Proper documentation of sources is a key component of scholarly research, and the need for documentation is no less important for data sources than for bibliographic sources. It is vital that scholars in the social sciences and elsewhere who use quantitative data in their work be clear and explicit as to the sources for their data. This is in part because of the importance of replication in the sciences - in principle, your analysis must be capable of being replicated by other scholars so as to better-assess the soundness of your work, which requires that those other scholars have access to the data used in your original analysis.
Greater transparency in sources also encourages greater accountability in research and increases confidence that the data used in your work are suited for the questions you are asking, which in turn will make others more confident in whatever conclusions your research presents. In addition, thorough documentation of data sources makes it easier for scholars to assess whether sources used in your research are appropriate for use in their own work.
More generally, the scholarly community is placing a growing emphasis on greater transparency in and documentation of data sources, to the point where some scholarly journals even require authors to submit their data to archives where others can download them. We have also included examples of data availability policies from different journals in different fields as examples of how different fields choose to operationalize disciplinary norms of transparency in empirical research.
http://www.icpsr.umich.edu/files/ICPSR/enewsletters/iassist.html - This guide will help you figure out how to cite your data in a way that is informative and useful to others. See http://www.icpsr.umich.edu/icpsrweb/content/datamanagement/citations.html and https://youtu.be/xTUiefaq128 for additional information on and discussion of data citation, including its benefits for both collectors/producers of data and users of data.
http://www.esrc.ac.uk/files/funding/guidance-for-grant-holders/data-citation-what-you-need-to-know/ - This .pdf on data citation from the UK's Economic and Social Research Council is another guide to help you figure out how to cite your data in transparent and useful ways. See http://www.esrc.ac.uk/funding/guidance-for-grant-holders/data-citation/ for additional discussion from the ESRC on the need for data citation.
http://www.dartstatement.org/ - "The Data Access & Research Transparency Joint Statement": A statement of principles and standards for data availability and replication that has been endorsed by multiple Political Science journals.
http://media.wix.com/ugd/fa8393_d55bef088ac44830bd194b5f80190479.pdf - "Openness in Political Science: Data Access and Research Transparency": A symposium published in PS: Political Science & Politics in January of 2014 that articulates the need for greater transparency in research practices, for both quantitative and qualitative data.
http://ajps.org/ajps-replication-policy/ - "AJPS Replication Policy": The replication policy for the American Journal of Political Science, as an example of expectations of transparency in quantitative research in Political Science.
https://www.insidehighered.com/blogs/rethinking-research/should-journals-be-responsible-reproducibility - "Should Journals Be Responsible for Reproducibility?" A commentary from the editor of the AJPS and data archivists discussing the journal's data availability policy.
http://duckofminerva.com/2014/07/an-open-letter-from-the-new-dgs.html - "An Open Letter from the New DGS": Guidance for new graduate students from Emory alum Dr. Amanda Murdie at the University of Georgia.
http://www.theguardian.com/news/datablog/2011/jul/28/data-journalism - "Data Journalism at the Guardian: What Is It and How Do We Do It?" - The Guardian newspaper is a first-rate practitioner of data journalism and has much practical advice on how to do data journalism well. It also has a nice visualization of its workflow for prepping data for stories. Their key, and very valuable, insight: working with data is "80% perspiration, 10% great idea, 10% output".