CITATION ANALYSIS OF THESES ON NIGERIAN LANGUAGES AND AVAILABILITY OF CITED SOURCES IN NIGERIAN UNIVERSITY LIBRARIES

Transcription

1 CITATION ANALYSIS OF THESES ON NIGERIAN LANGUAGES AND AVAILABILITY OF CITED SOURCES IN NIGERIAN UNIVERSITY LIBRARIES BY JONAS IFEANYICHUKWU EZEMA (PG/PhD/07/43038) DEPARTMENT OF LIBRARY AND INFORMATION SCIENCE UNIVERSITY OF NIGERIA, NSUKKA MARCH, 2012

2 i TITLE PAGE CITATION ANALYSIS OF THESES ON NIGERIAN LANGUAGES AND AVAILABILITY OF CITED SOURCES IN NIGERIAN UNIVERSITY LIBRARIES PRESENTED TO THE DEPARTMENT OF LIBRARY AND INFORMATION SCIENCE IN FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (Ph.D) IN INFORMATION SCIENCE BY JONAS IFEANYICHUKWU EZEMA (PG/PhD/07/43038) MARCH, 2012

3 ii CERTIFICATION EZEMA, JONAS IFEANYICHUKWU, a postgraduate student of the Department of Library and Information Science with the registration number PG/PhD/07/43038 has satisfactorily completed the requirements of the research work for the award of Doctor of Philosophy. The work embodied in this thesis is original and has not been submitted in part or full for any other diploma or degree of this or any other university Ifeanyi Jonas Ezema Prof. V.W Dike Student Supervisor

4 iii APPROVAL PAGE This thesis has been approved for the Department of Library and Information Science, University of Nigeria, Nsukka. By Prof. V.W Dike Internal Examiner External Examiner Prof. V.W Dike Head of Department Prof. S.A Ezeudu Dean, Faculty of Education

5 iv DEDICATION This work is dedicated to first of all God Almighty for His love and protection and then my wife Uju and children, Oluchukwu, Nnanna, Nneoma and Chukwubuikem for their patience with me during the process of completing the work.

6 v ACKNOWLEDGEMENTS This research would not have been possible without immense contributions of my supervisors, colleagues, family members and friends. I am highly indebted to my supervisor Prof V.W Dike who painstakingly directed this research, despite her very tight schedule. I will not forget her patience with me. I must posthumously appreciate Late Prof. B.E Edoka who was my first supervisor before his death in His contributions in the choice of the research topic and encouragement cannot be ignored. I will ever remember his fatherly roles in my life. The University Librarian, Nnamdi Azikiwe Library UNN, Dr. C.O Omekwu also played very useful roles to see the success of this research. His mentoring strategy exposed me a lot in research techniques which facilitated this study. I remain indebted to him. I also want to acknowledge the contributions of Prof M Afolabi and Prof F.A Okwor in the instrumentation and their very useful inputs in the work. I am also indebted to Dr. B.M Mba of the Department of Linguistics, Igbo and Other Nigerian Languages UNN for providing me with research materials and source data from his department. I am equally indebted to Dr. R. E Ozioko, Dr. E.O Omeje, Dr. V.N Nwachukwu, all of Department of Library and Information Science UNN, Dr. D.U Ngwoke, Dr. G.G Ezugwu, Okechukwu Attama, C.I Ugwu and Dr. (Mrs) V.N Okafor for the useful advice which assisted in enriching the work. I also recognize the contributions of my research assistants Uchenna and Onyinye Isiani, Mr. Chima Ugwuowo, Mr. S.A Obadera of University of Ife,

7 vi Mrs Ngozi Ukachi of University of Lagos, and Mr. G Kasa of Ahmadu Bello University Zaria for their assistance in the data collection. I am deeply indebted to my wife and children for their support and patience during the period. My brothers and sisters Regina, Edwin, Titus, Bernadine, Benson, Elizabeth and Cordelia were very supportive. I equally recognize the fatherly encouragement of my uncle Mr., C.I.A Isiani and Chief V.D Agbo who continuously reminded me of the need to complete this work. I will not forget to appreciate the support and prayers from Alpheus, Adolphus, Amechi, Nnaemka, Chibueze, Ogochukwu, Thaddeus Omeke, and many others too numerous to mention here. Ifeanyi J. Ezema

8 vii TABLE OF CONTENTS TITLE PAGE ii CERTIFICATION iii APPROVAL PAGE iv DEDICATION v ACKNOWLEDGEMENTS vi TABLE OF CONTENTS vii LIST OF TABLES x LIST OF FIGURES xi ABSTRACT xii CHAPTER ONE: INTRODUCTION Background to the Study Statement of the Problem Purpose of the Study Research Questions Hypotheses Significance of the Study Scope of the Study CHAPTER TWO: LITERATURE REVIEW Conceptual Framework of Citation Analysis Source Data for Citation Analysis Patterns of Citations and Information Seeking in the Humanities - 37 Pattern of citations in language and linguistics Theoretical Framework Language Research in Nigeria Availability of Nigerian Language Sources in Nigerian University Libraries Empirical Studies on Citation Analysis Summary of Literature Review CHAPTER THREE: RESEARCH METHODS Design of the Study Area of the Study Population of the Study Sample and Sampling Technique Sample Size Instrument for Data Collection Validation of the Instrument Procedure for Data Collection Method of Data Analysis Decision Rule for Selection of the Frequently Cited Sources

9 viii CHAPTER FOUR: PRESENTATION AND ANALYSIS OF DATA Research Question Research Question Research Question Research Question Research Question Research Question Research Question Research Question Research Question Research Question Research Question Research Question Summary of Major Findings CHAPTER 5: DISCUSSION OF FINDINGS, CONCLUSION, IMPLICATIONSANDRECOMMENDATIONS Discussion of Findings Implications of the study Recommendations Suggestions for Further Research Limitations of the study Conclusion REFERENCES APPENDICES Appendix I List of Cited Journals Appendix II Periods of journals acquisition in the university libraries 178 Appendix III: Citation Analysis of Nigerian Language Research Document Analysis Guide (CANLRDAG) Appendix IV: Availability of Frequently Cited Journals in Nigerian language Document analysis guide (AFCJNLDAG) Appendix V: Validators Comments Appendix vi Language Map of Nigeria

10 ix LIST OF TABLES Table 1: Type of cited sources Tabled 2: Language spread of theses Table 3: Language in which the theses were written Table 4: Citation to E-resources Table 5: Language of cited sources Table 6: Timeliness of sources cited in theses in Nigerian languages - 99 Table 7: Summary of timeliness of cited sources by periods Table 8: Frequently cited journals Table 9: Application of Bradford Law Table 10: Most frequently cited authors Table 11: Most frequently cited books Table12: Availability of frequently cited journals in Nigerian university libraries Table 13: Distribution of the Availability of Frequently cited Journals Table 14: Availability of most frequently cited books Table 15: Correlation between the Frequently Cited Journals Table 16: Test of Significance of Association Table 17: Correlation between Frequently Cited Books Table 18: Test of Significance of Association

11 x LIST OF FIGURES Figure 1: Citation Model Figure 2: Pie chart showing types of cited sources Figure 3: Bar chart showing subject spread of theses Figure 4: Pie chart showing the language in which the theses were written Figure 5: Pie chart showing the languages of the cited sources Figure 6: A line graph showing the timeliness of cited sources Figure7: line graph showing the scattering of the journals

12 xi ABSTRACT This study was conducted to find out the types of information sources used by the Nigerian language researchers and the availability of these sources in Nigerian university libraries. The population of the study was four hundred and forty nine (449). Theses from six Nigeria s first generation universities namely; University of Nigeria, Nsukka, University of Ibadan, University of Lagos, Obafemi Awolowo University Ife, Ahmadu Bello University Zaria and University of Benin were used for the study. Twelve research questions concerned with types of information sources, subject spread of theses language of the theses, use of e-resources, languages of cited sources, timeliness of sources, frequently cited journals, frequently cited authors, frequently cited books, and availability of cited sources in libraries; and two hypotheses guided the study. Descriptive survey design was used for the study. Data was collected by using researcher s developed document analysis guide where manual counts of citations in the theses were done. The data collected were analyzed using descriptive statistics of frequencies and percentages and then presented in tables and charts. Twenty-seven thousand, six hundred and eighty-six (27,686) citations were harvested from the 449 theses giving an average of 61.6 citations per thesis. The findings of the study showed that citations to books were 63.3% as against 16.2% to journals; Yoruba languages produced more theses than other Nigerian languages; over 80% of the theses were written in English; there were very few citations to electronic resources; English language sources were more cited than any other languages, materials published in the 1980s were cited more than other periods. Twenty-three journals were identified as most frequently cited journals while 74 authors with up to 50 citations were identified as the most frequently cited authors. The findings also identified 37 books as the most frequently cited books and inadequate availability of the most frequently cited books and journals in Nigerian university libraries. The two null hypotheses tested were retained indicating that there is no significant relationship between the number of frequently cited journals and their availability in Nigerian university libraries. In the same way there is no significant relationship between the number of frequently cited books and their availability in Nigerian university libraries was accepted. The findings have implications drawn for the governments of Nigeria, librarians and Nigerian linguists in developing policies that would address these inadequacies concerned with the study and availability of research materials in Nigerian languages. Based on these implications, some recommendations were made among which are: acquisition of current research materials in Nigerian languages by Nigerian university libraries, the Nigerian universities should make it mandatory for theses in Nigeria languages to be written in local languages; development of ICT infrastructure by the Nigerian university libraries; Nigerian government should sponsor publication of more research materials using indigenous Nigerian languages; acquisition of frequently cited research materials by Nigerian university libraries.

13 1 CHAPTER ONE INTRODUCTION Background to the Study A critical aspect of human socio- cultural evolution has been the development of language. While communication occurs among various animals, the most complex of all communication is the human communication embedded in language. This has made a remarkable difference between human beings and the rest of the animal world. Trudgill (1983) has noted that language is a very important means of establishing and maintaining relationship with other people. Heavily rooted in all aspects of human life, language is the store of ideas that provides us with different ways of looking at things around us. It is the encyclopedia of knowledge about the cultural life of a people, their religion, medicine, philosophy, literary tradition, and verbal arts. Rufai (2000:57) defines language as the spoken means by which human beings communicate and/or interact. However, Trudgill (1983) believes that the purpose of language goes beyond communication since other animals can as well communicate without the use of a language. Thus, Le Page cited in Rufai (2000:57) identified four functions of language for an individual. First, by acquiring the speech habits of his community, an individual becomes a full member of the society, contributing his response to the common stock in such a manner that they will be easily recognized and stimulate reactions from others which he himself can understand. Language also offers the individual greater

14 2 possibilities of self expression than he could have without it. In addition, it provides him with a tool for exploring and analyzing his own conceptual processes. In this sense, it helps him understand his own thoughts and at the same time imposes upon him the thinking habits of the community as crystallized in the lexicon and grammar of his language. It also gives him access to the experiences of others depending on the adequacy of his responses to the experiences and those of others. For any language to perform these functions adequately, it requires careful study of the language. The scientific study of Nigerian languages, numbering more than 450 (Adegbija, 2004), began in the 19 th century through the colonization and evangelization of Africa. Through the study of these languages, the Christian missions advanced their missionary activities, while the colonial government encouraged them in order to gain more political dominion over the people. The government used it as an opportunity to train clerks and interpreters who assisted in the propagation of colonial agenda. According to Hair (1967), the Roman Catholic Mission (RCM), the Church Mission Society (CMS), the Methodist and the Presbyterian missions laid the foundation for the study of Nigerian languages. From the 1840s, the Presbyterian mission had started work on Ibibio language in Calabar. Later in the century, the Roman Catholic Mission and the Methodist mission began the study of Yoruba and Igbo languages. However, the agents of Church Missionary Society undertook the earliest, the most thorough and the most extensive research in Nigerian languages (covering a dozen or more

15 3 languages). As was observed by Hickey (2000), unlike the Roman Catholic Mission, the C M S paid more attention to the development of indigenous languages in the areas they operated. From 1850, the Church Missionary Society was moving its missionaries from their main base in Freetown to other parts of West Africa. A place for the rehabilitation of the ex-slaves, Freetown provided the CMS with personnel for the study of many Nigerian languages that were widely spoken in the area then. The Nigerian language studies in the pre- independent era were in two phases: the missionary-based and the collaboration between the Christian mission and the government (Afolanya, 1989). According to him the missionary-based period started with missionary activities in Nigeria in the 1840s and continued until 1882 when the British government began the management of education and linguistic policies in Nigeria. The primary objectives of colonial education were the introduction of three Rs (Reading, Writing and Arithmetic), which would facilitate the colonial government s tripartite objectives of colonization, evangelization and commerce. The second phase was the period between 1882 and 1960 when Nigeria gained political independence. From 1882, there began collaboration between the missions and colonial government in the establishment of schools. Other parts of the country outside the coastal areas began to have schools. As the government began to formulate education and linguistic policies, the need to train clerks, interpreters, civil servants, teachers etc rose dramatically. Emphasis began to shift from knowledge of indigenous languages to

16 4 knowledge of English and other languages. This led to the introduction of multilingualism in educational systems. This emphasis was made prominent when the policy demanded that for one to obtain a government certificate, he must pass English language. This was also the requirement for securing whitecollar jobs (Ahmed, 1989). In spite of this policy, the colonial government did not entirely neglect the study of indigenous languages. Ahmed (1989) recalled that the colonial government gave the development of Hausa language considerable attention. For example, the government established a number of organizations for the promotion and development of Hausa language. Some of these organizations, according to him, include the Hausa Translation Bureau (1930), Hausa Literature Bureau (1939), North Regional Literature Agency (NORLA) (1954) and the Hausa Language Board. These organizations made serious and remarkable impacts on the development and study of Hausa language. Unfortunately, for other indigenous languages, there is no record that they enjoyed similar attention. Within this period, interest in the development and study of African languages in general and Nigerian languages in particular extended beyond the colonial government and the missionary societies. Through interest from international organizations and agencies, individual countries began the formulation of language policies as many African countries were gradually gaining independence from their colonial masters. There is no doubt that language policy formulation then was of significant importance for the growth of indigenous languages. Adegbija (2004) noted that the need for language policy arose partly because of the dangerous incursion foreign

17 5 languages were making into the country s socio-cultural values. Added to this was the need for a policy guide in identification of official, national, educational, inter-ethnic and international languages for the country. This policy formulation attracted the attention of regional and international organizations to Nigeria language study. Since 1951, according to Emenanjo (2006), UNESCO has committed considerable resources to the study of African languages. The period following independence in 1960 offered Nigeria the opportunity of intensifying studies in Nigerian languages. This opportunity regrettably was squandered, as the interest of the government was more on the study of foreign languages. Though government policies favoured the three major indigenous Nigerian languages, namely Hausa, Igbo and Yoruba, Ahmed (1989) noted that English language was respected and accorded the status of official language. The Regional Government made useful efforts in the development of the three languages-hausa in the North, Igbo in the East and Yoruba in the West. Afolayan (1989) remarked that the knowledge of Hausa was officially recognized as one of the criteria for appointment and promotion in the Northern Civil Service. This revolutionary measure to sustain people s interest in the study of the languages lacked national outlook, as neither Eastern nor Western Regions had such policies. The period of military rule was another critical phase in the development and study of Nigerian languages (Afolanyan, 1989; Ahmed, 1989; Emenanjo, 2006). The major effort during the period was the formulation of

18 6 national language policy that was incorporated into the National Policy on Education (revised in 2004). This policy recognizes the mother tongue as the language of instruction in primary schools in Nigeria. The major flaw of the policy, according to Emenanjo, was lack of specification on how the policy would be implemented. For instance the following implementation-related questions were not addressed: How would a child learn another language after the mother tongue? How would the necessary logistics be provided? What are the roles of mother-tongue after the junior primary school? The civilian administration of 1979 to 1983 paid little attention to the study and development of Nigerian indigenous languages. The implementation of the language policy as contained in the National Policy on Education was totally absent. The National Assembly could only appropriate money for the compilation of a glossary of political terms in Nigerian languages (Afolanyan, 1989). As a result of this, Ahmed (1989) contends that it seems the colonial government paid more attention to the development and study of Nigerian languages than the Nigerian government. He was worried that in the Northern Region, the use of Hausa as the medium of instruction in primary school was dropped after Nigeria s independence. This was replaced by a policy tagged Straight for English, which later failed. In the West, the popular Ife language project was abandoned by states in that region. Igbo language equally suffered serious neglect within the period.

19 7 Thus, Nwadike (2006) contends that most of the indigenous Nigerian languages are highly endangered. UNESCO (2009) in its publication confirmed this fear when it released an alert that about 2500 world languages are endangered and these include many Nigerian languages. Nwadike believes that a language is endangered when it has less than 5000 speakers. In line with this, Ugwoke s (1999) revelation that more than 150 Nigerian languages are endangered calls for serious concern since these languages according to him had less than 1000 speakers. Nwadike s fear is a reflection of Ohiri-Aniche s (2001:61) warning that, at the present time, all Nigerian languages are endangered. If Nigerians remain blind to this fact or prefer to look on unconcerned, then one must be prepared to see most if not all of them becoming extinct in the 21 st century. This development has generated renewed interest in the study of Nigerian languages. Many seminars, conferences and workshops have been organized to intensify studies in Nigerian languages. A 2006 workshop sponsored by UNESCO and National Commission for Arts and Culture (Obiosio, 2006) was aimed at mapping out strategies for safeguarding endangered Nigerian languages. The institutions of higher learning, particularly universities with departments of linguistics and Nigerian languages, have been mandated by the workshop to intensify studies on Nigerian languages. Consequently, several universities have begun to offer courses in Nigerian languages. Such mandates have also been extended to the Linguistics Association of Nigeria and other organizations that are interested in

20 8 sustaining research in Nigerian languages (Emenanjo 2006). Similarly, JAMB (2009) indicated that about twenty-nine universities in the country offer courses in Nigerian languages. The National Institute of Nigerian Languages (NILAN), now part of University of Nigeria, Nsukka offers courses in up to ten Nigerian languages. On the whole, these universities and NILAN offer courses in sixteen Nigerian languages. These languages are Hausa, Igbo, Yoruba, Edo, Kanuri, Fulfulde, Idoma, Izon, Tiv, Ngizim, Efik, Ibibio, Ikwere, Kalabari, Khana, and Urhobo. Though the universities in the country have been striving to live up to their responsibilities in the conduct of research in Nigerian languages as posited by Bamgbose (2006), the present interest in the sciences to the detriment of humanities scholarship has resulted in low intake of humanities students, particularly in Nigerian languages, and has consequently reduced research materials which libraries provide to Nigerian language researchers. By implication this development may have reduced the quantity and quality of research in Nigerian languages. In spite of this problem, Nigerian universities have continued to make efforts in research in indigenous Nigerian languages. For instance, Bamgbose (2006) reported that the Department of Linguistics and African Languages of the University of Ibadan has completed more than 26 doctoral theses since establishment in Similarly, Onyegegbu and Otagburuagu (2004) reported that students of the Department of Linguistics and Other Nigerian Languages of the University of Nigeria, Nsukka as at 2004, had completed ten

21 9 doctoral theses. Similar research outputs have been observed in other universities with departments of Nigerian languages. This is in line with King ei s (1999) postulation that Africa s development in the present millennium must be research driven since socio-cultural and economic investigation would provide the critical data which will propel sound informed policies governing development for all facets of the society. Language research particularly in developing countries, according to King ei,(1999:1) is a driving force for development research because language is a vital tool of communication. He contends that Africa has failed to make substantial improvements in the quality of life for the majority, mainly because development theory and practice have failed to exploit local languages as media for research and development work. In line with this, Emenanjo (2000) has written that language research is necessary in Nigeria to provide relevant facts, figures and statistics without which adequate planning and implementation will be difficult. He noted that it is important to know how many Nigerian languages are available. What are their exact locations? How many people speak which language as language 1 (L1) and/or language 2 (L2)? What are the exact stages of development of these languages? What is the commitment of government towards the development of these languages and language research generally? This situation calls for a more pragmatic approach to research in local languages which must be guided with adequate information.

22 10 The library has often been strategically placed as the major source of information for such research. As a result, librarians have always been interested in creation and use of information resources to enhance research and scholarly productivity. As information experts, they have usually been challenged in the ability to generate, organize, make accessible, analyze and evaluate information resources in a way the researchers would find useful. Consequently, librarians have recognized bibliometric studies as a very useful technique for proper analysis and evaluation of research outputs, leading to its growing acceptance for research evaluation. Sharada (1993) defined bibliometrics as the quantitative study of various aspects of literature of a subject. Such study can be undertaken at national or international levels. This has been put to test by several works in bibliometric studies, (Garfield, 1979; Karisson, 1994; Omekwu and Atinmo, 1998; Ajala, 2001; Ibeun, 2001; Lasca and Mendeson, 2001; Thompson, 2002; Herzel, 2003; Cameron; 2005; Lar Bonte, 2005; Leydesof and Olger, 2006; Tonta and Al, 2006). Wikipedia free encyclopedia defines citation analysis as the examination of the frequency and pattern of citation in articles and books and using these citations to establish links to other works or other researchers. Gooden (2001) defines citation analysis as an excellent unobtrusive method to determine which resources researchers are using. In the same vein, Meho (2007:1) noted that citation analysis is a branch of information science in which researchers study the way articles in a scholarly field are accessed and referenced by others He points out that it involves counting how many times a paper or a

23 11 researcher is cited and at the same time assuming that influential scholars and important works are cited more than the others. Citation analysis assumes that references to a particular journal reflect a scholarly impact of that article on the author of the citing work. Another assumption is that the accumulated citations to a given author s work in some sense reflect the impact of that author on scholarship and research. In their contribution, Ibeun and Madu (2001) contend that citation analysis is the investigation of literature cited by writers in a subject field. It is the analysis of the information resources used by authors in composing the contents of their texts. The assumption is that there is an intellectual link between the citing sources and the reference article. Therefore, citation analysis is used to determine the characteristics of materials used by authors in a given area and this provides the citers link with the authors they have cited. Karison (1994) observed that citation analysis has been found to be useful in investigating the coverage of library acquisitions and for research evaluation. The objective of citation analysis is to scrutinize how researchers make references to earlier works they have read. It is believed that a work s relevance to scholars is determined by the frequency of its citations in later works. Harter (1996) has remarked that citation analysis is a generic term for a set of well-known techniques that have a long history in bibliometric studies of scholarly communication. Citation is used to give credit to related works, to substantiate claims, to criticize previous works. Garfield (1979:245) characterized the basic nature of citation analysis as a very general measure of

24 12 contribution an individual makes to the growth of knowledge. In line with this, Casserly and Bird (2003) have noted that through citation, an author generally acknowledges a researcher before him. They argue that it is a very good method of tracing the historical development of a given field of study and lays a proper foundation for future research in the field. The relevance of citation analysis has been a burning issue in bibliometric research (Bichteteler 1980; Karisson 1994; Garfield, 1998; Gooden 2001; Brian, 2005; Moed, 2005; Meho 2007). Owing to this Bitcheteler (1980) has argued that a good reason for citing an author is to give credit to the intellectual owner of the work. Through this process, it becomes possible to establish the citation pattern of a given discipline depending on the types of sources cited by researchers. The dominant sources could be books/monographs, journals, theses, government documents, news publication and other kinds of research materials. The fifteen reasons given by Weinstock (1971), adequately summarized why citation is relevant in research and scholarship: 1. Paying homage to pioneers; 2. Giving credit for related works 3. Identifying methodology, equipment, etc; 4. Providing background reading; 5. Correcting one s own work; 6. Correcting the work of others; 7. Criticizing previous work;

25 13 8. Substantiating claims; 9. Alerting researchers to forth-coming works; 10. Providing leads to poorly disseminated, poorly indexed, or uncited works; 11. Authenticating data and classes of facts physical, constant etc; 12. Identifying original publications in which an idea or a concept was discussed; 13. Identifying the original publication describing an epongnic concept or term as, e.g. Hodgkin s disease, Pareto s law, Friedel-Grafts Reaction; 14. Disclaiming work or ideas of others; 15. Disputing priority claims of others. Beyond these, it has been discovered that citation analysis provides a good method of in-house evaluation which librarians employ in collection evaluation and maintenance of library resources (Gooden, 2001). It assists in determining the most frequently cited sources and through this, guides in identifying core journals in given field of study. Citation analysis is therefore useful in various disciplines to eliminate costly low use/unused journals, and ascertain journals needed for patron use and in determination of most active research and volume of research in a particular area of study. Similarly, Karisson (1994) and Lawani (1982) contend that citation analysis is a useful instrument for investigation of coverage of library acquisition; investigation of how the cognitive content and social structures of scientific theories developed;

26 14 and evaluation of research in terms of individual output, institutional output or geographical output. Garfield (1998a) has provided historical bases for citation analysis. He remarked that citation analysis has been useful to ascertain scholars who may be behind a particular subject area. When a data-base of a subject is created, the authors whose titles are included in that subject or related area will show the history of that subject and those who have created impact on the subject. In addition to this, citation analysis has also been found useful in identifying the dominant languages of cited sources with the intention of assisting librarians in the choice of languages of materials to acquire and equally assist publishers in determining the languages in which very important books would be translated. Consequently, Meho (2007) believes that citation analysis is useful for the following reasons; it provides information on how much impact a particular article has had by showing which other authors found the work relevant in writing their own papers. It is also a useful tool to investigate more about a field or topic (by reading papers that cite several works in that area); it is also used to find out how much impact a particular author has had (by looking at his/her total citations). Apart from this, it is a useful tool in ascertaining the timeliness of information sources used by researchers in their scholarly communication. This concern here is to examine the age of the sources cited by authors in order to identify sources that have been consistently relevant despite their age.

27 15 Attempts to measure the impact of journals or an author in a particular field have always been an issue in the library and information science profession. As far back as 1927 Gross and Gross made a bold attempt to conduct a study of citation patterns in order to determine the most valuable journal in chemistry by manually counting the number of citations in a Journal of American Chemical Society and ranking them in frequency of citation (Herzel, 2003). Since then, there have been several efforts to establish the relevance of journals, the pedigree of researchers to determine which journal should be selected, to determine the volume of research in a given field, to evaluate the performance of an institution in terms of its contribution to knowledge and to determine the relationship existing between and/or among some articles through citation analysis. Unfortunately for humanists, most bibliometric studies have been associated with impact assessment of scientific journals. The neglect of the humanities and monographs has created a vacuum, which must be filled to accommodate the entire scholarly community. Since the scientific community has dominated bibliometric research, a common method of handling citation analysis is by collecting source data from science citation index. This approach, though has gained international recognition, but has hardly taken care of journals and other information sources that are not abstracted or indexed by international abstracting and indexing agencies dominated by the Institute for Scientific Information (ISI) now Thompson Scientific.

28 16 Sadly enough, most journals coming out of Africa and other developing countries belong to the group that is hardly indexed or abstracted internationally. As a result of this, La-Bonte, (2005) has called for a citation analysis using faculty publications instead of science citation index to determine the relevance of journals in structural biology. This shift from using science citation index is equally reflected by McCain (1981). In an evaluative study, he attempted to find out the level of use of the journals in the departmental library by faculty members and graduate students, using scholarly publications of full time faculty members, dissertations of graduating doctoral students, and preliminary doctoral qualifying graduate students. Through this method, scholars who are excluded by abstracting agencies are given the required recognition. As Karison (1994) has argued, low citation rate of a work does not make it substandard; particularly when the work is published in place not prominent in the global scholarly environment. According to him, several variables may reduce the citation rate of an author or published work. These include: an extra ordinary pioneering piece of work, which had not been widely accepted and absence of a scientific/scholarly community in a very rare field. These variables are characteristics of developing countries such as Nigeria where very many researchers face the problem of their work being widely read by peers. This problem localizes even an excellent work and therefore such works are less cited. This is evident in citation studies that generated their source

29 17 materials from Science Citation Index, Social Science Citation Index or Arts and Humanities Index, which have Euro-American bias. Electronic publications that could have afforded researchers from developing countries opportunity to globalize their publication have not taken root in Nigeria and so many other developing countries. Apart from this, studies have revealed that citations from electronic sources have been very low, particularly in the humanities (Hartar, 1996; Kushkowski, 2005; Zharig, 1998; Casserly and Bird, 2003). In most of these works, citation analysis conducted revealed that only a very insignificant percent of cited materials are from electronic sources. For example in Zharig s (1998) work, 74 peer reviewed scholarly journal were examined and only 1.9% of the total journals were electronic journals. He therefore concluded that electronic journals do not play a significant role in scholarly publications. In a later work by Herring (2002), it was revealed that electronic citation accounted for only 16% of selected set electronic journals. In view of this development, it is usually more convenient to use the theses as source data for citation analysis coming in most African countries. The relevance of theses as very important information sources in research literature have been underscored in several studies. Theses and dissertations have been found to contain results of preliminary studies or discussions of future research directions that would be very valuable to upcoming researchers in that area (Eaton, cited in Kiondo, 2004). The doctoral theses in particular form the bulk of the academic programme and play very significant roles in

30 18 higher education. In The Role and Nature of the Doctoral Dissertation: A Policy Statement, the Council of Graduate Schools (1991:3) states There is no question that, in the view of the faculty, students, and administrators participating in this study, the doctoral dissertation, as a demonstration of a student s ability to carry out research independently, defines the essence of the Ph.D. degree. In corroborating this, Boyer cited in Gooden (2001:2) described the dissertation as the capstone to formal academic training process. Similarly, Barry (1997) argued that a successful doctoral student tends to be comprehensive and up to date in reviewing the literature. Doctoral theses therefore are invaluable primary literature which cannot be ignored in the scholarly environment. This is because the doctoral thesis is an embodiment of primary research finding that has passed through a meticulous and rigorous peer review process through series of panels of experts. Kiondo (2004) has equally underscored the importance of theses and dissertations in the scholarly communication circle. According to her they are considered as valuable sources of research materials for not only students and scholars, but also for policy makers and other people who benefit from research findings. Librarians are not only interested in the type of information sources researchers cite in their work, but also on whether these sources are available in the libraries. This is because it is believed that the availability of these sources determine the extent use and citations. Availability of information resources in libraries particularly academic and research libraries has been a source of concern to library and information professionals leading to several

31 19 studies on the problem of scarcity of information sources in libraries. Therefore, Ugah (2008) reasoned that availability of information sources entails the presence of such sources in the library for immediate use. The moment a library user finds it difficult to identify a material he requires, frustration sets in and this might affect the quality of research being generated. As a result of this, Aguolu and Aguolu (2002) were of the opinion that availability or non availability should be a yardstick in measuring the state of library resources in Nigerian university libraries. They argued that the poor state of most university libraries in Nigeria in terms of availability of information sources was as a result of proliferation of universities in the country with no corresponding match to human and material resources. This position corroborates an earlier study by Dike (1992) on scarcity of books in Nigeria as a threat to academic excellence. She posited that non availability of information resources has led to poor library services. Availability of information sources in libraries can be seen in four perspectives; namely physical (which is the main concern of this study), bibliographic, intellectual and online (Omeje, 2010). Physical availability is concerned with the presence of the information sources (preferably in hard copies) in the library. This means that the users can physically see and make use of the source. Bibliographic availability refers to the presence of which were made to the sources or their contents (indexing and abstracting and compilation of bibliography on them) without the physical presence of the document in the library. Intellectual availability entails the presence of

32 20 individuals (library staff) with the knowledge of the information or other relevant sources of getting the information which the library user requires. Finally, online availability implies the presence of the required information sources on the Internet and to a greater extent the presence of information and communication infrastructure in the library for enhanced access of the sources. The problem of finding information requested by users made availability studies in libraries very critical in evaluation of library services quality. This motivated Kantor (1984), using patrons requests and branching analysis to conduct a study on why some items were not available in libraries. His methods were tested at four research libraries and published by Association of Research Libraries in Later other studies on availability of information sources began modifications of this method (Bachman-Derthick & Spurloch, 1989; UNC Health Sciences Library, 1997). As the information and telecommunication technology began to propel availability and access to information sources, availability studies started getting interested on the electronic availability of sources resulting to some interesting studies (Canick, 2002; Baldwin, 2009). Another study by Marama and Ogunrombi (1996) regretted that poor availability of information resources in the field of Library and Information Science in many Nigerian university libraries has put to question on where and how Nigerian Library and Information professionals get access to their research materials. The argument is that it will be difficult for them to conduct research if materials are not available in the libraries. This study though limited to

33 21 Library and Information Science can be used for generalization. This is because other studies such as Aina (1985),Unomah (1987), Okiy (2000), Iyaro (2004) revealed similar cases of poor availability of information resources in Nigerian university libraries. According to Iyaro s study the unavailability rate of serials publications in University of Ibadan was 94%. Most often this leads to frustration among library users while others lost interest in going to the library for information. In another study by Kumar and Shah on availability of information sources in Vikran university library in India, about 68 percent of the respondents were satisfied with the reading materials available, but science faculty members were not satisfied. This points to the fact that a library may be rich in collections of information sources in some disciplines but poor in some others. Statement of the Problem Language is a vital instrument for national development. It has been recognized as an indispensable means of communication through which individuals express ideas, feelings and opinions to one another. Language has pivotal roles to play in all aspects of human endeavour ranging from education, religion, politics, culture and socio-economic life of a people. The effective performance of these roles demands the viability of the language. As a result of this, King ei (1999) argues that African indigenous languages as vehicles of social integration and orientation should be continuously strengthened to compete favourably with other international languages. Language studies and

34 22 research are pragmatic approaches in doing this, but such requires the availability of research materials in Nigerian languages. Paradoxically, observations have shown that universities tend to favour scientific and technological studies rather than language studies. Therefore, languages generally are neglected and Nigerian languages usually suffer more from this neglect. This is also reflected in the pattern of acquisition of research materials in university libraries where a greater percentage of the library budget is devoted to the acquisition of scientific publications. Consequently, the university libraries with already depleted budget concentrate their efforts in the acquisition of research materials more in the sciences than the languages and whenever there is the need for cancellation of journal subscriptions the humanities particularly languages are usually the victims. This has been the general pattern in many Nigerian university libraries. Should this be allowed to continue, the study and research in Nigerian languages will continue to suffer terribly and may discourage the younger generation from developing interest in the study of Nigerian languages. Through this, Nigeria would have lost one of the most valuable cultural heritages as UNESCO (2009) had already given a serious alert that many of the languages are endangered. The UNESCO s alert is so serious that Igbo, one of the major indigenous Nigerian languages may go into extinction in the next fifty years. This alert requires an investigation to examine the availability of research materials in Nigerian languages for it is already becoming doubtful whether Nigerian university libraries are meeting

35 23 their statutory objectives of supporting teaching, learning and research particularly in Nigerian languages. In the light of the above development, this study attempted to examine the kinds of information sources used in Nigerian language research and their availability in Nigerian university libraries. Therefore, the problem which this study addressed posed as a question is: could citation analysis be carried out to investigate the types of resources the Nigerian language researchers cite and the availability of these cited sources in Nigerian university libraries? Purpose of the Study The main purpose of this study was to conduct a citation analysis of theses on Nigerian languages and determine the availability of cited sources in Nigerian university libraries. Specifically, the study sought to find out the following; 1. the types of sources cited in Nigerian language research; 2. the language spread of theses; 3. the language in which these theses are written; 4. the quantity of electronic citations in the cited sources; 5. the language of cited sources in the theses of Nigerian languages research; 6. the timeliness of the sources cited in the theses of Nigerian language research;

36 24 7. the most frequently cited journals in the theses of Nigerian language research; 8. the most frequently cited authors in the theses of Nigerian language research; 9. the most frequently cited books in the theses of Nigerian language research; 10. the availability of the cited sources in the university libraries. 11. The relationship between the number of frequently cited journals and their availability in Nigerian university libraries; 12. The relationship between the number of frequently cited books and their availability in Nigerian university libraries. Research Questions The following research questions guided the study: 1. What types of sources are cited in Nigerian language research? 2. What is the language spread of the theses? 3. What are the languages of the theses? 4. What is the quantity of electronic sources cited in the theses? 5. What are the languages of cited sources in theses on Nigerian languages? 6. What is the timeliness of sources cited in theses on Nigerian languages? 7. Which are the most frequently cited journals in the theses on Nigerian languages? 8. Who are the most frequently cited authors in the theses of Nigerian languages?

37 25 9. Which are the most frequently cited books in the theses of Nigerian languages? 10. Which of the most frequently cited journals are available in Nigerian university libraries? 11. What is the relationship between the number of frequently cited journals and their availability in Nigerian university libraries? 12. What is the relationship between the number of frequently cited books and their availability in Nigeria university libraries? Hypotheses This study tested the following null hypotheses using Pearson s correlation coefficient at 0.05 significance: 1. There is no significant relationship between the number of frequently cited journals and their availability in Nigerian university libraries. 2. There is no significant relationship between the number of frequently cited books and their availability in Nigerian university libraries. Significance of the Study This study will make some significant contributions to evaluation of information resources used in Nigeria language research and assessment of research out puts in Nigerian languages. This is a major research in the types of information resources used in Nigerian language research and their availability in Nigerian university libraries through citation analysis. Through the findings of the study, the quality of information resources used in Nigerian

38 26 language research will be revealed. The lapses observed will guide future researchers and students in the field. In the era of budget cuts in libraries, the study will be a useful guide to librarians in initiating sustainable strategies in selection and acquisition of relevant monographs, journals and other information resources for language research. Beyond this, the findings of this study will be useful to Nigerian linguists and organizations concerned with the development of Nigerian languages. It will help formulate strategies that will deepen the study of Nigerian languages in schools so as to combat the problem of endangered languages in Nigeria. The revelation of the study will also be of significant interest to the National Institute of Nigerian Languages (NILAN) that has been empowered to conduct research and train manpower for the development of Nigerian Languages. The findings will help the institute re-evaluate its activities and policies so as to strengthen interest in the study of Nigerian languages. Publishing houses will also benefit from the findings of the study since the findings will assist them in identifying areas that lack publications in Nigerian languages in terms of research materials. For example the findings of the study will help to identify the kinds of books, journals and other research materials that are of scholarly value to publish. Finally, the findings of the study will be of interest to scholars with research interest in bibliometrics and language study. It will lay a very good foundation in bibliometric study of Nigerian language research.

39 27 Scope of the Study There are about five hundred languages in Nigeria, (Egbokhare, 2004). This constitutes about 20% of more than 2000 African languages. However, the languages covered by this study are Hausa, Igbo Yoruba, Edo, Efik, Fulfulde, Urhobo, Nupe, Ibibio, Isoko, Tiv and Ngas. Theses submitted to the departments of Nigerian languages in Nigeria s first generation universities from 1998 to 2007 were used for the study. Ten year-period was chosen for the study so as to accommodate enough theses and dissertations for the study. For the purpose of this study, theses imply all postgraduate research reports (both doctoral and masters) accepted by the universities studied.

40 28 CHAPTER TWO LITERATURE REVIEW subheadings: The study reviews literature under the following headings and Conceptual framework of citation analysis Source Data for Citation Analysis Patterns of Citations and Information Seeking in the Humanities Patterns of Citations in Languages and Linguistics Theoretical framework of citation analysis Bradford s Law of Scattering Garfield s Law of Concentration Language Research in Nigeria Availability of Nigerian language sources in university libraries. Empirical works on Citation Analysis Summary of Literature Review Conceptual Framework Concept of Citation Analysis Citation analysis is an aspect of bibliometric studies which deals with analysis of references listed at the end of a research work, usually done through citation

41 29 counts. It is a wide ranging area of bibliometrics that studies the relationship between a work and documents citing the work or an author and other authors citing him. The assumption is that there is a relationship between a work and the source it is citing. Figure 1 (Conceptual framework of citation) is a self developed diagrammatic representation of these relationships in citations. A close look at the diagram in fig.1 shows that the cited information sources in the library are generated by the authors, publishers, vendors and other stake holders in the book production. The library acquires these resources which are in different languages including Nigerian languages for the benefits of its users. These sources include books/monographs, theses, journals, reports, proceedings, government documents among others. The availability of these sources in the library is dependent on the acquisition policy and financial ability of the library. Publishers Authors Vendors theses, books journals proceedings, reports, lectures etc. Available information sources User (Citer) Citation Theses Books Journals Proceedings Reports Lectures etc Input Process Fig 1: Conceptual Framework of Citation Output

42 30 Library users including the students, lecturers and other staff members depend on these resources to conduct research. One practical use of the resources is citation for the researchers usually cite what they have consulted. Naturally therefore, there is an interaction between the cited sources and the citer and by extension, the author of the cited sources and the citer. The outputs of the researchers are likely to come out in form of theses, books/monographs, journal articles, proceedings, reports, lectures among others which will go back to populate the information sources in the library. Citations are references made to textual work from the perspective of the citing article. This implies that in order to have citations, there must be a cited-citing pair (Leydesdorff, 1998). This cited-citing pair is usually relational since the cited work has an interaction with the citing article. Leydesdorff noted that by adding a dynamic perspective, these relations can be considered as relational operations. This is because the relational operation is recursive since citations may refer to other texts in which citations still refers to other texts. Thus a network of citation relations is built up at each moment in time and this is reproduced over time. The result of this network is a form of citation architecture in the area under investigation. Citations are therefore the result of interaction between network of authors and between networks of their scholarly communications. The cited source is the intellectual work of the cited author which has been referenced by the citing source (the intellectual work of the citing author). This interaction is displayed in such a manner that there is a social network

43 31 between the citing author and the cited author as researchers with interest in a particular area which may result in personal communication or research collaboration. Apart from this social relationship, there is also intellectual relation between the cited source and the citing source which facilitates scholarly communication process globally. Citation analysis is therefore, the study of the citation process in scholarly research in form of citation counting to determine the frequency of citations made to authors, journals, books and other forms of intellectual works so as to evaluate their impacts in scholarship. The analysis of these citations can equally be used to determine the timeliness of cited sources (Ogunronbi, 1988; Nkiko & Adetoro, 2007), authorship pattern in a field (Omekwu & Atinmo, 1998), frequently cited sources (Okiy, 2003; Sam & Tackie, 2008) and even citations to electronic resources. The Wikipedia Free Encyclopedia explains that citation analysis involves examination of an item s referring documents for the purpose of searching for material and analyzing their merits. Citation analysis is associated with Garfield (1955), who proposed a citation index that offered a new approach to subject control of the scientific literature. He was of the view that this new approach would assist in bridging the gap of subject approach to indexing. Garfield argued that there exist some relationships between a particular article and articles in its references. Citation analysis is a method based on the principle that articles citing the same references also have much of their contents in common. This similarity between an article and its references

44 32 is the cornerstone of citation analysis since this relationship is of great interest to scholars. The reference to sources of ideas was developed after the Renaissance. The use of footnotes, a method of making reference to the materials used in the process of research first appeared in the 16th century (White 1985). However, efforts in the use of these footnotes for research analysis were not explored until 1873, when Frank Shepard published his Shepard s Citation. This project was the compilation of a list of American court cases and judgments, with the complete history of each being recorded in a simple code. Under every listing was a record of other court cases that had since referred to it, subsequent judgments that had affected it as well as anything else of potential value to the lawyer (Tonta & Al 2006, Smith 2007). Shepard s Citation, though a novelty and highly valuable was not interested in the publication or citation counts. In 1917, F.J. Cole and Nellie B. Eates produced the first paper that would give a description of literature using publication counts and applying graphic illustrations to them by year and country. This paper, The history of comparative anatomy: a statistical analysis of literature, was considered a classic as a bibliometric research (Hertzel, 2003). The paper reported 6,436 recorded publications dealing with animal anatomy for the period 1543 to 1860 in chronological charts. The fluctuations shown were explained by some minor influences which the authors admitted. The work of E. Wyndham Hulme in 1922 also made significant impact on bibliometric research (Hertzel 2003). Hulme used 13 annual issues of

45 33 International Catalogue of Scientific Literature from 1901 to 1913 to list author entries for various subjects. He further did an analysis which resulted in the tabulation of the number of journals indexed, arranged by countries and listing countries in order of total productivity of the number of journals indexed. However, the first known complete departure from publication counts to analyzing citations was taken by Gross and Gross cited in Hertizel (2003). They tabulated the references in single volume of the Journal of American Chemical Society to find out the journals relevance in the study of chemistry, (Hertizel 2003, Cameron 2005, and Smith 2007). The result according to Hertzel was 3,633 references for 247 different periodicals or journals. The tabulation was made for five year periods from 1871 to 1925, as found in the 1926 volume and trends were traced from the tabulations. Gross in another work cited in Hertzel (2003: 298) had written: Consideration of the method of investigation (citation counts) here employed, will show that we are concerned not merely with the quantity of work published during this period ( ) but in reality we are concerned only with the good work which has survived and which has proved of value to the investigators who followed. The method, therefore, has a distinct advantage over any method which counts pages or number of papers published in various journals for its basis of comparison. The implication of this is that papers which are continually cited are the most valuable for researchers. Another great contributor to the development of bibliometric research is Samuel Bradford. His special article Sources of information on specific subjects (Bradford cited in Hertzel, 2003) described

46 34 how scientific articles on a given topic were being unevenly distributed across the journal literature. This paper was a landmark on the growth of bibliometric studies for it provided it with theoretical framework. Source Data for Citation Analysis Over the years citation analysis has often relied on data from the Institute of Scientific Information (ISI) which was founded by Eugene Garfield in This Institute has three citation indexes namely, Science Citation Index (SCI), Social Science Citation Index (SSCI) and Arts & Humanities Citation Index (A&HCI). These citation indexes provide available information on systematic analysis of impact and influence of scholarly literature. They have long been the most common source data for citation analysis. These indexes serve both as a bibliographic and citation data base which accounts for their uniqueness over other data bases such as Chemical Abstracts, American chemical society s Mathscinet and PSYCINFO by American Psychological Association (Neuhaus & Daniel, 2006). Some of these databases have introduced citation indexing into their data bases and also provided detailed information on download frequencies. According to Neuhaus & Daniel, this additional information assists in the evaluation of the resonance and the impact of the publication in the scientific community. They are discipline-oriented databases and therefore have very narrow coverage. The Institute of Scientific Information with its three broad-based databases provides wider coverage of scholarly publications. The indexes

47 35 cover approximately 6,500 publications for Science Citation Index, (1,900 cover to cover), and 3,300 partial coverage for Social Sciences Citation Index, and 1,100 publications and 700 partial coverage for Arts and Humanities Citation Index, (Neuhaus & Daniel 2006). However, Moed (2005) and Meho (2007) have observed the following limitations on these indexes. They have limited coverage of citation indexes. Secondly, the coverage is restricted only to the journal literature and therefore marginalizes other kinds of scholarly publications such as books, chapters in edited books, conference proceedings, technical reports, patents, etc. This implies that their concentration is mainly on selected scientific fields covering mainly English language titles from North America and Western Europe. The restriction of their coverage only to journal literature is a serious flaw, particularly to the researchers in humanities and the social sciences. The Institute of Scientific Information had all the databases in print form until 1988 when a C-D-ROM version of them was developed (Meho 2007). With the launch of Web of Science by the same Institute in 1997, the databases became online. The Web of Science (multidisciplinary citation index) gathers scholarly contents from institutional and subject-based repositions and adds cited references searching to electronic documents such as conference proceedings, technical reports, preprints, dissertations and other grey literature. Although Thomson Scientific has dominated citation indexing for more than 40 years now, other web-based citation indexing services are now challenging this dominance. Some of these databases, as identified by Moed

48 36 (2005) include Chemical Abstracts, Google Scholar, Google Scopus among others. Chemical Abstracts according to him covers 1,500 scientific journals since 1907 while Google Scholar covers books, journals, conference proceedings, dissertations, technical reports, preprints and post prints. Scopus covers 15,000 peer reviewed journals including coverage of 500 open access journals, 700 conference proceedings and 125 books series (Moed 2005). This data base was launched in 2004 by Elsevier. It indexes more journals than Thomson Scientific; however, it lacks the depth in coverage in years of journals of Thompson Scientific. According to Meho (2007) each of these indexing services produces slightly different results when used for citation analysis. This therefore underscores the need for several citation sources to give the true evaluation of scholarly productivity. The web, through Scopus and Google scholar is really revolutionizing citation research to the extent that Meho has predicted that this may mark the end of about 40 years monopoly held by the US-based firm. With the web-based sources now, a more comprehensive coverage of world literature is now guaranteed, which makes it possible for a new era of citation analysis based on multiple sources. The web has provided several new citation measures and methods, including critical download counts, link analysis, goggles page rank, web citation and h-index recently developed by United States physicist Jorge Horseh. Other popular sources of data for citation analysis are theses and dissertations. Citation analysts rely on them because of their

49 37 comprehensiveness in terms of contents and literature documentation. The earliest citation analysis drawn from students dissertations or theses was that of Fussler, cited in Lawani (1981), which was submitted to the University of Chicago Graduate Library School in Fussler applied the citation analysis in selection of chemistry and physics journals at three different periods (1899, 1919 and 1939) to determine the research literature in the United States then. Another earlier citation analysis relying on theses and dissertations as source data was conducted by Emerson cited in Sam and Tackie (2007). In the study, 23 engineering doctoral theses at the Columbia University between 1950 and 1954 were used to determine the percentage of references available or otherwise in other campus libraries in the university. Since then several citations studies have relied on theses and dissertations and undergraduates projects in generating data for citation analysis. Typical examples are McCain (1981), Zainab & Goi (1997), Gooden (2002), Nkiko & Adetoro (2007), Sam & Tackie (2007). The present study equally generated source data from theses and dissertations since Nigeria lacks the required data base which would have provided the source data for the study. The theses equally provide the most reliable and comprehensive literature search compared to other sources. Patterns of Citations and Information Seeking in Humanities Humanities research deals with the creative, imaginative and subjective world. Contributions to the discipline are usually based on personal analysis and interaction with a wide variety of primary information sources. Sukovic (2008) remarked that this reliance on primary literature sources has been one of

50 38 the basic characteristics of humanities research. Thompson (2002) has also observed that early bibliometric studies in humanities were primarily concerned with making distinctions on the research approach of the humanists and the scientists and developing an accurate picture of the humanists scholarly attitudes. In trying to establish the distinction of the humanists and the scientists as researchers, Peters (1990) applied citation analysis to determine the patterns of citations in sciences, social sciences and humanities and came up with mean citations preferences in these areas. Sciences according to him have 15% citations to books, 84% to serials; social sciences have 58% citations to books, and 37.7% to serials and humanities has 61.4% citations to books and 29.8% serials. With this, he described science as a hard discipline and humanities as a soft discipline and social sciences in between the two. This study corroborates an investigation by Wiberly and Jones (1989) who had earlier observed that unlike the scientist, the humanist prefers monographs to journals. They equally rely heavily on primary information sources such as interviews, archival materials, and visits to places for research. The humanist is more likely to work alone and do more reading than the scientist who spends much of his time in the laboratory working with collaborators. In the same vein, the social scientist spends much time with coinvestigators planning and executing field work, survey and data analysis. These characteristics of humanities research are also reflected in the works of Herubel and Buchman (1994), Delgadillo and Lynch (1999).

51 39 Thompson (2002) has also highlighted the importance of monographs to the humanists. According to her the soul of the humanist lies between the covers of a scholarly monograph. However she regrets that these monographs are gradually disappearing in libraries, since library collections now lay more emphasis on the scientific publications. Regular library budget cuts have always been on the humanities publications and the acquisition of monographs and humanities scholarship has continued to suffer (Sewell, 2001; Case, 1997 and Waters, 2000). Thompson therefore fears that this paucity of materials for humanities research may have contributed to few citations studies in the humanities, unlike what is obtainable in the sciences. Garfield (1983) has also demonstrated that the citations pattern in the humanities is different from that of the sciences. He provided evidence showing that the humanists are more likely to cite monographs than journals and that they cite older publications than the scientists. In his earlier studies, Garfield (1978a and 1978b) showed that the most cited author in the sciences was born in 1899, while similar study in the arts and humanities shows that Homer, one of the most cited authors in humanities, lived around 9 th century. The studies also revealed that approximately 10% of the hundred most cited authors in humanities lived before 14 th century and 60% of them were born before The works of these authors (many of them more than 500 years old) are still heavily cited. This shows that humanities publications do not become obsolete for great scholarship and criticism endures.

52 40 The prominence of books and monographs over journals in humanities research has also been elaborated by East (2006) in his ranking of journals in humanities. He argued that since journals do not have much influence in humanities scholarship, evaluation of humanists through journal ranking is improper. Council for Humanities, Arts and Social Sciences (2005:9) strengthened this argument when it remarked that the standard bibliometric practices do not capture the variety of research outputs (for instance books, documentaries, policy reports) in the humanities and social sciences and are clearly absurd in the creative arts. Similarly, a recent Canadian review of bibliometric practice in the social sciences and humanities cited in East (2006) concludes that the research article clearly does not play as central a role in the social sciences and humanities (SSH) as it does in the natural sciences and engineering (NSE). Several citation analyses and other investigations have demonstrated these inherent differences in research patterns of the humanities and the sciences. In these studies, monographs and books dominate journals, unlike the sciences where the reverse is the case. In a study by Herubel and Buchanan (1994), a citation analysis on Garden History as a sub field of Art history revealed that citations from books and monographs constituted 45.21%, while serials had 11.64%. In another study by Zainab and Goi (1997) using master s and doctoral dissertations as source data, it was revealed that 52% citations were made from books and 23.55% were from journals while, another 9.4% of the citations were from book chapters. Theses, government publications,

53 41 conference papers and newspapers contributed 6.24%, 3.8%, 2.9% and 0.8% respectively. The age of citations used spread over 170 years. The core authors in humanities are usually many, as demonstrated in the study, with English language accounting for about 66% of all the publications. A study by Ileperuma, (2002) categorized information sources of the arts into formal, informal and semi-formal channels. The study surveyed 218 scholars and findings revealed that journals were ranked first as sources of information, followed by monographs and books. More than 90% of the respondents indicated that printed information is the most useful source. This result however contradicts the usual pattern of information seeking behavior of the humanists. A further investigation is required for proper verification. Mendez and Chapman (2006) also conducted a study on the use of scholarly literature on Journal of Latin American History which equally disagrees with Ileperuma s study. Acknowledging that monographs are popular vehicle of scholarly communication in the humanities, the study confirms that there was a decrease of monographs in humanities research between 1985 and 1995 but this increased after the period. It equally reaffirmed that monographs continue to dominate humanities research literature. In the study greater proportion of cited publications were between 11 and 20 years old. However materials over 50 years of age constitute 6% of the citations, showing that age of publication does not matter in humanities. Classics which covers broad areas of language, architecture, other fields of humanities and history, was also investigated (Dabrishus, 2005). Three

54 42 journals dealing with classical studies were selected as source data and the findings showed that classical scholars cite monographs more than periodicals. There was no evidence that classists use any other sources than edited monographs and periodicals. However, the study seems not to be comprehensive as only three journals were used as source data. In an earlier study of English history (Jones, Chapman and Wood, 1972), over 7000 references drawn from a sample of articles to find out the nature of materials the researchers use, also shows a high reliance on non-serial publications. The pattern of age distribution of references suggests that secondary historical works become fairly rapidly obsolete, but the primary sources retain their value over time. Hemming (2008) also examined the information-seeking behavior of the visual artists to determine if a consistent model exists. Findings from the study show a consistency with the information seeking pattern of other humanities fields. Perhaps the most comprehensive citation study in humanities is the work of Knievel and Kellsey (2005), who conducted citation analysis in eight humanities fields, namely, Arts, Classics, History, Linguistics, Literature, Music, Philosophy and Religion. The study analyzed 9,131 citations from journals in each of these fields. Citations from newspapers, reviews and unpublished manuscripts were excluded from the study because of its focus. The results of the study revealed that in all the fields, books and monographs accounted for 74.3%, journal articles accounted for 25.3% while e-resources accounted for only 3%. The average citation per publication was A

55 43 further breakdown of the results by discipline indicates that books/monographs were cited more with literature leading with 83%, followed closely by religion (82.8%), and music (80.6%). Arts, classics and history were the next with 76.9%, 76.8% and 76.4% respectively. Books and monographs are least cited in philosophy (51.4%) and linguistics (60.8%). Journal citations are in reverse as philosophy leads with 48.2% followed by linguistics and finally literature (16.6%). Though this study was conducted in the United States and all the journals used were published there, it has significantly demonstrated that most humanities fields follow the same citation pattern. Another shortcoming of the study, however, is the exclusion of other relevant information sources in the humanities such as manuscripts, newspapers and reviews. Their inclusion would have assisted in establishing the impact of such materials in humanities scholarship. Over the years, the humanists have been described as very aversive to electronic sources. Interestingly however, a recent study by Sukovic (2008), reports evidence that the situation is now changing. There has been an increasing reliance on electronic resources due to democratized access to electronic publications, which the new information technology has provided. For instance Graham (2000) has demonstrated that the humanists now cite electronic publications. Another trend in the use of electronic information sources in humanities research is that though the humanists may still prefer to work alone, they now engage in networking and participation in invisible

56 44 colleges which are important ways of keeping in touch with colleagues. They equally use s to communicate with colleagues. A related work by Delgadillo and Lynch (1999) provides evidence that electronic information sources are significantly used by history graduate students. Another good testimony of the use of e-resources in humanities research is documented in an elaborate study by Toms and O Brien (2008) in an investigation which attempted to understand the needs of the humanists with respect to information and communication technology (ICT) so as to prescribe the design of an e-humanist work bench. A web-based survey comprising over 60 questions obtained data from 169 humanists. Findings revealed that the humanists rely on the availability of expensive, quality-controlled e-resources for their research. The existence of primary sources in digital forms influenced their research pattern. The humanists were unaware of existing tools for conducting text analyses but search engines are gradually replacing the library catalogues as the key access tool for sources. However, research continues to be solitary with little collaboration among scholars. Patterns of Citations in Language and Linguistics The classification of language and linguistics as a field of study has been a contentious issue. Tannen (1989) has argued that linguistics (the science and study of languages) can be scientific, humanistic and aesthetic and therefore has been regarded as the most scientific of the humanities and the most humanistic of the sciences. Though Zwaan and Nederhof (1990) in their

57 45 investigation concluded that theoretical linguistics was showing a publication pattern closer to the social sciences than the literary and historical studies, linguistics generally has been classified in the humanities, (Geogas & Cullars, 2005; Knievel and Kellsey, 2005). Some linguists for example are historians of languages; others work quite closely with philosophers, anthropologists, sociologists and literary artists. This controversy about its classification may have contributed to a very scanty study on the citation patterns of the linguists, as has been reported by Karisson (1994) and Georgas & Cullars (2005). Karisson contends that humanities in general have not been favoured by citation analysis, but linguistics is one of the lowest of the humanities disciplines. His argument is that the culture and language bound fragmentation of the humanities is a serious problem from the perspective of citation analysis as most small cultures and languages are not covered by standard citation indexes such as A&HCI from Institute for Scientific Information (ISI). Another reason for this lack of coverage is that like other humanities fields linguistics relies mainly on books and monographs, which are not in the data base of ISI. For example, in the investigation on the citations of Finnish linguists using a popular linguistics journal Virittaja, Karisson (1994) demonstrated that the first ten most cited works were books. Linguists also cite older materials like other fields in humanities. This is shown from the study where a large proportion of publications cited were published between 1745 and 1910 and more than 60% of them were to books published between 1940 and 1970.

58 46 Paradoxically, a later investigation (Georgas and Cullars, 2005) using Language and Linguistics Behavioural Abstracts (LLBA) and MLA bibliography indicates that the linguistics citation pattern mirrors that of social sciences, with 49.7% citations to books and 42.8% to journal articles. The remaining proportions are to theses and other materials. This is contrary to humanities citations which range from 60% to 80% citations to books and monographs. The ages of citations were between one and fifteen years, though, more than 5% of the citations were published between 1600 and 1960, which suggests that old materials were significantly cited. The study did not identify any set of core journals in the field contrary to Zwaan and Nederhof s (1990) findings. An earlier study (Heinzkill, 1980) equally faults Georgas and Cullars investigation. Using a total of 9556 references from 15 journals in English language and literature, Heinzkill reveals that 75% of the citations in the study were to books and 20% to journals while the remaining 5% were to other materials. Only very few core journals were identified in the study with the publications of Modern Language Association dominating the citations. The investigation also shows that citations in language and literature do not conform to Bradford s law of scattering, but the indication is that language and literature seem to value current materials. From the available literature, language and linguistics have some significant variations in their citation patterns. However, Knievel and Kellsey (2005) and Karisson s (1994) works seem to have put to rest which discipline

59 47 language and linguistics belong. The field may be rightly described as relying heavily on books and monographs, citing older publications, making use of essentially primary information sources and lacking core journals all of which are characteristics of humanities scholarship. Theoretical Framework of Citation Analysis The theory of citation has been of concern to scholars. In 1981 two authors independently raised the issue of theory of citation (Cozzens 1981 and Cronin 1981). Cozzens provided a review of citation theories from a sociological perspective while Cronin based his theory from information retrieval viewpoint, arguing that citation is a function of scientific communication among texts. Cozzens is concerned with the relationships that exist among citing authors who have influenced the development of the discipline. These authors have social links in the history of the field. Lawani (1981) also contends that the sociological basis of citation lies in its use in assessing the quality of individual publications, the eminence of individual researcher and the prestige of research departments or institution. He further identified another sociological basis of citation in connection with research collaboration in a given discipline or institution or geographical area. Cronin however viewed theory of citation as it concerns information retrieval which implies communication among scientific texts. This historical deconstruction of citations to interacting networks of authors and texts provides another attempt of citation theory which Leydesdorff (1998) believes has a position in multi-dimensional space constituted by other citations. Citation analysis studies

60 48 the traces of operation of distributed network system and therefore plays a dual role a link between social and textual information. Citations therefore, link an article to another article and also an author to another one. In relation to this, Garfield (1998) argues that a work cited by another author is regarded to be relevant to the study he is working on. This may not be so when closely looked at, but his argument is based on the fact that a high degree of bibliographic coupling will produce a high degree of relevance. Several bibliometric laws such as Lotka s law of scientific productivity, Bradford s law of scattering and Zipf s law of word occurrence have provided some insight into the theory of citation analysis. At last, Eugene Garfield s law of concentration deepened the theory of citation analysis. However, in this work, Bradford and Garfield s laws were given more detailed review because they are more suitable to the frame work of the study. Bradford s Law of Scattering Samuel Bradford, in 1927 published Bibliography of Cooperation where he indicated his concern for the accelerating accumulation of useful scientific and technical literature, (Hertzel, 2003). Bradford s classic article published his law of scattering where he propounded that if scientific journals are arranged in order of decreasing productivity of articles, in a given subject, they may be divided into a nucleus of periodicals more devoted to the subject and several groups or zones containing the same number of articles as the nucleus,

61 49 when the number of periodicals in the nucleus and succeeding zones will be a 1: n: n 2... Bradford was concerned with duplication of efforts in scholarly publications. In his study, 300 abstracting and indexing journals captured 750,000 articles each year and only 250, 000 of them were dealt with, missing 500,000. A statistical analysis was therefore carried out and the result revealed three rough zones or groupings which he graded as: those producing more than four references a year; those producing more than one but not more than four a year; and those producing one or less than one a year. This statistical analysis is the theoretical basis of his law of scattering. Drott (1981) noted that the law begins with a regularity which is observed in the retrieval or use of published information. This regularity according to him is characterized by both concentration and dispersion of specific items of information over several sources of information. The implication is that for a search on a particular item of information, a large percentage of relevant articles will be concentrated on a small number of journal titles while the rest will be dispersed over a large number of titles. The law explains that a small core of journals has as many papers on a given subject as a much larger number of journals n which again has as many papers on the subject as n2 journals. The law does not however claim that scattering is the same from one subject to another. The belief is that soft disciplines such as humanities field will be more scattered than hard disciplines (Meadow cited in Hjorland, 2007). Some problems have been

62 50 associated with the law. For instance it appears that the dispersion of articles over ranked journals is mathematically regular rank by rank instead of being regular only for groups. Similarly, there is no hint on the formular or the process that create the scattering and how one can determine the size of the core journals. These underlining questions led to several studies on the law to clarify its theoretical implications (Vickery, Kendal, Leimkuhler and Brookes cited in Drott, 1981). Brookes hinging on the work of Vickery derived a formular for the law which did not depend on the grouping of journal titles, thus R(n) = k log (n), where n is the rank of each journal. This formalar is used to calculate the number of articles contributed by a journal at any rank. An important use of the law is that it assists in the determination of the most productive periodical or core journals using rank list which librarians employ in journal subscription or cancellation of subscription. The reader could also identify and locate more useful and resourceful journals for their proposed studies. Hjorland (2007) noted that the law is not only about quantitative issues but also about qualitative matters that the most productive journals on a subject are also the best journals which of course should be selected by libraries and users. This seems controversial since there is interplay of several factors in the quality of journals. For instance, journals which at a time represent a minority view and are thus regarded as of less quality in relation to Bradford s distributions may later represent a majority view. Bradford s law therefore, provided a foundation for studies in the relevance of journal publications through citation studies and also serves as a

63 51 general guideline to librarians in determining the number of core journals in any given field. Garfield s Law of Concentration Eugene Garfield later examined Bradford s law of scattering and with some modifications came up with law of concentration (Garfield 1971). The law is an expansion of Bradford s law. His examination of Institute of Scientific Information (ISI) data bases showed that a list of 1000 journals will contain all the leading journals on any specialty list, and at the same time account for a large percentage of articles published in that field. This implies that what Bradford postulates in a single discipline, Garfield has proposed for the whole body of science. This law holds that a basic list of journals will account for % of all journal references (Garfield 1971:223). Garfield (1976) also developed the theory of constant where he demonstrated that the average number of references cited per paper (R/S) and the utilization factor determine the impact of a journal or author and not the quantity of the literature. He therefore, argues that the size of the literature will determine the number of papers that can exceed a particular citation threshold. In Garfield s constant he postulates that the references processed each year have a ratio of 1:7 constant, with the number of different items cited by those references. He however, remarked that as the number and types of journal covered increased, the ratio changes slightly. Mathematically, Garfield s constant is the product of growth rate and number of references per paper

64 52 divided by the utilization factor (C = GR/U) where G = Growth rate; R = number of references per paper and U = Utilization factor and C = Constant. This means that a field like biochemistry will produce a larger number of papers whose citation frequency exceed 400 or 500 citation reaching to what Garfield (1998) called citation classic. The theories of citation analysis usually lean on several assumptions which are articulated thus: citation of a document implies the use of that document by the authors; it reflects the merits of the documents, citations are made to the best works, a cited document is related in context to the citing document (Aina and Ajiferuke 2002). The validity of these assumptions has been criticized by researchers. For example Ibeun and Madu (2002) identified the following odds against citation analysis. One, there is no assurance that all materials used are cited, which could lead to over-estimation; the problem of self citation could result in citing documents that are irrelevant; and there is the problem of availability and accessibility which implies that an author can only cite documents accessible to him. In relation to this, Meho (2007) has identified several flaws of citation analysis, which limit, to some extent, the validity of the method as a form of research evaluation. The phenomenon of homographs makes it difficult to separate citations of two or more unrelated researchers. There is also the problem of cronyism, where friends or related colleagues reciprocally cite each other to mutually build their citation counts. Another problem is self-citation, where people deliberately cite themselves or journals they are involved with.

65 53 Related to this is the problem of ceremonial citation, where an author may cite an authority in a field without ever having consulted the work itself and finally, there is the problem of negative citation, where incorrect research results were pointed out, yet they are still counted in favour of the document. However proponents of citation analysis argue that though it may have its shortcomings, it is one of the most reliable research methods. Aina and Ajiferuke (2002) therefore, remarked that it is a popular and reliable method in bibliometrics because citations are both unobtrusive and readily available unlike a questionnaire and interviews which rely heavily on the co-operation from respondents before collection of valid data. Moed (2005) and Ezema (2010) have also argued that there is no mechanism for evaluation of research that is flawless; therefore, even with the limitations of citation analysis, it is a most valuable tool in policy studies. Koenig cited in Meho (2007:3) also argues that despite the shortcomings of citation analysis, it provides an objective measure of what is variously termed productivity, significance, quality, utility, influence, effectiveness or impact of scientists and their scholarly products. After all, most citations found in articles and books are useful as they pay homage to pioneers, identify original publications, provide background reading and alert readers about forth-coming works. They also substantiate claims and give credit to related works and through this, provide leads to poorly disseminated, poorly indexed and un-cited publications, (which were previously impractical). It is based on these identified merits of citation analysis that this study applied it in examining the

66 54 characteristics of literature in Nigerian language research. The conceptual framework of the present study is in the perspective of evaluative bibliometrics, which Stevens has noted concerns the analysis of references cited by authors in a field to ascertain the utility value of publications in the field and equally find out relevant information sources and researchers contributions to the field. Language Research in Nigeria Language research in Africa is heavily associated with the sociopolitical development of the continent as a geographical area. According to Emenanjo (2006), the balkanization of Africa by the Berlin conference hampered the natural growth and research in major African languages. Instead of the languages natural to Africans, the people have modern linguistic divisions, namely the Anglophone, Francophone and Lusophone nations. There have been several though uncoordinated attempts in research into African languages between the Berlin conference and independence era in Africa. These significant strides, according to Emenanjo include more insightful description of various aspects of African languages such as typology as seen in Crowder and Yoruba language, Ida Ward on Igbo and Efik languages; the transcription translation and adoption of African folklore in African languages, standardization and stabilization of several orthographies of African languages, the emergence and codification and propagation of standard varieties of major African lingua francas and the beginning of creative writing and literary tradition in African languages. Other notable strides in African language research were the establishment of the Institute of African Languages and

67 55 Culture in 1962 and the enunciation of language policies in many countries among others. Students and researchers in African languages in the pre-independence era were dominated by European traders, missionaries and expeditionary agents. However, post-independence research in African languages began to witness indigenous scholars who have made indelible marks in language research in Africa. Bamgbose (2006) remarks that some of the research conducted in Nigerian languages often have to do with languages that have not been written at all. This is because the foundation of research in any language has to do with the development of orthography of the language. In his review of language research in Nigeria, he emphasized the importance of theoretical research in Nigerian languages, regretting that many young researchers at postgraduate levels usually avoid research in the core areas, particularly syntax According to him, majority of them go into socio-linguistics which they regard as a softer option. He equally identified several factors that impede language research in Nigeria. Such factors include funding and paucity of research materials in libraries, such as books, relevant and current journals and lack of publication outlets. Writing on language research as a social responsibility of Nigerian linguists, Bamgbose, (2006) noted that the linguists should be involved not only in research but also in other issues such as language teaching, devising of orthographies, writing text books or even primers. It is through this way that the linguist can show that he is genuinely concerned with the problems of the

68 56 society in which he lives. He further enumerated areas of language research in the language corpus to include orthography, school texts, data collection, research on endangered languages, terminology, translation and literature. He argues that much of the research should deal with languages that have not been written at all. Therefore, it is not enough to do an elegant phonology without paying some attention to practical orthography so that speakers of the languages can write and read in it. Even the written languages may have inadequate orthography in which case the orthography would need some modifications or harmonization in cases where there is much orthography. Vocabulary expansion is another area of language research which needs urgent attention. The development of technical vocabularies for teaching in schools and Meta languages in various professions and relevant fields is very critical to the overall national development. In the same way, King ei (1999) had earlier posited that research in African languages is critical in propelling other kinds of research because it is mainly through the use of languages that vital information relevant for the conduct of research could be obtained. According to him, given that most of Africa remains rural and illiterate, the issue of which language is to be a medium of research is especially crucial. Verbal communication is often the only option that trained researchers obtain information and attain their research goals. As a result it is basic that the need to speak to people in their own native language cannot be ignored if this necessary information for research must be obtained.

69 57 The issue of indigenous language research in Africa has been of great interest to linguists resulting in a number of national and international conferences. According to Emenanjo, these conferences propelled the formation of Pan African Linguistic Association that was mandated to implement the language plan that was earlier adopted by 44 th ordinary session of OAU Council of Ministers in Addis-Ababa in July Furthermore, the use of indigenous languages in national development was the major theme of the 1998 Johannesburg Conference where discussions focused on Thabo Mbeki s heralded African Renaissance initiative. Perhaps it is because of this that Chumbow (2005) remarked that development of human resources is essential for African socio-economic and technological growth. Chumbow blamed the failure of education policies in Africa on the neglect of indigenous language in pedagogy. Such neglect is contrary to UNESCO s 1953 findings that learning process is more effective using the mother tongue at an early stage (Ansre, 1979; Bamgbose 1991). Accordingly, the vital roles of indigenous languages in education have been underscored in several works (Chumbow, 1990; Afolanya; 1989; Rufai, 2000; Ohiri-Aniche, 2004; Adegoju, 2006; Ohiri- Aniche, 2006). Availability of Nigerian Language Sources in Nigerian Libraries Nigerian languages publications began to circulate in Nigerian libraries when publishing houses got interested in books in the local languages. Although religious writings have existed in Nigerian languages (Isibidi in Efik/Ibibio and the Ajami Script in northern Nigeria), the rise of missionary

70 58 activities in Nigeria resulted in formal publications in Nigerian languages, which made more books and other published materials to be acquired by libraries. Missionary activities in many parts of Nigeria in the 17th century influenced language studies generally and publishing in particular. The aim of missionaries, particularly the CMS which set the pace of publishing books in indigenous languages, was to produce hymnal books, Bible commentaries, catechisms in the language of the people so as to drive the message of the gospel home. Secondly the missionaries used the publications to train converts who would assist them in their missionary work. According to Hair (1967), one of the major results of the scientific study of Nigerian languages is publishing in the indigenous languages and the circulation of these publications in Nigerian libraries. He noted that the earliest collections of Yoruba words were put in print in 1819 by the English diplomatic agent Bowdich. This publication appeared to have introduced the language to other linguists such as Mrs. Hannah Kilham and Rev John Rabon. Mrs. Kilham collected the first Yoruba words to appear in print in 1828 while Roban came out with three Yoruba language publications in 1830, 1831 and Hair (1967) remarked that these formed the first ever published book in a Nigerian language. Later, other publications in Yoruba language began to appear. Notable among them is Bishop Crowder s translation of the Holy Bible into Yoruba language. Henry Townsend also worked seriously in Yoruba publications but most of his works were published anonymously. Hair (1967) observed that his contributions to Yoruba language publications included his editorship of the

71 59 first indigenous periodical Iwe Irohim. These publications which were made available in Nigerian libraries encouraged teaching learning and research in Nigerian languages. While these linguists were making waves in Yoruba language literature, several publishing works had earlier started in Hausa language. Hair (1967) reported that the earliest material collected from the northern part of the country on language was made by Niebhr during his expedition to northern Nigeria. He was a Dane sent by the king of Denmark for an expedition to northern Nigeria. Between 1773 and 1850, other travelers published original materials of the area, most of them providing information about Hausa and Kanuri languages. These travelers included Simon Lucas on behalf of African Association (1789), who collected numerals of Bornu and Katsina and published it in Publication in Igbo language was spearheaded by Bishop Ajayi Crowder. In preparation for Onitsha Mission Station, Crowder had in 1857 printed the first book in Igbo language (the primer in Igbo). This 17 page book gave words and sentences in Igbo and concluded with a few prayers and verses of scripture in translation (Hair 1967). Although there is no record of libraries acquiring these earlier publications because of paucity of libraries then, these publications were later acquired for the promotion of reading in local languages. School and public libraries were the greatest buyers of these materials as they provided the information needs of the library patrons.

72 60 Olaofe s (1990) study nevertheless, offers a more general view of publication in Nigerian languages. Of the seven reputable publishing houses surveyed, 55.88% of academic publications in Nigerian languages are in Yoruba, 18.63% in Hausa and 16.67% in Igbo. At the primary level, nine Nigerian languages have books published in them. They are Yoruba, Hausa, Igbo, Efik, Urhobo, Edo, Ebira, Igala and Tiv. At the secondary level, they came down to five namely; Igbo, Hausa, Yoruba, Efik, Edo. A breakdown of the texts published for primary schools shows that 55.86% are in Yoruba, 16.82% in Hausa and 25.4% in Igbo while the other six languages shared about 9%. The major Nigerian language publishers are Longman (50%) UPL (24%) Macmillan (20%) and Evans (10%) Heinemann and Africana FEP have one each out of the 111 published works; while AUP has 56.35% in Yoruba, Igbo 28.57%, Hausa 11.91%, Efik 2.78% and Edo 0.39%. Observations reveal that libraries particularly public libraries got interested in the acquisition of these publications as an enrichment of their collections with books in local languages and authored by Nigerians. The acquisition of these books was important because the then government encouraged its officers to study the local languages (Tamuno, 1980). A recent study by Abdulkareem (2010) on public libraries in the promotion of indigenous culture reveals that Nigerian language books are available to some extent in Nigerian libraries. The languages involved are mainly the major Nigerian languages, Hausa, Igbo and Yoruba. The degree of availability of these books was however not examined in the study. The

73 61 available books may not have a good coverage of academic books that would support research in Nigerian languages. However a very critical variable in the availability of these publications in Nigerian libraries is the marketing strategy adopted by the publishing houses. According to Orimalade (2005), the acquisition of publications in Nigerian languages to a large extent depends on a cohesive book chain, where there is mutual understanding among the stake-holders in the book industry such as the publishers, the book-sellers and the libraries. He regretted that the greatest challenge the book sellers face is lack of awareness of what has been published in Nigerian languages since publishers catalogues and other mechanisms for promotion are usually not available. Empirical Studies on Citation Analysis Several empirical studies have been conducted on citation analysis within and outside Nigeria. Many of these studies are interested more on the types of sources cited in the area of their investigations, while a few others moved further to find out how much electronic resources are cited. Yet other studies were interested in the timeliness of cited sources or the most frequently cited sources and authorship patterns in the citations. Types of Sources Citation studies that investigated the type of sources cited dominate the empirical literature. These studies were interested on how much

74 62 books/monographs, journals, grey literature and other information sources are cited in a give discipline. These studies reviewed are: Aina (1983) applied citation studies to determine the characteristics of literature used by agriculturalists in Nigeria. Source data were obtained from journals published in Nigeria that reported original articles with evidence of utilization of published sources. The journals were also representative of agriculture as a whole. The source journals contributed a total of 961 citations and findings reveal that serials titles contributed 516 (53%), monographs contributed 427 (44%) while theses contributed 18 (1.8%). It was also discovered from the study that nearly half of the titles used by agriculturists in Nigeria were published in the United States and England. The two countries account for about 50% of the publications. This study relied on journal articles as source data unlike the present study which relied on theses and dissertations. Aina s study also focused on agriculture, unlike the present study which was on languages. Royahi (1997) in another citation analysis reviewed 20 doctoral dissertations in the Arts and Humanities field. These universities in Tehran are Tarbiat Modares University and Islamic Azad University. This study analyzed 11,848 citations, and concluded that in Tarbiat Modares University, 58.6 per cent of all citations were books, 27.5 percent were periodicals, 8.4 per cent dissertations, and 5.5 per cent for other materials. In this case, 46.2 per cent of resources were Persian, 21.6 per cent were English, and 32.2 per cent were Arabic. The dissertations submitted during at Islamic Azad

75 63 University yielded 4,899 citations out of which 41.9 per cent were books, 39.5 per cent periodicals, 9.5 per cent dissertations, and 9.1 per cent were other sources per cent of citations were to Persian resources, 20 per cent to English, and 30 per cent to Arabic. This study was conducted in Iran unlike the present study which has Nigeria as the area of study. Though the study used theses as the source data and focused on humanities field like the present study, the population of the theses is very small compared to the present study. A population of 20 theses is very small for the purpose of generalization. Also, Ajala (2001) employed citation analysis to examine science, engineering and technology students in a Nigerian University using manual investigative technique of citation counts. Results show that students in these disciplines cited more monographs than journals in their final year project reports. Engineering and technology students cited more literature than the basic science students. This study has the flaw of using inexperienced researchers that are yet to be exposed to information literacy and research writing skills which may have resulted in more books and monographs being cited contrary to citation patterns of science and technology where journals are cited more, and therefore, the author recommended library user education to expose the students. Theses and dissertations which were used by the present study may generate more reliable results. Apart from this, the study also has a narrow scope for it used only one university, unlike the present study that studied six universities.

76 64 In a related study, Gooden (2001) conducted citation analysis of dissertation accepted in the Department of Chemistry at the Ohio State University between 1996 and 2000 to determine the materials used. The 30 dissertations studied generated a total of 3,704 citations. Type of materials cited, currency of literature and dissertation topics were all examined. The results show that journal articles were cited more frequently than monographs. Journal articles constitute 85.8% of the citations while 8.4% were monographs, dissertations, theses, conference proceedings, newspapers and annual reports constitutes 2.2% and miscellaneous publications are 3.6%. The study concludes that the revelations will help to improve the collection development of Ohio State University Library. Though this study used similar source data (theses) like the present study, it suffers from inadequate population of the theses and may be deceptive to use such a population to generalize. The number of university used in the study may not be adequate for a study like this. Otubelu (2003) also conducted citation analysis of post graduate research in life sciences in University of Nigeria using 128 research reports. The findings show that the preferred literature are journals, which constituted 73.6% of the citations, followed by books with 15.4%, grey literature has 9.6% and unpublished sources 1.4%. She equally observed that there are variations in the patterns of citations among the departments studied. In addition to this, most of the heavily cited materials were foreign. Though this study has similar source data like the present one, the focus of the study is in the sciences unlike

77 65 the present study which is interested in the languages. A major flaw of Otubelu s study is that the number of university surveyed is inadequate for a reliable generalization. Ezeani (2005) applied citation analysis to examine the use of grey literature in Agricultural research of University of Nigeria, Nsukka. Doctoral theses submitted to the Faculty of Agriculture from 2000 to 2005 were analyzed and the result revealed that grey literature was not fully utilized in agricultural research in the university. Of the twenty doctoral theses used for the study only 19.4% of the citations harvested were grey literature. This study has similarity with the present study on the method since the same kind of source data were used but like some other studies reviewed it equally suffers from the same problem of inadequate theses population and number of universities surveyed which may affect generalization. Also unlike the present study, the study is only interested in finding out how much grey literature was used in agricultural research. Kellsey (2005) also conducted a citation analysis in eight humanities fields, namely, Arts, Classics, History, Linguistics, Literature, Music, Philosophy and Religion. Source data were taken from journals and 9,131 citations were analyzed. The findings showed that in all the fields, books and monographs accounted for 74.3% of the whole citations, journal articles accounted for 25.3% while e-resources accounted for only 0.3%. `Although this study has a focus in humanities, the method is not similar with the present study. Several subject areas were studied which might make it difficult to

78 66 understand the citation pattern of a particular field among the eight disciplines. The use of journals as source is another area of difference between the study and the present one. Citations in the theses have been found to be more comprehensive than the journal and that may affect the result of the study. Carlson (2006) examined the citation behaviour of undergraduate students using data collected from 583 references from student s research papers. The study analyzed the effect on the class years of the student, the academic discipline of the course, and the level the course have on the type of and the mean number of sources cited by undergraduates. The results of the study showed that the mean number of sources cited was Overall, books made up 50 percent of the total number of citations, reference books accounted for 3 percent, journal articles for 19 percent, magazine articles for 7 percent, newspaper articles for 2 percent while web citation was 16 percent. Other sources accounted for 3 percent and unknown sources for 1 percent. The study concludes that all the three variables examined had some significant effects on student s citation behaviour. However a major problem of this study is that it relied on undergraduate research projects as source data. Their citations would not be comprehensive enough as the authors of the research projects are yet to be grounded the necessary research skills. Also using theses references as source data, Tonta and Al (2006) did a citation analysis of 100 theses and dissertations of library and information science and discovered that 50% of the citations are from monographs 42% from journals, 3% from dissertations, 3% from electronic publications and 2%

79 67 from other sources. The findings also showed that more than half of the cited sources are in English language (55%) the rest mainly are in Turkish language. This study like the present study obtained source data from theses but the focus of the study is on Library and Information Science. The population of the theses seems comprehensive but only one university was used for the study unlike the present one that studied six universities. The result may not be comprehensive enough for generalization. Another related work conducted by Zandian et al. (2007) analyzed the PhD Arts and Humanities dissertations defended in Tarbiat Modares University during They reported print materials citations at 96 per cent, and book citations as percent. Persian resources received 60.1 per cent of total citations followed by English 19.2 per cent and Arabic 20.6 per cent. They concluded that because Arts and Humanities courses are book based disciplines that in Iran there were enough resources in Persian for doctoral students. Swanepoel (2008) applied it to determine the nature and extent of information sources used by postgraduate students of health and biomedical sciences with special reference to the use of humanities publications. This study covered the period from 2004 to Generating source data from theses and dissertations and covering a period of four years, the findings indicate that the students make use of limited information from humanities in preparing their theses despite their availability. Only about 3% of the citations are from social science sources. This study suffers from narrow scope that would be used for generalization. Similarly, the period covered is small unlike

80 68 the present study. However the result has given insight on the citation patterns of health and biomedical researchers. Ezema and Asogwa (2012) conducted a citation study using two main line journals in animal health and production namely; Nigerian Journal of Animal Production and Nigerian Veterinary Journal (for seven years) were used for the study. Citations in the journals yielded a total of 8328 out of which 58% of them are journal citations and books and monographs accounted for 24%. The findings also revealed that more than 31% of the cited materials were published between 1991 and 2000 while 25.9% were published from 1981 to Materials of 2001 to date constituted 20.2 percent. There was also a very low citation to electronic resources which would have provided citations to more current materials. Similarly authorship pattern revealed that multiple authors dominate the cited sources. About 21 journals were identified as the most frequently cited journals while most frequently researched animal is poultry. This study though has wide coverage in terms of period like the present study, but has its focus in Agriculture while the present study is interested in Nigerian languages. Also the source data were from two journals only which may not provide reliable results for generalization. A very recent citation analysis was conducted by Riahinia, (2010) to examine the citation pattern of Masters Dissertations of library and information science graduate students using 259 dissertations from five universities in Tehran submitted between 2003 and The result yielded a total of citations with a total mean citation per dissertation of Findings revealed

81 69 that 92.6 percent of the citations were to print resources. Citations to English language resources were percent. Citations to books were 5.3 percent as against 29.7 percent for the journals while citations to dissertations accounted for 3.7 percent. Other resources accounted for 61.2 percent. Though source data for the citation analysis is theses like the present study and the population of the theses is quite encouraging, the study has its focus on Library and Information Science unlike the present study that has interest in the study of Nigerian languages. Similarly, the study did not go further to identify cited sources and their availability in libraries. This is a short coming in this study for it did not provide the researchers and libraries the opportunity of finding out whether libraries are meeting the research needs of the scholars as the present study has done. Electronic Resources Citation analysis has also been applied to investigate the rate at which researchers use electronic resources. For example, Harter (1996) conducted a citation analysis to determine the impact of electronic journals on scholarly communication. Thirty-nine scholarly journals that began electronic publications not later than 1993 were selected for the study. The result of this study shows that great majority of scholarly e-journals have had essentially no impact on scholarly communication in their respective fields. Only eight of the 39 e-journals (20.5%) have been cited ten or more times over their lifetimes. The findings also reveal that the top eight journals are from the sciences, medicine and social sciences. Harter s investigation has the limitation of

82 70 considering only electronic journals which many researchers do not publish in because of their questionable peer review mechanism. Another limitation of this study is that access to these journals is restricted to those who are connected to the Internet. Unlike the present study, Harter failed to determine the availability of these journals to the database of the researchers who are stake holders in the study. Another study was conducted by Zhang (1998) to investigate the impact of internet-based electronic resources on formal scholarly communication in the area of library and information science. The result of the study reveals that of the 1,175 citations analyzed, 7.49% articles have electronic references while the average e-references are 3.78%. The investigation included articles with pointers like foot notes and the percentage went up 8.94%. The study concludes that the impact of e-citation is very low when compared with print types. Though there is increase over the years in the e-resources, it is not significant. It was also observed that e-journals are more likely to cite e- resources than the print journal articles. This study suffers from the same limitation with that of Harter for the scope of the study is only on electronic citations. Apart from this, the focus of the study is Library and Information Science unlike the present one. The study also failed to look at rate of availability of the cited sources as the present study did. In a similar vein, Herring (2002) did a study to investigate the use of electronic resources in scholarly electronic journals through citation analysis. Scholarly electronic journals published in were used for the study.

83 71 The analysis focused on the extent to which scholars were using electronic resources that are being referenced. The findings of the study show that more than half of the articles studied included electronic references and that 16% of the total references were to electronic resources. In the study 345 (8%) of the electronic resources were to such unpublished or ephemeral resources and over one quarter (27%) of the electronic resources cited in the study were categorized as interdisciplinary or outside the discipline, area of the journals or the author s affiliations. This study took source data from journals published within two years unlike the present study which collected data from theses within ten years period. The period involved was not enough to make reliable conclusions on the use of electronic resources among researchers in the field. It did not also go into the details of finding out the availability of these cited materials to the researchers like the present study did. Jan (2009) also conducted a citation analysis of all the journal articles published in the Library Trends from A total of 593 articles were published in the journal during 14 years. Highest number (52) of articles was published in The Journal contained 15,662 references for the study period of which 13,783 were print citations and 1,879 are electronic citations. Every issue published approximately 11 articles, and each article has an average of 23.2 print references and 3.1 electronic citations. It was found that % print books were consulted by the authors and 0% electronic books were accessed. Authors have consulted 44.04% print journals as against 11.82% electronic journals. Though the period covered by this study is very

84 72 comprehensive, the source data were drawn from journals citations unlike the present study that generated its source data from theses. Similarly, the study was more interested in the sources cited and no attention was paid to frequently cited sources and their availability of cited sources in the libraries. This does not give clear picture to librarians and researchers on whether the libraries are supporting research activities. Frequently Cited Sources Udofia (1997) conducted citation analysis for selection of principal veterinary medical journals. Using the data base of 105 journals for a period of five years, the result of the study shows that Bulletin of Animal Health and Production in Africa is the most cited journal with 605 citations representing 66.2% of the entire citations. Udofia s work is comprehensive enough in terms of years covered but has the limitation of using only one journal in conducting the study. The result might not be reliable enough for generalization. In the same way the study failed to identify frequently cited journals that would be useful for library subscription. Another citation analysis is that of Okiy (2003) who examined educational dissertations at Delta State University using 70 postgraduate research reports submitted from 1992 to These dissertations yielded 4,012 citations and findings reveal that books (60.3%) were cited more than other forms of library materials. A total of 12 out of the 18 most cited journals are from the United States. Though this study has similar approach in terms of

85 73 source data, the number of university used is not adequate. Therefore it might be deceptive to use the result for generalization. Like the present study, Okiy s study identified most frequently cited journals but failed to check for the availability of the journals in the university library to ascertain whether the library collections is adequate enough to support research in the field investigated. Another study by (Dabrishus, 2005) analyzed the citation pattern of Classics. Three journals were used as source data and the result revealed that classical scholars cite more monographs than the periodicals. In the study, a total of 63 most frequently cited journals in Classics were identified. Each of these journals has two or more citations. Journals with one citation which were not included in the most frequently cited were 55 in number. Ranking first in the list of the most frequently cited journals is Classical Quarterly with 32 citations. The study like the present one focused on humanities field but differs in the source data. It used data taken from journals while the data for the present study are from theses. This might have affected the number of citations to the identified frequently cited journals. Again the study failed to check the availability of frequently cited sources to determine whether the libraries are meeting the research needs of the users. LaBonte (2005) used citation analysis to determine if the Sciences and Engineering Library at the University of California is meeting the information needs of 60 faculty members at the new California Nanosystems Institute. The latest three publications of each faculty members (published within the last two

86 74 years) were analyzed in two ways using the Science Citation Index: first, the journals they were published in were analyzed, and then the there was an analysis of journals where cited articles were published. The results of the investigation show that a total of 163 source articles were published in 81 journals. A breakdown reveals that 55 journals (67.9%) contained one article each, and one journal, Applied Physics Letters contained 19 articles, when grouped into 3 zone based on journal most frequently published in, 6 journals occur in the 1 st zone with at least 6-19 article published in each of them, 19 titles occur in zone 2 with 2-5 articles and 54 titles in zone 3 all with one article, of the 4,023 citations analyzed from 643 journals of which 318 journals (49.5%) occurred in one citation, one journal accounted for 267 citations (6.6%). The mean number of citations per bibliography was of which 90.2 percent were journals, 3.4 percent were conference proceedings and 6.4 percent were other formats such as books, patents, and personal communication. The study concludes that the library is adequately meeting the needs of the faculty members. The source data unlike the present study were from publications of Faculty staff which may not be adequate enough for generalization. Another limitation of the study is that the number of years investigated may not give reliable result for drawing conclusions. Citation analysis was equally applied in Ghana (Sam & Tackie, 2007) to investigate the research pattern of library and information science graduates. A total of 67 dissertations yielded 2,212 citations averaging 33 citations per dissertation. Findings indicate that books and monographs (43.8%) were cited

87 75 more frequently, followed by journals (24.9%). Other materials accounted for the remaining percent. Fifteen core journals were identified and they generated more than 50% of journal citations. Journals published between 1990 and 2000 accounted for nearly 60% of journal citations. This study like the present one drew its data from theses, but has a narrow scope since only one university was covered which may not be adequate for a reliable conclusion. Beside this, the study is more interested in identifying the most frequently cited journals without investigating the availability of these sources in the libraries like the present study did. Authorship Pattern The authorship patterns and author collaborative research has been studied empirically by a number of researchers through citation analysis. For example, Vimala and Reddy (1996) did a study on the authorship pattern and collaborative research in zoology with a sample of 19,323 journal citations figured in the theses on Zoology accepted for the award of the doctoral degree by Sri Venkateswara University, Tirupati, India. Findings from the study revealed that although multiple authorships were dominant, solo research also existed. The proportion of single authored papers has shown a declining trend during the period Observation showed that the proportion of single authorship would likely be insignificant after the year The degree of collaboration in research was 0.75 in Zoology as a whole. Though this study like the present one got source data from theses, it suffers from narrow scope for only one university was used for the study unlike the present study that used

88 76 six universities. The result of the study may not be adequate for generalization because of this narrow scope. Omekwu and Atinmo (1998) conducted a citation analysis on patterns of agricultural communication research at the University of Ibadan. In the study bibliographic references of 37 theses purposefully selected were analyzed and authors of cited references were classified into two: Nigerian and non Nigerian authors. Journals cited were grouped into three: national, regional and international journals. The citation analysis indicated a very significant difference in the citation of national and non national authors. Non Nigerian authors were cited more than Nigerian authors. The study also reveals no significant difference between the citation of regional and international sources but a very significant difference when compared with the citation of national journal sources. Though the source data for this study is similar to the present study, it has narrow scope since only one university was covered. Another limitation of the study is that the number of theses analyzed may not be sufficient for the purpose of generalization. Another study by Pradhan, Panda and Chandrakar (2011) examined the trends and authorship pattern and author collaborative research in Indian chemistry literature with sample 53,977 articles downloaded from SCI- Expanded database in Web of Science during the period The average number of authors per article is 3.55 %. In the study the degree of collaboration (C) during the overall 10 years ( ) is 0.03, but the year wise degree of collaboration is almost same in all the years of mean value 0.97.

89 77 In the period under review, the multi-authorship articles were higher than single authorship. The study found out that the researchers in chemistry were keen towards team research or group research rather than solo research. This study though has a very wide scope in terms of the periods covered but relied on electronic source data which are usually skewed against authors who publish their works in journals that were not indexed by SCI Web of Science. This may affect the result of the data in terms of generalization. It is quite unlike the present study which relies on theses as the source data. Kumar and Kumar (2011) also analyzed 8,093 citations given in the Journal of Oilseed Research (JOR) published during 1993 to Out of these citations 5642 were given in main articles and 2551 in communications of the JOR. The paper covered the analyses of authorship patterns of citations in line with calculation of collaboration coefficient. Findings revealed that collaboration coefficient is very high during all the years ranging from 0.76 to It is equally revealed that the higher the value of collaboration coefficient, the less the number of single authored papers. This trend showed that collaboration among scientists was very high. Only 18.9% references were single authored while 81.1% papers referred were written jointly. The table also revealed that 39.3% papers referred were two authored papers. The paper concluded that only 20 core periodicals could cover more than 50% references and also indicated that collaborative research was new trend in oil seeds research. This work though comprehensive in terms of period of coverage has narrow scope. Only one journal was analyzed which may not be a good

90 78 representation of the field. The focus of the study is different from the present study which was on Nigerian languages. Timeliness of Cited Sources A citation study on the timeliness of cited sources was done by Ogunronbi s (1988) he examined the citations of Ilorin Journal of Education. In the study, 80 articles were drawn from five issues of the journal which yielded 965 citations. The findings showed that books and monographs constituted 56.4% of the citations, journal articles 25.9% while conference papers, government documents and unpublished dissertations shared the remaining proportion. Publications that were 10 to 15 years old dominated the citations. However, materials published between 1960 and 1979 accounted for 78.4% of the entire citations. Only 20% of the journals cited were published in Nigeria. A major limitation of this study was that it had a narrow scope in terms of population of source data and this might seriously affect the result of the study. Nkiko and Adetoro (2007) also applied citation analysis in examining the citation pattern of undergraduate students at Covenant University. The study found out that citations to books were 53.3% and that of journals 25.1%. The documents cited were current since more than 60% of citations were not up to ten year old. School of Social Sciences generated more than 53% of the citations. Though, the source data of this study are similar with that of the present study, the use of undergraduate research project might not provide

91 79 adequate result since their research experiences were still in infancy stage. Theses would likely provide more reliable result that can be used for generalization. Apart from this the coverage of the study was narrow since only one university was investigated unlike the present study that investigated up to six universities. Availability of Information Sources Some empirical evidences on availability of information sources in Nigerian university libraries were equally reviewed. Many of them revealed the poor state of Nigerian university libraries in terms of information availability. A study by Canick (2002) investigated a group of recent law articles to determine the proportion of cited resources that are easily available online. Searches were conducted in databases such as LEXIS and Westlaw, and on the Internet, for full-text versions of every source cited in seven articles. The results have been broken down into 13 categories, including federal cases, books, foreign law, and legal periodicals pre Overall, the study found out that 77% of the 1,984 citations in the articles reviewed were available online. The article was concluded with a general discussion of why the percentage was likely to increase in the future. This study unlike the present study was interested in online journal availability and the focus was law sources. The study however provided evidence that online sources is likely to be more reliable in terms of availability than print sources.

92 80 Okafor and Ukwoma (2007) investigated the availability and use of information sources in the library by faculty members in Science and Engineering. Using a descriptive survey design and a population of 291 academics in universities in Southern Nigeria, they found out that the information sources required by the faculty members are not often available in the university libraries. Sources that are usually available in the libraries according to the findings were books theses and dissertations while scientific and technical journals were not available. The method applied in this study is quite interesting but it is different from the present study which used what the researchers frequently cite to determine what the researchers required instead of using questionnaire which may not be reliable enough. Apart from this, the study was applied in Science and Technology field as against the language discipline of the present study. A study by Squires, Moore and Keesee (2007) used a set of 400 citations as a sample of what might be valuable to students, researchers, faculty and staff of University of North Carolina to determine the availability of these sources online. Findings revealed that 78% of these sources are available online. Only 88 of these sources were not found online. The reason for this unavailability was that the library did not have access license to the electronic journals within the period. This study was mainly concerned with online availability of the sources unlike the present study that investigated the physical availability of the sources. Besides this, the study had a narrow scope that would be inadequate for generalization since only one university was used for the study.

93 81 Another study by Ugah (2008) on the availability and accessibility of information sources in Michael Okpara University of Agriculture, using 168 respondents made up of students and faculty members revealed that more than 72% of the respondents indicated that information sources they required were not available. However analysis conducted found out that there was no relationship between availability of the information sources and the use of library services. While this study has been able to establish availability problems in Nigerian university libraries, it has a narrow scope by using only one university for the study. This may not be adequate for generalization. The present study has a wider scope by using six university scattered within the country. In another study conducted by Baldwin (2009) Thomson Reuters database Web of Science was used to collect the citations of engineering faculty publications from five colleges. The availability of each citation was determined using Google Scholar as well as the Registry of Open Access Repositories and OAI protocol. Result indicated that 13% of the citations in mechanical and nearly twice that figure for chemical engineering were available. This study though has a large scope, is concerned with online sources unlike the present study that is more interested in print sources. Apart from this, the study like the other study reviewed has its focus on engineering while the present study focuses on Nigerian languages. A very recent study (Okafor & Ukwoma, 2011) investigated the availability of online and printed journals where academics in Science and

94 82 Engineering of universities in Southern Nigeria publish their articles using descriptive survey method. Questionnaires addressing the availability rate and currency of 57 journals identified were used. Findings from the study reveal that University of Lagos has the highest availability rate of 63% of the online journals and 46% of the printed journals followed by Niger Delta University with 43.9% of the online journals and 75.9% of the printed ones. Other universities do not have the journals online while none has them on CD-Rom. The researchers recommended regular acquisition of these journals. Like the present study, this has interest in the availability of information sources in Nigerian university libraries. However the focus of the study is in Science and Engineering discipline while the present study has its focus on languages. Also the methodology adopted for the study is questionnaire unlike the present study which adopted citation analysis to identify the frequently cited journals. Interestingly however, the study has shown that availability of resources is really a source of concern in Nigeria university libraries. Summary of Literature Review Citation analysis is explained as an aspect of bibliometric studies which is concerned with the examination of citations made to a work as a way of evaluation of publications, authors, institutions or even a geographical entity. The theoretical basis of citation analysis is anchored more from Bradford s law of scattering and Garfield s law of concentration developed from Bradford s law. Bradford was concerned more with the quantity of publication output while Garfield took interest on the quality of scholarly productivity.

95 83 Reference to sources of information is not entirely a new development. The use of foot notes in referring to consulted documents first appeared in the 16 th century. Shepard s citation a compilation of a list of American court cases and judgments was established in In citation analysis the commonest sources of data have been the citation indexes published annually by Institute of Scientific Information (ISI). However these citation indexes have a number of flaws ranging from limited coverage of journals, Euro- American biases to its selection of mainly science-based journals. Citation analysis is a very useful instrument in the evaluation of research out puts. It can be applied in the investigation of coverage of library acquisition, historical development of a discipline and those behind it, development of scientific theory, and evaluation of research out puts of authors, institutions or geographical regions. Information seeking behaviour of the humanities and the citation patterns differ greatly from the scientists. The humanists rely heavily on books and monographs and cite older materials than the scientists. They also value primary sources such as interviews, archival materials and visit to places. Investigations have revealed that the citation patterns of the linguists and their information seeking behaviour follow that of the humanities rather than that of the social sciences as many had believed. Several citation analyses in language and linguistics show that the linguists like other humanities scholars rely heavily on books and monographs, cite older materials and more likely to work alone.

96 84 So many empirical works on citation analysis are identified both locally and internationally in fields of social sciences and science and technologies mainly. These works were reviewed under these sub headings: types of cited sources, electronic resources, frequently cited sources, authorship pattern and timeliness of cited sources. Some of them used references to theses and dissertations as their source data. Other empirical works obtained their source data from mainly print and electronic journals. From the resources available to this researcher, there seems to be no study that has been conducted on the citation and availability Nigerian language sources. This has created a gap in scholarship which the present study attempted to fill.

97 85 CHAPTER THREE RESEARCH METHODS This chapter discusses the detailed procedures which were used in the study as follows: research design, area of the study, population of the study, sample and sampling techniques, instrument for data collection validation of instrument, method of data collection and method of data analysis. Design of the Study The design of the study was a descriptive survey design. This design was considered appropriate for this work because there was the need to collect statistical data for accurate description of the conditions as they exist. Citation analysis was applied in analyzing the documents (theses) which were used as the source data. Citation analysis was used for the study because it is a method that will generate actual data on the information required for the study. Aina & Ajiferuke (2002) noted that citation analysis is a popular and reliable research method because citations are both unobtrusive and readily available unlike questionnaire and interview which rely heavily on the co-operation from the respondents before collecting the valid data. A check of frequently cited books and journals was conducted in the university libraries to determine the availability of the sources.

98 86 Area of the Study The area of the study is Nigeria and specifically, Nigerian universities. These institutions are spread in the Western, Eastern and Northern parts of Nigeria. The universities are: Ahmadu Bello University, Zaria; University of Benin; University of Ibadan; University of Lagos; University of Nigeria, Nsukka and Obafemi Awolowo University, Ife. These first generation universities were chosen because of their pioneering roles in Nigerian language research. It is assumed that the universities with their experiences in the study of Nigerian languages would have generated sizeable numbers of theses in the departments of Nigerian languages. Population of the Study The population of the study comprised all the completed theses submitted to the departments of Linguistics and Nigerian languages in the six first generation universities from which have been acquired by the libraries of the universities. From the records of the Acquisition Departments of these university libraries the population of the theses was 508 summarized as follows. University of Nigeria, Nsukka (98), University of Ibadan (224), University of Benin (2), Ahmadu Bello University, Zaria (44), University of Lagos (67) and Obafemi Awolowo University, Ife (73).

99 87 The period covered was ten years, and this period was chosen because it was considered adequate in examining the most recent theses in Nigerian languages so as to ascertain the present trend in terms of information availability and use in the field which will serve as a guide for library acquisition for Nigerian languages. Ten year period (a decade) is equally remarkable to determine the growth of literature in a given field in terms of the visibility and citation of the literature. Sample and Sampling Technique The sampling technique used for the study was incidental sampling technique. This was applied in order to sample all the available theses submitted to the departments of Nigerian languages and Linguistics from 1998 to 2007 since not all the theses that formed the population of the study were available during the sampling. Sample Size With the incidental sampling of available theses, the sample size of the study was 449 theses as indicated below: University of Nigeria Nsukka (87); University of Ibadan (208); University of Benin (2); Ahmadu Bello University Zaria (36); University of Lagos (54) and Obafemi Awolowo University Ife (62).z

100 88 Instrument for Data Collection The instrument for data collection were the citations documented as references in the theses and dissertations which were analyzed and the researcher s two sets of self designed document analysis guide entitled Citation Analysis of Nigerian Language Research Document Analysis Guide (CANLRDAG) and Availability of Frequently Cited Journals in Nigerian Languages Document Analysis Guide (AFCJNLDAG). The first document analysis guide consists of eight items on number of citations from different sources such as books, journals, government publications theses, electronic resources, conference/seminar papers, news publications, reports, lecture notes, inaugural lectures, mimeographs, interviews, diary and radio/televisions. Other items are number of theses by Nigerian languages, number of electronic citations, languages of cited sources, the timelines of cited sources and frequently cited journals in Nigerian language research, the frequently cited authors and the frequently cited books. The second guide has six items on the six university libraries to be studied. Each of the items was used to obtain data on frequently cited sources (journals and books) and their availability in the university libraries (see Appendices III and IV pp ). Validation of the Instrument The instruments were subjected to two forms of validations namely; face validity and content validity. Three experts in the field of Library and Information Science, Education and Statistics were requested to thoroughly

101 89 examine the items in relation to content relevance, clarity and relevance of items in the guide. Through this process useful corrections were made by these experts. The language in which the theses were written was included as item number three in the first part of the document analysis guide. Similarly in other to be exhaustive, others specify was added at the end of 1, 2, 3 and 5 in the first part of the guide. Also, in item number 5 in the first part of the instrument specific periods of time were removed so as to determine the number of cited sources per year instead of per period. In addition to these, the second part of the instrument on the availability of frequently cited journals, number of issues was replaced with number of issues cited while latest issues was replaced by number of issues available Procedure for Data Collection The researcher personally visited the libraries in the institutions selected for the study. Photocopies of title and reference pages of the theses and dissertations that were sampled were made. The researcher s self designed document analysis guide was used to record the citations made in the theses and dissertations in relation to the research questions that guided the study. The citations embodied in the theses were manually counted for the analysis. The researcher equally used the second document analysis guide to find out whether the frequently cited journals and books were available in the university libraries studied.

102 90 Method of Data Analysis The data collected were analyzed using descriptive statistics of frequency and percentages and presented in tables and charts. The two null hypotheses were analyzed using Pearson s correlation coefficient and the significant of the relationship was tested using t-test. Decision Rule on Selection of Most Frequently Cited Sources The decision rule to determine the most frequently cited items (journals, books and authors) was reached by using the mean citations for the most frequently cited item in question as the cut off marks. For example the total number of journals cited in the study was 98 giving a mean citation of 50. Therefore, the cut off mark for the most frequently cited journals becomes 50 citations.

103 91 CHAPTER FOUR RESULTS The data collected for this study are presented and analyzed in this chapter. The research questions formulated guided the presentation. Research Question 1: What types of sources are cited in Nigerian language research? Table 1: Type of Cited Sources Types of Sources ABU UNILA OAU UI UNN BEN Total % G Books Journals E-Resources Theses Govt. Pub Conference/Semin , News Pub Reports Lecture Notes Inaugural lectures Mimeograph Interview Dairy Radio/TV TOTAL 1,550 2,728 6,333 11,933 5, , (5.6%) (9.9%) (22.%) (43.1%) (18.4%) (0.21%) 99.9%

104 92 Lecture Notes, 0.9 Govt. Pub, 0.7 Reports, 0.4 Conference/Semina r, 4.9 E-Resources, 0.6 Theses, 10.8 Journals, 16.2 News Pub., 0.9 PERCENTAGE Inaugural Interview, lectures, Mimeograph, 0.1 Radio/TV, 0.1 Books, 63.6 Dairy, 0.01 Books Journals E-Resources Theses Govt. Pub Conference/Seminar News Pub. Reports Lecture Notes Inaugural lectures Mimeograph Interview Fig 2: Pie chart showing types of cited sources. Table 1 shows the type of information sources used by Nigerian language researchers. It could be observed from table 1 and fig 2 that books and monographs are the dominant sources of information for the researchers. These account for about 17,616 (63.6%) of the entire citations. Journals are second with a total citation of 4,521 or 16.3% of the citations. Theses equally constitute significant percentage in the citations with 2,977 or 10.8%. This is closely followed by conference/seminar papers with a total citation of 1,347 or 4.9%. Citations to news publications are 240 (0.9%) while citations to government publications are 184 (0.7%). It is however note worthy to observe from the table that citations to electronic publications are very low 174 citation (0.6 %.) Lecture notes, inaugural lectures and reports have 244 (0.9%), 49 (0.2%) and 118 (0.4%) respectively while citations to oral interviews are 142 (0.5%). The least citations are to diaries with only 3 citations followed by

105 93 radio/ televisions with 33 citations or 0.1% of the total. University of Ibadan has the highest contributions to the citation with 43.1% of the citations followed by Obafemi Awolowo University with 22.9% of the citations. University of Benin has the least citation with only 58 citations. Research Question 2: What is the language spread of the theses by Nigerian languages? Table 2: Language Spread of Theses LANG ABU UNILAG OAU UI UNN BENIN TOTAL % Hausa Igbo Yoruba Edo Efik Fulfulde Urhobo Nupe Ibibio Isoko Tiv Ngas TOTAL 36 (8.0%) 54 (12.0%) 62 (13.8%) 208 (46.3%) 87 (19.4%) 2 (0.45%) % PERCENT PERCENT Fig. 3: Bar chart showing language spread of theses Table 2 shows the subject spread of theses by Nigerian languages. The table indicates that the most studied Nigerian language is Yoruba which

106 94 generated 256 theses or 57% of the total number of theses. This is followed by Igbo language with 119 theses or 26.5% of the total while Hausa language has 45 theses (10%). Ibibio has 10 theses or 2.2% of the total number of theses while Edo has 7 theses representing 1.6% of the entire theses studied. Urhobo generated 5 theses representing 1.2% while Fulfulde and Tiv have 2 theses each. The least number of theses are generated by Nupe, Isoko and Ngas with 1 thesis each. University of Ibadan contributed the highest number of theses with 46.3%, followed by University of Nigeria Nsukka with 19.4% of the theses. The least number of theses came from University of Benin with only 0.45% of the theses. Research Question 3: What are the languages of the theses? Table 3: Language of the Theses LANG ABU UNILAG OAU UI UNN BENIN TOTAL Percent English Hausa Igbo Yoruba Edo Efik TOTAL % % % % % % % 99.8%

107 95 Percent Hausa 3% Igbo 0% Yoruba 16% Edo 0% Efik 0% English 81% English Hausa Igbo Yoruba Edo Efik Fig. 4: Pie chart showing the languages in which the theses were written Table 3 presents the languages in which the theses studied were written. It is observed from the table that all the theses were written in just three languages with English language dominating. Of the entire 449 theses, 363 or 81% were written in English. Yoruba language is the next with 71 theses or 15.8% of the theses, while Hausa language was used in writing 15 of the theses representing 3.2 % of the whole. It can be seen from the table that none of the theses under study were written in Igbo language.

108 96 Research Question 4: What is the quantity of electronic sources cited in the theses? Table 4: Citations to e-resources YEAR ABU UNILG OAU UI UNN BENIN TOTAL Percent TOTAL 3 (1.7%) 7 (4.0%) 45 (26.0%) 85 (49.1%) 33 (19.0%) - (0%) % 99.9% Table 4 presented citations to electronic resources within the period under study. There was no citation to electronic resources in the first six years within the period ( ). Citations to electronic resources began in 2004 with 18 of such citations. However the highest citation to electronic resources was in 2006 with 80 citations representing 46.2% while 2007 followed with 67 citations (38.7%). The least citations to electronic resources were recorded in 2005 with only 8 citations representing 4.6% of the total. University of Ibadan produced the highest citations to electronic resources with 49.1% of the electronic resources followed by Obafemi Awolowo University with 26% of the electronic citations. University of Benin did not contribute any citation to electronic resources.

109 97 Research Question 5: What are the languages of cited sources in the theses on Nigerian languages? Table 5: Language of Cited Sources LANG. ABU UNILAG OAU UI UNN BENIN TOTAL Percent English French German Hausa Igbo Yoruba Edo Efik TOTAL 1550 (5.6%) 2728 (9.9%) 6333 (22.9%) % % % % 100 Percent Igbo, 506 Hausa, 456 German, 24 French, 45 Yoruba, 3843 Efik, Edo, 1 5 English, English French German Hausa Igbo Yoruba Edo Efik Fig. 5: Pie chart showing the languages of the cited sources

110 98 Table 5 presents the language in which the cited sources were written. It is evident from the table that citations to sources written in English language are the highest with citations representing 82.4% of the total followed by Yoruba language sources with 3843 (13.9%). Hausa language sources have 456 citations or 1.6%, while Igbo language sources have 506 citations representing1.8% of the entire citations. French and German language citations are 45 (0.2%) and 24 (0.08%) respectively while Efik language citations are 5 (0.02%). The least is Edo with only 1 (0.003%)

111 99 Research Question 6: What is the timeliness of sources cited in the theses on Nigerian languages? Table 6: Timeliness of Cited Sources DATE ABU UNILAG OAU UI UNN BENIN TOTAL Percent Before No date TOTAL 1550 (5.6%) 2728 (9.9%) 6333 (22.9%) % % % % 100

112 100 It can be observed from the table that 1981 recorded the highest citations for a single year with 1143 (4.1%) of the total. Materials published between 1900 and 1960 generated 1597 (5.8%) citations. Citations to resources produced in 1982 are 957 representing 3.5% of the total citations, while citations to materials of 1985 are 835 or 3.0%. It is noteworthy to observe that citations without date of publication generated up to 363 (1.3%) citations. The least citations were recorded in 2007 with 44 citations or 0.2% of the total citations Table 7: Summary of Timeliness of Cited Sources by Periods. Period No Percent Before No date Total

113 Before PERCENT No date PERCENT Fig. 6: A Line graph showing the summary of timelines of cited sources. Table 7 is a summary of timeliness of cited sources by periods. Materials published before 1900 generated the least citations with 105 citations or 0.4% of the whole. The period between 1980 and 1989 recorded the highest number of citations (7757 or 28%). This is followed by sources produced between 1970 and 1979 which has 6668 citation representing 24.1% of the whole, while the period has 5957 citations representing 21.5% of the entire citations. The table also shows that the period from generated 2282 citations (8.2%) while the period has 1767 citations or 6.4%. citations without date are 373 in number representing 1.4%

114 102 Research Question 7: Which are the frequently cited journals in the theses on Nigerian languages? Table 8: Frequently Cited Journals Rank Journals ABU LAG OAU UI UNN BEN Total % of freq. cited Jnl. 1 Linguistics Inquiry Nigerian Magazine Journal of W/African languages 4 Language Studies in African Linguistics 6 Yoruba: a Jnl of Yoruba Studies Ass. 7 International Journal of the Sociology of Language 8 Odu: Journal of African Studies 9 Language Learning Research in African Literature 11 Journal of Afr. Lang & Linguistics 12 Linguistics Analysis Anthropological Linguistics 14 Phonotica Word Journal of Linguistics Circle 16 Linguistics Langbasa Journal of Ling. Ass. of Nigeria 19 Lingua Yoruba Gbode African Notes Journal of Linguistics Journal of Nig. Lang & Literature Total Percent % of total Jnl. citations (4521) Table 8 presents data on the most frequently cited journals. Twenty three journals with 50 citations or more were identified as frequently cited journals.

115 103 From the table, the most cited journal is Linguistics Inquiry with 375 citations representing 8.3% of the total journal citations or 14.6% of the most frequently cited journals followed by Nigerian Magazine with 247 citations representing 5.5% of the entire journal citations or 9.6% of the most frequently cited journals. Journal of West African Languages ranked third in the list with 195 citations translated to 4.3% of the whole journal citations or 7.6% of the most frequently cited journals while Language ranked fourth with 163 citations or 3.6% of the whole journal citations. Studies in African Linguistics has 146 citations representing 6.1% of the most frequently cited journals or 3.5% of the overall journal citations. We can also see from the table that a Nigerian indigenous language journal Yoruba ranked sixth in the table with 131citations. This figure represents 5.1% of the most frequently cited journals or 2.9% of the total journal citations. International Journal of the Sociology of Languages, Odu: Journal of African Studies and Language and Learning ranked seventh, eighth and ninth with 127, 125 and 102 citations respectively. The rest of the journals in the list have less than 100 citations with Journal of Nigerian Language and Literature taken the rear with 50 citations translated to 1.9% of the most frequently cited journals or 1.1% of the overall journal citations. The twenty three journals in the table yielded a total of 2575 citations or 57% of the whole journal citations. University of Ibadan has the highest number of citations in the most frequently cited journals with 41.9% followed by Obafemi Awolowo University with 24.8% while the least contribution came from University of Benin with 0.04%.

116 104 Application of Law of Scattering on the Frequently Cited Journals Using Brookes mathematical explanation of Bradford s law of scattering with the formular R(n) = k log (n) the following result was obtained. Table 9: Application of Bradford s Law of Scattering Rank Journals No of citations (N) Partial sum (p) 1 Linguistics Inquiry Nigerian Magazine Journal of W/African languages Language Studies in African Linguistics 156 1, Yoruba: A Journal of Yoruba Studies 131 1, Association. 7 International Journal of the Sociology of 127 1, Language 8 Odu: Journal of African Studies 125 1, Language Learning 102 1, Research in African Literature 84 1, Journal of Afr. Lang & Linguistics 79 1, Linguistics Analysis 77 1, Anthropological Linguistics 76 1, Phonotica 73 2, Word Journal of Linguistics Circle 72 2, Linguistics 71 2, Langbasa 68 2, Journal of Ling. Ass. of Nigeria 64 2, Lingua 63 2, Yoruba Gbode 61 2, African Notes 60 2, Journal of Linguistics 56 2, Journal. Of Nig. Lang & Literature 50 2, Log of p Table 9 presents result for the application of Bradford s law of scattering. According to the law if scientific journals are arranged in order of decreasing productivity of articles, in a given subject, they may be divided into a nucleus of periodicals more devoted to the subject and several groups or zones containing the same number of articles as the nucleus, when the number

117 105 of periodicals in the nucleus and succeeding zones will be a 1: n: n 2... Following this law, it can be observed from the table above that eight journals produced one third of the total journal citations making the first nucleus, while 33 journals produced two thirds of the total journal citation making the second nucleus (see appendix I for details). The ratio of the scattering is 8: 33: 91. This does not follow exactly Bradford s ratio because this is a humanity field which has low citations to the journals Scattering of journals 2500 Vertical (value) axis Partial sum (p) Log of p Fig. 7: Graph showing the scattering of the journals

118 106 Research Question 8: Who are the most frequently cited authors in the theses of Nigerian languages? Table 10: Most Frequently Cited Authors Rank AUTHORS ABU UNILAG OAU UI UNN BENIN Total Percent 1 Chomsky, N Bamgbose, A Olatunji, O.O Awobuliyi, O Ogundeji, P.A Babalola, A Ambimbola, W Isola, A Owolabi, K Emenanjo, E.N Crystal, D Uwalaka, M.A McCharty, J Olabimitan, A Finnegan, R William, E Yusuf, O Goldsmith, J Ogunsina, J.A Ilesanmi, T.M Clement, G.N Halliday, M.A Adedeji, J.A Beier, Ulli Nwachukwu, P.A Eagleton, T Banjo, A Radford, A Williamson, K Nwadike I.U Baberk Awoyalale, Y Elugbue, B.O Olukoju, E.O Greenberg, J.H Uzochukwu, Olajubu, O Hyman, L.M Abraham, R.C Bloomfield, L Adetugbo, A Agbaje, J.B Labov, W Egbokhare, F.O Baker, C.L Fagunwa, D.O Ikekeonwu, C.I Hymes, D.H Okedeji, O

119 107 Rank AUTHORS ABU UNILAG OAU UI UNN BENIN Total Percent 50 Idowu, E.B Galandici, Abram, M.H Enekwe, O.O Trudgill, P.T Ubesie, T Ofomata,,E C Fabunmi, M.A Ogbalu, F. C Ahmed, U Bernstein, B.B Abiodum, M.A Olubunle, A Akinlabi, A.M Ubahakwe, E Nwoga, D.I Yai, O.B Mba, B.M Akinkugbue, O.O Macnamara, J Olunlade, T Agbedo,.C.U Afolayan, A Oluikpe, Okebalama, C.N Total Percent Table 10 presents the result of the most frequently cited authors from the analysis. Seventy four authors with 50 citations and above were identified. From the table it can be seen that Noam Chomsky is the most frequently cited author with 435 citations or 5.1% of the citations in the list. He is followed by Bamgbose, A with 390 citations or 4.6% of the frequently cited authors. Olatunji, O.O ranked third in the table with 278 citations representing 3.3% of the citations in the table while Awobuliyi, O ranked fourth with 266 citations or 3.1% of the citations in the table. Ranking fifth and sixth respectively are Ogundeji, P.A and Babalola, A with 233 (2.8%) and 229 (2.7%), while Ambimbola, W and Isola, A followed them with 212 or 2.5% and 206 or 2.4%

120 108 respectively. Other authors that are highly cited are Owolabi, K who has 187 (2.2%), Emenanjo, E.N with180 citations (2.1%), Crystal, D who recorded 158 citations (1.9%) and Uwalaka, M.A who has 152 citations representing 1.8% of the most cited authors. As can be seen from the table, 34 authors have one hundred citations and above, however at the rear are Okabalama and Oluikpe with 50 citations each. University of Ibadan also has the highest number of contribution to the citations with 37.3% followed by OAU with 19.3% while UNN contributed 19.1% of the citations. UNIBEN has the least contribution to the citations with 0.08%. The 74 authors identified yielded 8,467 citations representing 30.6% of the total citations in the study.

121 109 Research Question 9: Which are the most frequently cited books in the theses of Nigerian languages? Table 11: Most Frequently Cited Books Rank Books Author Date of pub ABU LAG OAU UI UNN BEN TOTAL % of freq cited bks % of bks cited 1 Aspect of the theory Chomsky, N of syntax 2 A short Yoruba Bamgbose, grammar A 3 Features of Yoruba Olatunji, oral poetry O.O 4 Essentials of Yoruba Awobuluyi, grammar O 5 Introduction to Ogundeji Yoruba oral poetry P.A 6 Knowledge of Chomsky, N language, its nature, origin and use. 7 The modern Yoruba Isola, A novels: an analysis of the contents art 8 Introduction to Ogundeji, Yoruba oral literature P.A 9 Contents and forms Babalola, A of Yoruba ijala 10 Language in Nigeria Owolabi, K Oral literature in Finnegan, R Africa 12 Yoruba oral tradition Abimbola, W 13 The development of Ogunsina, Yoruba novels J.A 14 The handbook of Goldsmith, J phonological theory 15 Yoruba theatre: Beier, Ulli introduction to African theatre 16 Igbo verbs: a semantic syntactic analysis Uwalaka, M.A Multilingualism, minority languages and language policy in Nigeria. 18 Towards an Igbo literary standard 19 Transformational grammar: a first course 20 Yoruba meta language 21 Igbo-English dictionary Emenajo, E.N Nwachukwu, P.A Radford, A Awobuliyi, O Williamson, K

122 110 Rank Books Author Date of pub ABU LAG OAU UI UNN BEN TOTAL % of freq cited bks % of bks cited 22 Literary theory: an Eagleton, T introduction 23 Language of Africa Greenberg, J.H 24 Exploration in the Halliday, functions of language M.A.K 25 Olodumare: Gods in Idowu, E.B Yoruba belief 26 Linguistics Crystal, D Transformational Yusuf, O generative grammar: an introduction. 28 Lectures on Chomsky, N government binding 29 Language, culture Hymes, D.H and society: a reader in linguistics and anthropology 30 Dictionary of Abraham, modern Yoruba R.C 31 Traditional Yoruba Olukoju, poetry E.O 32 Phonology: theory Hyman, L and analysis 33 Awon oriki orile Babalola, A The language of Greenberg, Africa J. 35 Language Bloomfield, L 36 Yoruba language and Afolayan, A literature 37 Sociolinguistics Trudgill, P TOTAL Percent Table 11 presents result for the most frequently cited books among Nigerian language researchers. The table shows that the most frequently cited book is Chomsky s Aspect of the theory of syntax with 234 citations or 6.6% followed by A short Yoruba grammar by Bamgbose which yielded 183 citations. Ranking third among the frequently cited books is Features of Yoruba oral poetry by O.O Olatunji with 162 citations while Essentials of Yoruba grammar by Awobuluyi ranks fifth with 134 (3.8%) citations. Another

123 111 book by Chomsky, Knowledge of language, its nature and use is sixth in the ranking with123 (3.5%) citations. This has the same number of citations with A Isola s The modern Yoruba novels: an analysis of the contents and art. Ogundeji s book Introduction to Yoruba literature ranks eighth in the list with 121 (3.4%) citation while Contents and forms of Yoruba ijala by Babalola, follows it with 107 (3.0%) citations. Owolabi s Language in Nigeria and Ruth Finnegan s Oral Literature in Africa ranked eleventh and twelfth with 106 (2.9%) and 104 (2.9%) respectively. Three books have 98 (2.8%) citations. These books are Yoruba oral tradition by Abimbola, W, The development of Yoruba novel by Ogunsina, J.A and The hand book of phonological theory by Goldsmith. Other books that are frequently cited are Ulli Beier s Yoruba theatre: introduction to African theatre with 97 (2.7%), Igbo verbs: a semantic-syntactic analysis by Uwalaka, M A with 95 citations, Multilingualism minority languages by E.N Emenanjo which has 94 citations, P.A Nwachukwu s Towards an Igbo literary standard with 93 citations and Transformational grammar: a first course by Radford which has 91 citations. Ranking twentieth in the list is Yoruba Meta language by O. Awobuliyi with 89 citations while the twenty first is Igbo-English dictionary by K Williamson with 86 citations. The remaining most cited books have less than eighty citations each. However at the rear of the most frequently cited books are Sociolinguistics by P. Trudgill with 50 citations, Yoruba language and literature by A Afolayan with 51 citations and Language by L. Bloomfield. The 37 most frequently cited books yielded 3,496 citations representing 20.6%

124 112 S/N of the citations to books and 12.6% of the total citations. From the table the oldest book among the most frequently cited books (Language by Bloomfield) was published in 1935 while the latest The modern Yoruba novels: an analysis of the contents art by A. Isola was published in The highest contribution to the citation is University of Ibadan with 35.0% followed by OAU with 32.9% while UNIBEN took the rear with 0.5%. Research Question 10: Which of the most frequently cited journals are available in Nigerian university libraries? Table 12: Availability of Frequently Cited Journals in Nigerian university libraries JOURNALS No of citations Universities UI UNN UNI LAG OAU ABU UNIB EN Total Percentage of total availability 1 Linguistics Inquiry Nigerian Magazine Journal of W/African languages Language Studies in African Linguistics Yoruba: A Journal of Yoruba Studies Association. 7 International Journal of the Sociology of Language 8 Odu: Journal of African Studies Language Learning Research in African Literature Journal of Afr. Lang & Linguistics Linguistics Analysis Anthropological Linguistics Phonotica Word Journal of Linguistics Circle Linguistics Langbasa Journal of Ling. Ass. of Nigeria Lingua Yoruba Gbode African Notes Journal of Linguistics Journal. Of Nig. Lang & Literature Total Percent Note: 1 means available and 0 means not available.

125 113 Table 12 presents results of the availability of frequently cited journals in Nigerian university libraries studied. A close observation from the table indicates that Research in African Literature and Journal of Linguistics are available in all the university libraries closely followed by Anthropological Linguistics and Studies in African Linguistics which are available in 5 university libraries. Up to 6 other journals are available in only four university libraries. The least in the availability of the journals are Language Journal of Nigerian and Literature and Yoruba: a Journal of Yoruba Studies Association which are available in only 1 library each. Obafemi Awolowo University has the highest journal availability with 23% followed by University of Ibadan with 20%. UNN and ABU have 19% each while the least journal availability is in UNILAG and UNIBEN with 9% each. The average availability percentage is 54.3%.

126 114 S/N Table 13: Distribution of Availability of Frequently Cited Journals in Nigerian university libraries JOURNALS No of Distribution of journal availability in the citation universities UI UN N UNI LA G OA U AB U UNI BE Total Percentage of total availability 1 Linguistics Inquiry Nigerian Magazine Journal of W/African lang Language Studies in Afr Ling Yoruba: a Jnl of Yoru Studies Asso. 7 Int. Jnl. of the Soc of Lang 8 Odu: Jnl of Afr. Studies Language Learning Research in Afr. Lit Jnl. of Afr. Lang & Ling Linguistics Analysis Anthropo Linguistics Phonotica Word Journal of Linguistics Circle 16 Linguistics Langbasa Jnl. of Ling. Ass. of Nig Lingua Yoruba Gbode African Notes Journal of Linguistics Journ. of Nig. Lang & Lit Total Table 13 presents data on the availability of the most frequently cited journals. The result indicates that only two journals namely; Research in African literature and Journal of linguistics are available in all the university libraries and the two have the highest number of issues available. The former has 194 issues available while the later has 158 issues available. Two journals namely; Yoruba Gbode and Journal of linguistics Association have the least number of issues available in the libraries. They have 8 issues each available in the libraries. University of Ibadan library has the

127 115 highest number of journal issues with 501(34.8%) issues while University of Benin library has the least number of journal issues available with only 41issues (2.8%). Table 14: Availability of most frequently cited books Rank Books Author Date of pub No of citations UI UNN UNILAG OAU ABU UNIBEN Total Percentag e of total availabilit y 1 Aspect of the theory Chomsky, N of syntax 2 A short Yoruba Bamgbose, A grammar 3 Features of Yoruba Olatunji, O.O oral poetry 4 Essentials of Yoruba Awobuluyi, O grammar 5 Introduction to Ogundeji P.A Yoruba oral poetry 6 Knowledge of Chomsky, N language, its nature, origin and use. 7 The modern Yoruba Isola, A novels: an analysis of the contents art 8 Introduction to Ogundeji, P.A Yoruba oral lit. 9 Contents and forms Babalola, A of Yoruba ijala 10 Language in Nigeria Owolabi, K Oral literature in Finnegan, R Africa 12 Yoruba oral Abimbola, W tradition 13 The development of Ogunsina, J.A Yoruba novels 14 The handbook of Goldsmith, J phonological theory 15 Yoruba theatre: Beier, Ulli introduction to African theatre 16 Igbo verbs: a Uwalaka, M.A semantic syntactic analysis 17 Multilingualism, Emenajo, E.N minority languages and language policy in Nigeria. 18 Towards an Igbo Nwachukwu, literary standard P.A 19 Transformational Radford, A grammar: a first course 20 Yoruba meta Awobuliyi, O language 21 Igbo-English dictionary Williamson, K

128 116 Rank Books Author Date of pub No of citations UI UNN UNILAG OAU ABU UNIBEN Total Percentag e of total availabilit y 22 Literary theory: an Eagleton, T introduction 23 Language of Africa Greenberg, J.H Exploration in the functions of language 25 Olodumare: Gods in Yoruba belief Halliday, M.A.K Idowu, E.B Linguistics Crystal, D Transformational generative grammar: an introduction. Yusuf, O Lectures on government binding 29 Language, culture and society: a reader in linguistics and anthropology Chomsky, N Hymes, D.H Dictionary of Abraham, R.C modern Yoruba 31 Traditional Yoruba Olukoju, E.O poetry 32 Phonology: theory Hyman, L and analysis 33 Awon oriki orile Babalola, A The language of Greenberg, J Africa 35 Language Bloomfield, L Yoruba language Afolayan, A and literature 37 Sociolinguistics Trudgill, P TOTAL Percent Table 14 presents data on the availability of frequently cited books. Results show that none of the university libraries have all the books identified as frequently cited. University of Ife library has the highest number of frequently cited books in its collections with 37 (25%) copies while University of Ibadan library follows with 32 (21%) copies of the books. University of Benin library has the least number of copies with14 copies while UNN, UNILAG and ABU have 22 (15%), 31 (20%) and 15 (10%) copies respectively. Three of the books namely; Contents and forms of Yoruba

129 117 Ijala, Oral literature in Africa and Multilingualism, minority languages and language policy in Nigeria are available in all the libraries studied. The first two of these books have the highest number of copies in terms of availability, while Phonology: theory and analysis has the least number of copies and is only available in one university library. Research Question 11: What is the relationship between the frequently cited journals and their availability in Nigerian university libraries? S/N Table 15: Correlation* between the number of frequently cited journals and their availability in the university libraries JOURNALS No of No Availa XY X 2 Y 2 Cit (X) (Y) 1 Linguistics Inquiry Nigerian Magazine Journal of W/African languages Language Studies in African Linguistics Yoruba: A Journal of Yoruba Studies Association. 7 International Journal of the Sociology of Language 8 Odu: Journal of African Studies Language Learning Research in African Literature Journal of Afr. Lang & Linguistics Linguistics Analysis Anthropological Linguistics Phonotica Word Journal of Linguistics Circle Linguistics Langbasa Journal of Ling. Ass. of Nigeria Lingua Yoruba Gbode African Notes Journal of Linguistics Journal. Of Nig. Lang & Literature Total , *r = 0.140; α = 0.05; df = 22; t-cal = 0.69; t-crit = 1.72; t-cal < t-crit. Table 15 presents data for the correlation between the number of frequently cited journals and their availability in the Nigerian university libraries. As can

130 118 be observed from the table the degree of freedom (df) is 22, t-calculated is 0.69 while t-critical is 1.72 and the relationship r is It can be seen that t- calculated is less than t-critical. There is a low but positive relationship between the number of frequently cited journals and their availability in the university libraries Hypothesis 1: There is no significant relationship between the number of frequently cited journals and their availability in Nigerian university libraries. Table 16: Test of significance of the association between the number of frequently cited journals and their availability in the university libraries Number of Ranked Journals R df t- calculated t- critical Decision t-cal. < t-crit. Test not significant Table 16 presents result of the test of significance of the association between the number of frequently cited journals and their availability in Nigerian university libraries. The result shows that number of ranked journals is 23, the relationship r is 0.140, degree of freedom is 22, t-calculated is 0.69, and t- critical is The result of the t-test shows that t-calculated is less than t- critical which is not significant, therefore retain the null hypothesis.

131 119 Research Question 12: What is the relationship between the frequently cited books and their availability in Nigerian university libraries? Table 17: Correlation** between the number of frequently cited books and their availability in the university libraries S/N No of Total Date citatio availa Books Author of ns bility pub XY X 2 Y 2 (X) (Y) 1 Aspect of the theory of Chomsky, N syntax 2 A short Yoruba grammar Bamgbose, A Features of Yoruba oral Olatunji, O.O poetry 4 Essentials of Yoruba Awobuluyi, o grammar 5 Introduction to Yoruba oral Ogundeji P.A poetry 6 Knowledge of language, its Chomsky, N nature, origin and use. 7 The modern Yoruba novels: Isola, A an analysis of the contents art 8 Introduction to Yoruba oral Ogundeji, P.A literature 9 Contents and forms of Babalola, A Yoruba Ijala 10 Language in Nigeria Owolabi, K Oral literature in Africa Finnegan, R Yoruba oral tradition Abimbola, W The development of Ogunsina, J.A Yoruba novels 14 The handbook of Goldsmith, J phonological theory 15 Yoruba theatre: introduction to African theatre Beier, Ulli Igbo verbs: a semantic syntactic analysis Uwalaka, M.A Multilingualism, minority Emenajo, E.N languages and language policy in Nigeria. 18 Towards an Igbo literary Nwachukwu, standard P.A 19 Transformational grammar: Radford, A a first course 20 Yoruba meta language Awobuliyi, O Igbo-English dictionary Williamson, K Literary theory: an Eagleton, T introduction 23 Language of Africa Greenberg, J Exploration in the functions of language Halliday, M.A

132 120 S/N Books Author Date of pub No of citatio ns (X) Total availa bility (Y) XY X 2 Y 2 25 Olodumare: Gods in Idowu, E.B Yoruba belief 26 Linguistics Crystal, D Transformational Yusuf, O generative grammar: an introduction. 28 Lectures on government Chomsky, N binding 29 Language, culture and Hymes, D.H society: a reader in linguistics and anthropology 30 Dictionary of modern Abraham, R.C Yoruba 31 Traditional Yoruba poetry Olukoju, E.O Phonology: theory and Hyman, L analysis 33 Awon oriki orile Babalola, A The language of Africa Greenberg, J Language Bloomfield, L Yoruba language and Afolayan, A literature 37 Sociolinguistics Trudgill, P Total 3, , , **r = 0.145; α = 0.05; df = 36; t-cal = 0.93; t-crit. = 1.69; t-cal < t-crit. Table 17 presents data for the correlation between the number of frequently cited books and their availability in the Nigerian university libraries. As can be observed from the table the degree of freedom (df) is 36, t-calculated is 0.93 while t-critical is 1.69 and the relationship r is It can be seen that t-calculated is less than t-critical. There is a low but positive relationship between the number of frequently cited books and their availability in the university libraries

133 121 Hypothesis 2: There is no significant relationship between the number of frequently cited journals and their availability in Nigerian university libraries. Table 18: Test of significance of the association between the number of frequently cited books and their availability in the university libraries Number of R df t- t- Decision Ranked Books calculated critical t-cal. < t-crit. Test not significant Table 18 presents result of the test of significance of the association between the number of frequently cited books and their availability in Nigerian university libraries. The result shows that number of ranked books is 37, the relationship r is 0.145, degree of freedom is 36, t-calculated is 0.93, and t- critical is The result of the t-test shows that t-calculated is less than t- critical which is not significant; therefore the null hypothesis is retained. Summary of Major Findings 1. The findings of the study shows that the theses studied yielded a total of 27,686 citations while books and monographs contributed more than half of the citations followed by journals with 16.2% of the citations. Government publications, reports, inaugural lectures, diaries and news publications were lowly cited. 2. The findings also reveal that of the 449 theses studied, Yoruba language contributed the highest number of theses while Urhobo contributed the least. Minority Nigerian languages made low contributions.

134 Most of the theses were written in English language even though the focus of the theses is purely on Nigerian indigenous languages. Over 80% of the theses were written in English language while none is written in Igbo, a major Nigerian language. 4. Citations to electronic resources in the theses are very low. Electronic publications contributed only 0.6% in the total citations. There was no citation to electronic publications from 1998 to The findings also revealed that 2006 witnessed the highest citation to electronic resources with 46.2% of the electronic citations 5. The findings also revealed that English language sources have the highest citations with 84.5% while the least is Edo with 0.003%. 6. The findings indicate that materials published between 1900 and 1960 contributed 6.4% of the total citations. The highest numbers of citations are to materials published in 1981 while the least are materials published in Twenty three journals were identified as the most frequently cited journals in Nigerian language research. Linguistics inquiry is leading these journals with 14.6% of the total journal while the least citations came from Journal of Nigerian Language and Literature with 1.9% of the total journal citations. 8. Seventy four authors that have up to 50 citations were identified as the most frequently cited authors. Findings revealed that Noam Chomsky tops these authors with 435 citations or 5.1% and the least citations

135 123 among these authors were from Okebalama, C.N with 0.7% of the citations. 9. The finding equally identified thirty seven most frequently cited books. Top among these books is Chomsky s Aspect of theory of syntax with 234 citations or 6.6% of the citations while the least in the list is Sociolinguistics by P Trudgill with 60 citations. 10. The findings indicate that only two journals namely; Research in African literature and Journal of linguistics are available in all the university libraries and the two have the highest number of issues available. 11. Findings show that none of the university libraries have all the books identified as frequently cited. University of Ife library has the highest number of frequently cited books in its collections with 37 copies while University of Benin has the least. 12. Findings reveal there is a non significant (p<.05) low but positive relationship (r = 0.14) between number of cited journals and their availability in the libraries (t-cal = 0.69; t-crit = 1.717). 13. Similarly, there is a non significant (p<.05) low but positive relationship (r = 0.15) between number of cited books and their availability in libraries (t-cal = 0.93; t-crit = 1.688). 14. The language journals follow Bradford s law of scattering however, the ratio of the scattering is 8: 33: 91. This does not follow exactly Bradford s ratio because this is a humanity field which has low citations to the journals.

136 124 CHAPTER FIVE DISCUSSION OF FINDINGS, IMPLICATIONS, ECOMMENDATIONS AND CONCLUSION This chapter presents discussion of findings of the study, the interpretation of the findings, implications, recommendations and conclusion. Discussion of Findings The Types of Cited Sources in the Theses The result of the findings showed that books and monographs are the dominant information sources cited in the theses of Nigerian languages. This finding is in line with earlier works on information seeking behavior of the humanists. For instance, Thompson (2002) had underscored the relevance of books and monographs to researchers in the humanities. The finding equally corroborates other citation studies such as Mendez and Chapman (2006); Zainab and Goi, (1997) among others. Therefore the prominence of books and monographs in humanities research from this finding continue to be consistent. However, there is a slight increase in the citations to books and monographs. While earlier citation studies recorded a little above fifty percent the present finding is more than sixty percent. The findings also reveal that journal citations are fairly represented with more than sixteen percent of the total citations. The implication is that the Nigerian language researchers consider journal as a very good vehicle for

137 125 scholarly research. This also shows that primary information sources are considered very seriously as research materials in the languages. The finding is in line with other citation studies reviewed in the literature. They demonstrated similar result on the use of journals in linguistics research. For example in the study carried out by Knievel and Kellsey (2005), journal articles accounted for over twenty five percent of the total citations. It equally corroborates earlier findings by Heinzkill (1980), who reported that journals contributed twenty percent in his work on English language and literature. Interestingly, findings from the study show that another important information source of particular interest to the Nigerian language researchers are theses and dissertations. Citation to these primary information sources accounted for over ten percent of the total citations. This development is not surprising for it is a known fact that theses and dissertations contribute in no small measure to research literature of all fields of study. Citations to theses and dissertations have usually been classified as others in most citation studies making it difficult to appreciate their value in scholarly research. However, the earlier study of Zainab and Goi (1997) specified the contributions of theses to research efforts to be more than six percent and the present findings has a lot of similarities with that of Zainab and Goi. Other types of information sources that are represented in the citations are conference and seminar papers. The findings show that these yielded close to five percent of the total citations. This finding is not out of place for even though these kinds of information source are not easily accessible because they

138 126 are often unpublished; researchers still find them as very useful source of information for scholarly publications. The findings also lay credence to studies cited in the literature. For example, in the work of Zainab and Goi (1997), citations to conference and seminar papers yielded nearly four percent of the total citations. Other studies usually group conference papers with other information sources, denying the readers knowledge of their actual contribution to research. The low citations to seminar and conference papers could be as a result of the problems associated in tracking down such information sources since they are not published. An important information source that has a surprising low citation is government documents. This finding to some extent contradicts earlier findings in citation studies. For example, Zainab and Goi (1997) found out that government publications contributed up to 3.8 percent of total citations in their analysis of theses of humanities researchers. Could it be that government documents are not accessible to Nigerian language researchers or that the researchers do not consider this type of information sources very relevant in their research process? Observation during the analysis even shows that citations to National Policy on Education contributed more than half of the citations to government publications. This means that without the National policy on education, citations to government documents could have been lower than the present figure. Another primary source of information as reflected in the findings is interview. Most citation studies do not give adequate consideration to this very

139 127 important information source in humanities research. Though the contribution to the overall citation is quite low, it is important to point out that as has been noted in the literature, interview has always been recognized as a very reliable source of information in the humanities as has been demonstrated by Wiberly and Jones (1989). Other information sources that have made great impact from the findings are lecture notes and reports. These sources were hardly recognized or even completely absent in earlier citation studies. Interestingly, findings have demonstrated that they should be given adequate attention in terms of organization, access and use, particularly as it concerns librarians. The Language Spread of Theses Studied Theses analyzed concentrated their studies in the three major Nigerian languages, namely Hausa, Igbo and Yoruba. Findings however reveal that theses on Yoruba language studies were dominant, contributing fifty-seven percent of the total number. Perhaps this is a reflection of the level of development of Nigerian indigenous languages. Yoruba arguably seems to be the most developed Nigerian language. This is not far from the contents of Hair s (1967) work who had revealed that publications in Yoruba language first appeared as far back as The early study of Yoruba language seems to have given it an edge over other indigenous Nigerian languages and equally propelled research in the language. Another reason that may be responsible for the large percent of theses focusing on Yoruba studies is that three of the first generation universities are located in Yoruba-speaking environment. Although observations during the analysis revealed that other Nigerian languages such as

140 128 Igbo, Hausa, Edo, Efik among others were also researched, the extent of such research cannot be compared with that of Yoruba, which is the native language of large percent of the catchment areas of the universities. From the findings, Igbo language is another Nigerian indigenous language that is highly researched. This is not surprising as literature has revealed that Igbo language seems to be the next most developed language after Yoruba. For instance a study carried out by Olaofe (1990) demonstrated that apart from Yoruba, Igbo language has more publications than any other Nigerian indigenous language. The number of theses studying Igbo language is also a clear testimony of the development of Igbo language. Besides this, observations made during the analysis revealed that out of the three major Nigerian languages, Igbo is the most studied outside the catchment areas of the location of the universities. Though Hausa language recorded an appreciable number of theses, one would have expected more than forty five considering the fact that Hausa is a major Nigerian language. The low number of theses in Hausa could be a reflection of the number of publications in the language as reflected in Olaofe s (1990) work. Evidently, the contributions of each of the languages portray the level of development of these languages in terms of publications, research, development of orthographies and Meta languages. On the whole the three major languages contributed over ninety three percent of the total theses analyzed.

141 129 Interestingly, the minority languages were fairly represented. From the revelation of the findings, Ibibio is the most researched of all the Nigerian minority languages. This minority language made a fair contribution to the theses studied with about 1.2 percent. Similarly Edo and Urhobo contributed to some extent in the theses population. One can therefore safely conclude that these three minority languages are emerging Nigeria indigenous languages with potential research focus. It is however, surprising that a popular Nigerian language Fulfulde contributed very little to the theses population. The reason behind this is not clear for we do know that Fulfulde is a language spoken in many African countries and has large speaking population in Nigeria, as can be seen in Nigerian language map (see appendix vi). This should give a great concern to Nigerian linguists, particularly this time that many languages are going into extinction. It is also a source of worry that of the over four hundred and fifty languages in Nigeria, Adegbija (2004), only twelve of them are represented in the study. Government and Nigerian language researchers should be concerned on what may be happening to the remaining that are completely absent. Could it be that no orthography has been developed for these languages or could it be that many of them are already going into extinction? The Language in Which the Theses Were Written Research in Nigerian languages may still be struggling to detach itself from colonial subjugation as can be clearly seen from the findings. Even when

142 130 the research is clearly on Nigerian indigenous languages, a large proportion of the research was written in a foreign language, English. Though English is the official language of the country and also language of instruction in most Nigerian schools, it is amazing that English is still being used in writing research conducted specifically in Nigerian indigenous language and culture. The over eighty percent dominance of English language is highly regrettable, leaving the remaining percent for other Nigerian languages put together. Paradoxically, Igbo, one of the major Nigerian languages, is completely absent from the list. This development is highly disturbing for languages are better studied by reading and writing in the very languages. The inability of Nigerian language researchers to write in Nigerian languages no doubt may have contributed to the poor development of most of the Nigerian languages and may equally affect the availability of information sources published in Nigerian languages. The low level of writing in Nigerian languages is equally antithetical to the argument of King ei (1999) that conducting research in African indigenous languages is capable of propelling other research, for according to him it is mainly through the use of languages that vital information relevant for research could be obtained. In addition to this, writing of theses in Nigerian languages would encourage the publication of text books in Nigerian languages and the development of orthographies in other minority languages which Bamgbose (2006) believes are among the social responsibilities of Nigerian linguists. It is encouraging however; that University of Ife is showing a good example for all the theses generated from the

143 131 university were written in Yoruba language. This is what other universities should be doing. Citations to Electronic sources in the Theses Against the present reliance on electronic resources for sourcing information for research, findings reveal low rate of electronic citations in the theses studied. There was no citation to electronic resources in the first six years out of the ten years period in the study. Perhaps this could be as a result of poor information and communication technology (ICT) infrastructure and general lack of internet search skills among researchers in Nigeria within the period. Internet connectivity in Nigerian universities even now is yet to take proper shape. This development does not encourage access to online resources which facilitates research in the country. Findings also show that citations to electronic resources in the theses began in It is however puzzling that instead of steady growth in electronic citations; it dropped from 10.4 percent in 2004 to 4.6 percent in the following year There was another sharp increase in 2006 to 46.2 percent and dropped again in It is not quite clear why there is no defined pattern in the growth of electronic citations over these years. The expectation would have been a steady increase because of improvement in ICT infrastructure within the period but that is not the case. A possible explanation may however be that a particular set of students may have poor internet skill and ICT awareness than another one and this could have affected their citations to

144 132 electronic resources. It is not surprising however, that University of Ibadan has more citations to electronic resources because they contributed nearly half of the theses analyzed. It is also of particular concern to see from the findings that even the four years of citation to e- resources, accounted for a very low citation rate compared with other types of cited sources. While available evidence from literature records a growing increase in citation to online resources, as have been demonstrated by Delgadillo and Lynch (1999), Graham, (2000) and Sukovic, (2008), Nigerian language researchers are yet to fully embrace the use of electronic resources in their research. The study of Graham recorded a steady increase in the citation of electronic resources over the years. The low citation of electronic resources by Nigerian language researchers may also be as a result of paucity of relevant materials on the Internet. Over the years there has been growing concern over lack of electronic resources from Nigeria and other African countries. This paucity of Nigerian resources on the Internet is even more glaring with resources published in indigenous languages. There is no doubt however, that the use of internet has impacted positively on research in the global research environment. For instance internet arguably provides the most current and up-to-date research materials to particularly information starved countries of Africa and Latin America.

145 133 The Languages of the Cited Sources in the Theses English language continues to dominate the languages of the cited information sources in the theses. The findings reveal that citations to English language sources contributed over eight four percent of the total citations. This perhaps is because the language commands a lot of respect in the global research community. In addition to this, English is not only the official language in Nigeria but also the language of instruction in almost all the schools in the country. This presupposes that most researches in Nigerian languages are often written in English language. However, one may have expected that since all the theses have their focus on Nigerian indigenous languages, more citations to Nigerian language sources should have been recorded. The findings reveal that it is only citations to Yoruba language sources that contributed fairly significantly to the total citations. Citations to Igbo and Hausa language sources regrettably fall below expectation in relation to the number of theses studied. This further points to the paucity of publications in the three major Nigerian languages, particularly Igbo and Hausa languages. German and French language sources seem to be the other sources that are cited by the Nigerian language researchers. It is puzzling that the contributions of these foreign languages to the overall citations are more than the Nigerian minority language sources. The reason for this could be that most of the minority Nigerian languages are yet to be developed. It is interesting to note that the dominance of English language sources in the citation is not far from earlier citation studies in the literature. For

146 134 example, the study of Zainab and Goi, (1997) found out that citation to English language sources accounted for about sixty six percent of the total citations recorded in the study. Similarly, the findings lay credence to Tonta and Al (2006) who found out that English language sources contributed over fifty percent of citation recorded in the study. The variation in the percentage could be that most of the journals and other books found in the libraries are published in English. The low citation to Nigerian indigenous language sources could also be as a result of lack of research culture generally in Nigeria and particularly in languages as has been demonstrated by King ei (1999). This is because the growth of any language is usually a function of the amount of research conducted in that language. The Timeliness of Cited Sources in the Theses Information emanating from the findings indicates that the time of publication of the cited sources ranges from pre 20 th century to Age therefore seem not be a major factor in the use of information resources by Nigerian language researchers. For example, materials published in the preindependence era contributed fairly to the overall citations despite the fact that scholarship and academic publications were still at infancy stage within the period. It is of interest to note that materials produced within that period contributed over six percent of the total citations. Though there are materials of 20 th century that were cited, they did not make significant contributions to the

147 135 entire citations. Such materials were publications of well known authors and archives originating from colonial masters which are found very useful in humanities research as has been observed by Wiberly and Jones (1989). Archives are primary information resources which have been found to be very authoritative in substantiating evidence in humanities and social science research. From the findings, citations to publications continue to grow with time until in the 1980s when it reached its peak. Specifically, 1981 witnessed the highest citations with over one thousand citations. One could safely say that 1980s is the golden period of publications in Nigeria. This is because the number of citations witnessed in a particular period could be a function of what was published and made available in the period. After the 1980s citations continued to decrease until the period of 2000s. One can equally argue that the pattern of citations over the period is a reflection of Nigerian s political and socio economic development within the period. Economically that was the period of oil boom when government was paying attention to education generally and publishing in particular. The period provided an enabling environment for academic activities. Politically, Nigeria was more of a civilian democracy within the period and these encouraged academic activities of which publications was a very critical component. Shortly after, there was military intervention which was not education friendly and citations to sources within the period began to drop. The drop reflected the attitude of the military dictatorship on education and publications. The decline

148 136 in the citations from 2000 to 2007 could be as a result of lack of accessibility of materials published within the period. It is obvious that materials become more accessible to researchers after few years of their publications when libraries would have acquired and advertised them, using the current awareness mechanisms available to them. In addition to this, the theses studied were produced between 1998 and 2007 which implies that most works published in 2007 were not likely to be cited and works published after 2007 would not appear in the citations. Findings also have shown that age of materials does not significantly affect their value in language research. It is evident from the findings that materials that are over fifty years old contributed more than six percent of the overall citations and materials that are over a hundred years contributed up to 105 citations. A close observation of the findings reveals that materials that are over twenty years of age account for up to seventy percent of the total citations. Similar findings have equally been recorded in Mendez and Chapman (2006) who reported that materials that were over fifty-five years contributed significantly in a citation study of Journal of Latin American history. Similarly Garfield, (1978a) and (1978b) had observed that humanities publications do not become obsolete, since works that are over five hundred years are still valuable to researchers in the area. This finding is equally in line with an earlier study of Karisson (1994) who discovered that linguistics like other humanities fields value old publications. Findings from his study revealed that large proportion of

149 137 publications cited were published between 1745 and 1910 and over sixty percent were citations to books published between 1940 and In the present findings materials published between 1970 and 1980 accounted for more than fifty percent of the total citations. It is also evident from the findings that sources without date of publication contributed more than one percent in the citation. Observations during the analysis indicate that some of these materials may have dates but lack proper documentation from the citers. Others are materials that naturally have no date of publication. The line graph (fig.v) clearly shows the pattern of growth in the citations over the period. The 1980s was the peak period in the citation after which it began to decline giving the graph a cone-shaped figure. The Most Frequently Cited Journals In the findings twenty-three most frequently cited journals were identified out of a total of ninety-one journals cited in the study. Surprisingly, Linguistics Inquiry ranked first as the most frequently cited journal, even though it is a foreign journal. This probably testifies to its reputation as an international journal. Another remarkable thing in the findings is that only eight out of the twenty-three most frequently cited journals are published in Nigeria. Most of the journals are published outside Africa. This may be as a result of the problems inherent in journal publication in Nigeria which has been described as irregular and uncoordinated.

150 138 It is however, of interest to observe from the findings that the twenty three journals identified generated close to sixty percent of the total journal citations, while the remaining forty three percent of the journal citations are scattered all over the remaining journals. This is in line with Bradford s law of scattering. The law explains that a small core of journals have as many papers on a given subject as much as larger number of journals. In this case, twentythree journals produced 57 percent of the citations which is more than what the 68 remaining journals (two thirds of the journals) produced. Therefore one third of the journals are the most productive journals in language and linguistics and can be safely regarded as the core journals. This finding also corroborates the observation of Hjorland (2007) that the law of scattering is not mainly about quantitative production of articles by journals but the determination of quality or the productivity level in terms of use among researchers in the area. The findings also made another baffling revelation. Only two journals published in Nigerian indigenous language made the list of the most frequently cited journals and as usual these two journals are published in Yoruba language. The journals Langbasa and Yoruba Gbode ranked 17 th and 20 th in the list. This does not give a fair representation of Nigerian indigenous languages in a study specifically focusing Nigerian languages. It therefore poses a huge challenge to Nigerian linguists in terms of publication of journals in the local languages. It is also a clear testimony that there is paucity of Nigerian language publications in libraries as has been observed by Tamuno (1980). Since 1980s

151 139 not much has changed in terms of improving the availability of Nigerian language publications in libraries. The Most Frequently Cited Authors About five hundred authors were cited in the theses, however findings revealed seventy-four most frequently authors. The authors that met the cut-off mark contributed over thirty percent of the total citations. Findings reveal that Noam Chomsky has the highest number of citations despite the fact that he is not a Nigerian author. The high citations he received in the theses testify to his international recognition in the field of language and linguistics. Interestingly apart from Chomsky, the first ten most cited authors are Nigerians with Ayo Bamgbose, a renowned Nigerian linguist, leading the authors. This evidently shows that Nigerian authors are not too backward in publications. However, it is important to note that since all the theses are research conducted specifically in Nigerian languages, the possibility of citing so many Nigerian authors is certain. Among Nigerians, authors of Yoruba extraction constitute greater percentage of the authors. For instance, among the first ten Nigerian authors only one author, Nolue Emenanjo is not from Yoruba extraction. Two reasons may have accounted for this. One is that greater proportion of the theses used were generated from universities cited in Yoruba environment which means that students are likely to cite authors from Yoruba extraction. Secondly, evidences from literature have shown that Yoruba is more developed in terms

152 140 of publications than other Nigerian languages. Another amazing discovery is that only one Hausa author, Galandici was able to make the list of the seventyfour most cited authors, ranking fifty-one in the list. Though the theses generated from Ahmadu Bello University, where Hausa studies has a very strong root are not so many compared to other universities, theses on Hausa language studies generated by other universities are sufficient to produce more frequently cited authors from Hausa extraction. Even authors from minority languages such as Egbokhare and Elugbe were among the most cited authors, despite the fact that minority languages generated very few theses. Therefore the low number of Hausa authors may not be attributed to the number of theses on Hausa language but on the quantity and quality of publications from the area. The findings revealed authors who have been actively involved in research in Nigerian languages over the years. This is in line with Garfield s (1998) position that citation analysis is useful to ascertain scholars who may be behind any subject area. Other foreign authors that have contributed significantly in the citations are David Crystal, J. McCharty, Ruth Finnegan, E William and J. Goldsmith. These authors may not have been published specifically in Nigerian languages but their publications may have laid a strong foundation for other Nigerian publications. For instance Ruth Finnegan s Oral literature in Africa gives a solid foundation and theoretical basis on any work in Nigerian literature. Similarly, Chomsky s Aspect of theory of syntax provides a very strong foundation on the grammar of any Nigerian language.

153 141 Most Frequently Cited Books The findings revealed a total of thirty seven most cited books, yielding 3,550 citations or 20.6 percent of the overall book citations. A total 294 books were cited, yielding 17,616 citations which give average citation of 60 per book. Findings clearly revealed the recognition of Chomsky in language and linguistics. Three of his books namely; Aspects of theory of syntax, Knowledge of language, its nature, origin and use and Lectures on government binding are in the list of the most frequently cited books ranking first, sixth and twenty eighth respectively. Similarly, books from Yoruba authors account for a large proportion of most frequently cited books. Two reasons may likely be responsible for this. First, a large number of the sources data (theses) are from universities cited in Yoruba dominated area. Therefore, it is most likely that citers, that are the researchers, may have access more to books written by Yoruba authors. Secondly, available evidence from literature points to the fact that of the three major Nigerian languages Yoruba is the most developed and has more publications than any other. Books published by Igbo authors are significantly represented in the list. On the contrary, there is a total absence of books written by authors from Hausa extraction. The reason for this is difficult to explain because theses focusing in Hausa language made high contribution in the total theses analyzed.

154 142 Findings also reveal that a greater percentage of the most frequently cited books are books published by Nigerians. Out of the thirty-seven books sixteen of them were written by non Nigerian authors. It is however necessary to observe that citations to these books contributed heavily in the overall citations. Some of these books written by non Nigerians are very important in the study of Nigerian languages for some of them are specifically interested in African languages in general and sometimes in Nigerian languages. For example, Ruth Finnegan s Oral literature in Africa, Kay Williamson s Igbo- English dictionary and J. H Greenberg s Language of Africa are all books that are very valuable for researchers in Nigerian language and literature. The findings also reflected the paucity of publications in Nigerian indigenous languages. Only one book (Awon oriki orile) by Ayo Babalola made the list of the most cited books, ranking 33 rd in the list. This finding corroborates earlier findings by Olaofe (1990) and Orimalade (2005) who had reported that there is scarcity of books written in Nigerian indigenous languages in Nigerian libraries. Observations have shown that the lack of availability of books in Nigerian languages could be attributed to three reasons. First, many writers do not have interest in publishing in Nigerian languages because according to them such publications do not attract wider readership. Secondly, publishers do not often accept manuscripts in local languages because such publications do not provide them the profit margin they require as businessmen. Finally, many Nigerians find it difficult reading materials published in Nigerian indigenous languages, no matter their level of education.

155 143 This therefore does not encourage authors and publishers in production of books in Nigerian languages. The age of the frequently cited books reflects the characteristics of materials used by humanities researchers. Available literature indicates that humanities researchers do not reject old research materials. In the present findings, the latest book among the most cited books is more than ten years old. The book (The modern Yoruba novels) written by A Isola was published in The oldest among the books (Language) by L Bloomfield was published in 1935, which means that it is more than seventy years old. Majority of the most cited books were published between 1960 and The reason behind this could be that these books are predominantly available in the university libraries. This finding is in line with that of Garfield (1983) and Herubel and Buchaman (1994) in the report that age of materials does not matter so much for humanities scholars. Though age of materials may not be a serious consideration in the materials used by humanities scholars, it is puzzling to observe the near absence of current publications among the most cited books. Could it be that valuable books are no longer being published in the area or the libraries lack the required funds to acquire recent materials? The later may actually be the problem because observations have shown that libraries in Nigeria lack the required finances to acquire research materials as most of them face huge budget cuts. It is even more difficult in the acquisition of humanities research

156 144 materials for attention of most libraries, particularly academic libraries, is focused on the acquisition of materials for science and technological research. Availability of Cited Sources in Nigerian University Libraries The sources identified as most frequently cited were checked in the university libraries under study and findings revealed that only two of the most cited journals are subscribed to by all the libraries. Paradoxically, these two journals Research in African literature and Journal of Linguistics rank tenth and 22 nd in the list. This suggests that citations to a journal may not be entirely a function of its availability but the impact of the journal in the field of study. This is equally reflected in Linguistics Inquiry, which is the most cited journal but is only available in only four of the libraries. Observations during the analysis indicate that the journal is highly cited across the theses in the universities. This finding therefore lays credence the claim of Meho (2007) that citation analysis reflects the impact of a journal or a particular article in a given field. Another finding worthy of discussion is that none of the university libraries has all the most cited journals identified. University of Ibadan library for instance subscribes to fifteen out of the twenty-three journals, while University of Nigeria Nsukka subscribes to fourteen of them and University of Lagos subscribes to only seven of the journals. The university with highest subscription to the journals is University of Ife with subscription to sixteen of the journals, while the university with the lowest subscription is University of

157 145 Benin. These subscriptions may not be too poor but observations during the analysis indicate a lot of gaps in the subscription, which does not guarantee regular access to the journals in these libraries (see appendix II). These irregularities in the subscriptions, according to all the serials librarians, are mainly as a result of poor funding which corroborates Chombow s (2005) remark that government over the years has neglected funding of research in Nigerian languages. Though there is general neglect of research and education in Nigeria, which of Nigerian languages seems more glaring. The unavailability of so many of the journals in the libraries studied corroborates earlier studies by Okafor & Ukwoma (2007) and Ugah (2008) in their investigation of availability of information sources in Nigerian university libraries. It is also very painful to note that almost all these journals are out of subscription in all the university libraries under study. It is only Journal of Linguistics that is currently been subscribed to by three of the university libraries. Subscriptions to all other journals in all the libraries have been cancelled in the 1980s and early 1990s (See Appendix II). This probably accounts for the drop in citations after 1980s. The result of this is that Nigerian language researchers do not have adequate access to current journals in the area and this will likely affect the quality of research in the area. This goes to confirm the earlier claim that when libraries are concerned with budget cuts, the humanities fields are usually the target. The cancellation of subscriptions by all the university libraries is equally a reflection of poor funding of libraries in Nigeria.

158 146 Another source of worry from the findings is that the two Nigerian indigenous language journals in the most cited journals have poor circulation and accessibility. They are subscribed in only two of the university libraries. Similarly very few issues are available in the libraries that subscribe to them. The subscription to the journals has been cancelled in these libraries for over six years now (See Appendix II). The poor subscription to these journals does not encourage readership and subsequently discourages publishers from publication. One is not even sure whether these journals are not out of publication, for one characteristic of Nigeria journals is irregularity in publication. Generally, only few journals have subscriptions in up to three universities. The journals with highest subscription rate amazingly are foreign journals. Research in African Literature, Journal of Linguistics and Anthropological Linguistics are among the highly subscribed journals. The poor subscription to local journals could be as a result of irregularity in journal publication in Nigeria. This is because observations from pattern of journal publishing in Nigeria reveal that most Nigerian journals end with the maiden edition. Some would start publication and have a break for up to two or four years before publication is resumed. Another check was conducted on the availability of the frequently cited books in the university libraries under investigation. Findings reveal that none of the libraries has all the frequently cited books identified. For instance a library with the highest rate of availability of the books is Obafemi Awolowo

159 147 University Ife which recorded twenty seven of the books available while the libraries of Ahmadu Bello University and that of University of Benin have the lowest with only twelve of the books in their holding. This is apparently similar to the findings recorded in the availability of journals. A probable reason for this may be that majority of the most cited books published by Yoruba authors and therefore the possibility of the libraries located in the areas acquiring them is more than other universities outside the environment. Three books are outstanding in terms of availability in all the libraries. These books are Contents and forms of Yoruba Ijala, Oral literature in Africa and Multilingualism, minority languages and language policy in Nigeria. The availability of these books in all the libraries is likely to be a reflection of their importance in Nigerian language research. A careful observation also shows that they have high citation rates in the analysis. Interestingly too, the most frequently cited book Aspects of the theory of language by Noam Chomsky has high availability rate only one of the libraries does not have it in its collection. This is in line with number of citations it received. It is a book that cuts across all linguistics background, which equally explains its high citation and availability rates. Another finding is that some of the books are only available in one or two university libraries. A typical example of such book is Phonology: theory and analysis by L Hyman which is available in only one of the libraries under study. The citations it received is encouraging but its availability is very low. The citation it received despite its scarcity in the libraries gives an impressive

160 148 testimony about its usefulness in linguistics research. Could it be that it is a core text in the area of phonetics? About five other books have significant citations but are available in only two libraries. Apparently researchers may have gotten access to these books through their private collections, loans from friends and through other libraries such as the public libraries. Implications of the Study Nigeria as a developing country requires additional efforts to compete in the global environment. A good way of improving the global competitive readiness of country is the placement of emphasis on research, particularly in the local languages, in terms of provision of adequate and timely information. Unfortunately, the findings of the study have revealed that the Nigerian language researchers lack the necessary information in the libraries for the conduct of their research. For instance, the findings reveal that the Nigerian language researchers rely heavily on the use of books, monographs and journals. Unfortunately, poor funding may have made it difficult for the university libraries to support research in Nigerian languages. This has serious implications for the university librarians, particularly now that emphasis is highly placed on the study of science and technological disciplines to the detriment of the humanities particularly the languages. If the country is really committed to the vision , the continuous neglect of research and the study of Nigerian languages in terms of provision of necessary research materials should not continue.

161 149 The findings also have implications to the Nigerian linguists who are major stakeholders in the study of Nigerian minority languages. A country with over 450 languages cannot afford to place emphasis on the study of very few of them. The findings have shown that only the major Nigerian languages are studied extensively. This does not show enough seriousness in the efforts of Nigerian linguists toward the development of Nigerian minority languages. The neglect of these minority languages poses a grave danger to the country for it could lead to the extinction of these languages, as has happened to so many African languages. The findings have revealed that the Nigerian indigenous languages are rarely used in the writing of theses. This has grave implications for Nigerian linguists particularly the language educators and supervisors. It is a source of worry that more than eighty percent of the theses were written in English language, yet the theses studied Nigerian languages. The best approach in the study of any language is the use of the language in reading and writing. Should this approach continue, the younger generation would not appropriately learn the skills of writing in the local languages and this is a great disservice to the cultural heritage of the country? Apart from this, it will be difficult for publication of books and other research materials in Nigerian languages. By extension Nigerian libraries would continue to lack books published in local languages. The findings reveal low citations to electronic resources. This has some implications for university administrators and librarians. This is because in the

162 150 present information environment emphasis has shifted from acquisition of research materials to accessibility. Information and telecommunications technology has seriously impacted on research in the global environment. It is surprising that the Nigerian language researchers are yet to embrace the internet as a source of research materials. The consequence of this is that access to current materials may continue to be a problem to researchers in Nigerian languages. This also has implications to the researchers themselves for they require some training in the use of information and telecommunication facilities for access to these electronic resources. The findings equally have some implications for authors, publishers and librarians in terms of provision of research materials published in local languages. The findings revealed very low citations to materials published in local languages. This has implications to publishers in terms of accepting manuscripts written in local languages on one hand and authors particularly Nigerian linguists who would be in the fore front of production of research materials in the local languages. Similarly, it has some implications for the Nigerian university libraries in formulation of acquisition policies that will address the research needs of those conducting research in Nigerian languages. The consequence of continuous reliance on materials published in English languages lies in the fact that such research may not adequately convey the message as would research materials published in the local languages. It is quite worrisome that the findings revealed that greater percentage of the citations were to materials about thirty years old. Of course this has some

163 151 implications to the librarians in terms of acquisition of current research materials. Though generally, language researchers make use of old materials, but evidence has shown that libraries lack the required funds to acquire the most current materials. It is very clear to university librarians that much needs to be done in addressing the research needs of language researchers in Nigeria. The continuous reliance on older materials by the language researchers may lead to the production of low quality research. By the findings of the study, librarians are already aware of the journals the Nigerian language researchers frequently cite. The absence of all or some of these journals would be an eye opener for them to begin to modify their acquisition policy so as to accommodate the research needs of the Nigerian linguists. The failure to subscribe to all these or some of these journals portends a grave danger to research productivity in Nigerian languages on one hand and the quality of libraries and library resources in Nigeria on the other hand. The findings on the most frequently cited authors have implications to Nigerian university libraries. Through the findings, the impact of authors in language and linguistics has been established. Since the university libraries are not adequately funded, the findings provide a clear method of selection of research materials in Nigerian languages based on the scholarly impact of the authors. The problem of spending scarce resources on books that may have low research impact would have been a thing of the past. In the same vein, it has implications for the researchers in terms of evaluation of the pedigree of the authors.

164 152 The findings on the most frequently cited books also have implications for librarians and publishers. They indicate to librarians that some books that may be regarded as dated may still be relevant to language researchers. This will equally guide the librarians on what kinds of books are to be retained in time of weeding. In addition to this, librarians may be guided in selection of books for language researchers. Ignorance of these findings may result in the selection of books that would not be relevant to language researchers. Acquiring such books would eat deep into the little funds libraries receive from the funding body. Similarly, this finding has implications for publishers in terms of understanding some research materials that may require reprint for the purpose of providing the research needs of the linguists in Nigeria. Part of the findings is that very few Nigerian local publications were identified in the most cited books. This has grave implications for the linguistics society of Nigeria. The provision of research materials in the local languages is a challenge which they have to face without delay. The findings from the check conducted to determine the availability of the most frequently cited research materials made very worrisome revelations that have serious implication to Nigerian university libraries. The low availability of these frequently research materials and the irregularity of the available journals are glaring indication of the state of Nigerian university libraries in terms of provision of the research needs of the students and the faculty staff. The university librarians and funding bodies would from these findings mainstream on a new funding policy of university libraries where

165 153 some percentage of the library budget is devoted to the acquisition of research materials in the languages. This is important if Nigeria is really preparing to compete favourably in the present knowledge economy. Recommendations Based on the findings of the study, the following recommendations are made: 1. The university libraries in Nigeria should improve on their acquisition of current books and journals since these two resources form the major sources of information resources used by the language researchers. 2. The Nigerian language researchers should step up efforts in research in the minority languages. In order to achieve this, government should institute research grants for linguists who have interest in the study of Nigerian minority languages. 3. The National University Commission should as a matter of policy make it mandatory that all research in Nigerian languages beginning from undergraduate to doctoral levels should be written in Nigerian indigenous languages. Individual universities on their own should emulate the example of University of Ife in presenting research findings in Nigerian languages. 4. Nigerian university libraries should embark on the development of information and communication technology infrastructure required to get them connected to the internet to improve access to electronic resources which is very critical to research and development in the these

166 154 days. The Nigerian students should engage themselves in training required for internet access to improve their literature search skills. 5. Nigerian linguists should as a matter of urgency publish books, journals and other information resources in local languages. Governments at all levels on their own part should sponsor some of the publications and make purchases through special grants to the libraries. 6. Nigerian university libraries should use the identified frequently cited journals as guide in the subscription of journals so as to meet the research needs of Nigerian language researchers. 7. The university libraries should acquire more books in language and linguistics placing emphasis on the authors that have great impact in the discipline. Other current books written by these authors should be given special preference in the selection of books to be acquired. 8. University libraries should improve on the availability of research materials in Nigerian languages by renewing subscriptions to frequently cited journals where subscriptions have been cancelled. This is very important because almost all the frequently cited journals are no longer being subscribed to. 9. In other to promote research in Nigerian languages, the Federal government through the National University Commission (NUC) should make bulk purchases of books on Nigerian languages and distribute same to all the universities that study Nigerian languages.

167 Efforts should be made to rekindle interests of young ones in the study of Nigerian languages in schools. This can be done by making it mandatory that students must have passed at least one Nigerian language at credit level before been admitted in any Nigerian university. Similarly, the study of Nigerian languages should be made part of General Studies programme in Nigerian universities. Suggestions for Further Research This study does not lay any claim for a comprehensive research on information resources used by language and linguistics researchers and availability of such resources in Nigerian university libraries. Based on this the following suggestions are made for further research: 1. A survey study could be carried out to investigate the availability of Nigerian language literature in Nigerian university libraries. 2. A webometric analysis could be conducted to determine the extent of Nigerian language resources available on the web. 3. An attitude study could be conducted to determine the dispositions of librarians toward acquisition of resources published in Nigerian languages. Limitations of the study The major limitations this study suffered are how to get access to the theses which were analyzed because of the unavailability of some of the theses in the university libraries. The reasons given for this are that some of them are being digitized, and in some cases one would be told that the theses were on

168 156 short loan. The researcher had to rely on only the theses available in the libraries at the moment. As a result of this, data collected may not have been as accurate as it would have been without these limitations. Conclusion Language remains a critical factor to the overall national development because it is through the use of languages that scientific and technological development is propelled. The level of development of the national language of a country to a large extent determines the global rating of such country. That is why language research is very important towards the development of any language. Apparently, research in Nigerian languages needs to be supported with reliable, up-to-date, and adequate information. The university libraries are critical to research and development generally and language research in particular. This study therefore sought to find out the kinds of information resources the Nigerian language researchers use in the conduct of their research and the availability of such information resources in Nigerian university libraries. It was found out that the Nigerian language researchers rely heavily on books and monographs in addition to use of journals and other kinds of grey literature such as theses and dissertations. The study also identified some journals and text books which are frequently cited by the researchers. Unfortunately, some of these identified resources are rarely available in Nigerian university libraries. Subscriptions to many of the frequently cited

169 157 journals have been cancelled by almost all the university libraries as a result of lack of funds. Citations to electronic information are very low among the researchers in Nigerian languages which to some extent affect the currency of materials cited. The belief is that many of the researchers lack the required skills for online research and that some of the universities lack the necessary ICT infrastructure to access electronic resources. Evidences from this study indicate that more English language sources are cited than any other languages. Similarly greater percentage of the theses studied was written in English languages irrespective of the fact that the focus of all the theses is on Nigerian languages. It is also interesting to note that age of the materials has little to do with their use in the language research. This is because many of the materials cited are more than twenty years old. The two null hypotheses tested show that there is no significant relationship between the cited sources and their availability in the university libraries. The cited journals were subjected to Bradford s law of scattering and it was discovered that language journals do not exactly follow the law in Bradford s ratio. The findings have several implications for government, the university administrators the university librarians, librarians, the Nigerian linguists, and research students. Based on the findings of the study recommendations were made on how to improve the availability of Nigerian languages research materials in Nigerian university libraries. Some of these recommendations include acquisition of current research materials in the university libraries, provision of ICT infrastructure in the Nigerian university libraries, renewal of

170 158 cancelled journal subscriptions particularly the frequently cited journals and institution of grants by the government for the bulk purchase of Nigerian language publications for the university libraries.

171 159 REFERENCES Abdulkareem, M.Y. (2010) Role public libraries and electronic information media in the promotion of indigenous culture: empirical evidence from Kwara state, Nigeria. Journal of Communication. 1(2), Retrieved on April from Adegbija, E. (2004). Language policy and planning in Nigeria. Current Issues in Language Planning. 5(3), Adegoju, A. (2006). African development: focus on the Nigerian milieu. In E.N Chia (ed.) African linguistics and the development of African communities. Dakar: CODESRIA Afolayan, A. (1989). Language and education: policies, problems and prospects. In T.N/Tamuno & J.A. Atanda (eds.) Nigeria since independence: the first 25 years, vol.3. Ibadan: Heinemann, Aguolu, C.C & Aguolu, I.E. (2002). Libraries and information management in Nigeria. Maiduguri: Ed-Linform Services. Aguyi, A.I. (2006). Preserving Igbo studies through publishing of Igbo language text books. Paper presented at Igbo Studies Association Annual Congress and Conference at University of Nigeria, Nsukka, September. Ahmed, U.B (1989). The cultural context in Nigerian education: the language curriculum. In P. Ekeh & G. Ashiwaju (eds.) Nigeria since independence: the first twenty-five years, vol.vii. Ibadan: Heineman, Ajala, I.O. (2001). Thesis literature citation analysis of Science, Engineering and and Technology students in a Nigerian university. Inter-World Library of Science and Technology, 1(1), Ajulo, S.B. (2000). Nigeria linguistic requirements for the twenty-first century. In S.B Ajulo (ed.) Language in Education and Society. Maiduguri: University of Maiduguri Press Aina, L.O. (1983). Citation characteristics of literature used by agriculturalists in Nigeria. Nigerian Journal of Library and Information Studies. 1(1), Aina, L.O. (1985). Availability of periodical titles cited in the literature of Nigerian s scientific research in Nigerian libraries. Nigerian Libraries 21(1), Aina, L.O. (1991). Directions of information profession in Africa as reflected in the literature. International Library Review. 23(4), Retrieved on 3 rd March 2010 from

172 160 Aina, L.O. (1993). The periodic literature in librarianship in Nigeria: a comparative analysis of author characteristics of local and foreign-based journals. Leading Libraries and Information Centres, 1(20, Aina, L.O & Ajiferuke, I.S.Y. (2002). Research methodologies in information science. In L.O Aina (ed.) Research in information science: an African perspective. Ibadan: Stirling-Horden Aina, L.O. & Mabawonku, I.M. (1998). Manuscripts submitted for publication in the information profession in Africa: a comparative analysis of rejected and accepted papers. Journal of Documentation. 54(2), Amayo, A. (1983). The search for national integration and national identity in Nigeria since independence: the linguistics aspects. Paper presented at the National Confernce on Nigeria since Independence. Ahmadu Bello University, Zaria, March. Ansre, G. (1979). Four rationalizations for maintaining European languages in education in Africa. African Languages. 5(2 ), Apeji, E.A. (1998). Book publishing in Nigeria: a discourses. Libri. 48, Astrom, F. (2008). Patterns in humanities research: scholarly communication exemplified by a bibliographic coupling analysis on comparative literature. Bibliometric Notes. 8(2). Awonusi, V.O. (1993). The English language, national development and unity in Nigeria: a sociolinguistic perspective. Education Today Bachmann-Derthick J., Spurlock S. (1989). Journal availability at the University of New Mexico. Advances in Serials Management 3, Retrieved on April from Baker, David. (1978). Characteristics of the literature used by English musicologists. Journal of Librarianship and Information Science. 10(3), Retrieved on 22/9/08 from Baldwin, V.A. (2009). Using Google Scholar to search for online availability of cited article in engineering disciplines. Faculty Publications, University of Nebraska-Lincoln Libraries. Retrieved on 3 rd April 2011 from Bamgbose, A. (1976). Language in national integration: Nigeria as a case study. Paper delivered at the 12 th West African Languages Congress, University of Ife, Ile Ife.

173 161 Bamgbose, A. (1982). Local languages development policy and practice. In B. Ikara (ed.) Nigeria languages and cultural development. Lagos: National Language Centre, Bamgbose, A. (1991). Language and the nation: the language question in sub Saharan Africa. Edinburgh: Edinburgh University press. Bamgbose, A. (1998). Language as a resource: an African perspective. Workshop paper on the role of African languages in democratic South Africa held at University of Pretoria, 5 th 6 th March. Bamgbose, A. (2006). Linguistics and social responsibility: the challenge for the Nigerian linguists. A key note address delivered on the occasion of the Twentieth Annual Conference of the Linguistics Association of Nigeria (LAN)/ Silver Jubilee held at the Nigeria Educational Research & Development Council, Abuja, November, 14. Barry, C.A. (1997). Information skill for an electronic world: training doctoral research students. Journal of Information Science, 23(3), Retrieved on May from jis.sagepub.com/content/23/3/225.full.pdf Bichteler, J. and Eaton, E. (1980). The combined use of bibliographic coupling and cocitation for document retrieval. JASIST, 31(4): Boyer, C.J. (1972). The Ph.D dissertation: an analysis of dissertation as an information source. Austin, Texas. Brian, D. C. (2005). Trends in the usage of ISI bibliometric data: uses, abuses, and implications. Portal: Library and the Academy, 5(1), Buchanan, A.L & Herubel, J.V.M. (1994). Interdisciplinarity in historical studies: citation analysis of journal of interdisciplinary history. Libres 4(2/3). Retrieved on 24/7/08 from Buttler, L. (1999). Information sources in library and information science doctoral research. Library and Information Science Research. 21, Retrieved on June 8, 2009 from Cameron, B.D. (2005). Trends in the usage of ISI bibliometric data: uses abuses and implications. Portal: Libraries and the Academy. 5(1), Canick, S (2002 Availability of Works Cited in Recent Law Review Articles on LEXIS, WESTLAW, the Internet, and Other Databases, Legal Reference. Service Quarterly. 21(2&3) Carlson, J. (2006). An examination of undergraduate student citation behavior. The Journal of Academic Librarianship 32(1):14-22.

174 162 Case, Mary M., ed Cutbacks in library materials purchasing: OSC Quick SPEC survey, Association of Research Libraries, Issues in Scholarly Communication. [Report on-line]; available from URL: arl.org/scomm/prices.html [viewed 2 March 2002]. Casserly, M.F. & Bird, J.E. (2003). Web citation availability: analysis and implications for scholarship. College and Research Libraries, 64(4), Retrieved on March from Chakava, H. (1996). Publishing in Africa: one man s perspective. Nairobi: Bellago publishing network. Chumbow, B.S (1990). The place of mother tongue in the national policy on education. In E.N Emenanjo (ed.) Multilingualism, minority languages and language policy in Nigeria. Agbor: Central books Ltd Chumbow, B.S. (2005). The language question and national development in Africa. In African intellectuals: rethinking politics, language, gender and development. Dakar: CODESRIA, Council for the Humanities, Arts and Social Sciences. (2005). Measures of quality and impact of publicly-funded research in the humanities, arts and social sciences: final report (draft). Retrieved on 3/5/08 from Council of Graduate Schools. (1991). The Role and Nature of the Doctoral Dissertation: A Policy Statement. Washington, D.C.: Council of Graduate Schools. Cozzens, S.E. (1981). Taking the measure of science: a review of citation theories. Newsletter of International Society for the Sociology of Knowledge. 8, p.16. Cronin, B. (1981). The need for a theory of citation. Journal of Documentation. 37, Cronin, B, Synder, H. & Atkins, H. (1997).. Comparative citation ranking of authors and monographic and journal literature: a study of sociology. Journal of Documentation. 53(3), Cullars, J. M. (1998). Citation characteristics of of English language monographs in philosophy. Library and Information Science Research. 20(1), Dabrishus, M. (2005). The forgotten scholar: classical scholar and periodical use. A Masters Paper Submitted to Faculty of School of Information and Library Science of the University of North Carolina at Chapel Hill.

175 163 Retrieved on 13 June 2008 from Delgadillo, R. & Lynch, B.P. (1999). Future historians: their quest for information. College and Research Libraries. 60(3), Retrieved on June 2008 from Diadoto, V. (1994). Dictionary of bibliometrics. Binghamton, New York: Haworth press. Dike, A. (1992). Scarcity of tertiary books in Nigeria: a threat to academic excellence and suggestions for action. Journal of Librarianship and Information Science, 24(2), Drott, M.C. (1981). Bradford s law: theory, empiricism and the gaps between. Library Trends, Summer, Retrieved on June 2008 from East, J.W. (2006). Ranking journals in the humanities: an Australian case study. AARL. 37(1). Egbokhare, F. (2004). Language and politics in Nigeria. In K. Owolabi (ed.) English and indigenous language in Nigeria. Ibadan: Group publisher Elugbe, B.O. (1990). National language and national development. In E.N Emenanjo (ed.) Multilingualism, minority languages and language policy in Nigeria. Agbor: Central books Ltd Emenanjo, N. E. (2000). Beyond the Rivers reader project. In E Nolue Emenanjo and Ozo-Mekuri Ndimelu (eds.) Issues in African languages and linguistics: Essays in honour of Kay Williamson,(pp ) Aba: National Institute for Nigerian Languages. Emenanjo, N.E. (2006). 25 years of the Linguistics Association of Nigeria: some reflections on the achievements and challenges facing Nigeria and African language studies. Paper presented at the 11 th Annual National Conference of the Association for Promoting Nigeria Languages and Culture(APNILAC) at the Imo State University, Owerri, 20 th 23 rd November. Ezeani, C.N (2005). A citation analysis of the use of grey literature in agricultural research in the Special Collection Division of the University of Nigeria, Nsukka Library. Borno Library, Archival and Information Science Journal. 4(2), 1 6.

176 164 Ezema, Ifeanyi.J. (2010). Journal impact factors and the crisis of scholarly communication in Africa: the dilemma of Nigerian academics. Library Review 59(5). Ezema, Ifeanyi J. & Asogwa, B.E. (2012). Analysis of cited information sources in Nigerian agricultural research with emphasis on animal health and production. International Journal of Library and Information Science, 4(1), 1-9. Retrieved on February from Federal Republic of Nigeria. (1979). The constitution of the Federal Republic of Nigeria Lagos: Government Printer. Federal Republic of Nigeria. (1988). The Nigerian Cultural Policy. Lagos: Government Printer Federal Republic of Nigeria. (1989). The constitution of the Federal Republic of Nigeria Lagos: Government Printer Federal Republic of Nigeria. (1999). The constitution of the Federal Republic of Nigeria Lagos: Government Printer Federal Republic of Nigeria. (2004). National Policy on Education. Lagos: Federal Government Press. Frost, C.O. (1979).The use of citation in literary research: a preliminary classification of citation functions. Library Quarterly. 49(4), Garfield, E. (1955). Citation index for science: a new dimension in documentation through association of ideas. Science. 122, Retrieved on 23/6/07 from Garfield, E. (1971). The mystery of the transposed journal list: where in Bradford s law of scattering is generalized according to Garfield s law of concentration. Essays of an Information Scientist. 1, Retrieved on 23/6/07 from Garfield, E. (1972). Citation analysis as a tool in journal evaluation. Science. 178, Retrieved on 23/6/07 from Garfield, E. (1976). Is the ratio between number of citations and publication cited a true constant? Essays of Information Scientist. 2, Retrieved on 23/6/07 from Garfield, E. (1978a) A&HCI most cited authors. Current Contents. Retrieved on 23/6/07 from

177 165 Garfield, E. (1978b). The 300 most cited authors, , including coauthors at last: how the names were selected. Current Contents. 28, Retrieved on 23/6/07 from Garfield, E. (1979). Citation indexing. Its theory and application in Science, Technology and Humanities. New York: Wiley and Sons. Retrieved on 26/1/08 from Garfield, E. (1983). Is information retrieval in arts and humanities inherently different from that in science? The effect that ISI s citation index for the arts and humanities is expected to on future scholarship. Reprinted in Essay of Information Scientist. Vol. 6, Retrieved on 23/6/07 from Garfield, E. (1993). What citations tell us about Canadian research. Canadian Journal of Library and Information Science. 18(4), 34. Retrieved on 23/6/07 from Garfield, E. (1998a). Mapping the world of science. Paper presented at the 150 th anniversary meeting of the AAAS, Philadelphia, February 14. Retrieved on 23/8/08 from Garfield, E. (1998b). Random, thoughts on citatiology; its theory and practice. Scientometrics. 43(1), Retrieved on 23/6/07 from Georgas, H. & Cullars, J. (2005). A citation study of the characteristics of the linguistics literature. College and Research Libraries. 66(6), Retrieved on 18/8/08 from Gooden, A.M. (2001). Citation analysis of chemistry doctoral dissertations: an Ohio State University case study. Issues in Science and Technology Librarianship, Fall Retrieved on 12may 2009 from Graham, S.R. (2000). Historians and electronic resources: a citation of the journal of the Association for History and Computing. 3(3). Retrieved on 23/8/08 from Hair, P.E.H. (1967). The early study of Nigerian languages: Essays and bibliographies. Cambridge: Cambridge University Press. Harter, S.P. (1996). The impact of electronic journals on scholarly communication: a citation analysis. The Public Access Computer Systems Review, 7(5), Retrieved on May from

178 166 Heinzkill, R. (1980). Characteristics of references in selected scholarly English literary journals. Library Quarterly. 50(3), Retrieved on June from Hemmig, W.S. (2008).The information seeking behaviour of visual artists: a literature review. Journal of Documentation. 64(3) Retrieved on 22/9/08 from Herring, S.D. (2002). Use of electronic resources in scholarly electronic journals: a citation analysis. College & Research Libraries. 63(4)., Hertzel, D.H. (2003). Bibliometric history. In M.A Drake (ed.) Encyclopedia of Library and Information Science vol.2. New York: Marcel Dekker inc Herubel, J.V.M. (1999). Historical bibliometrics: its purposes and significance in the history of disciplines. Library and Culture. 34(4). Herubel, J.V.M. & Buchanan, A.L (1994). 'Interdisciplinarity in Historical Studies: Citation Analysis of the _Journal of Interdisciplinary History_', LIBRES v4n02-3 URL Hickey, R. (2000). Language policy of the Roman Catholic Church in Northern Nigeria: In S.B Ajulo (ed.) Language in Education and Society. Maiduguri: University of Maiduguri, Hicks, D. (1999). The difficulty of achieving full coverage of international social science literature and the bibliometric consequences. Scientometrics. 44(2), Hjorland, B. (2007). Scattering (with Bradford s law). Retrieved on 12/12/08 from Ibeun, M.O. (2001). Applied bibliometrics and cooperative acquisition, as tools for selecting journals and sharing in Nigerian fisheries libraries. African Journal of Libraries, Archives and Information Science. 11(1), Ibeun, M.O & Madu, I.D. (2002). Application of analysis to library management with reference to Nigeria fisheries and aquatic sciences literature. 2(2) Ileperuma, S. (2002). Information gathering behaviour of arts scholars in Srilankan universities: a critical evaluation. Collection Building.21 (2),

179 167 Institute for Scientific Information. (2004). The ISI data base: The journal selection process. Retrieved on 23/9/08 from Iyaro, A.O. (2004). Serial availability, accessibility and utilization: perception of in-training information professionals in Nigerian university. The Nigerian Library Link 11(2), JAMB (2009). Guidelines for admission to programmes/courses in universities. Abuja: Joint Admission and Matriculation Board. Jan, R. (2009). Citation analysis of Library Trends. Webology, 6(1). Accessed on 24/11/09 from Jones, C., Chapman, M. & Woods, P.C. (1972). The characteristics of literature used by historians. Journal of Librarianship and Information Science. 4(3), Kantor, P.B. (1984) Measurement of availability using patron requests and branching analysis. In: Objective performance measures for academic and research libraries. Washington, D.C., Association of Research Libraries Karisson, F. (1994). Linguistics in the light of citation analysis. Publication No. 23. Retrieved on July 16, 2008 from King ei, K. (1999). Language in development research in 21 st century Africa. African Studies Quarterly. 3(3), 1 5. Retrieved on 11/12/08 from Kiondo, E. (2004). Historical practice in managing theses and dissertations at African universities and university libraries Paper presented at the Association of African Universities DATAD workshop on Intellectual Property, Governance, Dissemination and Funding Strategies, Accra, Ghana, February 19-20, Knievel, J.E.& Kellsey, C. (2005). Citation analysis for collection development: a study of eight humanities fields. Library Quarterly. 75(2), Retrieved on 22/9/08 from Koroma, D.S.M (2000). Aspects of language and culture in the defunct Soviet Union: some lessons for Nigeria. In S.B Ajulo (ed.) Language in Education and Society. Maiduguri: University of Maiduguri Press Kumar, S & Kumar, S. (2001). Citation Analysis of Journal of Oilseed Research. Proceedings of 8th International CALIBER , Goa University, Goa, March 02-04,

180 168 Kumars, S & Shah, (1999). Faculty attitude towards library services in Vikran. In J.L Sardema (ed.) Challenge before the university libraries in India in the 21 st century. New Delhi: Indian Library Association, Kushkowski, J.D. (2005). Web citation by graduate students: a comparison of print and electronic theses. Portal: Libraries and the Academy. 5(2), Kushkowski, J.D., Parsons, K.A. & Wiese, W.H. (2003). Masters and doctoral theses citations: analysis and trends of a longitudinal study. Portal: Libraries and the Academy. 3, LaBonte, K.B. (2005). Citation analysis: a method for collection development for a rapidly developing field. Issues in Science and Technology Librarianship. Summer Lascar, C. and Mendelsohn, L.D. (2001). An Analysis of Journal Use by Structural Biologists with Applications for Journal Collection Development Decisions. College Research Library, 62 (5) LaBonte, K.B. & Kristen, B. (2005). Citation analysis: a method of collection development for rapidly developing field. Issues in Science and Technology Librarianship. Summer2005, 1 9. Lawani, S.M. (1981). Bibliometrics: its theoretical foundations, methods and applications. Libri 31(4), Lawani, S.M. (1982). On the relationship between quantity and quality of a country s research productivity. Journal of Information Science. 5(4), Leydesdorff, L. (1998). Theories of citation? Scientometrics 43(1), Marama, I.D. & Ogunrmbi, S.A. (1996). Availability of library and information science collections in Nigerian university libraries. Library Bulletin: Nigerian University Library System 1(2). McCain, K.W. (1981).Pattern of journal use in a departmental library: a citation analysis. Journal of the American Society for Information Science. 32(4), Mc Cain, K.W. (1987). Citation patterns in the history of technology. Library and Information Science Research. 9, Meho, L.I. (2007). The rise and rise of citation analysis. Physics World. Retrieved on 23/7/08 from

181 169 Meho, L.I & Sonnenwald, D.H. (2000). Citation ranking versus peer evaluation of senior faculty research performance: a case study of Kurdish scholarship. Journal of America Society for Information Science. 51(2), Retrieved on 13/9/08 from Mendez, M. & Chapman, K. (2006). The use of scholarly monographs in the journal of Latin American history. Electronic Journals of Academic and Special Librarianship. 7(3). Moed, H.F. (2005). Citation analysis in research evaluation. Berlin: Springer. Narin, F. (1972). Evaluative bibliometrics: the use of publication and citation analysis in the evaluation of scientific activity. Cherry Hill: computer Horizons. Nederhof, A.J. & Zwaan, R.A. (1991). Quality judgments of journals as indicator for research performance in the humanities and the social and behavioural sciences. Journal of the American Society for Information Science. 42, Neuhaus, C. & Daniel, H. (2006). Data sources for performing citation analysis: an overview. Journal of Documentation Nkiko, C. & Adetoro, N. (2007). Pioneer bachelor degree citation analysis of Covernant University students research projects. Library Philosophy and Practice (e-journal). Retrieved on 6/8/08 from commons.unl.edu/libphi/prac/150 Nwadike, I.U. (2006). Language, culture and endangerment in Nigeria. Paper presented at the 11 th Annual National Conference of the Association for Promoting Nigeria Languages and Culture (APNILAC) at the Imo State University, Owerri, 20 th 23 rd November. Nwadike, I.U. (2008). Agumagu Igbo: iduuzi, akuko nkeke, ejije na abu, Paper delivered at the Conference of Igbo Studies Association at University of Nigeria, Nsukka. August Nwadike, I.U. (2008). Igbo studies: from the plantation of West Indies to the forests of West Africa, An inaugural lecture of the University of Nigeria, Nsukka, delivered on June 12, Nsukka: University of Nigeria Senate Ceremonial Committee. Obioha, R. (2007). Igbo language and the extinction theorists. Daily Sun, Thursday 18 th January. Obiosio, E. (2006). UNESCO/NCAC unites to safeguard endangered Nigerian languages: summary of one day workshop on the best practice to safeguard Nigerian languages held in Enugu, August 3.

182 170 Ocholla, D.N & Ocholla, L. (2007). Research in library and information science in South Africa: an analysis of journal research output from Paper presented at the World Library and Information Congress: 73 rd IFLA General Conference and Council rd August, Durban, South Africa. Retrieved on 22/8/08 from Ogunrobi, S.A. (1988). Citation analysis in the Ilorin Journal of Education: an exploratory study. Retrieved on 30/8/08 from Ohiri-Anichi, C. (2001). Nigerian languages in the 21 st century: What vision? In Ogbulogo et al (eds.) Advances in African Languages, Literatures and Cultures: Essays in memory of Nnabenyi Ugonna (pp ). Ohiri-Anichi, C. (2004). Parents choice of language to be first acquired by children in Lagos metropolis, Nigeria. Paper presented at the 24 th West Africa Languages Congress held at University of Ibadan, Nigeria, August 1 6. Ohiri-Anichi, C. (2006). Marginalization of Nigerian languages in nursery and primary schools: path to indigenous language death in Nigeria. Paper presented at the 20 th Annual Conference of the Linguistics Association of Nigeria held at Abuja, 13 th 17 th November. Okafor,V.N & Ukwoma, S.C. (2007). Availability and use of information resources by academics in science and engineering faculties in Southern Nigerian universities. The Information Technologist. 4(2), Retrieved on 3 rd April 2011 from Okafor, V.N & Ukwoma, S.C. (2011). Scholarly communication of academics in science and engineering in Nigerian university libraries. International Journal of Library and Information Science. 3(3), Retrieved on 3 rd April 2011 from journals.org/ijlis Okiy, R.B. (2000) Assessing students and faculty use of academic libraries in Nigeria: the case of Delta State University, Abraka. Frontiers of Library and Information Science 1(1), Okiy, R.B. (2003). A citation analysis of educational dissertations at the Delta State University, Abraka, Nigeria. Collection Building. 22(4), Okwudishu, A.U. (2000). Projecting the nascent national language profile. In S.B Ajulo (ed.) Language in Education and Society. Maiduguri: University of Maiduguri Press Okwudishu, A.U. (2006). Of the tongue-tied and the vanishing voices: implications for African development. In E.N Chia (ed.) African linguistics and the development of African communities. Dakar: CODESRIA

183 171 Olagoke, D.O. (1982). Choosing a national language for Nigeria. Journal of Linguistics Association of Nigeria. No. 1, Olaofe, I.A. (1990). The relevance of linguistics teaching and learning to the implementation of the national language policy. In E.N Emenanjo (ed.) Multilingualism, minority languages and language policy in Nigeria. Agbor: Central books Ltd Omeje, E.O. (2010). Availability and use of grey literature for science and technology research in university libraries in South East Nigeria. Unpublished doctoral thesis, Department of Library and Information Science, University of Nigeria, Nsukka. Omekwu, C O. & Atinmo, M.I. (1998). Author and journal citation patterns of agricultural communication researchers at the University of Ibadan, Nigeria. African Journal of Libraries, Archives and Information Science. 8(2), Onyegegbu, S.O & Otagburuagu, E.J. eds. (2004). University of Nigeria abstracts of theses Nsukka: School of Postgraduate Studies. Orimalade, O. (2005). Challenges of marketing books in indigenous languages. In: Publishing in Indigenous languages: paper presented at the 4 th Nigeria International book fair Osaji, D. (2000). Place of Nigeria in the theory of language choice in education and society. In S.B Ajulo (ed.) Language in Education and Society. Maiduguri: University of Maiduguri Press Osundare, N. (1982). Caliban s curse: the English language and Nigerian s underdevelopment. Ufabamu: Journal of African Activist Association. 10(2), Otubelu, N.J. (2003). Citation analysis of postgraduate researches in the life sciences: the case of University of Nigeria, Nsukka. MLS Dissertation submitted to the Department of Library and Information Science, University of Nigeria Nsukka. Oyelaran, O.O. (1990). Language marginalization and national development in multilingualism. In E.N Emenanjo (ed.) Multilingualism, minority languages and language policy in Nigeria. Agbor: Central books Ltd Peritz, B.C. (1992). On the objectives of citation analysis: problems of theory and method. Journal of the America Society for Information Science. 43(6),

184 172 Peters, S.H. (1990). Characteristics of the sources used by American historians writing on the history of modern Germany. Doctoral Dissertation, Indiana University. Dissertation Abstract International, 52, 330A. Potter, W.G. (1988). Of making many books there is no end: bibliometrics and libraries. The Journal of Academic Librarianship. 14, Pradhan, P, Panda, S & Chandrakar, R. (2011). Authorship pattern and degree of collaboration in Indian Chemistry literature. Proceedings of 8th International CALIBER , Goa University, Goa, March 02-04, Pritchard, A. (1969). Statistical bibliography or bibliometrics? Journal of Documentation. 25(4), Rahm, E. & Thor, A. (2005). Citation analysis of data base publications. SIGMOD Records. 34(4), Retrieved on 12 Dec from Riahinia, N. (2010). A citation analysis study of MA dissertations in the library and information science field in universities in Tehran. Library Review, 59(1), Retrieved on May 6, 2010 from Royahi, H. (1997), Citation analysis study of doctoral dissertations in Arts and Humanities field among two universities in Tehran during Unpublished project, Tarbiat Modares University, Tehran. Rufai, A. (2000). Society, culture and language in Nigeria. In S.B Ajulo (ed.) Language in Education and Society. Maiduguri: University of Maiduguri Press Sam, J. & Tackie, S. N. B. (2007). Citation analysis of dissertation accepted by the department of information studies, university of Ghana, Legon. African Journal of Library, Archives and Information Science, 17 (2): Sassen, C.J. (1992). Citation accuracy in the journal literature of four disciplines: chemistry, psychology, library science and English and American literature. Ph..D Theses, University of North Texas. 1CCBA3C3CC83C8153?id=chamo: Sewell, Robert G Library Materials Budget Survey, Association of Research Libraries, Issues in Scholarly Communication. [Report on-line]; available from URL: 2 March 2002].

185 173 Sharada, B.A (1991). Multiple authorship patterns in linguistics. Library Science with a slant to Documentation and Information Studies. 28(3), Sharada, B.A (1993). Bibliometric studies in linguistics and bibliometric laws. Library Science 30, Smith, D.R. (2007). Historical development of the journal impact factors and its relevance to occupational health. Industrial Health. 45, Retrieved 6/8/08 from Sofunke, B. (1990). National language policy for democratic Nigeria. In E.N Emenanjo(ed.). Multilingualism, minority languages and language policy in Nigeria. Agbor: Central books Ltd Squires, S.J, Moore, M.E & Keesee, S.H. (2007). Electronic journal availability study. Retrieved on January from Stern, M. (1983). Characteristics of the literature of literacy scholarship. College & Research Libraries. 44, Sukovic, S. (2008). Convergent flows: humanities scholars and their interactions with electronic texts. Library Quarterly, 76(3), Retrieved on Feb from t%20flows.pdf Swanepoel, A. (2008). Using citation analysis to determine the use of information sources in the humanities by postgraduate students in the health and biomedical science: a case study. Paper presented at the World Library and Information Congress: 74 th IFLA General Conference and Council, August, Retrieved on 18/8/08 from Tamuno, O.G. (1980). Book production in Lagos, In S.B Aje ed. Library services in metropolitan area of Lagos: background and sociological framework. Lagos: National Library of Nigeria, Thompson, J.W. (2002). The death of scholarly monograph in the humanities: citation patterns in literary scholarship. Libri. 52, Retrieved on June 5, 2006 from Toms, E. & O Brien, H.L. (2008). Understanding the information and communication technology needs of the humanists. Journal of Documentation. 64(1), Retrieved on 22/9/08 from

186 174 Tonta, Y. & Al, U. (2006). Scatter and obsolesce of journals cited in theses and dissertations of librarianship. Library and Information Science Research. Retrieved on June from Trudgill, P. (1983). Sociolinguistics: an introduction to language and society. New York: Penguin. Udofia, U.I. (1997). Selecting veterinary medical periodicals through citation analysis. Library Review. 46 (1-2), Ugah, A.D. (2008). Availability and accessibility of information sources and the use on library services at Michael Okpara University of Agriculture. Library Philosophy and Practice May, 1 9. Ugwuoke, I. (1999). Nigerian languges in danger of disappearing. In E.N Emenajo & P.K Bleambo (eds.) Language endangerment and language and language empowerment. Aba: National Institute of Nigerian Languages UNESCO (1953). The use of vernacular languages in education: monograph on fundamentals of education. Paris: UNESCO. UNESCO (2010). Atlas of the World s Languages in Danger, 3rd ed. Paris, UNESCO Publishing. Online version University of North Carolina at Chapel Hill Health Sciences Library (2007, February). Usability testing final report. Vimala, V & Reddy, V.P. (1996). Authorship pattern and collaborative research in the field of Zoology. Malaysian Journal of Library and Information Science, 1(2), Waters, Lindsay.( 2000). A modest proposal for preventing the books of the members of the MLA from being a burden to their authors, publishers, or audiences. PMLA 115(3): Retrieved on 12 Dec from Waugh, C.K. & Ruppel, M. (2004). Citation analysis of dissertation, theses and research paper references in work force education and development. Journal of Academic Librarianship. 30, Retrieved on 5 th July 2010 from Weinstock, M. (1971). Citation index. In Allen Kent and Harold Lancour (eds.) Encyclopedia of Library and Information Science vol.5 New York: Marcel Dekker,

187 175 West African Examination Council. (1997). Senior certificate examination (Nigeria): statistics of entries and results, Unpublished Report. WAEC: Yaba, Lagos. White, E.C. (1985). Bibliometrics: from curiosity to convention. Special Libraries. 76, Retrieved on 5 th July 2010 from White, H.D. (1981). Co cited author retrieval online: an experiment with social indicator literature. Journal of American Society for Information Science. 32(1), Wiberly, S. & Jones, W.G. (1989). Patterns of information seeking in the humanities. College and Research Libraries. November, Williamson, K. (1990). Development of minority languages: publishing problems and prospects. In E.N Emenanjo(ed.). Multilingualism, minority languages and language policy in Nigeria. Agbor: Central books Ltd Zainab, A.N & Goi, S.S. (1997). Characteristics of citations used by humanities researchers. Malaysian Journal of Library and Information Science. 2(2), Retrieved on March 6, 2009 from Zandian, F., Rezazadeh, E. & Hassanzadeh, M. (2007). Citation analysis study of doctoral Arts and Humanities dissertations in Tarbiat Modares University during Unpublished dissertation, Tarbiat Modares University, Tehran. Zhang, Yin. (1998). The impact of internet-based electronic resources on formal scholarly communication in the area of library and information science: a citation analysis. Journal of Information Science.24 (4), Retrieved on January 2009 from jis.sagepub.com/content/24/4/241.full.pdf Zipp, L.S. (1996). Thesis and dissertation citations as indicator for faculty research use of university library journal collections. Library Resources and Technical Services. 40, Retrieved on January 10, 2008 from Zwaan, R.A & Nederhof, A.J. (1990). Some aspects of scholarly communication in linguistics: an empirical study. Language. 66, Retrieved on December from

188 176 APPENDIX I List of Cited Journals Rank Journals Frequency of Citations Cumulative citation 1 Linguistics Inquiry Nigerian Magazine Journal of W/African languages Language Studies in African Linguistics 156 1,136 6 Yoruba: A Journal of Yoruba Studies Association ,267 7 International Journal of the Sociology of Language 127 1,394 8 Odu: Journal of African Studies 125 1,519 9 Language Learning 102 1, Research in African Literature 84 1, Journal of Afr. Lang & Linguistics 79 1, Linguistics Analysis 77 1, Anthropological Linguistics 76 1, Phonotica 73 2, Word Journal of Linguistics Circle 72 2, Linguistics 71 2, Langbasa 68 2, Journal of Ling. Ass. of Nigeria 64 2, Lingua 63 2, Yoruba Gbode 61 2, African Notes 60 2, Journal of Linguistics 56 2, Journal. Of Nig. Lang & Literature 50 2, Research in Yoruba Language and Lit 49 2, Phonology Applied Linguistics Current Trends in African Linguistics African Lang Studies Nsukka Jnl of Afr Lan & Ling TESOL Quarterly International Rev of Applied Linguistics Ihafia: a Journal of African Studies Conch Ling. & Philosophy Anu Journal of Igbo Culture Inquiry in African Language & Literature Canadian Journal of Linguistics International Journal of American Linguistics Journal of American Language and Literature Ibadan Journal of Semantics Studies in Linguistics English Linguistics Linguistics Today Linguistics Science Papers on Linguistics Working Papers in Linguistics Applied Linguistics Linguistics Theory

189 177 Rank Journals Frequency of Citations Cumulative citation 50 Nigerian Language Studies Journal of Asian and African Languages Current Trends in Linguistics International Journal of General Linguistics West African Languages in Education Linguistics Review Current Trends in African Languages Advances in African Language and Culture Igbo Journal of SPILC Nsukka Studies in African Literature Journal of Phonetics Canadian Language Review Journal of Language Art and Communication Ilorin Journal of Language and Literature Nigerian Language Teacher African Journal of Language Research Word Stress Language and Speech Journal of American Linguistics Nigerian Field The Conch Nordic Journal of African Studies Ife: African Language and Literature World Englishes Work in Progress African Language and Culture Research Notes Journal of Literary Studies Black Orpheus Review of English and Literary Studies Proverbium Review of National Literature Igede: Journal of Igbo Studies Olukun: Journal of Yoruba Culture Lagos Notes Odenigbo Nka: Journal of the Art Ikenga: Journal of African Studies Uwa Ndi Igbo Journal of American Folklore Nsukka Journal of Humanities Journal of Igbo Life and Culture Total 4521

190 178 APPENDIX II Periods of journals acquisition in the university libraries S/N Journals No cit Universities UI UNN UNILAG OAU ABU BEN 1 Linguistics Inquiry Nigerian Magazine Journal of W/African languages Language Studies in African Linguistics Yoruba: a Jnl of Yoruba Studies Assoc Int Jnl of the Sociology of Lang Odu: Journal of African Studies Language Learning Research in African Literature Journal of Afr. Lang & Linguistics Linguistics Analysis Anthropological Linguistics Phonotica Word Journal of Linguistics Circle Linguistics Langbasa Journal of Ling. Ass of Nigeria Lingua Yoruba Gbode African Notes Journal of Linguistics date date date Journal. Of Nig. Lang & Literature

191 179 APPENDIX III CITATION ANALYSIS OF NIGERIAN LANGUAGE RESEARCH DOCUMENT ANALYSIS GUIDE (CANLRDAG) Type of sources cited in Nigerian language research. MATERIALS NUMBER OF CITATIONS Books/monographs Journals Theses/dissertations Government publications Conference/seminar papers News papers/magazines Others (specify) 2. The language spread of theses studied Hausa Igbo Yoruba Edo Efik Fulfulde Ibibio Idoma Igala Kanuri Nupe Languages Others (specify) Number of Theses

192 The languages in which the theses are written Languages Number of Theses English Hausa Igbo Yoruba Edo Efik Fulfulde Ibibio Idoma Igala Kanuri Nupe Others (specify) 4. Yearly distribution of citations to electronic resources in Nigerian language research Total Year Number of electronic citations 5. Frequency of languages of cited sources in Nigerian languages. English Hausa Igbo Yoruba French Languages Other Nigerian Languages (specify) Total Number of Citations

193 The timelessness of cited sources in theses on Nigerian language research. Year Number of Citations Total 7. Frequency of cited journal titles in Nigerian language research. Journals Number of Citations. 8. Most frequently cited authors Authors Number of Citations.

194 Frequently cited books in Nigerian language research Books Author Year of pub. Number of Citations.

195 183 APPENDIX IV AVAILABILITY OF FREQUENTLY CITED JOURNALS IN NIGERIAN LANGUAGE DOCUMENT ANALYSIS GUIDE (AFCJNLDAG) A. UNIVERSITY OF NIGERIA LIBRARIES S/N FREQUETLY CITED JOURNALS NUMBER OF CITATIONS B. UNIVERSITY OF IBADAN LIBRARIES S/N FREQUENTLY CITED JOURNALS NUMBER OF CITATIONS C. UNIVERSITY OF LAGOS LIBRARIES S/N FREQUENTLY CITED JOURNALS NUMBER OF CITATIONS NUMBER OF ISSUES AVAILABLE NUMBER OF ISSUES AVAILABLE NUMBER OF ISSUES AVAILABLE

196 184 D. OBAFEMI AWOLOWO UNIVERSITY LIBRARIES S/N FREQUENTLY CITED JOURNALS NUMBER OF CITATIONS E. AHMADU BELLO UNIVERSITY LIBRARIES S/N FREQUENTLY CITED JOURNALS NUMBER OF CITATIONS NUMBER OF ISSUES AVAILABLE NUMBER OF ISSUES AVAILABLE F. UNIVERSITY OF BENIN S/N FREQUENTLY CITED JOURNALS NUMBER OF CITATIONS NUMBER OF ISSUES AVAILABLE

197 185 S/N Observation of the validators 1 The languages in which the theses were written are necessary to be included. 2. It is very important to be exhaustive in items numbers 1,2,3 and 5 in the first part of the instrument. 3. It is necessary to find out the citations on each year instead of grouping the years into ten years period in item number 5 of the first part of the instrument. 4 The use of number issues and latest issues in the in the second part of the instrument is inadequate. APPENDIX V VALIDATORS COMMENTS S/N Modification suggested 1 The researcher should include the languages in which the theses were written as item number 3 in the first part of the instrument. 2 The researchers should add others, please specify in items numbers 1,2,3 and 5 of the instrument to make the list exhaustive. 3. In order to find out the yearly citations, the researcher should include individual years instead of grouping them into periods. 4. The researcher should replace number of issu with number of issues cited while latest issues should be replaced with number of issues available in the second part of the instrument. Remarks This was done as suggested. This was done as suggested This suggestion was accepted. This was done as suggested.

198 APPENDIX VI LANGUAGE MAP OF NIGERIA 186