Date of this Version
Published by Renew Training, a trading name of Simon Inger Consulting Ltd Published March 2016
Fernhill, Church Lane, Drayton
Abingdon, OX14 4JS, United Kingdom
This report is the output of a large-scale survey of readers of scholarly publications (n=40439) and their behaviour in the discovery of journal articles and online books. The survey was conducted during October, November, and December of 2015. While usage statistics and analytics gathered by publishers, libraries and intermediaries can give us a partial view of discovery behaviour, there are many gaps in the knowledge that these can provide which we have endeavoured to fill by asking readers what tools they use in discovery.6
This survey builds upon previous surveys conducted by the authors in 2005, 2008 and 2012. For four key questions in the survey, it allows for longitudinal analysis over the ten-year span, and for a further three questions allows for trend analyses between 2012 and 2015. The subtle shifts over time in reader preferences provide a valuable insight into reader navigation, the features that they find useful in publisher web sites, and the role and effectiveness of library technologies. For the first time, the 2015 survey includes three new questions regarding discovery of online books. Please refer to section 8 Methodology and section 9 Demographics, for a full discussion of the survey methodology and the demographics of those responding to the survey.
The discovery of journal content is certainly more refined than for online books, with a range of discovery methods available for most resources. Historically, journal articles tended to be available on a limited number of platforms, usually the publisher’s official web site and any sanctioned aggregation of its content, such as EBSCO and ProQuest collections. However, further incarnations of articles are increasingly discoverable in institutional repositories, subject repositories (especially PubMedCentral), as well as a range of other sites including ResearchGate, Academia.edu, and Mendeley. This has further complicated discovery since (at the time of writing) no single discovery service indexes all of these incarnations, and no single discovery service seems to index at least one incarnation of all of the content available. This limitation is partly one of business relationships, partly a lack of understanding of metadata distribution, and sometimes political constraints.
As a starting point for search, A&Is seem to be in a slight decline when looked at in aggregate across all regions and sectors, but remain the most important.
Academic researchers in high income countries now rate library discovery as highly as A&Is, and rate academic search engines as the most important discovery resource when searching for journal articles.
Library discovery services have made significant advances in importance in search for academic researchers, and for all roles in hard sciences in the academic sector. As an average across all subjects and sectors, however, they have not grown in importance in since 2012.
More than half of all journal content delivery appears to be from free incarnations of articles. There appears to be a clear PubMedCentral effect in the medical sector. Social media sites appear to be a significant source of free articles in lower income countries.
In academic STM in higher income countries, academic search engines are now more important than general search engines.
Table of Contents alerts have reduced in popularity in all measures across the survey.
There appears to be an increased role for social media in discovery.
Online book discovery varies significantly by sector, with academics preferring library web pages marginally over general web search engines, the medical sector preferring A&I services and library web over search engines, but all other sectors preferring search engines over other forms of discovery.
Publisher web sites are becoming more popular as a search resource, although this is less true for people in wealthier countries.
Google Scholar is used more than Google in the academic sector, but less than Google in all other sectors.
A perceived lack of awareness of Google Scholar in poorer nations appears to be leading to a reduced use of free incarnations of content in institutional repositories from these regions. Page 40
Readers in low income countries use their mobiles to access journals more than their counterparts in richer countries. However, access by phone still accounts for only about 10% of the use.
A&Is continue to be the most important search method in the medical sector.
The primary method of journals discovery is search, but even more so for online books.
App use for journal discovery is still low.
The most highly sought-after features of journal web sites are changing. Figure
Access to supporting data from a publisher website is more important to people in high income countries than people in lower income countries.
Across all demographics there is no significant appreciation of the availability of social media sharing or article-level metrics, even though most publishers feel that these are essential features.
Librarians behave quite differently to everyone else in search, preferring professional search databases and library-acquired resources.