Open access 2(1)Editor’s Note: This article was written by former ADV author Danilo Freire.

Cooperation is one of the main drivers of modern science. Co-authorships and peer reviews are so common in the academia that it is not necessary to point out their merits. However, not only much of the scientific knowledge currently being produced is still behind paywalls, but also many authors do not adopt procedures to make their research replicable and reproducible. What should we do in order to move to a more open science? 

The first thing to consider is the academic content providers. Most academic journals are available only in subscription bundles, whose prices has increased dramatically over the last decades. As noted by Priceonomics, in 2002 subscription costs are on average 600% higher than in 1984, and Harvard now spends about 3.75 million dollars per year for online journal content, a situation that the faculty committee qualifies as “fiscally unsustainable and academically restrictive”. When the world’s wealthiest university says it cannot afford journal subscriptions, you can guess there is something very wrong with the academic publishing model. It is indeed a bizarre scenario: universities and public agencies pay scholars to carry out their research, pay the salaries of peer reviewers, and they still have to spend another considerable sum just to read the final result of what they had funded. It is very hard to argue that this is fair.

One possible solution is to publish your papers in open access journals. The premise behind the open access publishing model is simple: scholarly research should be available free of charge for those who want to read it. While several open access journals are entirely free, some of them charge a fee to the authors after the paper has been accepted in order to cover its publishing costs (academics from developing countries usually pay less). Nevertheless, the distribution of content remains open to all, usually under Creative Commons’ licenses. It is true, however, that some open access journals have poor peer-review systems, so a preliminary research about the quality of the publication to which you want to send your article is always necessary.

Another possibility is to carve a breach in the paywalls. Prestigious journals are rarely free of charge, but as noted by Gary King and Stuart Shieber, two Harvard professors, it is often possible to negotiate with publishers to make your article publicly available. You could then publish it on your personal webpage where people can download it for free. Since publishing in a heavyweight journal is a must for every aspiring scholar, this might be a feasible, if not necessary, compromise between closed and open access.

Finally, researchers should pay more attention on the tools they use to produce and store their empirical results. Data repositories and open source software are important means for widening your academic audience and guaranteeing that your work can be properly replicated. As this week’s The Economist points, out, replication is an underestimated and overlooked pillar of science. There are relatively few scholars who make their datasets available to the public with an accompanying script, and even when they do so, they usually keep their data in proprietary formats (SPSS, Stata, Matlab). While this is not bad per se, it ends up forcing someone to buy expensive software licenses to replicate the results. As a consequence, the replication process turns to be not only unnecessarily costly but also dependent of the estimating techniques embedded in those softwares, which are not always well-documented or reviewed by the academic community. In this sense, why not make use of free and open source software? Instead of using Stata or SPSS, political scientists would make their research easier to replicate by using R, an extremely powerful statistical language whose source code can be checked by everyone and which also has nice packages targeted at beginners. R use has risen significantly over the past decade, and political scientists can be part of that trend too.

In sum, a few small steps can help scientists to break the vicious circle of closed access science. Making academic research more transparent is crucial for fostering collaboration (and competition) amongst scholars, and that is what science is all about. Piece by piece we can tear down the paywalls.