Exchange of research data remains low and increases slowly

Photo: Tripp

Photo: Tripp.

Over the past 10 years noticeable changes in open data policies and data sharing were produced, especially in the biological sciences. A survey conducted in 20131 with 1,600 researchers in biological sciences manifests that this shift toward data sharing was produced by three main causes:

  • The policies of funding agencies such as the NIH in US and also the Human Genome research projects.
  • The tools and infrastructure available to share data openly.
  • The publication policies of high impact scientific journals.

Despite significant changes, the effectiveness of these policies is still far from the desired goal, because one third of funders do not consider important in reviewing projects plans to share data, and another third of the agencies ignore these requirements. The results of other research areas outside the biological sciences, however, have poorer outcomes.

Analyzing from the standpoint of the paradigm of scientific research – as the dissemination of results and reproducibility – it seems strange this “resistance” to share data. A reasonable explanation is that there are tensions between what public interests are and what the private interests of individual scientists are. Science as a whole moves through the dissemination of information… but the careers of scientists advances by personal contributions, priorities and individual credits.

Available open data is a constitutive part of the scientific method as it allows the verification of the results and progress in research. Many scientists are willing to make available their data online, but cannot for lack of financial resources or infrastructure. Funding agencies that consider this point in their design reviews take into account the preservation of data in short-term, but not long-term.

In order to encourage data sharing policies the NIH in the United States, from 2003, began to require that requests for funds for research projects to include plans to preserve the data openly and recently also effected a policy for the Genome project, which will begin in 2015, in the same direction2.

According to the above mentioned survey in second place and minor relevance, publication policies of specialized journals had some positive influence in encouraging the sharing of research data. A recent study3 conducted with 50 high-impact journals shows that 88% of these have some sort of instruction to authors concerning the public availability of research data for sharing, however the requirements varied widely, ranging from the availability of primary data to a statement indicating that the data will be sent upon request. However, despite the statement in the instructions of the journals, the reality of the study shows that a substantial proportion of articles published in high impact journals do not adhere to the data availability policies.

Moreover, while policies and technological and infrastructure tools have been developed to share data, also contrary policies have been developed. Consequently, difficulties and barriers raised in the intellectual property to protect the financial interests of universities and companies that fund research, imposing conditions and restrictions on what and how much material can be shared, clauses on technology transfers, trade secrets, and patents, including national security reasons.

While journals and funding agencies have been proactive in expanding open data, universities and medical research centers have endeavored to preserve the institutional intellectual property.

Another problem that manifests itself in the research is the instrumentation policies of funding agencies. Its greatest weakness is that it seems that there are very few formal or informal sanctions for those not complying with the requirements.

The main recommendations of the scientists interviewed are that it should be established as an implicit norm that research data, rather than being regarded as private property, should be considered as public goods. They also suggest that when equally competitive projects are presented, should be favored those who establish open data and if the assumptions are not met they shall be penalized.

Another important recommendation is to find less bureaucratic and costly procedures to deposit data in open repositories. One way to reduce costs is standardization of procedures for storage and preservation of data.

The obvious question then is whether there is a genuine willingness to share data among the scientific community or it is only one of many rules of conduct that we expect to be others to obey them, not us.

Also interesting are the reflections of an article published by WICHERTS, BAKKER, MOLENAAR in 2011 in PLoS ONE4 where the they point out:

Widespread reluctance to share research data is often attributed to the fear of the authors of that reanalysis may expose errors in their work or may produce conclusions that contradict their own…

and, after analyzing the statistical results of important works of psychology, the article concludes that:

We found certain reluctance to share data associated with weaker evidence (against the null hypothesis of no effect) and higher prevalence of apparent errors in the presentation of statistical results. Unwillingness to share data was particularly clear when reports of errors affect statistical significance.

The authors’ conclusion is that mandatory policies should be established to archive the data.

Importantly, SciELO has defined as a criterion for indexing5 the positioning of the journals regarding the availability of research data.

Notes

1 PHAM-KANTER, G., ZINNER, D.E., and CAMPBELL, E.G. Codifying Collegiality: Recent Developments in Data Sharing Policy in the Life Sciences. PLoS ONE. 2014, vol. 9, nº 9. doi:10.1371/journal.pone.0108451. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0108451

2Final NIH statement on sharing research data. National Institutes of Health: Office of Extramural Research. 2003. Available from: http://grants.nih.gov/grants/guide/notice-files/NOT-OD-03-032.html

3 ALSHEIKH-ALI, A.A., et al. Public Availability of Published Research Data in High-Impact Journals. PLoS ONE. 2011,vol. 6, nº 9. doi:10.1371/journal.pone.0024357. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0024357

4 WICHERTS, J.M., BAKKER, M., and MOLENAAR, D. Willingness to Share Research Data Is Related to the Strength of the Evidence and the Quality of Reporting of Statistical Results. PLoS ONE. 2011, vol. 6, nº 11. doi:10.1371/journal.pone.0026828. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0026828

5 Critérios, política e procedimentos para a admissão e a permanência de periódicos científicos na Coleção SciELO Brasil. SciELO. 2014. Available from: http://www.scielo.br/avaliacao/Criterios-SciELO-Brasil-20141003.pdf

References

ALSHEIKH-ALI, A.A., et al. Public Availability of Published Research Data in High-Impact Journals. PLoS ONE. 2011,vol. 6, nº 9. doi:10.1371/journal.pone.0024357. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0024357

PHAM-KANTER, G., ZINNER, D.E., and CAMPBELL, E.G. Codifying Collegiality: Recent Developments in Data Sharing Policy in the Life Sciences. PLoS ONE. 2014, vol. 9, nº 9. doi:10.1371/journal.pone.0108451. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0108451

PIWOWAR, H.A. Who Shares? Who Doesn’t? Factors Associated with Openly Archiving Raw Research Data. PLoS ONE. 2011, vol. 6, nº 7. doi:10.1371/journal.pone.0018657. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0018657

TENOPIR, C., et al. Data Sharing by Scientists: Practices and Perceptions. PLoS ONE. 2011, vol. 6, nº 6. doi:10.1371/journal.pone.0021101. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0021101

WICHERTS, J.M., BAKKER, M., and MOLENAAR, D. Willingness to Share Research Data Is Related to the Strength of the Evidence and the Quality of Reporting of Statistical Results. PLoS ONE. 2011, vol. 6, nº 11. doi:10.1371/journal.pone.0026828. Available from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0026828

 

Ernesto SpinakAbout Ernesto Spinak

Collaborator on the SciELO program, a Systems Engineer with a Bachelor’s degree in Library Science, and a Diploma of Advanced Studies from the Universitat Oberta de Catalunya (Barcelona, Spain) and a Master’s in “Sociedad de la Información” (Information Society) from the same university. Currently has a consulting company that provides services in information projects to 14 government institutions and universities in Uruguay.

 

Como citar este post [ISO 690/2010]:

SPINAK, E. Exchange of research data remains low and increases slowly [online]. SciELO in Perspective, 2014 [viewed ]. Available from: https://blog.scielo.org/en/2014/11/12/exchange-of-research-data-remains-low-and-increases-slowly/

 

Leave a Reply

Your email address will not be published.

Post Navigation