Published on November 25, 2016
1. Exploring how user-generated content in Canadian public libraries can impact readers’ advisory services Louise Spiteri. Dalhousie University. Halifax, NS Laurel Tarulli. Sacred Heart School. Halifax, NS Jen Pecoskie. Independent researcher
2. The context 2
3. The social environment of the public library Public libraries are social environments that encourage the interaction, sharing, and communication of ideas, opinions, and many other types of information. In Readers’ Advisory (RA) services information professionals help clients find items that match their reading interests, traditionally via a face-to-face discussion initiated by the reader. 3
4. Reader expectations Social media websites like Goodreads and LibraryThing provide popular platforms for people to share and discuss their reading interests. Various online public access catalogues (OPACs) are integrating social discovery platforms such as BiblioCommons, SirsiDynix, and Encore, which allow users to connect with each other through user-generated content such as reviews, comments, recommendations, or tags. 4
5. Social information and readers’ advisory With the increasing popularity and use of social websites, such as Goodreads and LibraryThing, it is necessary to explore the implication of user-generated content for RA, to strengthen RA services, and to benefit readers who have become accustomed to sharing and benefiting from other users’ suggestions. 5
6. 6 Amazon Canada http://amzn.to/1TaO4GX
7. 7 Amazon Canada http://amzn.to/1TaO4GX
8. 8 Ottawa Public Library http://bit.ly/1TaOgG5
9. 9 Ottawa Public Library http://bit.ly/1TaOgG5
10. 10 Defining readers’ advisory (RA) Readers’ advisory services foster an environment where reading is a valued activity and readers’ advisory staff advocate the importance and joy of reading within the community. Successfully helping the user to find the next great fiction read or connecting a reader to what interests them in the non-fiction collection involves collection knowledge, readers’ services skills and good conversation. Southern Ontario Library Services (http://bit.ly/1TaNUzh)
11. Aspects of RA RA is traditionally a face-to-face discussion initiated by the reader or library staff, and is based on the reference interview. RA can be intimidating to readers: Shyness A lack of awareness of RA services Perception of library staff as authority figures Fear of being judged Perceived differences in age, gender, or culture 11
12. The value of user-generated content to RA Readers can discuss their views on books Readers can make recommendations for future reading Readers can provide tags (and thus access points) to supplement or complement Library of Congress Subject Headings Readers can discuss the emotional impact of books upon them RA staff can share their own experiences with the books and enter into a conversation with readers. 12
13. Two studies We will discuss the findings of two studies that examined user-generated content in the bibliographic records for 22 fiction titles. The same set of data was used for both studies: The first study looked more broadly at the nature of Library of Congress Headings, tags, and reviews that were found in the records. The second study examined in depth the affective content extracted from the user reviews. 13
14. Study 1: Content Analysis of LCSH, Tags, and Reviews 14
15. Study 1 Research Questions 1. What kind of content do users contribute about adult fiction titles, i.e., tags and reviews/comments? 1. What categories of access points do users provide about the content of adult fiction titles, e.g., location, subject, genre, and so forth? 1. How do user-contributed access points parallel those in traditional face-to-face RA model? 15
16. Methodology User-generated content was examined for 22 adult fiction titles in 43 Canadian public libraries that use the BiblioCommons, SirsiDynix, and Encore social discovery systems. The titles were selected from the following lists: Giller Prize Shortlist Canadian Governor General’s Literary Awards Man Booker Prize Shortlist Pulitzer Prize Fiction Finalists Commonwealth Writer’s Prize Winners 16
17. Grounded theory The derived LCSH and Tags and Reviews were analysed and coded into categories independently by four researchers (2 each for LCSH/Tags and Reviews) A fourth researcher examined all the sets of categories and created final sets. 17
18. What content do users provide about books? 18 No. of LCSH No. of Tags LCSH per title Tags per title Records without LCSH Records without tags Total 3501 4541 89.40 108.85 68 259 Average 159.14 206.41 4.49 4.95 3.09 11.77
19. What content do users provide about books? 19 Unique LCSH Unique tags Total 339 192 Average per title 15.41 8.73 No. of records with reviews Unique reviews Total 678 632 Average per title 30.82 28.73
20. 20 LSCH Tags Awards 2 (0.31%) 40 (19.23%) Format 4 (0.62%) 9 (4.32%) Genre 320 (48.84%) 30 (14.42%) Historical event 6 (0.93%) 4 (1.92%) Location 75 (11.68%) 18 (8.65%) People 101 (15.73%) 16 (7.69%) Period 140 (21.81%) 4 (1.92%) Tone 0 (0%) 21 (10.09%) Topic 116 (18.06%) 59 (28.36%) Main categories represented by LCSH and Tags (Abbreviated)
21. 21 Categories in user reviews
22. User access points vs. RA models 22 Tag categories RA categories Administrative note No equivalent Author No equivalent Awards Award/recognition Format No equivalent Genre Genre Historical event Setting; Real Events; Factual information Location Setting Pace Pacing Paratext Paratext People Specific characters; Charac. occupations Period Time
23. User categories vs. RA categories 23 Reviews categories RA categories Personal note No equivalent Plot Plot development; Ending Protagonists Specific characters Readability Readability Recommendation Award/recognition; Advice to reader Tone Emotional experience Topic Subject No equivalent Characters’ relationships No equivalent Intended audience No equivalent Literary influences No equivalent Size or length of book
24. Conclusions: Study 1 24 Cataloguers provide an objective stance about a book, but readers provide what it contains in a factual sense and what it contains in an emotional or reading experience sense. LCSH denote the “who, what, where, when, and type” of a work; while the tags do contain these elements, what is noticeable is their description of the tone or mood of a work.
25. Conclusions: Study 1 Tags express important aspects of a work not easily expressed by subject headings. Tags and reviews place heavy emphasis on affective aspects of work (readability, tone, mood, experience). “Aboutness” or thematic emphasis indicates a distinction between objectivity vs. “the feel” of a work. Readers provide a more complete picture of a title: what it contains factually as well as what it contains in an emotional and reading experience sense. 25
26. Study 2: Building affective taxonomies from reader reviews 26
27. Goal of the study 27 To create useful taxonomies of emotions, moods, and associations that could be used to assist readers as they narrow the focus of their searches for works of fiction, either through facets supplied by the social discovery system layer on a library catalogue (e.g., narrow the results by a specific types of emotions, such as sadness, joy, and so forth), or through interactions with readers’ advisory staff.
28. Research questions 28 1. What emotions are discussed as part of the reader’s reading experience? 1. What tones did the reading experience elicit for the reader? 1. What associations to external factors do readers make as part of their reading experience?
29. Methodology The reviews obtained from the first study were explored in depth. Using Grounded Theory, two researchers independently coded the reviews into three pre-determined categories based on observations from the first study: Emotions, Tones, and Associations (Memories). A third researcher reviewed the derived sets and created taxonomies for Emotions, Tones, and Associations. 29
30. Theoretical bases for the taxonomies Emotions Appraisal theory Basic emotions theory Prototype theory Social psychology Tones Library & Information Science Appeals theory RA studies Associations Cognitive psychology Neurobiology 30
31. 31 Basic Emotions Sub-emotions Anger Anger; Annoyance, Displeasure; Entrapment; Frustration Disgust Disgust, Dislike Engagement Captivation; Curiosity; Engagement; Reflection Fear Confusion; Difficulty; Disorientation; Fear, Stress; Uncertainty Happiness Anticipation; Excitement; Joy; Pleasure Love Admiration; Attraction; empathy; Enchantment; Lust; Sensitivity 9 basic emotions 44 unique emotions Taxonomy of Emotions
32. 32 Engagement 168 (26%) Happiness 154 (23.69%) Sadness 148 (22.76%) Fear 55 (8.46%) Surprise 38 (5.84%) Anger 28 (4.3%) Love 21 (3.23%) Uncategorized 21 (3.23%) Disgust 17 (2.61) Total 1633 Number of basic emotions in reviews
33. 33 Taxonomy of tones
34. 34 Realistic 125 (22.77%) Imaginative 92 (16.76) Frightening 72 (13.13%) Dramatic 62 (11.29%) Charming 49 (8.93%)` Humourous 35 (6.38%) Sad 34 (6.19%) Conventional 29 (5.28%) Complex 23 (4.19%) Optimistic 16 (2.91%) Cerebral 12 (2.18%) Total 549 No. of basic tones in reviews
35. 35 Taxonomy of associations
36. 36 Object 58 (43.93%) Agents 53 (40.15%) Experiences 29 (21.97%) Places 6 (4.55%) Events 3 (2.27%) Activities 2 (1.52%) Periods 1 (0.76%) Total 132 Number of basic associations in the reviews
37. 37 Conclusions: Study 2 The presence of 141 unique emotions, which occurred 650 times in the reviews serves as a strong indicator of the importance of affect in readers’ interactions with the books they read. The MARC record provides only the bare-bone description of the content of the work; the user reviews provide the added richness and nuances of the work that can help provide other readers with a greater understanding of the work.
38. Conclusions: Study 2 Reading experience cannot be expressed well in the MARC record. Affective access points can serve as an important addition to the bibliographic records for works of fiction. The derived taxonomies for Emotions, Tones, and Associations could be used as facets by which to narrow the results of a search, e.g., narrow the results by books that are humourous and surprising, or omit books that are frightening or sad. 38
39. Conclusions: Study 2 The broader categories could be used as suggested placeholders for users who wish to add tags to fiction titles (e.g., what Emotions did this book evoke?). RA staff could use these taxonomies to assist readers in selecting items to read, or to generate suggested reading lists that correspond to these taxonomies (e.g., books that are imaginative and cerebral). These taxonomies can help readers define more clearly their reading experience and why they enjoy (or not) reading certain works. 39
40. Looking for partners We would be very interested in partnering with any libraries to explore further the possibilities of how user-generated metadata could be used to enhance the RA and search experience. This would allow us to test our theories in a real environment 40
41. Contact information Louise Spiteri Louise.Spiteri@dal.ca @cleese6 About Me: https://about.me/louisespiteri 41
OPACs, Users, and Readers’ Advisory: Exploring the Implications of User-Generated Content for Readers’ Advisory in Canadian Public Libraries
Home Education OPACs, users, and readers’ advisory: Exploring the implication of user-generated content for readers’ advisory in Canadian public libraries 3 ...
... Catalog Records: Implications for Readers’ Advisory ... platforms and user-generated content (UGC) on readers ... services in Canadian public libraries.
Information Management Public ... of User-generated Content for Readers' Advisory in ... generated content to RA services in Canadian public libraries. ...
GLIS 614 - Public Libraries ... Users, and Readers’ Advisory: Exploring the Implications of User-Generated Content for Readers’ Advisory in Canadian Public ...
OPACs, users, and readers’ advisory: Exploring the implication of user-generated content for readers’ advisory in Canadian public libraries
Authors: Laurel Tarulli; OPACs, users, and readers’ advisory: Exploring the implications of user-generated content for readers’ advisory in Canadian public libraries
Check out the schedule for APLA 2016. ... Exploring How User-Generated Content in Canadian Public Library Catalogues Can Impact Readers’ Advisory Services ...