It has been shown (S. Lawrence, 2001, Nature, 411, 521) that journal articles which have been posted without charge on the internet are more heavily cited than those which have not been. Using data from the NASA Astrophysics Data System (ads.harvard.edu) and from the ArXiv e-print archive at Cornell University (arXiv.org) we examine the causes of this effect.Recently Lawrence (2001) and Brody, et al. (2004) have demonstrated that articles which are available on-line at no charge are cited at substantially higher rates than those which are not. Kurtz (2004) has shown that restrictive access policies can cut article downloads to half the free access rate.There are (at least) three possible, and non-exclusive, explanations for the effect noted by Lawrence (2001) and Brody, et al. (2004).

Abstract. The NASA Astrophysics Data System Abstract Service has become a key component of astronomical research. It provides bibliographic information daily, or near daily, to a majority of astronomical researchers worldwide.We describe the history of the development of the system and its current status. Urania (Boyce 1996), and the ADS role in the emerging electronic astronomical data environment are discussed. Astronomy is unique in that it already has a fully functional data resource, where several of the most important data sources exist on-line and interoperate nearly seamlessly. The ADS and the Strasbourg Data Center (CDS; Genova et al. 2000) form the core of this resource.We show several examples of how to use the ADS, and we show how ADS use has increased as a function of time. Currently it is still increasing exponentially, with a doubling time for number of queries of 17 months.Using the ADS logs we make the first detailed model of how scientific journals are read as a function of time since publication. We find four distinct components. We directly compare the readership rate with the citation rate for scientific articles as a function of age. Citations generally follow reads, but there are some differences.The main journals of astronomy have differences in the ways they are read and cited. We discuss these from a number of different aspects.The impact of the ADS on astronomy can be calculated after making some simple assumptions. We find that the ADS increases the efficiency of astronomical research by 333 Full Time Equivalent (2000 hour) research years per year, and that the value of the early development of the ADS for astronomy, compared with waiting for mature technologies to be adopted, is 2332 FTE research years.A full technical description of the ADS is in three companion articles:

Digital libraries such as the NASA Astrophysics Data System (Kurtz et al., 2005) permit the easy accumulation of a new type of bibliometric measure, the number of electronic accesses ("reads") of individual articles. We explore various aspects of this new measure. We examine the obsolescence function as measured by actual reads and show that it can be well fit by the sum of four exponentials with very different time constants. We compare the obsolescence function as measured by readership with the obsolescence function as measured by citations. We find that the citation function is proportional to the sum of two of the components of the readership function. This proves that the normative theory of citation is true in the mean. We further examine in detail the similarities and differences among the citation rate, the readership rate, and the total citations for individual articles, and discuss some of the causes. Using the number of reads as a bibliometric measure for individuals, we introduce the read-cite diagram to provide a two-dimensional view of an individual's scientific productivity. We develop a simple model to account for an individual's reads and cites and use it to show that the position of a person in the read-cite diagram is a function of age, innate productivity, and work history. We show the age biases of both reads and cites and develop two new bibliometric measures which have substantially less age bias than citations: SumProd, a weighted sum of total citations and the readership rate, intended to show the total productivity of an individual; and Read10, the readership rate for articles published in the last 10 years, intended to show an individual's current productivity. We also discuss the effect of normalization (dividing by the number of authors on a paper) on these statistics. We apply SumProd and Read10 using new, nonparametric techniques to compare the quality of different astronomical research organizations.

, has developed a distributed online digital library which has become the dominant means by which astronomers search, access, and read their technical literature. Digital libraries permit the easy accumulation of a new type of bibliometric measure: the number of electronic accesses ("reads") of individual articles. By combining data from the text, citation, and reference databases with data from the ADS readership logs we have been able to create second-order bibliometric operators, a customizable class of collaborative filters that permits substantially improved accuracy in literature queries. Using the ADS usage logs along with membership statistics from the International Astronomical Union and data on the population and gross domestic product (GDP), we have developed an accurate model for worldwide basic research where the number of scientists in a country is proportional to the GDP of that country, and the amount of basic research done by a country is proportional to the number of scientists in that country times that country's per capita GDP. We introduce the concept of utility time to measure the impact of the ADS/ URANIA and the electronic astronomical library on astronomical research. We find that in 2002 it amounted to the equivalent of 736 full-time researchers, or $250 million, or the astronomical research done in France.

In this report we examine the change in citation behavior since the introduction of the arXiv e-print repository (Ginsparg, 2001). It has been observed that papers that initially appear as arXiv e-prints get cited more than papers that do not (