Data citation and peer review

Abstract

A scientific publication is fundamentally an argument consisting of a set of ideas and expectations supported by observations and calculations that serve as evidence of its veracity. An argument without evidence is only a set of assertions. Consider the difference between the statement "The hairy woodpecker population is declining in the northwest region of the United States" and the statement "Hairy woodpecker populations in the northwest region of the United States have declined by 11% between 1992 and 2003, according to data from the Institute for Bird Populations (http://www.birdpop.org/)." Both or neither of these statements could be true, but only the second one can be verified. Scientific papers do, of course, present specific data points as evidence for their arguments, but how well do papers guide readers to the body of those data, where the data's integrity can be further examined? In practice, a chasm may lie across the path of a reviewer seeking the source data of a scientific argument.