A bibliometric study of the publication patterns of scientists in South Africa 1992-96, with particular reference to status and funding

Daisy Jacobs
Dept of chemistry, University of Transkei,
Umtata, Republic of South Africa

This paper is a bibliometric study of the publication patterns of a selected group of academic and research scientists of ten universities of South Africa for a period of five years, 1992-96. The subject fields surveyed are Physics, Chemistry, Plant and animal Sciences, and Microbiology/Biochemistry. These ten universities vary considerably with respect to standards of education, quantity of research and publication and overall progress. The general purpose of the study is 1) to determine whether academic status and prestige have any impact on the level of productivity, 2) to study the productivity within different areas of science, 3) to investigate whether or not the level of funding and/or the prospects of getting funded has any influence on the level of productivity in each area of science and their pattern of publication. The study collected two sets of data through a scientometric analysis of science citation index and a questionnaire. The study demonstrates that there is a direct relationship between status and publication productivity. The study further shows that there are significant differences in productivity between areas of sciences but that there is no direct relationship between institutional funding and productivity.

Printer-friendly version

Introduction

Studies of publication patterns, also known as bibliometric or quantitative studies are useful indicators of scientific productivity, trends, emphasis of research in various disciplines, and of researchers’ preferences for publication outputs. Results of such studies may be very useful in decision making in research administration and planning, in collection development and use in libraries. These results further enable policy makers in different organisations and funding agencies such as Foundation for Research and development (FRD), and Centre for scientific Information and Research (CSIR) to evaluate their decisions on the awarding of grants to individuals and institutions.

Scientific productivity has been linked to various factors, such as age and subject specialisation, and economic indicators, such as government expenditure on civil research and development. According to Budd and Seavey (1996) prestige and productivity go together and can be summed up as a "scientist’s lifework". A study conducted by Pouris (1989) showed that the criteria used for ranking academic institutions in South Africa range from opinion surveys to measures of research productivity as proxied by publication counts in a sample of reputable journals. According to Price and Beaver (1966), prestige seems to be one of the driving forces that encourages scientists to publish profusely. There is a relationship between the importance of the scientist and the logarithm of the number of papers he has published during his life. Prestige is a driving force that prompts authors to publish in foreign journals. Many of the bibliometric studies which have been conducted in other countries, provide a theoretical context for this research. Lancaster (1982), argues that many scientists in developing countries prefer to publish in foreign journals rather than in their native journals for the sake of prestige and recognition and half of the papers of Indian scientists are published in the United States. Garfield (1983) agreed with Lancaster when he said that only 17 percent of Latin American research articles were published in local languages.

According to Alabi (1989) there has been tremendous growth in scientific activities in most of the third world countries. Saracevic (1977) also expressed similar sentiments when he said that the rate of scientific activities have been tripling in most developing countries compared with the doubling tempo in the developed countries. Most scientists in the West are not aware of the research and publication productivity that goes on in the third world countries. The main reason for this could be that scientists from some of these third world countries fail to publish the results of their research in reputed international journals (Jacobs and Ingwersen, 2000). This is also true about South Africa which, although has a lot of scientific research and production, the country having been "closed off" meant little was "known" outside. One of the legacies of the apartheid system was the discrepancy in funding and support for various activities, including research work in science and technology, based on racial grounds, some institutions of higher learning and research institutes were more favoured than others in terms of resources.

The historically black universities were at a disadvantage as they had to cope with many additional problems such as upheavals and high drop-out rates. There was a high failure rate among students due to students spending most of their study time in fighting against the "system", which discouraged investment in these institutions. Most of the students who joined these institutions came from a poor schooling system, resulting inadequate knowledge of subjects such as Mathematics, Science and English.

Most of these institutions were isolated physically, politically and academically, hence it was far more difficult to establish an infrastructure for research, or attract renowned academics to work in these institutions (Jacobs and Ingwersen, 2000: 76). This in turn provided very little opportunity to expose staff and students to other institutions. Consequently, the productivity in some of the science fields and in some institutions lagged behind others. One of the key areas for the enhancement of development in South Africa is improved productivity in Science and Technology.

One of the ways to cover this is through regular independent bibliometric studies. In South Africa few bibliometric studies have been done especially among scientists at both previously advantaged and disadvantaged institutions. Bibliometric studies afford investigators to study the quality and quantity of work done by scientists in the various fields. A recent study by Ovens (1995), on the citation patterns of the scientists of the University of Orange Free State, was noted but represented only a single institution and the work stressed the method of obtaining the relevant information; hence a more comprehensive study was considered relevant. I have therefore undertaken a bibliometric investigation of a five year period 1992-96.

The present study aims to investigate the publication patterns of the South African scientists in the four departments of Physics, Chemistry, Botany and Zoology, and Biochemistry and Microbiology. The special aim of the study is to see whether there is a relationship between academic status and productivity; prestige and productivity; funding and productivity. In order to achieve this, the study further looks at the various channels used for publishing research results such as national and international journals, presenting papers at conferences both national and international, and guiding postgraduate students.

This paper is organised as follows. First, I present the data collection and analysis methods. In this section the survey questionnaire is described as well as the use of statistics. The next section presents the response rate of the survey 1992-96 and the distribution of scientists over academic status. This section includes publication of papers in national and international journals, presentation of papers at conferences and supervision of postgraduate students. The third section contains a detailed analysis and comparison of research funding and the productivity of academics in the various departments. The paper is concluded with a discussion of findings.

Methodology

The Science Citation Index (SCI) online was used to collect the initial data. All the scientists from the departments of Physics, Chemistry, Botany, Zoology, Biochemistry/Microbiology were chosen as the survey population. These scientists belonged to the three provinces of Eastern Cape, Western Cape and KwaZulu-Natal in South Africa. These ten universities are representative of scientists from all the universities of South Africa as they include the two main types of universities, the historically advantaged institutions as well as historically disadvantaged institutions.

The first data collection instrument thus used to elicit data from respondents was a personal letter addressed to scientists individually requesting them to supply a copy of their publication list. A letter explaining the purpose of the research and a self-addressed, stamped envelope accompanied it. Out of three hundred and fifty letters that were sent, only 189 scientists responded. Since the information received was unsatisfactory, a second set of data needed to be collected.

Three hundred and fifty letters together with a structured questionnaire and self addressed and stamped envelopes were sent, requesting the scientists to complete the questionnaire. Out of the three hundred and fifty letters that were sent out, 174 scientists or 50% replied and completed the questionnaire. Only 153 questionnaires were analysable, i.e., 44%.

Collecting data through a questionnaire has both advantages and disadvantages, the latter being very low response rate. According to Babbie (1973), a response rate of 60% is good, but a response rate of 45-50% is adequate for analysing and reporting. The questionnaire was divided into a section on demographic data such as work, area of specialisation, experience, teaching and administration. The second section dealt with the research and publication activities. The third section asked for the opinion associated with research ratings and grading of senior scientists.

Inferential statistics (one-way ANOVA) was used to determine the differences between and among demographic characteristcs of the study population. Pearson’s correlation coefficient was used to test the correspondence between academic status and forms of productivity. Bonferroni comparisons test was used in order to confirm the significant differences between productivity of various categories of scientists.

Distribution of population and respondents according to departments

Table 1 shows the distribution of scientists from the ten universities that have been surveyed. Chemistry and zoology stand out as the two largest departments as far as response rate is concerned, but is not representative of the population of scientists of the various universities.

The objective of the first part of this research is to investigate whether there is a correlation between academic status (qualification such as Ph.D, M.Sc) and the productivity of the academic scientists. Within the context of this study, academic status is defined as a scientist’s position in the academic arena with regards to his/her qualification.

Table 2 below represents the publication productivity of scientists according to his/her status in the department. Out of the 16 items analysed with regard to the relationship between status of the scientists and their productivity, seven items were found to be significant. They were the following:

1. Attendance at overseas conferences
2. Presentation of papers overseas
3. Papers published in the last five years
4. Papers in S.A journals
5. Papers published in overseas journals
6. Evaluation by the FRD
7. Influence of "A" grade scientists on the productivity of other scientists

The literature reviewed showed that status and reward are two very important variables in the productivity of scientists in the academic institutions. A study conducted by Cooper (12) showed that promotion was the driving force behind faculty research and publication, as this upgraded the faculty members in status and pay. Therefore, it is thought that most of the lecturers reviewed in this study attend conferences and presented papers in the hope of receiving promotion, while professors publish profusely in order to maintain their status amongst their peers in the same field. In this study, those scientists with a higher status are referred to as those with doctoral and post-doctoral qualifications, whereas those without such qualifications are referred to as being at a lower status. A similar result is obtained with regard to presentation of papers at overseas conferences.