Data Sharing Is Growing but Looks Different for Qualitative and Quantitative Methods

Different expectations • Ethical motivations and challenges • How to share

As a qualitative researcher, Ruth Abrams, a senior lecturer at the University of Surrey, doesn’t usually share her data.  

“It feels like a private moment between the researcher and the participant,” she said in an interview with the Observer. 

But psychological scientists, particularly quantitative researchers, are increasingly recognizing the value of data sharing, one tenet of open science. Data sharing leads to less redundancy, more efficiency, and more transparency, but must be done carefully to maintain participant confidentiality. Researchers working with qualitative data are not as well versed in how to effectively, easily, and safely share data and have additional concerns over protecting privacy, revealing different challenges for different types of researchers and highlighting long-standing differences between quantitative and qualitative methods.

A 2025 Advances in Methods and Practices in Psychological Science paper by Moin Syed and Dulce Wilkinson Westberg discussed how, through both explicit and implicit means, psychological science has preferred quantitative methods over qualitative ones.  

“The allegiance to one methodological family and the rejection of another means that, at least in part, our methods are constraining the universe of research questions we are willing to ask,” Syed and Westberg wrote. “Moreover, questions that focus on the what, how, and why of psychological phenomena—questions that are of core interest to many psychologists—are particularly well-suited to qualitative methods.” 

As a solution, the authors urged psychological scientists to do more mixed-methods research—a combination of both qualitative and quantitative practices—and outlined how to go about incorporating both types.  

Abrams said there are more similarities than differences between the approaches, and the field has moved beyond the debate of which is best. “It ultimately just is about your research question and what methods and approaches are going to answer that best.” 

But, as she discovered, there are differences in emerging challenges and expectations for each approach, particularly when it comes to data sharing.  

Different expectations  

In a 2025 Advances in Methods and Practices in Psychological Science article, Abrams surveyed 14 researchers at her university and found that qualitative researchers were less likely to share their data, had less knowledge about how to do so, and didn’t have as established expectations around data sharing as quantitative researchers.  

“There’s quite a difference across research communities in terms of expectations about data sharing, ease, access, knowledge, ability, all of those sorts of things,” said Abrams. “Quantitative disciplines, or the more hard sciences, generally have a much more embedded approach to sharing data in that it’s often set up right from the start of a project.” 

“For qualitative research, it’s quite a different premise,” she said. Researchers try to connect with their participants and develop trust that conversations and answers are private. “The concept of then sharing that data requires quite a mindset shift.” 

For quantitative researchers, it’s often a given they will share data, and it may be required by journals and funders. “To not do that signals something about you as a researcher and how you engage with the research agenda and are seen as a credible, legitimate scientist,” said Abrams.  

But qualitative researchers are becoming more interested in sharing data.  

“Until a few years ago, in qualitative research, the idea was, it just can’t be done,” said Don van Ravenzwaaij, a professor at the University of Groningen. “It’s a few years behind, but people are thinking about this. They’re seeing that it’s different. You can’t just copy the things that people in quantitative sciences are doing, but there are ways to do this and there is certainly an interest.”

Abrams agreed. “There is an increase in awareness about why we might want and need to [share data], the role it can play in reducing research waste, increasing innovation, and generally just enhancing public trust in research, as well.” 

Ethical motivations and challenges  

With data sharing, researchers with similar questions may be able to answer their questions without having to collect new data, granting huge savings of time and resources. It also helps to not have to contact the authors; even if data is said to be available upon request, when researchers contact authors for their data, they are only successful between 7% and 26% of the time (Gabelica et al., 2022; Wicherts et al., 2006). 

Data sharing also allows other researchers to verify and expand results, as well as reviewers to detect mistakes.

“We should be motivated to share our data so that other people who want to evaluate our claims are actually able to do that and to help us identify when we’ve made mistakes,” said APS Fellow Heather Urry, a professor at Tufts University.

That can be a deterrent if researchers are afraid of being discovered to have made an error, but Urry thinks such fears are overblown.  

“Obviously nobody wants that, but I think as scientists we should even more want to be sure that we’re protecting the scientific record,” she said. “Making your data public is one way of doing that.” 

But researchers, especially qualitative and clinical researchers, also have both ethical and logistical challenges to sharing data.  

“The last thing we would want is for our participants to be identifiable,” said Abrams. 

“The challenge is always to make sure that the data that you are planning to upload to something like the open science framework or another repository can’t identify people,” said Urry. “With qualitative research, there’s just more text to go through in order to find out whether or not any participants have actually said something that would be potentially identifying.”  

If many different demographic fields are collected and reported, it may be possible to single out individuals with a combination of those characteristics. Quotes might contain a name or location. Audio and video recordings need to be manipulated or blurred. The sheer volume of material—protocols, participant information sheets, consent forms, topic guides, data—is a lot to wade through. Additionally, answers in different languages may need to be translated. The time it takes to go through this process requires sufficient funding, a barrier to data sharing both quantitative and qualitative researchers relate with.  

Abrams shared her data for her paper on data sharing, because, she said, it would have been hypocritical not to. In the process, she learned how much effort it took. “That requires a lot of time, a lot of attention to detail, double checking, triple checking, even across the team as well, just to make sure that whatever you share can’t be traced back to an individual,” she said. 

Even with a smaller sample sized appropriately for qualitative studies (Abrams’s had 14 participants), this is a time-consuming process.  

“We are looking much more for information power, so the depth of the information rather than the breadth,” said Abrams.  

But depth means a lot of data.  

“When you deal with anything that has more open questions that you ask participants, I think that sort of mandates either coding everything so that you don’t put the actual answers out there, or at least going over the answers one by one to screen them,” said van Ravenzwaaij. “There is a point where the amount of effort required to do this kind of thing is so large that people are maybe no longer interested in doing that.”  

At the same time, there’s more to get out of these dense data. “The researchers that are making this effort probably create something maybe immortal,” van Ravenzwaaij said. “It is more effort, but also more reward.” 

How to share 

Another challenge Abrams identified in her study was the lack of knowledge about how to go about sharing data.  

This turned out to be the case for van Ravenzwaaij. When he started trying to convert files to make data sets shareable, he realized it wasn’t straightforward. “It was certainly harder than I thought it was at the outset,” he said.  

He talked to scientists who want to make their data publicly available, but realized the resources and infrastructure aren’t there. To try to help, van Ravenzwaaij used two real data sets to model the process both safely and effectively in a 2025 Advances in Methods and Practices in Psychological Science paper. 

Using behavioral and social sciences data sets (one on age differences in maintaining boundaries while working from home and one on the consequences of exposure to traumatic films), van Ravenzwaaij showed how to first systematically de-identify the data to ensure participants’ privacy. He then went step-by-step through guidelines for making the data FAIR (findable, accessible, interoperable, and reusable).  

The FAIR principles provide a structured approach to ensure data sets are easy for other researchers to understand and reuse. Data is assigned a unique identifier and registered in a searchable database (findable); retrievable by using an open, free protocol (accessible); machine-readable (interoperable); and both accurately and richly described and legally able to be reused (reusable). The paper provides a detailed example of how to follow these steps and covers the considerations and choices that may come up in the process. 

There’s a gray area between the ideal balance of sharing and protecting privacy, but as long as it’s done in good faith and choices are justified, “Don’t be afraid to not find the optimal solution,” said van Ravenzwaaij. “There is not one way to do this.” 

Urry agreed. “My best piece of advice is you don’t have to do everything all at once. You can start somewhere,” she said. “Carefully figure out what to share and why, and which pieces are the most important in order for others to be able to evaluate the work.” 

It’s also essential to start thinking about how to do this from the start of the project, such as with consent forms and ensuring there is time and money allotted for this part of the work. Abrams said funders should welcome paying someone to do this work, as it ultimately gives back to science, reduces waste, and encourages innovation.  

Funders and journals could both play a big part in encouraging open science. Though Urry would love to rely on the goodwill of researchers, data sharing remains a rare practice. “People simply aren’t doing it,” she said. “I suspect policies are going to be a big part of that solution.” 

But Abrams worries about blanket approaches or strict mandates for qualitative research.  

“If we take it too far in terms of making these things public, we could lose the voices that we’re trying to highlight,” she said. 

Fortunately, Abrams said she thinks sharing data in qualitative research is at an early enough stage to shape approaches and practices. In her study, the drive to share came from the researchers themselves: The main reason people shared their data was to be seen as trustworthy, credible researchers who help move the field forward.  

Indeed, once Urry learned how to do it and how helpful it is to the field and protecting the scientific record, she said she feels really good sharing her data now.  

“Yes, it’s a hassle. Yes, I’m sometimes sorry that I have to do it, and yes, it’s still worth it,” she said. “It really does feel like the right move.” 

Back to top