2

Descriptive analyses

The easiest way to begin a bibliometric study is through descriptive analyses of the main characteristics of the studies that will be included. These will offer a quick and even visual impression of certain aspects related to productivity in the given field of study.

Temporal evolution

Once all the documents that will form part of the study have been collected it is then possible to analyse how they have evolved over time. In order to conduct this analysis it is necessary to know the year of publication of each study and to calculate the frequency of studies published each year. It will then be possible to identify the trend of scientific productivity in a given area of study. All the significant information about temporal evolution, such as year, number of studies (frequency), percentage and cumulative percentage is usually collated in a table. These data, along with a graphical representation, will show the trend in output.

A good example of the temporal evolution of scientific productivity can be seen in the area of the human immunodeficiency virus (HIV), which leads to acquired immunodeficiency syndrome (AIDS). Table 2.1 shows, over the years, the number of publications found in the Medline database containing the acronym HIV in the topic.

Table 2.1

Publications in Medline about HIV

1970 1 < .001 < .001 1975 1 < .001 < .001 1982 29 0.01 0.01 1983 696 0.33 0.34 1984 1,153 0.54 0.89 1985 1,759 0.83 1.71 1986 2,944 1.39 3.10 1987 4,649 2.19 5.29 1988 7,433 3.50 8.79 1989 7,627 3.59 12.38 1990 8,531 4.02 16.39 1991 9,277 4.37 20.76 1992 9,546 4.49 25.25 1993 9,543 4.49 29.75 1994 9,275 4.37 34.11 1995 10,241 4.82 38.93 1996 10,568 4.97 43.91 1997 10,331 4.86 48.77 1998 10,999 5.18 53.95 1999 10,638 5.01 58.96 2000 10,190 4.80 63.76 2001 10,366 4.88 68.64 2002 10,173 4.79 73.42 2003 10,571 4.98 78.40 2004 10,987 5.17 83.57 2005 11,352 5.35 88.92 2006 11,514 5.42 94.34 2007 12,019 5.66 100.00

Although a couple of articles about immune deficiency diseases were published during the 1970s, it was not until the 1980s that a clear interest in this topic emerged. From this point on, one can say that an information explosion took place. This result is consistent with the fact that it was in 1983 that the human immunodeficiency virus was found to lead to AIDS. A graphical representation of the temporal evolution of HIV research (Figure 2.1) enables us, at a glance, to understand the data shown in

Figure 2.1 Temporal evolution of scientific productivity about HIV research in Medline

Table 2.1. Figure 2.1 clearly shows a sharp increase in research activity up until 1990, followed by a constant increase up to the present day.

Another example of the temporal evolution of productivity is that provided by Miguel-Dasit et al. (2008). These authors analysed how productivity in the field of magnetic resonance (for Spanish researchers) progressed from 2001 to 2007 in the Medline database. As can be seen in Table 2.2 they divided the articles found into different categories according to topics within magnetic resonance research. They also provide a general count of the number of papers published during this period.

Table 2.2

Temporal evolution of productivity in magnetic resonance production in Spain from 2001 to 2007

Abdominal 4 (17%) 1 (3%) 7 (27%) 3 (7%) 7 (17%) 5 (17%) 2 (3%) 29 (100%) Breast – 1 (11%) – 2 (22%) 3 (33%) 2 (22%) 1 (11%) 9 (100%) Cardiac 1 (5%) 3 (14%) 3 (14%) 1 (5%) 6 (28%) 4 (19%) 3 (14%) 21 (100%) Chest – 1 (25%) – – – 3 (75%) – 4 (100%) Contrast media 1 (17%) 1 (17%) 3 (49%) 1 (17%) – – – 6 (100%) Genitourinary 5 (31%) 2 (6%) 1 (6%) 4 (19%) 2 (12%) 3 (19%) 1 (6%) 17 (100%) Head and neck 3 (75%) – – 1 (25%) – – – 4 (100%) Musculoskeletal 8 (13%) 10 (12%) 12 (20%) 11 (16%) 8 (14%) 8 (14%) 10 (13%) 67 (100%) Neuroradiology 18 (13%) 27 (19%) 15 (11%) 18 (13%) 14 (9%) 25 (19%) 22 (14%) 139 (100%) Pediatrics – 2 (22%) 2 (22%) 2 (22%) 2 (22%) 1 (11%) 2 (11%) 11 (100%) Vascular 1 (12%) 2 (12%) – 4 (37%) 1 (12%) 2 (25%) – 10 (100%) Computer applications – – – 1 (33%) – 1 (33%) 1 (34%) 3 (100%) Miscellaneous – – 3 (27%) 2 (18%) 1 (9%) 3 (27%) 2 (18%) 11 (100%) Total no. of papers and mean IF of MR publications 41 (12%) 50 (15%) 47 (14%) 51 (15%) 44 (13%) 57 (17%) 42 (12%) 332 (100%)

Reprinted with permission from Miguel-Dasit et al. (2008).

Depending on the number of studies, it may also be possible to represent the temporal evolution graphically in terms of ranges of years rather than specific years.

The temporal evolution of productivity can, in fact, be consulted via online databases. One database that offers this resource is the ISI Web of Knowledge. Thus when performing a search in any of the ISI Web of Knowledge databases it is possible to obtain information about the number of publications in a given period of time by means of a single click. When a search is carried out of a given topic, a list of articles is shown, along with the option ‘Analyse Results’. Selecting this option yields the number of publications in a field according to certain criteria, such as the year of publication.

The same analysis regarding temporal evolution could, alternatively, focus on productivity over time in a given journal. In this case, the number of publications will refer to those published in a specific journal.

By way of example Table 2.3 considers the report by Kirchler and Hölzl (2006), who conducted a bibliometric study of the Journal of Economic Psychology. It can be seen that the temporal evolution of productivity in this journal is analysed in terms of the number of articles published over time. Other descriptive data regarding certain characteristics of the articles are also included in this table.

Table 2.3

Temporal evolution of publications in the Journal of Economic Psychology

Articles published 127 135 182 190 220 854 Page count, 5% trimmed...