like wastewater data, which are a bit more reliable than cases right now, we're not seeing a massive jump yet nationwide. So for me, that kind of requires adjusting my assumptions to say, like, I think we're still probably going to expect outbreaks around the holidays. But I have to adjust how I write about this current moment in the pandemic and not just say, okay, there's going to be a surge. Like, we definitely know that because we never know that. We can make hypotheses and we can prepare for that to happen, but that doesn't mean it's definitive until we see the data in the next few weeks?

Okay, that's really awesome perspective, especially how it ties into, like, preemptively trying to making sure that the narrative has enough caveats to a certain extent, that you bake in in your reporting.

Yeah, I mean, Covid newsletter, one of the sections that I do every week is a national update, which is just like a short couple hundred words that's like, here are the COVID patterns right now. And literally anybody who reads my newsletter could probably tell you that for the last month and a Half, it's been like fall stirred. Maybe. We're not sure yet. Here are some signs why. And also, why not, you know, and that's just in the situation. And I'll continue to caveat it as best I can until we have a clearer pattern.

That's definitely great to hear, and I think this marks a great segue to discuss your overall work and experience covering complex topics such as COVID 19. In a lot of ways, there's a lot of weight on one's shoulder when discussing, visualizing and writing about, like, complex health stories such as Covid maybe. Walk us through first the challenges of covering COVID 19. What have you had? What would you consider are the main learnings from having covered it?

Probably the biggest challenge is just how many unknowns there are. Like, I think the COVID pandemic has been really interesting from a data perspective. Also from a science and health perspective, wherein we have more information about the coronavirus than we have had about probably any other disease. I'm not sure that I can say that really definitively, but I know that, for example, I just did a piece for my newsletter about comparing Covid tracking to the flu and rsv, which are both kind of having large outbreaks in the US Right now. And I was reflecting on how we have never tried to count every flu case. We have never tried to count every case of these other common viruses that we're used to dealing with. But Covid was such a huge crisis that there was an impetus to try and track it really precisely and track it through novel methods and try out all these new things for treatments leading to the development of MRNA vaccines and all of this stuff. You would think we would be able to answer any question, but that's actually not true. All case numbers are underestimates. All official sources have gaps. We don't have basic demographic data in the United States for a lot of things. I could go on about this all day, but the basic point is that we still have a lot of unknowns, and it can be hard to explain what those are when people think, oh, surely we've answered all the questions and we know exactly what's going to happen and Covid is over when we actually, we have no definitive information like that.

So let's start maybe talking about, I think how in a lot of ways, the inconsistencies and the gaps and the challenges that you've talked about here has trickled down into also inconsistent reporting and coverage when it comes to COVID 19 data. One thing that we've seen during the COVID 19 surge, especially in the first year and a half of the pandemic, is an explosion of data visualizations showcasing different angles and flavors of how COVID 19 is spreading. Right. Could have been on a local level, where a local municipality, municipalities or local government is like reporting on how COVID 19 is spreading in their local area. Or it could be national or international news outlets covering the spread of the virus globally. However, in a lot of ways, this has been hit and miss. Right. I think mainly due to the lax use of data visualization, lax use of narrative, or employment of narrative when it comes to shaping the story. How do you think we can avoid this in the future? And what are failsays that we can think of to ensure that this doesn't necessarily happen?

Yeah, this is such a good question, and this is definitely something I think about a lot, especially as I consider the fact that I work in two niches within journalism that I really wish weren't niches. I wish that every general assignment reporter at a local outlet was able to make charts and was able to read scientific papers and was able to closely follow every update from the CDC or from their local public health agency. And I think local journalism in the US from what I've seen and from talking to friends who are in those roles, is just at a huge capacity problem where there are not enough people to deliver the information that needs to be delivered. And so I'm really thinking about how can we improve education on data literacy, on science and health literacy, and kind of help your average reporter do the things that I do without it being a super specialized skill set. I would love for my role to not be as unusual or whatever as it is. And I think it would also be great to have more resources for those local outlets, whether that's like, oh, here's an organization that has made charts for every state and you can just use the one for your state if you want. There are some groups that start to do this. Climate Central is one example of a nonprofit that does this kind of work in the climate environmental space. Stacker, which is a company I used to work at, did some of this stuff, creating like a local news wire with data driven stories. And I think this goes to not just local news, but also local public health agencies and other kinds of local agencies that are tackling these crises. Like, they also need to have infrastructure to communicate to their audiences or communicate to their communities and also address misinformation, which we know has been such a huge problem during the pandemic. A friend of Mine gave a presentation at a conference recently talking about how misinformation has been so rampant. And she mentioned asking the US CDC at one point if they had a plan for Covid misinformation. And the CDC saying, not really. We're going to rely on journalists. It's like, well, you maybe shouldn't. This is a huge problem and you should have your own kind of infrastructure. So that's another thing to think about.

I think that's really great. I love the holistic answer, maybe focusing on the skills component of it. What do you think? If you were to design a basic data literacy or data skills upskilling program for the industry, what would the essential principles be that you would teach?

I think I would probably have to do more research myself to make sure that I'm creating something comprehensive. But going off of what I know now, I think that being able to critically interpret statistics, whether that's from a survey or a scientific paper or from a health agency, that is super critical. And then thinking about how to make charts, how to interpret charts, how to explain where data are coming from, like what is a methodology, what goes into a methodology. And then maybe after that one could get into the basics of doing your own analysis. But I think those sorts of just getting used to treating data as a source, that must be questioned rather than, oh, I see numbers. So I'm just going to assume the numbers are right. I think that's kind of a key mindset shift that might need to happen.

Yeah, it's kind of a data determinism that people fall into whenever they see a chart. They're like, okay, this is a higher level of truth just because it's like visualized on a chart on like some website.

Yeah, yeah. And that's actually not the case at all.

Yeah, exactly. So I'm sure another challenge of COVID 19, and I think, even though it shouldn't be necessarily is. Objection. Handling criticism right. This is a highly controversial topic. It's highly politicized. As a data journalist, how have you approached this? Especially when there's a lot of feedback that must be bad faith. Feed feedback and criticism.

When I get feedback like this, I definitely try to separate out or identify what is in good faith and what is in bad faith. For example, if I have a story that's getting popular on Twitter and I'm getting a lot of replies, I can usually tell pretty quickly by just checking somebody's profile whether they are a concerned reader who has a question, or even somebody with some data expertise or somebody with some science expertise, who has good faith feedback, or if they are just spreading misinformation. And if it's the latter, then I'm probably not going to engage with it because I have better things to do with my time. But I always try to answer questions when they are honest questions. And I try to explain complexity, especially if and this happens all the time in journalism. But it can be especially challenging when you have a data story or sort of a story in a complicated niche, when you have pieces of complexity that get cut out in the editing process. And then somebody asks a question and you're like, oh, I wrote that, but that paragraph was cut. So sometimes you can kind of use a bit of your reporting notes or your material that didn't make it into the story to answer a question. And this is why also, one thing I like to do with my newsletter is to share full interviews. Not like entirely full, but share sort of transcribed, edited versions of interviews that I do with sources and tell people like, here's the finished story. And here you can read this 20, 25 minute long conversation that I had with a scientist and you can see all of these complexities that didn't make it into the piece for kind of a more general audience. And I think that's like a nice thing to do to share a little bit of the reporting process for people.

That's really great. I love this insight and I love especially and it must be very frustrating to have a piece cut out. Indeed. That makes it back into the questions.

It happens. I am definitely one of those people, like any editor I've worked with can tell you I always write over my word count and I have to cut stuff back. Whether that's me doing it or an editor doing it. It's just part of the process.

Yeah. Indeed. As we wrap up this conversation, Betsy, which I really enjoyed. I'd love to end on a more future looking note, but I'd love if you can outline maybe in your own words what the future of data journalism and storytelling looks like.

Definitely. Making stuff more accessible, I think has been a big theme of our conversation and that's something I anticipate seeing more of going forward. I know right now we have tools for visualization like Flourish and Data wrapper are two I use pretty frequently that are so much easier to get into than if you were somebody starting out in data journalism like 10 or 20 years ago. I know like some of the older reporters in ire, they came from an era of what they call computer assisted reporting, which just feels so much more technical than what we're able to do now. So really, anybody who's interested in getting into data journalism can make an account and start making charts. And I think those platforms are going to get easier and there are going to be more platforms like that. I'm also kind of interested to see what happens with newer formats. Like, are we going to see 3D data visualizations that are incorporated into. I'm not going to say the Metaverse, because I don't know how much I'm excited about the Metaverse. But platforms like that, or even exploring other kinds of ways to engage with data. There's one visualization expert I follow who is big into data sonification, which I think is so cool, like making a visualization, but it's through sound, so you listen to it.

That's the first thing.

Yeah, I just think that I'd love.

To check that out.

Yeah, I'll send a link. Maybe you can put it in the show notes or something. But I find that stuff so cool. I'm also thinking about like, newer platforms. Like, I don't know. I'm on TikToks as of a few weeks ago because Twitter seems to be not in a good place. So I was like, I need to expand my social media footprint a little bit and I'm still getting used to it. But I like how TikTok allows you to do visual explainers. You can put a chart behind you and then point stuff out, explain the trends and you get a lot more space than you do in a tweet. Obviously people might not watch the whole video, so I think that comes with its own challenges. But I am interested to see how more journalists or more data viz, people getting onto those platforms changes how we think about data journalism.

Yeah, that's really great. I'm very excited to see what's in store for the field. One additional thing that I wanted to ask you about is with the rise of AI generating tools, right from Dall E to like GPT3, even like Codex and Coding Assistants, and these are going to be probably relatively mature to use within the next two years or so. How do you anticipate these technologies as well to impact data journalism?

Yeah, I don't have a ton of experience with them myself, but I know Muckrock, where I work, has done some work with AI for analyzing documents. If investigative reporters know, sometimes you get a trove of documents back from a public information request and it can be like a thousand pages that you have to sort through. And so Muckrock has been working on an AI tool that can help journalists do that more quickly and more efficiently. So I think there's obviously folks also work on like machine learning for data analysis. And yeah, this is not something that I have a ton of experience with myself, but. But definitely I think that will similarly help, if not improving access, then improving the efficiency of analysis. Like how much are you able to do in one workday or in one work week? I think is probably going to change a lot. Although of course AI kind of analysis can come with its own caveats and stuff too.

Definitely that's something to cover for a future episode. Now Betsy, as we wrap up, is there any final call to action you have before we end today's episode?

I think just educate other people about data or about your sort of chosen niche. People can handle more complexity than you think they can. You just have to trust your readers and be reliable and answer questions and that can go a long way.

Okay, that is awesome. Thank you so much, Betsy, for coming on Data Framed.

Yeah, thanks for having me. You've been listening to Data Framed, a podcast by datacamp. Keep connected with us by subscribing to the show in your favorite podcast player. Please give us a rating, leave a comment, and share episodes you love. That helps us keep delivering insights into all things data. Thanks for listening. Until next time.