Measuring Verifiability in Online Information (1509.05631v2)

Abstract: The verifiability of online information is important, but difficult to assess systematically. We examine verifiability in the case of Wikipedia, one of the world's largest and most consulted online information sources. We extend prior work about quality of Wikipedia articles, knowledge production, and sources to consider the quality of Wikipedia references. We propose a multidimensional measure of verifiability that takes into account technical accuracy and practical accessibility of sources. We calculate article verifiability scores for a sample of 5,000 articles and 295,800 citations, and compare differently weighted models to illustrate effects of emphasizing particular elements of verifiability over others. We find that, while the quality of references in the overall sample is reasonably high, verifiability varies significantly by article, particularly when emphasizing the use of standard digital identifiers and taking into account the practical availability of referenced sources. We discuss the implications of these findings for measuring verifiability in online information more generally.