Keywords

1 Overview

The use of source material is essential in academic writing. By integrating sources in the body of their text, writers contextualize their ideas, demonstrate the breadth and depth of their critical engagement with the available literature, and acknowledge the work of other authors. They underpin their own ideas, arguments, and opinions with evidence (Cumming et al., 2016). Sources are typically included in a text by in-text citations in the form of summaries, paraphrases or quotations that are linked to the list of works cited at the end of the text (i.e. reference list or bibliography). Referencing entails thus acknowledging a source in the text and in the bibliography. A reference indicates a scholarly source by providing a standard set of information (i.e. citation information) that allows readers to easily identify, search, and retrieve the source (Kali, 2016). In order to ensure consistency in the format and layout of citations, it is common for writers to use a specific citation style.

In the past, source material often was manually documented and organized on small-sized index cards and stored in boxes. Each card contained information related to a source, such as summaries and paraphrases of important information, quotations, personal comments, or author data and bibliographic information. Location information, when included, allowed for a quick return to a specific passage in the source material. Keywords made the documented information be reused. This manual management of references was tedious and time consuming (Fenner et al., 2014).

Reference management systems, also referred to as reference/citation managers, or as bibliographic management systems or software, allow for the computer-assisted management of sources. Today, they enable the personal collection, organization, and use of citation information and support the management, analysis, and further utilization of the corresponding source material (i.e. full text, e.g., Murphree et al., 2018).

Originally introduced in the 1980s to reduce the effort of editing citation information, early desktop applications for reference management (e.g., Endnote, ProCite, Reference Manager) mainly provided functionalities for collecting references and integrating citation information into one's own text in a formally correct manner (Kali, 2016; Murphree et al., 2018; Steele, 2008). Writers had to manually add the citation information as individual entries into the respective digital reference library, but could enrich them with personal notes. Full texts could usually not be saved or accessed directly.

In the 2000s, the desktop applications for reference management evolved into web-based systems (e.g. Refworks, online versions of desktop applications, Zotero), making it possible to manage references online. In addition, it was now possible to automatically import citation information and full texts from online bibliographic databases such as Web of Science, Scopus, PubMed, or Google Scholar, but also directly from web pages.

The most recent development since the 2010s has been in the direction of bibliographic social bookmarking (Fenner et al., 2014). Online social bookmarking tools such as CiteULike and Bibsonomy, as well as reference management systems such as Mendeley or ReadCube focus on the public sharing of references. This functionality helps writers capture, manage, and reuse source material, as well as complete and accurate citation information (Giglia, 2010). It also enables, for example, the generation of usage statistics as a novel means of measuring scientific impact (Chen et al., 2018).

2 Core Idea of the Technology and Functional Specifications

Many researchers agree that the quality of writers’ engagement with the source material is an important predictor of the quality of a scientific text (e.g., Plakans & Gebril, 2013; Spivey & King, 1989). Thus, one’s text product is built through the purposeful interplay of source comprehension and text production processes (e.g., Cumming et al., 2016; Jakobs, 2003; Parodi, 2006). Reading and writing processes are blended in which writers actively collect, select, analyze, interpret, organize, and connect information of different sources (Drake, 2013; Parodi, 2006; Spivey, 1990). This requires writers to deal with a variety of source material and to organize this information in a consistent way for their text (Jakobs, 2003). Yet this also makes it necessary to clearly identify where ideas from source material have been used in one's own text (Jakobs, 2003). Not only is this a way to avoid plagiarism, it also allows readers to distinguish between writer's own words and ideas and those of others. Reference management systems can assist in many of these source material engagement activities (Francavilla, 2018). They help capture, store, organize, use, cite, annotate and share source material for writing.

Unfortunately, reference management systems are not explicitly designed to support writing, but to facilitate the organization and management of bibliographies (Francavilla, 2018). Therefore, some functionalities are not completely aligned with a writing process (Vaidhyanathan et al., 2012). Moreover, many of the functionalities that reference management systems offer to support writing are not visible to the writer at first glance. Often, an in-depth knowledge of the respective reference management system is necessary to profitably use these functionalities for one's own writing process. Familiarization with a reference management system requires a considerable amount of time and effort that writers may be able to invest in one, but usually not in several reference management systems. As a result, many of the system functionalities that can support the writing process beyond the formally correct preparation of citation information are not or only very rarely used (Melles & Unsworth, 2015).

Nowadays, reference management systems are mainly organizational tools (Perkel, 2020). They are typically desktop applications with an associated web-based interface that allow writers to remotely access their individual libraries (i.e., self-compiled, self-structured, and annotated collections of source material). Further, these reference management systems often include browser plug-ins, which facilitate the simultaneous import of citation information and the corresponding full text from scientific databases, journal web pages, or other online sources. Most reference management systems also provide mobile apps that allow writers to add, read, and annotate sources from smartphones or tablets.

The organizational functionalities of reference management systems facilitate the analysis and elaboration of source material in earlier stages of writing projects as well as an alternating work on writers’ text produced so far, their annotations, and the sources during actual text production. Most reference management systems allow writers to organize their individual source material collection into folders. Tags can also be used in most reference management systems to organize the source material. A tag is a kind of individually generated label or keyword that the writer can add to a reference entry to classify and remember it (Giglia, 2010). With most reference management systems, it is also possible to add notes to each source. All sources can be searched by author, keywords, text and notes. PDF viewers—in most cases built-in—allow the writer to access and read the full text of a source. After accessing, for example by double-clicking, the writer can annotate the full text and highlight important passages using the PDF viewer tools. In other words, the full texts are typically not processed via the reference management system, but only accessed. The annotations are saved in such a way that they are available for subsequent work on the full text even without access via the reference management system. Some reference management systems (e.g., Endnote, Zotero) can even import further sources or missing meta-data and full text via included search functions.

The editing and correct formatting of citation information (i.e. in-text citations and the corresponding reference list) is supported in most reference management systems through integration with word processing software, for example, in the form of add-ins. These allow the writer to insert in-text citations (i.e. the citation information acknowledging the source of quotations, paraphrases, or summaries) into documents as they are written. Reference lists are automatically created in the required citation style and reformatted on the fly when the citation style changes. Finally, reference management systems typically allow writers to share their individual libraries with colleagues so that it can be used in co-writing situations or as a shared knowledge base. In this way, not only collaborative writing but also collaborative thinking and exchange of ideas is supported.

3 Main Products

There exist more than 30 different reference management systems at present. A continuously updated overview can be found in the Wikipedia article on comparison of reference management software: https://en.wikipedia.org/wiki/Comparison_of_reference_management_software. This overview contains a description of the basic functional features of the respective system as well as information on whether a reference management system is regularly updated. Using a reference management system without regular releases is not recommended.

By now, all common reference management systems provide basic functionalities for capturing and storing source material, for the insertion of citations into one’s text, as well as for web-based reference management and social bookmarking. Nevertheless, the various systems are developing so quickly and in part unpredictably that an overview of the system features could look quite different in some time.

In the following, we will discuss unique features and/or functionalities of five important reference management systems, illustrating the diversity of those on the market. These five systems are frequently used, and usually referred in literature and research (e.g., Lorenzetti & Ghali, 2013; McMinn, 2011).

By now, all five systems offer the basic functionalities described above. These include functionalities for

  • collecting references with full texts and organizing them into folders and/or (sub)collections,

  • tagging reference entries,

  • annotating and highlighting full text via (built-in) PDF viewers,

  • citing references in different citation styles via (a) add-ins for word processing software or (b) (automatically) creating and updating BIB files for (La)Tex,

  • synchronizing references between desktop app and mobile version as well as between different computers, and

  • sharing references with colleagues.

3.1 Citavi

Citavi is a good example that the border between reference management systems, word processors, outliners, and idea management is fluent. Citavi is a proprietary software of the developer Swiss Academic Software. It was developed in 1994 as a project named LiteRat. Since 2006, it is known as Citavi. In February 2021, Citavi was purchased by QSR International. Citavi is compatible with MS Windows and the interface is available in different languages. Citavi offers an add-in for MS Word to insert references from a Citavi project into a document in a specific citation style. It also helps creating manuscripts with the TeX typesetting system.

In addition to the typical basic functionalities, Citavi offers two unique tools – the knowledge organizer and the task planner. The knowledge organizer supports writers’ engagement with the collected sources and working on one's own text product using the annotated source material. Using the knowledge organizer quotations, paraphrases, summaries, and notes can be automatically collected and structured. A built-in full text editor assists in automatically extracting highlighted passages from the full texts into the Citavi system. The extracted elements are called “knowledge items” and can be assigned to user-defined hierarchical categories in Citavi. These categories can then be exported to a text as chapter headings via the Citavi MS Word add-in. Using this add-in, the knowledge items can also be further adapted. In addition, the corresponding citation information and a reference list in the required citation style are automatically integrated into the manuscript.

The task planner supports the planning and definition of (sub-)tasks related to the writing project and time management. Unfortunately, there is no export option or exchange with MS Outlook or any other task management software.

Team work was already possible since 2018, but the cloud-based collaboration required the Citavi desktop application for each team member and thus the use of MS Windows. Since 2021, Citavi is available via a web interface, so that now cross-platform team work with different operating systems is possible.

3.2 EndNote

EndNote is one of the earliest and most widely used reference management systems (Childress, 2011; Karavaev, 2016; McMinn, 2011). EndNote Desktop is a proprietary software, but available for MS Windows and MAC OS. It offers an add-in for MS Word, Apple Pages and Open Office that allows writers to insert in-text citations and create a reference list at the same time.

EndNote was launched in 1988 as desktop application by the developer Thomson Reuters; currently it is produced by Clarivate. A free web version “EndNote Web” is available since 2006. Although EndNote Web does not have all the features of Endnote Desktop, it is a good alternative to the proprietary version. “EndNote Web serves as an online supplement to EndNote Desktop, though it can be fully functional on its own. One does not have all the features of the other, and vice versa; they are complementary to each other” (Zhang, 2012, p. 47). For sharing references and collaborative writing, EndNote Web is necessary.

EndNote is one of the few reference management systems that does not offer an extension to collect references directly via web pages. There are extensions for data transfer, but these require a roundabout via import files. It is not a one-click process as in Zotero, Mendeley or Citavi. However, EndNote Desktop offers many import formats for a wide range of databases, catalogues and other platforms. In contrast to most other reference management systems, Endnote does not import citation information and full texts simultaneously, but rather sequentially. It also offers writers some degrees of freedom. For example, writers can define entirely new reference types to add non-typical entries such as photos or paintings with their special meta-data.

In addition, EndNote offers a function for searching in external databases. This search function should not be used in early stages of a writing project to locate and read relevant literature. Fitzgibbons and Meert (2010) compared the integrated EndNote search with direct searches in different databases. They showed that the comparability of the hits depends on both the database searched and the technique used for the search. Therefore, in early stages of a writing project it seems to be more meaningful to use the direct search in databases to be able to gain an overview of the current state of research. Furthermore, Endnote does not support a parallel search in multiple databases. Thus, the integrated Endnote search function is helpful, for example, to add citation information of already known sources to writers’ reference library.

3.3 Mendeley

German students developed the first version of Mendeley in 2008. Mendeley is free of charge, but not open source. In 2013, Elsevier purchased Mendeley. In the following years Mendeley was continuously developed. At the moment, Mendeley provides a desktop app for MS Windows, MAC OS and Linux, as well as a web version. It also offers a browser plugin to collect references and full texts directly from web pages and an add-in for MS Word and LibreOffice. Using Mendeley requires a personal account. Besides Mendeley Free a proprietary premium version “Mendeley Institutional Edition” is available. This edition provides more cloud storage space and more collaboration features, for example to create an unlimited number of shared groups.

Now, a radical change seems to be pending, as, according to statements on the product homepage, users will no longer be able to download and install Mendeley Desktop software after September 1, 2022. Mendeley's new reference management suite consists of three fully integrated applications. Mendeley Reference Manager can be used to organize and share source material, the Mendeley Cite add-in for MS Word to generate and format in-text citations and bibliographies, and the Mendeley Web Importer browser extension to create customized collections of source material when searching online.

A unique feature of Mendeley is that it automatically checks the correctness and completeness of the meta-data after importing new reference entries. To this end, a huge data collection is used, which is derived from the collected and corrected references meta-data of other Mendeley users. In this way, the laborious and time-consuming individual assessment and correction of inaccurate meta-data can be substantially facilitated and supported (Salem & Fehrmann, 2013).

3.4 RefWorks

RefWorks is a proprietary, web-based reference manager founded in 2001. In 2008, RefWorks was purchased by ProQuest (now Clarivate). Like EndNote, it is one of the earliest and most widely used reference management systems (e.g., McMinn, 2011). RefWorks offers add-ins for MS Word and Google Docs.

A unique feature of Refworks is that the collected source material is only saved in the cloud, not on a local computer. Thus, syncing and sharing references is very simple. The interface is intuitive to use, an integrated search in selected external databases is possible, as well as a subsequent addition of full texts.

3.5 Zotero

Zotero has similarly advanced features to support the use of source material in writing as Citavi, but is free and open source, supports all major operating systems, as well as various word processors in more than 30 languages. It was developed by a group of Librarians at the Roy Rosenzweig Center for History and New Media at George Mason University in 2006. The first version of Zotero was only a browser extension for Mozilla Firefox that collected and organized references. Since 2011, Zotero is available as a desktop application for MS Windows, Mac OS and Linux, as well as a web-based interface. Zotero still offers a browser plugin for Mozilla Firefox, Google Chrome, Safari and MS Edge to collect references and full texts directly from databases, catalogues and websites. For synchronization and collaboration, a Zotero account is needed. Zotero offers plugins for MS Word, LibreOffice and Google Docs.

As of March 2022, Zotero also includes an integrated full-text editor for annotating, tagging, and extracting full-text passages and images. Zotero saves these as notes, similar to the “knowledge objects” in Citavi. Using the word processor plugin, the writer can insert the notes into their own text; the corresponding citation information is automatically added and formatted according to the selected citation style.

With Zotero, writers have also a unique opportunity of expressing relationships between sources, as different references can be linked to each other using the “related items” feature. The writer decides for themselves according to which criteria they want to establish the relationship (for example, according to the content of the sources or according to formal criteria). In this way, the identification of relations between different sources and/or the sources and the writer's own positions is also supported, a very important activity when engaging with source material.

Zotero offers further a function for searching in external databases to import meta-data by standard identifiers such as ISBN, DOI or arXivID. This search function uses databases such as WorldCat, CrossRef and PubMed. In this way, the citation information of already known sources can be added quickly to writers’ reference library.

Finally, due to its open source character, there are numerous third party extensions that make Zotero even more powerful. For example, “zotfile” renames the related full text PDF-files according to a constant scheme. This renaming makes it easier to find sources, for example on a local computer, because the new PDF name includes the authors’ name, the publication year, and the title of the publication. Other extensions such as “Zotero Citation Counts Manager” track the citation counts of publications. Citation counts show the impact of publications in the scientific community. Note that they are provided for Zotero by free platforms such as Google Scholar and CrossRef, not by proprietary databases like Web of Science or Scopus. In addition, extensions like “Better Bibtex” or “LyZ” support writing with LaTex or further Tex-editors. All Zotero plugins can be found at Zotero’s plugin documentation web page: https://www.zotero.org/support/plugins.

4 Research

Over the last two decades, much has been published on reference management systems (e.g., Emanuel, 2013; Fourie, 2011; Tramullas et al., 2015). Most of these publications compare different reference management systems (see, for example, Tramullas et al., 2015 for a review). In addition, much has been said and written about the decision to use a reference management system. There is agreement that this decision depends on the conditions and preferences of the writer, as well as the writing context (e.g., Perkel, 2020).

There are very few peer-reviewed publications (e.g., Emanuel, 2013). The research in most cases focuses on how bibliographic data are captured, edited, and generated (Tramullas et al., 2015). In addition, most of the research focuses exclusively on locating and using new sources (Whittaker, 2011). Research on how the source material is organized for future (re-)use is almost not existent (Drake, 2013; Whittaker, 2011). In general, it appears that the body of literature on reference management systems can be broadly divided into four areas (see also Emanuel, 2013; Fourie, 2011):

  • Non-empirical, mostly narrative analyses and comparisons of reference management systems features, trends, and criteria of selection (e.g., Kali, 2016; Karavaev, 2016; Perkel, 2020; Steele, 2008; Vaidhyanathan et al., 2012; Zhang, 2012).

  • Empirical studies on the accuracy of automatic citation information (e.g., Fitzgibbons & Meert, 2010; Homol, 2014; Wyles, 2004).

  • Empirical, mostly survey-based, studies on user perceptions of reference management systems and behavior related to collecting and managing sources (e.g., Chen et al., 2018; Nitsos et al., 2021; Speare, 2018).

  • Best practices on training and support (e.g., Childress, 2011; Dovey, 2010; Mead & Berryman, 2010).

From this list, it becomes evident that as far as the impact of reference management systems on the theory and practice of writing in higher education is concerned, there is clearly a need for further research. Knowing what and how writers use functionalities of reference management systems can have a significant impact on the development of instructional strategies for writing with sources. However, there is virtually no research on how the use of reference management systems affects writers’ activities and/or quality, productivity, or creativity (e.g., Melles & Unsworth, 2015; Vaidhyanathan et al., 2012). One exception, for example, is the study by Rokni et al. (2010), whose study compared inserting and reformatting references in a paper manually with inserting and reformatting references using EndNote software. The study showed that significantly more time was required for manually editing and reformatting citation information. Lorenzetti and Ghali (2013) surveyed authors of clinical reviews and meta-analyses on the frequency and ease of use of reference management systems. Their study revealed that the majority of authors of systematic reviews use reference management systems, but do not report this in their articles. Furthermore, Lorenzetti and Ghali (2013) found no clear links to exist between the choice of a reference management system and its perceived functionality or ease of use. On the contrary, empirical, survey-based studies consistently show that users value many of the functionalities of reference management systems (e.g., Salem & Fehrmann, 2013; Setiani et al., 2020, 2021).

McMinn (2011) systematically analyzed the Association for Research Libraries (ARL) academic library websites for information on reference management system support. It was shown that support for reference management systems in the major academic libraries is substantial. He concludes significant time savings could be achieved if the staff responsible for supporting these reference management systems could share their expertise, possibly by establishing a central repository for training and instructional materials. Essentially, however, this study raises the question of how libraries, perhaps in cooperation with other institutions such as writing centers, can most effectively provide support. Even more than 10 years later, the current state of research makes it difficult to propose recommendations on this issue. Much of the literature on reference management instruction is exploratory or descriptive. More research, including experimental studies, is needed to understand effective instructional methods for utilizing reference management systems in academic writing.

In this regard, there appears to be a positive development for survey-based studies. Recent studies are using more focused methods and study designs, for example by adopting established frameworks such as the UTAUT model (Venkatesh et al., 2003) to examine the use and acceptance of reference management systems (e.g., Nurkhin et al., 2019; Rempel & Mellinger, 2015; Setiani et al., 2020). Other studies systematically examine user characteristics to better understand indicators of referencing in the scientific community such as citation counts (e.g., Chen et al., 2018). If this trend continues, the results of survey-based studies may also help shed light on the specific conditions under which reference management systems support academic writing.

The discussion so far indicates that reference management systems can support and optimize the utilization of sources in writing projects. However, few writers use the core features of reference management systems for their writing activities (Melles & Unsworth, 2015). Thus, how exactly, and to what extent reference management systems influence writing with sources is not well understood (e.g., Drake, 2013). An important goal of university education and training should be to raise awareness among writers (i.e. lecturers, theses supervisors, students, etc.) for reference management systems and their functionalities to support writing. In addition, further research is needed, for example, to examine how reference management systems can support writers’ alternate reading and writing activities during text production (e.g., Drake, 2013; McGinley, 1992). In addition, it is necessary to understand how specific features of reference management systems help to distribute writing activities between writers and tools in a beneficial way, and which features are suitable to support which activities of a writing project (e.g., O’Hara et al., 2002). To this end, it is necessary to adopt a writer's perspective rather than searching for the perfect bibliographic software.

Within their current state of functionality, reference management systems can assist in source reading and writing activities, but their use can also have risks. For example, using reference management systems may also weaken writers’ knowledge and understanding about referencing (Kali, 2016). Furthermore, the accuracy of reference management systems and/or the meta-data automatically imported from journal websites or other online sources can be a concern (Salem & Fehrmann, 2013). Therefore, it is important for writers to be aware of both the power and the limits of reference management systems.

Looking back to the beginning, reference management has become easier, faster and more social over the years (Fenner et al., 2014). Collecting sources and automatically updating citation information has become more comfortable. In the near future, reference management systems will increasingly have to deal with alternative scientific content such as presentation slides, blog posts, or web links. Standards of open science will play an increasingly important role. From a technical point of view, issues of portability of reference libraries, interaction possibilities with co-writers, opportunities to share annotated libraries with the academic community, as well as the integration of reference management systems into writing-to-learn, as well as learning-to-write environments, for example by using AI technologies, will be significant.

At present, only advanced users can take advantage of the support that most reference management systems provide for writing. Moreover, this support is not completely aligned to the demands of academic writing. For this reason, many authors use a wide variety of tools (e.g., note-taking tools, mind-mapping tools, etc.) for the different stages of their writing project. Thus, for future research and development, it would be valuable to bridge tools for reference collection, analysis, and annotation to tools that focus on knowledge management, processing, and production in the writing process. As a result, applications and tools that facilitate academic writing with sources can be developed evidence-based. First promising approaches in this regard are, for example, commercial software-as-a-service solutions such as Auratikum (https://auratikum.com/), Sciflow (https://www.sciflow.net/), or raxter (https://www.raxter.io/). Such solutions allow academic writers to have the most important functionalities for their writing conveniently in one place.

5 Implications of this Technology for Writing Theory and Practice

More research is needed to understand the exact nature of the relationships between searching, reading, and analyzing sources and using reference management systems in the context of academic writing. Moreover, the functionalities of reference management systems are not fully aligned with the writing process (Vaidhyanathan et al., 2012). More advanced reference management systems might even provide more support for writing projects. We would like to illustrate this with three examples below.

First, writers—particularly inexperienced writers—often pay limited attention to evaluating the trustworthiness of a source by examining the author, the publisher, or the type of publication (e.g., Britt & Aglinskas, 2002; Drake, 2013; Goldman et al., 2012). Using more and better evaluation strategies is associated with less reliance on unreliable sources in one's own text (e.g., Anmarkrud et al., 2014; Britt & Aglinskas, 2002; Goldman et al., 2012). Thus, reference management systems could more explicitly support writers in paying attention to such information (e.g., by color-coding). Currently, for example, different publication types, are difficult to distinguish from each other at first glance in reference management systems.

Second, writers rarely follow a linear sequence of systematically searching databases, downloading and entering references into their libraries, saving and analyzing the full texts of selected sources, and then producing their text including citations in the correct style (Drake, 2013; McGinley, 1992; Mead & Berryman, 2010). Rather, writing activities interact in a recursive way as the writing process unfolds. This requires the writer to keep track of what they have already studied, what sources are relevant, and how these sources relate to their text and/or the overall topic. Functions that facilitate this integration process are hardly offered in reference management systems so far (Vaidhyanathan et al., 2012). For example, it seems to be important for writers to be able to view, annotate, and use multiple documents at the same time (O’Hara et al., 2002). Graphical overviews or perhaps creativity software and idea mapping technology could also make connections between different sources more visible (Goldman et al., 2012, see https://docear.org/ how such a graphical organizer could look like). Including such functions into reference management systems or combining reference management systems with such tools would make cognitive resources available for the writer to focus on integration processes across the source material that should be beneficial for quality of one's own text (Goldman et al., 2012; O’Hara et al., 2002; Proske & Kapp, 2013).

Third, in particular inexperienced writers have difficulty comprehending and summarizing multiple sources and integrating this source into their own line of writing (Cumming et al., 2016; Wette, 2010). Thus, there is a great deal of variability in how writers incorporate sources into their text (e.g., Goldman et al., 2012). Many reference management systems provide the option to copy verbatim citations from the full texts (i.e. quotations). Doing so encourages writers to repeat the content of a single source, but not to summarize sources in one's own text, and certainly not to develop an integrated argumentation. Moreover, copying and reusing quotations in this way might even promote instances of plagiarism. Here, at least, a functionality would be desirable that not only automatically extracts the selected passages from the full text, but simultaneously also the corresponding citation information.

6 List of Tools

Citavi

Description: Citavi is a proprietary software. The desktop applications “Citavi for Windows” and “Citavi for DB Server” are originally only compatible with MS Windows. Citavi Web closes this gap, as Citavi Web is browser-based and thus operating system independent. In addition to literature management, Citavi also offers areas for knowledge management and task planning

Developer: QSR International; previously Swiss Academic Software

URL: https://www.citavi.com/

EndNote

Description: Endnote is a proprietary software that is widely used. Endnote is available for MS Windows and MAC. It is complemented by a free web version. Endnote offers a wide range of functions that support reference management

Developer: Clarivate; previously Thomson Reuters

URL: https://endnote.com/

Mendeley

Description: Mendeley is a free software, but not open source. In addition to the free version, a paid variant is available which offers additional features, such as increased storage space for (group) projects outsourced to the cloud. Mendeley is available for the operating systems Windows, MAC and Linux and is supplemented by a web version. In addition to the typical functions of a reference management system, Mendeley offers the best conditions for a comprehensive collaborative exchange. The user interface of Mendeley is only available in English

Developer: Elsevier

URL: https://www.mendeley.com/

RefWorks

Description: RefWorks is a proprietary, web-based reference manager. Both the meta data and the collected full texts are stored in the web cloud. Thus, location-independent use is possible and no local updates are necessary

Use of RefWorks requires an institutional subscription; individual user accounts are not offered

Developer: ExLibris, a Part of Clarivate; previously ProQuest

URL: https://refworks.proquest.com/

Zotero

Description: Zotero is a free and open source reference management system. In addition to versions for the MS Windows, MAC and Linux, Zotero is also available as a web-based tool. Starting as a simple browser extension for Firefox, Zotero has evolved into a comprehensive reference management software. Due to its open source character, numerous extension are offered for Zotero. The Zotero interface is available in more than 30 languages

Developer: Corporation for Digital Scholarship; previously Roy Rosenzweig Center for History and New Media, George Mason University, Virginia

URL: https://www.zotero.org/