research paper on library science

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals

Library science articles from across Nature Portfolio

Latest research and reviews.

The effect of prosocial behavior and its intensity on doctors’ performance in an online health community

• Yuguang Xie
• Shuping Zhao

What factors influence the intention to adopt blockchain technology in accounting education?

• Hamood Mohammed Al-Hattami

Digital financial inclusion in micro enterprises: understanding the determinants and impact on ease of doing business from World Bank survey

• Mohammad Asif
• Mohammad Wasiq

Big data visualisation in regional comprehensive economic partnership: a systematic review

What determines digital accounting systems’ continuance intention? An empirical investigation in SMEs

• Faozi A. Almaqtari

What makes deceptive online reviews? A linguistic analysis perspective

News and Comment

Pandemic publishing poses a new COVID-19 challenge

The scientific community’s response to COVID-19 has resulted in a large volume of research moving through the publication pipeline at extraordinary speed, with a median time from receipt to acceptance of 6 days for journal articles. Although the nature of this emergency warrants accelerated publishing, measures are required to safeguard the integrity of scientific evidence.

• Adam Palayew
• Ole Norgaard
• Jeffrey V. Lazarus

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Open Access is an initiative that aims to make scientific research freely available to all. To date our community has made over 100 million downloads. It’s based on principles of collaboration, unobstructed discovery, and, most importantly, scientific progression. As PhD students, we found it difficult to access the research we needed, so we decided to create a new Open Access publisher that levels the playing field for scientists across the world. How? By making research easy to access, and puts the academic needs of the researchers before the business interests of publishers.

We are a community of more than 103,000 authors and editors from 3,291 institutions spanning 160 countries, including Nobel Prize winners and some of the world’s most-cited researchers. Publishing on IntechOpen allows authors to earn citations and find new collaborators, meaning more people see your work not only from your own field of study, but from other related fields too.

Brief introduction to this section that descibes Open Access especially from an IntechOpen perspective

Want to get in touch? Contact our London head office or media team here

Our team is growing all the time, so we’re always on the lookout for smart people who want to help us reshape the world of scientific publishing.

Home > Books > Qualitative versus Quantitative Research

Research Methods in Library and Information Science

Submitted: 28 October 2016 Reviewed: 23 March 2017 Published: 28 June 2017

DOI: 10.5772/intechopen.68749

Cite this chapter

There are two ways to cite this chapter:

From the Edited Volume

Qualitative versus Quantitative Research

Edited by Sonyel Oflazoglu

To purchase hard copies of this book, please contact the representative in India: CBS Publishers & Distributors Pvt. Ltd. www.cbspd.com | [email protected]

Chapter metrics overview

5,992 Chapter Downloads

Impact of this chapter

Total Chapter Downloads on intechopen.com

Total Chapter Views on intechopen.com

Overall attention for this chapters

Library and information science (LIS) is a very broad discipline, which uses a wide rangeof constantly evolving research strategies and techniques. The aim of this chapter is to provide an updated view of research issues in library and information science. A stratified random sample of 440 articles published in five prominent journals was analyzed and classified to identify (i) research approach, (ii) research methodology, and (iii) method of data analysis. For each variable, a coding scheme was developed, and the articles were coded accordingly. A total of 78% of the articles reported empirical research. The rest 22% were classified as non‐empirical research papers. The five most popular topics were “information retrieval,” “information behaviour,” “information literacy,” “library services,” and “organization and management.” An overwhelming majority of the empirical research articles employed a quantitative approach. Although the survey emerged as the most frequently used research strategy, there is evidence that the number and variety of research methodologies have been increased. There is also evidence that qualitative approaches are gaining increasing importance and have a role to play in LIS, while mixed methods have not yet gained enough recognition in LIS research.

• library and information science
• research methods
• research strategies
• data analysis techniques
• research articles

Author Information

Aspasia togia *.

• Department of Library Science & Information Systems, Technological Educational Institute (TEI) of Thessaloniki, Greece

Afrodite Malliari

• DataScouting, Thessaloniki, Greece

*Address all correspondence to: [email protected]

1. Introduction

Library and information science (LIS), as its name indicates, is a merging of librarianship and information science that took place in the 1960s [ 1 , 2 ]. LIS is a field of both professional practice and scientific inquiry. As a field of practice, it includes the profession of librarianship as well as a number of other information professions, all of which assume the interplay of the following:

information content,

the people who interact with the content, and

the technology used to facilitate the creation, communication, storage, or transformation of the content [ 3 ].

The disciplinary foundation of LIS, which began in the 1920s, aimed at providing a theoretical foundation for the library profession. LIS has evolved in close relationship with other fields of research, especially computer science, communication studies, and cognitive sciences [ 4 ].

The connection of LIS with professional practice, on one hand, and other research fields on the other has influenced its research orientation and the development of methodological tools and theoretical perspectives [ 5 ]. Research problems are diverse, depending on the research direction, local trends, etc. Most of them relate to the professional practice although there are theoretical research statements as well. LIS research strives to address important information issues, such as these of “ information retrieval, information quality and authenticity, policy for access and preservation, the health and security applications of data mining ”(p. 3) [ 6 ]. The research is multidisciplinary in nature, and it has been heavily influenced by research designs developed in the social, behavioral, and management sciences and to a lesser extent by the theoretical inquiry adopted in the humanities [ 7 ]. Methods used in information retrieval research have been adapted from computer science. The emergence of evidence‐based librarianship in the late 1990s brought a positivist approach to LIS research, since it incorporated many of the research designs and methods used in clinical medicine [ 7 , 8 ]. In addition, LIS has developed its own methodological approaches, a prominent example of which is bibliometrics. Bibliometrics, which can be defined as “ the use of mathematical and statistical methods to study documents and patterns of publication ” (p. 38) [ 9 ], is a native research methodology, which has been extensively used outside the field, especially in science studies [ 10 ].

Library and information science research has been often criticized as being fragmentary, narrowly focused, and oriented to practical problems [ 11 ]. Many authors have noticed limited use of theory in published research and have advocated greater use of theory as a conceptual basis in LIS research [ 4 , 11 – 14 ]. Feehan et al. [ 13 ] claimed that LIS literature has not evolved enough to support a rigid body of its own theoretical basis. Jarvelin and Vakkari [ 15 ] argued that LIS theories are usually vague and conceptually unclear, and that research in LIS has been dominated by a paradigm which “ has made little use of such traditional scientific approaches as foundations and conceptual analysis, or of scientific explanation and theory formulation ” (p. 415). This lack of theoretical contributions may be associated with the fact that LIS emanated from professional practice and is therefore closely linked to practical problems such as the processing and organization of library materials, documentation, and information retrieval [ 15 , 16 ].

In this chapter, after briefly discussing the role of theory in LIS research, we provide an updated view of research issues in the field that will help scholars and students stay informed about topics related to research strategies and methods. To accomplish this, we describe and analyze patterns of LIS research activity as reflected in prominent library journals. The analysis of the articles highlights trends and recurring themes in LIS research regarding the use of multiple methods, the adoption of qualitative approaches, and the employment of advanced techniques for data analysis and interpretation [ 17 ].

2. The role of theory in LIS research

The presence of theory is an indication of research eminence and respectability [ 18 ], as well as a feature of discipline’s maturity [ 19 , 20 ]. Theory has been defined in many ways. “ Any of the following have been used as the meaning of theory: a law, a hypothesis, group of hypotheses, proposition, supposition, explanation, model, assumption, conjecture, construct, edifice, structure, opinion, speculation, belief, principle, rule, point of view, generalization, scheme, or idea ” (p. 309) [ 21 ]. A theory can be described as “ a set of interrelated concepts, definitions, and propositions that explains or predicts events or situations by specifying relations among variables ” [ 22 ]. According to Babbie [ 23 ], research is “ a systematic explanation for the observed facts and laws that related to a particular aspect of life ” (p. 49). It is “ a multiple‐level component of the research process, comprising a range of generalizations that move beyond a descriptive level to a more explanatory level ” [ 24 ] (p. 319). The role of theory in social sciences is, among other things, to explain and predict behavior, be usable in practical applications, and guide research [ 25 ]. According to Smiraglia [ 26 ], theory does not exist in a vacuum but in a system that explains the domains of human actions, the phenomena found in these domains, and the ways in which they are affected. He maintains that theory is developed by systematically observing phenomena, either in the positivist empirical research paradigm or in the qualitative hermeneutic paradigm. Theory is used to formulate hypotheses in quantitative research and confirms observations in qualitative research.

Glazier and Grover [ 24 ] proposed a model for theory‐building in LIS called “circuits of theory.” The model includes taxonomy of theory, developed earlier by the authors [ 11 ], and the critical social and psychological factors that influence research. The purpose of the taxonomy was to demonstrate the relationships among the concepts of research, theory, paradigms, and phenomena. Phenomena are described as “ events experienced in the empirical world ” (p. 230) [ 11 ]. Researchers assign symbols (digital or iconic representations, usually words or pictures) to phenomena, and meaning to symbols, and then they conceptualize the relationships among phenomena and formulate hypotheses and research questions. “ In the taxonomy, empirical research begins with the formation of research questions to be answered about the concepts or hypotheses for testing the concepts within a narrow set of predetermined parameters ” (p. 323) [ 24 ]. Various levels of theories, with implications for research in library and information Science, are described. The first theory level, called substantive theory , is defined as “ a set of propositions which furnish an explanation for an applied area of inquiry ” (p. 233) [ 11 ]. In fact, it may not be viewed as a theory but rather be considered as a research hypothesis that has been tested or even a research finding [ 16 ]. The next level of theory, called formal theory , is defined as “ a set of propositions which furnish an explanation for a formal or conceptual area of inquiry, that is, a discipline ” (p. 234) [ 11 ]. Substantive and formal theories together are usually considered as “middle range” theory in the social sciences. Their difference lies in the ability to structure generalizations and the potential for explanation and prediction. The final level, grand theory , is “ a set of theories or generalizations that transcend the borders of disciplines to explain relationships among phenomena ” (p. 321) [ 24 ]. According to the authors, most research generates substantive level theory, or, alternatively, researchers borrow theory from the appropriate discipline, apply it to the problem under investigation, and reconstruct the theory at the substantive level. Next in the hierarchy of theoretical categories is the paradigm , which is described as “ a framework of basic assumptions with which perceptions are evaluated and relationships are delineated and applied to a discipline or profession ” (p. 234) [ 11 ]. Finally, the most significant theoretical category is the world view , which is defined as “ an individual’s accepted knowledge, including values and assumptions, which provide a ‘filter’ for perception of all phenomena ” (p. 235) [ 11 ]. All the previous categories contribute to shaping the individual’s worldview. In the revised model, which places more emphasis on the impact of social environment on the research process, research and theory building is surrounded by a system of three basic contextual modules: the self, society, and knowledge, both discovered and undiscovered. The interactions and dialectical relationships of these three modules affect the research process and create a dynamic environment that fosters theory creation and development. The authors argue that their model will help researchers build theories that enable generalizations beyond the conclusions drawn from empirical data [ 24 ].

In an effort to propose a framework for a unified theory of librarianship, McGrath [ 27 ] reviewed research articles in the areas of publishing, acquisitions, classification and knowledge organization, storage, preservation and collection management, library collections, and circulations. In his study, he included articles that employed explanatory and predictive statistical methods to explore relationships between variables within and between the above subfields of LIS. For each paper reviewed, he identified the dependent variable, significant independent variables, and the units of analysis. The review displayed explanatory studies “ in nearly every level, with the possible exception of classification, while studies in circulation and use of the library were clearly dominant. A recapitulation showed that a variable at one level may be a unit of analysis at another, a property of explanatory research crucial to the development of theory, which has been either ignored or unrecognized in LIS literature ” (p. 368) [ 27 ]. The author concluded that “explanatory and predictive relationships do exist and that they can be useful in constructing a comprehensive unified theory of librarianship” (p. 368) [ 27 ].

Recent LIS literature provides several analyses of theory development and use in the field. In a longitudinal analysis of information needs and uses of literature, Julien and Duggan [ 28 ] investigated, among other things, to what extent LIS literature was grounded in theory. Articles “ based on a coherent and explicit framework of assumptions, definitions, and propositions that, taken together, have some explanatory power ” (p. 294) were classified as theoretical articles. Results showed that only 18.3% of the research studies identified in the sample of articles examined were theoretically grounded.

Pettigrew and McKechnie [ 29 ] analyzed 1160 journal articles published between 1993 and 1998 to determine the level of theory use in information science research. In the absence of a singular definition of theory that would cover all the different uses of the term in the sample of articles, they operationalized “theory” according to authors’ use of the term. They found that 34.1% of the articles incorporated theory, with the largest percentage of theories drawn from the social sciences. Information science itself was the second most important source of theories. The authors argued that this significant increase in theory use in comparison to earlier studies could be explained by the research‐oriented journals they selected for examination, the sample time, and the broad way in which they defined “theory.” With regard to this last point, that is, their approach of identifying theories only if the author(s) describe them as such in the article, Pettigrew and McKechnie [ 29 ] observed significant differences in how information science researchers perceive theory:

Although it is possible that conceptual differences regarding the nature of theory may be due to the different disciplinary backgrounds of researchers in IS, other themes emerged from our data that suggest a general confusion exists about theory even within subfields. Numerous examples came to light during our analysis in which an author would simultaneously refer to something as a theory and a method, or as a theory and a model, or as a theory and a reported finding. In other words, it seems as though authors, themselves, are sometimes unsure about what constitutes theory. Questions even arose regarding whether the author to whom a theory was credited would him or herself consider his or her work as theory (p. 68).

Kim and Jeong [ 16 ] examined the state and characteristics of theoretical research in LIS journals between 1984 and 2003. They focused on the “theory incident,” which is described as “an event in which the author contributes to the development or the use of theory in his/her paper.” Their study adopted Glazier and Grover’s [ 24 ] model of “circuits of theory.” Substantive level theory was operationalized to a tested hypothesis or an observed relationship, while both formal and grand level theories were identified when they were named as “theory,” “model,” or “law” by authors other than those who had developed them. Results demonstrated that the application of theory was present in 41.4% of the articles examined, signifying a significant increase in the proportion of theoretical articles as compared to previous studies. Moreover, it was evident that both theory development and theory use had increased by the year. Information seeking and use, and information retrieval, were identified as the subfields with the most significant contribution to the development of the theoretical framework.

In a more in‐depth analysis of theory use in Kumasi et al. [ 30 ] qualitatively analyzed the extent to which theory is meaningfully used in scholarly literature. For this purpose, they developed a theory talk coding scheme, which included six analytical categories, describing how theory is discussed in a study. The intensity of theory talk in the articles was described across a continuum from minimal (e.g., theory is discussed in literature review and not mentioned later) through moderate (e.g., multiple theories are introduced but without discussing their relevance to the study) to major (e.g., theory is employed throughout the study). Their findings seem to support the opinion that “ LIS discipline has been focused on the application of specific theoretical frameworks rather than the generation of new theories ” (p. 179) [ 30 ]. Another point the authors made was about the multiple terms used in the articles to describe theory. Words such as “framework,” “model,” or “theory” were used interchangeably by scholars.

It is evident from the above discussion that the treatment of theory in LIS research covers a spectrum of intensity, from marginal mentions to theory revising, expanding, or building. Recent analyses of the published scholarship indicate that the field has not been very successful in contributing to existing theory or producing new theory. In spite of this, one may still assert that LIS research employs theory, and, in fact, there are many theories that have been used or generated by LIS scholars. However, “ calls for additional and novel theory development work in LIS continue, particularly for theories that might help to address the research practice gap ” (p. 12) [ 31 ].

3. Research strategies in LIS

3.1. surveys of research methods.

LIS is a very broad discipline, which uses a wide range of constantly evolving research strategies and techniques [ 32 ]. Various classification schemes have been developed to analyze methods employed in LIS research (e.g., [ 13 , 15 , 17 , 33 – 35 , 38 ]). Back in 1996, in the “research record” column of the Journal of Education for Library and Information Science, Kim [ 36 ] synthesized previous categories and definitions and introduced a list of research strategies, including data collection and analysis methods. The listing included four general research strategies: (i) theoretical/philosophical inquiry (development of conceptual models or frameworks), (ii) bibliographic research (descriptive studies of books and their properties as well as bibliographies of various kinds), (iii) R&D (development of storage and retrieval systems, software, interface, etc.), and (iv) action research, it aims at solving problems and bringing about change in organizations. Strategies are then divided into quantitative and qualitative driven. In the first category are included descriptive studies, predictive/explanatory studies, bibliometric studies, content analysis, and operation research studies. Qualitative‐driven strategies are considered the following: case study, biographical method, historical method, grounded theory, ethnography, phenomenology, symbolic interactionism/semiotics, sociolinguistics/discourse analysis/ethnographic semantics/ethnography of communication, and hermeneutics/interpretive interactionism (p. 378–380) [ 36 ].

Systematic studies of research methods in LIS started in the 1980s and several reviews of the literature have been conducted over the past years to analyze the topics, methodologies, and quality of research. One of the earliest studies was done by Peritz [ 37 ] who carried out a bibliometric analysis of the articles published in 39 core LIS journals between 1950 and 1975. She examined the methodologies used, the type of library or organization investigated, the type of activity investigated, and the institutional affiliation of the authors. The most important findings were a clear orientation toward library and information service activities, a widespread use of the survey methodology, a considerable increase of research articles after 1960, and a significant increase in theoretical studies after 1965.

Nour [ 38 ] followed up on Peritz’s [ 37 ] work and studied research articles published in 41 selected journals during the year 1980. She found that survey and theoretical/analytic methodologies were the most popular, followed by bibliometrics. Comparing these findings to those made by Peritz [ 37 ], Nour [ 38 ] found that the amount of research continued to increase, but the proportion of research articles to all articles had been decreasing since 1975.

Feehan et al. [ 13 ] described how LIS research published during 1984 was distributed over various topics and what methods had been used to study these topics. Their analysis revealed a predominance of survey and historical methods and a notable percentage of articles using more than one research method. Following a different approach, Enger et al. (1989) focused on the statistical methods used by LIS researchers in articles published during 1985 [ 39 ]. They found that only one out of three of the articles reported any use of statistics. Of those, 21% used descriptive statistics and 11% inferential statistics. In addition, the authors found that researchers from disciplines other than LIS made the highest use of statistics and LIS faculty showed the highest use of inferential statistics.

An influential work, against which later studies have been compared, is that of Jarvelin and Vakkari [ 15 ] who studied LIS articles published in 1985 in order to determine how research was distributed over various subjects, what approaches had been taken by the authors, and what research strategies had been used. The authors replicated their study later to include older research published between 1965 and 1985 [ 40 ]. The main finding of these studies was that the trends and characteristics of LIS research remained more or less the same over the aforementioned period of 20 years. The most common topics were information service activities and information storage and retrieval. Empirical research strategies were predominant, and of them, the most frequent was the survey. Kumpulainen [ 41 ], in an effort to provide a continuum with Jarvelin and Vakkeri’s [ 15 ] study, analyzed 632 articles sampled from 30 core LIS journals with respect to various characteristics, including topics, aspect of activity, research method, data selection method, and data analysis techniques. She used the same classification scheme, and she selected the journals based on a slightly modified version of Jarvelin and Vakkari’s [ 15 ] list. Library services and information storage and retrieval emerged again as the most common subjects approached by the authors and survey was the most frequently used method.

More recent studies of this nature include those conducted by Koufogiannakis et al. [ 42 ], Hildreth and Aytac [ 43 ], Hider and Pymm [ 32 ], and Chu [ 17 ]. Koufogiannakis et al. [ 42 ] examined research articles published in 2001 and they found that the majority of them were questionnaire‐based descriptive studies. Comparative, bibliometrics, content analysis, and program evaluation studies were also popular. Information storage and retrieval emerged as the predominant subject area, followed by library collections and management. Hildreth and Aytac [ 43 ] presented a review of the 2003–2005 published library research with special focus on methodology issues and the quality of published articles of both practitioners and academic scholars. They found that most research was descriptive and the most frequent method for data collection was the questionnaire, followed by content analysis and interviews. With regard to data analysis, more researchers used quantitative methods, considerably less used qualitative‐only methods, whereas 61 out of 206 studies included some kind of qualitative analysis, raising the total percentage of qualitative methods to nearly 50%. With regard to the quality of published research, the authors argued that “ the majority of the reports are detailed, comprehensive, and well‐organized ” (p. 254) [ 43 ]. Still, they noticed that the majority of reports did not mention the critical issues of research validity and reliability and neither did they indicate study limitations or future research recommendations. Hider and Pymm [ 32 ] described content analysis of LIS literature “ which aimed to identify the most common strategies and techniques employed by LIS researchers carrying out high‐profile empirical research ” (p. 109). Their results suggested that while researchers employed a wide variety of strategies, they mostly used surveys and experiments. They also observed that although quantitative research accounted for more than 50% of the articles, there was an increase in the use of most sophisticated qualitative methods. Chu [ 17 ] analyzed the research articles published between 2001 and 2010 in three major journals and reported the following most frequent research methods: theoretical approach (e.g., conceptual analysis), content analysis, questionnaire, interview, experiment, and bibliometrics. Her study showed an increase in both the number and variety of research methods but lack of growth in the use of qualitative research or in the adoption of multiple research methods.

In summary, the literature shows a continued interest in the analysis of published LIS research. Approaches include focusing on particular publication years, geographic areas, journal titles, aspects of LIS, and specific characteristics, such as subjects, authorship, and research methods. Despite the abundance of content analyses of LIS literature, the findings are not easily comparable due to differences in the number and titles of journals examined, in the types of the papers selected for analysis, in the periods covered, and in classification schemes developed by the authors to categorize article topics and research strategies. Despite the differences, some findings are consistent among all studies:

Information seeking, information retrieval, and library and information service activities are among the most common subjects studied,

Descriptive research methodologies based on surveys and questionnaires predominate,

Over the years, there has been a considerable increase in the array of research approaches used to explore library issues, and

Data analysis is usually limited to descriptive statistics, including frequencies, means, and standard deviations.

3.2. Data collection and analysis

Articles published between 2011 and 2016 were obtained from the following journals: Library and Information Science Research, College & Research Libraries, Journal of Documentation, Information Processing & Management, and Journal of Academic Librarianship ( Table 1 ). These five titles were selected as data sources because they have the highest 5‐year impact factor of the journals classified in Ulrich’s Serials Directory under the “Library and Information Sciences” subject heading. From the journals selected, only full‐length articles were collected. Editorials, book reviews, letters, interviews, commentaries, and news items were excluded from the analysis. This selection process yielded 1643 articles. A stratified random sample of 440 articles was chosen for in‐depth analysis ( Table 2 ). For the purpose of this study, five strata, corresponding to the five journals, were used. The sample size was determined using a margin of error, 4%, and confidence interval, 95%.

Table 1.

Profile of the journals.

Table 2.

Journal titles.

Each article was classified as either research or theoretical. Articles that employed specific research methodology and presented specific findings of original studies performed by the author(s) were considered research articles. The kind of study may vary (e.g., it could be an experiment, a survey, etc.), but in all cases, raw data had been collected and analyzed, and conclusions were drawn from the results of that analysis. Articles reporting research in system design or evaluation in the information systems field were also regarded as research articles . On the other hand, works that reviewed theories, theoretical concepts, or principles discussed topics of interest to researchers and professionals, or described research methodologies were regarded as theoretical articles [ 44 ] and were classified in the no‐empirical‐research category. In this category, were also included literature reviews and articles describing a project, a situation, a process, etc.

Each article was classified into a topical category according to its main subject. The articles classified as research were then further explored and analyzed to identify (i) research approach, (ii) research methodology, and (iii) method of data analysis. For each variable, a coding scheme was developed, and the articles were coded accordingly. The final list of the analysis codes was extracted inductively from the data itself, using as reference the taxonomies utilized in previous studies [ 15 , 32 , 43 , 45 ]. Research approaches “ are plans and procedures for research ” (p. 3) [ 46 ]. Research approaches can generally be grouped as qualitative, quantitative, and mixed methods studies. Quantitative studies aim at the systematic empirical investigation of quantitative properties or phenomena and their relationships. Qualitative research can be broadly defined as “ any kind of research that produces findings not arrived at by means of statistical procedures or other means of quantification ” (p. 17) [ 47 ]. It is a way to gain insights through discovering meanings and explaining phenomena based on the attributes of the data. In mixed model research, quantitative and qualitative approaches are combined within or across the stages of the research process. It was beyond the scope of this study to identify in which stages of a study—data collection, data analysis, and data interpretation—the mixing was applied or to reveal the types of mixing. Therefore, studies using both quantitative and qualitative methods, irrespective of whether they describe if and how the methods were integrated, were coded as mixed methods studies.

Research methodologies , or strategies of inquiry, are types of research models “ that provide specific direction for procedures in a research design ” (p. 11) [ 46 ] and inform the decisions concerning data collection and analysis. A coding schema of research methodologies was developed by the authors based on the analysis of all research articles included in the sample. The methodology classification included 12 categories ( Table 3 ). Each article was classified into one category for the variable research methodology . If more than one research strategy was mentioned (e.g., experiment and survey), the article was classified according to the main strategy.

Table 3.

Coding schema for research methodologies.

Methods of data analysis refer to the techniques used by the researchers to explore the original data and answer their research problems or questions. Data analysis for quantitative researches involves statistical analysis and interpretation of figures and numbers. In qualitative studies, on the other hand, data analysis involves identifying common patterns within the data and making interpretations of the meanings of the data. The array of data analysis methods included the following categories:

Descriptive statistics,

Inferential statistics,

Qualitative data analysis,

Experimental evaluation, and

Other methods,

Descriptive statistics are used to describe the basic features of the data in a study. Inferential statistics investigate questions, models, and hypotheses. Mathematical analysis refers to mathematic functions, etc. used mainly in bibliometric studies to answer research questions associated with citation data. Qualitative data analysis is the range of processes and procedures used for the exploration of qualitative data, from coding and descriptive analysis to identification of patterns and themes and the testing of emergent findings and hypotheses. It was used in this study as an overarching term encompassing various types of analysis, such as thematic analysis, discourse analysis, or grounded theory analysis. The class experimental evaluation was used for system and software analysis and design studies which assesses the newly developed algorithm, tool, method, etc. by performing experiments on selected datasets. In these cases, “experiments” differ from the experimental designs in social sciences. Methods that did not fall into one of these categories (e.g., mathematical analysis, visualization, or benchmarking) were classified as other methods . If both descriptive and inferential statistics were used in an article, only the inferential were recorded. In mixed methods studies, each method was recorded in the order in which it was reported in the article.

Ten percent of the articles were randomly selected and used to establish inter‐rater reliability and provide basic validation of the coding schema. Cohen’s kappa was calculated for each coded variable. The average Cohen’s kappa value was κ = 0.60, p < 0.000 (the highest was 0.63 and lowest was 0.59). This indicates a substantial agreement [ 48 ]. The coding disparities across raters were discussed, and the final codes were determined via consensus.

3.3. Results

3.3.1. topic.

Table 4 presents the distribution of articles over the various topics, for each of which a detailed description is provided. The five most popular topics of the papers in the total sample of 440 articles were “information retrieval,” “information behavior,” “information literacy,” “library services,” and “organization and management.” These areas cover over 60% of all topics studied in the papers. The least‐studied topics (covered in less than eight papers) fall into the categories of “information and knowledge management,” “library information systems,” “LIS theory,” and “infometrics.”

Table 4.

Article topics.

Figure 1 shows how the top five topics are distributed across journals. As expected, the topic “information retrieval” has higher publication frequencies in Information Processing & Management, a journal focusing on system design and issues related to the tools and techniques used in storage and retrieval of information. “Information literacy,” “information behavior,” “library services,” and “organization and management” appear to be distributed almost proportionately in College & Research Libraries. “Information literacy” seems to be a more preferred topic in the Journal of Academic Librarianship, while “information behavior” is more popular in the Journal of Documentation and Library & Information Science Research.

Figure 1.

Distribution of topics across journals.

3.3.2. Research approach and methodology

Of all articles examined, 343 articles, which represent the 78% of the sample, reported empirical research. The rest 22% (N = 97) were classified as non‐empirical research papers. Research articles were coded as quantitative, qualitative, or mixed methods studies. An overwhelming majority (70%) of the empirical research articles employed a quantitative research approach. Qualitative and mixed methods research was reported in 21.6 and 8.5% of the articles, respectively ( Figure 2 ).

Figure 2.

Research approach.

Table 5 presents the distribution of research approaches over the five most famous topics. The quantitative approach clearly prevails in all topics, especially in information retrieval research. However, qualitative designs seem to gain acceptance in all topics (except information retrieval), while in information behavior research, quantitative and qualitative approaches are almost evenly distributed. Mixed methods were quite frequent in information literacy and information behavior studies and less popular in the other topics.

Table 5.

Topics across research approach.

The most frequently used research strategy was survey, accounting for almost 37% of all research articles, followed by system and software analysis and design, a strategy used in this study specifically for research in information systems (Jarvelin & Vakkari, 1990). This result is influenced by the fact that Information Processing & Management addresses issues at the intersection between LIS and computer science, and the majority of its articles present the development of new tools, algorithms, methods and systems, and their experimental evaluation. The third‐ and fourth‐ranking strategies were content analysis and bibliometrics. Case study, experiment, and secondary data analysis were represented by 15 articles each, while the rest of the techniques were underrepresented with considerably fewer articles ( Table 6 ).

Table 6.

Research methodologies.

3.3.3. Methods of data analysis

Table 7 displays the frequencies for each type of data analysis.

Table 7.

Method of data analysis.

Almost half of the empirical research papers examined reported any use of statistics. Descriptive statistics, such as frequencies, means, or standard deviations, were more frequently used compared to inferential statistics, such as ANOVA, regression, or factor analysis. Nearly one‐third of the articles employed some type of qualitative data analysis either as the only method or—in mixed methods studies—in combination with quantitative techniques.

3.4. Discussions and conclusions

The patterns of LIS research activity as reflected in the articles published between 2011 and 2016 in five well‐established, peer‐reviewed journals were described and analyzed. LIS literature addresses many and diverse topics. Information retrieval, information behavior, and library services continue to attract the interest of researchers as they are core areas in library science. Information retrieval has been rated as one of the most famous areas of interest in research articles published between 1965 and 1985 [ 40 ]. According to Dimitroff [ 49 ], information retrieval was the second most popular topic in the articles published in the Bulletin of the Medical Library Association, while Cano [ 50 ] argued that LIS research produced in Spain from 1977 to 1994 was mostly centered on information retrieval and library and information services. In addition, Koufogiannakis et al. [ 42 ] found that information access and retrieval were the domain with the most research, and in Hildreth and Aytac’s [ 43 ] study, most articles were dealing with issues related to users (needs, behavior, information seeking, etc.), services, and collections. The present study provides evidence that the amount of research in information literacy is increasing, presumably due to the growing importance of information literacy instruction in libraries. In recent years, there is an ongoing educational role for librarians, who are more and more actively engaging in the teaching and learning processes, a trend that is reflected in the research output.

With regard to research methodologies, the present study seems to confirm the well‐documented predominance of survey in LIS research. According to Dimitroff [ 49 ], the percentage related to use of survey research methods reported in various studies varied between 20.3 and 41.5%. Powell [ 51 ], in a review of the research methods appearing in LIS literature, pointed out that survey had consistently been the most common type of study in both dissertations and journal articles. Survey reported the most widely used research design by Jarvelin and Vakkari [ 40 ], Crawford [ 52 ], Hildreth and Aytac [ 43 ], and Hider and Pymm [ 32 ]. The majority of articles examined by Koufogiannakis et al. [ 42 ] were descriptive studies using questionnaires/surveys. In addition, survey methods represented the largest proportion of methods used in information behavior articles analyzed by Julien et al. [ 53 ]. There is no doubt that survey has been used more than any other method in LIS research. As Jarvelin and Vakkari [ 15 ] put it, “it appears that the field is so survey‐oriented that almost all problems are seen through a survey viewpoint” (p. 416). Much of survey’s popularity can be ascribed to its being a well‐known, understood, easily conducted, and inexpensive method, which is easy to analyze results [ 41 , 42 ]. However, our findings suggest that while the survey ranks high, a variety of other methods have been also used in the research articles. Content analysis emerged as the third‐most frequent strategy, a finding similar to those of previous studies [ 17 , 32 ]. Although content analysis was not regarded by LIS researchers as a favored research method until recently, its popularity seems to be growing [ 17 ].

Quantitative approaches, which dominate, tend to rely on frequency counts, percentages, and descriptive statistics used to describe the basic features of the data in a study. Fewer studies used advanced statistical analysis techniques, such as t‐tests, correlation, and regressions, while there were some examples of more sophisticated methods, such as factor analysis, ANOVA, MANOVA, and structural equation modeling. Researchers engaging in quantitative research designs should take into consideration the use of inferential statistics, which enables the generalization from the sample being studied to the population of interest and, if used appropriately, are very useful for hypothesis testing. In addition, multivariate statistics are suitable for examining the relationships among variables, revealing patterns and understanding complex phenomena.

The findings also suggest that qualitative approaches are gaining increasing importance and have a role to play in LIS studies. These results are comparable to the findings of Hider and Pymm [ 32 ], who observed significant increases for qualitative research strategies in contemporary LIS literature. Qualitative analysis description varied widely, reflecting the diverse perspectives, analysis methods, and levels of depth of analysis. Commonly used terms in the articles included coding, content analysis, thematic analysis, thematic analytical approach, theme, or pattern identification. One could argue that the efforts made to encourage and promote qualitative methods in LIS research [ 54 , 55 ] have made some impact. However, qualitative research methods do not seem to be adequately utilized by library researchers and practitioners, despite their potential to offer far more illuminating ways to study library‐related issues [ 56 ]. LIS research has much to gain from the interpretive paradigm underpinning qualitative methods. This paradigm assumes that social reality is

the product of processes by which social actors together negotiate the meanings for actions and situations; it is a complex of socially constructed meanings. Human experience involves a process of interpretation rather than sensory, material apprehension of the external physical world and human behavior depends on how individuals interpret the conditions in which they find themselves. Social reality is not some ‘thing’ that may be interpreted in different ways, it is those interpretations (p. 96) [ 57 ].

As stated in the introduction of this chapter, library and information science focuses on the interaction between individuals and information. In every area of LIS research, the connection of factors that lead to and influence this interaction is increasingly complex. Qualitative research searches for “ all aspects of that complexity on the grounds that they are essential to understanding the behavior of which they are a part ” (p. 241) [ 59 ]. Qualitative research designs can offer a more in‐depth analysis of library users, their needs, attitudes, and behaviors.

The use of mixed methods designs was found to be rather rare. While Hildreth and Aytac [ 43 ] found higher percentages of studies using combined methods in data analysis, our results are analogous to those shown by Fidel [ 60 ]. In fact, as in her study, only few of the articles analyzed referred to mixed methods research by name, a finding indicating that “ the concept has not yet gained recognition in LIS research ” (p. 268). Mixed methods research has become an established research approach in the social sciences as it minimizes the weaknesses of quantitative and qualitative research alone and allows researchers to investigate the phenomena more completely [ 58 ].

In conclusion, there is evidence that LIS researchers employ a large number and wide variety of research methodologies. Each research approach, strategy, and method has its advantages and limitations. If the aim of the study is to confirm hypotheses about phenomena or measure and analyze the causal relationships between variables, then quantitative methods might be used. If the research seeks to explore, understand, and explain phenomena then qualitative methods might be used. Researchers can consider the full range of possibilities and make their selection based on the philosophical assumptions they bring to the study, the research problem being addressed, their personal experiences, and the intended audience for the study [ 46 ].

Taking into consideration the increasing use of qualitative methods in LIS studies, an in‐depth analysis of papers using qualitative methods would be interesting. A future study in which the different research strategies and types of analysis used in qualitative methods will be presented and analyzed could help LIS practitioners understand the benefits of qualitative analysis.

Mixed methods used in LIS research papers could be analyzed in future studies in order to identify in which stages of a study, data collection, data analysis, and data interpretation, the mixing was applied and to reveal the types of mixing.

As far as it concerns the quantitative research methods, which predominate in LIS research, it would be interesting to identify systematic relations between more than two variables such as authors’ affiliation, topic, research strategies, etc. and to create homogeneous groups using multivariate data analysis techniques.

• 1. Buckland MK, Liu ZM. History of information science. Annual Review of Information Science and Technology. 1995; 30 :385-416
• 2. Rayward WB. The history and historiography of information science: Some reflections. Information Processing & Management. 1996; 32 (1):3-17
• 3. Wildemuth BM. Applications of Social Research Methods to Questions in Information and Library Science. Westport, CT: Libraries Unlimited; 2009
• 4. Hjørland B. Theory and metatheory of information science: A new interpretation. Journal of Documentation. 1998; 54 (5):606-621. DOI: http://doi.org/10.1108/EUM0000000007183
• 5. Åström F. Heterogeneity and homogeneity in library and information science research. Information Research [Internet]. 2007 [cited 23 April 2017]; 12 (4): poster colisp01 [3 p.]. Available from: http://www.informationr.net/ir/12-4/colis/colisp01.html
• 6. Dillon A. Keynote address: Library and information science as a research domain: Problems and prospects. Information Research [Internet]. 2007 [cited 23 April 2017]; 12 (4): paper colis03 [6 p.]. Available from: http://www.informationr.net/ir/12-4/colis/colis03.html
• 7. Eldredge JD. Evidence‐based librarianship: An overview. Bulletin of the Medical Library Association. 2000; 88 (4):289-302
• 8. Bradley J, Marshall JG. Using scientific evidence to improve information practice. Health Libraries Review. 1995; 12 (3):147-157
• 9. Bibliometrics. In: International Encyclopedia of Information and Library Science. 2nd ed. London, UK: Routledge; 2003. p. 38
• 10. Åström F. Library and Information Science in context: The development of scientific fields, and their relations to professional contexts. In: Rayward WB, editor. Aware and Responsible: Papers of the Nordic‐International Colloquium on Social and Cultural Awareness and Responsibility in Library, Information and Documentation Studies (SCARLID). Oxford, UK: Scarecrow Press; 2004. pp. 1-27
• 11. Grover R, Glazier J. A conceptual framework for theory building in library and information science. Library and Information Science Research. 1986; 8 (3):227-242
• 12. Boyce BR, Kraft DH. Principles and theories in information science. In: W ME, editor. Annual Review of Information Science and Technology. Medford, NJ: Knowledge Industry Publications. 1985; pp. 153-178
• 13. Feehan PE, Gragg WL, Havener WM, Kester DD. Library and information science research: An analysis of the 1984 journal literature. Library and Information Science Research. 1987; 9 (3):173-185
• 14. Spink A. Information science: A third feedback framework. Journal of the American Society for Information Science. 1997; 48 (8):728-740
• 15. Jarvelin K, Vakkari P. Content analysis of research articles in Library and Information Science. Library and Information Science Research. 1990; 12 (4):395-421
• 16. Kim SJ, Jeong DY. An analysis of the development and use of theory in library and information science research articles. Library and Information Science Research. 2006; 28 (4):548-562. DOI: http://doi.org/10.1016/j.lisr.2006.03.018
• 17. Chu H. Research methods in library and information science: A content analysis. Library & Information Science Research. 2015; 37 (1):36-41. DOI: http://doi.org/10.1016/j.lisr.2014.09.003
• 18. Van Maanen J. Different strokes: Qualitative research in the administrative science quarterly from 1956 to 1996. In: Van Maanen J, editor. Qualitative Studies of Organizations. Thousand Oaks, CA: SAGE; 1998. pp. ix‐xxxii
• 19. Brookes BC. The foundations of information science Part I. Philosophical aspects. Journal of Information Science. 1980; 2 (3/4):125-133
• 20. Hauser L. A conceptual analysis of information science. Library and Information Science Research. 1988; 10 (1):3-35
• 21. McGrath WE. Current theory in Library and Information Science. Introduction. Library Trends. 2002; 50 (3):309-316
• 22. Theory and why it is important - Social and behavioral theories - e-Source Book - OBSSR e-Source [Internet]. Esourceresearch.org. 2017 [cited 23 April 2017]. Available from: http://www.esourceresearch.org/eSourceBook/SocialandBehavioralTheories/3TheoryandWhyItisImportant/tabid/727/Default.aspx
• 23. Babbie E. The practice of social research. 7th ed. Belmont, CA: Wadsworth; 1995
• 24. Glazier JD, Grover R. A multidisciplinary framework for theory building. Library Trends. 2002; 50 (3):317-329
• 25. Glaser B, Strauss AL. The discovery of grounded theory: Strategies for qualitative research. New Brunswick: Aldine Transaction; 1999
• 26. Smiraglia RP. The progress of theory in knowledge organization. Library Trends. 2002; 50 :330-349
• 27. McGrath WE. Explanation and prediction: Building a unified theory of librarianship, concept and review. Library Trends. 2002; 50 (3):350-370
• 28. Julien H, Duggan LJ. A longitudinal analysis of the information needs and uses literature. Library & Information Science Research. 2000; 22 (3):291-309. DOI: http://doi.org/10.1016/S0740‐8188(99)00057‐2
• 29. Pettigrew KE, McKechnie LEF. The use of theory in information science research. Journal of the American Society for Information Science and Technology. 2001; 52 (1):62-73. DOI: http://doi.org/10.1002/1532‐2890(2000)52:1<62::AID‐ASI1061>3.0.CO;2‐J
• 30. Kumasi KD, Charbonneau DH, Walster D. Theory talk in the library science scholarly literature: An exploratory analysis. Library & Information Science Research. 2013; 35 (3):175-180. DOI: http://doi.org/10.1016/j.lisr.2013.02.004
• 31. Rawson C, Hughes‐Hassell S. Research by Design: The promise of design‐based research for school library research. School Libraries Worldwide. 2015; 21 (2):11-25
• 32. Hider P, Pymm B. Empirical research methods reported in high‐profile LIS journal literature. Library & Information Science Research. 2008; 30 (2):108-114. DOI: http://doi.org/10.1016/j.lisr.2007.11.007
• 33. Bernhard, P. In search of research methods used in information science. Canadian Journal of Information and Library Science. 1993;18(3): 1-35
• 34. Blake VLP. Since Shaughnessy. Collection Management. 1994; 19 (1‐2):1-42. DOI: http://doi.org/10.1300/J105v19n01_01
• 35. Schlachter GA. Abstracts of library science dissertations. Library Science Annual. 1989; 1 :1988-1996
• 36. Kim MT. Research record. Journal of Education for Library and Information Science. 1996; 37 (4):376-383
• 37. Peritz BC. The methods of library science research: Some results from a bibliometric survey. Library Research. 1980; 2 (3):251-268
• 38. Nour MM. A quantitative analysis of the research articles published in core library journals of 1980. Library and Information Science Research. 1985; 7 (3):261-273
• 39. Enger KB, Quirk G, Stewart JA. Statistical methods used by authors of library and infor- mation science journal articles. Library and Information Science Research. 1989; 11 (1): 37-46
• 40. Jarvelin K, Vakkari P. The evolution of library and information science 1965-1985: A content analysis of journal articles. Information Processing and Management. 1993; 29 (1):129-144
• 41. Kumpulainen S. Library and information science research in 1975: Content analysis of the journal articles. Libri. 1991; 41 (1):59-76
• 42. Koufogiannakis D, Slater L, Crumley E. A content analysis of librarianship research. Journal of Information Science. 2004; 30 (3):227-239. DOI: http://doi.org/10.1177/0165551504044668
• 43. Hildreth CR, Aytac S. Recent library practitioner research: A methodological analysis and critique on JSTOR. Journal of Education for Library and Information Science. 2007; 48 (3):236-258
• 44. Gonzales‐Teruel A, Abad‐Garcia MF. Information needs and uses: An analysis of the literature published in Spain, 1990‐2004. Library and Information Science Research. 2007; 29 (1):30-46
• 45. Luo L, Mckinney M. JAL in the past decade: A comprehensive analysis of academic library research. The Journal of Academic Librarianship. 2015; 41 :123-129. DOI: http://doi.org/10.1016/j.acalib.2015.01.003
• 46. Creswell JW. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 3rd ed. Thousand Oaks, CA: SAGE; 2009
• 47. Strauss A, Corbin J. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Newbury Park, CA: SAGE Publications; 1990
• 48. Neuendorf KA. The Content Analysis Guidebook. 2nd ed. Thousand Oaks, CA: SAGE Publications; 2016
• 49. Dimitroff A. Research for special libraries: A quantitative analysis of the literature. Special Libraries. 1995; 86 (4):256-264
• 50. Cano V. Bibliometric overview of library and information science research in Spain. Journal of the American Society for Information Science. 1999; 50 (8):675-680. DOI: http://doi.org/10.1002/(SICI)1097‐4571(1999)50:8<675::AID‐ASI5>3.0.CO;2‐B
• 51. Powell RR. Recent trends in research: A methodological essay. Library & Information Science Research. 1999; 21 (1):91-119. DOI: http://doi.org/10.1016/S0740‐8188(99)80007‐3
• 52. Crawford GA. The research literature of academic librarianship: A comparison of college & Research Libraries and Journal of Academic Librarianship. College & Research Libraries. 1999; 60 (3):224-230. DOI: http://doi.org/10.5860/crl.60.3.224
• 53. Julien H, Pecoskie JJL, Reed K. Trends in information behavior research, 1999-2008: A content analysis. Library & Information Science Research. 2011; 33 (1):19-24. DOI: http://doi.org/10.1016/j.lisr.2010.07.014
• 54. Fidel R. Qualitative methods in information retrieval research. Library and Information Science Research. 1993; 15 (3):219-247
• 55. Hernon P, Schwartz C. Reflections (editorial). Library and Information Science Research. 2003; 25 (1):1-2. DOI: http://doi.org/10.1016/S0740‐8188(02)00162‐7
• 56. Priestner A. Going native: Embracing ethnographic research methods in libraries. Revy. 2015; 38 (4):16-17
• 57. Blaikie N. Approaches to social enquiry. Cambridge: Polity; 1993
• 58. Johnson RB, Onwuegbuzie AJ. Mixed methods research: A research paradigm whose time has come. Educational Researcher. 2004; 33 (7):14-26
• 59. Westbrook L. Qualitative research methods: A review of major stages, data analysis techniques, and quality controls. Library & Information Science Research. 1994; 16 (3):241-254
• 60. Fidel R. Are we there yet?: Mixed methods research in library and information science. Library and Information Science Research. 2008; 30 (4):265-272. DOI: http://doi.org/10.1016/j.lisr.2008.04.001

© 2017 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution 3.0 License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Continue reading from the same book

Edited by Sonyel Oflazoglu Dora

Published: 28 June 2017

By Seyma Demir and Yasemin Yildirim Usta

1960 downloads

By Maria Cecília de Souza Minayo

2165 downloads

By Yusuf Bilgin

3259 downloads

IntechOpen Author/Editor? To get your discount, log in .

Discounts available on purchase of multiple copies. View rates

Local taxes (VAT) are calculated in later steps, if applicable.

Support: [email protected]

To read this content please select one of the options below:

Please note you do not have access to teaching notes, analysis on the research progress of library and information science since the new century.

Library Hi Tech

ISSN : 0737-8831

Article publication date: 15 December 2020

Issue publication date: 25 August 2023

Library science and information science, two subdisciplines of library and information science (LIS), are developed independently but interconnectedly. In this information age, LIS is in a special period of transformation and development, which has caused some changes in both library science and information science. By accurately capturing these changes and analyzing them, the authors can effectively map the development of LIS in the new century, thus providing a reference for the evolution and development of the field. The purposes of this paper are to explore the mainstream research fields and frontiers of library science and information science, respectively, since the new century, and to make a comparative analysis of the two subdisciplines.

Design/methodology/approach

By using CiteSpace to visualize LIS journals, this study draws knowledge maps of the two subdisciplines of LIS through the co-occurrence descriptors network. Using burst detection algorithm, this study detects words of high frequency variation by investigating the time frequency distribution.

The results show that the research focus of library science has experienced a change from traditional to digital library while information science has moved from information to data focus. This study also finds the similarities and differences between mainstream areas of library science and information science.

Originality/value

This study focuses on the evolution of library science and information science, and explores their mainstream research fields and frontiers in the 21st century. These findings will promote the transformation and development of LIS as well as provide research directions for scholars in the field.

• Library science
• Information science
• Research frontier
• Development trend

Acknowledgements

This study was funded in part by major project of National Social Science Foundation of China (19ZDA348), the Natural Science Foundation of Zhejiang Province (LY20G030011) and supported by the Fundamental Research Funds for the Provincial Universities of Zhejiang (GK209907299001-201) and National Social Science Foundation of China (61702009).

Song, Y. , Wei, K. , Yang, S. , Shu, F. and Qiu, J. (2023), "Analysis on the research progress of library and information science since the new century", Library Hi Tech , Vol. 41 No. 4, pp. 1145-1157. https://doi.org/10.1108/LHT-06-2020-0126

Emerald Publishing Limited

Copyright © 2020, Emerald Publishing Limited

Related articles

All feedback is valuable.

Please share your general feedback

Report an issue or find answers to frequently asked questions

Contact Customer Support

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Springer Nature - PMC COVID-19 Collection

Forecasting the future of library and information science and its sub-fields

Zehra taşkın.

Scholarly Communication Research Group, Adam Mickiewicz University in Poznań, Poznań, Poland

Forecasting is one of the methods applied in many studies in the library and information science (LIS) field for numerous purposes, from making predictions of the next Nobel laureates to potential technological developments. This study sought to draw a picture for the future of the LIS field and its sub-fields by analysing 97 years of publication and citation patterns. The core Web of Science indexes were used as the data source, and 123,742 articles were examined in-depth for time series analysis. The social network analysis method was used for sub-field classification. The field was divided into four sub-fields: (1) librarianship and law librarianship, (2) health information in LIS, (3) scientometrics and information retrieval and (4) management and information systems. The results of the study show that the LIS sub-fields are completely different from each other in terms of their publication and citation patterns, and all the sub-fields have different dynamics. Furthermore, the number of publications, references and citations will increase significantly in the future. It is expected that more scholars will work together. The future subjects of the LIS field show astonishing diversity from fake news to predatory journals, open government, e-learning and electronic health records. However, the findings prove that publish or perish culture will shape the field. Therefore, it is important to go beyond numbers. It can only be achieved by understanding publication and citation patterns of the field and developing research policies accordingly.

Introduction

Price ( 1963 p. 19, 1974 , p. 166), predicted more than half a century ago that if the exponential growth of big science continued, we could have two scientists for each person and dog in the population in the future, and we could have one million academic journals by the 2000s. Today, an average of 2.3% of worldwide gross national product is devoted to research and development activities (World Bank 2018 ), and 8.5 out of every 1000 workers is employed as a researcher (Organisation for Economic Co-operation and Development 2020 ). The current total number of active journals published worldwide is 380,299 ( ULRICHSWEB Global Serials Directory 2020 ), and at least the 73,299,923 articles have been published since Price published Little Science, Big Science in 1963. 1 One of Price’s biggest concerns was that if the growth of big science continued in this way, there would be no scientist who would be able to read every paper ( 1974 , p. 165). Even though we have not reached the number of journals estimated by Price, scientific outputs have still been increasing rapidly, and science is more difficult to follow than ever. In fact, the 90% of the research papers are never cited, and 50% of published research papers are never read by anyone else than the authors, reviewers and editors (Tripathy and Tripathy 2017 , p. 198).

One of the most important problems caused by big science is the inequality of scientific practices in various fields. Big science requires large budgets, diverse research groups with numerous staff members and big laboratories. The high costs of big science create a continuous interplay between the status system, which depends on honour and esteem, and class (Merton 1968 , p. 57). According to Allison and Stewart ( 1974 , p. 599), several publications and citations are affected by this inequality. One of the problems that creates this inequality is disciplinary differences: authors’ productivity depends on their work discipline, popularity and experience (Allison 1980 ; Merton 1968 ). Even today, big science provides a cumulative advantage for some scientists and disciplines. This cumulative advantage, in turn, affects the distribution of science funds (Bol et al. 2018 ) and other scientific career decisions (Petersen and Penner 2014 ). Scientific rewards are much more unequally distributed than other well-being outcomes (Xie 2014 , p. 810). For these reasons, the general characteristics of each discipline should be understood, and decisions should be made according to these characteristics to be able to make the right decisions in research evaluations.

Through the examination of the development of the LIS field, the same inequality can be seen. Over the years, studies have revealed that although the field is relatively small in the social sciences, it has several sub-fields, and the characteristics of these sub-fields are different from each other in terms of publication and citation patterns, authorship structures, production frequencies, etc. (Åström 2010 ; Moya-Anegón et al. 2006 ; White and McCain 1998 ). Besides, the development of sub-fields is directly affected by time and trends. For example, the number of articles written using terms such as ‘information technology’, ‘social network analysis’ or ‘citations’ has increased in recent years, but traditional librarianship topics such as librarianship, archiving or cataloguing have shown a decreasing trend (Larivière et al. 2012 , pp. 1006–1009). While this can be advantageous for some sub-fields, it negatively affects the visibility of more traditional fields and causes an unequal distribution of funds and resources.

The main aim of this study is to determine the sub-fields of the LIS field, reveal the potentials of these fields and make predictions of each sub-field. This will highlight the different scientific practices within the same discipline, which must then be taken into consideration when making decisions. The research questions are as follows:

• What is the current structure of the LIS field and its sub-fields? Is there a significant difference between the sub-fields and publication/citation patterns?
• Based on a 10-year forecast using the publication information produced in the LIS field, what size increase might be expected in the number of future publications?
• Is it possible to predict the number of future citations? What are the citation potentials of the sub-fields?
• How will the number of references cited in LIS papers change in the future?
• Will the co-authorship patterns in the LIS field change in the future?
• Are the quantitative predictions consistent, and do they provide valid insights for the future?
• What are the emerging topics of the LIS field? Is it possible to predict future topics of LIS?

Literature review

The literature review is organized into two main parts. The first part presents the subject distribution of papers published in the LIS field which use time series analyses. In the second part, various studies using time series analysis in scholarly communication and research evaluation fields are summarized. The explanation about the use of time series analysis is given in the Methodology section.

Time series analysis studies in LIS

Time series analysis has been applied in the LIS literature to provide forecasts on four different sub-topics: Bibliometrics, health sciences, management and social media. To define main application areas of time series analysis in the field of LIS, 452 papers published in LIS and indexed in Web of Science were evaluated. 2 (see Fig. ​ Fig.1 1 ).

Most used keywords of the time series analysis studies in LIS (The sunburst graph was created by using Flourish Studio ( https://app.flourish.studio/ ). Keyword occurrences were calculated by using VOSviewer. Before the calculation, the keyword standardization process was conducted.)

An article by Bates et al. ( 1999 ) is the most cited paper with its 768 citations in the dataset, which includes publications indexed in Web of Science’s Information Science and Library Science category. The article evaluated the impact of computerised physician order entries to reduce the number of medication errors. The authors used prospective time series analysis to calculate the effectiveness of computerised systems for medications. As a result, it is found that computerised systems resulted in a large decrease in medication errors. The second-most cited paper (372 citations) evaluated time series data for online product reviews to understand the effects of word of mouth on online shopping (Li and Hitt 2008 ). The third-most cited article (283 citations) was written by the founder of CiteSpace and his colleagues (Chen et al. 2010 ). The authors used time series analysis to introduce a new multi-perspective co-citation analysis method for information science literature. The most-cited articles from three different sub-topics prove the subject diversity of publications which used time series analysis methods and techniques.

The bibliometric studies using time series analysis are focused on research evaluations, bibliometric indicators and scientometric visualisations. These studies have sought to reveal the relation between early citations and cumulative advantage (Adams 2005 ), evaluate the effectiveness of monetary support systems (Tonta 2018 ), understand the citation trajectories of Nobel prize winners in economics (Bjork et al. 2014 ), visualise or discover the intellectual structure of disciplines (Ma 2012 ) or events (Clausen and Wormell 2001 ), analyse the evolution of research topics (Wu et al. 2014 ), predict citation counts (Abrishami and Aliakbary 2019 ), observe the effects of science policy changes on the number of publications (Baskurt 2011 ), forecast research activities (Bildosola et al. 2017 ), detect emerging/leading papers (Iwami et al. 2014 ) and evaluate research metrics (Liu and Rousseau 2008 ; Ye and Rousseau 2008 ). The time series analysis techniques have been used in bibliometric studies since the early 1990s, and it is still one of the preferred methods in the literature. The main reason for this choice might be explained by the policymaking mission of research evaluations. Following the impact of research policy changes or detecting number of future citations provide important findings to the policymakers to enhance evaluation processes.

Time series analyses have also been used in the papers on health information. In recent years, the studies in health information have focused on evaluating electronic health records, predicting health risks (Perrote et al. 2015 ), optimising drug-drug interaction alert rules using electronic health records (Simpao et al. 2015 ), understanding information-seeking behaviours on health subjects (Huerta et al. 2016 ) and monitoring mental health discussions on Twitter (McClellan et al. 2017 ). The whole world has witnessed how long- and short-term predictions on health issues important during COVID-19 times. It is expected to see a publication explosion in this field in the future. Studies that make predictions on various issues related to the COVID-19 have started to be published in the literature (e.g. Jiang et al. 2020 ; Salgotra et al. 2020 ). Although there are many “unknown unknowns” exists about the virus, time series analysis is likely to be more popular among policymakers by providing a range of scenarios (Grogan 2020 ).

Economics and management sub-subjects of LIS field are also conducted research by using time series analysis. The papers have focused on telecommunication infrastructure and its relations to economic growth/activity (Cronin et al. 1991 ; Dutta 2001 ), disseminating economic census data (Zeisset 1998 ) and early detection of an economic bubble (Dmitriev et al. 2017 ). The last subject category, social media, can be accepted as a part of management subject. During the social media age, the predictions on big data (Niu et al. 2017 ; Saboo et al. 2016 ), social media analyses (Luo and Zhang 2013 ; Zhang et al. 2019 ), word of mouth (Li Hitt 2008 ) and election analyses on Twitter (Conway et al. 2015 ) are some of the important research topics.

The thematic diversity of LIS studies that have used time series analyses demonstrates that this is an essential method for scholars working in this field and is not limited to forecasting. In this study, the main aim of using a time series analysis was to make predictions about research outputs for the LIS field.

Prediction types in the field of scholarly communication and bibliometrics

Forecasting the future is one of the most frequently discussed subjects in bibliometrics and research evaluation studies. Predictions are often made to estimate Nobel Prize Laureates by considering publication and citation patterns. The Web of Science group has provided this well-known prediction mechanism for Nobelists since 2002 (Bourke-Waite 2019 ). Since 1970, millions of indexed publications and citations to these papers have been evaluated and estimations made. Until 2019, 50 Nobel prize winners who were on the list of citation laureates won the Nobel Prize. Of these, 29 researchers received the prize within 2 years of being nominated. Besides the Web of Science Group, there have been other numerous papers published in the literature to predict Nobel Prize winners (e.g., Ashton and Oppenheim 1978 ; Claes and De Ceuster 2013 ; Siegel 2019 ); however, Gingras and Wallace ( 2010 ) warned against the limits of bibliometric tools for predicting Nobel Prize winners due to the rapid growth of disciplines and the halo effect.

Another important area of predictive research is estimating the future number of publications and citations using different tools, techniques and perspectives. Leydesdorff ( 1990 ) sought to estimate the national performance of EEC (European Economic Community) countries and the US using time series analysis models. He found that it is possible to predict the following year’s publication statistics. In Rousseau ( 1994 ) proposed a double exponential model for first citation processes. He aimed to find a model for first citations, and he suggested two models to predict the total number of articles in a fixed group that would ever be cited. In Burrell ( 2003 ) developed the theory of stochastic models to predict the future citation patterns of individual papers. He found that expected citation count was a linear function of the current number, thus proving the idiom ‘success breeds success’.

Chen ( 2012 ) proposed a theoretical and computational model to predict future citations using three metrics: modularity change rate, cluster linkage and centrality divergence. The results indicate that the model could successfully predict future citations. Also, authors’ collaboration statistics and the number of references were found to be good predictors of global citations.

From the citation perspective, Abbasi et al. ( 2011 ) created a model to identify the effects of co-authorship networks on scholars’ performance. As a result, they recommended using researchers’ networks to predict scholars’ future performance. Tahamtan et al. ( 2016 ) reviewed the literature and presented 28 factors affecting the number of citations, these factors were then sorted into three main categories: paper-related factors (such as quality of papers, document type, etc.), journal-related factors (such as the impact factor or journal’s language) and author-related factors. The authors indicated that it is possible to predict the frequency of citations by considering these factors. Similarly, Chakraborty et al. ( 2014 ) developed a two-stage prediction model that produced better results for highly cited papers, and the authors suggested using this model to predict seminal papers in the scientific fields. The authors indicated that although the publication’s authors and venue are crucial for gathering citations, the features related to the papers’ content are more effective for long-term citation predictions. Another study on estimating the factors affecting the number of citations received by articles published in 12 crime psychology journals showed that author impact might be a more powerful predictor of how many times an article is cited than the venue (journal) of publication (Walters 2006 ).

Brody et al. ( 2006 ) examined the relationship between the number of early downloads and the number of citations received for the publications on Arxiv. The study showed that there was a correlation between early downloads and citation impact. Besides, the longer the period for which downloads were counted, the higher the correlation between downloads and citation impact. The authors concluded that the 2 year citation impact should be estimated using 6 months of download statistics.

One of the most recent studies on citation data and forecasting investigated whether the number of volumes that the journals published affected the impact factors of the journals (Zhang 2020 ). The results showed that if the increase of volumes is consistent and significant, a decrease of impact factors is unlikely.

Unlike the other studies mentioned above, some of the studies in the literature did not aim to estimate the number of citations using different statistical data but rather to predict future technologies using citation data. Small ( 2006 ) proposed using clustering, mapping and string formation to track and predict growth areas in science. Érdi et al. ( 2013 ) developed a new model to detect new technological hot spots by clustering patent citation data. Similarly, the Bass and ARIMA models, which are time series analysis models, were utilised to forecast development trends based on patent data (You et al. 2017 ). Kwon and Geum ( 2020 ) indicated that promising inventions can be identified by considering the number of backward citations as the link with previous knowledge. All these studies demonstrate that time series analysis can not only be used to predict the number of outputs in the literature but also to forecast technological developments.

Considering the number of forecasting studies in the literature, predictions provide important findings for scholars, policymakers and managers working in LIS and its sub-fields. Through these findings, it is possible to develop policies, identify the problematic practices and measure the effects of policy changes.

Methodology

Data structure.

To achieve the aims of the study, an advanced search of the Web of Science core indexes (SCI, SSCI and A&HCI) was conducted on 12 December 2019 using the search string WC = ( “Information science and Library Science” ) AND LA = ( English ) AND PY = (1921–2018) AND DT = ( article ). Although the Information Science and Library Science category is only indexed in SSCI, up to 5000 articles were indexed in SCI and A&HCI but not SSCI. Therefore, all three core indexes were included in the study to cover all studies in the field.

The oldest paper within the author’s subscription limits was from 1921, so that year became the starting point. Since the research was carried out before the end of 2019, the year 2019 was excluded from the scope of the research to avoid manipulation of the data and findings. However, the publication and citation data for 2019 were used to validate the success of the predictions made in this study. Also, only articles written in the English language were considered to avoid manipulation due to document type or language differences.

A total of 123,742 articles were analysed and evaluated within the context of this study. The metadata of all articles was downloaded as tab-delimited text using the Web of Science exporting features. A total of 248 different .txt files were downloaded because of the download limits of Web of Science (500 records per download). Then, all the .txt files were combined using the command prompt. 3 After creating one data file, a deep data cleaning and unification process was conducted. The main characteristics of the dataset are shown in Fig. ​ Fig.2 2 .

The main characteristics of the dataset.

The articles in the dataset were published in 174 different journals. To answer the research questions, the dataset was divided into four different sub-fields using social network analysis and clustering methods.

Clustering and determination of LIS sub-fields

Two different networks were created for subject clustering. One was a co-cited journal network and the other was a co-occurrence of keywords network. The creation phases of the networks were:

• Co-cited journals The VOSviewer visualisation tool was used to create a co-citation network. Before creating the network, the names of the cited journals were standardised. During the standardisation process, different variations of journal names (e.g., Libr Trends, Lib Trends and Library Trends) and title changes (e.g., American Documentation, JASIS and JASIS&T) were considered. All journal names were unified. As a result, 537,227 sources were listed in our dataset. The limit for the minimum number of citations for a source was set at 20; 11,253 sources met this threshold. The co-citation network shown in Fig. ​ Fig.1 1 presents the top 1000 co-cited journals in the network.

Clustering for journals in the dataset (networks of co-cited journals and keyword co-occurrence)

The main reason for creating two different network maps was to cross-validate the subject distribution of the dataset. Based on the clustering results, five clusters were determined for each network map. The clusters determined by most-occurred keywords were parallel with the co-cited journal network. It provided the opportunity to verify the accuracy of the classification. For both networks, the purple clusters were considered to be part of the green cluster. Therefore, the main subjects were classified into four main clusters for our study: librarianship and law librarianship (traditional library studies), health information in LIS , scientometrics and information retrieval and management and information systems .

Although some authors have argued that the journal citation reports (JCR) subject classification is problematic because it covers management information system (MIS) journals, which are different from other sub-fields (Larivière et al. 2012 , p. 999; Ni and Sugimoto 2011 ), our classification results for this field align with previous studies in the literature (e.g., Moya-Anegón et al. 2006 ; Ni and Sugimoto 2011 ; Tseng and Tsay 2013 ) that the field is generally divided into four sub-fields: information science (including information retrieval and information seeking), library science (practical and research-oriented), MIS and scientometrics. In this study, we also added health information to these classifications.

The main limitation of the classification used in this study was the journal-based approach. Some problems were determined for the journals which publish papers on two or more different topics. For example, the journal Health Information and Libraries was classified into the librarianship and law librarianship cluster by co-cited journal analysis, however, the main subject field of the journal is health libraries (Overview - Health Information and Libraries Journal 2020 ). To avoid that kind of problems, an expert control mechanism was conducted and content information from the articles published in that journal was used to decide the journal’s main focus. Additionally, if a journal was not listed in the network map, the same process was applied. For example, African Journal of Library Archives and Information Science was classified into the librarianship and law librarianship cluster using this method. The distribution of the articles into classes is shown in Fig. ​ Fig.4 4 .

Distribution of journals into subject clusters

Each of the subject fields has different features even though they are all in the same subject category—LIS. Therefore, it is important to understand the structures of these sub-fields and their potentials. Although the librarianship and law librarianship category contains up to 50% of the articles, it is the field with the lowest citation rate. Furthermore, collaboration is more common for health information in the LIS literature. To understand the differences between the sub-fields, the Kruskal-Wallis test was conducted. The test showed that:

• The sub-categories of the articles significantly affect the number of publications that the articles cite ( H [3] = 17951.379, p < 0.001),
• The sub-categories of the articles significantly affect the number of times an article is cited ( H [3] = 19807.543, p < 0.001) and
• The sub-categories of the articles significantly affect the number of authors per paper ( H [3] = 20557.826, p < 0.001).

The test results demonstrate that even if the study focused on a specific category, the sub-fields of that category could have different structures, and thus, evaluations must consider these differences.

Time series analysis and time series forecasting

Many systems that we use today produce time-based data, which can be used to make various inferences. By using the data produced as a result of observations or experiments, problems with the system can be revealed, and predictions can be made about the future. The systematic approach to answering mathematical and statistical questions posed by time correlations is called time series analysis (Shumway and Stoffer 2006 , p. 1). This method of analysis has been used in various fields, from economics to geographical sciences, and it has a wide range of applications. The literature review section summarized different variations of time series analyses in the LIS literature to achieve different aims.

Forecasting is one method of time series analysis and is used to provide the t + 1 value of future time by evaluating the t number of available observations (Box et al. 2008 , p. 2). The forecasting process includes seven phases: (1) problem definition, (2) data collection, (3) data analysis, (4) model selection and fitting, (5) model validation, (6) forecasting and model deployment and (7) monitoring forecasting model performance (Montgomery et al. 2008 , p. 12). SPSS Statistics 23 (IBM) was used to conduct the model selection, fitting, validation, deployment and monitoring phases of this study.

There are different types of time series data, and this must be considered when choosing the analysis method. The well-known data types in time series analyses are trend data, seasonal data and cyclical variations. As seen in Fig. ​ Fig.2, 2 , our dataset shows a linear trend, and thus the analyses were conducted to predict the future of this trend. The only exception for our data was the citations. Any publication requires a certain period to gather citations, and this period varies from discipline to discipline. The decrease in the number of citations over the last 8 years (Fig. ​ (Fig.2) 2 ) indicates that the half-life of citations in the LIS field is 8.3 (Incites Journal Citation Reports 2018 ). To prevent this decline from adversely affecting the results of the forecasting, only citation data up to 2010 were used. Thus, time series forecasting was applied using the period from 1921 to 2010.

Unusual events, disturbances or errors that might affect time series data are known as outliers (Box et al. 2008 , p. 536). There are different methods to remove outliers from the data or to normalise the data to provide strong predictions. Removing or normalising citation data was vital for this study because there were too many extreme values, and without processing the data to remove outliers, it would have been impossible to provide a powerful forecast for research outputs in the LIS field. To achieve this aim, median scores of the number of references and the number of citations per year were used to normalise the data. Additionally, autocorrelation and partial autocorrelation plots were created ( Appendix ).

The results of the forecasting analyses are presented in this section according to the number of publications, number of citations, number of references and number of authors per title.

Number of publications

As shown in Fig. ​ Fig.5, 5 , it is predicted that the number of publications in the LIS field will increase in the future. The average number of English language articles published per year in these 97 years was 1262; however, 50% of these articles were published in the last 20 years. 4 Thus, an increasing publication pattern can be easily seen in Graph 1 in Fig. ​ Fig.5 5 (all LIS fields). Forecasting the number of publications for the whole LIS literature produced significant results [ Ljung Box Q (18) = 30.286, df = 18, p = 0.035, ARIMA (0, 1, 0) = 0.539, SE = 0.162, p = 0.001], and according to the results, 3974 publications were predicted for 2019 and 4632 for 2027.

Forecasting for number of publications

Because the expected number of articles for 2019 was estimated at 3974, and most of the articles published in 2019 are indexed in the Web of Science, it was possible to compare the forecast to the actual number of publications in 2019. A total of 4412 English language articles were published in 2019 and indexed before October 2020. 5 This shows that the number of publications will likely increase beyond the prediction, as that many articles are not expected until 2024 in the time series analysis. However, this number is still within the limits of the upper confidence level. If we follow the upper confidence level of the forecast to estimate the near future, there may be 6069 published articles in 2028. This means that if the upper confidence levels are actualised, a total of 52,807 articles could be published between 2019 and 2028.

Although the forecasting tests for sub-fields produced meaningful results, the data were not sufficient to make predictions. It is possible to follow the data from the trend lines and Ljung-box scores. Results of the analysis suggest that increases are expected in the number of publications that will be produced in all sub-fields. This is evidenced by the fact that the forecasts and the actual numbers are quite similar (see Table ​ Table1), 1 ), indicating that estimating the number of publications in the LIS field and its sub-fields is possible using time series analysis.

Expected and actual number of publications in 2019

Number of citations

Approximately 17% of the articles published in the LIS field have received 80% (1,209,824) of the citations for the whole literature. These statistics are important in terms of showing the existence of core articles in the LIS field. It is important to note that some publications receive numerous citations while others do not. Fig. ​ Fig.6 6 shows the distribution of citations received by sub-fields.

Distribution of citations according to sub-fields

Analysis of the dataset shows that two articles received 10,000 citations. These articles were classified into the sub-fields of management and information systems and health information in LIS. The citation potentials are different for each category. For example, papers published in the sub-field of management and information systems are more likely to be cited than those published in the sub-field of librarianship and law librarianship. One of the main features of citation data is their skewness (Bornmann and Leydesdorff 2017 ), and my dataset was no exception. This skewness makes it difficult to produce accurate forecasts for the number of citations in the future.

In addition to the skewness of citation data, other problems are literature obsolescence and citation half-lives. Since the cited half-life of the LIS field is 8 years, it is not possible to make an accurate prediction using data from the last 8 years. For this reason, forecasting only covered the years 1921–2010, and the last 8 years were excluded.

Fig. ​ Fig.7 7 presents the forecasting results which show that the most consistent prediction could be obtained by analysing the entire discipline. However, for the field-based analyses, the predictions did not produce meaningful results. The results indicate that half of the publications could be cited 20 or more times per year in the future. Considering that the median number of citations currently is 10 per year, this prediction of a major increase in citation counts is possible. However, it is estimated that the number of citations received in the LIS field might exceed 100,000. If we assume the same upper confidence level as we did for the number of publications, the upper confidence for the total number of citations is estimated to be 141,000. Since the distribution of median values does not offer a linear trend for sub-fields, it is difficult to predict which sub-fields will receive more citations. Furthermore, the half-life may be different for each sub-field. This is one of the factors that makes forecasting difficult. Considering all these factors, future analyses might produce more meaningful results.

Forecasting for citations

Number of references per paper

While the number of references that could be cited in publications was more limited in the past, with the increase in the number of publications, there has also been a significant increase in the number of references made in studies. It is possible to monitor this increase from the trendlines in Fig. ​ Fig.8. 8 . The forecast predicts that a total of 300,000 references will be listed in the LIS literature in 2028. Half of the publications are expected to cite at least 63 sources. In 2018, this number was 47. The tests for forecasting produced significant results, and upper and lower confidence level scores were very close, indicating the accuracy/consistency of the future prediction.

Forecasting of the number of references

Despite the success in forecasting the future number of references, it is difficult to make a similar forecast for the sub-fields because of the differences between the fields and the skewness of the reference/citation data. Although they are in the same main subject category, the citation patterns are different for each sub-field. The average number of references per article is 15 (median = 21) in the field of librarianship and law librarianship, 26 (median = 21) in scientometrics and information retrieval, 30 (median = 26) in health information in LIS and 42 (median = 37) in the management and information systems fields.

Forecasting author collaborations

The average number of authors per paper in the LIS literature is two, and the median is one. Thus, scholars in the LIS field generally prefer to work alone. However, health information in LIS is the most collaborative sub-field of LIS literature. The article entitled ‘Academic domains as political battlegrounds: A global enquiry by 99 academics in the fields of education and technology’ 6 . is the most collaborative paper with 99 authors. The article is classified as part of the librarianship and law librarianship sub-field in our dataset. The main statistics for authorship patterns are shown in Table ​ Table2 2 .

Co-authorship patterns of LIS field

The forecast results show that the average number of authors per paper is three. In the next 10 years, this number is expected to increase to 3.6. Considering the trendline over the past 97 years, this expected result is reasonable. Fig. ​ Fig.9 9 presents the forecasts for collaboration patterns in the LIS literature.

Forecasting of collaboration patterns

Possibility of consistent forecasting in the publish or perish world

The analyses above demonstrate the difficulty of predicting research outputs. Every year, the number of publications increases. Since no regular trend can be seen in this increase in the number of publications, any predictions we make today are minimum values for the future. Table ​ Table1 1 presents one example of this. All three graphs in Fig. ​ Fig.10 10 show the estimated increase per year in comparison with the previous period. The periods are determined by considering the years that had increases in the number of publications.

Forecasts by different periods

Figure ​ Figure10 10 demonstrates that the number of publications does not have a regular trend. Thus, there is the possibility that no prediction will accurately forecast the number of future publications. If the trend until 1950 had continued to today, the number of publications in the LIS literature today would be 37,049 (30% of today’s actual number). If the data from 1970 were used, the number would be 76,612 (61% of today’s actual number), and if the data from 2000 were used, the number would be 117,336 (94% of today’s actual number). While it is possible to say that forecasts in recent years have been more accurate, that is, the publication trends have been similar in recent years, the unpredictability should be expected to continue regardless of any changes in research performance evaluation systems. Besides, it should be kept in mind that the number of publications may be indirectly affected by unexpected emerging issues such as COVID-19 that significantly affect the publishing patterns of the authors.

It is difficult to estimate the total number of citations using the data up to 1970 because of the cumulative nature of citations. However, predictions using data up to 2000 produced forecasts that are close to reality. Using data up to 1970, it was estimated that the average number of references per paper would be 67 in 2018. However, the data after 1970 changed the situation. Using the data until 2000, the estimated average number of references per paper in 2018 was 22, while the actual average number of references in 2018 was 51. Thus, the number of references in publications have increased far beyond the predictions made using more recent data.

Forecasting the research subjects

The findings confirm that the entire LIS field will face many more publications in the future with the spread of publish or perish culture. Therefore, the key to following the developments in LIS is to go beyond numbers. Although it is difficult to forecast the potential future of LIS subjects by using just numbers, making inferences by looking at the emerging subjects in recent years is possible. Figure ​ Figure11 11 shows the most-used keywords of the papers published in the last two years in the LIS field. 7

Emerging subjects of the LIS field (Flourish Studio was used to create the radial tree)

Figure ​ Figure11 11 shows a network of keywords that includes five clusters. The four clusters are parallel with the classification of this study. However, a new cluster named “COVID-19” has been added to the LIS literature as expected. The emerging subjects of each sub-field are:

• COVID-19 All the countries have been fighting with COVID-19 since December 2019. According to WHO’s COVID-19 Global Research Database ( Global Research on Coronavirus Disease ( COVID- 19) 2020 ) a total of 123,959 publications were published from the day of the outbreak to November 7, 2020. The subject is also popular in the LIS field. Social media, fact-checking, governmental responsibilities during the pandemic such as political communication, transparency and participation, digital journalism and fake news are the important subjects of LIS field recently. The cluster proves the importance of LIS research focusing on open and correct information all over the world during the pandemic.
• Bibliometrics and information retrieval The keywords of this cluster show that bibliometrics and information retrieval issues converge to each other with the developments of computational techniques. Machine learning, text mining, topic modelling and sentiment analyses are used for bibliometric studies such as content-based citation analyses and digital humanities. Also, scholarly communication subjects like peer-review, societal impact, incentives, predatory journals, language (multilingualism) and rankings are important keywords of this cluster. As indicated in the literature review part, predictions are still important for this sub-field.
• Librarianship and law librarianship The effects of COVID-19 is also observed in this cluster (e.g. e-learning, e-resources). Information literacy plays a vital role among researchers, students and the public during COVID-19 times. Therefore, traditional librarianship subjects will be important to solve information problems of individuals in the future. Besides, digitization and preservation of archival materials are the other popular subjects of the cluster.
• Health information in LIS Many studies in this sub-field focus on disadvantaged groups in recent years. Studies on inequality, refugees and genders can be evaluated in this content. Also, public access to health information, health communication and electronic health records are popular subjects and related to COVID-19 pandemic.

Discussion and conclusion

The study suggested a forecasting mechanism for research outputs in the LIS field. The main aim of the study was to inform scholars and policymakers about the future of research in this field. Nowadays, articles are often only read by a few people (Eveleth 2014 ; Simkin and Roychowdhury 2015 ; Tripathy and Tripathy 2017 ), and the main purpose of publishing is to achieve a numerical advantage rather than further the development of science. Although many researchers have emphasised that the current system should change, there have not yet been any concrete changes.

First, we revealed that publishing, citation and collaboration patterns differ between the sub-fields in the LIS literature. It is a well-known fact that apples and oranges are incomparable in research evaluations (Johnes and Johnes 1992 ); however, this study shows that it is also difficult to compare apples to each other because there are different types of apples (e.g., red, granny smith, honeycrisp, etc.). According to the results of the study more articles are published in traditional librarianship journals, and these journals tend to be cited less than others in the field. Articles published in the management dimension of the LIS field have greater citation potential than other sub-fields. This explains why management journals tend to have the highest impact factor in JCR among the LIS journals. This study shows the sub-field differences in LIS, and any evaluations based on categories should consider the sub-fields and their different characteristics.

The findings of this study indicate the number of publications and citations will continue to increase each year unless there is a change in research evaluation systems. This could lead to an uncontrollable mass of publications in the LIS field. The upper confidence levels estimated by the forecasting model produced in this study were already realised in 2019, demonstrating that this increase will be huge. However, it is difficult to forecast the future of sub-fields because the publication trends in sub-fields differ greatly from the general framework. If the existing systems continue, the inequality between LIS sub-fields will continue to grow. The meaning of following the current research evaluation systems is that the production of papers will continue to increase, and some of the sub-fields will not be able to benefit from future opportunities due to their disadvantages. For this reason, decision-makers and managers must consider field- and time-based differences in their research evaluation tasks.

One of the important results revealed in this study is the predicted increase in the number of publications, citations and references. Given that evaluations are made using citation data, the growing amount of data will make future evaluations more difficult. For this reason, supporting programmes such as the Initiative for Open Citations, which aims to promote the unrestricted availability of scholarly citation data, may also be useful for managing data in the future.

The results show that a lot of papers which have long reference lists will be produced, they will cite each other, more authors will work together to write papers. However, their contents and levels will be different from each other. Many of the studies have predicted that publishing will change in the future as a consequence of these differences. For example, Priem ( 2013 ) claimed that publishing forms, reward systems, measurement tools and peer-review systems will soon change. Similarly, Waldrop ( 2008 ) and Kendall ( 2015 ) stated that open science will be the new norm and that we will experience many changes to authorship and research evaluation systems in the next years. The predictions for the future of the publishing system is also the subject of the LIS field. This study proves the astonishing diversity of research subjects of the LIS field and tries to show the importance of looking beyond numbers.

Acknowledgements

This research was supported by a research grant from the Polish National Agency for Scientific Exchange, NAWA Poland (PPN/ULM/2019/1/00062). I would like to thank Emanuel Kulczycki, Güleda Doğan, Ola Swatek and the anonymous peer-reviewers for their meticulous reading of the paper and their invaluable suggestions.

Residual autocorrelation function (ACF) and residual partial autocorrelation function (PACF) plots

1 The search was conducted on 7 July 2020 using the term PY = (1963–2020) in Web of Science’s core indexes: social sciences citation index (SCI), social sciences citation index (SSCI), Arts and humanities citation index (A&HCI), Emerging sources citation index (ESCI), Conference proceedings citation index (CPCI) and Book citation index (BKCI).

2 To access papers on/using time series analyses, the search string TS=(“time series” OR “forecasting analys*s” OR ARIMA OR “Exponential smoothing”) AND WC=(“Information science & library science”) was used. The search was conducted on 15 May 2020 using the Web of Science core indexes.

3 The text files were combined into a single file using the following steps: (1) open command prompt, (2) enter the folder, (3) use the code >> copy *.txt join.txt .

4 Although half of the articles were published in the last 20 years, using the entire past provides advantages for time series analyses using linear data (Jones 1964 , p. 47). The rate of increase in the number of publications over the years is one of the important factors for the success of the forecast. To provide accuracy on forecasts, the entire 97-year period was used for prediction.

5 A total of 137 articles were identified as early access. These articles might be covered by volumes/issues published in 2020. The total number of articles excluding early access articles was 4275.

6 https://journals.sagepub.com/doi/full/10.1177/0266666916646415 .

7 An advanced search was conducted on 5 November 2020 using the search string WC= ( “Information Science and Library Science” ) AND PY = (2019–2020) to gather the publication data for the last two years. All Web of Science indexes including ESCI, CPCI and BKCI were included to be able to cover more papers. Articles, reviews and proceedings were considered. A total of 13,856 papers were evaluated to analyze the emerging subjects of the field.

• Abbasi A, Altmann J, Hossain L. Identifying the effects of co-authorship networks on the performance of scholars: A correlation and regression analysis of performance measures and social network analysis measures. Journal of Informetrics. 2011; 5 (4):594–607. doi: 10.1016/j.joi.2011.05.007. [ CrossRef ] [ Google Scholar ]
• Abrishami A, Aliakbary S. Predicting citation counts based on deep neural network learning techniques. Journal of Informetrics. 2019; 13 (2):485–499. doi: 10.1016/j.joi.2019.02.011. [ CrossRef ] [ Google Scholar ]
• Adams J. Early citation counts correlate with accumulated impact. Scientometrics. 2005; 63 :567–581. doi: 10.1007/s11192-005-0228-9. [ CrossRef ] [ Google Scholar ]
• Allison PD. Inequality and scientific productivity. Social Studies of Science. 1980; 10 (2):163–179. doi: 10.1177/030631278001000203. [ CrossRef ] [ Google Scholar ]
• Allison Paul D, Stewart JA. Productivity differences among scientists: Evidence for accumulative advantage. American Sociological Review. 1974; 39 (4):596–606. doi: 10.2307/2094424. [ CrossRef ] [ Google Scholar ]
• Ashton SV, Oppenheim C. A method of predicting Nobel Prizewinners in chemistry. Social Studies of Science. 1978; 8 (3):341–348. doi: 10.1177/030631277800800306. [ CrossRef ] [ Google Scholar ]
• Åström F. The visibility of information science and library science research in bibliometric mapping of the LIS Field. Library Quarterly. 2010; 80 (2):143–159. doi: 10.1086/651005. [ CrossRef ] [ Google Scholar ]
• Baskurt OK. Time series analysis of publication counts of a university: What are the implications? Scientometrics. 2011; 86 :645–656. doi: 10.1007/s11192-010-0298-1. [ CrossRef ] [ Google Scholar ]
• Bates DW, Teich JM, Lee J, Seger D, Kuperman GJ, Ma’luf N, Boyle D, Leape L. The impact of computerized physician order entry on medication error prevention. Journal of the American Medical Informatics Association. 1999; 6 (4):313–321. doi: 10.1136/jamia.1999.00660313. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bildosola I, Gonzalez P, Moral P. An approach for modelling and forecasting research activity related to an emerging technology. Scientometrics. 2017; 112 :557–572. doi: 10.1007/s11192-017-2381-3. [ CrossRef ] [ Google Scholar ]
• Bjork S, Offer A, Söderberg G. Time series citation data: The Nobel Prize in economics. Scientometrics. 2014; 98 :185–196. doi: 10.1007/s11192-013-0989-5. [ CrossRef ] [ Google Scholar ]
• Bol T, de Vaan M, van de Rijt A. The Matthew effect in science funding. Proceedings of the National Academy of Sciences. 2018; 115 (19):4887–4890. doi: 10.1073/pnas.1719557115. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bornmann L, Leydesdorff L. Skewness of citation impact data and covariates of citation distributions: A large-scale empirical analysis based on Web of Science data. Journal of Informetrics. 2017; 11 (1):164–175. doi: 10.1016/j.joi.2016.12.001. [ CrossRef ] [ Google Scholar ]
• Bourke-Waite, A. (2019, September 24). The Web of Science Group reveals annual citation laureates of ‘Nobel class’. https://clarivate.com/news/the-web-of-science-group-reveals-annual-citation-laureates-of-nobel-class/
• Box GEP, Jenkins GM, Reinsel GC. Time series analysis: Forecasting and control. 4. NewJersey: John Wiley; 2008. [ Google Scholar ]
• Brody T, Harnad S, Carr L. Earlier web usage statistics as predictors of later citation impact. Journal of the American Society for Information Science and Technology. 2006; 57 (8):1060–1072. doi: 10.1002/asi.20373. [ CrossRef ] [ Google Scholar ]
• Burrell QL. Predicting future citation behavior. Journal of the American Society for Information Science and Technology. 2003; 54 (5):372–378. doi: 10.1002/asi.10207. [ CrossRef ] [ Google Scholar ]
• Chakraborty T, Kumar S, Goyal P, Ganguly N, Mukherjee A. Towards a stratified learning approach to predict future citation counts. IEEE/ACM Joint Conference on Digital Libraries. 2014 doi: 10.1109/JCDL.2014.6970190. [ CrossRef ] [ Google Scholar ]
• Chen C. Predictive effects of structural variation on citation counts. Journal of the American Society for Information Science and Technology. 2012; 63 (3):431–449. doi: 10.1002/asi.21694. [ CrossRef ] [ Google Scholar ]
• Chen C, Ibekwe-SanJuan F, Hou J. The structure and dynamics of cocitation clusters: A multiple-perspective cocitation analysis. Journal of the American Society for Information Science and Technology. 2010; 61 (7):1386–1409. doi: 10.1002/asi.21309. [ CrossRef ] [ Google Scholar ]
• Claes AGP, De Ceuster MJK. Estimating the economics Nobel Prize laureates’ achievement from their fame. Applied Economics Letters. 2013; 20 (9):884–888. doi: 10.1080/13504851.2012.758836. [ CrossRef ] [ Google Scholar ]
• Clausen H, Wormell I. A bibliometric analysis of IOLIM conferences 1977–1999. Journal of Information Science. 2001; 27 (3):157–169. doi: 10.1177/016555150102700305. [ CrossRef ] [ Google Scholar ]
• Conway BA, Kenski K, Wang D. The rise of Twitter in the political campaign: Searching for intermedia agenda-setting effects in the presidential primary. Journal of Computer-Mediated Communication. 2015; 20 (4):363–380. doi: 10.1111/jcc4.12124. [ CrossRef ] [ Google Scholar ]
• Cronin FJ, Parker BP, Colleran EK, Gold MA. Telecommunications infrastructure and economic growth: An analysis of causality. Telecommunications Policy. 1991; 15 (6):529–535. doi: 10.1016/0308-5961(91)90007-X. [ CrossRef ] [ Google Scholar ]
• Dmitriev A, Dmitriev V, Sagaydak O, Tsukanova O. The application of stochastic bifurcation theory to the early detection of economic bubbles. Procedia Computer Science. 2017; 122 :354–361. doi: 10.1016/j.procs.2017.11.380. [ CrossRef ] [ Google Scholar ]
• Dutta A. Telecommunications and economic activity: An analysis of granger causality. Journal of Management Information Systems. 2001; 17 (4):71–95. doi: 10.1080/07421222.2001.11045658. [ CrossRef ] [ Google Scholar ]
• Érdi P, Makovi K, Somogyvári Z, Strandburg K, Tobochnik J, Volf P, Zalányi L. Prediction of emerging technologies based on analysis of the US patent citation network. Scientometrics. 2013; 95 :225–242. doi: 10.1007/s11192-012-0796-4. [ CrossRef ] [ Google Scholar ]
• Eveleth, R. (2014, March 24). Academics write papers arguing over how many people read (and cite) their papers. Smithsonian Magazine . https://www.smithsonianmag.com/smart-news/half-academic-studies-are-never-read-more-three-people-180950222/?no-ist
• Gingras Y, Wallace ML. Why it has become more difficult to predict Nobel Prize winners: A bibliometric analysis of nominees and winners of the chemistry and physics prizes (1901–2007) Scientometrics. 2010; 82 :401–412. doi: 10.1007/s11192-009-0035-9. [ CrossRef ] [ Google Scholar ]
• Global research on coronavirus disease (COVID-19) (2020). https://www.who.int/emergencies/diseases/novel-coronavirus-2019/global-research-on-novel-coronavirus-2019-ncov
• Grogan, M. (2020, September 22). COVID-19 From A Time Series Perspective. Medium . https://towardsdatascience.com/covid-19-from-a-time-series-perspective-a5082903d836
• Huerta TR, Walker DM, Johnson T, Ford EW. A time series analysis of cancer-related information seeking: Hints from the health information national trends survey (HINTS) 2003–2014. Journal of Health Communication. 2016; 21 (9):1031–1038. doi: 10.1080/10810730.2016.1204381. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Incites Journal Citation Reports (2018). Category profile: Information science & library science (2018). https://jcr.clarivate.com/JCRCategoryProfileAction.action?year=2018&categoryName=INFORMATION%20SCIENCE%20%26%20LIBRARY%20SCIENCE&edition=SSCI&category=NU
• Iwami S, Mori J, Sakata I, Kajikawa Y. Detection method of emerging leading papers using time transition. Scientometrics. 2014; 101 :1515–1533. doi: 10.1007/s11192-014-1380-x. [ CrossRef ] [ Google Scholar ]
• Jiang F., Zhao, Z., & Shao, X. (2020). Time series analysis of COVID-19 infection curve: A change-point perspective. http://arxiv.org/abs/2007.04553 [ PMC free article ] [ PubMed ]
• Johnes G, Johnes J. Apples and oranges: The aggregation problem in publication analysis. Scientometrics. 1992; 25 (2):353–365. doi: 10.1007/BF02028091. [ CrossRef ] [ Google Scholar ]
• Jones RH. Spectral analysis and linear prediction of meteorological time series. Journal of Applied Meteorology. 1964; 3 (1):45–52. doi: 10.1175/1520-0450(1964)003<0045:SAALPO>2.0.CO;2. [ CrossRef ] [ Google Scholar ]
• Kendall, G. (2015, October 15). The future of scientific publishing: Let’s make sure it’s fair as well as transparent. The Conversation . https://theconversation.com/the-future-of-scientific-publishing-lets-make-sure-its-fair-as-well-as-transparent-48900
• Kwon U, Geum Y. Identification of promising inventions considering the quality of knowledge accumulation: A machine learning approach. Scientometrics. 2020 doi: 10.1007/s11192-020-03710-3. [ CrossRef ] [ Google Scholar ]
• Larivière V, Sugimoto CR, Cronin B. A bibliometric chronicling of library and information science’s first hundred years. Journal of the Association for Information Science and Technology. 2012; 63 (5):997–1016. doi: 10.1002/asi.22645. [ CrossRef ] [ Google Scholar ]
• Leydesdorff L. The prediction of science indicators using information theory. Scientometrics. 1990; 19 (3–4):297–324. doi: 10.1007/BF02095353. [ CrossRef ] [ Google Scholar ]
• Li X, Hitt LM. Self selection and information role of online product reviews. Information Systems Research. 2008; 19 (4):456–474. doi: 10.1287/isre.1070.0154. [ CrossRef ] [ Google Scholar ]
• Liu Y, Rousseau R. Definitions of time series in citation analysis with special attention to the h-index. Journal of Informetrics. 2008; 2 (3):202–210. doi: 10.1016/j.joi.2008.04.003. [ CrossRef ] [ Google Scholar ]
• Luo X, Zhang J. How do consumer buzz and traffic in social media marketing predict the value of the firm? Journal of Management Information Systems. 2013; 30 (2):213–238. doi: 10.2753/MIS0742-1222300208. [ CrossRef ] [ Google Scholar ]
• Ma R. Discovering and analyzing the intellectual structure and its evolution of LIS in China, 1998–2007. Scientometrics. 2012; 93 :645–659. doi: 10.1007/s11192-012-0702-0. [ CrossRef ] [ Google Scholar ]
• McClellan C, Ali MM, Mutter R, Kroutil L, Landwehr J. Using social media to monitor mental health discussions—Evidence from Twitter. Journal of the American Medical Informatics Association. 2017; 24 (3):496–502. doi: 10.1093/jamia/ocw133. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Merton RK. Thematthew effect in science: The reward and communication systems of science are considered. Science. 1968; 159 (3810):56–63. doi: 10.1126/science.159.3810.56. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Montgomery DC, Jennings CL, Kulahci M. Introduction to time series analysis and forecasting. NewJersey: John Wiley; 2008. [ Google Scholar ]
• Moya-Anegón F, Herrero-Solana V, Jiménez-Contreras E. A connectionist and multivariate approach to science maps: The SOM, clustering and MDS applied to library and information science research. Journal of Information Science. 2006; 32 (1):63–77. doi: 10.1177/0165551506059226. [ CrossRef ] [ Google Scholar ]
• Ni C, Sugimoto CR. Four-facets study of scholarly communities: Artifact, producer, concept, and gatekeeper. Proceedings of the 2011 ASIS&T Annual Meeting. 2011 doi: 10.1002/meet.2011.14504801343. [ CrossRef ] [ Google Scholar ]
• Niu N, Liu X, Jin H, Ye X, Liu Y, Li X, Chen Y, Li S. Integrating multi-source big data to infer building functions. International Journal of Geographical Information Science. 2017; 31 (9):1871–1890. doi: 10.1080/13658816.2017.1325489. [ CrossRef ] [ Google Scholar ]
• Organisation for Economic Co-operation and Development (OECD) (2020). Researchers (indicator). https://data.oecd.org/rd/researchers.htm
• Overview—Health Information & Libraries Journal . (2020). Wiley Online Library. https://doi.org/10.1111/(ISSN)1471-1842
• Perrote A, Ranganath R, Hirsch JS, Blei D, Elhadad N. Risk prediction for chronic kidney disease progression using heterogeneous electronic health record data and time series analysis. Journal of the American Medical Informatics Association. 2015; 22 (4):872–880. doi: 10.1093/jamia/ocv024. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Petersen AM, Penner O. Inequality and cumulative advantage in science careers: A case study of high-impact journals. EPJ Data Science. 2014 doi: 10.1140/epjds/s13688-014-0024-y. [ CrossRef ] [ Google Scholar ]
• Price DJ. Little science, big science. NewYork: Columbia University Press; 1963. [ Google Scholar ]
• Price DJ. Science since Babylon. Enlarged. London: Yale University Press; 1974. [ Google Scholar ]
• Priem J. Beyond the paper. Nature. 2013; 495 :437–440. doi: 10.1038/495437a. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rousseau R. Double exponential models for first-citation processes. Scientometrics. 1994; 30 (1):213–227. doi: 10.1007/BF02017224. [ CrossRef ] [ Google Scholar ]
• Saboo AR, Kumar V, Park I. Using big data to model time-varying effects for marketing resource (re)allocation. MIS Quarterly. 2016; 40 (4):911–939. doi: 10.25300/MISQ/2016/40.4.06. [ CrossRef ] [ Google Scholar ]
• Salgotra R, Gandomi M, Gandomi AH. Time series analysis and forecast of the COVID-19 pandemic in india using genetic programming. Chaos, Solitons & Fractals. 2020; 138 :109945. doi: 10.1016/j.chaos.2020.109945. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Shumway RH, Stoffer DS. Time series analysis and its applications with R examples. 2. NewYork: Springer; 2006. [ Google Scholar ]
• Siegel, K. (2019, October 10). Can we predict which biologists are likely to win a Nobel Prize? The Startup . https://medium.com/swlh/can-we-predict-which-biologists-are-likely-to-win-a-nobel-prize-6a748e40e207
• Simkin M, Roychowdhury V. Do you sincerely want to be cited? Or Read before you cite. In: Cronin B, Sugimoto CR, editors. Scholarly Metrics Under the Microscope From Citation Analysis to Academic Auditing. NewJersey: Information Today; 2015. [ Google Scholar ]
• Simpao AF, Ahumada LM, Desai BR, Bonafide CP, Gálvez JA, Rehman MA, Jawad AF, Palma KL, Shelov ED. Optimization of drug-drug interaction alert rules in a pediatric hospital’s electronic health record system using a visual analytics dashboard. Journal of the American Medical Informatics Association. 2015; 22 (2):361–369. doi: 10.1136/amiajnl-2013-002538. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Small H. Tracking and predicting growth areas in science. Scientometrics. 2006; 68 (3):595–610. doi: 10.1007/s11192-006-0132. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tahamtan I, Afshar AS, Ahamdzadeh K. Factors affecting number of citations: A comprehensive review of the literature. Scientometrics. 2016; 107 :1195–1225. doi: 10.1007/s11192-016-1889-2. [ CrossRef ] [ Google Scholar ]
• The World Bank (2018). Research and development expenditure (% of GDP). UNESCO Institute for Statistics . https://data.worldbank.org/indicator/GB.XPD.RSDV.GD.ZS?view=chart
• Tonta Y. Does monetary support increase the number of scientific papers? An interrupted time series analysis. Journal of Data and Information Science. 2018; 3 (1):19–39. doi: 10.2478/jdis-2018-0002. [ CrossRef ] [ Google Scholar ]
• Tripathy P, Tripathy PK. Fundamentals of research: A dissective view. Germany: Anchor Academic Publishing; 2017. [ Google Scholar ]
• Tseng Y-H, Tsay M-Y. Journal clustering of library and information science for subfield delineation using the bibliometric analysis toolkit: CATAR. Scientometrics. 2013; 95 :503–528. doi: 10.1007/s11192-013-0964-1. [ CrossRef ] [ Google Scholar ]
• ULRICHSWEB Global Serials Directory. (2020). https://ulrichsweb.serialssolutions.com
• Waldrop, M. M. (2008, May). Science 2.0—Is open access science the future? https://www.scientificamerican.com/article/science-2-point-0/
• Walters GD. Predicting subsequent citations to articles published in twelve crime-psychology journals: Author impact versus journal impact. Scientometrics. 2006; 69 (3):499–510. doi: 10.1007/s11192-006-0166-1. [ CrossRef ] [ Google Scholar ]
• White HD, McCain KW. Visualizing a discipline: An author co-citation analysis of information science, 1972–1995. Journal of the American Society for Information Science. 1998; 49 (4):327–355. doi: 10.1002/(SICI)1097-4571. [ CrossRef ] [ Google Scholar ]
• Wu QW, Zhang C, Hong Q, Chen L. Topic evolution based on LDA and HMM and its application in stem cell research. Journal of Information Science. 2014; 40 (5):611–620. doi: 10.1177/0165551514540565. [ CrossRef ] [ Google Scholar ]
• Xie Y. ‘Undemocracy’: Inequalities in science. Science. 2014; 344 (6186):809–810. doi: 10.1126/science.1252743. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ye FY, Rousseau R. The power law model and total career h-index sequences. Journal of Informetrics. 2008; 2 (4):288–297. doi: 10.1016/j.joi.2008.09.002. [ CrossRef ] [ Google Scholar ]
• You H, Li M, Hipel KW, Jiang J, Ge B, Duan H. Development trend forecasting for coherent light generator technology based on patent citation network analysis. Scientometrics. 2017; 111 :297–315. doi: 10.1007/s11192-017-2252-y. [ CrossRef ] [ Google Scholar ]
• Zeisset PT. Disseminating economic census data. Government Information Quarterly. 1998; 15 (3):303–318. doi: 10.1016/S0740-624X(98)90005-3. [ CrossRef ] [ Google Scholar ]
• Zhang T. Will the increase in publication volumes “dilute” prestigious journals’ impact factors? A trend analysis of the FT50 journals. Scientometrics. 2020 doi: 10.1007/s11192-020-03736-7. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Zhang Y, Shah D, Foley J, Abhishek A, Lukito J, Suk J, Kim SJ, Sun Z, Pevehouse J, Garlough C. Whose lives matter? Mass shootings and social media discourses of sympathy and policy, 2012–2014. Journal of Computer-Mediated Communication. 2019; 24 (4):182–202. doi: 10.1093/jcmc/zmz009. [ CrossRef ] [ Google Scholar ]

A lexical and syntactic study of research article titles in Library Science and Scientometrics

• Published: 16 May 2021
• Volume 126 , pages 6041–6058, ( 2021 )

Cite this article

• Junli Diao   ORCID: orcid.org/0000-0002-9603-8278 1  

828 Accesses

5 Citations

Explore all metrics

Title of a research article is an abstract of the abstract. Titles play a decisive role in convincing readers at first sight whether articles are worth reading or not. Not only do research article titles show how carefully words are chosen by authors, but also reflect disciplinary differences in terms of title words and structure between hard sciences and soft sciences. This study examined the lexical density and syntactic structure of 690 research article titles chosen from five Library Science and Scientometrics journals, aiming to reveal disciplinary differences. The result suggested both Library Science and Scientometrics have almost the same title length and the prevalent usage of Nominal Phrase (NP) to govern the title structure. The result also stated some disciplinary differences: Library Science demonstrates more punctuation complexity, particularly a greater frequency in using colons; but Scientometrics shows more involvement of words related to research methods, which is an indicator to papers’ scientific value, and more usage of declarative Full Sentence (FS) structure, which were mostly discovered in the research articles in hard sciences.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

A diachronic comparative study of research article titles in linguistics and literature journals

Titles in research articles and doctoral dissertations: cross-disciplinary and cross-generic perspectives

Scientometric analysis of social science and science disciplines in a developing nation: a case study of Pakistan in the last decade

Ávila-Argüelles, R., Calvo, H., Gelbukh, A., & Godoy-Calderón, S. (2010). Assigning Library of Congress Classification codes to books based only on their titles. Informatica, 34 , 77–84

Google Scholar  

Adams, W. M. (1967). Relationship of keywords in titles to references cited. American Documentation, 18 , 26–32

Article   Google Scholar  

Anthony, L. (2001). Characteristic features of research article titles in computer science. IEEE Transactions on Professional Communication, 44 , 187–194. https://doi.org/10.1109/47.946464

Appiah, K. R., Ankomah, C., Osei, H. Y., & Hattoh-Ahiaduvor, T. (2019). Structural organisation of research article titles: A comparative study of titles of business, gynaecology and law. Advances in Language and Literary Studies, 10 (3), 145–154. https://doi.org/10.7575/aiac.alls.v.10n.3p.145

Archibald, A. J. B. (2017). A linguistic analysis of conference titles in Applied Linguistics. International Journal of Foreign Language Teaching & Research, 5 (18), 11–25

Arsenault, C., & Ménard, E. (2011). Searching titles with initial articles in library catalogs: A case study and search behavior analysis. Library Resources & Technical Services, 51 , 190–203. https://doi.org/10.5860/lrts.51n3.190

Baicchi, A. (2003). Relational complexity of titles and texts: A semiotic taxonomy. In L. Merlini Barbaresi (Ed.), Complexity in language and text. (pp. 319–341). Edizione Plus-Universidad de Pisa.

Berkenkotter, C., & Huckin, T. N. (1995). Genre knowledge in disciplinary communication: Cognition, culture, power . Lawrence Erlbaum Associates.

Busch-Lauer, I. A. (2000). Titles of English and German research papers in Medicine and Linguistics. Analysing Professional Genres, 74 , 77–94

Buxton, A. B., & Meadows, A. J. (1977). The variation in the information content of titles of research papers with time and discipline. Journal of Documentation, 33 , 46–52. https://doi.org/10.1108/eb026633

Casson, L. (2001). Libraries in the ancient world . Yale University Press.

Cheng, S. W., Kuo, C.-W., & Kuo, C.-H. (2012). Research article titles in Applied Linguistics. Journal of Academic Language and Learning, 6 (1), A1–A14

MathSciNet   Google Scholar  

Diers, D., & Downs, F. S. (1994). Colonizing: A measurement of the development of a profession. Nursing Research, 43 , 316–318

Dillon, J. T. (1981). The emergence of the colon: An empirical correlate of scholarship. American Psychologist, 36 , 879–884. https://doi.org/10.1037/0003-066X.36.8.879

Dillon, J. T. (1982). In pursuit of the colon: A century of scholarly progress: 1880–1980. The Journal of Higher Education, 53 , 93–99. https://doi.org/10.1080/00221546.1982.11780427

Fortanet, I., Coll, J. F., Palmer, J. C., & Posteguillo, S. (1997). The writing of titles in academic research articles. In R. M. Chamorro & A. R. Navarrete (Eds.), Lenguas aplicadas a las ciencias y la tecnología. Aproximaciones. (pp. 155–158). Universidada de Extremadura, Servicio de Publicaciones.

Fortanet, I., Posteguillo, S., Coll, J. F., & Palmer, J. C. (1998). Linguistic analysis of research articles: Disciplinary variations. In I. Vazquez & I. Camilleu (Eds.), Perspectivas praguietices en linguistica aplicada, zaragoza. (pp. 443–447). Anubar Ediciones.

Gómez, I. F., Gómez, S. P., García, J. F. C., & Silveira, J. C. P. (1998). Linguistic analysis of research article titles: Disciplinary variations. In I. V. Orta & I. G. Galve (Eds.), Perspectivas pragmáticas en lingüística aplicada. (pp. 443–448). Anubar Ediciones.

Gesuato, S. (2008). Encoding of information in titles: Academic practices across four genres in linguistics. In C. Taylor (Ed.), Ecolingua: The role of e-corpora in translation and language learning. (pp. 127–157). EUT.

Goodman, R. A., Thacker, S. B., & Siegel, P. Z. (2001). What’s in a title? A descriptive study of article titles in peer-reviewed medical journals. Science Editor, 24 , 75–78

Haggan, M. (2004). Research paper titles in literature, linguistics and science: Dimensions of attraction. Journal of Pragmatics, 36 , 293–317. https://doi.org/10.1016/S0378-2166(03)00090-0

Hartley, J. (2007). Planning that title: Practices and preferences for titles with colons in academic articles. Library & Information Science Research, 29 , 553–568. https://doi.org/10.1016/j.lisr.2007.05.002

Jahoda, G., & Stursa, M. L. (1969). A comparison of a keyword from title index with a single access point per document alphabetic subject index. American Documentation, 20 , 377–380. https://doi.org/10.1002/asi.4630200422

Lewison, G., & Hartley, J. (2005). What’s in a title? Numbers of words and the presence of colons. Scientometrics, 63 , 341–356. https://doi.org/10.1007/s11192-005-0216-0

Maiti, D. C., & Dutta, B. (2013). Comparative study between words in titles and keywords of some articles on knowledge organisation. DESIDOC Journal of Library & Information Technology, 33 , 498–508

Michelson, G. (1994). Use of colons in titles and journal status in industrial relations journals. Psychological Reports, 74 , 657–658. https://doi.org/10.2466/pr0.1994.74.2.657

Milojević, S. (2017). The length and semantic structure of article titles—Evolving disciplinary practices and correlations with impact. Frontiers in Research Metrics and Analytics, 2 (2), 1. https://doi.org/10.3389/frma.2017.00002

Milojević, S., Sugimoto, C. R., Yan, E., & Ding, Y. (2011). The cognitive structure of library and information science: Analysis of article title words. Journal of the American Society for Information Science and Technology, 62 , 1933–1953. https://doi.org/10.1002/asi.21602

Moattarian, A., & Alibabaee, A. (2015). Syntactic structures in research article titles from three different disciplines: Applied linguistics, Civil Engineering, and Dentistry. Journal of Teaching Language Skills, 34 , 27–50

Morales, O. A., Perdomo, B., Cassany, D., Tovar, R. M., & Izarra, É. (2020). Linguistic structures and functions of thesis and dissertation titles in Dentistry. Lebende Sprachen, 65 , 49–73. https://doi.org/10.1515/les-2020-0003

Nagano, R. L. (2015). Research article titles and disciplinary conventions: A corpus study of eight disciplines. Journal of Academic Writing, 5 (1), 133–144. https://doi.org/10.18552/joaw.v5i1.168

O’Connor, J. (1964). Correlation of indexing headings and title words in three medical indexing systems. American Documentation, 15 , 96–104. https://doi.org/10.1002/asi.5090150207

Perry, J. A. (1985). The Dillion hypothesis of titular colonicity: An empirical test from the ecological sciences. Journal of the American Society for Information Science, 36 , 251–258. https://doi.org/10.1002/asi.4630360405

Sahragard, R., & Meihami, H. (2016). A diachronic study on the information provided by the research titles of Applied Linguistics journals. Scientometrics, 108 , 1315–1331. https://doi.org/10.1007/s11192-016-2049-4

Salager-Meyer, F., Ariza, M. A. A., & Briceño, M. L. (2013). Titling and authorship practices in medical case reports: A diachronic study. Communication & Medicine, 10 (1), 63–80. https://doi.org/10.1371/journal.pone.0197775

Soler, V. (2007). Writing titles in science: An exploratory study. English for Specific Purposes, 26 , 90–102. https://doi.org/10.1016/j.esp.2006.08.001

Soler, V. (2011). Comparative and contrastive observations on scientific titles written in English and Spanish. English for Specific Purposes, 30 , 124–137. https://doi.org/10.1016/j.esp.2010.09.002

Swales, J. (1990). Genre analysis: English in academic and research setting . Cambridge: Cambridge University Press.

Slougui, D. (2018). Dissertation titles in EFL and UK-based contexts: How much do they differ? ASp, 74 , 135–161. https://doi.org/10.4000/asp.5466

Trosborg, A. (2000). Introduction. In A. Trosborg & J. Benjamins (Eds.), Analysing professional genres. (pp. vii–xvi). ProQuest Ebook Central.

Chapter   Google Scholar  

Wang, Y., & Bai, Y. (2007). A corpus-based syntactic study of medical research article titles. System, 35 , 388–399. https://doi.org/10.1016/j.system.2007.01.005

Xie, S. (2020). English research article titles: Cultural and disciplinary perspectives. SAGE Open, 10 (2), 1–12. https://doi.org/10.1177/2158244020933614

Yakhontova, T. (2002). Titles of conference presentation abstracts: A cross-cultural perspective . Retrieved on October 9, 2020 from, https://www.academia.edu/36296603/Titles_of_conference_presentations_abstracts_A_cross_cultural_perspective .

Yitzhaki, M. (1997). Variation in informativity of titles of research papers in selected humanities journals: A comparative study. Scientometrics, 38 , 219–229. https://doi.org/10.1007/BF02457410

Ziebland, S., & Pope, C. (1995, September/October) Use of the colon in titles of British Medical Sociology Conference papers, 1970 to 1993. Annals of Improbable Research , 7–9.

Download references

Author information

Authors and affiliations.

York College Library, The City University of New York, New York, USA

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Junli Diao .

Rights and permissions

Reprints and permissions

About this article

Diao, J. A lexical and syntactic study of research article titles in Library Science and Scientometrics. Scientometrics 126 , 6041–6058 (2021). https://doi.org/10.1007/s11192-021-04018-6

Download citation

Received : 27 December 2020

Accepted : 26 April 2021

Published : 16 May 2021

Issue Date : July 2021

DOI : https://doi.org/10.1007/s11192-021-04018-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Library Science
• Scientometrics
• Lexical density
• Syntactic structure
• Find a journal
• Publish with us
• Track your research

newjerseystatemuseum.org |

Research paper topics in library and information science

A systematic approach is best when undertaking research in the library and information science. Not only should you have an in-depth knowledge of major themes in the area, but you should also be aware of current research methods and topics of influence, such as library systems, cooperation between libraries, and the flow of information between libraries.

Finding a good research paper topic can greatly depend upon your interests and what you took away from your coursework. Paying attention in classes and taking adequate notes makes it easier to assimilate that knowledge into a coherent research paper topic. Take a look at the following research paper topics for some ideas:

• A critical analysis of student attitudes towards cataloguing and classification in college campus libraries
• The Impact of Public Libraries at the state level
• The implementation of information and communication technology in academic libraries in Brazil
• Evaluating the effect of feminization and professionalization on librarianship
• The challenges involved in running private libraries in Nigeria
• Defining comparative and international library and information science
• An assessment of international cultural exchange through libraries
• The role of international librarianship in promoting freedom of information and expression
• International issues faced by librarians and information science professionals with regard to the knowledge society
• Exploring the relationship between government schools and public libraries in the context of South Asia
• The importance of resource-sharing in an international library network: bridging gaps using modern technology
• Tackling indigenous knowledge by adopting innovative tools and strategies
• The influence of library aid in developing countries during globalization
• A critical comparison of American librarianship and information science research in European countries
• Learnings from major book acquisitions in American academic libraries
• The expanding purview of American ideas in German public libraries
• The British Council and its critical role in building bridges across the developing world

Browsing through sample topics in library and information science can help you brainstorm your own ideas more effectively. Take the time to scan such resources and choose a topic that you can convincingly discuss and analyze. A good source for potential research paper topics and paper help is mypaperwriter.com , also papers written by past students as well as reputed works in the field.

Copyright ©2017 - newjerseystatemuseum.org

• USC Libraries
• Research Guides

Organizing Your Social Sciences Research Paper

• Reading Research Effectively
• Purpose of Guide
• Design Flaws to Avoid
• Independent and Dependent Variables
• Glossary of Research Terms
• Narrowing a Topic Idea
• Broadening a Topic Idea
• Extending the Timeliness of a Topic Idea
• Academic Writing Style
• Applying Critical Thinking
• Choosing a Title
• Making an Outline
• Paragraph Development
• Research Process Video Series
• Executive Summary
• The C.A.R.S. Model
• Background Information
• The Research Problem/Question
• Theoretical Framework
• Citation Tracking
• Content Alert Services
• Evaluating Sources
• Primary Sources
• Secondary Sources
• Tiertiary Sources
• Scholarly vs. Popular Publications
• Qualitative Methods
• Quantitative Methods
• Insiderness
• Using Non-Textual Elements
• Limitations of the Study
• Common Grammar Mistakes
• Writing Concisely
• Avoiding Plagiarism
• Footnotes or Endnotes?
• Further Readings
• Generative AI and Writing
• USC Libraries Tutorials and Other Guides
• Bibliography

Reading a Scholarly Article or Research Paper

Identifying a research problem to investigate requires a preliminary search for and critical review of the literature in order to gain an understanding about how scholars have examined a topic. Scholars rarely structure research studies in a way that can be followed like a story; they are complex and detail-intensive and often written in a descriptive and conclusive narrative form. However, in the social and behavioral sciences, journal articles and stand-alone research reports are generally organized in a consistent format that makes it easier to compare and contrast studies and interpret their findings.

General Reading Strategies

W hen you first read an article or research paper, focus on asking specific questions about each section. This strategy can help with overall comprehension and with understanding how the content relates [or does not relate] to the problem you want to investigate. As you review more and more studies, the process of understanding and critically evaluating the research will become easier because the content of what you review will begin to coalescence around common themes and patterns of analysis. Below are recommendations on how to read each section of a research paper effectively. Note that the sections to read are out of order from how you will find them organized in a journal article or research paper.

1.  Abstract

The abstract summarizes the background, methods, results, discussion, and conclusions of a scholarly article or research paper. Use the abstract to filter out sources that may have appeared useful when you began searching for information but, in reality, are not relevant. Questions to consider when reading the abstract are:

• Is this study related to my question or area of research?
• What is this study about and why is it being done ?
• What is the working hypothesis or underlying thesis?
• What is the primary finding of the study?
• Are there words or terminology that I can use to either narrow or broaden the parameters of my search for more information?

2.  Introduction

If, after reading the abstract, you believe the paper may be useful, focus on examining the research problem and identifying the questions the author is trying to address. This information is usually located within the first few paragraphs of the introduction or in the concluding paragraph. Look for information about how and in what way this relates to what you are investigating. In addition to the research problem, the introduction should provide the main argument and theoretical framework of the study and, in the last paragraphs of the introduction, describe what the author(s) intend to accomplish. Questions to consider when reading the introduction include:

• What is this study trying to prove or disprove?
• What is the author(s) trying to test or demonstrate?
• What do we already know about this topic and what gaps does this study try to fill or contribute a new understanding to the research problem?
• Why should I care about what is being investigated?
• Will this study tell me anything new related to the research problem I am investigating?

3.  Literature Review

The literature review describes and critically evaluates what is already known about a topic. Read the literature review to obtain a big picture perspective about how the topic has been studied and to begin the process of seeing where your potential study fits within the domain of prior research. Questions to consider when reading the literature review include:

• W hat other research has been conducted about this topic and what are the main themes that have emerged?
• What does prior research reveal about what is already known about the topic and what remains to be discovered?
• What have been the most important past findings about the research problem?
• How has prior research led the author(s) to conduct this particular study?
• Is there any prior research that is unique or groundbreaking?
• Are there any studies I could use as a model for designing and organizing my own study?

4.  Discussion/Conclusion

The discussion and conclusion are usually the last two sections of text in a scholarly article or research report. They reveal how the author(s) interpreted the findings of their research and presented recommendations or courses of action based on those findings. Often in the conclusion, the author(s) highlight recommendations for further research that can be used to develop your own study. Questions to consider when reading the discussion and conclusion sections include:

• What is the overall meaning of the study and why is this important? [i.e., how have the author(s) addressed the " So What? " question].
• What do you find to be the most important ways that the findings have been interpreted?
• What are the weaknesses in their argument?
• Do you believe conclusions about the significance of the study and its findings are valid?
• What limitations of the study do the author(s) describe and how might this help formulate my own research?
• Does the conclusion contain any recommendations for future research?

5.  Methods/Methodology

The methods section describes the materials, techniques, and procedures for gathering information used to examine the research problem. If what you have read so far closely supports your understanding of the topic, then move on to examining how the author(s) gathered information during the research process. Questions to consider when reading the methods section include:

• Did the study use qualitative [based on interviews, observations, content analysis], quantitative [based on statistical analysis], or a mixed-methods approach to examining the research problem?
• What was the type of information or data used?
• Could this method of analysis be repeated and can I adopt the same approach?
• Is enough information available to repeat the study or should new data be found to expand or improve understanding of the research problem?

6.  Results

After reading the above sections, you should have a clear understanding of the general findings of the study. Therefore, read the results section to identify how key findings were discussed in relation to the research problem. If any non-textual elements [e.g., graphs, charts, tables, etc.] are confusing, focus on the explanations about them in the text. Questions to consider when reading the results section include:

• W hat did the author(s) find and how did they find it?
• Does the author(s) highlight any findings as most significant?
• Are the results presented in a factual and unbiased way?
• Does the analysis of results in the discussion section agree with how the results are presented?
• Is all the data present and did the author(s) adequately address gaps?
• What conclusions do you formulate from this data and does it match with the author's conclusions?

7.  References

The references list the sources used by the author(s) to document what prior research and information was used when conducting the study. After reviewing the article or research paper, use the references to identify additional sources of information on the topic and to examine critically how these sources supported the overall research agenda. Questions to consider when reading the references include:

• Do the sources cited by the author(s) reflect a diversity of disciplinary viewpoints, i.e., are the sources all from a particular field of study or do the sources reflect multiple areas of study?
• Are there any unique or interesting sources that could be incorporated into my study?
• What other authors are respected in this field, i.e., who has multiple works cited or is cited most often by others?
• What other research should I review to clarify any remaining issues or that I need more information about?

NOTE:   A final strategy in reviewing research is to copy and paste the title of the source [journal article, book, research report] into Google Scholar . If it appears, look for a "cited by" reference followed by a hyperlinked number under the record [e.g., Cited by 45]. This number indicates how many times the study has been subsequently cited in other, more recently published works. This strategy, known as citation tracking, can be an effective means of expanding your review of pertinent literature based on a study you have found useful and how scholars have cited it. The same strategies described above can be applied to reading articles you find in the list of cited by references.

Reading Tip

Specific Reading Strategies

Effectively reading scholarly research is an acquired skill that involves attention to detail and an ability to comprehend complex ideas, data, and theoretical concepts in a way that applies logically to the research problem you are investigating. Here are some specific reading strategies to consider.

As You are Reading

• Focus on information that is most relevant to the research problem; skim over the other parts.
• As noted above, read content out of order! This isn't a novel; you want to start with the spoiler to quickly assess the relevance of the study.
• Think critically about what you read and seek to build your own arguments; not everything may be entirely valid, examined effectively, or thoroughly investigated.
• Look up the definitions of unfamiliar words, concepts, or terminology. A good scholarly source is Credo Reference .

Taking notes as you read will save time when you go back to examine your sources. Here are some suggestions:

• Mark or highlight important text as you read [e.g., you can use the highlight text  feature in a PDF document]
• Take notes in the margins [e.g., Adobe Reader offers pop-up sticky notes].
• Highlight important quotations; consider using different highlighting colors to differentiate between quotes and other types of important text.
• Summarize key points about the study at the end of the paper. To save time, these can be in the form of a concise bulleted list of statements [e.g., intro provides useful historical background; lit review has important sources; good conclusions].

Write down thoughts that come to mind that may help clarify your understanding of the research problem. Here are some examples of questions to ask yourself:

• Do I understand all of the terminology and key concepts?
• Do I understand the parts of this study most relevant to my topic?
• What specific problem does the research address and why is it important?
• Are there any issues or perspectives the author(s) did not consider?
• Do I have any reason to question the validity or reliability of this research?
• How do the findings relate to my research interests and to other works which I have read?

Adapted from text originally created by Holly Burt, Behavioral Sciences Librarian, USC Libraries, April 2018.

Another Reading Tip

When is it Important to Read the Entire Article or Research Paper

Laubepin argues, "Very few articles in a field are so important that every word needs to be read carefully." * However, this implies that some studies are worth reading carefully if they directly relate to understanding the research problem. As arduous as it may seem, there are valid reasons for reading a study from beginning to end. Here are some examples:

• Studies Published Very Recently .  The author(s) of a recent, well written study will provide a survey of the most important or impactful prior research in the literature review section. This can establish an understanding of how scholars in the past addressed the research problem. In addition, the most recently published sources will highlight what is known and what gaps in understanding currently exist about a topic, usually in the form of the need for further research in the conclusion .
• Surveys of the Research Problem .  Some papers provide a comprehensive analytical overview of the research problem. Reading this type of study can help you understand underlying issues and discover why scholars have chosen to investigate the topic. This is particularly important if the study was published recently because the author(s) should cite all or most of the important prior research on the topic. Note that, if it is a long-standing problem, there may be studies that specifically review the literature to identify gaps that remain. These studies often include the word "review" in their title [e.g., Hügel, Stephan, and Anna R. Davies. "Public Participation, Engagement, and Climate Change Adaptation: A Review of the Research Literature." Wiley Interdisciplinary Reviews: Climate Change 11 (July-August 2020): https://doi.org/10.1002/ wcc.645].
• Highly Cited .  If you keep coming across the same citation to a study while you are reviewing the literature, this implies it was foundational in establishing an understanding of the research problem or the study had a significant impact within the literature [either positive or negative]. Carefully reading a highly cited source can help you understand how the topic emerged and how it motivated scholars to further investigate the problem. It also could be a study you need to cite as foundational in your own paper to demonstrate to the reader that you understand the roots of the problem.
• Historical Overview .  Knowing the historical background of a research problem may not be the focus of your analysis. Nevertheless, carefully reading a study that provides a thorough description and analysis of the history behind an event, issue, or phenomenon can add important context to understanding the topic and what aspect of the problem you may want to examine further.
• Innovative Methodological Design .  Some studies are significant and should be read in their entirety because the author(s) designed a unique or innovative approach to researching the problem. This may justify reading the entire study because it can motivate you to think creatively about also pursuing an alternative or non-traditional approach to examining your topic of interest. These types of studies are generally easy to identify because they are often cited in others works because of their unique approach to examining the research problem.
• Cross-disciplinary Approach .  R eviewing studies produced outside of your discipline is an essential component of investigating research problems in the social and behavioral sciences. Consider reading a study that was conducted by author(s) based in a different discipline [e.g., an anthropologist studying political cultures; a study of hiring practices in companies published in a sociology journal]. This approach can generate a new understanding or a unique perspective about the topic . If you are not sure how to search for studies published in a discipline outside of your major or of the course you are taking, contact a librarian for assistance.

* Laubepin, Frederique. How to Read (and Understand) a Social Science Journal Article . Inter-University Consortium for Political and Social Research (ISPSR), 2013

Shon, Phillip Chong Ho. How to Read Journal Articles in the Social Sciences: A Very Practical Guide for Students . 2nd edition. Thousand Oaks, CA: Sage, 2015; Lockhart, Tara, and Mary Soliday. "The Critical Place of Reading in Writing Transfer (and Beyond): A Report of Student Experiences." Pedagogy 16 (2016): 23-37; Maguire, Moira, Ann Everitt Reynolds, and Brid Delahunt. "Reading to Be: The Role of Academic Reading in Emergent Academic and Professional Student Identities." Journal of University Teaching and Learning Practice 17 (2020): 5-12.

• << Previous: 1. Choosing a Research Problem
• Next: Narrowing a Topic Idea >>
• Last Updated: Aug 27, 2024 1:14 PM
• URL: https://libguides.usc.edu/writingguide