Lifecycle Open Science in Action: Researcher Q&A with Rodrigo Alonso Reyes Cordova

At the Center for Open Science (COS), we use the term lifecycle open science (LOS) to describe research with publicly accessible plans, outputs—such as data, materials, and code—and outcomes that are linked and findable. When those elements are connected across the research lifecycle, the work becomes easier for others to evaluate, reuse, and build on—not just the final paper.

Rodrigo Alonso Reyes Cordova is a PhD candidate in sociology at the médialab at Sciences Po in Paris, France, where his research focuses on the intersection between politics and the public's understanding and perceptions of science and expertise. His study, Political Identification-Driven Trust in Research Fields and Scientific Claims, examines whether political identity shapes how people perceive scientific claims depending on the field the expert comes from—specifically, whether liberals and conservatives respond differently to the same information when it comes from a sociologist versus an economist, or an environmental scientist versus an agricultural scientist. 

Reyes Cordova's project is a strong example of lifecycle open science in practice: he used OSF to preregister the study and afterwards linked his research record with data shared on Zenodo and the resulting published article in the journal Public Understanding of Science.

COS spoke with Reyes Cordova about the benefits and importance of preregistration, how keeping plans, outputs, and outcomes linked and discoverable strengthens the research record, and what he sees as the value of making research materials publicly accessible.

Q: Can you share a brief overview of this study—what question(s) were you trying to answer, and what motivated it?

A: My main motivation was a feeling of discomfort with the usual narrative—at least, the one coming from the United States—that liberals are pro-science and conservatives are anti-science. I don't think that things are so simple—or at least, I didn't think that things were so simple. I felt that it rather depends on how knowledge is used, and whether it aligns with people's political leanings.

I would say that at least since the 1970s, science has become more entangled with regulation, and this is why there is the perception that science is more liberal, as you would call it in the United States—more pro-regulating the economy and regulating industry, particularly because of the climate crisis that we’re in right now.

So, I wanted to ask: if I present people with the same piece of information, will they react differently to it when it comes from an economist versus a sociologist, or from an agricultural scientist versus an environmental scientist? I thought that if political identification matters—and if, for example, a sociologist is seen more as a leftist—the same piece of information might be better received by someone who's left-wing if it comes from a sociologist than if it comes from an economist. This was the main question.

What I found is that it didn't matter. What mattered the most was the claim itself. So, I’m afraid that the simple narrative that I described earlier is very much true. [laughs] Liberals trust every expert that I examined more than conservatives, and they also trust every claim that I examined more than conservatives.

Q: What led you to make your preregistration and study materials publicly available on OSF?

A: I did my undergraduate and my graduate studies at Tilburg University. In the Netherlands, there was a big scandal involving a professor who was basically inventing data and publishing it. This professor came from Tilburg, and I think this is why, in my program, we were very hammered on the fact that you should be very open with everything that you do as a researcher. We had several courses on questionable research practices, and this is where I learned that you should definitely preregister everything—from the design to the analysis that you will do, and most importantly, your hypotheses.

I'm also very familiar with the replication crisis in social science, and with p-hacking in general. So I think this was my main motivation. My partner told me: if you want to be seen as a credible researcher, you need to pregister everything. I didn't want to seem like I was hiding anything. This is why I preregistered the hypotheses mainly, and my analysis and scripts are also there.

Q: It sounds like you came from a research culture where open science practices like preregistration were very much encouraged. But along the way, did you encounter any challenges in implementing practices like preregistration?

A: Not so much. I must say that I was, at least in my research environment, perhaps the person who pushed the most for preregistering. I strongly believe that even if you have results that don't line up with what you hypothesized, they should still be publishable. And this was a fear of some of the people that I have met.

But at the end of the day, I think it's very clear that my results did not align with my hypotheses. And luckily, the journal where my article was published, Public Understanding of Science, didn't really care about this. They cared more about the integrity of the research.

Q: You preregistered this study before collecting data. For you, what's the broader value of preregistration in terms of how you plan, document, and carry out research?

A: If you preregister, you have to think about every single detail before launching the experiment, and then again before analyzing. Of course, there is some leeway in that. I have seen cases where a preregistration was written with a certain analysis in mind, but then the data came and the analysis didn't fit, so the researchers decided to go with another approach, which is okay to do so long as that’s disclosed. But I think it's very nice that these things are made visible. Preregistering makes you put everything on paper. If anybody wants to see the process, it's there on the internet, and it will always be there.

When you publish the study, you need to be very transparent about the decisions that do not match the preregistration. It can be very time-consuming, yes. But at the end of the day, now things are out there. People can contest my analysis and the way that I designed things, but they cannot say, "You saw the data before and you came up with the hypotheses and then decided to publish." At least things are visible. And I think that lends credibility to research.

It can definitely be overwhelming at first—but it's like with anything. If I start reading a book and in the first chapter I'm completely lost, eventually I start getting the rhythm. I think it's the same with these types of forms. The preregistration template on OSF was very, very helpful. It's so structured that, at the end of the day, it not only gives you credibility later on, but it also gives you a template to follow when you do the analysis. You can go back to it and say, "Okay, I said I would do this, so I will just do it." The analysis decisions, the design decisions—you're forced to make them beforehand, but then you just have to follow a template, which is very nice.

Q: Your project brings together a preregistered study plan, research outputs, and a published paper. What are the benefits of keeping your research outputs connected and publicly accessible throughout the research lifecycle?

A: Apart from what I said about credibility and the fact that it makes your workflow easier in the later stages of the project, I think it's also very good to have examples—not textbook examples, but real-life examples—of how to conduct this type of research for other people. Not only for other researchers who might want to do something similar, but also for students.

When I was a student, I would have been very grateful to see something like what I have done out there: real-life examples of how you can do research, and maybe even more importantly, the data and the scripts.

Anybody can go on Zenodo, download my dataset, run my script, and they will get the same results. If a student wants to go through the script and see the analysis themselves, and then compare it to my interpretation in the article, they can do that. And perhaps this is also a good exercise for someone who wants to learn how to interpret statistical data.

Q: What advice would you give to researchers who are just beginning their open science journey—particularly around preregistration and making their work findable and connected?

A: Preregister everything. It might be time-consuming, but it makes life easier later on. And even if you're overwhelmed by the templates available on OSF, you can just write a document of what you plan to do—your hypotheses, the type of analysis you plan to make. The more detailed, the better. But it can also be something rougher. My advice would be to just do it.

If you don't want to do it all the way, just do something. It's always better to have something on paper and be able to say, "I thought about this before. It didn't come out of looking at the data."