In Summer 2022, Mitchell Ransden ('23) and Leyla Jacoby ('25) participated in collaborative research on the Democracy Erosion Project. Check out their interview with PEPLC members Tea Wallmark ('25) and Quinn Allred ('26) below!
The transcript of this interview, which occurred on Jan. 10, 2023, has been edited for clarity.
Leyla Jacoby Bio: Leyla Jacoby ‘25 is a Quantitative Social Science major focusing her studies on Government, Linguistics, and Computer Science. She has previously worked on political campaigns and to advocate for political accountability, digital privacy rights, and freedom of information through technology. She intends to pursue the quantitative study of political discourse.
Mitchell Ransden Bio: Mitchell Ransden '23 is a Government major with minors in Public Policy and Psychology. In addition to his time as a Research Fellow at the Democratic Erosion Consortium, he has worked as a Research Assistant with James O. Freedman
Presidential Professor Brendan Nyhan, and is currently writing his undergraduate honors thesis in the Government Department. He intends to pursue policy analysis in the nonprofit sphere after graduation.
Political Economy Project Leadership Council Interviewers: Tea Wallmark ’25 and Quinn Allred ‘26
Tea: What led you to apply and how did you find out about the Democracy Erosion internship?
Mitchell: I found out about this internship because I had previously done some undergraduate research with Professor Nyhan in the Government Department here at Dartmouth. He forwarded the Political Economy Project email to a list of students who had done research with him before. At the time, I was starting to apply for internships (late winter/early spring). So I applied because it sounded interesting, aligned with my interests, and paid better than the other opportunities that I was looking at. And then, a couple of months later I got another email back in my inbox. I’d almost forgotten about it, but I was delighted.
Quinn: That’s awesome, that’s really cool! And Leyla, what about you?
Leyla: I had a similar process. I’m sure that you all have or will experience this but it’s very hard to get a job after freshman year; nobody really wants to hire you. So I also heard about the opportunity through Professor Nyhan. I was in his class last spring and asked him for help with finding opportunities for freshmen. He gave me a couple of things and I applied to all of them very much as long shots. I was going to be doing off campus research through Dartmouth over the summer but when I got the email I was very excited! It seemed absolutely fascinating and I was very glad I got the internship. We got to do really interesting data science work with real world applications and work on an innovative project. Very enjoyable experience!
Quinn: That’s really cool! How would you describe your experience as an intern?
Leyla: To start off, the project draws students from a consortium of schools, I think Dartmouth, Brown, Texas A.M., and University of Chicago. All the interns were generally interested in government and/or data science and there were a bunch of different projects that the interns had to get through. The intern manager, Hannah Baron, was a wonderful Phd. student from Brown who coordinated discussions about what specific projects we wanted to do and if we wanted more leadership positions, what skills we could bring to the table. I’m a QSS major and am really interested in data science, data wrangling, and data visualization. Mitchell also has a lot of these skills while other interns didn’t have as much experience in data science. So we did more of the data specific work which was really really fun. It was fun to bring in what I’d learned in the classroom, especially from one class last Spring, and apply it all summer which was really enjoyable. QSS 17 was super helpful for this internship, for me at least. But then there were other students who were more on the Government side of the spectrum from other schools who developed different analysis frameworks. Overall, there was a wide range of experiences and we both seemed to approach it from a more data perspective in addition to the government perspective because we both had those skills.
Mitchell: Yes, Leyla and I worked closely for the whole summer because we both had more data-heavy backgrounds. The main project that we were part of consisted of building out this dataset of instances of democratic erosion around the world. For example, tracking government repression and the collapse of civil liberties. We were helping to build out that dataset and then analyze trends. We were also pretty heavily involved with the project that the consortium also works with, Bright Line Watch, which is run by Professor Nyhan and Professor Carey. In the project, we were tracking claims of voter fraud and instances of candidates refusing to concede races in and around the 2020 election. The internship was fully remote but we spent a lot of time on Zoom meetings together with our larger cohort of eight people. It was great.
Leyla: So everyone in our cohort of eight helped with the coder work of turning papers into data, and then there were smaller roles in addition to that. So Mitchell and I managed the dataset and did a lot of the back end work of turning everyone’s entries into a usable dataset. We helped make visualizations out of that, making sure it looked good. And like Mitchell mentioned, we also took the lead on a project on concession statements for congressional races in the US from 2014 to 2022. We’ll see how far back we go. So we got to do a lot of the preliminary stuff for that big project and actually just presented it at a conference over winterim, which was great. So this was all part of the Bright Line Watch project and now I’m continuing with the work this term!
Mitchell: While we weren’t presenting at the conference in the fall—the largest gathering of North American political scientists—we did help, which was cool. I drove up to Montreal and saw the paper and the tables that we made. There was this kind of constant sense that we were doing work that was sometimes very difficult and somewhat tedious, but it did have these real world applications. Our main dataset was being used by academics. The U.S. State Department takes the data into account when allocating foreign aid to USAID because the government has interest in not sending money to repressive regimes. So there was definitely both the sense of real world applications and also making sure that we were double-checking all of our calculations.
Quinn: That’s so cool that the project had so many long-term applications. And for you Leyla, you’re still working on something that’s very much related. And Mitchell, you were talking about your thesis and it’s all gotta be interconnected.
Mitchell: Oh, absolutely!
Tea: What was it like working with these longer-term projects? I get the sense here at Dartmouth that the quarters go by so fast so it's hard to really dive into research for a one-term class. What was it like starting an internship and realizing you have all these months ahead of you? What was it like planning it all out? Did it go as you expected? What was it like working in a team on this longer project?
Leyla: That’s a really good question. It’s a long term project and when we came in they were already in version four of the dataset and our job was to make version five. I think the intern manager, Hannah, did a really good job of breaking the project down into more bite-sized pieces. She started off by saying that our job was to come up with version five this year and our goal for version five was to expand it to these countries, go through all of these case studies, and develop a U.S. version of this. She gave us very tangible goals and we got to pick which ones we wanted to work on. Because of this structure we were able to come up with some really good results. The project would have been more overwhelming and less continuous if we didn’t have that structure and Hannah who had been working on the project this whole time and would be working on the project in the future. It was a lot easier for us to jump in and determine our role in the project.
Mitchell: Yeah, it was nice not having to build the project from the ground up. But we definitely did face some challenges of having people work across multiple time zones because, again, it was a fully remote internship. In our smaller group, Leyla and I worked with one guy from Egypt so we had some interesting meeting times.
Leyla: You would hand over things and then they would be done by the time you work up the next morning. It was great. It was 24hr work.
Mitchell: It was awesome. There was this overarching structure of laying out the longer-term goals, but at the same time being able to pursue research questions of our own interest and also take some leadership among this small cohort. Whether that was deciding who’s turn it was to run the weekly meetings and deciding how we stay on track and what needs to get done by next week, or the larger leadership of what kind of analysis we are going to do with these data. Multiple times, I remember having to set different deadlines for different parts of this project, because it was very large and there were a lot of moving pieces. But for the most part, it worked out alright. People were able to do what we needed them to do, and we were able to adjust as we went along. There was a little bit of rush near the end as we prepared for the hard deadline at the conference in Montreal, but otherwise I feel like the self-direction really allowed us to pace things out. We knew where we were headed from the beginning, so we were able to chart things out ahead of time.
Leyla: Yes, and we also all didn’t have to chart out all of it. Mitchell and I were working on the analysis portion so we got to make the plan for that portion. We made the plan and we changed the plan as needed on that portion and then other interns would be in charge of the timing for their portion of the project. So we all got to work on the larger project, but it was also really helpful to split up the timing and delegation for each project.
Quinn: That’s really cool. I’m really glad you both brought up the structure of it since it’s hard to get a sense of the day to day of the internship. Could go into more detail about either specific data that you collected, or just anything about the very tangible part of the project?
Mitchell: So the data we were collecting is part of this very cool and unique project. Before there was this dataset, some faculty members got together and they decided they would teach a shared curriculum about democratic erosion and government repression. I don’t know exactly how many universities.
Leyla: Over 60, I think?
Mitchell: Yeah, it started out small, but it’s grown to a lot of professors who are teaching the same class and have the same final assignment. The assignment is a term paper where every student picks a case study and discusses trends and individual events with regards to democratic erosion in that country. So our job was to take all of these papers—and there were a lot of them of varying quality—and use a fairly comprehensive coding strategy that had been developed by previous interns to pull out individual events from these papers and coded them by type, by impact, by dates. All of the data points underwent a very specific coding scheme that turned into the fifth version of the dataset.
Leyla: This class is offered at Dartmouth, it’s called Democratic Erosion, and it seems to be very good. We read all of their term papers and they were very good. Once we found all of these events we kept track of them and put them into a qualtrics form. The qualtrics form spit it out as a spreadsheet, as a CSV, and then we downloaded that CSV and turned it into a comprehensive dataset. We then merged it with a previous version and corrected some smaller issues. We also had this big step part of the way through where we had to go through and get rid of all the duplicate events because we didn’t want them showing up twice in the dataset. So after we were done with the coding, there was the big step of making sure that the dataset made sense. After that, we could start working on the paper and the analysis for the memo. We used linear models to present specific types of events that led to others and didn’t lead to others. And we did time series analysis of which ones occurred first and which occurred later. That kind of thing.
Mitchell: It was definitely a great experience to manage the entire data pipeline. We had both had experience in the limited context of knowing what to do with data but now we had to actually pull it all together and turn that into quality observations, which has definitely been helpful for what I’m working on now.
Tea: Have you used any of the skills you learned this summer in your classes or in your research?
Mitchell: Sure, it was a great refresher over the summer to dust off my old GOV 10 notes and as I move into text analysis in my research. Fortunately, that is not a manual process for the thesis that I’m writing. The organizational component was also very helpful. The internship was an opportunity to be a part of a longer-term research project that was still very self-directed. Charting out this project, starting in the fall and up through the deadline in April, it was great to develop those organizational skills to reach the deliverables of the project.
Leyla: That makes sense. I haven’t written my thesis yet, but I’m sure that these skills will come up when I do. Like I said before, I was really lucky that I did this internship right after the class that taught me so much of what I used during the internship. The internship certainly solidified those skills and I’m sure it will make it easier for me to apply those skills in the future. I know I will use the skills as I watch Mitchell write his thesis and hear about work that other upperclassmen are doing. I’m already starting to use them in my QSS class that I’m taking now, which is nice. I also agree about the remote work aspect; the internship taught me a lot about working remotely with a team and how to coordinate amongst yourselves using slack and Zoom to do our work. Mitchell and I worked synchronously on Zoom, but it sounds like other intern teams worked more asynchronously. Learning how to manage deadlines, give deliverables, keep each other updated through those asynchronous tools was definitely a good skill.
Mitchell: Near the end of the internship we did get the opportunity to go out to Providence, Rhode island and meet up with the other interns and that was a great time. The Democratic Erosion consortium and the PEP helped pay for that. I think it was very productive to meet in person over the course of three days at a key point in the project where we were aligning everything for the next big step. That was a great experience.
Quinn: Our next question is more logistical in nature. You’ve both mentioned the asynchronous work aspect. Where were you during the internship? What was the funding used for? Was the funding sufficient? Do you have anything to say about the setup, the setting, and how much it costs?
Leyla: I lived at home for most of the internship because it was an asynchronous job. I visited some family a bit because we got to work remotely. I think the funding mostly went towards keeping ourselves able to do the work instead of having to do something else. I worked on my computer so there weren’t really any resources I needed. But, like Mitchell mentioned, the Democratic Erosion consortium did pay for us to take that trip to Providence, which was very nice of them. It was sufficient for me and it was very generous of them to let us focus on the work and not have to worry about other factors.
Mitchell: Leyla, I echo your sentiments all the way down the board. I was also living at home and were it not for this opportunity, I probably would have been living in a tight apartment in Hartford, so I was very fortunate to work at home and be compensated for the work because it was a full time job.
Leyla: Yeah, definitely.
Mitchell: We had weekly check-ins where we posted the hours that we worked and between the two of us we were racking up some serious time. It’s always nice to do research that you know is going to be important, but I also really appreciated being able to take that trip without worrying about the cost and spending several months doing the work that we did.
Leyla: As a full-time position, too, which would not have been possible without the funding.
Tea: That’s awesome. Do either of you have any advice or tips for an incoming democracy erosion intern?
Leyla: Figure out what interests you about the project and try to do as much of that as you can because there was so much different work to go around. We were all able to find work that we were really really interested in. Think about what you want to do for the whole summer and advocate for yourself so that you find your work interesting and can really make a difference, learn more, and have fun doing it.
Mitchell: Similarly, it is a project that has a specific focus and I, personally, had no experience with comparative politics, which it was it falls under. But don’t let that dissuade you, and I know we’re both data people, but if you’re not a data person don’t let that dissuade you either because it is a big project. Working with people who weren’t as data savvy illustrated the importance of thinking hard about how to interpret the data and how to construct that initial pipeline. A lot of different skills can be used—it’s a great opportunity to build new ones and build on the skills that you already have—all in an environment that has real world importance. I was never panicking about deadlines, and it was very doable and spaced out. I would encourage people to apply for it if they’re at all interested in the subject or just doing research with a group of other undergrads. Great way to meet new people.
Leyla: I didn’t have much of a comparative politics background, either, and we learned so much about comparative politics through the internship, which was great.
Quinn: That sounds really really cool. And then just with backgrounds and interests, we have one of our final questions, just for the media you consume. Is there one thing you can name whether it be a podcast, an ebook, is it a social media account? Anything that you would want or you think other people should listen, read, interact with at all?
Leyla: That is so interesting. Let me think.
Mitchell: Is it personal or related to the subject?
Quinn: Anything goes, whatever media you think is worth plugging.
Leyla: Yeah, I was gonna say I don't know if a lot of television shows focus on the intricacies of government repression.
Mitchell: Oh, yeah, I will say, I have done some work. I have done some work previously, and how I got into all of this, I was doing research about misinformation with Professor Nyhan in a paper that is hopefully going to be accepted for publication. And so I have recently been getting into a podcast called “Oh no, Ross and Carrie!” who investigate things that people tend to be misinformed about. So things like fringe science or paranormal type stuff and really all sorts of conspiracy theories. It’s really fun and they take it apart from beginning to end themselves, and it’s so entertaining because they are both really fun people. But it's also illuminating as to the sorts of things that some people can believe. So I would definitely recommend that to anyone. Friends told me about a couple months ago, and I'm just, I'm all in.
Quinn: I think I saw everyone go to their phone and write that down, too.
Leyla: Yeah, that's so cool. I am also very interested in misinformation slash research with Professor Nyhan. Same with (Mitchell). If you got this from that, I'm sure I will like it. I will give my two personal favorite podcasts a plug which are not related to this at all, but they can be, I guess, if you are interested in these things you'll probably be interested in the material. John Green’s podcast “The Anthropocene Reviewed” is wonderful. The book is also really good. It's just about what it means to be a person. There are 5 star reviews of different facets of the human-centered planet. You know what, that does relate to this because it's about turning real world experiences into data. And he's kind of making fun of the whole “imposing reviews on an experience” situation. But he does it very well. So that's great. And then one of my favorite tech podcasts, shrouded in a bit of drama at the end, is “Reply All.” They're very good journalists who do really interesting stories about things happening in the tech world. My favorite one is them trying to track down a song that somebody remembers from childhood, but nobody else seems to have ever heard of except for this one guy on Facebook in Africa, so it obviously exists. But they try to find out what happened to the song and why you can't find it anywhere online. I recommend that podcast, too.
Quinn: Those are two very different, but very fascinating answers.
Tea: Yeah, yeah, I love that. Is there anything else either of you would like to share before we wrap this up?
Leyla: I’d like to say thank you for the experience that we both had this past summer and for doing this interview today! I remember before our internship started, I really did not know what to expect. I had lunch with someone who did this internship three years ago but his experience had been pretty different. So I’m glad that you’re doing this for future students doing this internship and grateful for the PEP for the experience this past summer.
Mitchell: I echo that wholeheartedly. The PEP is an example of one of many great organizations on campus. There’s a lot of great people here doing awesome things.
Quinn: Thank you both for your time and your contributions to future interns!