How readers-discover-content-in-scholarly-journals-summary-edition
Upcoming SlideShare
Loading in...5

How readers-discover-content-in-scholarly-journals-summary-edition






Total Views
Views on SlideShare
Embed Views



0 Embeds 0

No embeds



Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
Post Comment
Edit your comment

How readers-discover-content-in-scholarly-journals-summary-edition How readers-discover-content-in-scholarly-journals-summary-edition Document Transcript

  • How Readers DiscoverContent in Scholarly Journals Summary Edition Comparing the changing user behaviour between 2005 and 2012 and its impact on publisher web site design and function. By Tracy Gardner and Simon Inger
  • Published by Renew Training Copyright © Renew Training, Tracy Gardner and Simon Inger ISBN 978-0-9573920-3-8 Renew Training Fernhill, Church Lane, Drayton Abingdon, United KingdomThe full version of this report (123 pages and 90 figures) is available in Kindle (for Kindle, iPad, PC and Mac) and PDF formats, the entire survey results data set uponwhich this report is based, and the analytical framework are available for purchase at
  • ACKNOWLEDGEMENTS This work would have not been possible without the support of the following organisations, who between them executed hundreds of thousands of invitations to journal readers asking them to participate in this survey. Our heartfelt thanks go out to them all. BMJ Group CABI Cambridge University Press IOP Publishing Nature Publishing Group Palgrave Macmillan Publishing Technology RSC Publishing SAGESummary Edition Page 1
  • TABLE OF CONTENTS Acknowledgements 1  Table of Contents 2  Introduction 3  Methodology 4  Demographics 5  Discovery Resource Preference 8  Discovery Resource Conclusions 16  The Role of the Library 18  Search Engine Preference 20  Device Preference 21  App Use 22  Publisher Web Site Features 23  Conclusions 24  About the Authors 25 Summary Edition Page 2
  • INTRODUCTION This summary report is the output of a large scale survey of journal readers (n=19064) about journal content discovery conducted during May, June and July of 2102. While statistics and analytics can tell us some of this information, there are many gaps in the knowledge that these can provide which we have endeavoured to fill by asking readers what how they discover journal content.Summary Edition Page 3
  • METHODOLOGY This research carries on from, and expands upon, previous research undertaken in 2005 and 2008 (also by Simon Inger and Tracy Gardner) and attempts to follow the trends in behaviour over that period of time. Naturally, each time the survey is repeated, the authors have sought to keep the questions as consistent as possible with the questions in earlier surveys whilst keeping terminology current and tracking new developments. For this reason the three key questions on reader behaviour were modified a little, some options being reclassified and additional options created. However, since those questions don’t limit how many starting points the reader acknowledges as being important, this approach should have minimal impact on the results for any option present in the survey all the way from 2005 to 2012. Other questions were dropped completely, since the conclusions from these in 2008 are now so widely accepted as fact (and easily checked with analytics) that these were not tested. These included asking readers where links from discovery products would take them in publisher web sites, the answer being predominantly at the article level. The full methodology is available in the full report available at: journals.htmSummary Edition Page 4
  • DEMO OGRAPH HICS Figure 1 spondents to the survey come from an excellent As shown in Figure 1, the res n s e regional s spread and this allow for sign d ws nificant regional (an in some cases nd e elow. 1330 people told us th eir country (and country) breakdowns, see be 08 hence reg gion). Figure 2 9406 peoople came from the t top 15 cou untries sho own above i.e. 71% of those e, % e who indic cated a cou untry.Summary Edition Page 5
  • Figure 3 The secto or-breakdo own is also very goo with su o od, ufficient nu umbers in all but Charity/N NGO to allo for furt ow ther break kdowns by subject, r region, inc come and 8958 people told us which sector they worked in. so on. 18 w Figure 4 The numb bers of res spondents by job role allows for meanin s f ngful comp parisons to be mad for all roles save for, perha de r aps, journalists and marketing g/PR/sales roles. Th s here are su ufficient re within academic esponses w researche lecturer and stud er, dent categ gories to allow for co onsiderable further e demograp phic analy ysis within these gro oups, including by su ubject, reg gion and income. 1 17403 peo ople told u s their job role. bSummary Edition Page 6
  • Figure 5 When em mbarking on this proj ject, our aim was to get aroun 1000 responses a o nd in each su ubject are so that detailed demographic analys would be ea, d sis b possible w minimal error-bars. This was achieved for six of within subject with m the subje areas, and three more sub ect a bject areas got aroun 600 res s nd sponses, which still allows fo some us or seful further sub-div vision. Com mputer Sc cience, Earth Scie ence, Environmenta Science and Mathe al ematics ar the leas well- re st represent ted in the data, alth ough in ab bsolute terms there are enough e individuals in these areas to allow for a useful su e ubject-bas sed analys even sis, if not whe combin en ned with a further de emographic.Summary Edition Page 7
  • DISCOVERY RESOURCE PREFERENCE In designing a journal’s online presence, a publisher needs to gain an understanding of how readers will navigate to the journal and at what part of the journal web site they will arrive. This will help inform decisions on which partners to work with, how to distribute essential data to them, and how to design web pages within a journal web site that meet the needs of readers wherever they arrive within the site. This research focuses on three main forms of reader behaviour with respect to journals; citation searching, core journal browsing, and subject searching. Given these different approaches to the literature, researchers select their most appropriate starting points on the internet (described extensively in the full edition of this report at discover-content-in-scholarly-journals.htm) and navigate to journal content. The combination of where readers want to perform certain functions (such as search) and on which pages within a journal web site that the reader “lands” as a consequence of their navigational behaviour sheds light on the design and feature-set of a journal’s web presence. It also helps inform publishers as to which kinds of starting points they should seek to enable first, for the greatest possible return in reader traffic. For each type of behaviour this research tested, the survey asked about a number of different starting points. An explanation of these starting points is included in the full report. See discover-content-in-scholarly-journals.htm CITATION SEARCHING In this question we asked people to state the importance to them of each starting point when following up on a citation.Summary Edition Page 8
  • Figure 6 We chose not to co e ontinue wit some of the optio from t th ons the earlier surveys. Terminolo ogy in the sector ha moved on and mo as o oreover th e boundar ries between what were consider e red to be “journals gateways” and librar web “ g ry pages has blurred considerab s c bly. Comm munity web sites hav come online b ve o since 200 In addi 08. ition, for t the first tim this ye me, ear’s surve separates out ey the use o the likes of Google from Goo of s e ogle Scholar, Bing f from Micro osoft Academic Search. To make a sensible compariso betwee the muc smaller c T on en ch r surveys o 2005 an 2008 an this one the 201 data ha been do of nd nd e, 12 as own- sampled using a random sele ection met thod to giv a samp of appr ve ple roximately y the same size as th 2008 st e he tudy and with the sa w ame subje mix as before. ect This samp is biased toward Life Scie ple ds entists in North Ame N erica and Europe. E The full re eport show the res ws sults for al responde ll ents to the 2012 sur e rvey and shows de emographic breakdo wns by region, income bracke subject area, c et, t job role a and sector See www r. w.renewtra m/how-rea aders-disco over- content-in n-scholarly y-journals s.htmSummary Edition Page 9
  • For this sample of respondents, we see that use of a specialist bibliographic database continues to climb. Given the comments made in the survey it is not surprising – so many life scientists commented that they use PubMed almost exclusively. As in previous years the survey shows that readers faced with a citation seem to know their subject areas well enough to go directly to the web site of the journal to follow up on the citation, whilst the use of library web pages in this regard is in steady decline over the period. Web pages managed by a key research group have increased since 2008. Given the margin for error, there is no significant difference in respondents using the publisher’s website or the journal’s homepage to look up a citation. More respondents are using a Scholarly Society web page to look up article citations than in 2005 and 2008. Another feature of this year’s results is that readers of online journals seem to have become much savvier about their information discovery. Some of the options that seemed oddly popular to us in the past, such as using an archive of ToC alerts to follow up on a citation, have declined in popularity in 2012. Academic search engines such as Google Scholar are more popular than general web search engines and are the second most popular source for looking up a citation, after the bibliographic databases. CORE JOURNAL BROWSING The second behaviour studied is the user who regularly reviews a few select journals that he considers worth scanning upon publication. In this question we asked people to state the importance to them of each starting point when reading their core content.Summary Edition Page 10
  • Figure 7 This chart shows th trend in behaviou for read t he n ur ders who w wish to dis scover the latest articles in the subject area. A& database continu to grow as a eir t &I es ue w resource for this ty ype of beha aviour. A publisher’s web site journal s e, homepages and sch holarly So ociety web pages hav all grow in popu ve wn ularity, perhaps indicating that reade are becoming more familia with jou ers ar urnal and publisher brands. Web pages managed by a key research group has also W s d y s grown, in ndicating perhaps th e growing success of some of these res p g o f sources. Showing a significa downw ant owever it is still the ward trend is journal alerts, ho second m most popular resourc for disco ce overing latest article es. The full re eport show the res ws sults for al responde ll ents to the 2012 sur e rvey and shows de emographic breakdo wns by region, income bracke subject area, c et, t job role a and sector See www r. w.renewtra m/how-rea aders-disco over- content-in n-scholarly y-journals s.htmSummary Edition Page 11
  • SUBJECT SEARCHIN NG The third form of user behav viour studied occurs when a us is sear ser rching for articles on a specifi subject. A user is likely to undertake a compre ic . u ehensive subject se earch prior to under rtaking res search in a specific f field or wh hen seeking to check, prior to pu blication, the precis state of the curre p se f ent literature. In this qu uestion we asked pe e eople to sta the im ate mportance to them of each o starting p point when reading t n their core content. Figure 8 This ques stion asked responde d ents about their beh haviour wh hen search hing for articles on a subjec The gra ct. aph shows quite diff s ponses to the other ferent resp two types of behav s viour we as sked abou – following a citat ut tion and vi iewing the latest articles. Specialist bibliograp t S phic databases (A&Is are still the most s) l t popular resource fo this type of activity and allo or e owing for a margin of error, o shows no significan change over time. Library web pages however have o nt w sSummary Edition Page 12 2
  • grown significantly in popularity, possibly due to the introduction of web scale discovery services. At the same time general web search engines have shown a slight downward trend, possibly because the additional alternative option of an academic search engine was added to the options in the survey. Web pages managed by a key research group and society web pages have both shown a slight upward trend which may be due to changes in publisher marketing strategies resulting in readers becoming more familiar with publisher and society brands. In absolute terms, searching within a journal aggregation is quite significant, as are publisher web sites and journal homepages, the latter of which has shown a significant change since 2005. As in previous years, all of these methods have a value to someone but the most popular method in this comparison is Abstract and Indexing databases. The full report shows the results for all respondents to the 2012 survey and shows demographic breakdowns by region, income bracket, subject area, job role and sector. In addition it shows the correlation in search behaviour between subjects. See content-in-scholarly-journals.htmSummary Edition Page 13
  • SUBJECT AREA BRE EAKDOWN S IN SEAR RCH Figure 9 Library w web pages are of mo importa ost ance to pe eople work king in Edu ucation Research and Humanities fol lowed by Social and Political S d Science an nd Agricultur Respon re. ndents in P Physics va alued library web pa ages a lot less than l many oth subject areas by quite a si her t y ignificant margin. The full re eport cont tains furth breakd her downs inclu uding aggr regators, A&Is, A communi web sit ity tes, and m more. See www.renewtraining. w .com/how-readers- discover-content-in n-scholarly y-journals. .htm MOST REC CENT ARTI ICLE ACCES SSED With the luxury of such a larg numbe of responses to th survey, it is also s ge er he a useful c check to ask how re aders disc covered the very las article th they st hat accessed. This is us seful since in part it can validate their p e t previous answers about the relative importance of variou starting points, b also indicates e e us g but which mo odes of use they are more com e e mmonly in. It is com mmon to im magineSummary Edition Page 14 4
  • that most people are in searc mode more than in browse mode, or in t ch m e r following- -up-on-a-c citation m ode. Figure 10 0 This grap compare behavio from Academic Researche in Nort ph es our A ers th America, Europe an Asia an shows that Asian researche are mu more nd nd t ers uch likely to b searching for arti be icles than anything else, and much mor likely e re to be doin so than their Eur ng n ropean and North Am d merican co ounterpart Links ts. in emails are also more frequ m uently use by Asian research ed n hers than those t from Nort America and Euro th a ope. There is no not e table differ rence in th he behaviour of acade emic resea rchers in Europe and North A America. The majority of researc chers in bo regions say they were sea oth s y arching for articles r when they last acce essed an a article. Ho owever, jou urnal alert are also well ts o used, as are article citations from othe sources. Social ne e er etworking links, journal bo ookmarks and saved search alerts are not used f d a n frequently by researche in any of these t ers three regio ons. The full re eport show demog raphic bre ws eakdowns by region, income bracket, , b subject area, job ro and se ole ector. See www.rene ewtraining w-readers- discover-content-in n-scholarly y-journals. .htmSummary Edition Page 15
  • DISCOVERY RESOURCE CONCLUSIONS ABSTRACTING AND INDEXING SERVICES (SPECIALIST BIBLIOGRAPHIC DATABASES) A&Is are seen to be the most important starting point for many subject areas across many sectors in search, and also an increasingly common way of following up on a known citation. The trend since 2005 is positive in all three modes of use studied, although some subject areas are much better served than others. LIBRARY WEB PAGES (PREVIOUSLY LIBRARY OPAC) Investment in library technology seems to be having a positive impact on the use of library web pages in search, while as a means of viewing the most recent articles in a subject niche, popularity remains flat. Readers seem to be turning away from these resources as a means of following up on a citation. A JOURNAL COLLECTION, OR AGGREGATION (NEW IN THIS SURVEY) Journal aggregations overall are important, but not as important as many other discovery methods, until one starts to look into some subject disciplines and other demographics where they are more popular. A COMMUNITY WEB SITE (NEW IN THIS SURVEY) Across the board, community web sites such as Mendeley and Researchgate are used much less than other starting points for all three behaviours. This could be because they are relatively new and have yet to build up the brand recognition and penetration needed to compete with some of the other resources. WEB PAGES MANAGED BY A KEY RESEARCH GROUP Key research groups are also used less than other starting points, although they are more important than community web sites. This could be because they generally appeal to smaller subject niches so when we look at the results more broadly, they are relatively unimportant. However, when looking at the trend graphs, they have grown in importance for all three types of behaviours studied.Summary Edition Page 16
  • A PUBLISHER’S WEB SITE A publisher’s web site has become more important for looking at latest articles in core journals over time. For the other two types of behaviours their importance has remained static. EMAIL BASED ALERTS For all three types of behaviour, email based alerts have decreased in popularity since 2005. However, when looking at them in comparison to other resources in absolute terms they are still important for looking at the latest issues of core journals. THE JOURNAL’S HOMEPAGE The journal’s homepage has remained important for looking up a citation; this hasn’t changed over time and is as important as an academic search engine, and more important than a general search engine. It has slightly grown in popularity for discovering latest articles and searching. GENERAL WEB SEARCH ENGINES General web search engines have grown very slightly in popularity for people looking up a citation, grown slightly more for people discovering latest articles but decreased in popularity for people searching. This could be because search engines have been separated out this year to general and academic. ACADEMIC SEARCH ENGINES (NEW FOR THIS SURVEY) When following a citation, academic search engines are the second most popular resource across the board. They are less important for people who want to discover latest articles; they are more likely to use the journal home page, journal alerts, a publisher’s web site or an A&I. When searching, an academic search engine is the second most popular resource. SOCIETY WEB PAGES Whilst still not as important as many of the resources listed, society web pages have grown in importance in all three behaviours studied.Summary Edition Page 17
  • THE ROLE OF THE LIBRAR RY Over the past fiftee years, l en libraries have been expending ever more time g and mone in the manageme of e-re ey m ent esources, and much of these efforts a e have been focussed on the d d developme and implementat ent tion of libr rary web pages, journal A to Z listings improvements to library cat o s, talogues, library l link-serve and no web sca discovery products. These web site tools are ers ow ale e ology vendors – often the sam organisations available from libra techno ary me responsib for the earliest li brary auto ble omation pr rojects of 30-40 yea ago, ars while others are mo recent innovators in the area. ore t a In additio to unde on erstanding library we pages as a starti ng point, the eb a t survey as sked reade about t ers the influen of libra techno nce ary ology on navigation n and the findings are shown b below. Figure 11 1Summary Edition Page 18 8
  • This plot shows the sum of those who said library technology affected navigation more than 50% of the time in each category, for the 15 countries providing the highest aggregate response to the survey. Readers in Malaysia are highly aware of library technology, much more so than any other country, with the possible exception of Australia. Awareness in Canada, US and Australia is more on a par. There is an indicative result that awareness is lowest in Germany, Japan and China which either indicates that take up in these regions is low or it is so unobtrusive, readers are not aware of how the library affects discovery. Perhaps somewhat surprisingly, awareness in Iran is relatively high which supports the view that the Middle East is becoming an important market for library technology. The full report shows further demographic breakdowns. See journals.htmSummary Edition Page 19
  • SEARC ENG CH GINE PREFERE ENCE Figure 12 2 Information manag gers use G Google search engine less tha student and es an ts researche ers. Presum mably this is due to their use of more s s specialist tools. Of t note perh haps, is that Informa ation Mana agers mak margina ke ally more use of Scirus tha student and res an ts searchers. Students use Goog le Scholar slightly r aps surprisingly academic res more than Google, and perha searchers use u Google m more than Google Sc G cholar - ma aybe because they a higher users of are r A&I datab bases and will use s search eng gines for a more gen neral searc ch, negating some of the need fo Google Scholar. or The full re eport show demog raphic bre ws eakdowns by region, income bracket, , b subject area, job ro and se ole ector. See www.rene ewtraining w-readers- discover-content-in n-scholarly y-journals. .htmSummary Edition Page 20 0
  • DEVIC PREFERENC CE CE Figure 13 3 The Academic secto uses mo or obile phon for jou nes urnal acces less tha the ss an medical, charity and corpora te sectors and a sim s, milar amo ount to the e governme sector. There is an indicat ent tive result that the g group usin tablets ng and phon to acce online journal ar nes ess rticles is th medica l sector. he ector make the most use of static devi ces and th The government se es s heir use of laptops is on a par with ac s p cademics. Charity an Corpora sector make nd ate rs the least use of des sktops for this purpo ose. The full re eport show further demograp ws r phic break kdowns. S See www.rene ewtraining w-readers- -discover-c content-in n-scholarly y- journals.h htmSummary Edition Page 21
  • APP USE Figure 14 4 In compa aring resea archers in academic and corpo orate envi ronments, we find , that the c corporate researche are bet ers ped with s martphones than tter equipp demic counterparts, but apparently don make a their acad , n’t any more use of u them for accessing journals v apps. via As an indicative res sult, inform mation ma anagers may be mak king more use of e apps in accessing journals, e especially in browsing. The res i sults show that, in w absolute terms, App use is q uite low. The full re eport show further demograp ws r phic break kdowns an reveals many of nd the intere esting com mments ma ade on thi topic. Se is ee www.rene ewtraining w-readers- -discover-c content-in n-scholarly y- journals.h htmSummary Edition Page 22 2
  • PUBLISHER WEB SITE FE EATURE ES In the survey, resp pondents w were invite to indicate which features, of a ed selected l list of feat ublisher we sites, they found useful. These tures of pu eb d ose tested in 2005 and 2008 t allow fo direct features a the same as tho are to or cifically tes some more mode inventions, such compariso and do not spec on, o st m ern as faceted browse/search. d Figure 15 5 A sample of the 2012 data w was taken at random in the bro a m ortions of oad propo region an subject area seen in the 20 nd n 008 resear and the results for this rch e f set compared with the 2005 and 2008 results. This can sh T how us the trends e over time but cannot be take as an in e en ndication of behavio on a global or o our indeed pa an-subject area basi t is.Summary Edition Page 23 3
  • CONCLUSIONS It is widely acknowledged that readers arrive on publisher web sites from a discovery resource at the article level, and for a few years now publishers have been building web sites that expose vital functionality to the reader on that landing page. However, search on the publisher web site is persistently popular, not just on external discovery platforms, and indeed the number of readers who bookmark key pages in publisher web sites, or simply remember them, shows that publishers need to keep a multitude of navigational paths open to their readers. Moreover, this research shows that the relative importance of all of these paths vary from subject to subject, from region to region, and by job function. Multidisciplinary publishers need to take special note of keeping an open mind to reader navigation while perhaps more specialist ones could take a more decisive approach. A key measure of publisher success is the usage of its e-journals, which can be maximised by influencing and enabling all the routes to its content. Library technology plays a key role in user navigation, as well as the more apparent starting points such as Google or major subject A&I databases. Publishers need to support all conceivable routes to their content through the web. This can best be achieved through the distribution of XML header information to as many discovery platforms as possible, through RSS feeds, collaboration with CrossRef, library technology vendors and through working with major gateways, A&Is and search engines. As metadata distribution is maximised and users are able to choose more freely their preferred routes to content, many of the advanced features that users require seem to be migrating to their chosen discovery platforms leaving the publisher site ever more as a content silo, without the need for quite so many of the advanced features that are currently present there. However, publishers remain under pressure to maintain a high level of functionality to ensure that they engage with content buyers, authors and editorial boards.Summary Edition Page 24
  • ABOUT THE AUTHORS Simon Inger has worked in the journals industry for over twenty-five years. In this time, he has worked for Blackwell, CatchWord, Ingenta and, since 2002, as an independent consultant. Simon was Founder and Managing Director of CatchWord Ltd, the world’s largest e-journal platform of the time, from its inception in 1995 to its sale to Ingenta in 2001. Simon has worked extensively in journal sales, marketing and pricing; e-journal delivery and platform selection; fulfilment and editorial systems selection; business reviews; management; financial planning; product development; market research; content development; and library technology. In addition, he runs training courses for librarians in the UK and Ireland on e-journal technology and management, as well as courses for publishers on best practice in e-journal dissemination. Simon’s clients include societies, university presses and commercial publishers from across Europe and North America. Tracy Gardner has over fifteen years’ experience in marketing and communications. She has a very broad view of the industry having worked for publishing technology companies (CatchWord and Ingenta), a not-for- profit publisher (CABI Publishing) and a consultancy company (Scholarly Information Strategies) where she worked on various projects for publishers, intermediaries and libraries. Throughout her career she has been focussed on improving the communication channels between publishers, intermediaries and librarians and understands the business of scholarly publishing from many different perspectives. She now has her own company and offers consultancy and training services to those involved in the scholarly publishing industry. Tracy has been a co-trainer on UKSG’s E- Resources Technical Update course since the beginning of 2007. For the full report see discover-content-in-scholarly-journals.htmSummary Edition Page 25