IFLA

As of 22 April 2009 this website is 'frozen' in time — see the current IFLA websites

This old website and all of its content will stay on as archive – http://archive.ifla.org

IFLANET home - International Federation of Library Associations and InstitutionsAnnual ConferenceSearchContacts

62nd IFLA General Conference - Conference Proceedings - August 25-31, 1996

A Bibliometric Study Of Library And Information Research In China

Cheng Huanwen
Department of Library and Information Science
Zhongshan University, Guangzhou, Guangdong 510275
People's Republic of China


ABSTRACT

This paper, which is modeled on the studies done by Jarvelin and vakkari in 1990 and 1993, is a content analysis of library and information science (LIS) research in China from 1979 to 1994. The aim is to study how research articles were distributed over various topics, what research methods were applied, and what the similarities and differences between LIS research in China and in the world are . The study samples comprise 1930, 2447, and 2665 articles published in the core LIS journals respectively in 1985, 1990 and 1994. The largest groups of articles from each year concerned the basic theory of LIS (26%-32%) and information service (20%-25%). The most popular research strategies were historical method (25%-19%) and not applicable (16%-14%), whereas the least popular were experiments (0.2%-0.5%) and survey method (4%-1.6%). The main difference is that LIS research in China trended towards theoretical research, but in the world towards applied research.


PAPER

1. INTRODUCTION

This paper analyzes formally published journal articles in library and information science (LIS) in China from 1985 to 1994. The aim of this paper is to study: (1) how the research articles are distributed over various topics; (2) what research methods were applied; (3) what the similarities and differences between LIS research in China and that in the world are (especially at the international l evel). Such an analysis reveals the foci of LIS research in China, their coherence, changes, and neglected areas and, the similarities and differences of between LIS research in China and abroad. Such a study contributes to a better understanding of what LIS in China is and how it may evolve.

Up to now, various contact analyses of LIS research publications have been made by many scholars around the world, such as Wersig, Neveling, Peritz, Nour, Schrader, Feehan, Gragg, Havener, Kester, Hauser, Atkins, Jarvelin, Vakkari, etc.. These content analyses of LIS research publications can be roughly divided into two types in several ways: One method distinguishes between analyses , analyzed s amples published in the short term (generally in one year) and in the long term; these analyses can also be divided at the national and international levels; or, in the comprehensive fields of LIS and in the limited fields of LIS. Although these studies have some deficiencies, they all contributed to the development of this research subject.

In the interests of continuity and comparability, this paper is mainly modeled on the research articles of two well-known scholars in this research field: Kalervo Jarvelin and Pertti Vakkari, in 1990[1], and 1993[2]. Because of different conditions in every country, the development of LIS research is different and uneven between developing countries and developed countries, even among countries a t the same level of development. In this article, therefore, I will try to supplement or correct the research model of Jarvelin and Vakkari according to the real conditions of China.

In general, the content analysis model of Jarvelin and Vakkari consists of three parts:

  1. the distribution of the articles over topics;
  2. the approaches: viewpoints on information dissemination and social levels; and,
  3. the methods: research strategies and data collection methods. It is difficult to identify the categories of the authors of LIS articles in China (i.e. whether they be librarians, producers, users, etc.) and research strategies naturally reflect the chosen method of data collection. Therefore, when analyzing journal articles in LIS this paper only uses two parts of Jarvelin and Vakkari's mode l:
    1. the distribution of the articles over topics, and,
    2. the research strategy.

The data for this paper was collected from both research and professional articles (non-research-based articles) published in most of the core LIS journals in China. However, the aim of this paper is to study the research articles; thus, the professional articles are only analyzed as compared with the analysis of research articles in Section 3.1.

2. METHODS

2.1 The Classification Scheme

In general, there are two kind of methods used in designing classification schemes for content analyses of LIS articles. The first method designs the scheme completely on the basis of the articles used, like as Atkins did, or mainly on the basis of the articles used, such as Jarvelin and Vakkari did. The other method designs it on the basis of the essence of LIS. When using the first method, th e classification scheme is created after analyzing articles, and can only provide an overall picture of LIS research articles in a limited period but not an overall picture of LIS research. Therefore, it is unable to accurately reveal the overall situation of LIS research, especially its neglected areas. When using the second method, the classification scheme is created before analyzing articles , so it is able to accurately reveal the overall situation of LIS research. Even so, so far there has not been a detailed classification scheme of LIS research for this previously created type which is universally accepted throughout the world. Therefore, using the second method of classification, I will try to design a simplified LIS classification scheme(cf. Appendix B), based mainly of my unde rstanding of the systematic structure of LIS, partly on the China Library Classification[3], partly on The Classification Scheme of Jarvelin and Vakkari in 1993, and partly on The Paper Index of Library, Information, Archive, Publication and Distribution Science(1949-1985)[4] and other similar indexes in China. The main classes of LIS topics are as follows:

Every class contains subclasses (cf. Appendix B). Each article in my sample was classified into a single subclass. When an article had many topics, I tried to determine its main topic.

Because "research strategy is an overall approach to the study within which, for example, the decisions concerning data collection and the type of analysis are made"[5], the description of methods only includes research strategies (cf. Appendix B) but not data collection methods, types of analysis, or types of investigation. The list of possible research strategies consists of historical method, survey method, qualitative method, comparative method, logical method, bibliometric method, other mathematical method, concept analysis, experiment, literature review, other method, and not applicable (no method). Each article was also classified into one of these research strategy classes.

2.2 The Data

Jarvelin and Vakkari verified in 1990 and 1993 that limiting data to journal articles may cause some bias in study results; however, it is generally true that even when a study is published as a monograph, it is at some point also reported in journals. Therefore, it is possible to make relatively reliable inferences concerning LIS research on the basis of journal articles. Because it is difficult to identify the publications which contain research fulfilling the criteria that determines what LIS research is, research publication in the field are ostensibly defined by choosing articles from the core scientific journals in LIS.
2.2.1 Periodization.
The history of LIS research in China in modern times, namely in the 20th century, can be divided into six periods:
  1. the embryonic period (1900-1924);
  2. the rising period (1925-1937);
  3. the period of decline (1938-1949);
  4. the period of transformation(1950-1965);
  5. the period of suspension (1966-1978);
  6. the golden period(1979~). The golden period can be also divided into three phases:
    1. the recovery phase(1979-1985);
    2. the flourishing phase (1986-1990); and
    3. the phase of further development (1991~).

For my content analysis focused on the golden period, which is the most representative period of LIS research in China, due to its relative temporal proximity to the present and, its comparatively high state of development. I took the samples from the years 1985, 1990 and 1994 for the analysis, because these years represent the research peak of each phase in the period, and basically reflect the changes in LIS research that have taken place in China over the past 16 years.

2.2.2 Selection of journals.
According to statistics from A Guide To The Core Journals Of China (GCJC)[6], there were 92 LIS journals in China (the real number is 89, which excludes 3 repeatedly-counted journals), including 62 journals in library science (LS) and 30 in information science (IS), in 1992. According to statistical analysis [7], the GCJC selected a total of 32 core journals in LIS (the real number is 29, which e xcludes 3 repeatedly-selected core journals), including 19 core journals in LS and 13 in IS.

On the other hand, the Library Society of China(LSC) chosen 12, 9 and 12 excellent journals in LS from 72 journals in LS, respectively in the years 1989, 1993, and 1995, by all kinds of statistical figures and secret balloting of the members of the Editorial and Publishing Committee and Journal Research Group of LSC[8]. The 12, 9, and 12 excellent journals in LS of the years 1989, 1993, and 1995 are just among the top 12 of the 19 core journals in LS selected by the GCJC (excepting 1 journal which was ranked 18th); these excellent journals from the years 1989, 1993, and 1995 have basically not changed (except for 1 journal). Therefore, I regard these 12 excellent journals of LS as the most representative core journals in LS research in China.

According to the above, 13 core journals out of 19 core journals in LS (which includes 12 excellent journals and 1 journal which was ranked fifth on the list of 19 core journals), and 10 core journals out of 13 core journals in IS (3 core journals in IS were repeatedly selected in 13 the core journals in LS), were selected for the study. Therefore, the study samples contain altogether 23 journals (cf. Appendix A). At the time of collecting the source data, altogether the 9 volumes from the 23 journals were not available, for various reasons.

2.2.3 Selection of articles and the basis of analysis.
From the journals selected, I included only full-length articles in the sample. Like Javelin and Vakkari, I excluded editorials, letters to the editor, newslike reviews, personal and conference news, and advertisements. All the materials which were analyzed were whole articles.

3. FINDINGS

The samples comprise 1930, 2447, and 2665 articles published in the core LIS journals respectively in the years 1985, 1990, and 1994, as indicated in Table 1.

Table 1. Division of the source data among article types of 1985,1990, and 1994
Not Available, please contact Author

In the three yearly data subsets, the proportion of research articles was above 50%, and the proportion of research articles in the last two yearly data subsets increased by nearly 10% over the year 1985, This finding is consistent with the conclusions of Jarvelin and Vakkari in their 1990 and 1993 studies, namely, that proportion demonstrates the effectiveness of this purposive selection of arti cles, and that the share of research articles has grown.

3.1 The Distribution of Research Articles over Topics

3.1.1 The distribution of the articles over the main classes of LIS topics (Table 2).

Table 2. Topic distribution among main classes in the research articles in 1985, 1990, and 1994
Not Available, please contact Author

The Table 2 shows that the largest body of the articles in each year pertained to the basic theory of LIS (25.7%-31.9%); the second largest was Information service (IS, 20.1%-25%). Their shares of the total increased steadily from 1985 to 1990 but decreased in 1994, a timeframe which corresponds exactly with the division of the three phases of the golden period of LIS research in China (Section 2.2.1). The proportion of these two largest groups in my study was nearly identical to that of the study of Jarvelin and Vakkari in 1993: that the largest body of articles belonged to the category of information storage and retrieval (IS & R, 26.2%-32.4%), and the second largest was library and information service activities (L & ISA, 25.4%-27.2%). The in class of the second largest group (IS) in my research was also identical with that (L & ISA) of the study of Jarvelin and Vakkari. However, the main class of the largest (the basic theory of LIS)was completely different from that (IS & R) of the study of Jarvelin and Vakkari. The differences demonstrate the differences between LIS research in the western world and in China; that is, the tradition of LIS research in China focuses mainl y on the study of so-called pure LIS theory, and in the western world on the study of L & I practice.

The shares and proportion of the third largest group (related disciplines) decreased steadily, which shows that, the traditional disciplines, such as documentation, book history, bibliography, and textual criticism, are still as important today in China they have been historically. These decreasing trend is due to a rise in other research fields of LIS after 1980s.

The shares of information process and L & I industry from each year seemed quite stable, which demonstrated that the two main topics appeared to be gradually interesting to researchers.

On other topics there were considerably fewer articles. The shares of information storage and preservation (4.2%-10.2%), education in LIS (2.9%-3.8%), other study (1%-2.2%), and other LIS topics (0.4%-0.5%) were also relatively steady, although the shares of information storage and preservation sharply increased from 4.2% in 1985 to 10.2% in 1990 because of the nation-wide program of library coll ection survey around 1990. In contrast to the study of Jarvelin and Vakkari in 1993, namely, that the growth of "other LIS topics" from 1% in 1965 to 10% in 1985 is remarkable and may indicate a gradual movement from traditional to new research topics in LIS in the world, the shares of "other LIS topics" was quite stable (0.4%-0.5%) from 1985 to 1994 and may either indicate a basically stable str ucture of LIS research topics, or a small increase in new research topics in China.

3.1.2 The topic distributions in the subclasses.
The subclasses of the top 4 main classes (Table 2) are analyzed as follows. The figures are drawn from Appendix C.

Within the basic theory of LIS, the most popular sub-fields have been studies of the principles of LIS (15.6%-15.8%-14.8%), and the principles of L& I management (4.3%-7.3%-5.5%). The branch disciplines of LIS (3.1%-4%-3.2%) and analyses of LIS (1.7%-3.4%-3.3%) appear to be more and more interesting to researchers. Compared with the study of Jarvelin and Vakkari in 1993, the principles of L & I m anagement was also one of the most popular sub-fields, and less attention is paid to methodology (1%-1.4%-0.9%) .

Within information service, the most popular sub-fields have been studies of automated information retrieval (5.5%-6.6%-5.6%), the general theory of information service (2.9%-5.3%-4%), and reference service (4.1%-4.6%-2.8%). While the new topic of automated information retrieval has gained in popularity since 1985 (5.5%-6.6%-5.6%), the traditional topics of conventional information retrieval and circulation and readers' services have lost (2.1%-2%-1.3%, and 4.4%-1.2%-1.3%). Studies of users and services have gradually increased (1.9%-2.5%-2.2%), whereas user education has gradually dropped (2.1%-1.7%-1.3%), less attention is paid to these two sub-fields. Library buildings and facilities has been a neglected area (1.1%-0.9%-1.2%).

Among analyses of related disciplines, the strongest emphases have been put on literature review (3.9%-4.7%-3.3%), book histories (5.1%-3.7%-2.1%), and bibliography science (6.9%-1.9%-2.4%), although the trend of these last two sub-fields is declining. Less attention has been paid to the most conventional sub-fields of documentation science (1.4%-0.6%-1.6%) and textual criticism (1.6%-1.3%-0.9%).

In the category information process, classification and indexing (8.3%-5.4%-7.5%) and cataloguing (2.7%-2.2%-3.5%), as both basic and conventional information processes, have been the most popular sub-fields. Although both bibliographic database (0.4%-1.6%-1.9%) and non-bibliographic database (0%-0.3%-1.3%) appear to be less interesting to researchers, clearly the trend is for them to increase, w hich indicates a gradual movement from traditional to new research topics.

3.2 Most frequent topics of research articles

The summary of the most popular topics of research articles in 1985, 1990, and 1994 is shown in average-ranked order as follows(Table 3).

It is clear that the principles of LIS was the most popular topic from 1985 to 1994, which further demonstrates the differences between LIS research in the western world and in China, namely, that much more attention is paid to the so-called pure theory of LIS in China than in the western world.

Similar to the study of Jarvelin and Vakkari in 1993, there are clear trends: the proportion of classification and indexing has decreased, while the proportion of automated information retrieval has grown steadily. Classification and indexing, automated information retrieval, and information collection and storage are continuously among the most popular topics. Compared with the analysis results of Jarvelin and Vakkari in 1993, which concluded that publishing and book history were popular only in 1965 (the periodization was from 1965 to 1985), the conventional research topics in China of bibliography science, and book history were popular only in 1985. This may indicate that the movement from traditional to new research topics in LIS in China has lagged behind that of the western world b y about 20 years.

Table 3. Summary on most popular topics in the articles in 1985, 1990, and 1994(%)
Not Available, please contact Author

Contrary to the analysis of Jarvelin and Vakkari in 1993, which stated that the analysis of LIS decreased popularity between 1965 and 1985, the analysis of LIS gradually gained in popularity between 1985 and 1994 in China, which may also indicate that LIS research in China attaches major importance to the so-called pure theory of LIS.

It is well worth noticing that the share of information industry in China grew sharply from its original rank of the fifteenth (1.4%) in 1985 to the rank of second (8.58%) in 1994, which may show the great impact of "the information superhighway" on LIS research in China.

3.3 Research strategies

The distribution of research strategies in the articles from the years 1985, 1990, and 1994 is shown in Table 4; the most popular research strategies are shown in Table 5.

Table 4. Research strategies in the articles in 1985, 1990, 1994
Not Available, please contact Author

Table 5. Most popular research strategies in the articles in 1985, 1990, and 1994
Not Available, please contact Author

The most popular strategy in each year was historical method (25%-18.9%-18%). The study of Jarvelin and Vakkari in 1993 proved that, for a long time, historical method was the only scientific method for dealing with problems in the field; only beginning in the 1950s did it begin to lose ground. It is still a fairly strong tradition in LIS (the fourth most popular) in the western world. Although h istorical method has decreased slightly in popularity, it is still the strongest in LIS in China, which indicates that historical method, the most conventional strategy, still plays a leading role in LIS research in China.

To LIS scholars in other countries, it is hardly conceivable that the second most popular research strategy in China was not applicable, or no method (15.5%-14.4%-20.9%)! This unexpected finding indicates that many scaled researchers in LIS in China have not mastered the methodologies of scientific research, and many research articles have remained at a low level.

What is surprising is that these findings are completely absolutely different from the analysis of Jarvelin and Vakkari in 1993, which found that the survey method was the most frequently used empirical strategy (22.5%-20.3%-22.9%) in the western world; however, it has been very rarely used in China (2.4%-4%-1.6%). On the other hand, the qualitative method was very rare (0.7%-0%-1.6%) in the worl d, but it was frequently used in China (5%-7.2%-10.9). These differences may indicate that much attention was paid to positive research in the western world, but that a considerably strong emphasis was put on uncritical theoretical research in China.

It is encouraging that this phenomena of the emphasis on uncritical theoretical research appears to be changing: the mathematical method, a new research strategy, rose in China in the early 1980s and is continuously among the most popular strategies. In addition, the share of Bibliometric method has grown steadily (2.2%-5.9%-4.7%).

Similar to the study of Jarvelin and Vakkari in 1993, logic method and concept analysis are continuously among the most popular strategies, and experiments were extremely rare in the research articles(0.3%-0.5%-0.2%)as well.

3.4 The application of strategies on topic

Historical, not applicable, and mathematical strategies were, as a rule, the most popular strategies within the different topics(Table 6). The notable exceptions to this rule are the basic theory of LIS with not applicable and bibliometriic method, and information process with concept analysis, historical and other methods as the main strategies.

The three topics of the basic theory of LIS (22%-17%-29%), information service (27%-24%-28%), and other LIS topics (50-%-43%-63%), are dominated by the not applicable strategy, which indicates that most researchers at lower levels are in these research fields.

The conventional topics of Library and information (68%-60%-41%), education in LIS (49%-56%-25%), and related disciplines (45%-35%-40%), are understandably dominated by the historical strategy.

Table 6. Most popular research strategies in LIS topics in 1985,1990, and 1994(%)
Not Available, please contact Author

In research on information storage and preservation, the use of strategies other than mathematical strategy, which was consistently the most frequent was completely different in each year of the study. In research on information processes, however, the frequent use of different strategies was in the stable order of concept analysis, historical, and other strategy.

4. DISCUSSION

The main finding from article samples from the years 1985,1990, and 1994 is the remarkable similarity of the distributions. The largest study areas, in order of popularity, were the basic theory of LIS, information service, and related disciplines. In combination they covered over 60% of the research articles. The relative size of any other field of LIS was basically below 10%. All of these findi ngs indicate that the foci of LIS have not changed greatly from 1985 to 1994 in China. On the other hand, the main fields of LIS research in China were different from those (IS & R, and L & I service activities) in the world, which suggests that LIS research in China focuses much on theory and history, whereas in the world the focus on practice. In a word, LIS is not a pure discipline but an appl ied one; therefore, the focus of LIS research in China should also move to practice.

Although the trends of research on bibliographic databases (0.4-%-1.6%-1.9%) and non-bibliographic databases (0%-0.3%-1.3%) are rising, the proportion of the articles contributed to these areas is still very small. With the development of library automation, LIS researchers in China should attach greater importance to research in these neglected areas.

Similar to the findings of the studies of Jarvelin and Vakkari, the methodology of research in LIS and the analysis of LIS have also received little attention in China. Jarvelin and Vakkari concluded in 1993 that both methodological discussion and analysis of the foundation of the discipline are prerequisites for a more varied use of research strategies and a more general articulation of research problems. These topics, therefore, should also receive more attention.

The most frequent research strategies were historical method and not applicable, whereas the least frequent were experiments and survey method, which is very different from the finding of Jarvelin and Vakkari in 1993. In spite of the fact that lots of research strategies from the natural sciences have been applied to social science research, and, as Jarvelin and Vakkari indicated in 1990, the fi eld of LIS research in the world is so survey-oriented that almost all problems are seen from a survey viewpoint, this is not the case in LIS in China. This may indicate that the selection of research strategies by LIS researchers in China trends towards convention and conservativeness. In order to the improve LIS research in China, it is essential to strengthen research on the methodology of LI S research, to use the experiences of other countries as a reference, and to introduce all sorts of new available methods from other disciplines at home and abroad into LIS research in China.

In a word, the best way of improving and enriching the quality of LIS research in China is to smash the bonds of tradition, and to go along with the world trends of LIS research.

Acknowledgment

The author is eternally grateful to Professor Beverly P. Lynch for her invitation to submit this article, and for providing reference materials.

REFERENCES

[1] Kalervo Jarvelin, Pertti Vakkari. Content Analysis of Research Articles in Library and Information Science. Library and Information Science Research, 12 (1991): 395-421.

[2][5] Kalervo Jarvelin, Pertti Vakkari. The Evolution of Library and Information Science 1965-1985: A Content Analysis of Journal Articles. Information Processing & Management, 29:1(1993): 129-144.

[3] The Editorial Board. China Library Classification (3rd edition). Beijing: Bibliography Press. 1990.

[4] The College of Library and Information Science, Wuhan University. The Paper Index of Library, Information, Archive, Publication and Distribution Science (1949-1985). Beijing: Scientific and Technological Literature Press. 1991.

[6] Zhuang Shoujing (editor-in-chief), A Guide To The Core Journal Of China. Beijing: Beijing University Press. 1992.

[7] The method used for selection of the core journals is as followers. First, for every Chinese journal over a period of three years(1988-1990), a statistical count of the total number of articles presented by the journal, the number of articles abstracted by relevant abstract journals, and the number of articles cited by other selected journals in the field was performed. Secondly, a weighted mean was calculated for these three statistical figures. Finally the journals in each field were arranged by weighted mean. Those journals at the top level of all ranks were selected as core journals in the corresponding field.

[8] Ling Tong. The Excellent Journals In Library Science In China. Library Tribune, 6(1995):76.

APPENDIX A: LIST OF JOURNALS

  1. The Journal of the Library Science in China, 1985, 1990, 1994
  2. Library and Information Service, 1985, 1990, 1994
  3. The Library Journal, 1985, 1990, 1994
  4. Journal of Academic Libraries, 1985, 1990, 1994
  5. Researches In Library Science, 1985, 1990, 1994
  6. Library Development, 1985, 1990, 1994
  7. Journal of Library and Information Science, 1985, 1990, 1994
  8. Knowledge of Library and Information Science, 1985, 1990, 1994
  9. Library, 1990, 1994
  10. Library Theory and Practice, 1985, 1990, 1994
  11. Library Work and Study, 1985, 1990, 1994
  12. Library and Information, 1985, 1990, 1994
  13. Library Tribune, 1985, 1990, 1994
  14. Journal of the China Society for Scientific and Technical Information, 1985, 1990, 1994
  15. Journal of Information Science, 1985, 1990
  16. Information Science, 1985, 1990, 1994
  17. Information Professional Research, 1990, 1994
  18. China Information Review, 1985, 1990, 1994
  19. Information Science and Technology, 1990, 1994
  20. Information Science and Technology, 1990, 1994
  21. Journal of Information, 1994
  22. Information Knowledge, 1985
  23. New Technology of Library and Information Service, 1985, 1990, 1994

APPENDIX B: CLASSIFICATION SCHEME

Library and Information Science

Research strategy

APPENDIX C: DETAILS OF TOPIC DISTRIBUTIONS

Not available, please contact Author