Comparing Research outputs of Nigeria Federal Universities Based on the Scopus Database Bashir S. Galadanci Sanah A. Muaz Maryam I. Mukhtar Bayero University Kano Bayero University Kano Bayero University Kano Software Engineering Department Software Engineering Department Software Engineering Department +2348032032211 +2348034143099 +2348034574409 bashirgaladanci@yahoo.com samaaz.cs@buk.edu.ng mimukhtar.se@buk.edu.ng ABSTRACT policy making, grant allocation, quality assurance, quality Research outputs are among the key indices in the world ranking assessment, quality improvement, benchmarking and sustainability among other factors [9][27]. For example, in Nigeria, the National of universities. The focus of Nigerian universities, is teaching and Universities Commission (NUC) presidential special scholarship research. This paper compares research outputs of Nigerian scheme for innovation and development adopted the QS World Federal Universities (NFU’s) based on their publication records in Universities ranking to select universities for first class honors the Scopus database. It was found that University of Ibadan (UI) degree scholars, for their postgraduate programs [16]. The QS has the highest total number of research outputs more than other World Universities Rankings of 2013/2014 indicated that no NFU’s. Bayero University Kano has the highest percentage of Nigerian university (NU) is listed among the ranked universities papers with international research collaborations than other NFU’s [24]. although there is a general growth pattern in publications in all of Research is one of the key indices in the ranking of universities the universities selected for the study. Researchers in NFU’s can and NU is focused on teaching and research. In this paper, our use these findings in creating collaborative efforts amongst each focus is on research because it constitutes one of the major criteria other. The National Universities Commission (NUC) can also use for ranking of universities [18]. Research has proven to be the these findings as the basis in developing and implementing a source of important breakthroughs regarding our knowledge of the framework that can significantly increase the visibility of research world and the evaluation of scientific research is considered to be outputs from NFU’s in the Scopus database and hence pave the of great importance. One of the main factors in the assessment of way for the NFU’s to be listed in the Quacquarelli Symonds (QS) research performance is its international influence which World Universities Rankings. represents a measurable aspect of scientific quality [12]. The various works such as that of [17] [19] [23] have compared research outputs of NU, however all of them were using CCS Concepts questionnaires as tools for their analysis. The main issue with • General and reference ➝Document types ➝Surveys and questionnaire analysis is that the accuracy of the responses cannot overviews be verified as database with such publications are not available to the researchers. Others such as [6] [13] have analyzed research Keywords outputs of NU based on some specific Nigerian journals and Research outputs; Patent; Scopus; QS World Universities Ranking; African Journals respectively. The issue with using only these two International collaborations journals is that the scope is narrow as international collaborations and patents were not investigated in their studies. Objective and 1. INTRODUCTION accurate research outputs, including citations, number of authors, Some of the best known university rankings are: The Academic journals, countries, etc. are typically provided by databases [17] Ranking of World Universities; the Quacquarelli Symonds (QS) which are arguably the most reliable sources of data for Ranking; the Times Higher Education World Universities bibliometric studies [10]. Clearly, there is the need for a verifiable Ranking; the SCImago Institution Ranking; the Leiden World and accurate comparison of NU research outputs in terms of Ranking; the Performance Ranking of Scientific Papers for world citations, h-index, research outputs, number of authors and universities; the ranking of the Web of World universities [1] and publication records in journals, international collaborations as well the Web of Universities of the Cybermetrics Ranking [20] [26]. as patent. However, the Times Higher Education Ranking is the only private independent ranking [1]. As such in this paper, we compare the research outputs of Nigeria Federal Universities (NFU’s) based on the Scopus database. The ranking of universities has recently gained interest as a result Scopus is a product of Elsevier and is the world’s largest abstract of the need to acquire tools which can be used for management, and citation database comprising of peer-reviewed research literature. The Scopus database contains more than 18,000 titles of international journals from over 5,000 international publishers. The objectives of the study are fourfold:(1) to identify the number of research outputs published by each of the NFU’s selected for the CoRI’16, Sept 7–9, 2016, Ibadan, Nigeria. study; (2) to compare the productivity of the NFU’s; (3) to determine the extent of international research collaborations of the NFU’s; (4) to identify the most prolific authors in the NFU’s. This analysis will create collaborative efforts between scholars in the Nigerian Universities, may guide the NUC to adapt their 79 publication policy with a view to significantly increasing the Research outputs in the context of NU’s based on the Scopus visibility of NU’s research outputs in the Scopus database, and can database are scarce in the literature. In fact, it can be seen that none be used to encourage international collaboration in NFU’s. of the cited studies have investigated the pattern of research The paper is organized as follows: section 2 present some related outputs of Universities in Nigeria based on the Scopus database. works, section 3 presents a detailed description of the research methodology including the selection criteria of the universities 3. METHODOLOGY The NUC indicated that there are 40 federal universities in Nigeria under study. Section 4 discusses the results of the study and makes [16]. The oldest University is the University of Ibadan (UI) as some concluding remarks and further research is discussed in shown in Table 1. A complete list and details of the universities section 5. can be found through the NUC official website 2. LITERATURE REVIEW (http://www.nuc.edu.ng/). Nigeria is divided into 6 geopolitical There are bibliometrics studies in the literature regarding research zones: North Central (NC), North East (NE), North West (NW), outputs in NU’s. For example, Nwagwu [13] studied the research SS, SE and SW. Each of these zones has 6 states, except the SE outputs of biomedical scholars in Nigeria based on publications in which has 5. The NFU’s are spread across these 6 zones. Nigerian journals indexed in Medline. The study covered research Data was collected from publication records (journals, conference published between 1967 and 2002 and found that the productivity proceedings and book chapters) as well as patents for each of the coefficient of first authors was higher than the productivity of any federal universities from the Scopus database and was analyzed other authors. using Scopus Analyzer. The Times Higher Education World Universities Ranking also used research publications in the Scopus Similarly, [17] conducted a study on the research outputs of some database for ranking the research outputs of universities [1].The selected NU’s located in three geopolitical zones of Nigeria Scopus database provides researchers a fast, simple, and virtually namely; South South (SS), South East (SE) and South West (SW). complete resource to support the research needs of the scientific, Data regarding the scholar’s publications in science and technical, medical, social science, and art and humanities engineering faculties in both indexed and non-indexed journals disciplines [28]. The year span for this research was limited to year were elicited through questionnaires and the period covered was 1990-2014, for fairness of comparison, since the universities were from 1997 to 2006. Results showed that the scholars published not established in the same year, as discussed in the study [11] on more in local journals than in international journals. Further Korean Universities. For the purpose of selection, publications of analysis of the results revealed that no significant differences each of the NFU’s were collected since inception of the existed in the research outputs of the scholars in local publications; universities. In each of the zones, the 2 universities with the however a significant difference existed in their international highest number of research outputs were selected for study (see publications [19]. Table 1) to have a reflection of the Universities in Nigeria. Popoola [23] investigated the impact of information sources and Numbers of publications were used as the criteria for the selection services on the research outputs of social scientists in 13 first and because the focus of this research is the research outputs of the second generation universities in Nigeria. The data for the study NFU’s, a major indicator of research reputation of a university [6]. were collected through questionnaires and analyzed. It was established that the use of information sources and services had an effect on the research outputs of social scientists, with on average Table 1. The profile of the selected universities each with a two outputs produced each year. corresponding total number of research outputs since the inception of the university as retrieved from Scopus (May 2014) Also [6] compared research outputs in Nigeria’s tertiary institutions by extracting data from 7 journals selected from the Established Research Generation University Official Website African Journals online database, for a period from 1999 to 2005. outputs It was found that polytechnics in Nigeria have very low research http://www.abu. 1948 14,197 First UI edu.ng/ outputs, although African Journals online is of course limited only to journals from Africa. http://www.oaui 1962 6,673 First OAU fe.edu.ng/ Furthermore, [19] analyzed the research outputs of 13 Federal Universities in Southern Nigeria. The data was collected through http://www.abu. 1962 5,209 First ABU edu.ng/ administering of questionnaires. Their results showed that there is no significant difference in the mean research output of academics http://www.buk. 1975 773 Second in the universities of southern Nigeria. However there is statistical BUK edu.ng/ difference in the mean research output between the universities http://www.uni 1975 1,498 Second when local journal publications were considered. Their results also UNIMAID maid.edu.ng/ showed that the University of Benin had the highest mean publication output in local journals while University of http://www.atbu 1988 350 Third ATBU .edu.ng/ Agriculture, Abeokuta had the highest in international journals. http://www.unil 1975 3025 Second Oliver and Miriam [29] also investigate the research output of UILORIN orin.edu.ng/ about 1,800 business economists working at Austrian, German and Swiss universities. The research rankings of the university http://www.unij 1975 1,610 Second UJ os.edu.ng/ departments were analyzed to determine each department’s research performance. Their results showed that individual http://www.unib 1970 4,271 Second research productivity and consequently departmental research UB en.edu.ng/ productivity is affected by institutional and personal http://www.unip 1975 2,121 Second UP characteristics. 80 ort.edu.ng/ Total 1990-2014 359 570 12t h http://www.unn. 1960 5,762 First UNN edu.ng/ BUK 1990-1994 41 41 http://www.uniz 1992 1,110 Third 1995-1999 32 54 UNIZIK ik.edu.ng/ 2000-2004 36 73 2005-2009 159 158 A total of 12 Universities were included in the study: the 2010-2014 361 158 University of Maiduguri (UNIMAID) (Borno State); Abubakar Tafawa Balewa University (ATBU) (Bauchi State); Bayero Total 1990-2014 629 484 11t h University Kano (BUK) (Kano State); Ahmadu Bello University (ABU) (Kaduna State); the University of Jos (UJ) (Plateau State); ABU 1990-1994 329 157 the University of Ilorin (UNILORIN) (Kwara State); the University of Ibadan (UI) (Oyo State); Obafemi Awolowo 1995-1999 361 158 University (OAU) (Osun State); the University of Port-Harcourt 2000-2004 457 156 (UP) (Rivers State); the University of Benin (UB) (Edo State); the 2005-2009 967 158 University of Nigeria Nsukka (UNN) (Enugu State); and Nnamdi Azikiwe University (UNIZIK) (Anambra State). The universities 2010-2014 1474 160 were classified into generations (Table 1) based on the Total 1990-2014 3588 789 4th classifications given by [14]. UNILORIN 1990-1994 219 154 4. RESULTS AND DISCUSSIONS 1995-1999 236 158 The results of the analysis provided by the Scopus analyzer are presented below and comprise the analysis of the NFU research 2000-2004 269 158 outputs based on subject area, author index, international 2005-2009 748 158 collaborations, patents, together with a comparison of research outputs of the UI and selected universities from Malaysia, Brazil, 2010-2014 1211 159 Indonesia, and South Africa. Total 1990-2014 2683 787 6th UJ 1990-1994 219 155 Research productivity by university affiliation 1995-1999 177 155 Table 2 presents the total number of research outputs for each of the NFU’s including the scholars involved in the research. The first 2000-2004 193 154 column is the name of the university, the second column represent 2005-2009 408 150 the period covered for the research outputs, the third column shows the number of research outputs within the latter and the fourth 2010-2014 458 150 column presents the number of authors involved in producing the Total 1990-2014 1455 764 8th research outputs. The last column of Table 2 shows the ranking of each of the universities in terms of the number of research outputs UI 1990-1994 524 160 published. The period of the publications are categorized into 4 1995-1999 633 159 subgroups in an interval of 5 years, similar to the work of Park 2000-2004 961 159 [21], in order to investigate any changes in research output over time [12]. 2005-2009 2458 160 Table 2. Research productivity of the universities and authors 2010-2014 3261 158 involved during 1990–2014 Total 1990-2014 7837 796 1st University Period Research Authors Ra OAU 1990-1994 549 158 covered outputs involved nk 1995-1999 521 160 UNIMAID 1990-1994 149 155 2000-2004 788 159 1995-1999 180 157 2005-2009 1699 159 2000-2004 179 156 2010-2014 1643 159 2005-2009 395 150 Total 1990-2014 5200 795 2nd 2010-2014 532 158 UB 1990-1994 261 157 Total 1990-2014 1435 776 9th 1995-1999 259 158 ATBU 1990-1994 34 58 2000-2004 390 158 1995-1999 71 127 2005-2009 1163 158 2000-2004 43 85 2010-2014 1210 158 2005-2009 105 156 Total 1990-2014 3283 789 5th 2010-2014 106 144 UP 1990-1994 173 157 81 1995-1999 148 153 Table 3: The most prolific researchers in the 12 selected 2000-2004 191 155 universities 2005-2009 533 158 Author Research Citations Author Name University 2010-2014 777 159 ID outputs (h-index) Total 1990-2014 1822 782 7th 70062 Mohammed, UNIMAID 47 274(4) 41848 Idris UNN 1990-1994 476 154 66034 Aliyu, 1995-1999 466 159 ATBU 30 144(6) 34330 Usman O. 2000-2004 489 155 Habib, 35569 2005-2009 1025 155 Abdulrazaq BUK 37 390(10) 54140 Garba 2010-2014 2030 157 70063 Total 1990-2014 4486 780 3rd Ameh E. A. ABU 148 822(15) 56993 UNIZIK 1990-1994 32 51 70036 Abdurrahma UNILORI 76 278(1) 1995-1999 120 156 08012 n, Murtala B. N 2000-2004 151 148 66026 Agaba, 54 206(8) UJ 2005-2009 256 144 79143 Emmanuel I. 2010-2014 597 155 70065 Ogunniyi, 151 2862(22) UI 18286 Adesola O. Total 1990-2014 1156 654 10t h 70038 Bassir, 113 241(0) OAU 46322 Olumbe 70035 Okieimen F. 117 1101(18) Table 2 clearly shows that in all of the universities included in UNIBEN the study, the pattern of research outputs are changing over time. 32113 Ebhodaghe The number of research outputs is increasing in each of the 70039 Orisakwe, O. 122 700(12) UP universities as well as the number of authors involved in those 98930 E. output. The increasing number of publications as well as authors 70048 Akah, Peter 119 1216(15) might be attributable to an increased awareness of the need to UNN 48496 Achunike publish in reputable indexed journals. The university with the highest number of research outputs between 1990 and 2014 is the 67016 Esimone C. UNIZIK 95 454(10) UI whereas ATBU has the lowest output. This is not surprising 61871 Okechukwu because ATBU is a third generation university and research outputs increases with the age of the university, as shown in Table In a similar manner to the work of Bar-Ilan [3], Table 3 shows the 1. The top 4 universities with the highest number of research most productive scholars in each university in terms of their ID, outputs are the first generation universities (see Tables 1 and 2). If name, institutional affiliation, and department, number of the researchers in NFU’s are motivated to publish research outputs publications, citations, and h-index. The performance index of the in journals indexed in Scopus, probably the number of the research researchers is based on the number of publications, citations, and outputs in Scopus from NFU’s can significantly increase over time. h-index [15] [7].In each of the universities, the scholar with the Even if not all the NFU’s is motivated, the motivation can be pilot highest number of publications based on the Scopus rankings was test with the first generation universities in view of the fact that selected for inclusion. The scholar that has the highest number of they have the potentials to compete with other world class research outputs, citations, and h-index is Ogunniyi, Adesola O., Universities if adequately motivated. It was found however that who can probably therefore be considered as the most influential most scholars in Nigerian universities publish in non-indexed local researcher in medicine in Nigeria. Bassir, Olumbe and journals [17]. This limits the visibility of the research outputs Abdurrahman, Murtala B. have 0 and 1 h-index respectively, internationally and hence also its impact. Figure 1 below depicts despite having high publications and citations. This is because the rankings of NFU’s based on Table 2. Scopus considers only articles published after 1995 for computation of the h-index [28]. International collaborations NFU'S Rankings International research collaborations are measured based on international co-authorship [11]. International collaborations in 1st 3rd 5th 7th 9th 12th research are essential for some types of research as they may be 11th necessary for funding purposes and to increase research 2nd 4th 6th 8th 10th productivity [10]. Moreover, Amin and Mabe [2] have shown that international collaborative papers have more impact than papers authored by a single author. Therefore, it was necessary to analyze the international research collaborations between the NFU’s and other universities outside Nigeria. As per Beaver [4], the collaborations were extracted from the Scopus categorization of Figure 1: Ranking of NFU’s international collaborating affiliations of institutions. Scopus is one 82 of the standard methods used in such an analysis of co-authorship generation universities respectively. The universities with the [10]. Table 5 shows the total number of international lowest number of international collaborations are the UNILORIN collaborations and percentages (bold) computed using Equation (1) and the UI as indicated in Table 4. The UI, despite having a very for each of the universities. high number of research outputs compared to the other NFU’s, still requires significant improvement in terms of international profile. tic Pic  x100 (1) tp 5. CONCLUSION Where Pic, tic, and tp are the percentage of international The paper compares the research outputs of the NFU’s in order to collaborations, number of research outputs with international allow policy makers to quickly identify areas of weakness. The collaborations (see Table 5) and the total number of research research outputs of the NFU’s were collected from the Scopus outputs (1990–2014 refer to Table 2). The rank column in Table 5 database and analyzed. The major findings are; The UI has the indicates the ranking for each of the universities based on highest number of research outputs among the universities under percentage of international collaborations. The highest rank is 1 study and was found to be leading in mathematics, computer and the lowest 12. science, medicine and the social sciences and BUK has the highest percentage of papers (31.64%) with international research Table 4. Percentage of university international collaborations collaborations. 1990–2014 Future work will include extension to other research subject areas. University International Rank We also suggest future researchers to conduct a study on the Collaborations comparison of first, second and third generation universities in UNIMAID 233 (16.23) 4th Nigeria. To investigate the annual research and development ATBU 112 (31.20) 2nd budgets of the universities and to look in more detail at citations per staff, international faculty, ratio of international students and UNILORIN 20 (0.75) 12th citation per faculty among other characteristics as derived from the UJ 276 (19.00) 3rd Scopus database. We also suggest that future research compare the productivity of the private, state and federal Universities in UB 204 (6.21) 7th Nigeria. UP 52 (2.85) 10th 6. REFERENCES ABU 251 (7.00) 6th [1] Aguillo, I.F., Bar-Ilan, J., Levene, M. and Ortega BUK 199 (31.64) 1st J.L.,2010.Comparing university rankings. Scientometrics, 85(1), 24-256. UI 61 (0.78) 11th [2] Amin M and Mabe M (2000) Impact factors: Use and abuse. OAU 198 (3.81) 9th Perspectives in Publishing 1:1–6. UNN 253 (5.64) 8th [3] Bar-Ilan, J., 2008.Which h-index?—A comparison of WoS, Scopus and Google Scholar. Scientometrics, 74(2), 257-271. UNIZIK 101 (8.74) 5th [4] Beaver, D.D., 2001.Reflections on scientific collaboration (and its study): Past, present, and future. Scientometrics, 52(3), 365–377. [5] Breitzman, A. F.,and Mogee, M. E.,2002. The many International Collaborations applications of patent analysis. Journal of Information Science , 28(3), 187-205. 40.00% [6] Chiemeke, S., Longe, O.B, Longe, F.A and Shaib, I.O., 30.00% 2009.Research Outputs from Nigerian Tertiary Institutions: 20.00% An Empirical Appraisal. Library Philosophy and Practice (e- 10.00% 0.00% journal). Paper 233. UNILO… UNIMA… http://digitalcommons.unl.edu/libphilprac/233. UNN UNIZIK BUK ATBU OAU UJ UB UP UI ABU [7] Huang, M.H and Lin, W.Y.C., 2011. Probing the effect of author self-citations on h index: A case study of environmental engineering. Journal of Information Science, Figure 2: NFU’s International Collaborations 0165551511412027. [8] International Monetary Fund. World Economic and financial Table 4 reveals that BUK and ATBU have the highest percentage surveys. http://www.imf.org/. Accessed on 23 August, 2014. of papers with international research collaborations, possibly a [9] Ioannidis,J.P., Patsopoulos,N.A., Kavvoura, F.K, Tatsioni,A., result of their own limited research resources, a practice similarly Evangelou,E., Kouri,I.,Contopoulos-loannidis,D.G and engage by Korean researchers (Kim, 1999). Another possible Liberopoulos,G.,2007. International ranking systems for reason for the high percentage of international collaborations in universities and institutions: a critical appraisal. Bmc BUK and ATBU is the increase in the number of scholars in these Medicine, 5(1), 1. universities who have gone overseas for their postgraduate [10] Kenekayoro,P., Buckley, K. and Thelwall,M.2014.Hyperlinks training. These universities (BUK and ATBU) might have possibly as inter-university collaboration indicators. Journal of motivated their scholars to collaborate with researchers in foreign Information Science, 40(4), 514-522. Universities. The first generations universities (see Table 1) have a [11] Kim,M.J., 1999. Korean international co-authorship in very low proportion of international research collaborations science 1994-1996. Journal of Information Science, 25(5), compared to BUK and ATBU which are second and third 403-412. 83 [12] Mingers,J.C., and Lipitakis,E.A., 2014.A Bibliometric [22] Perkmann, M. and Walsh, K., 2007.University-industry Comparison of the Research of Three UK Business relationships and open innovation: towards a research agenda. Schools. In Proceedings of the International MultiConference International Journal of Management Reviews, 9 (4), 259– of Engineers and Computer Scientists. Vol. 2, Vol. II, March 280. 12 – 14. [23] Popoola, S.O., 2008. The Use of Information Sources and [13] Nwagwu, W., 2006.A bibliometric analysis of productivity Services and Its Effect on the Research Output of Social patterns of biomedical authors of Nigeria during 1967- Scientists in Nigerian Universities. Library Philosophy and 2002. Scientometrics, 69(2), 259-269. Practice (e-journal).Paper [14] Nwagwu,W.E. and Agarin,O., 2008. Nigerian University 183.http://digitalcommons.unl.edu/libphilprac/183. websites: a webometric analysis. Webology, 5(4), 1-20. [24] Quacquarelli Symonds Times Higher Education World [15] Nwagwu, W.E., 2010.Cybernating the academe: Centralised University rankings 2013/2014. scholarly ranking and visibility of scholars in the developing http://www.topuniversities.com/university-rankings Accessed world. Journal of Information Science 36(2), 228-241. 1 September, 2014. [16] NUC. National Universities Commission presidential special [25] Report on Visit to Universidade Sao Paolo, Brazil, 12 scholarship scheme for innovation and development. November 2012 http://scholarship.nuc.edu.ng/# Accessed on 24 August, 2014. https://docs.google.com/a/siswa.um.edu.my/file/d/0B7U5- [17] Okafor,V.N. and Dike,V.W.,2010.Research output of rw_vemBekZRNTYtWnR0TjQ/edit?pli. Accessed on 25 academics in the science and engineering faculties of federal August, 2014. universities in Southern Nigeria. African Journal of Library, [26] Stratilatis, C., 2014.University rankings and the scientification archives and information science, 20(1), 41-51. of social sciences and humanities. Ethics in Science and [18] O’Connell, C., 2013.Research discourses surrounding global Environmental Politics, 13(2), 177-192. university rankings: exploring the relationship with policy and [27] STolZ,I., Hendel, D.D. and Horn, A.S., 2010.Ranking of practice recommendations. Higher Education, 65(6), 709-723. rankings: Benchmarking twenty-five higher education ranking [19] Okafor, V.N., 2011.Comparative Analysis of Research Output systems in Europe. Higher education, 60(5),507-528. of Federal Universities in Southern Nigeria. Library [28] Scopus.http://www.elsevier.com/online-tools/scopus. Philosophy and Practice (e-journal) Paper 498. Accessed 25 May, 2014. http://digitalcommons.unl.edu/libphilprac/498. [29] Fabel,O.,Hein,M. and Hofmeister,R.,2008.Research [20] Pagell, R.A., 2009. University Research Rankings: From Page productivity in business economics: An Investigation of Counting to Academic Accountability. Evaluation in Higher Austrian, German and Swiss Universities. German Economic Education, 3(1), 71-101. Review, 9(4), 506-531. [21] Park, W.G., 2008.International patent protection: 1960– 2005. Research policy, 37(4), 761-766. 84