Development Research in Practice: The DIME Analytics Data Handbook. By Kristoffer Bjärkefur, Luíza Cardoso de Andrade, Benjamin Daniels, and Maria Jones
On behalf of the Development Data Partnership, I'm delighted to submit comments on Chapter 7: Publishing Research Outputs.
Overall, the Chapter is well written, with its justified, practical advisory. Following are minor comments for the team's consideration.
LaTeX
I like how the authors not only present the pros and cons of using LaTeX, but also present options to help overcome some of the challenges, such as introducing cloud-based implementations like overleaf.com to broaden the scope of potential users.
Risk of Disclosure
In the introduction, there is a diagram on page 13 (no title or figure number – the authors may consider adding these to support future reference), where under “publication”, one of the steps is to “Asses the risk of disclosure.” While Chapter 7 introduces different methods to reduce risk of data re-identification, the chapter doesn’t address other ethical issues related to risk of disclosure of final results, which may impact whether research should be provided directly to its intended audience for action or published as a global public good.
Chapters 1 and 2 of the handbook address privacy and basic ethical risks in project design. But the authors may consider a framework for Chapter 7 for reviewing not just the data, but also the analytical results prior to publication, to evaluate the risk of misconstrued findings or the potential harm findings could have on a given set of stakeholders.
Licensing
Chapter 7 provides a number of helpful options for publishing reproducibility packages but doesn’t include any mention of licensing options for analytical work or derived datasets that will be published. This may be outside the scope of the handbook, but the authors may consider including a link to an existing resource (if any) to guide researchers through the different standard open-source licenses available and under what conditions different licenses may be selected.
Dear DIME Team,
On behalf of the Development Data Partnership, I'm delighted to submit comments on Chapter 7: Publishing Research Outputs.
Overall, the Chapter is well written, with its justified, practical advisory. Following are minor comments for the team's consideration.
LaTeX I like how the authors not only present the pros and cons of using LaTeX, but also present options to help overcome some of the challenges, such as introducing cloud-based implementations like overleaf.com to broaden the scope of potential users.
Risk of Disclosure In the introduction, there is a diagram on page 13 (no title or figure number – the authors may consider adding these to support future reference), where under “publication”, one of the steps is to “Asses the risk of disclosure.” While Chapter 7 introduces different methods to reduce risk of data re-identification, the chapter doesn’t address other ethical issues related to risk of disclosure of final results, which may impact whether research should be provided directly to its intended audience for action or published as a global public good.
Chapters 1 and 2 of the handbook address privacy and basic ethical risks in project design. But the authors may consider a framework for Chapter 7 for reviewing not just the data, but also the analytical results prior to publication, to evaluate the risk of misconstrued findings or the potential harm findings could have on a given set of stakeholders.
Licensing Chapter 7 provides a number of helpful options for publishing reproducibility packages but doesn’t include any mention of licensing options for analytical work or derived datasets that will be published. This may be outside the scope of the handbook, but the authors may consider including a link to an existing resource (if any) to guide researchers through the different standard open-source licenses available and under what conditions different licenses may be selected.
Holly Krambeck