What is wrong with obsolescence?

The growth of scientific output in recent years has meant that fewer libraries are able to offer the entire range of journals, with the others being forced to make a selection. The objective of the present work is to describe criteria to regulate the selection of these journals to provide the researcher with the information that is most being used in research. One form of quantifying this information is by way of the citations that papers receive over a period of time following their publication. Obsolescence, expressed in terms of an annual aging factor, does not reflect the real behaviour of most papers. An alternative is the use of “topicality,” considered as a latent variable, with the Rasch model as the measuring instrument. We considered 45 physics journals, and found the results of applying the Rasch model to be more satisfactory than those obtained with the annual aging factor.