When developmental scientists share research data (e.g., CHILDES, AddHealth, ECLS, Early Head Start), the field grows and prospers. In addition, data sharing:
- meets funder and journal requirements,
- reflects our duty to participants and other scientists,
- improves scientific practices by increasing transparency and reproducibility, and
- expands our influence as researchers.
Funder and Journal Requirements
Federal funding agencies encourage or mandate data sharing, and frequently request "data management plans" in grant proposals where investigators must delineate the ways they will share data upon study completion. As one example, NSF's Data Sharing Policy states that, "Investigators are expected to share with other researchers, at no more than incremental cost and within a reasonable time, the primary data, samples, physical collections and other supporting materials created or gathered in the course of work under NSF grants. Grantees are expected to encourage and facilitate such sharing." As a second example, NICHD states, "Data sharing is essential to speed translation of research results into knowledge, therapies, and procedures to improve human health" (https://www.nichd.nih.gov/research/resources/Pages/index.aspx). These policies make it clear that data sharing is often an expectation for PIs with federal funds, and an obligation of their institutions (the Grantees).
Many journals also require data sharing. Over 5,000 journals and publishers have signed onto the Center for Open Science's Transparency and Openness Promotion (TOP) guidelines at various levels of data sharing, including Elsevier, Springer, and Wiley.
In short, the data sharing train has "left the station" – with funders and journals increasingly requiring data sharing as a condition of support.
Duty to Participants…and Other Scientists
Adherence to the foundational Belmont principles that govern research ethics suggests that researchers must consider data sharing and reuse as a duty to participants (Brakewood & Poldrack, 2013). Data sharing increases the beneficence of research studies by giving more "bang for the buck" – by sharing and reusing data, researchers maximize participants' contribution by continuing to make scientific use out of data that would otherwise sit dusty on the shelf.
We also have a duty toward our fellow research scientists, and data sharing increases justice. Not all institutions and researchers have access to the necessary resources for primary data collection. Data sharing allows early career scholars and those with limited resources to bring their unique perspectives to new research questions that will accelerate the growth of scientific knowledge designed to improve the human condition.
Transparency and reproducibility are foundational to the behavioral sciences, and video data sharing is the primary vehicle for ensuring we meet these goals. By allowing other researchers access to our data and our procedures, we foster replication, allowing us to be more certain of the broader applicability of our findings, or the ways in which contextual variation can lead to different findings. Data sharing increases standardization across researchers and labs, both in procedures (e.g., template language for data sharing releases; consensus on how to ask participants for permission to share) and content (e.g., definitions of developmental terms; break points of categorical variables), to enable comparisons of results across studies.
Data sharing also greatly increases our ability to connect and collaborate with researchers across the world, because data can be uploaded prospectively from one lab and coded remotely by researchers in another. Data sharing is a powerful vehicle for establishing and promoting cross-national collaboration, and providing tools to researchers in other countries. Through the process of data sharing, researchers establish infrastructures for transparency in protocols and providing valuable data with colleagues and students in other countries.
Finally, collaboration and reuse greatly expand our datasets, multiplying sample size (and power), and allowing us to examine cohort effects, and assess possible differences due to geographical location or population characteristics. For example, there are new, emerging international efforts to "crowd-source" psychological science, like the Many Labs project, the Many Babies project, TalkBank, the "CERN for Psychological Science", and the PLAY (Play and Learning Across a Year) project, which hold promise for producing larger and more diverse data sets than most individual PIs could possibly organize on their own.
Expanding Your Influence
Data sharing also allows you to expand your influence in the field. When you opt to share your dataset, you receive a persistent DOI and, as others reuse those data, you get credit in the currency of our field – citations. In addition, when you opt to share your procedures, your expertise in obtaining consent, collecting data, coding, training staff to code or collect data…all of these become the foundation upon which other researchers, particularly students and early career scholars, can build.
For data sharing to become the norm requires researchers to commit to the scientific value and benefits of data sharing, as outlined above. The question is how to make data sharing no more than an "incremental" cost to researchers while maximizing the benefits to the field as a whole. Databrary provides many of the resources to reduce the cost to researchers: In addition to a user-friendly repository, Databrary includes template language for obtaining participants' permission to share their data, boilerplate data management plans, and suggestions for obtaining IRB approval to share data.