Imagine you’re a researcher in Tbilisi. You’ve designed a study, collected your data, run your analyses, and you’re ready to write up your findings. You go to read the latest literature on your topic – and you can’t. The paper you need is paywalled. Your university doesn’t have a subscription to that journal. The article costs $35 to download, and there are dozens more like it. You hit the wall.
Now imagine the reverse. You finish your study, publish your paper, and another researcher, somewhere in the world, wants to verify your findings. Or maybe they have a slightly different question they could answer with your data. They reach out, but your data isn’t shared anywhere. They can’t replicate your work, can’t build on it, can’t ask new questions of it. Your study sits frozen in time, useful only to those who happened to read it before moving on.
This is the everyday reality in much of academic psychology. And it’s exactly what open science is trying to change.
What Open Science Actually Means
When I talk about open science, I mean two things specifically. First, that academic papers should be accessible to anyone, not locked behind paywalls. Second, that the datasets behind those papers should be shared alongside them, available for other researchers to examine, replicate, and reuse.
That’s it. Two principles. But the implications are enormous.
The Access Problem
For researchers in Georgia, accessing academic literature is genuinely difficult. Most major journals are paywalled. APA journals, which I rely on as a psychologist, are particularly hard to access. Many of the foundational papers in my own areas of work sit behind subscriptions that cost more than entire research budgets in smaller countries.
This isn’t a minor inconvenience. It’s a structural barrier to scientific participation. When researchers in well-funded institutions can read everything and researchers elsewhere can read only what’s free, the result is a two-tier system. Knowledge accumulates faster in some places than others, not because researchers there are more talented, but because they have access to the inputs.
The solution exists. Open-access publishing, preprint servers like PsyArXiv, and open-access mandates from major funders are all moving in the right direction. There’s still the problem of predatory journals, outlets that charge fees but provide no real peer review or quality control. But predatory journals will always exist; they’re not an argument against open access. They’re an argument for being careful about where you publish. And as more reputable journals adopt open-access models, the predatory ones will become less appealing by comparison.
The Data Problem
The second issue is even bigger, and it hits Georgia hard. Despite a significant amount of psychological and social research being conducted here, almost none of the underlying data is publicly available. Studies are published, results are reported, and then the data disappears into the researcher’s personal hard drive, never to be seen by anyone else.
What’s lost when data stays private?
The first thing is replicability. The fundamental promise of science is that anyone with access to the same evidence should be able to reach the same conclusions. When data is shared, anyone can run the analyses themselves and verify whether the results hold. They can check for errors, examine the data quality, and see whether the statistical methods were applied correctly. Without shared data, peer review can only go so far. We’re trusting researchers’ word that their analyses are correct, and that’s not how science is supposed to work.
The second thing is what I’d call secondary discovery. When a researcher collects data for one study, they answer the questions they were interested in. But the data often contains information relevant to questions they never thought to ask. If that data is shared, other researchers can come along with different hypotheses, different theoretical frameworks, different methods and find new insights that the original author missed. The dataset becomes a resource, not just an artifact. And importantly, the original author still gets cited every time someone uses their data, which means data sharing is also good for individual researchers’ careers.
The Real Reasons People Don’t Share
To be fair, there are real reasons researchers hesitate to share their data. The most common one is also the most mundane: it’s a lot of work.
If you’re just analyzing data for your own paper, you can get away with messy spreadsheets, undocumented variable names, and shortcuts that only you understand. But if you want to share data with others, it needs to be cleaned properly, accompanied by a clear codebook, with all variables documented and all decisions explained. That’s a substantial additional task on top of writing the paper itself. It’s no different from the difference between informally telling colleagues about your findings and going through the formal process of publishing a peer-reviewed article. The published version is more polished, more rigorous, more useful to others, and it takes more time.
Other concerns include fear of having mistakes discovered, or fear of being scooped on follow-up papers. I think those concerns are mostly overstated. Most researchers aren’t going to dig through your data hoping to embarrass you, and the academic world has plenty of mechanisms to protect priority on follow-up work.
More importantly, if data sharing became standard practice across the field, the overall quality of research would improve dramatically. Errors would be caught earlier. Methods would be refined faster. Theories would face stronger empirical tests. The temporary discomfort of having your work scrutinized is a small price to pay for a more reliable scientific literature.
What Access Has Done For My Work
I want to be specific about what open data has enabled for me, because abstract arguments only go so far.
I’m currently writing my PhD using PISA 2018 data. That work is possible only because the OECD makes PISA data publicly available to anyone, anywhere, for any purpose. The same goes for TIMSS and PIRLS data from the IEA. These large-scale international assessments are among the most valuable data sources for educational and psychological research, and they’re available because the organizations that produce them have made open access a core principle.
Within Georgia, the Caucasus Research Resource Centers (CRRC) is the standout example. They share their survey data publicly, and that openness has enabled countless research projects. But beyond CRRC and the international assessment data, the landscape of openly available Georgian research data is sparse. That has to change.
What Georgia Needs
Building a stronger open science culture in Georgia will require effort on multiple fronts, but I think individual researchers can lead the way. If even a small group of researchers commits to making their data available with every publication, others will follow. Practices spread by example.
Universities can also play a role. Imagine if Georgian universities created data repositories, something modeled on Harvard Dataverse, where students could deposit their thesis datasets, where faculty could archive their research data, and where everything would be freely accessible to other researchers. The technology exists. The platforms are well-developed. What’s needed is the institutional will to set them up and the cultural shift to use them.
Funding mandates would help too. If grants required data sharing as a condition, compliance would follow quickly. But waiting for top-down mandates means waiting a long time. The faster path is for researchers to simply start.
Where to Begin
If you’re reading this and thinking “I’d like to be more open, but I don’t know where to start,” here’s the simplest answer: start by making your data available.
There are many platforms designed exactly for this. The Open Science Framework (OSF) is free, well-supported, and integrates with most workflows. Zenodo, Figshare, and Dataverse are also excellent. Choose one, upload your data with a basic codebook, and you’ve taken the most important step.
Beyond data, you can also pre-register your studies. Platforms like OSF let you write down your hypotheses, methods, and analysis plans before you collect data. This protects you against the temptation of post-hoc rationalization and makes your research more transparent. It’s another small step that has a big impact.
Be as open as possible. That’s the principle. Share your data. Share your code. Pre-register your hypotheses. Use preprints. Publish in open-access venues when you can.
None of this is easy at first. It requires extra work, and the rewards are partly diffuse, you’re contributing to a healthier scientific ecosystem rather than getting a direct personal payoff. But over time, open practices become normal, the work gets faster, and the field gets better.
Science is a collective enterprise. Treating it that way isn’t optional anymore.
—
Giorgi Tchumburidze
April 2026