PaperlayerView exact revision on arXiv

Source-linked AI summary

Scholarly use of social media and altmetrics: a review of the literature

Cassidy R. Sugimoto, Sam Work, Vincent Larivière, Stefanie Haustein

arXiv:1608.08112v1cs.DL

TL;DR

Research on scholarly social media use and altmetrics lacked a comprehensive synthesis connecting platforms, academic practices, and derived indicators. This review integrates that literature and finds substantial heterogeneity, with social media opening channels for informal researcher discussion but not guaranteeing broader conversation about research.

  • Problem

    A comprehensive review linking scholarly social media use with metrics derived from that activity was lacking, limiting understanding of their users, uses, and communities.

  • Method

    The paper reviews empirical literature on academic social media use, institutional practices, and altmetric indicators, including their roles, interpretations, and methodological limitations.

  • Results

    Social media opened new channels for informal discussion among researchers, but its presence alone did not guarantee broader conversation about research.

  • Takeaways & Limitations

    Social media and altmetrics represent a substantive transformation of scholarly communication rather than a passing trend.

  • Takeaways & Limitations

    Altmetric evidence is constrained by DOI-focused aggregation and inconsistencies between curated datasets and API-mined data, complicating coverage and cross-dataset comparisons.

Abstract

from arXiv · show

Social media has become integrated into the fabric of the scholarly communication system in fundamental ways: principally through scholarly use of social media platforms and the promotion of new indicators on the basis of interactions with these platforms. Research and scholarship in this area has accelerated since the coining and subsequent advocacy for altmetrics -- that is, research indicators based on social media activity. This review provides an extensive account of the state-of-the art in both scholarly use of social media and altmetrics. The review consists of two main parts: the first examines the use of social media in academia, examining the various functions these platforms have in the scholarly communication process and the factors that affect this use. The second part reviews empirical studies of altmetrics, discussing the various interpretations of altmetrics, data collection and methodological limitations, and differences according to platform. The review ends with a critical discussion of the implications of this transformation in the scholarly communication system.

1 Introduction

Scholarly communication is being transformed by social media, which increases the visibility and dissemination of scholarship while creating demand for indicators of research activity. This review examines academic social media use and altmetrics, including their functions, influencing factors, interpretations, data limitations, and platform differences.

  • Social media is discussed as increasing scholars’ and scholarship’s visibility and providing new vehicles for dissemination.
  • Research funders and administrators increasingly demand indicators demonstrating research value to broader audiences.
  • Understanding social media uses and users is central to identifying communities engaging with scholarly content through altmetrics.
  • The review’s first part examines academic social media use, platform functions in scholarly communication, and factors affecting use by researchers and institutions.
  • Its second part examines altmetrics as research indicators based on social media activity, including interpretations, data collection and methodological limitations, and platform differences.

2 Social media use in academia

Social media has spread rapidly and nearly ubiquitously, but scholars and professional researchers remain critical of its use, while evidence about scholarly adoption and perceptions is limited by common survey-method weaknesses.

  • Professional perceptions: Health policy researchers rated social media the poorest dissemination method, citing professional risk, uncertain efficacy, incompatibility with research, and unfamiliarity.Their assessment came from a survey of dissemination methods.
  • Evidence on scholarly use: Studies of scholarly social media use mainly measure self-reported personal use, institutional use, and perceived usefulness through surveys, interviews, and focus groups.The evidence base emphasizes professional purposes and attitudes toward usefulness.
  • Evidence on scholarly use: Because these studies frequently use non-probabilistic samples, their findings should be generalized with caution.The passage identifies sampling as a recurring methodological limitation.

2.1 Scholarly use by researchers · Social networking

Researchers’ overall social media use is high, but survey estimates are broad and individual-platform use varies considerably. Social networking sites support scholarly communication through diverse platforms, motivated mainly by connection, dissemination, and following research, despite time-related adoption barriers.

  • 2.1 Scholarly use by researchers: 75–80% of researchers reported using social media in large-scale surveys, although those surveys used a broad definition of social media.The definition included video conferencing tools such as Skype and collaborative writing tools such as Google Docs.
  • 2.1 Scholarly use by researchers: Less than 10% of scholars reported using Twitter, illustrating considerable variation in use across individual platforms.
  • Social networking: Social networking sites let users create profiles, identify connections, and view or navigate those connections within a bounded system.
  • Social networking: Scholars use both general-public networks, including Facebook and LinkedIn, and scholar-targeted sites such as ResearchGate, Academia.edu, and VIVO.
  • Social networking: Facebook has repeatedly been reported as the most frequently used social networking platform among scholars.Facebook’s prominence is associated with its scale of 1.5 billion users, representing half of the world’s online population.
  • Social networking: Connecting with researchers, disseminating research, and following others’ research output are primary motivations for scholarly use of social networking sites.
  • Social networking: A barrier to professional use is weighing the time required to learn a new tool against its expected advantages.

Social bookmarking and reference management

Social bookmarking and reference-management tools let scholars save, organize, and share publications, with journal articles the most commonly bookmarked genre. Mendeley is promising for altmetrics but remains relatively lightly adopted, and its users and uses vary by research status.

  • Social bookmarking and reference management: Bookmarking and reference-management tools enable users to save, organize, and share research publications across all genres, although journal articles are most widely used.Examples include Mendeley, Zotero, CiteULike, BibSonomy, and Connotea.
  • Social bookmarking and reference management: Mendeley is considered a promising altmetrics data source, but surveys generally find adoption below 10%.More than one-quarter of bibliometricians report using Mendeley, indicating higher adoption in that subgroup.
  • Social bookmarking and reference management: Doctoral students and junior researchers form Mendeley’s largest reader group, while faculty members publicize publications and students search for them.These differences indicate that Mendeley use varies by research status and purpose.

Social data sharing

Social data sharing platforms support sharing scholarly objects, but social functionality and scholarly participation remain limited. Available evidence indicates that researchers’ awareness, regular use, and sharing activity are generally low.

  • Infrastructure: Data sharing is required by several funders and journals to enhance scientific verifiability and replicability, while platforms support sharing scholarly objects.Many platforms focus on specific fields or individual communities.
  • Infrastructure: Only a few repositories, including Figshare, SlideShare, and Github, combine social functionalities with substantial scholarly participation.These platforms are identified as having reached a certain level of participation from scholars.
  • Adoption: Social data sharing and reuse are not yet common, and surveys of platform adoption and use remain rare.Evidence includes low awareness, regular visitation, uploads, and engagement among researchers.
  • Adoption: Less than 10% of surveyed researchers were aware of Figshare, and as few as 0.5% visited the site regularly.These figures indicate limited awareness and recurring use of the platform.
  • Adoption: Less than 5% of highly-cited researchers had SlideShare uploads, which received relatively few views, likes, or shares.The study therefore found limited participation and engagement among highly-cited researchers.

Video

Video functions as a genre for social interaction and scholarly communication, with YouTube the most popular video-sharing platform and science and technology prominent among highly discussed videos.

  • Video: Video provides a genre for social interaction and scholarly communication.The passage identifies video as part of scholarly communication.
  • Video: YouTube, launched in 2005, is by far the most popular video-sharing platform.The passage contrasts YouTube with other video-sharing platforms.
  • Video: Science & Technology videos are prominent among highly discussed videos on the platform.YouTube offers a broad spectrum of content, while Science & Technology is prominent among highly discussed videos.

Blogging

Scholarly blogs broaden research communication by disseminating and critiquing published work, enabling academic culture critique and communal engagement. However, their credibility, institutional recognition, permanence, and evidence base remain contested.

  • Infrastructure: ResearchBlogging.org, launched in 2007, provides a relatively stable structure that aggregates posts referencing peer-reviewed research and credentials scholarly blogs through its icon.This stability makes the platform suitable for empirical study.
  • Functions: Scholarly blogs commonly disseminate, comment on, and critique published research while also supporting critical discussion of academic systems.Their informality facilitates academic culture critique.
  • Functions: Blogs’ inherently social features enable communal scholarship, academic discourse, and greater public engagement with science through linking and commenting.Commenting was associated with a shift from public understanding toward public engagement.
  • Use: 86% of surveyed medical bloggers read blogs to find medical news, while 23% of British doctoral students followed blogs and 9% contributed to them.Academic blogging communities have also exhibited small-world network properties.
  • Limitations: Blogging’s scholarly status remains uncertain: it was rated among the least efficacious research-dissemination methods, and most institutions reportedly do not recognize it consistently in promotion and tenure.Blogs are also vulnerable to abrupt cancellation and deletion, challenging their permanence as scholarly communication.
  • Evidence limitations: Generalization is difficult because many studies use convenience samples, small samples, or single case studies that may overrepresent highly active online users and strong opinions.These sampling limitations may bias conclusions about scholarly blogging.

Microblogging

Microblogging is widely known among scholars but used less than other social media, with Twitter serving both professional and personal purposes. Scholarly microblogging supports dissemination, networking, conference communication, journal clubs, and rapid critique of research.

  • Microblogging: Microblogs are well-known among scholars, but reported use ranges from 5% to 32% and is lower than use of other social media.Studies rarely distinguish between personal and professional use.
  • Microblogging: Most Twitter users identify themselves by full name and profession, and professionally identified users have more followers.Follower networks can help identify influential participants in scholarly dissemination.
  • Microblogging: Although academics commonly use Twitter for research discussions and field communication, a large share of their activity remains personal.Microblogging therefore blurs professional and personal boundaries.
  • Microblogging: Scholarly tweets often indirectly link to recent journal articles and blogs, while direct links are more frequently used for open access articles.Tweets linking to scholarly articles tend to be limited to certain patterns of dissemination.
  • Microblogging: Twitter enables scholarly interaction beyond conferences, including journal clubs and rapid critique or correction of the scientific record.Health professionals have used Twitter to discuss relevant papers, and research data have been re-analyzed and posted there.

Wikis

Wikis, especially Wikipedia, support scholarly information seeking and research dissemination, but scholars contribute to them at much lower rates than they use them for searching. Their broader academic uses include collaborative drafting, research documentation, and enhanced presentation of research materials.

  • Wikipedia is viewed as a credible source and has been suggested as a useful starting point for research.
  • 30-45% of scholars use Wikipedia for searching information, whereas contribution rates are much lower.
  • Low contribution rates may reflect Wikipedia’s lack of explicit authorship, which conflicts with academic reward systems.
  • Wikis are used internally to write early article drafts and document research processes as online lab notebooks.
  • Web-based wiki platforms can enhance research dissemination through functionality such as rendering three-dimensional images.

Social recommending, rating, and reviewing

Social platforms and dedicated systems use recommendations, ratings, and discussion to filter or assess scholarly material, blurring the boundary between recommending and reviewing. Although early open-review efforts attracted limited participation and poor perceptions, newer platforms and communities have renewed interest in reviewing and scholarly discussion.

  • Recommendation and rating: Dedicated systems have been developed to filter relevant scientific content through recommendations and ratings, with F1000Prime identified as the most popular platform.These systems complement the filtering function that commenting and discussion can provide on social media platforms.
  • Recommendation and rating: Online platforms blur the boundary between recommending and reviewing, while peer review remains a quality-control and gatekeeping mechanism.Traditionally closed gatekeeping has become more transparent and visible online through practices such as signed open reviews.
  • Open peer review: Post-publication commenting has been limited, with most PLOS ONE comments written by authors and editors, while scholars have rated open peer review poorly.Despite these early shortcomings, F1000Research and PeerJ show increasing interest in open peer review.
  • Informal discussion and rating: Scientific journal mentions on Reddit were historically rare, but new science-focused subreddits and Ask Me Anything sessions have expanded discussion of scientific information.Reddit users can submit, discuss, and rate online content.
  • Informal discussion and rating: Amazon and Goodreads have been mined as potential sources for compiling impact indicators from comments and ratings of books.Both sites allow users to comment on and rate books.

2.2 Scholarly use by institutions and organizations

Scholarly institutions and organizations increasingly use social media to disseminate and promote research, support researchers, and connect with wider audiences. Adoption varies across organizational types and is shaped by institutional characteristics, publishing models, resources, and platform-specific practices.

  • Institutional dissemination: Universities, libraries, professional societies, and publishers increasingly use social media to diffuse and promote research.Universities also use these tools for student recruitment, while libraries use them to promote services; those marketing uses are outside this review’s scope.
  • Institutional dissemination: Higher education institutions disseminate research to local and wider communities, with use varying by institutional size and structure.Faculty and student use has also prompted studies of social media’s integration into higher education’s pedagogical mission.
  • Researcher support: Libraries create social media accounts and provide services, tools, and metrics to support researchers’ social media use.They also use products tailored to libraries, including Mendeley Institutional Edition, which mines documents, annotations, and behavior.
  • Publishing organizations: Journal adoption varies widely: Facebook presence ranges from 9% to 80%, while Twitter presence ranges from 15% to 44%.The reported variability suggests that organizational social media adoption remains heterogeneous.

2.3 Factors affecting social media use

Scholarly social media use varies by age, academic rank, gender, discipline, country, and language, but findings are often mixed across populations and platforms. Social media can reproduce existing visibility disparities, while access and credibility also differ geographically.

  • 2.3 Factors affecting social media use: Use and acceptance vary by age, academic rank, gender, discipline, country, and language, although results differ across populations and studies.These differences mirror established patterns in scholarly communication.
  • 2.3 Factors affecting social media use: Age-related findings are inconsistent: younger scholars view social media as more credible, but studies report no age difference, higher use among older scholars, or higher use among younger scholars.The cited studies therefore do not establish a single age gradient for scholarly communication use.
  • 2.3 Factors affecting social media use: Rank and platform shape use in divergent ways, with doctoral students prominent on social bookmarking services and Twitter use highest among scholars with 7 to 9 years of academic experience.Assistant professors were also more likely than associate and full professors to use social media for creating and maintaining research connections.
  • 2.3 Factors affecting social media use: Disciplinary differences lack consensus, despite evidence of field-specific barriers and higher platform use in biomedical sciences, Computer Science, and Mathematics.The literature suggests that disciplinary context affects acceptance and use, but findings are not uniform.
  • 2.3 Factors affecting social media use: Social media often reproduces traditional visibility disparities: scholarly blogs are dominated by English, and publications from emerging countries receive lower visibility on social networking platforms.Platform use and credibility also vary by country, while restricted access to tools such as Twitter affects scholarly communication.

3 Social media and research evaluation

Altmetrics offer multidimensional signals of scholarly attention, but their use in research evaluation is constrained by conceptual, coverage, data-quality, and user-transparency limitations. Empirical evidence shows platform- and field-dependent relationships with citations, with Mendeley providing relatively broad coverage and moderate correlations.

  • Conceptual foundations: Altmetrics face criticism because they lack validation and a clear conceptual framework, and may measure attention and popularity rather than research impact.These concerns motivate caution in interpreting altmetrics as impact indicators.
  • Conceptual foundations: Altmetrics are classified by action—viewed, saved, discussed, cited, or recommended—and by audience, such as scholarly or general-public users.These distinctions reflect the multidimensional nature of altmetrics.
  • Data limitations: Identifier-based aggregators neglect document types without unique identifiers, while DOI-focused collection effectively narrows many studies to journal articles.Altmetric.com’s DOI focus excludes some documents and journals from measurement.
  • Data limitations: Curated datasets and API-mined data differ, making cross-dataset comparisons difficult because platforms use different retrieval strategies and coverage rules.For example, Altmetric.com tracks only public Facebook wall posts, whereas PLOS ALM also includes private posts, shares, and likes.
  • Empirical evidence: Facebook coverage is low, ranging from less than 1% to around 10%, whereas Mendeley covers 60-80% of articles with at least one reader.Coverage varies by research domain, genre, and platform.
  • Empirical evidence: Mendeley reader-count correlations with citations range from .2-.7, usually around .5 or .6, while CiteULike correlations are lower at about .2 to .4.Other platform-specific findings include no significant relationship between tag counts and citedness, and higher readership for F1000-reviewed papers.

4 Conclusion and outlook

The review concludes that social media and altmetrics have transformed scholarly communication by enabling new ways to diffuse, discuss, and organize scholarship and by producing new research indicators. Their future significance remains uncertain because altmetrics depend strongly on the platforms and services underlying them.

  • Review contribution: The review synthesizes empirical literature on scholars’ and organizations’ use of social media platforms and on the strengths and limitations of associated indicators.It focuses on platforms’ roles in scholarly communication and on indicators measuring related activities.
  • Drivers of adoption: The popularity of social media and altmetrics reflects technological innovation alongside funders’ and researchers’ demand for indicators beyond publications and citations.Start-ups and scholarly publishers supplied the technological push, while policy demands created a complementary pull.
  • Platform dependence: Altmetric indicators are strongly dependent on the underlying social media platforms, whose affordances shape scholarly online acts and generate platform-specific measures.Many of these acts cannot exist outside the particular platform on which they occur.
  • Implications: The review argues that scholarly social media use and altmetrics are more than fads because they enable broader discussion than earlier digital communication tools.The earlier wave included email, listservs, and electronic journals, whereas the second wave supports broader discussion.
  • Outlook: Whether social media and altmetrics become central to research dissemination and evaluation remains unresolved, as platforms and indicators may disappear when they lack relevance or their underlying services end.This possibility underscores their strong, potentially entire, dependence on platforms and services.
Loading 1608.08112v1…