Estimation of some missing bibliometric indicators at INDUSTRIAL HEALTH.

(ISI) in 1987 and received its inaugural impact factor in the 1989 Journal Citation Reports ® (JCR ®) 1). For various reasons however, no bibliometric indicators were issued between 1995 and 1997 2). In this article we present some estimates of the 'missing' three years, so that a continuous bibliometric history for INDUSTRIAL HEALTH can be analyzed for the first time. Impact factors are derived from the Science Citation Index ® (SCI ®), which was first proposed by Eugene Garfield in 1955 as a means for improving the dissemination and retrieval of scientific literature 3). Calculation is still undertaken using Garfield's original formula, whereby journal articles published in a given two-year period are tracked to see if they are subsequently cited in the following year. The number of citations received is then divided by the total number of citable articles published in that same two-year period 4). Not all articles are counted however, with the selection of those considered to be 'citable' being ascertained by Thomson Reuters, formerly known as the ISI. While official impact factor scores for ISI-listed journals are published in the yearly JCR ® , the list is not all-inclusive. BioMed Central (BMC) for example , contains many new titles which have not yet had sufficient time for their citations to be counted or processed, and for these reasons, calculates their own Unofficial Impact Factors using the SCI ®5). Other methods for predicting impact factors have also been recently described 6). To estimate the 'missing' or 'unofficial' bibliometric indicators at INDUSTRIAL HEALTH, it is necessary to examine articles published over a 4-yr period. This occurs because the 1995 score relates to articles published dur-Secondly, it is necessary to ascertain which articles should actually be included in the denominator, that is, which articles might be deemed 'citable'. Eugene Garfield published some insight as to how such criteria were originally evaluated 7) , based on what types of journal articles the ISI had found were being regularly cited. In his article, Garfield listed five criteria for which points would be awarded and explained how a paper receiving four or more points at the time was considered to be 'substantial' 7). For the current investigation, all articles published by INDUSTRIAL HEALTH between 1993 and 1996 (Volumes 31 to 34) were manually examined and evaluated using the afore-mentioned criteria. From a total of 109 articles published during these four years, 41 …