Talking data journalism and upskilling with Gurman Bhatia
"People don't remember numbers, people remember people," Gurman tells us.
Gurman Bhatia is an independent data journalist and trainer. Before going independent, she worked as a Data Visualization Developer at Reuters out of Singapore. She trains professional journalists and students in data journalism.
Gurman has been a friend since we all worked together at Hindustan Times, and we’re so thankful she agreed to sit down with us to tell us about her journey.
Earlier this year, she held a series of trainings for the Meta (formerly Facebook) Journalism Project India, which you can watch here. Her list of accomplishments is long, and we can’t fit it all in, so you can check out some of her work on her website. If you’re interested in connecting with Gurman, she hosts weekly office hours that you can sign up for here.
This interview has been edited for brevity, as we spoke to her for over 40 minutes.
In Old News: Do you remember any moment or any memory that you have where you're like: yeah, data journalism feels like something that I would want to do?
Gurman: I 100% have a moment and a memory. So, let’s go back to 2015, January 2015. I was a student at Columbia Journalism School. And I was taking this class, which I was 100% not convinced about. The class was called Interactive Design and Storytelling. And that class was being taught by two people, Michael Keller and Clarissa Dias. Michael Keller was at that time an interactive data journalist at Al Jazeera America. And the thing about why I was not convinced about taking that class was that these two were fairly young. And tiny, quite literally. So when they came to convince people that this is the course that you should do, I was like, not really convinced. But because the course description sounded like something that would be fun, it was about how we can tell good stories on the web. And using design. I was like, this sounds interesting. So after shuffling back and forth, I ended up in that class. The first class that I took, they took us through a lot of data visualisations, data stories, gave examples of stuff that people were doing around the world, and also gave us a roadmap: this is what we'll cover in the class. And up until that point, I hated being in New York. I was like, the city is so expensive, this course is so expensive, I spent so much money to come here. What is this? I will not get a job, this, that... And I was miserable. But when that class happened, I was like, this is what I want to do! I called my parents and like, you know, today, I felt like this is why I came to New York. And that class was quite the changing point for me, honestly. They gave me the building blocks to everything that I do today. They would stay up after class and just like teach me things that were above and beyond what they were supposed to teach.
In Old News: Has your vision of what data journalism is changed since then?
Gurman: 100%, I was also like, barely 21 at that time, so now as I've experienced more things, I've worked in newsrooms, I'm no longer a naive 21-year-old. I'm a naive 28-year-old. So there's a difference. So more towards the reporting side of things, when you start out you feel enamoured by everything that's glossy and beautiful and attractive. But gradually you learn that what matters really is what comes before the ‘makeup.’ The actual content, what the story is, how do you find that story? And that's majorly the challenging part. So I think not falling in just for the coolness of it all, is something that I learned with time.
In Old News: Were you always inclined towards certain types of stories when you were thinking that, okay journalism is maybe something that I want to check out?
Gurman: I think I have always been more inclined towards stories about India. So when I was in the US, there were a lot of problems that I was reporting on that seemed relevant, but a lot of it also seemed like first-world problems to me. So when I learned the skills and came back here to work at HT [Hindustan Times], it was a more fulfilling experience, I would say, because these were stories I cared about more.
Data is something that can come across as ‘“Oh, this is about numbers. And this is about increases and decreases in trends, patterns and statistics,” and all of that. I think finding ways of how I can make this about people is something that I constantly find myself thinking about as well. Something that somebody told me very early when I was an intern at the Palm Beach post, was that the best data stories have the least amount of numbers, because people don't remember numbers, people remember people.
In Old News: Humanising a story or an angle is already hard. So does it feel a little harder when the story is coming from data instead of an interview?
Gurman: Data journalism at the end of the day is just like, more empirical approach, right? Like, I'm not just talking to five people, 10 people and trusting what they're saying, I'm seeing actual statistics, caveat supplied, that the statistics might be right or wrong. But then going on and seeing “Oh, this is what the statistic experience is, what is the lived experience?,” and then merging the two in combining the two. It's not either, or, it's just like, how we can combine those.
In Old News: Sometimes you’ll go through data and not find a story. Do you have a rough success rate?
Gurman: I’ve spent months working on stories that never, ever publish. And that's okay. It's part of the learning experience. I think everybody learns that way. Having a good editor helps.
One approach that I think works better than most is that if you start with a hypothesis, a very specific hypothesis, and then you go on to use data as a way to prove or disprove that hypothesis. I think that approach has a much higher success rate, then going into a data set with a very open-ended approach where you might spend a lot of time looking at it and not finding anything. So I think context that is outside the data set is extremely helpful. To start approaching a data set.
There's a story, for example, that I did on Bollywood songs, we did a video on that one, of how the female solo was declining. And how that story idea came about was that I was watching this interview where Anupama Chopra was sitting with four or five female singers, contemporary singers, and they were just sitting and chatting. And there is this anecdote, where they say that, you know, we lived in an age where every album was filled with Lata and Asha, to the point where every other song is sung by Arijit Singh. Like, it's just an anecdote in an interview. And I was like, I wonder if this can be proved? The point being that that hypothesis can come from anywhere. And that hypothesis, plain and simple that the female singers are singing fewer songs than they used to in the 40s and 50s. And then you just find the data to prove it. And then as you do more reporting, that hypothesis becomes more nuanced. And you're like, female singers are singing fewer songs than in the 1950s because... the storyline is changing, because there is no lip sync. So like, you know, your hypothesis starts becoming more nuanced. And the more nuanced it is, the more specific your story and the better your story is.
In Old News: When you’re writing up the findings of your reporting, do you incorporate how you found the information? Do you show your proofs for every story or do you find it’s only necessary in some cases?
Gurman: So in some cases, you will make a lot of decisions in the process. Like say, for example, I was talking about this Bollywood story. Within the story, there were several things that I did. So for example, I counted songs only from the top 50 (and some of this might be wrong, and I'm just telling from my memory), but like top 50 movies, by box office grossings for every year, I picked those. And then I found the songs in those movies. And then I found the gender of the singers in that movie. And then that's the data set that I built. That's my subset that I analysed. So it's my job to explain that to the reader. Because what if somebody wants to check my work?
I think doing that also increases readers' trust in you. Your credibility improves, if you're entirely transparent about what you did.
In Old News: As you're going into independent journalism, do you have tips for other people who are considering going independent?
Gurman: I was fortunate in the sense that I didn't have student debt. I also took this decision at a very different point in my life, I wouldn't have done it three years ago. So I was at a point where I was debt free. I'm also at the point where my parents are not financially dependent on me, nobody is financially dependent on me. I am privileged enough that my parents would support me if I ask them. I don't want to if I can. I had some savings. So before you take the plunge, see where you're placed financially, if you can make it work for yourself. With a little bit of buffer, make sure you have like a year's worth of savings in the bank so that if nothing works out, you're not stressed about survival. That's first, see where you are financially and personally.
Having a strong emotional network to support you through freelancing, and like figuring it out is important. So having supportive friends and family was helpful too.
I have been in the industry for a while. So I had a network that somehow helped me through it all. And oftentimes, there were people who I reached out to for advice. And then when they were too busy, they just passed on their work to me. So that was also good in terms of tapping the network that it helped you.
I also like shamelessly applied to anything and everything. Now, I think I've sent out my CV more as a freelancer than I ever did before. I am part of Google groups and Slack channels where freelance gigs might pop up. And when they do, I just like dropping in a line, “Here's my portfolio.” I've done that a lot. Teaching also helps. Teaching is interesting because you think more critically about things than you would if you were professionally just doing it. So for example, if I am talking about the accessibility in databases, and I am thinking of it practically I will think of it very differently than when I'm teaching to an audience. Because I have to be better prepared now. So because they're gonna ask hard questions. So if they're gonna ask hard questions, I have to ask those hard questions before they do.
In Old News: Is it gratifying helping people learn data journalism?
Gurman: It's amazing to see people do things. It actually happens all the time that you give an assignment and some people do it, some people don’t. But of the few people that do do it, some will do such an amazing job. I experienced this for the first time when I was teaching at IIMC four and a half years ago. I used to teach a semester in data journalism at IIMC in Delhi. And when I was doing that, there was this one student who was super smart, super interested. And as a part of the course, the idea was that in the end they would have a full-fledged well-reported data story. And the student’s story was great. And I think at that point in time, I said that you should pitch it. It was about elections, because we use that data set in our exercises. And he did pitch it to a news room. And he did get it published. And seeing that published story that came out of a class was so gratifying. And it was the first time I experienced that feeling. And that student is perhaps the only student from the entire class that went on to pursue data journalism, like full time, he did jobs here [in India]. He recently graduated from the masters in data journalism at Columbia, on a full-ride scholarship, and is now doing a fellowship. But to see that entire journey was so gratifying. Something that started in the classroom is his career now, and I'm not saying it's me. But even if I had like 1% contribution to it, I feel immensely happy by that 1%. One of the reasons I quit my job was also that teaching with a full time job is hard. I was also an immigrant in Singapore, figuring out the logistics of it all was hard. And I wanted teaching to be a part of my life. I don't want it to only be what I do, but I want it to be a good 40-50% of what I do. So regulating my own time helps me do that too — figuring out how I share information. And I've been thinking about this — what are the different ways you can mentor people? What are the different ways you can share the knowledge that you have? Is it that you do YouTube videos? Is it that you have a newsletter with a set of tutorials? Is it that you have a mentorship program? That's maybe for a very select community. Maybe it's only for women in non-English speaking news rooms? I don't know. I don't know what the answers are. But even thinking about these things makes me super excited and [is] gratifying for sure. I do these office hours every Wednesday where anybody can sign up and show up. And I will chat with you one to one. Answer whatever questions you have about anything. And the conversations that I come out of is an incredible learning experience for me too. I'm oftentimes repeating the same things, because people have similar questions. But at the same time, sometimes I'm not. And most of the time, I end up learning something new from the conversation itself. So, I look forward to Wednesday evenings because of that. I'm going to talk to strangers and maybe make someone's day a little nice and learn from them in the process, too.
TL;DR it is gratifying.
*Correction: A previous version of this newsletter mistakenly stated that Gurman’s student received a Fulbright scholarship. It was a full-ride scholarship.
If you'd like to share your journalism experience, or you'd like to suggest who we should interview next, reply to the newsletter or reach out at firstname.lastname@example.org