Scientific Publishing in West Africa: A Comparison of Benin with Senegal and Ghana Eustache Mêgnigbêto eustachem@gmail.com Bureau d études et de recherches en science de l information (BERSI), 09 BP 477 Saint Michel, Cotonou (Bénin) Abstract We compared scientific indicators related to Benin, Senegal and Ghana. We collected data from Web of Science and used indicators like the yearly productivity, the language of publication, the type of publication, the citable documents, the publication fields, and the main international partners as well as the percentage of papers in collaboration. Results showed that Benin productivity is the lowest one; Ghana and Senegal competed over the period; depending on the type of documents under consideration, the positions of the three countries vary. Citable documents had an increasing trend for all the countries. There is less cooperation between African countries and Benin, Senegal and Ghana; colonial ties count much in their international partnership. Cooperation among the three countries is negligible. Introduction Research activities are a sustained economic and social development condition; therefore, policies are conceived and applied to improve research, to enlarge its frontiers and to make it contribute to the development process. Very often, research activities are assessed either at a geographical area, at an institutional, or at a scientist level. One specific objective is to compare an actor's (scientist, group of scientists, institution, city or country) profile with those of other actors from the same region or the same economic, historical and social conditions. Mahbuba and Rousseau (2010) compared selected bibliometric indicators for Bangladesh, Pakistan and Sri Lanka to those of India. They found a high correlation between GDP, GDP per capita and absolute data publication. Toivanen and Ponomariov (2011) studied the whole Africa s research collaboration pattern both in term of countries and cities; he distinguished three separate and weakly linked scientific regions in the continent which are Northern, Western and Southern- Eastern. Tijssen (2011) studied Africa s contribution to the worldwide research literature and computed indicators for comparison purpose. Pouris (2010) dealt with the scientific production in the Southern Africa Development Community (SADC); he made comparison among the community s countries and found that South Africa is responsible for 79% of the regional production. Pouris and Pouris (2009) reported the results of an effort to calculate scientometric indicators research publications and patents- for the African Continent, compared them with those of other regions and discussed their implications. Boshoff (2009) examined collaboration ties in Central Africa's scientific output; he revealed that colonial and cultural ties play a major role in international collaboration. Boshoff (2010) dealt with collaboration among the SADC countries as well as with the rest of Africa; he concluded that South Africa acted as the political North, i.e., it dominated the regional scientific research. Hence, none of the above mentioned papers studied specifically scientific output in West Africa nor compared two or more countries from this region.
On the basis of data from Web of Science, Gaillard (2010) ranked African countries by the volume of their scientific publication; Nigeria was 6 th, Ghana and Senegal 12 th and 13 th respectively, and Benin 25 th ; at West Africa region level, Nigeria was ranked 1 st, Ghana 2 nd, Senegal 3 rd, and Benin 8 th. SCImago Journal & Country Rank (SJR), with data from Scopus over the period 1996 to 2010, ranked at the continental level Nigeria 3 rd, Ghana 12 th, Senegal 14 th and Benin 22 th ; at West Africa region level, Nigeria is ranked 1 st, Ghana 2 nd, Senegal 3 rd and Benin 6 th. Table 1. More productive countries than Benin in West Africa. Rank Web of Science Scopus 1 Nigeria Nigeria 2 Ghana Ghana 3 Senegal Senegal 4 Burkina Faso Cote d'ivoire 5 Cote d'ivoire Burkina Faso 6 Gambia Benin 7 Mali 8 Benin Source: Gaillard (2010) and SJR. The African Union divided the Africa continent into five cultural, historical and geographical parts: the Northern, the Western, the Central, the Eastern and the Southern. 1 The West African region groups together 15 countries. Eight of them are former French colonies namely Benin, Burkina Faso, Cote d Ivoire, Guinea, Mali, Niger, Senegal and Togo; Guinea-Bissau and Cap Verde are former Portuguese colonies; and Nigeria, Ghana, Liberia, Sierra Leone, and the Gambia are former British colonies. Hence, three international language communities are distinguished in the region that are respectively French speaking, Portuguese speaking and English speaking. Senegal has been the Headquarter of the French colonial administration in West Africa; as such, it was the central point for transport and commerce, and hosted well-equipped hospitals and educational infrastructures. Pupils from other parts of the empire had to travel to Senegal to pursue school once they had reached a certain level in the education system. Therefore, the country received the first higher education institutions long time before independence; we assume that scientific institutions and skills should take root more in this country than in any other French speaking country in the region. The original idea that led to this paper was to study Benin s scientific literature and the contribution of this country to the global science system. One specific objective is to compare the country to the bigger science producers in the region. Our research question is: How does Benin compare with Senegal and Ghana, the two bigger science producers the former French speaking, and the latter English speaking in West Africa? 1 Resolution CM/Res.464 (XXVI) adopted by the African Unity Organisation (now African Union) Council of Ministers meeting in its twenty sixth ordinary session in Addis-Ababa (Ethiopia) from 23 February to 1 March 1976. The countries list was updated in 2004. 590
Methods and data Data source For bibliometric studies, data are often collected from bibliographic databases (Meho & Yang, 2007; Bal-Ilan, 2009); nowadays, there are three suitable international databases namely Scopus, Web of Science and Google Scholar. Web of Science is the largest and the eldest; however, Scopus and Google Scholar are its potential competitors (Meho & Yang, 2007). In this study we used Web of Science. Benin is the actual name of the former Dahomey; Senegal experienced between 1982 and 1989 an integration process with the Gambia under the name of Senegambia. Web of Science was searched in December 2011 and all the publications co-authored by at least one scientist from Benin, Senegal or Ghana were extracted. The databases searched were Science Citation Index Expanded (SCI-EXPANDED), Social Sciences Citation Index (SSCI) and Arts & Humanities Citation Index (A&HCI). All languages and all document types were selected. The advanced search option was used and searches were run separately for each country (cu=benin for Benin, cu=ghana for Ghana, cu=senegal for Senegal). Benin and Senegal data were complemented/completed by the results of two others searches, respectively cu=dahomey and cu=senegambia. Particularly, the results from Senegambia were browsed and records containing at least one Senegal city in Affiliation or Corresponding author fields were extracted and added to Senegal data. Data treatment Records were imported into a CDS/ISIS 2 database thanks to a program written in Pascal CDS/ISIS. 3 Another Pascal CDS/ISIS program was used to extract country names from affiliations data and to add them into the inverted file for research purpose. Thanks to the CDS/ISIS formatting language 4 and search function, some data were printed into a file and analyzed with statistical software or computed directly. The records cover the period from 1966 to 2011; over this period of time, many political changes occurred that had resulted in the renaming of countries. For example, Upper Volta became Burkina Faso, Dahomey became Benin, the Federal Republic of Germany and the Democratic Republic of Germany merged, Senegal and the Gambia experienced an integration process under the name of Senegambia, Congo Kinshasa became Zaire and changed again to Democratic Republic of Congo, etc. The record edition function of CDS/ISIS was used to standardize countries names affected by changes over the period in order to compute statistics on international collaboration. 2 CDS-ISIS is a text database management software developed and distributed by UNESCO. (http://www.unesco.org/isis). 3 CDS/ISIS provides a programming language designed to develop CDS/ISIS applications requiring functions which are not readily available in the standard package (UNESCO, 1989a). This programming language enables users to extend functions of the standard package, to make it more robust and in order to meet users specific needs. 4 The formatting language allows you to define precise formatting requirement for data base records. Through this language, you may select one or more specific data elements in the order you want and optionally insert constant text of your choice, e. g. to label some or all the fields, as well as specify vertical or horizontal spacing requirements ( ). The formatting language is therefore the core of many operations, and an efficient use of CDS/ISIS requires a thorough knowledge of this techniques (UNESCO, 1989b). 591
Although no date limits were given while searching WoS, the analyses below are related to the period up to 2010. The total number of bibliographical references considered in this paper is 15,184 of which 1,906 for Benin, 6,580 for Senegal and 6,698 for Ghana. Selected indicators We intended to compare Benin, Senegal and Ghana s scientific output. We used scientometric methods and techniques for this purpose. The comparison was done throughout indicators like the yearly productivity, the average number of authors per publication, the language and type of publications, the yearly number of citable documents and the international collaboration pattern. Results Scientific productivity Data downloaded from WoS revealed that Benin publications cover the period from 1977 to 2011. The scientific productivity curve (1973-2010, Figure 1) shows an increasing trend more pronounced starting from 2003. We fitted the part of the curve starting from 1991 to 2010 (a 20-0.7432 years period) with the power law. The resulting equation is y = 14.719t (R 2 = 0.77), where y is the number of publications and t the period of time (t = 1 in 1991). Senegal publications cover the period from 1966 to 2011. The scientific productivity curve (1966-2010, Figure 1) shows a regular increasing trend until 2000, followed by a decreasing trend starting from 2001. The equation of the power law that best fits Senegal production over the period 1991-2010 is 0.2313 y = 139.07t with t=1 in 1991 and R 2 =0.45. Ghana publications also cover the period from 1966 to 2011. The scientific productivity curve (1966-2010, Figure 1) shows an irregular form: some periods of increasing trends, and some others periods of decreasing trend. The part of the 0.4544 curve starting from 1991 is best fitted by a power law of which the equation is y = 91.36t where y is the yearly number of publications and t the period of time (R 2 = 0.63, t = 1 in 1991). Benin's scientific output has always been lower than that of Ghana and Senegal. From 1966 to 1974, Ghana and Senegal had about the same yearly output; but over the 20-years period that followed, Ghana scientific production was lower. Ghana took the advantage for only four years before catching up Senegal and taking the first place starting from 2005 when an exponential growth is registered. In the opposite, Senegal started a decreasing trend. Over all the period, Benin production was the lowest. The proportion of Ghana production to Benin varies from 2 to 30. 592
Yearly productivity of Benin, Senegal and Ghana Number of publications 500 450 400 350 300 250 Ghana Senegal Benin 200 150 100 50 0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Years Figure 1. Joint publication curves of Benin, Senegal and Ghana. Table 1. Quartiles of Benin, Senegal and Ghana s scientific output. Statistic Benin Senegal Ghana Minimum 1973 1966 1966 1 st quartile 1999 1988 1993 2 nd quartile 2005 1998 2000 3 rd quartile 2008 2004 2007 Maximum 2010 2010 2010 The quartiles of the scientific output curves for the three countries are summarized in Table 1. Benin has published between 1973 and 1999 (a 26-years period) as much as it had over 1999 to 2005 (a 6-years period), as much as over 2005 to 2008 (a 3-years period), and as much as over 2008 to 2010 (a 2-years period). Benin has published over 2005 to 2010 (a 5-years period) as many papers as it had published over 1973 to 2005 (a 32-years period). Senegal produced 25% of its total publication over the period 1966-1988 (a 22-years period), and also 25% of its publications over 1988-1998 a 10-years period; the two remaining 25% were produced over a 7- years period (respectively 1998-2004 and 2004-2010) (cf. Table 1), this suggests a stagnation or a decrease in the Senegalese research productivity. From 1966 to 1993 (23 years), Ghana has produced 25% of its total number of publications; the following 25% were produced between 1993 and 2000 (a 7-year period); again one fourth was produced over the next 7-years period (2000-2007) followed the last quarter of the production over a 3-years period. Languages of publication Ghana is an English speaking country; Benin and Senegal are French speaking. Almost all Ghanaian production is published in English (99.58%); others language are French (0.25%), German (0.11%) and Spanish (0.04%). Ghanaian papers in French were written by authors from the Alliance Française in Accra and are published in journals in France. Benin published in English (85.99%), in French (13.58%) and other European languages (0.43%); Senegal published less in English (76.07%) and more in French (23.80%) than Benin (Figure 2). 593
Percentage of publications 0 20 40 60 80 100 Languages of publication in Benin Senegal and Ghana Benin Ghana Senegal English French Others languages Figure 2. Languages of publication in Benin, Senegal and Ghana scientific literature (1966-2010). Types of publication We considered hereafter only citable documents. For each of the three countries the large majority of the publications are article type. Benin produced a higher percentage of Article type papers than Senegal that in turn produced relatively more than Ghana. However, Ghana takes the first place in terms of number of Meeting abstract, followed by Benin then by Senegal. Senegal however, is in first place as Articles, Proceedings papers type is concerned (Figure 3). Percentage of publications 0 20 40 60 80 100 Types of publication in Benin, Senegal and Ghana scie Benin Ghana Senegal Article Article, Proceedings Paper Letter Meeting Abstract Note Review Type of documents Figure 3. Types of publication in Benin, Senegal and Ghana s scientific literature. 594
Co-authorship Benin s papers have been co-authored by a total of 9,870 scientists, Senegal s by 32,331 scientists and Ghana s by 28,773 scientists. In Benin s literature, the number of authors per paper varies from 1 to 48; Senegal s papers have 1 to 115 coauthors and Ghana s have up to 160 coauthors. The number of papers per (co)author is 0.19 for Benin, 0.21 for Senegal and 0.23 for Ghana. These indicators are summarized in Table 2. Citable documents In Figure 4, the positions of the countries are clear: Benin is in the lowest position; Ghana and Senegal competed until 2003 where Ghana took the first place. Citable documents in Beninese, Senegalese and Ghan Number of publications 500 450 400 350 300 250 Ghana Senegal Benin 200 150 100 50 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Years Figure 4. Citable documents in Benin, Senegal and Ghana s scientific publication (1996-2010). Production per field Each of the three countries published paper in 125 to 189 ISI fields, Benin in 125, Senegal in 169 and Ghana in 189. Agriculture is ranked the first field in case of Benin, Public, environment and occupational health in case of Senegal and Tropical medicine in case of Ghana. Agriculture is in the top five fields for Benin and Senegal, and rankes 6 th in case of Ghana. For the three countries, Tropical medicine, and Public, environment & occupational health appear in the top 5 fields list (Table 4). International collaboration From the CDS/ISIS database and for each of the three countries, we searched for all records with at least one co-authors not affiliated to any institution in the country under consideration. The repartition of the result per year and per country is shown in Figure 5. It appears that Benin has the highest percentage of collaboration, followed by Senegal and Ghana over the period from 2000 on. Table 5 presents the top ten collaborating countries for the three countries. France was ranked the first collaborator country for Benin and Senegal, while USA was for Ghana. Belgium was ranked second for Benin, USA for Senegal and UK for Ghana. Nigeria appeared to be the 1 st African collaborator country for Benin and Ghana (7 th in each case). The first African contributor 595
for Senegal was Cameroun, ranked 5 th with 4.50%, followed by Ghana (4.9%). Senegal was ranked 11 th in Benin partner countries, and Ghana 15 th ; Benin was ranked 18 th in Senegal s partner countries and Ghana 24 th. Neither Benin nor Senegal appeared in the top ten partner countries of Ghana. The top 20 partner countries list 10 African countries for Benin, 10 for Senegal and 6 for Ghana. The partner countries Benin, Senegal and Ghana share (from the top ten) are USA, UK and Germany. We can conclude that for each of the three countries we study, the main research partners are non-african countries; the former colonial powers are still prominent partner (ranked first case of Benin and Senegal and second in case of Ghana). Part of international collaboration in Benin, Senegal an 100 Percentages of publications 90 80 70 60 50 40 Ghana Senegal Benin 30 20 10 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Years Figure 5. Percentage of international collaboration in Benin, Senegal and Ghana s scientific publications (1996-2010). Table 2. Synthesis of indicators. Indicators Benin Senegal Ghana Number of papers 1,906 6,580 6,698 Period 1973-2010 1966-2010 1966-2010 Period of time (in years) 38 45 45 Average number of paper per year 50.16 146.22 148.84 Number of co-authorships 9,870 32,331 28,773 Number of partner countries 106 146 154 Number of papers in collaboration 1,528 4,318 3,612 Percentage of papers in collaboration. 80.17 65.62 53.93 Percentage of papers in English 85.99 76.07 99.58 Number of field (ISI) 125 168 189 596
Discussion and conclusion Benin's yearly scientific research output volume is lower than that of Ghana and Senegal. Indeed, according to Table 3, the country has the lowest number of universities, the lowest rate of adult literacy, the lowest HDI and the lowest GDP per capita. Furthermore, the country is ranked by the UNDP HDI the poorest among the three. These data explain the position of Benin in comparison with Senegal and Ghana. Data also confirms Mahbuba and Rousseau's (2010) conclusion that there is a positive relation between HDI, GDP and research development. Table 3. Human development indicators. Indicators Benin Senegal Ghana HDI (Human Development Index, 2010) 0.427 0.459 0.541 Adult literacy (2007) 41.7 61.8 66.6 Number of Universities (Public / Private) 9 (2/7) 10 (5/5) 24 (17/7) GDP per capita (USD, 2010) 750 1042 1280 Rank (by HDI - out of 187 countries) 165 155 35 Over the period from 2001 to 2010, an increasing trend was noticed in Benin's and Ghana's production; however, Senegal's curve showed a decrease. This decline may be explained by: i) a real decrease in the Senegal scientific output, ii) a diversification in journals Senegalese scientists published in beyond those indexed by WoS, iii) a diminution of the number of publications Senegalese published in Sciences largely covered by WoS than Social Sciences and Humanities. However, a search undertaken in Scopus in March 2012 invalidated the decreasing trend. Thomson Reuters increased the coverage of the journals in the second part of the 2000s and this may contribute to the increasing trend in the number of publications. Benin published more in English than Senegal and has the highest percentage of international collaboration. This illustrates Boshoff's (2009) remark that the less productive a country is in terms of scientific output, the greater is its dependence on international collaboration. Thus, the three countries depend much on international collaboration. European and American countries are their first partners. Cooperation among them or with other (West) African countries is very weak. Indeed, research needs heavy and sustained investment; the three countries are categorized in the developing countries group and hence have not much means or have other priorities like basic infrastructure, education and health. Therefore, research is not an autonomous activity but relies on external cooperation. The former colonial powers contribute much to the literature of the three countries (France: 34.11% in case of Benin and 52.80% in case of Senegal; United Kingdom: 23.62% in case of Ghana). Pouris (2010) states that the colonial past and cultural ties play an important role in the structure of collaborative linkages. Boshoff (2009) remarks are also confirmed: African research collaboration appears to be driven by European and American investments and collaboration between African and other developing counties remained a relatively weak phenomenon. Even though cooperation among Benin, Ghana and Senegal is negligible, it is interesting to notice that Benin shared 1.6% of Senegal papers and 1.5% of Ghana's; Senegal shared 4.6% of Benin papers and 1.25% in Ghana output; in turn, Ghana shared 4% of Benin papers and 1.3% of Senegalese. Ghana and Senegal contribute much more to Benin scientific output than Benin to theirs. 597
Among the three countries, Ghana is ranked the first by the volume of publications, followed by Senegal and Benin. All the three countries published mainly in English. Ghana, which is an English speaking country, has the highest rate of papers in English followed by Benin and then Senegal. The main type of publication encountered is the article, irrespective to the countries. Indeed, this form of publication represents nowadays the final stage of research. However, Benin has published relatively more articles than Senegal that, in turn, has published more than Ghana. The mains fields of research vary with the country under consideration. As citable documents are concerned, Senegal comes first, Ghana second and Benin third. It appears that the three countries have a high collaborative percentage and their main partner countries are in Europe and America. Benin has the highest collaboration percentage and Ghana the lowest; Senegal lies between the two countries. Overall, whatever criteria is taken, Benin is ranked behind Ghana and Senegal. 598
Table 4. Top ten ISI fields in Benin, Senegal and Ghana production. Benin Ghana Senegal Rank Fields % Fields % Fields % 1 Agriculture 14.01 Public, environmental & occupational health 14.54 Tropical medicine 12.00 2 Entomology 13.53 Tropical medicine 11.78 Public, environmental & occupational health 11.94 3 Tropical medicine 10.78 Agriculture 9.58 Infectious diseases 11.00 4 Public, environmental & occupational health 9.67 Environmental sciences & ecology 6.45 Immunology 8.61 5 Plant sciences 7.61 General & internal medicine 4.79 Microbiology 7.76 6 Physics 6.18 Parasitology 4.55 Agriculture 7.45 7 Infectious diseases 5.55 Plant sciences 4.18 Parasitology 5.92 8 Biotechnology & applied microbiology 5.44 Infectious diseases 4.15 Virology 5.24 9 Parasitology 4.92 Immunology 3.99 General & internal medicine 4.95 10 Environmental sciences & ecology 4.70 Chemistry 3.72 Environmental sciences & ecology 4.66 Table 5. Top ten collaborating countries in Benin, Senegal and Ghana scientific literature. Benin Senegal Ghana Rank Countries # papers % (total) % (papers in collaboration) Countries # papers % (total) % (papers in collaboration Countries # papers % (total) % (papers in collaboration 1 France 498 26.32 34.11 France 2.281 34.33 52.80 USA 1107 16.53 30.65 2 Belgium 219 11.58 15.00 USA 852 12.82 19.72 United Kingdom 853 12.74 23.62 3 USA 145 7.66 9.93 United Kingdom 560 8.43 12.96 Germany 357 5.33 9.88 4 United Kingdom 136 7.19 9.32 Belgium 344 5.17 7.95 Netherlands 290 4.33 8.03 5 Netherlands 115 6.08 7.88 Cameroon 195 2.93 4.50 Canada 214 3.19 5.92 6 Germany 114 6.03 7.81 Switzerland 181 2.72 4.19 Denmark 202 3.02 5.59 7 Nigeria 101 5.34 6.92 Gambia 175 2.63 4.05 Nigeria 178 2.66 4.93 8 Cameroon 95 5.02 6.51 Burkina Faso 171 2.58 3.96 Japan 177 2.64 4.90 9 Cote d Ivoire 73 3.86 5.00 Germany 138 2.08 3.20 Switzerland 173 2.58 4.79 10 Switzerland 73 3.86 5.00 Italy 138 2.08 3.20 Kenya 136 2.03 3.77 599
References Bar-Ilan, J. (2009). A Closer Look at the Sources of Informetric Research. In B. Larsen, J. Leta (Eds.), Proceedings of ISSI 2009 the 12th International Conference of the International Society for Scientometrics and Informetrics, vol. 1 (pp. 1-9). Rio de Janeiro, Brazil, 14-17 July 2009, BIREME/PAHO/WHO and Federal University of Rio de Janeiro. Boshoff, N. (2009). Neo-Colonialism and Research Collaboration in Central Africa. Scientometrics, 81 (2), 416-434. Boshoff, N. (2010). South-South Research Collaboration of Countries in the Southern African Development Community (SADC). Scientometrics, 84 (2), 481-503. Gaillard, J. (2010). Le système national de recherche scientifique et technique au Bénin : un état des lieux. Paris: Unesco. Retrieved November 30, 2011 from: http://horizon.documentation.ird.fr/exl-doc/pleins_textes/divers11-02/010051127.pdf Mahbuba, D. & Rousseau, R. (2010). Scientific Research in the Indian Subcontinent: Selected Trends ad Indicators 1973-2007 Comparing Bangladesh, Pakistan and Sri Lanka with India, the Local Giant. Scientometrics, 84 (2), 403-420. Meho, L.I. & Yang, K. (2007). Impact of Data Sources on Citation Counts and Rankings of LIS Faculty: WoS versus Scopus and Google Scholar. Journal of the American Society for Information Science and Technology, 58 (13), 2105-2125. Pouris, A. (2010). A Scientometric Assessment of the Southern Africa Development Community: Science in the Tip of Africa. Scientometrics, 85 (1), 145-154. Pouris A. & Pouris, A. (2009). The State of Science and Technology in Africa (2000-2004): A Scientometric Assessment. Scientometrics, 79 (2), 297-309. Stefaniak, B. (2001). International Co-Operation in Science and Social Sciences as Reflected in multinational papers indexed in SCI and SSCI. Scientometrics, 52 (2), 193-210. SCImago. (2007). SJR SCImago Journal & Country Rank. Retrieved January 30, 2012 from: http://www.scimagojr.com Toivanen, H. & Ponomariov, B. (2011). African Regional Innovations Systems: Bibliometric Analysis of Research Collaboration Patterns 2005-2009. Scientometrics, 88 (2), 471-493. Tijssen, R.J.W. (2007). Africa s Contribution to the Worldwide Research Literature: New Analytical Perspectives, Trends and Performance Indicators. Scientometrics, 71 (2), 303-327. UNESCO. (1989a). Mini-micro CDS/ISIS: Pascal CDS/ISIS. Paris, Unesco. UNESCO. (1989b). Mini-micro CDS/ISIS: Pascal CDS/ISIS. Paris, Unesco. 600