Pág. 492-

of a discipline or scientific field by means of counting and analysing the different aspects of written communication, and as such they are considered to be highly useful in the world of research and science in general. They provide valuable information about the research activity of a discipline, country, institution or journal in terms of quantitative indicators (i.e., number of publications, of authors, of journals) and impact (i.e., number of citations for a study, journal or author); they also enable comparative studies of these indicators to be carried out. It is not surprising, therefore, that governments use these data, among others, when making decisions about the awarding of grants or offering promotion to personnel; for example, in the USA, bibliometric indicators have been used since the 1970s in the process of distributing research funding, and in both Scandinavia and Switzerland researchers have drawn up detailed maps of scientific output in a wide range of knowledge areas (Ball & Tunger, 2006). Bibliometry may also be used to suggest «future trends» in a given area of research, that is, to provide information about possible developments in the field in terms of productivity. As Ball and Tunger (2006) point out, the past of a discipline may be evaluated by counting the number of articles published during a given period (which may be extensive), its present by calculating the number of citations for the various studies, and its future by considering fluctuations in the number of publications and the number of times the work is cited. Scientific production on the Mantel-Haenszel procedure as a way of detecting DIF