Use of databases at Carlos III University, Mardid, 1995-1998

Carlos Suárez Balseiro, Elías Sanz-Casado, Pilar Vergara González
Info Technology Supply Ltd., Londyn-Warszawa, School of Communications, University of Havana. Dept. of Library & Information Science. Carlos III University of Madrid. Dept. of Library & Information Science. Carlos III University of Madrid.
3 Database Access Service. Carlos III University Library.


The processes involved in the consumption of information are essential to the continuous creation of knowledge in university environments. Hence the vital importance of studies on the use of the information resources available in academic institutions to enable them to attain their objectives, thereby contributing to the ongoing enhancement of the quality of higher education. A number of techniques is used in this paper to identify and highlight the features characterising the development of the Database Access Service at the Carlos III University Library in Madrid from 1995 to 1998. Trends in the use of network databases are evaluated and the indicators that typify tendencies and patterns by kind of user are analysed. Certain criteria on the interaction between users, new technologies and electronic sources of information in university libraries are likewise discussed.

Key words: User studies, indicators, information use, databases


One of the features of present research trends in general and in universities in particular is that in today’s multifunctional universities all processes rely heavily on information and knowledge. In this regard any study on information production and consumption processes in academic environments plays an essential role in improving institutional performance and the quality of higher education in general (ARTILES VISBAL and GONZÁLEZ GARCÍA, 1997; WILSON, 1998).

At this time, the steep rise in the volume of information, along with diversification of sources, constant and increasingly complex changes in user community information needs and demands, not to mention the preponderance of information and communication technologies, oblige libraries to constantly raise their efficiency levels. Studies on information consumption are, then, of primary importance, since they constitute an excellent management tool to ensure the most suitable library development in any context. This kind of studies is conducted applying a wide variety of techniques, including bibliometric techniques, "which make it possible to measure different user characteristics, from the standpoints of both their scientific production and their consumption of information... which has proved to be of great use in defining the information systems and services best suited to them" (SANZ CASADO and MARTÍN MORENO, 1997).

The high-geared development of information technologies over the last two decades has brought on a substantial increase in the quantity and quality of available resources. In this regard, databases are one of the sources whose accessibility and use have grown most rapidly. According to the Task Force Pro Libra (TFPL), whereas in the early eighties, for instance, there were 48 CD-ROM databases, by 1997 there were over 20,000. This same source indicates that Germany and France are driving the extraordinary development observed in Europe, where the number has been growing at a rate of 90% per year (KEENAN, 1998; LEACH, 1998; MCCARTHY, KRAUSSE and LITTLE, 1997).

In Spain, systems for accessing CD-ROM databases were first instituted in university environments in 1988. That year, ten academic libraries began to provide their users with access to the information in their CD-ROM databases. By 1991, use of these services had outpaced on-line access service indicators (MOSCOSO, NOGALES and CARIDAD, 1995). By the mid-nineties the use of this technology in Spanish university libraries came to 47.1% and the application of new information and communication technologies in general was on the rise: plans for the automation of library services, purchase of CD-ROMs and institutional networking were among the most prominent effects of these trends (ARRUDA RAMALHO, 1994; RAITT and CHEN, 1990).

The present survey aims to characterise the evolution of the various types of Carlos III University library users in terms of information consumption, measured on the grounds of the database queries made through the library’s Database Access Service, accessible to users over the university’s intranet. The results of the study show fundamental differences in access patterns among the various communities of university users: professors, students, library personnel and administration and service personnel (P.A.S.).


The databases selected for the study together accounted for 80% of all database visits in the period 1995-1998. This guarantees reliability, since these bases, as the core information sources, satisfy the vast majority of university users’ information needs.

Data were taken from the output generated by the CD-NETTM system between 1995 and mid-1997, when the university replaced its NOVELLTM LAN configuration with the Windows NTTM configuration and the INFOWARETM system, whose integrated CD-ROM database management module is the software presently used at the university for networked electronic information sources management. The data obtained from these two systems was processed using EXCELTM files.

The data were analysed using unidimensional indicators obtained from the resources available at the Carlos III University Library Database Access Service. The results were applied to determine yearly trends and draw conclusions on the grounds of the changes observed over the period 1995-1998, since these underlying dynamics are what inform the modifications and trends taking place among the community of users. The indicators chosen were:

  • Yearly evolution of the number of database visits.
  • Typology of users querying databases.
  • Databases queried.
  • Information consumption (in terms of total Mbytes downloaded from databases).

A descriptive analysis was run on the data obtained. Multivariate analysis techniques, namely simple correspondence and correlation analysis, were likewise used to take simultaneous account of different variables and perceive any interrelations among them. Functionalities associated with these techniques were used to map the variables.

For the intents and purposes of the multivariate technique, users were classified into four different groups i.e., Professors, Administration & services (PAS), Library and Students, each one of which was studied separately.

The computer tools used to process the data were integrated in accordance with their respective capabilities. EXCELTM 7.0 for Windows supports table processing, the application of different statistical functions and the formulation of (bi-dimensional) graphs. XLStatTM 3.5 software was used in conjunction with EXCELTM to expand the statistical tools supported by the spread sheet by incorporating functions that develop a number of multivariate analysis techniques. More complex analyses were performed using the Statistical Package for Social Sciences (SPSSTM 7.5 for Window).


Table 1 shows the total number of times network databases were accessed in each of the years analysed; such values include data for both the CD-NETTM system, in place through February-April 1997 and those recorded in the INFOWARETM system between that date and the month of December 1998.


Total DB visits

% of total













Table 1. No. of times all databases were accessed in the period 1995-1998.

As the table shows, the number of database queries grew constantly over the period surveyed, with 1998 accounting for 43.92% of the total. This trend is the more obvious in the time series given in Figure 1, in which the sharp and constant upward trend in access rates can be clearly perceived from the slope of the curve. Indeed, the final figure shown for the end of the period was 510.1% higher than the total recorded in 1995

 Growth in number of times networked databases were accessed per year, 1995-1998.

Fig. 1. Growth in number of times networked databases were accessed per year, 1995-1998.

The substantial increase in the number of database visits beginning in 1996 was due to simpler access to and greater availability of CD-ROM databases for the entire community of university users. Moreover, the implementation of the new INFOWARETM system in 1997 and the ensuing increase in the number of databases and simplified access were instrumental in the evolution and consolidation of query rate trends, as can be seen in the values for both 1997 and 1998.

An evaluation of group behaviour can be drawn from the analysis of the number of database visits by each user category. Figure 2 shows how the use of this type of resource grew among most users year after year, such as students, for instance, who made over 50% of all the database queries in the period studied. However, comparing the two most active groups (students and professors), the number of queries clearly grew much faster between 1996 and 1997 among the latter, indicating the enormous benefit the academic staff derived from the convenience of being able to access the CD-ROM network from their own offices, in lieu of physical presence in the library to access databases (as was formerly the case under the CD-NETTM system).

Another striking feature is the greater use of databases by library personnel in 1998, with the number of queries up 127.7% over the preceding year. Administration and services personnel made much more moderate use of the bases, with a slight dip (down 1.6%) in 1998 compared to 1997.

Trends in number of database visits, by user category

Fig. 2. Trends in number of database visits, by user category.

As far as the use of each different database is concerned, the figures on the number of times they were accessed over the period 1995-1998 are given in Table 2. INSPEC, one of the most outstanding examples, was visited much more frequently in 1997 and 1998, eloquent proof of the intense use that school of engineering users made of this resource. The number of queries recorded for the IBERLEX database, accessed essentially by Law Department professors and students, also grew at a brisk pace. Another development to be highlighted was that there was only a slight rise in the number of visits to the ABI/INFORM database, of particular interest to social science users. This may be partially explained by the introduction of the Business Sources Elite facility in late 1997, a database similar in profile to the ABI/INFORM, but which features full texts in its records, as opposed to bibliographic information only.

Another widely and rapidly accepted database is ACAD, containing multi-disciplinary, full-text information, which had surprisingly high access rates in 1998, for a base only recently added to the network.

















































































Table 2. Network databases with the highest access rates, 1995-19981.

Multivariate analysis of the figures provided a fuller understanding of database use patterns among professors in the different university departments. Figure 3 shows the correlation between the number of times bases were accessed and the number of professors per department. A Pearson correlation coefficient of 0.84 was found, showing the two variables to be highly inter-dependent, i.e., the number of times databases were accessed was closely related to the number of professors in the department.

Teaching staff (X axis) vs. number of times databases were accessed (Y)

Fig. 3. Teaching staff (X axis) vs. number of times databases were accessed (Y).

The graph shows that in various departments, such as Physics (FIS), Business Administration (EMP), Public Law (DPU) and Economics (ECO) the number of queries per professor was higher than the overall rate. Academic staff in other departments, by contrast, such as Statistics (EST), Humanities (HUM), Computer Science (INF) and Private Law (DPR) queried databases less often than the global rate.

Another important parameter studied was the amount of information professors obtained from the databases they queried per visit. Figure 4 shows the positioning of each department on the basis of the results obtained when these two variables were compared.


 Amount of information (Mbytes) obtained by departments (Y) vs number of times databases were accessed (X axis)

Fig. 4.  Amount of information (Mbytes) obtained by departments (Y) vs number of times databases were accessed (X axis).

As the graph shows, a number of departments is over the trend line, namely: Private Law (DPR), Business Administration (EMP) and Public Law (DPU). This may be due to the fact that most of the databases used by these departments contain full text records. Smaller amounts of information, in terms of Mbytes, were downloaded per visit to the different databases by Mathematics (MA), Biblioteconomics (BIB) and Engineering (ING) professors, in all likelihood because the bases most used by such departments contain bibliographic records only.

A tool that provides useful complementary information to more accurately determine the databases used by each department is the three-dimensional illustration obtained from correspondence analysis, as shown in figure 5. This map was obtained by relating the number of visits by professors in each university department to the databases queried. The closer the name of a department is to a given base, the greater use it made of the base in question. It will be noted, for instance, that the Biblioteconomics (BIB) department followed a substantially different pattern than the rest of the departments in respect of the use of the LISA and BIBESP databases. The Private Law (DPR) Department, in turn, differed considerably from Public Law (DPU), highlighting the differences in departmental database preferences, with the latter using primarily DERJUR, MAPLEX and IBERLUE. Another interesting development was the proximity between the Economics (ECO) and Business Administration (EMP) Departments, both of which are very close to a number of shared bases, such as ECONLIT and ABI/INFORM.

The Mathematics (MAT), Engineering (ING), Computer Science (INF) and Physics (FIS) Departments, in turn, are located close to one another because the database they all use most commonly is INSPEC.

This map was obtained by relating the number of visits by professors in each university department to the databases queried

Figure 5. This map was obtained by relating the number of visits by professors in each university department to the databases queried


The Carlos III University, Madrid, Library Database Access Service has made a wealth of information available to users, in over 100 compact disks containing thirty-some databases that provide extensive national and international coverage. The use of these databases grew considerably over the four years studied.

The rise in the number of database visits was the result of readier access to CD-ROM database hardware and software across the entire community of university users. Indeed, the replacement of the CD-NETTM system with INFOWARETM in 1997 as the system for accessing CD-ROM databases made it possible to use these resources from any computer on either of the university campuses, as well as to expand the number of databases and compact disks in the main server, all of which had a notable impact on the number of queries.

The use of CD-ROM databases has increased not only as a source of teaching and research information, but for students and internal library use as well, as is borne out by the high access rates among these users.

Professors in the Physics, Business Administration, Public Law and Economics Departments visited databases most often, whereas Statistics, Humanities, Computer Science and Private Law professors queried bases least frequently.

Certain departments also followed very clear patterns of specialisation by subjects, querying certain databases almost exclusively. Such was the case of the Engineering Department with respect to INSPEC or Biblioteconomics with respect to LISA.

Surveys of this nature are useful tools for internal university library management. In this regard, conducting such studies on a routine basis is recommendable to contribute to the enhancement of the quality of information services in university environments.


1. ARRUDA RAMALHO, F. La biblioteca universitaria ante las nuevas tecnologías de la información: estudio comparativo entre Espańa y Brasil. Documentación de las Ciencias de la Información, 94, nş 17, p. 11-26.

2. ARTILES VISBAL, S. y GONZÁLEZ GARCÍA, F. La biblioteca universitaria como factor de cambio en las instituciones de educación superior: el caso de la Universidad de Camagüey en Cuba. Ciencias de la Información, 97, vol. 28, nş 1, p. 11-15.

3. KEENAN, S. CD-ROMs - a trasient medium? Online & CD-ROM Review, 98, vol. 22, nş 2, p. 103-105.

4. LEACH, K. A constant demand for new services: a survey of CDROM acquisitions in UK academic institutions, 1996 [Página Web]. Revisada: 98. Disponible en:

5. MCCARTHY, C. A.; KRAUSSE, S. C. y LITTLE, A. A. Expectations and effectiveness using CD-ROMs: What do patrons want and how satisfied are they? College & Research Libraries, 97, p. 128-142.

6. MOSCOSO, M.; NOGALES, T. J. y CARIDAD, M. The use of online CD-ROM databases in spanish university centralised bibliographic services. Online & CD-ROM Review, 95, vol. 19, nş 1, p. 3-12.

7. RAITT, D. y CHEN, C. Use of CD-ROMs in spanish libraries. Revista Espańola de Documentación Científica, 90, vol. 13, nş 2, p. 710-716.

8. SANZ CASADO, E. y MARTÍN MORENO, C. Técnicas bibliométricas aplicadas a los estudios de usuarios. Revista General de Información y Documentación., 97, vol. 7, nş 2, p. 41-68.

9. WILSON, T. D. Redesigning the university library in the digital age. Journal of Documentation, 98, vol. 54, nş 1, p. 15-27.

Elías –

Los números que van entre corchetes son notas mías, a saber:

[1]: no está c laro a qué se refiere este porcentaje – al número total de consultas, al total de consultas a bases de datos, al número de bibliotecas...

[2]: no hay figura 5.

[3]: no estoy segura si se dice así eninglés lo de la "ingeniería automática"

varia 24.jpg

Zapytaj bibliotekarza

Telewizja UŚ

Więc Jestem. Studencki serwis rozwoju