Research Article

 

Exploring Topics and Genres in Storytime Books: A Text Mining Approach

 

Soohyung Joo

Associate Professor, School of Information Science

University of Kentucky

Lexington, Kentucky, United States of America

Email: soohyung.joo@uky.edu

 

Erin Ingram

Research Assistant, School of Information Science

University of Kentucky

Lexington, Kentucky, United States of America

Email: erin.ingram@uky.edu

 

Maria Cahill

Associate Professor, School of Information Science

University of Kentucky

Lexington, Kentucky, United States of America

Email: maria.cahill@uky.edu

 

Received: 15 Apr. 2021                                                                  Accepted: 14 Oct. 2021

 

 

cc-ca_logo_xl 2021 Joo, Ingram, and Cahill. This is an Open Access article distributed under the terms of the Creative CommonsAttributionNoncommercialShare Alike License 4.0 International (http://creativecommons.org/licenses/by-nc-sa/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one.

 

 

DOI: 10.18438/eblip29963

 

 

Abstract

 

Objective – While storytime programs for preschool children are offered in nearly all public libraries in the United States, little is known about the books librarians use in these programs. This study employed text analysis to explore topics and genres of books recommended for public library storytime programs.

 

Methods – In the study, the researchers randomly selected 429 children books recommended for preschool storytime programs. Two corpuses of text were extracted from the titles, abstracts, and subject terms from bibliographic data. Multiple text mining methods were employed to investigate the content of the selected books, including term frequency, bi-gram analysis, topic modeling, and sentiment analysis.

 

Results – The findings revealed popular topics in storytime books, including animals/creatures, color, alphabet, nature, movements, families, friends, and others. The analysis of bibliographic data described various genres and formats of storytime books, such as juvenile fiction, rhymes, board books, pictorial work, poetry, folklore, and nonfiction. Sentiment analysis results reveal that storytime books included a variety of words representing various dimensions of sentiment.

 

Conclusion – The findings suggested that books recommended for storytime programs are centered around topics of interest to children that also support school readiness. In addition to selecting fictionalized stories that will support children in developing the academic concepts and socio-emotional skills necessary for later success, librarians should also be mindful of integrating informational texts into storytime programs.

 

 

Introduction

 

Storytime programs for preschool children are a common provision of public libraries worldwide. With a long history (Albright et al., 2009) and the highest rates of attendance among public library program offerings (Miller et al., 2013), storytimes serve as the backbone of public library programming. Preschool storytimes are typically designed to be highly interactive, learning-focused, thirty-minute enjoyable endeavors that incorporate shared book readings, props, songs, and activities intended to engage both children and their adult caregivers (Diamant-Cohen & Hetrick, 2014; Goulding et al., 2017; Mills et al., 2018).

 

Book selection is one task among many that storytime providers undertake when planning a public library storytime program. Other planning tasks may include deciding on a theme, incorporating educational tips for caregivers, gathering supplies for an activity or craft, creating handouts for caregivers, integrating technology resources, and choosing songs, interactive rhymes, fingerplays, movement activities, or flannel board activities (Diamont-Cohen & Hetrick, 2014; Ghoting & Martin-Diaz, 2006). To aid in planning these tasks, storytime providers may rely on professional resources for storytime development such as conferences, books published by the American Library Association and other library publishers, trainings offered through state library agencies and consortia, and webinars or courses from commercial websites such as Webjunction or Library Juice Academy. However, these formal resources may not be accessible to all storytime providers due to financial costs or time constraints. Therefore, storytime providers at a public library with a limited budget may prefer to plan with the aid of a free informal, online resource, such as a librarian’s blog about storytime programs or a library’s website with recommended reading lists. 

 

In this study, we explored the subject matter of books recommended by online resources for public library preschool storytime programs. We extracted bibliographic records of books recommended for storytimes and analyzed them to explore topics, genres, and sentiment of those books. We applied text mining methods for data analysis, such as term frequency and network analysis, LDA topic modeling, and sentiment analysis.

 

Literature Review

 

Theoretical Framework

 

Picture books serve as unique multimodal sources of information in which verbal and visual elements combine and interact to convey meaning (Martens et al., 2012). The tension between visual and verbal modes—the tendency to slowly gaze upon images and layout versus the forward momentum of reading text—make the picture book reading experience distinct and synergistic (Sipe, 1998). Serafini (2012) has posited four roles essential for picture book reading: 1) the navigator role entails proficiently employing conventions of print and cognitive strategies and skills to move through the written text while simultaneously making sense of the visual design and images; 2) the interpreter role consists of meaning making; 3) in acting as a designer, the reader determines the path through which a text is experienced by determining the order of attention to and importance of textual and visual elements; and 4) in enacting the interrogator role, the reader serves as a critical analyst acknowledging that messages and interpretations are socially constructed and politically, socially, and culturally powerful. Using Eisenberg and Small’s (1993) framework for information-based education, Campana (2018) demonstrated that information resources, information process, and individuals interact within the storytime context to provide learning opportunities for all involved. As a primary information resource within storytime programs, picture books are an important element of the storytime learning environment, and the storytime librarians’ processes facilitate children’s picture book reading and meaning making.

 

General Content of Children’s Books

 

Nearly half of all picture books published in English in the United States feature animals as leading characters (Horning, 2016). Preschool-aged children have shown an aesthetic preference for artwork, such as picture book covers, that feature animals (Danko-McGhee & Slutsky, 2011). Markowsky (1975) pointed to four reasons why children’s book authors and illustrators might choose animal characters: a) void of the accoutrements associated with specific groups of people, animal characters are relatable to children of all stripes and circumstances; b) animal characters produce an element of whimsy, enabling escape, inspiration, and imagination; c) animal characters serve as a form of shorthand or symbolic representation; and d) animal characters with exaggerated features and characteristics lend an element of humor.

 

Book Choice in Preschool Education

 

Several studies have examined characteristics of books read aloud in preschool classrooms and centers. Mesmer (2018) interviewed the staff of 31 preschool centers in the southwestern United States who expressed that the characteristics they considered most frequently were illustrations, rhyme, length, simple content, and a topic relevant to children. When examining the genres of such books, researchers have noted a lack of informational books (Pentimonti et al., 2011; Thoren, 2016). This is concerning because sharing informational books provides many educational benefits to children, such as content knowledge and preparation for successfully reading this genre in school (Lennox, 2013; Neuman et al., 2016), and because children are highly interested in informational books (Baldwin & Morrow, 2019; Price et al., 2012). In addition, studies of preschool classrooms revealed a lack of concept books, or books that focused on foundational academic topics such as counting and the alphabet (Gou et al., 2013; Pentimonti et al., 2011).

 

Book Choice in Public Library Storytimes

 

Diamant-Cohen and Hetrick (2014) contended that “the main goal of preschool storytime is to help children develop a positive connection with books and illustrations, which will later translate into a positive attitude toward books in general” (p. 4). To accomplish this goal, librarians typically incorporate three to four books in each storytime program (Diamant-Cohen & Hetrick, 2014; Goulding et al., 2017; Kociubuk & Campana, 2019). Naturally, the topics and sentiment of books shared in public library storytime programs will affect the extent to which children develop a positive connection; thus, librarians’ book choices are important.

 

Despite the importance of book choice, only a few studies have examined the characteristics of books chosen for storytime programs. One study examined the books shared in 69 baby, toddler, preschool, and family storytimes in public libraries in the state of Washington during 2013 (Kociubuk & Campana, 2019). The book characteristics collected were genre (fiction or nonfiction) and publication date; thus, this study does not provide insight into the topic or sentiment of storytime books. In contrast, two surveys of professionals who choose storytime books revealed that they did take into account the topic and other aspects of a book’s content (Carroll, 2015; Fullerton et al., 2018). Carroll (2015) identified seven influential factors in book selection: length and complexity; illustrations; subjects, concepts, and themes; use of language such as rhyme and repetition; how easily the book could be used to invite audience participation; elements such as suspense or humor that could emotionally engage the children; and personal preference. Similarly, respondents to Fullerton et al.’s (2018) survey chose language use and illustrator’s craft as top considerations when selecting books.

 

Objectives of Public Library Storytimes

 

One of the important program objectives of storytimes lies in learning and education (Campana, 2018; Fehrenbach et al., 1998; Peterson, 2012). Storytime providers have long emphasized early literacy development of young children when designing programs (Albright et al, 2009), but scholars have recently noted the broader role libraries play in supporting school readiness (Campana, 2018). Storytime serves as a valuable opportunity for children to build knowledge structures of colors, numbers, singing, alphabet, and more (Cahill et al., 2018). According to Celano and Neuman (2001), “by reading books, telling stories, and reciting rhymes, librarians offer children a ‘leg up’ in developing emergent reading skills” (p. 39).

 

While parents and caregivers certainly value storytimes for the learning opportunities they afford, many view these programs as family entertainment venues (Khoir et al., 2017) and  worthy experiences simply because of the joy they bring to children (Cahill et al., 2020). Most storytime programs offer opportunities for children and adults to interact, play, and sing (Celano & Neuman, 2001). Stories told in rhymes and picture book versions of songs, in addition to encouraging playful interaction, may aid children in developing early literacy skills such as phonological awareness (Giles & Fresne, 2015).

 

Additionally, storytimes serve as opportunities to stimulate and extend children’s feelings and emotional experiences because reading children’s literature can be a source of emotional learning for children (Short, 2018; Thoren, 2016) and shared book reading has been found to be a viable activity to help children build socio-emotional competence (Schapira & Aram, 2019). Further, a majority of literacy educators, including librarians, regard social emotional learning as a responsibility of literacy educators but one for which they need further support (International Literacy Association, 2020).

 

The Aim and Research Questions of the Study

 

Storytime programs for young children are offered in nearly all public libraries. However, few researchers have investigated the topics and content of the books that librarians use in such programs. It is important to have a better understanding of the content of the books to produce best practices and suggestions for book selection in storytimes for young children. Our overarching aim through this study was to explore topics and genres of the books that are read in storytimes for young children in public libraries. In addition, we aimed to examine the nature of sentiment represented in books widely used in storytimes for young children. The methodological contribution of this study is that it employed computational text analysis methods to investigate the content of a large sample size of books recommended for library storytime programs. The following two research questions guide the investigation of this study:

 

RQ 1. What are the topics and genres of books recommended for public library storytime programs designed for preschool children?

 

RQ 2. What is the nature of sentiment represented in books recommended for public library storytime programs designed for preschool children?

 

Table 1
Online Resources for Storytime Chosen for This Study

Resource

Number of Total Themes

Number of Themes in Stratified Random Sample

Esther Storytimes

74

15

Jbrary

19

4

MCLS kids

158

32

Silly Librarian Preschool

100

20

Storytime Katie

195

39

Storytime Secrets

70

14

Total

616

124

 

Methods

 

To identify informal sources that storytime providers were likely to encounter and use, we conducted a simple Internet search for “storytime resources for librarians.” We then selected the first six sources that recommended books for storytime use based on theme: Esther Storytimes, Jbrary, MCLS Kids, Silly Librarian, Storytime Katie, and Storytime Secrets. We next created a list of all preschool storytime themes shared on each site, which totaled 616 in all (Table 1). Because each theme contained multiple book recommendations, including the entire population would be laborious and unfeasible from a data preparation standpoint; thus, we randomly selected 20% of the themes, resulting in 124 themes in the sample. For each theme, we recorded the name and author of each book recommended as supporting the theme and appropriate for storytime.

 

In this study, we analyzed largely two types of textual information collected about each book from the WorldCat database: “titles and abstracts” and “subject terms.” We excluded any books that did not have an abstract available from WorldCat. After removing them from the book list, 429 books were used for analysis. Two sets of text corpus data were made for text mining: (1) titles and abstracts and (2) subject terms.

 

Multiple text mining techniques were employed to explore the content of storytime books. First, a term frequency analysis was conducted to identify the most frequent terms that occurred in each corpus. The collected text underwent a pre-processing process, including tokenization, stopword elimination, and stemming. In addition, bi-grams were further investigated to identify key concepts or popular genres in storytime books. Second, we analyzed the relationships among the terms based on term co-occurrence analysis. A term co-occurrence map was created to identify key topics and genres in storytime books. Third, Latent Dirichlet Allocation (LDA) topic modeling was applied to uncover prevailing topics underlying the books commonly recommended for storytimes. LDA topic modeling is an unsupervised machine learning method to discover hidden themes or topics from unstructured text (Blei, 2012). We extracted 20 topics from the corpus of titles and abstracts. Fourth, we explored the nature of emotion reflected in storytime books by analyzing sentiment terms. The sentiment lexicon constructed by Liu (2015) was adopted to investigate emotional aspects of the text from storytime books. The bing lexicon classifies selected words into the binary categories, i.e., positive and negative sentiment categories (https://www.cs.uic.edu/~liub/FBS/sentiment-analysis.html).

 

Textual analysis was conducted using various R software packages. R is a software tool that can be used for statistical analysis and data science, and it includes a multitude of packages for natural language processing and text mining. For term frequency analysis, two packages were mainly used: R tm and R tidy. For the LDA topic model, the R topic models package was employed.

 

Results

 

RQ 1. What are the topics and genres of books recommended for public library storytime programs designed for preschool children?

 

First, we calculated the most frequent terms from the corpus of titles and abstracts. It exhibited 2,289 unique terms and 7,718 tokens after removing stopwords. In textual analysis, a unique word (also called type) refers to a distinct term in a corpus while a token indicates an occurrence of a unique type (Jackson & Moulinier, 2007). We investigated the top 96 stemmed terms that appeared more than 15 times (Appendix A). The two most frequent terms are “book” (1st, 1.35%) and “anim” (2nd, 1.17%). Abstracts usually summarize the books, so it is not a surprise that the term “book” occurred most frequently from the corpus. Interestingly, “anim,” which indicated the term of “animal(s),” was observed second most frequently. In addition, there were other animal-related terms observed among top ranked terms; for instance, “bear” (3rd), “cat” (15th), and “dog” (21st), among several others. The results also included several terms associated with early learning, such as “color” (8th), “rhyme” (12th), “alphabet” (26th), “read” (43rd), and “count” (82nd). Another distinct group of frequent terms can be classified as book audiences or characters, such as “young” (7th), “babi” (9th), “children” (10th), “boy” (12th), and “reader” (23rd). In addition, several terms were related to actions or movements, such as “play” (3rd), “find” (3rd), and “danc” (37th). We also observed terms depicting emotion, such as “love” (3rd), “fun” (43rd), and “enjoy” (82nd).

 

We next turned attention to the “subject terms” corpus and tallied frequencies of the terms observed in it. The corpus consisted of 538 unique terms and 3,642 tokens. Subject terms were much shorter than the combinations of titles and summaries. Also, this corpus mostly consisted of nouns with only few adjectives or verbs observed. The top 82 most frequent terms that appeared more than five times are listed in Appendix B. The analysis showed that subject terms are likely to provide genre information or more condensed topic terms. The two top-ranked terms indicated genres of books: “fiction” (1st, 25.95%) and “juvenil” (2nd, 13.40%). WorldCat organizes books by genre using subject headings. For example, A Fairy-Tale Fall by Apple Jordan has three subject terms: “Autumn -- Juvenile fiction,” “Princesses -- Juvenile fiction,” and “Halloween -- Juvenile fiction,” and represents a typical format of WorldCat subject terms, which is a combination of a topical term and a genre. Thus, genre related terms were a frequent observation from the corpus. Other genre classification terms that ranked highly included “stori” (3rd), “rhyme” (6th), “literatur” (7th), “pictori” (10th), “picture” (15th), “movabl” (17th), and several others. This finding implied that most storytime books could be categorized as juvenile fiction, stories in rhyme, pictorial works, picture books, or movable books. We can also infer that storytime books involved other genres, such as “folklore” (24th), “poetri” (26th), and “nonfict” (55th).

 

In addition, subject terms depicted prevalent topics in storytime books. Interestingly, similar to the results from the titles and abstracts, the term “anim” is ranked highly at 5th, revealing the popularity of animal related topics in storytimes. Animal related terms included “dog” (19th), “bear” (20th), “cat” (24th), “rabbit” (46th), and several others. Several of the terms implied topics related to knowledge and skills important for young children; for example “count” (20th), “song” (28th), “alphabet” (34th), “sound” (46th), “scienc” (46th), “read” (68th), and “languag” (83rd). Storytime books also reflected social and behavioural topics, such as “friendship” (46th), “behavior” (61st), and “social” (68th). Other notable topics or concepts that children can learn included: family (e.g., “famili” (32nd), “mother” (34th), “parent” (46th), “son” (61st), and “father” (55st)), nature (e.g., “natur” (32nd), “snow” (46th), “moon” (61st), and “tree” (83rd)), settings (e.g., “farm” (39th), “garden” (46th), and “zoo” (83rd)), Halloween (e.g., “Halloween” (34th) and “pumpkin” (68th)), and other objects/characters (e.g., “monster” (55rd), “pirat” (61st), and “dinosaur” (68th)).

 

We further analyzed bi-gram terms, which indicated two adjacent tokens, from the titles and abstracts corpus. In total, 9,018 unique bi-grams were observed after removing stopwords. Appendix C lists the top 44 bi-gram terms that were counted four times or more. The top two bi-grams are “picture book” and “rhyming text.” Also, among the ranked bi-grams, there were bi-grams that indicated the types of books, including “board book,” “simple text,” and “illustrations rhyming.” This finding highlighted that the nature of storytime books related to pictures, rhymes, and simple text to be shared with young children. The bi-gram analysis result also showed popular topics and characters in storytime books. As shown in Figure 1, a term co-occurrence network revealed relationships among key terms in storytime books. The term “book” showed close associations with “picture,” “children,” and “animal,” which reveals popular types of books in storytimes. Another notable linking group is made of “text,” “illustrations,” and “rhyming.”

 

The same bi-gram analysis was conducted for the subject terms (Appendix D). In total, 3,250 bi-grams (1,312 unique bi-gram types) were observed. The most frequent bi-grams indicated the genres or types of storytime books; for example “juvenile fiction,” “juvenile literature,” “fiction stories,” “board books,” “pictorial works,” and so on. The visualization of term co-occurrence analysis highlighted the genres of storytime books as it placed “juvenile” and “fiction” among other genre terms in the center of the network diagram (Figure 2).

 

To explore popular topics in storytime books, we derived 20 topics from the titles and abstracts corpus using the LDA topic model (Table 2). Not all topics exhibit distinct and coherent thematic terms, but we found that these topics can be classified largely into several categories. There are several learning components observed from the extracted topics: for example, T7 (e.g., alphabet, letter) and T4 (e.g., color, crayon, count). Animals and creatures were another prominent topic in storytimes, such as T6 (e.g., cat, dog), T2 (e.g., mouse, spider, pigeon), and T11 (e.g., bear, elephant). The LDA topic model also detected topic terms relevant to book types, such as T9 (e.g., book, illustration), T14 (e.g., rhyme, illustration, and interactive) and T20 (e.g., illustration, rhyme). The components of activities, actions, and movements were also reflected in storytime books: for example, T5 (e.g., dance, swim, discover) and T4 (e.g., play). Moreover, topic terms obtained from the LDA algorithm represented other aspects of topics covered in storytime books, ranging from audiences (e.g., children, young reader), friends and families (e.g., friend, family, mother), nature (e.g., tree, sky, flower), settings (e.g., farm, garden), and emotions (e.g., fun, love, happy).

 

Figure 1

A network of term co-occurrence relationships: titles and abstracts.

 

Figure 2

A network of term co-occurrence relationships: subject terms.

 

Table 2

LDA Topic Model (20 Topics)

T1

T2

T3

T4

T5

T6

T7

T8

T9

T10

night

fun

stop

home

busi

visit

pig

differ

follow

pictur

mous

spider

includ

pigeon

back

web

mommi

cooki

long

varieti

hat

magic

text

rabbit

classic

pop

best

name

treasur

chicken

play

color

crayon

children

creat

heart

zoo

bestsel

count

time

danc

pumpkin

halloween

swim

discov

five

favorit

board

pirat

grade

cat

love

dog

shoe

imagin

pete

around

white

kitti

sing

alphabet

letter

featur

differ

time

pictur

world

full

pea

everi

grow

follow

mani

plant

turn

tree

eat

forest

celebr

librari

book

illustr

life

old

artist

time

cut

journey

color

easi

egg

girl

back

salli

bug

butterfli

ice

big

chick

bring

T11

T12

T13

T14

T15

T16

T17

T18

T19

T20

bear

big

boy

brown

bus

eleph

tree

celebr

around

cake

find

boy

tri

song

pictur

school

describ

best

everyth

join

farm

bunni

mother

text

dinosaur

tale

bed

amaz

world

readi

young

reader

rhyme

garden

illustr

moon

children

interact

reveal

edit

stori

give

introduc

train

blanket

surpris

children

includ

sit

answer

perfect

famili

crocodil

keep

three

full

live

water

know

show

anim

tail

page

cloth

enjoy

everyon

wear

parent

louis

aloud

book

babi

love

penguin

toe

red

along

delight

flap

page

friend

read

simpl

snow

tri

sheep

monster

kiss

text

hous

appl

ten

illustr

tree

rhyme

leav

gingerbread

pie

simpl

trick

 

RQ 2. What is the nature of sentiment represented in books recommended for public library storytime programs designed for preschool children?

 

Sentiment analysis was conducted to explore the emotional aspects of storytime language. We identified the top 25 positive and negative terms respectively (Figure 3). Overall, there were more positive terms than negative terms observed from the titles and abstracts corpus. Frequently observed positive terms include “like,” “love,” “perfect,” “fun,” “classic,” “favorite,” and others. These positive terms described storytime books as likable, easy, fun, enjoyable, playful, happy, and other upbeat descriptions. On the contrary, there were fewer negative terms observed. The top ranked negative term turned out to be “fall.” According to our further observation, however, there were more cases when the term “fall” was used to indicate the season, rather than an act of falling or moving down: for example, “a little girl spends a glorious fall day picking apples and searching for the perfect pumpkin in this edition of a timeless favorite” in the book Apples and Pumpkins by Anne Rockwell. Other negative terms did not necessarily have any negative nuance in the context of children's stories. For instance, pigs and bugs are likely to be featured as friendly characters rather than unpleasant objects in storytime books. Despite the nuance based on context, other terms appeared with obvious negative connotations, such as “die,” “tired,” “trouble,” “lonely,” “skeptical,” “fear,” and others.

 

Figure 3

Positive and negative terms in storytime books.

 

Discussion

 

By analyzing the most frequent terms, we explored different aspects of storytime books, such as popular topics, audiences, and styles or techniques for sharing the books. Not surprisingly, animals were a major topic in these recommended books just as they are in picture books in general (Horning, 2016). In our results, various types of animals were ranked highly, such as bear, cat, dog, mouse, and rabbit. Because preschool-age children have shown a preference for artwork that features animals (Danko-McGhee & Slutsky, 2011), our results indicated that sharing these recommended books would appeal to children’s interests.

 

The frequent terms also pointed to opportunities to integrate school readiness and other elements of early learning into storytime programs. Several different concepts related to children’s learning were observed, such as color, rhymes, alphabet, read, song, and count. The use of concept books may be especially important in public library storytimes to give children additional support in these content areas because studies have found that preschool classrooms and centers may offer only small numbers of concept books (Guo et al., 2013; Pentimonti et al., 2011). Additionally, the findings coincided with the storytime providers’ preference for concept books that may engage and empower children with familiar, repetitive content (Carroll, 2015). The frequency of terms related to early learning concepts indicated that providers can use these recommended books to achieve the school readiness related objectives of storytimes.

 

The term frequency analysis also pointed to opportunities for librarians to structure storytimes as emotionally positive and fun. Frequent terms included “play,” “love,” “fun,” “enjoy,” and “dance.” Based on attendees’ expectations that storytimes are a place for entertainment (Khoir et al., 2017) and joyful experiences (Cahill et al., 2020), sharing these recommended books can be part of meeting attendees’ expectations and accomplishing storytime objectives.

 

The analysis of subject terms illuminated popular genres of storytime books as they tended to include controlled heading terms of book genres/categories. The top 10 terms identified the most common genres of books shared in storytimes: fiction, juvenile, story books, rhymes, children's literature, board books, and pictorial works. We found that storytime books also included poetry, folklore, and nonfiction, although to a much smaller extent. These findings were similar to those of Kociubuk and Campana (2019), who found that less than 10% of stories shared in storytimes were in the narrative informational or non-narrative informational genres. 

 

The term-level analysis revealed both visual and sound modes of story delivery in storytimes. Terms related to visual mode, such as pictures and illustrations, were highly ranked in both the corpuses. Also, storytime books actively involved sound modes, in particular, rhymes and songs. That is, storytime books are not simply text of stories, they promote interactivity through multiple modes by utilizing both visual and sound sensory channels to deliver stories to children. The importance of visual elements as a rationale for choosing books for read-alouds is reflected in research with children, storytime providers, children’s literature experts, and early childhood education teachers (Carroll, 2015; Danko-McGhee & Slutsky, 2011; Fullerton et al., 2018; Mesmer, 2018).

 

Movements and actions are another mode to facilitate interactions and are often integrated into storytimes; however, not many action/movement terms were observed among the top terms. This may be because storytime providers plan to get the audience moving before or after reading books aloud through activities such as songs, rhymes, or fingerplays (Giles & Fresne, 2015; Kociubuk & Campana, 2019).

 

We also explored emotional aspects of storytime books based on sentiment analysis. Overall, the sentiment nature of the storytime context is very positive, showing more positive vocabulary proportionately in this study. That is, the fundamental atmosphere of storytimes would be very positive and enjoyable. In addition, we observed that storytime books included various words representing emotion. They can be sources for children to learn and understand a diverse spectrum of emotion and sentiment, thus giving providers the opportunity to contribute to children’s social-emotional learning, another objective of storytime programs.

 

This study was not without limitations. First, the analysis was done with the bibliographic records from the WorldCat database. The dataset consisted of titles, abstracts, and subject terms of the selected books, but it did not include full-text. Second, the sample of 429 books might not represent the entire book selection in storytime practices. The books included are recommended widely across the storytime community, but there is no information on which books are actually adopted and read in the field. Third, several of the terms were not interpreted correctly in the sentiment analysis as the computational tool did not catch the meaning or nuance currently in the context. For example, the word “fall” was categorized into the negative terms even though it indicated the season. These limitations indicated a need for further research, which can investigate full-text content of an enlarged sample of storytime books. Future research can also develop a sophisticated method to better analyze topical terms and sentiment based on text mining. Additionally, the complexity of language can be investigated to assess the levels of vocabulary exposed to children via storytimes.

 

Conclusion

 

In this study, we employed a text mining approach to explore topics of storytime books. We identified 429 books recommended for use in public library storytime sessions designed for preschool age children, and we collected two corpuses of text from the WorldCat database based on these books: a) titles and abstracts, and b) subject terms. To investigate the nature of theme and sentiment in storytime books, we applied multiple text mining techniques for the collected text, such as term frequency analysis, bi-grams analysis, term co-occurrences network analysis, LDA topic modeling, and sentiment analysis. The findings revealed popular topics and genres as well as emotional terms that would likely be addressed in storytimes drawing from these sources.

 

So what? Why does it matter what books librarians may choose to share with children and caregivers during storytime? According to Sipe (2008), “literature … allows us to perceive our lives, the lives of others, and our society in new ways, expanding our view of what is possible, serving as a catalyst to ignite our capacity to imagine a more just and equitable world. To understand stories and how they work is thus to possess a cognitive tool that not only allows children to become comprehensively literate, but also to achieve their full human potential” (p. 247). Our findings suggested that the books recommended for storytime programs hold the promise of preparing children for school and for life.

 

Author Contributions

 

Soohyung Joo: Conceptualization, Data collection, Data analysis, Investigation, Writing – original draft, Writing – revision and editing Erin Ingram: Conceptualization, Data collection, Literature review, Investigation, Writing – original draft, Writing – revision and editing Maria Cahill: Conceptualization, Data collection, Literature review, Investigation, Discussion, Writing – original draft, Writing – revision and editing.

 

Acknowledgements

 

This work was supported by the Institute of Museum and Library Services (Federal Award Identification Number: LG-96-17-0199-17). We are grateful to the anonymous reviewers for their constructive comments.

 

References

 

Albright, M., Delecki, K., & Hinkle S. (2009). The evolution of early literacy. Children & Libraries, 7(1), 13–18.

 

Baldwin, C. G., & Morrow, L. M. (2019). Interactive shared book reading with a narrative and an informational book: The effect of genre on parent-child reading. Preschool & Primary Education, 7(2), 81–101. https://doi.org/10.12681/ppej.20909

 

Blei, D. M. (2012). Probabilistic topic models. Communications of the ACM, 55(4), 77–84. https://doi.org/10.1145/2133806.2133826  

 

Cahill, M., Joo, S., & Campana, K. (2018). Analysis of language use in public library storytimes. Journal of Librarianship & Information Science, 52(2), 476–484. https://doi.org/10.1177/0961000618818886 

 

Cahill, M., Joo, S., Howard, M., & Walker, S. (2020). We’ve been offering it for years, but why do they come? The reasons why adults bring young children to public library storytimes. Libri, 70(4), 335–344. https://doi.org/10.1515/libri-2020-0047 

 

Campana, K. (2018). The multimodal power of storytime: Exploring an information environment for young children (Publication No. 10825573) [Doctoral dissertation, University of Washington]. ProQuest Dissertations and Theses Global.

 

Carroll, V. (2015). Preschool storytime in Auckland’s public libraries: A qualitative study of book selection practices [Master’s thesis, Victoria University of Wellington]. Victoria University of Wellington Research Archive. https://researcharchive.vuw.ac.nz/xmlui/handle/10063/4690 

 

Celano, D., & Neuman, S. B. (2001). The role of public libraries in children’s literacy development: An evaluation report. Pennsylvania Library Association.

 

Danko-McGhee, K., & Slutsky, R. (2011). Judging a book by its cover: Preschool children’s aesthetic preferences for picture books. International Journal of Education through Art, 7(2), 171–185. https://doi.org/10.1386/eta.7.2.171_1

 

Diamant-Cohen, B., & Hetrick, M. A. (2014). Transforming preschool storytime: A modern vision and a year of programs. American Library Association Neal-Schuman.

 

Eisenberg, M. B., & Small, R. V. (1993). Information-based education: An investigation of the nature and role of information attributes in education. Information Processing and Management, 29(2), 263–275. https://doi.org/10.1016/0306-4573(93)90010-B

 

Fehrenbach, L. A., Hurford, D. P., Fehrenbach, C. R., & Brannock, R. G. (1998). Developing the emergent literacy of preschool children through a library outreach program. Journal of Youth Services, 12(1), 40-45.

 

Fullerton, S. K., Schafer, G. J., Hubbard, K., McClure, E. L., Salley, L., & Ross, R. (2018). Considering quality and diversity: An analysis of read-aloud recommendations and rationales from children’s literature experts. New Review of Children's Literature and Librarianship, 24(1), 76–95. https://doi.org/10.1080/13614541.2018.1433473

 

Giles, R. M., & Fresne, J. (2015). Musical stories infusing your read-alouds with music, movement, and sound. Public Libraries, 54(5), 31–34. http://publiclibrariesonline.org/2015/11/musical-stories-infusing-your-read-alouds-with-music-movement-and-sound/

 

Ghoting, S., & Martin-Diaz, P. (2006). Early literacy storytimes @ your library: Partnering with caregivers for success. American Library Association.

 

Goulding, A., Dickie, J., & Shuker, M. J. (2017). Observing preschool storytime practices in Aotearoa New Zealand's urban public libraries. Library & Information Science Research, 39(3), 199–212. https://doi.org/10.1016/j.lisr.2017.07.005

 

Guo, Y., Sawyer, B. E., Justice, L. M., & Kaderavek, J. N. (2013). Quality of the literacy environment in inclusive early childhood special education classrooms. Journal of Early Intervention, 35(1), 40–60. https://doi.org/10.1177/1053815113500343

 

Horning, K. T. (2016, July 28). Drilling down on diversity in picture books. CCBlogC. https://ccblogc.blogspot.com/2016/07/drilling-down-on-diversity-in-picture.html 

 

International Literacy Association. (2020). What’s hot in literacy report. https://www.literacyworldwide.org/docs/default-source/resource-documents/whatshotreport_2020_final.pdf

 

Jackson, P., & Moulinier, I. (2007). Natural language processing for online applications: Text retrieval, extraction and categorization. John Benjamins.

 

Khoir, S., Du, J. T., Davison, R. M., & Koronios, A. (2017). Contributing to social capital: An investigation of Asian immigrants’ use of public library services. Library & Information Science Research, 39(1), 34-45. https://doi.org/10.1016/j.lisr.2017.01.005

 

Kociubuk, K. J., & Campana, K. (2019). Sharing stories: An exploration of genres in storytimes. Journal of Librarianship and Information Science, 52(3), 905–915. https://doi.org/10.1177/0961000619882751

 

Lennox, S. (2013). Interactive read-alouds – An avenue for enhancing children's language for thinking and understanding: A review of recent research. Early Childhood Education Journal, 41(5), 381–389. https://doi.org/10.1007/s10643-013-0578-5

 

Liu, B. (2015). Sentiment analysis: Mining opinions, sentiments, and emotions. Cambridge University Press.

 

Markowsky, J. K. (1975). Why anthropomorphism in children’s literature? Elementary English, 52(4), 460-462, 466.

 

Martens, P., Martens, R., Doyle, M. H., Loomis, J., & Aghalarov, S. (2012). Learning from picturebooks: Reading and writing multimodally in first grade. The Reading Teacher, 66(4), 285-294. https://doi.org/10.1002/TRTR.01099

 

Mesmer, H. A. (2018). Books, read-alouds, and voluntary book interactions: What do we know about centers serving three-year-olds? Literacy Research and Instruction, 57(2), 158–182. https://doi.org/10.1080/19388071.2017.1347220 

 

Miller, C., Zickuhr, K., Rainie, H., & Purcell, K. (2013). Parents, children, libraries, and reading. Pew Research Center. https://www.pewresearch.org/internet/2013/05/01/parents-children-libraries-and-reading-3/

 

Mills, J. E., Campana, K., Carlyle, A., Kotrla, B., Dresang, E. T., Urban, I. B., Capps, J. L., Metoyer, C., Feldman, E. N., Brouwer, M, & Burnett, K. (2018). Early literacy in library storytimes, part 2: A quasi-experimental study and intervention with children’s storytime providers. Library Quarterly: Information, Community, Policy, 88(2), 160–176. https://doi.org/10.1086/696581

 

Neuman, S. B., Kaefer, T., &, Pinkham, A. M. (2016). Improving low-income preschoolers’ word and word knowledge: The effects of content-rich instruction. The Elementary School Journal, 116(4), 652–674. https://doi.org/10.1086/686463

 

Pentimoti, J. M., Zucker, T. A., & Justice, L. M. (2011). What are preschool teachers reading in their classrooms? Reading Psychology, 32(3), 197–236. https://doi.org/10.1080/02702711003604484

 

Peterson, S. S. (2012). Preschool early literacy programs in Ontario public libraries. Partnership: The Canadian Journal of Library and Information Practice and Research, 7(2), 1–21. https://doi.org/10.21083/partnership.v7i2.1961

 

Price, L. H., Bradley, B. A., & Smith, J. M. (2012). A comparison of preschool teachers’ talk during storybook and information book read-alouds. Early Childhood Research Quarterly, 27(3), 426–440. https://doi.org/10.1016/j.ecresq.2012.02.003

 

Schapira, R., & Aram, D. (2019). Shared book reading at home and preschoolers’ socio-emotional competence. Early Education and Development, 31(6), 819–837. https://doi.org/10.1080/10409289.2019.1692624

 

Serafini, F. (2012). Expanding the four resources model: Reading visual and multi-modal texts. Pedagogies: An International Journal, 7(2), 150–164. https://doi.org/10.1080/1554480X.2012.656347

 

Short, K. G. (2018). What’s trending in children’s literature and why it matters. Language Arts, 95(5), 287–298. https://old.coe.arizona.edu/sites/default/files/pages/files/whats-trending-childrens-literature.pdf

 

Sipe, L. R. (1998). How picture books work: A semiotically framed theory of text-picture relationships. Children's Literature in Education, 29(2), 97–108. https://doi.org/10.1023/A:1022459009182

 

Sipe, L. (2008). Storytime: Young children’s literary understanding in the classroom. Teachers College Press.

 

Thoren, J. E. (2016). Exploring preschool teachers’ perceptions regarding the barriers to selecting literature genres and utilizing extension activities: A qualitative multiple case study (Publication No. 10240705) [Doctoral dissertation, Northcentral University]. ProQuest Dissertations and Theses Global.

 

Appendix A

Most Frequent Terms from the Titles and Abstracts Corpus 

 

Rank

Term

Freq.

Percent

Rank

Term

Freq.

Percent

1

book

104

1.35%

43

full

19

0.25%

2

anim

90

1.17%

50

back

18

0.23%

3

love

48

0.62%

50

bed

18

0.23%

3

bear

48

0.62%

50

everyon

18

0.23%

3

find

48

0.62%

50

magic

18

0.23%

3

play

48

0.62%

50

garden

18

0.23%

7

young

45

0.58%

50

old

18

0.23%

8

color

42

0.54%

56

follow

17

0.22%

9

babi

39

0.51%

56

celebr

17

0.22%

10

children

38

0.49%

56

toe

17

0.22%

11

illustr

37

0.48%

56

just

17

0.22%

12

boy

36

0.47%

56

farm

17

0.22%

12

rhyme

36

0.47%

56

egg

17

0.22%

14

text

35

0.45%

56

board

17

0.22%

15

tri

33

0.43%

56

monster

17

0.22%

15

cat

33

0.43%

56

home

17

0.22%

17

stori

32

0.41%

65

surpris

16

0.21%

18

tree

31

0.40%

65

girl

16

0.21%

19

friend

30

0.39%

65

busi

16

0.21%

19

pictur

30

0.39%

65

learn

16

0.21%

21

dog

29

0.38%

65

shoe

16

0.21%

21

time

29

0.38%

65

featur

16

0.21%

23

reader

28

0.36%

71

child

15

0.19%

23

mani

28

0.36%

71

crayon

15

0.19%

23

big

28

0.36%

71

goe

15

0.19%

26

alphabet

25

0.32%

71

keep

15

0.19%

26

page

25

0.32%

71

kiss

15

0.19%

28

simpl

24

0.31%

71

beauti

15

0.19%

28

night

24

0.31%

71

pumpkin

15

0.19%

28

world

24

0.31%

71

stop

15

0.19%

28

hat

24

0.31%

71

give

15

0.19%

32

grow

23

0.30%

71

school

15

0.19%

32

mother

23

0.30%

71

snow

15

0.19%

32

includ

23

0.30%

82

crocodil

14

0.18%

35

ten

22

0.29%

82

count

14

0.18%

35

appl

22

0.29%

82

best

14

0.18%

37

perfect

21

0.27%

82

around

14

0.18%

37

danc

21

0.27%

82

creat

14

0.18%

37

red

21

0.27%

82

enjoy

14

0.18%

40

penguin

20

0.26%

82

share

14

0.18%

40

know

20

0.26%

82

everyth

14

0.18%

40

famili

20

0.26%

82

introduc

14

0.18%

43

fun

19

0.25%

82

spider

14

0.18%

43

water

19

0.25%

82

imagin

14

0.18%

43

read

19

0.25%

82

heart

14

0.18%

43

discov

19

0.25%

82

bath

14

0.18%

43

differ

19

0.25%

82

song

14

0.18%

43

mous

19

0.25%

82

duck

14

0.18%

 

Appendix B

Most Frequent Terms from the Subject Terms

 

Rank

Term

Freq.

Percent

Rank

Term

Freq.

Percent

1

fiction

945

25.95%

39

appl

10

0.27%

2

juvenil

488

13.40%

39

domest

10

0.27%

3

stori

126

3.46%

39

farm

10

0.27%

4

book

110

3.02%

39

play

10

0.27%

5

anim

99

2.72%

46

scienc

9

0.25%

6

rhyme

81

2.22%

46

rabbit

9

0.25%

7

literatur

60

1.65%

46

american

9

0.25%

8

children

55

1.51%

46

parent

9

0.25%

9

board

50

1.37%

46

garden

9

0.25%

10

pictori

42

1.15%

46

sound

9

0.25%

11

work

41

1.13%

46

friendship

9

0.25%

12

humor

27

0.74%

46

text

9

0.25%

13

toy

23

0.63%

46

snow

9

0.25%

13

specimen

23

0.63%

55

penguin

8

0.22%

15

pictur

22

0.60%

55

fictiti

8

0.22%

16

color

21

0.58%

55

monster

8

0.22%

17

english

19

0.52%

55

nonfict

8

0.22%

17

movabl

19

0.52%

55

duck

8

0.22%

19

dog

18

0.49%

55

father

8

0.22%

20

count

17

0.47%

61

hat

7

0.19%

20

bear

17

0.47%

61

moon

7

0.19%

22

subject

16

0.44%

61

individu

7

0.19%

22

child

16

0.44%

61

behavior

7

0.19%

24

folklor

15

0.41%

61

pirat

7

0.19%

24

cat

15

0.41%

61

son

7

0.19%

26

bedtim

14

0.38%

61

magic

7

0.19%

26

poetri

14

0.38%

68

train

6

0.16%

28

life

13

0.36%

68

read

6

0.16%

28

imagin

13

0.36%

68

autumn

6

0.16%

28

school

13

0.36%

68

pumpkin

6

0.16%

28

song

13

0.36%

68

bath

6

0.16%

32

famili

12

0.33%

68

butterfli

6

0.16%

32

natur

12

0.33%

68

dinosaur

6

0.16%

34

halloween

11

0.30%

68

social

6

0.16%

34

chicken

11

0.30%

68

cloth

6

0.16%

34

alphabet

11

0.30%

68

danc

6

0.16%

34

mother

11

0.30%

68

dress

6

0.16%

34

infanc

11

0.30%

68

easter

6

0.16%

39

charact

10

0.27%

68

state

6

0.16%

39

babi

10

0.27%

68

unit

6

0.16%

39

mice

10

0.27%

68

emot

6

0.16%

 

Appendix C

Top Bi-grams: Titles and Abstracts

 

Rank

Bi-gram

Freq.

Rank

Bi-gram

Freq.

1

picture book

24

18

illustrations rhyming

5

2

rhyming text

17

18

letter alphabet

5

3

young boy

14

18

grades k-3

5

4

board book

11

18

york times

5

5

full color

10

18

one day

5

5

pete cat

10

18

new york

5

7

simple text

9

18

one one

5

8

young readers

8

30

past four

4

9

many things

7

30

times bestseller

4

9

little girl

7

30

stop kissing

4

11

young children

6

30

brown bear

4

11

ten little

6

30

letters alphabet

4

11

definitely wear

6

30

heart like

4

11

babba zarrah

6

30

like zoo

4

11

ice cream

6

30

spin web

4

11

minerva louise

6

30

text reveal

4

11

wear clothing

6

30

new way

4

18

apple pie

5

30

animals definitely

4

18

magic hat

5

30

book celebrates

4

18

white shoes

5

30

minutes past

4

18

board pages

5

30

old lady

4

18

gingerbread man

5

30

old macdonald

4

 

Appendix D

Top Bi-grams: Subject Terms

 

Rank

Bi-gram

Freq.

Rank

Bi-gram

Freq.

1

juvenile fiction

412

33

fictitious character

8

2

stories rhyme

76

33

fiction dogs

8

3

juvenile literature

59

33

juvenile nonfiction

8

4

fiction stories

51

33

fiction domestic

8

5

board books

50

38

mother child

7

6

fiction animals

42

38

apples juvenile

7

7

pictorial works

41

38

fiction color

7

8

animals fiction

38

38

fiction family

7

9

works juvenile

34

38

fiction imagination

7

10

animals juvenile

27

38

fiction bears

7

11

humorous stories

26

38

child juvenile

7

12

books specimens

21

38

dogs juvenile

7

13

picture books

20

38

fiction mother

7

14

children's stories

19

38

rabbits fiction

7

14

toy movable

19

38

mice fiction

7

14

fiction humorous

19

38

sons fiction

7

14

movable books

19

38

life fiction

7

14

fiction board

19

38

imagination fiction

7

14

fiction juvenile

19

38

color fiction

7

20

books children

14

53

clothing dress

6

21

fiction children's

13

53

united states

6

22

bedtime fiction

12

53

parent child

6

23

animals infancy

11

53

baby books

6

23

fiction picture

11

53

infancy juvenile

6

23

bears fiction

11

53

cats juvenile

6

26

domestic animals

10

53

fiction schools

6

27

counting juvenile

9

53

fiction alphabet

6

27

fiction counting

9

53

fiction friendship

6

27

chickens fiction

9

53

fiction halloween

6

27

child fiction

9

53

fiction social

6

27

dogs fiction

9

53

pirates fiction

6

27

rhyme juvenile

9

53

play fiction

6

33

children's songs

8

53

books juvenile

6