wattpad addiction research paper

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

Wattpad as a resource for literary studies. Quantitative and qualitative examples of the importance of digital social reading and readers’ comments in the margins

Federico Pianzola

1 Department of Human Sciences for Education “R. Massa”, University of Milano-Bicocca, Milan, Italy

2 School of Media, Arts and Science, Sogang University, Seoul, South Korea

Simone Rebora

3 Department of Foreign Languages and Literatures, University of Verona, Verona, Italy

4 Digital Humanities Lab, University of Basel, Basel, Switzerland

Gerhard Lauer

Associated data.

All legally sharable data files are available from the OSF database "Wattpad Titles Corpus" ( https://doi.org/10.17605/OSF.IO/5GXMN ). Data of the Teen Fiction corpus cannot be shared publicly because the copyright is held by the authors. In accordance to the European GDPR, data of the comments corpus cannot be shared publicly because this will expose underage users to possible identification. The code (R markdown file) needed to reproduce the results is available as Supporting Information ( S1 File ) and at the following link: https://doi.org/10.5281/zenodo.3375654 A guide to access the data in the same way the authors did is available at the following link: https://github.com/SimoneRebora/RSelenium_scraping_tutorial .

The end of deep reading is a commonplace in public debates, whenever societies talk about youth, books, and the digital age. In contrast to this, we show for the first time and in detail, how intensively young readers write and comment literary texts at an unprecedented scale. We present several analyses of how fiction is transmitted through the social reading platform Wattpad, one of the largest platforms for user-generated stories, including novels, fanfiction, humour, classics, and poetry. By mixed quantitative and qualitative methods and scalable reading we scrutinise texts and comments on Wattpad, what themes are preferred in 13 languages, what role does genre play for readers behaviour, and what kind of emotional engagement is prevalent when young readers share stories. Our results point out the rise of a global reading culture in youth reading besides national preferences for certain topics and genres, patterns of reading engagement, aesthetic values and social interaction. When reading Teen Fiction social-bonding (affective interaction) is prevalent, when reading Classics social-cognitive interaction (collective intelligence) is prevalent. An educational outcome suggests that readers who engage in Teen Fiction learn to read Classics and to judge books not only in direct emotional response to character’s behaviour, but focusing more on contextualised interpretation of the text.

1. Introduction

There are at least 30 million books almost always ignored when talking about how and how much people read nowadays. This bibliographic treasure is a fast and steadily growing collection of short stories, novels, and poems published on the online platform Wattpad. As a comparison, the US Library of Congress—one of the biggest libraries in the world—has around 39 million catalogued books and “nonclassified prints”, and more than 72 million manuscripts, collected over 200 years of activity [ 1 ]. Wattpad was founded in 2007. In short, there will soon be more books in computers and digital shelves than in our material literary collections.

If literary studies want to understand the reading culture of the youngest generations in the 21st century, we need to consider the phenomenon of digital social reading [ 2 , 3 ]. This is because of two reasons: first, there is a huge volume of reading happening on digital social platforms like Wattpad on a worldwide scale; secondly, users’ comments in the margins can be an extremely valuable resource for empirically studying readers’ responses [ 4 ]. On Wattpad, readers share their thoughts and emotional reactions to specific paragraphs, not just to the whole book as in the case of online reviews. This means that researchers can access detailed data about readers’ response: comments that readers write while they are reading, briefly (and spontaneously) interrupting their activity before continuing it. It is a kind of thinking-aloud or real-time data [ 5 , 6 ], which has never been available on a scale of millions of readers. They “represent the actual responses of actual readers” [ 7 ]. Notes written on e-readers like the Amazon Kindle have a similar function and are an example of how the social affordance of digital media strongly affect the reading experience: many people use Kindle notes to engage in conversations unrelated to the book in which they appear [ 4 ].

Research on reader response has started to show interest in online reading groups and book reviews to test theoretical hypotheses, complement stylistic analysis, and as a source for ethnographic inquiry [ 8 – 10 ]. A striking difference from first-generation reader response theory concerns the concept of reader: works of critics like Jauss, Iser, and Eco focused mainly on abstractions and non-testable constructions—like Erwartungshorizont —rather than on actual readers [ 11 – 13 ]. Empirically-driven reader response studies, on the other hand, bring back the actual reader at the centre of the analysis, by recognising the relevance of their responses for the study of literature [ 14 ]. With a phenomenon like Wattpad, we also have the opportunity to bring readers into computational literary studies and cultural analytics, which usually had to divert towards publishing history, when trying to reconstruct the reception of books [ 15 ]. Moreover, so far databases of reader responses did not offer a large amount of data, e.g. the UK Reading Experience Database (RED) hosts about 30,000 entries [ 16 ]: a precious trove for reading historians, but quite modest when compared to the millions of comments available on Wattpad.

An additional advantage of studying reader response on Wattpad is that the ecological validity of the experimental method is not affected by the researcher who interrupts the reading experience or manipulates the text [ 17 ]. It is a “naturalistic study of reading” [ 18 ]. In a historical perspective, glosses in ancient manuscripts and marginalia on printed books can offer a kind of insight similar to Wattpad’s comments, but their amount, cultural diversity, and circulation is very limited [ 7 , 19 ] compared to the extent of the social interactions enabled by an online medium. With about 70 million readers, Wattpad publishes in more than 50 languages and nearly 300,000 writers from 35 countries take part every year in the largest writing competition [ 20 ].

Wattpad is an example of how the circulation of literature has changed due to digital technology, leading to a literary landscape in which the transmission of stories happens through complex sociotechnical digital networks [ 21 – 24 ]. Although it has been operating since more than a decade, so far there are only a few studies exploring the social dynamics of authors and readers on Wattpad [ 25 – 31 ], and even fewer are focusing on the texts and comments [ 32 – 34 ]. A similar phenomenon—in terms of popularity among readers and autonomy from the publishing industry—is that of fanfiction. Fanfiction has received greater attention but has been mostly studied with qualitative approaches [ 35 – 38 ], which did not aim at offering a large-scale description of its variety and circulation. Among the few quantitative investigations of fanfiction online platforms, interesting results have been achieved by Milli and Bamman [ 39 ], showing how fan-fiction deprioritizes main protagonists in comparison to canonical texts, and by Yin et al. [ 40 ], who created and analysed metadata to identify the most popular fanfiction genres. However, these cultural domains are understudied by researchers in comparison to their pivotal role among reading practices in today’s society.

Our aim is broader than what has been achieved by already existing research. With this paper we both contribute to research on the social aspect of reading and advocate for the role of distant reading in contemporary literary studies. Digital methods in literary studies have been deemed effective mostly for literary history and large-scale investigations [ 41 – 44 ], although there are also examples of computer-assisted close reading [ 45 ]. We present several analyses of how quantitative methods can be used to study reader response, namely using Wattpad to understand how fiction is read nowadays and to gain unprecedented insight into readers’ interaction with texts and with other readers. Recently, Rowberry [ 46 ] has shown how even the most sophisticated and pervasive sociotechnical reading ecosystems like Amazon are currently very limited for the study of reading practices. However, our contribution presents a methodology that successfully uses big data to understand the new world of reading and writing literature.

We explored both the social and the textual dimensions for books written in various languages and spread worldwide. Our analyses are guided by three hypotheses: 1. analysing the themes of stories published on Wattpad we can highlight differences and similarities of reading interests across different languages, contributing to the study of world literature; 2. analysing users’ behaviours with different genres we can show that readers’ engagement changes according to the literary prestige of texts; 3. comparing the emotional valence of stories and comments we can provide empirical evidence to link textual features to readers’ emotional response.

To explore the research aims, we used digital mixed methods and scalable reading [ 47 – 49 ], presenting analyses of different scope, and showing how quantitative analysis can be fruitfully used to identify parts of text for subsequent qualitative analysis, including close reading of excerpts from novels. Starting from describing the themes of the stories published on Wattpad (sections 2.2 and 3.2), we then move to mapping the geographical distribution of readers (sections 2.3 and 3.3). The next step is to show the difference between born-digital stories (originally published on Wattpad) and institutionalized literature (Classics), quantitatively describing readers’ engagement with the two categories (sections 2.4 and 3.4). Then we show an original application of computational methods for text analysis to empirically study reader responses to literary texts, namely focusing on two novels. Guided by computational results, we qualitatively analysed the content of comments and the readers’ interactions in relation to the text paragraphs that prompted them (sections 2.5 and 3.5).

To show how Wattpad can be used as a resource for understanding literary dynamics in the 21st century and for investigating reader response, we designed our research combining quantitative and qualitative analysis in a mixed methods approach suitable for computer-assisted literary studies [ 49 ]. One of the main characteristics of this approach is scalable reading , that is the combination of natural language processing and hermeneutics to “zoom in and out of things and discover that different properties of phenomena are revealed by looking at them from different distances” [ 47 ].

Perhaps literary modelling is a better term than scalable reading , since it does not suggest a practice limited to reading texts and stresses more the need to make the assumptions explicit about the nature of reading (literary) texts [ 50 , 51 ]. This includes a quantitative sociological investigation of reading behaviours. However, recent advocates of literary modelling use the term mainly with respect to its application to literary history [ 43 , 44 ]. We agree that it is a good term that can help position distant reading in a constructive dialogue with close reading, but we also propose that this inclusive effort should be wider, aiming at taking into account other computational methods that work with literary texts but do not engage directly with literary history. Reader response studies can surely debate the historical dimension of reading practices, but they also synchronically investigate the emotional and cognitive aspects of reading [ 12 , 14 , 52 ].

Adopting this perspective, we tested hypotheses, explored the results generated from data analysis looking for large scale patterns, and selected case studies to close read, in order to understand how the identified patterns emerge from smaller scale phenomena. In other words, we started from the relationship between readers and text, and only then began to look for textual features. For an example of an approach starting from textual features to determine their “emotion potential”, see Jacobs [ 53 ].

Our investigation can be divided into five groups of reflections. 1. The extent of the Wattpad phenomenon, i.e. how many stories are published, what are the different languages, and how many stories per language there are. 2. The most popular themes, including differences and similarities between stories written in different languages. 3. Stories’ worldwide reception based on readers’ comments. 4. Readers’ engagement in terms of number of written comments and interaction with other readers. 5. The relations between text and reader’s emotional response, confronting Teen Fiction and Classics.

In the following subsections we explain in detail the methods we applied and the procedures for the analyses.

2.1 Corpus building

Wattpad is mainly a community of readers and emerging authors who can read and publish texts via website or mobile app. It hosts a great variety of editorial forms and genres, but there is one crucial distinction that is of interest for literary scholars: most of the stories are born-digital, i.e. published for the first time on Wattpad and often written on smartphones [ 54 ]. However, there is also a small group of literary Classics, either imported from Project Gutenberg or published directly by Wattpad [ 55 , 56 ]. We did not perform a preliminary selection of the corpus. Instead, we collected data for all the stories indexed in Wattpad’s sitemap. We then defined a smaller corpus, based on language popularity, for which to gather progressively higher density of data and analyse themes and reader response. Another reduction of scale brought us to define an even smaller corpus of Teen Fiction (TF) and Classics, categories that can be considered as representative of popular and prestigious literature respectively, according to common opinions in literary studies [ 44 , 57 ]. It is worth noting that the Classics category existed in January 2018 but it doesn’t exist anymore in the website’s frontend (November 2019), although all the books are still available on Wattpad and the category can be accessed directly typing the URL https://www.wattpad.com/stories/classics . However, the reliability of the current categorization is now compromised, displaying all sort of titles. Overall, we collected data about the number of reads, votes and comments for 1,000 titles in each category. Then we collected the number of comments to each paragraph for the 20 most read stories in each of the two categories, and the content of comments for 6 titles in each category.

To build the corpus of titles, we downloaded all links to Wattpad stories listed in the Wattpad sitemap (dated January 2018). The links are in a subsection of the sitemap, comprised between http://s3.amazonaws.com/wattpad/sitemaps/2018-01-23/sitemap1.xml and http://s3.amazonaws.com/wattpad/sitemaps/2018-01-23/sitemap144439.xml . Story titles were extracted directly from the links, by decoding all special characters, substituting dashes with single spaces, and removing unique identifiers (e.g. “ https://www.wattpad.com/story/21-about-my-character%27s ” was transformed into “about my character’s”).

The dataset for TF and Classics stories has been expanded from a previous study [ 58 ]. It was divided into three sub-corpora: general metadata, fine-grained metadata, and full text/comments. The general metadata corpus hosts a total of 2,000 titles (1,000 for category), with information on the number of reads, votes, and the tags assigned by users. The fine-grained metadata corpus hosts a total of 300 titles (145 for the Classics, 155 for TF), selected from the general metadata corpus through an “importance” score that multiplies number of reads and of votes per title. This sub-corpus provides more detailed information on the number of reads, votes, and comments per chapter, together with the distribution of comments per paragraph. Finally, the full text/comments corpus hosts a total of 12 titles, selected among the most commented for each category (6 for Classics, 6 for TF). This corpus is structured as a table, with a unique identifier for commented book, chapter, and paragraph, the full text of comment and commented paragraph, and metadata on the comment (name of the commentator, date of publication, indication if the comment is a reply to a previous comment). All data was collected between January and July 2018, through a scraping algorithm written in R language and developed on the Remote WebDriver Selenium 2.0 (based on the Docker platform). For more information on the algorithm and on the corpus structure, see Rebora and Pianzola [ 34 ].

We informed Wattpad Corp. of our intention to scrape the website for research purposes and they did not object, pointing out that we needed to consider that users own the copyright of stories and comments. After consulting with the legal offices and the data protection boards of the institutions with which we are affiliated, we proceeded to collect the data implementing procedures to protect the users’ identity. All usernames have been anonymised before analysis and the dataset of comments has not been shared publicly. Regarding TF stories’ copyright, all procedures have been carried out through servers hosted in Germany, where an exemption for research is in place to ensure the possibility of mining copyrighted works for research purposes [ 59 , 60 ].

Once the datasets were ready, we used automatic language recognition to understand the extent of the cultural diversity of stories available on Wattpad. We preliminarily compared the performances of three different R packages on a sample of around 100,000 titles: the Naïve Bayesian classifier Compact Language Detector v2 (CLD2) [ 61 ], the neural network model Compact Language Detector v3 (CLD3) [ 62 ], and the n-gram text categorizer TextCat [ 63 ]. TextCat was clearly unreliable, whereas CLD2 and CLD3 were not able to identify 24.6% and 29.1% of the titles respectively. Therefore, we decided to use CLD2 for the entire corpus. We then manually checked a random list of 200 titles whose language was not identified and, having determined that the majority of them was in English (78%), we merged them with the English subset in order not to lose 38% of data. We did a second manual check on the 15 most common languages and found that around 100,000 titles were misidentified as Galician, so we merged them with the Spanish titles. In the subset identified as Danish, however, inaccuracy was too high, mixing titles in English, Danish, German and other languages, so we excluded it, since it was only 0.3% of data. As a last step, we decided to use for the analysis of themes only the languages with more than 100,000 titles, that is the first 13 languages of the list.

2.2 Stories

Before we started mapping the corpus, we first estimated the number of fanfiction titles by counting terms that refer typically to fanfiction, like “fanfic”, “ff”, but also names of characters and public figures like “styles” for the singer Harry Styles, “bts” for the Korean band, or “potter” for Harry Potter, to get a first understanding of the prevalence of this genre in Wattpad. We checked for these terms in the first 1,000 most frequent words in English titles. Then we identified the themes of the stories published analysing the most frequent words for all of the 13 languages with most stories. Stopwords lists for each language have been created starting from those provided by the Stopword ISO repository [ 64 ]. We modified the lists in order to remove words specifically related to the writing process (“editing”, “completed”, “story”, etc.) and to keep words that we deemed useful for identifying themes. We did not stop words like “poetry”, “poems”, “romance”, even though they do not convey information about themes, because they can still help in comparing genres popularity between languages. We aggregated records for conjugated lemmas (e.g., the Russian noun “love”: “любовь“, “любви”; the Spanish verb “to love”: “amo”, “amas”), plurals, gendered adjectives and related nouns (e.g., “amigo”, “amiga”, “amigos”, “amistad”). We also manually checked the top 50 words for each language and aggregated synonyms (e.g., Portuguese “girl”: “garota”, “menina”). For aggregated words we show in the list the one with most occurrences. To interpret the results, we searched the most frequent words directly on Wattpad, in order to understand co-occurrences, clichés, and terms specific of teen fiction self-published stories. For the languages that we do not speak, we used two online dictionaries: translate.google.com and glosbe.com .

2.3 Readers

From the subcorpus of 12 stories in English, we extracted and aggregated data about 381,473 users by collecting the number of comments they wrote and which books and paragraphs they commented. A more fine-grained analysis was performed on a sample of 300,000 user profiles. When registering to Wattpad, in fact, users are offered the opportunity to add a short description of themselves and their location. However, this information is optional and fully unstructured, leaving room for creative and unreliable statements. In our sample, only 34.3% of the users provided locations, many of which are fictitious (e.g. “Hogwarts” and “Wonderland”). We converted these locations into standardized state names with the support of the Google Maps Place Autocomplete API [ 65 ]. The API produced multiple errors (e.g. “Hogwarts” was located in the United States, “Hell” in Denmark), thus we manually checked and corrected the list. At the end of the process, 61.1% of the locations were correctly georeferenced, while the remaining locations were categorized as fiction (15.8%), house (e.g. “My bed” and “At home”, 3.7%), space (e.g. “Pluto” and “Mars”, 3.7%), and unidentified (15.6%). Final results were visualized through a map created with Flourish [ 66 ].

2.4 Engagement

To measure readers’ engagement, we used data available on Wattpad but also generated new information by analysing the commenting activity of users extracted from our smallest corpus and doing network analysis.

In a previous study we showed that readers’ engagement with TF and Classics is quite different in terms of absolute reads, votes and comments [ 58 ]. Here we looked for variations in the engagement with the stories over time, analysing the distribution of reads and comments across paragraphs for Classics and TF. To control whether engagement is affected by personal differences, we also checked whether there are users who read both TF and Classics and behaved differently in their commenting activity in each category. If this is the case, then we can assume that genre has somehow affected the commenting activity. To test this hypothesis, we reduced the scale of our observation, comparing the number of comments by users who read both Classics and TF with the top commentators of Classics. We selected a first sample who wrote at least 200 comments to Classics, and a second sample who wrote at least 100 comments in each of the two categories. We then created Sankey diagrams to represent the flow of comments these users wrote for each book. The aim was to identify the reading preferences of the most active users and to test whether genre is affecting the scale of the commenting activity.

To explore how readers engaged in social interaction with other readers, we generated user-book network graphs for each category. The goal was to visualize the activity of users who read more than one book and whether users did interact with each other across different books. To understand this, we considered both books and users as nodes, aggregating the links between users and paragraphs into user-book links. We limited the text nodes to the level of books not to complicate the network graph excessively. We also generated user-chapter networks graphs for Pride and Prejudice and The Bad Boy’s Girl , which are the two most popular stories in each category. In this case, we considered chapters and users as nodes, aggregating the links between users and paragraphs into user-chapter links. The networks thus generated were still too dense to be analysed (e.g. the full TF network contained about 1.7 million links), thus we reduced their dimensions by cutting less frequent links. The final, simplified networks contained between 1,586 and 3,038 links. The R Markdown file explaining the network analysis preprocessing can be found in the supplementary information S1 File .

Visualizations were realized through the Gephi platform [ 67 ]. Among the available layouts, after testing various options, we selected OpenOrd [ 68 ] to better highlight distinctions between communities. One of the consequences of its use, is the fact that the weakest links were deleted from the graph, thus visualizations contain only strong connections. Dimensions of the nodes were determined via weighted degree of connection, i.e., by considering both the number of connected nodes and the weight of the links.

The results helped us understand whether readers commented more on the text or replied to other users’ comments, and how texts are linked/separated by the commenting activity of users.

2.5 Emotions

Further reducing the scale of our investigation, we used quantitative criteria to select excerpts from stories and comments to read closely.

The first criterion used is the number of comments: we identified and read the paragraphs eliciting most comments in each category. The goal was to show what textual elements—story events, style, or else—are more effective in prompting readers’ response and what kind of responses they elicit. After that, to understand the link between readers’ emotional response and textual features, we used sentiment analysis to search for patterns in the relation between comments and text. Since one of the guiding criteria of our investigation was popularity, we selected the 6 stories that received most reads and votes for each category. Then, for TF, we preprocessed the texts removing paragraphs with paratextual information: mainly author’s notes thanking readers, self-promoting, and giving details about story updates. Once the texts have been cleaned, we used the R package Syuzhet [ 69 ] to do sentiment analysis [ 70 ] and identify positive and negative emotional values in the stories and comments. The purpose of this analysis was to explore the relations between the two emotional arcs, looking for points of convergence and macro-differences between the emotional arc of the story and readers’ response.

Syuzhet calculates the sentiment as follows: given a sentiment dictionary that couples a selection of words with negative or positive values, Syuzhet checks if those words are present in the analysed text and sums their values. For longer narrative texts, Syuzhet offers the possibility to visualize the “emotional arcs” of plots by processing the values attributed to each sentence. Such processing works through the application of a series of filters, from the simplest (moving window) to the most advanced (Fourier transform). Syuzhet has been criticized both for its simplicity and unreliability [ 71 – 73 ]. However, in this application the accuracy of the sentiment analysis with respect to the plot development is not really important, since the focus of our observations is the relations between text’s sentiment and comments’ sentiment, as a possible signal of remarkable reader response. Moreover, we are working here on a collective scale: the sentiment values are calculated on the sum of all readers’ comments. The results express the emotional valence of a collective act of reading, something similar to what Jonathan Rose [ 11 ] imagined achieving by studying working-class autobiographies.

We adopted Syuzhet with its default features, by applying the modifications already described in Rebora and Pianzola [ 34 ]. Instead of using it to tokenize sentences, we kept the division in paragraphs already provided by Wattpad, in order to have a perfect matching between the text and the associated responses. We then plotted the emotional arcs through the moving window procedure, to keep to a minimum the distortion of the original sentiment values: the scores for each text chunk, in fact, are simply re-calculated as the average of all the surrounding scores (by default, 10% of the total). Finally—and differently from our previous study—we decided not to group all comments in a single text chunk, but to calculate the mean of the sentiment values for each comment. This modification seemed necessary, because Syuzhet does not take into account the frequency of words to calculate sentiments: the scores for longer chunks are statistically more reliable than those for shorter ones, but the impact of words repeated over multiple comments could also get lost, when collapsing them in a single sentence.

The use of Syuzhet for the analysis of Wattpad has also some intrinsic limitations. In the case of Classics, text and comments are from different historical periods, so the use of the same dictionary for sentiment values might cause the misinterpretation of some language uses. For instance, the characters’ “decency” and “civility” in talking, typical of Jane Austen’s novels, is interpreted as positive, even when the conveyed meaning is negative. In the spirit of literary modelling, comparison could be done using different personalized dictionaries to set the sentiment values of words, adapting them to the language of 19th century English novels or to the specific Wattpad lingo that is found in the comments. Just to refer to a specific case, the word “thanks” occurs in comments to Classics when readers express their gratitude to other users who wrote a comment paraphrasing the paragraph, thus helping other readers who are struggling to understand the text. Syuzhet default dictionary assigns a positive value to “thanks”, but in this context it is actually a sign that the text is difficult. If we want to assign consistent values to words expressing the same readers’ response, terms like “difficult”, “thanks,” and “understand” should have similar negative values, because they all represent issues in text comprehension. However, this would create problems in identifying positive uses like, for instance, the hypothetical comment “I understand her feelings” in response to a character’s emotions. Moreover, the use of different dictionaries for text and comments would compromise our effort to compare the sentiment values of the two arcs. All considered, we decided to use Syuzhet default lexicon, since it is based on a corpus of novels and it is thus more appropriate for application to literary studies. The R Markdown file explaining the sentiment analysis procedure can be found in the supplementary information S1 File .

Looking for patterns in this relationship between textual features and the emotional response elicited, we first checked for correlation between the two emotional arcs. Then we performed linear regressions to check whether the effect of the story’s emotional valence on the readers’ emotional response varied between the two genres.

Then, to better understand the patterns spotted and confronting the two genres, we looked at text and comments of the two most popular stories for each category, Pride and Prejudice and The Bad Boy’s Girl . We did an in-depth qualitative analysis of the parts in which we identified bigger gaps and strongest convergence between the sentiment of story and comments, and analysed the contexts in which the words determining the sentiment values for that segment of text occur. We report here only some examples, but it should be kept in mind that the sentiment value of each segment is determined by all the words to which a value is attributed, sometimes being more than 1,000. For instance, if we want to look at a segment of 10 paragraphs that has an extremely high positive value, we have to consider that Syuzhet, to determine the sentiment, takes words from the segment but also from the 5% of text preceding and following the segment (10% moving window). It may be the case that words in the surrounding text have higher values than words in the selected segment. In this case, for instance, the overall value of the segment may be determined by the combination of 50 positive words with very small values occurring in the selected segment, plus a few words with higher score that occur in the surrounding text. Therefore, interpreting the sentiment values may not always be straightforward for humans, although the sentiment dictionaries have been originally created by human annotators [ 74 , 75 ]. Luckily, this is a kind of effort literary critics are used to: close reading texts paying attention to small details. Once the sentiment outputs are generated and notable segments have been identified, the duty of the critic is to look at the evidence brought by the machine and verify with the text if the argument is solid, even when the machine-generated interpretation is based on a variety of apparently insignificant words. The interpretation of comments’ sentiment values is usually more straightforward, since they are often short and a direct expression of readers’ thoughts and emotions. However, we already mentioned that the sentiment value for comments is an aggregated datum of the individual values for all comments to one paragraph. Therefore, even with comments it may be the case that a collective negative sentiment is the expression of many negative different words occurring just once, and not the result of a particular story event that elicits clear negative responses, like “I hate this and that”.

The highest record number that we identified in the sitemaps was 136,209,997. This is probably the total number of uploads mapped since Wattpad’s opening, but we did not find a complete linear progression in the record numbers, since many items may have been deleted by users or not included in the sitemap by Wattpad. The total number of titles we collected is 31,469,668. Wattpad’s official statements report that the platform hosts 565 million uploads [ 76 ] but we could not verify this number.

Regarding linguistic and cultural diversity, Table 1 shows how many stories have been written for each language and what is their percentage in the corpus. English is by far the most represented language, amounting to around 79% of the Wattpad corpus. However, it should be noted that this is an overestimation: browsing Wattpad, it can be seen that English titles are sometimes also used for stories written in other languages. The extent of the corpus is such that even languages constituting less than 1% of the corpus, like Malaysian, still have more than 100,000 titles.

The complete lists of detected languages can be found in the data repository.

3.2 Stories

Within the English titles subcorpus, there are around 2.3 million fanfiction stories (9.3%), a rough estimation due to the complexity of telling fanfiction apart from other fiction through the analysis of titles only. Thus, the majority of stories published on Wattpad are works of fiction creating original storyworlds. With respect to popular themes, Tables ​ Tables2 2 – 4 show the most frequent words in book titles for the 13 most popular languages on Wattpad.

Words in green cells refer to genres, words in blue cells are fanfiction themes, words in yellow cells are some themes that are unique for that language or unusually high ranking with respect to other languages. The complete lists of words can be found in the data repository. Relative word frequencies are normalised per 10,000 words.

“Love” is the top word for every language, except German, for which “life” ( leben ) is more frequent. There are words that occur in almost every language, with different proportional frequency: “girl”, “life”, “friend”, “heart”, and “dream”. There are also some notable exceptions: Indonesian does not mention “girl” and “life” often; Russian, Vietnamese, and Turkish do not mention “heart”. Only Western languages mention “world”, “new” and “never”. “Without” is in the 20 most frequent words for many languages but in English does not occur within the 300 most frequent words. Only a few languages talk about “death” frequently and among them for Polish authors it is quite an important topic. Polish are also the only ones talking about “hope”, together with Filipinos. Only in Russian titles “suicides” are quite frequent. There are also unique themes, i.e. words that are quite frequent in only one language. For Tagalog (Filipino): “runaway”, which is the third most frequent word, “rich”, and a word expressing the surprise for something previously unknown, pala . For Malaysian: “memories”, “child/son/daughter”, and “boss/leader”, which can occur both to indicate a superior at work or the leader of a gang. In this second meaning, Malaysian is unique with English in having many stories about “bad boys”. Beside Malaysian, family relationships are important for Vietnamese too, as attested by the word “family”. Moreover, Vietnamese also has a supernatural positive theme, attested by “celestial”, “constellation” and “god”. Tagalog (Filipino) is the only language, together with Turkish, to use “beautiful” quite often. Italian and Spanish pay more attention than other languages to physical details, especially the “eyes”. Regarding female roles: “princess” is frequent only for English, Polish, and Russian.

In many languages, there are two groups of words that give us more information about reading preferences, those about fanfiction and those about literary genres. Inspiration for fanfiction is different worldwide, in some languages is dominant while in others does not occur at all in the 50 most frequent words. References that are extremely popular in English, like Harry Potter and Harry Styles (a singer), do not occur in other languages. The only fanfiction theme which is common across more languages is K-pop, with stories about the two most famous boy bands, “BTS” and “EXO”. In Indonesian, Korean words like saranghae and oppa may refer to fanfiction but have also been absorbed in language uses outside the genre, given the influence of Korean culture in South East Asian countries. The analysis of words also revealed preferences for genres. Collections of very brief compositions are quite popular in many languages (“thoughts”, “lines”, “sentences”, “quotes”, “words”, “remarks”). “Diary” (including “rantbook”) and “oneshot”, too, have a transnational success. The latter are stories consisting of only one chapter, whose length can vary from a few hundred words up to a few thousands. English shows a peculiar preference for the “x reader” genre, which refers to stories written in second or first person and whose protagonist is the reader [ 77 ]. In Malaysian, “roleplay” and “memes” are clearly the dominating genres. Turkish have a preference for “horror” stories. Vietnamese and Spanish readers like “danmei” and “yaoi”, that is stories about male homosexual relationship targeted to female audience [ 78 ].

3.3 Readers

Some statistics about Wattpad readers can be obtained from the website itself: more than 70 million people use it every month, 90% of whom are “either Generation Z or Millennial”, i.e. younger than 35 years old [ 79 ]. More precisely, according to a survey conducted by Wattpad with 650 English speaking users worldwide, 80% of the respondents are female between 13 and 24 years old [ 80 ]. The average reading session is 37 minutes long and 85% of the times from a mobile phone [ 79 , 81 ]. The survey also states that “Almost half of respondents bought one or two books a month”, widely preferring paperbacks to ebooks, because readers “see themselves as curators and love to feel the paper, annotate their books, and financially support authors”. Beside the fact that paper books have the advantage of not running out of battery [ 80 ].

The sample of respondents to Wattpad’s survey is similar to the one we extracted from our comments corpus based on 12 titles from TF and Classics, in terms of language spoken (English) and location (worldwide). The sample we created includes 300,000 users, with real geographical location information available only for 35,208 of them. Fig 1 is a map of users’ distribution, showing that stories written in English are read by users living in many different countries. Most readers are from countries where English is either the first language (USA, UK, Canada, etc.) or a second language (India, Philippines, Indonesia, etc.), but there are also many non-English speaking countries, each of them amounting to around 0.5% of the readership (e.g., almost all European countries, Saudi Arabia, Egypt, South Korea, Argentina, etc.). Among the 12 stories in our corpus, Classic authors are all from the UK and TF authors come from the USA, Canada, and one from South Africa, but the works by all of them have been read in all continents. There is a worldwide community of readers who read, comment, and discuss books over the years, engaging with the reflections and opinions of people living in various countries, and meeting other culture-specific responses to the same stories.

Only details for countries with more than 300 users are written. An interactive version of the map with data for all countries is available as supporting information ( S1 Fig ). Reprinted from http://flourish.studio under a CC BY license, with permission from Flourish, Inc., original copyright 2019.

3.4 Engagement

Rebora and Pianzola [ 34 ] have shown that the number of reads, votes, and comments for TF and Classics is quite different, reaching a maximum of 197 million reads and 2.5 million comments for TF, but only 7.4 million reads and 42,000 comments for Classics ( Table 5 ).

Originally published in Rebora and Pianzola [ 34 ].

Absolute values can only suggest a first idea of readers’ engagement, similar to knowing how many people bought a book, without actually knowing whether they read it or just skimmed the first pages to see whether they may like it. Indeed, even among the commentators, there is a substantial amount of readers who only wrote one comment, usually to the book title or the first paragraph. Figs ​ Figs2 2 and ​ and3 3 report the distribution of number of reads and comments across paragraphs for 20 TF and Classics titles, an information that can help us better understand how many readers actually read the whole books, commenting while they progressed in reading. It is evident that very often users begin to read Classics but abandon them after the initial chapters. This phenomenon is mirrored by the commenting activity, with the only exception of Jane Eyre . On the other hand, TF stories are able to engage readers for a long part of the plot development, in some cases even increasing the number of reads and comments in later chapters.

The red line indicates the reads, the black line indicates the comments. The scale of the Y axis is the proportion of the total of reads and comments per title, to make the plots comparable: all chapters’ values have been divided by the total sum of reads or comments in the book.

The next step was to look for quantitative variations in the commenting activity of the most active users. Among all users, 19 wrote at least 200 comments to Classics and 18 wrote at least 100 comments in both categories. The two Sankey diagrams below (Figs ​ (Figs4 4 and ​ and5) 5 ) show the different behaviour of these two samples. Users are in the centre with flow lines going left and right to book titles: TF on the left, Classics on the right. The width of the lines connecting users and books is proportional to the number of comments written on that story by the user. The numbers on the links indicate how many comments a user has written to that specific story. All data and interactive versions of the figures, in which all link values can be seen, are available as supporting information.

Users are in the centre with flow lines going left and right to book titles; TF on the left, Classics on the right. Users represented with grey tones also wrote at least 100 comments to Teen Fiction. The width of the lines connecting users and books is proportional to the number of comments written on that story by the user. The number on the links indicate the number of comments written by a user to that specific story. An interactive version of this figure, in which all values can be seen, is available as supporting information ( S2 Fig ). Reprinted from http://flourish.studio under a CC BY license, with permission from Flourish, Inc., original copyright 2019.

Users ( n = 18) are in the centre with flow lines going left and right to book titles: TF on the left, Classics on the right. Users represented with grey tones also wrote at least 200 comments to Classics. The width of the lines connecting users and books is proportional to the number of comments written on that story by the user. The number on the links indicate the number of comments written by a user to that specific story. An interactive version of this figure, in which all values can be seen, is available as supporting information ( S3 Fig ). Reprinted from http://flourish.studio under a CC BY license, with permission from Flourish, Inc., original copyright 2019.

The results of our analyses of the flow of comments show quite different results for the samples of commentators of both categories and for the sample of commentators of Classics. Fig 4 shows the flow of comments for the 19 readers who wrote at least 200 comments to Classics. There are only 3 users who also wrote at least 100 comments to TF. The attention of strong commentators of Classics is quite polarized: 84% of them did not read TF. Among the Classics, Pride and Prejudice , Jane Eyre , and Wuthering Heights were favourite reads, whereas Emma and Alice’s Adventures in Wonderland had only one very devoted commentator each (users FF and JJ, respectively).

Fig 5 shows which books, and how much, have been commented by users that were very active in reading both Classics and TF socially. Among them, only 5 readers out of 18 wrote more comments to Classics (users Q, G, H, R, P), but without showing a dominant preference for this genre. They focused mainly on three titles— Pride and Prejudice , Jane Eyre , and Wuthering Heights— only user G also commented on Emma 5 times. Overall, the majority of the 18 readers read and commented on TF more.

Beside the reading interests of top users, we also wanted to get a sense of the reads shared by a broader sample of users and to understand how important social interactions are when reading. The two users-books networks below (Figs ​ (Figs6 6 – 8 ) visualise the relative vicinity of books—based on the readers they have in common—and the level of interaction between users across different books and within each genre. Yellow and blue nodes are books (yellow for Classics, blue for TF), pink nodes are users. The size of the nodes is proportional to the number of connections the node has. The colour of the edges is yellow/blue to connect users to the books they commented and pink to connect users that replied to each other’s comments. The width of the edges is proportional to the number of interactions. Before interpreting the results, it is worth reporting some general statistics for the two categories: for TF, just 21.7% of the comments are answers to other commentators, while the percentage increases to 30.6% for the Classics.

The graph contains 1,433 nodes, 12.4% of the total in the category (11,583). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 5 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of weighted degree. We chose this option to better highlight distinctions. Due to the large dimension of the network, some nodes remained out of the figure. A full-scale figure, as well as an interactive Gephi file are available as supporting information ( S4 Fig and S2 File).

The graph contains 2,423 nodes, 0.7% of the total in the category (347,884). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 250 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of the weighted degree. We chose this option to better highlight distinctions. Due to the large dimension of the network, the quality of the image is limited. A full-scale figure, as well as an interactive Gephi file are available as supporting information ( S5 Fig and S3 File ).

The users-Classics network ( Fig 6 ) shows the isolation of Emma and Alice’s Adventures in Wonderland : they do not share many readers with other stories and there are no strong interactions between their readers. There is a central conglomerate with Pride and Prejudice , Jane Eyre , and Wuthering Heights , which share many readers with a remarkable amount of interactions ( Fig 7 ; see the pink edges connecting the pink nodes). For instance, the node USER_394 has a weighted degree of 11.73 (corresponding to just 2% of the maximum, 547.97 for Pride and Prejudice ) and distances itself from Pride and Prejudice , notwithstanding the strong connection it has with this title. Its position depends mainly on the wide range of relationships it establishes with readers of other novels (the thinner pink edges). However, it should be noted that the strongest user in this graph (USER_80, with a weighted degree of 19.02) is represented by the node placing itself immediately above Pride and Prejudice , despite having strong connections with Jane Eyre and Wuthering Heights , too.

The full graph contains 1,433 nodes, 12.4% of the total in the category (11,583). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 5 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of the weighted degree. We chose this option to better highlight distinctions.

The isolated clusters of a few pink nodes indicate that there are some readers who have stronger connections with other readers than with the text. For them the conversational aspect of social reading is more important than commenting on the text, as suggested by their distance from the yellow book nodes. The reason for this became clear reading the comments: since Classics are written in an English difficult to understand for many teenagers, and depict worlds with social norms not familiar to them, conversations in the comments are often requests for help in understanding what is happening in the story.

In the TF graph ( Fig 8 ), it can be noted that a title like I Sold Myself to the Devil (Eigenvector centrality = 1), even if less commented overall, is a stronger attractor than The Bad Boy’s Girl (Eigenvector centrality = 0.90), because it has more very active commentators than other titles. The pink clusters at the margins of the graph are very sociable users who engaged repeatedly with 1–5 other readers more than 250 times. One peculiar exception to this trend is represented by the fact that one of the strongest user nodes (USER_2411, with a weighted degree of 23.99) places itself at the border of the network, sustained almost uniquely by relations with other users (the only connection with a book is directed towards I Sold Myself to the Devil ). The relevance of this node is only slightly surpassed by that of USER_358, that connects itself with almost all the TF titles (and with no other user). Note that the node places itself very close to The Cell Phone Swap but has no connections with it. The position is determined by the relations it has with the other five titles.

After having analysed the connections within the two genres, we then looked at how users commented on two stories— Pride and Prejudice and The Bad Boy’s Girl— and engaged in social reading. Figs ​ Figs9 9 – 12 show the users-chapters network graphs for each title. In the Pride and Prejudice network graph (Figs ​ (Figs9 9 and ​ and10), 10 ), the first chapters are isolated and clearly surrounded by many users who do not have connections to other chapters. This means that they did not continue reading past the beginning of the story, as it is also confirmed by the higher number of comments to these chapters in comparison to the others (see comments distribution in Fig 3 above). The graph confirms the high user-user activity seen in the Classics network (Figs ​ (Figs6 6 and ​ and7): 7 ): readers talk between each other more than in The Bad Boy’s Girl (Figs ​ (Figs11 11 and ​ and12). 12 ). Reading the comments, we found that, in the case of Pride and Prejudice , the comments to the first chapters are often questions like “Is this a complete copy?” or “i dont understand … is this the real book?”, but also “Is this for real? Jane Austen has a wattpad account … Do you have a copy right or whatever? Just asking..”. These are paratextual questions likely to start conversations between users before they actually begin to engage with the story. There are two major clusters: on the right, until chapter 15; on the left, chapters from 16 to 61. From chapter 16 the group of commentators is more stable: the left area is dense with user-chapter connections (yellow edges) and there are more big user nodes compared to the first 14 chapters. This means that the activity of individual users intensified after the first quarter of the story. The right cluster is stretched because of the attraction of a very sociable user (USER_277, with a weighted degree of 87.27), who had many social interactions from the beginning of the book and pulls the connections of the clusters towards the top.

The full graph contains 1,067 nodes, 17% of the total in the book (6,278). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 3 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of the weighted degree. We chose this option to better highlight distinctions. Due to the large dimension of the network, the quality of the image is limited. A high-resolution figure, as well as an interactive Gephi file are available as supporting information ( S6 Fig and S4 File).

The full graph contains 1,067 nodes, 17% of the total in the book (6,278). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 3 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of the weighted degree. We chose this option to better highlight distinctions.

The full graph contains 735 nodes, 0.6% of the total in the book (130,615). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 30 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of the weighted degree. We chose this option to better highlight distinctions. Due to the large dimension of the network, the quality of the image is limited. A full-scale figure, as well as an interactive Gephi file are available as supporting information ( S7 Fig and S5 File).

The full graph contains 735 nodes, 0.6% of the total in the book (130,615). This reduction was obtained by taking into consideration only nodes that interacted between each other at least 30 times. Weighted degree of the nodes (i.e. their dimension) is visualized through a logarithmic scale, thus half diameter corresponds to 1/10 of the weighted degree. We chose this option to better highlight distinctions.

The Bad Boy’s Girl graph ( Fig 11 ) confirms two trends seen in the whole TF category ( Fig 8 ), namely that there are many very active commentators, as attested by the size of the user nodes, but also a few single users who form separate clusters of readers. The biggest of these nodes (USER_682) even reaches a weighted degree (63.69) comparable to that of the strongest chapter node (CHAP_6, 119.93) ( Fig 12 ). Users clusters are so relevant in the network that they can even attract chapters isolating them from the main cluster, as it is the case for Chapter 5. Its position is not due to low engagement (it ranks 32 nd out of 42, with 46,700 comments vs. 16,900 of the least commented chapters; see dataset), but rather to the attraction of a user who commented it a lot and had another user as a privileged discussant ( Fig 12 ).

3.5 Emotions

To understand what text features elicit the strongest emotional response, we began by identifying the paragraphs that generate more interest, in terms of number of comments. Table 6 shows the most commented paragraph in the categories Classics and TF.

Within Classics, as expected, the most commented paragraphs belong to Pride and Prejudice , since there is a big difference in the total number of readers for this book and the others of the category. Previously, Fig 2 showed that for Classics the first paragraphs are usually the most commented of the book. This is because Wattpad readers like to announce publicly that they started reading a story, often stating whether it is their first time reading it (“ftr”, first time reader) or they are rereading it (“rr”, rereading). Browsing through the comments to the incipits of Pride and Prejudice and Wuthering Heights , it became clear that the success of these two books is due to the fact that they are mentioned in the TF novel After [ 82 ], in which two characters discuss Emily Brontë’s book and, in another scene, one male character says: “Elizabeth Bennet needs to chill”. Many readers of After claim that they have been attracted to the two Classics for this reason. There are also many other readers that use Wattpad for their English class required readings. Something similar happens for Romeo and Juliet : many readers claim to either read it for school or because of the song “The Prologue” by the American singer Halsey, which begins with Shakespeare’s prologue. The second most commented paragraph of the tragedy, however, is a first example of how content, rather than external factors, can spark readers’ reaction. It is the line by Sampson—of the house of Capulet—who says that he will rape Montague women. Readers react to this claim insulting the character and explaining to other readers the meaning of the allusion “thrust his maids to the wall”, since 17th century English may not be easily understood by everybody.

The two other highly commented paragraphs in the first chapter of Pride and Prejudice are examples of a trend that later continues in the whole book. First, Readers dislike Mrs. Bennet—especially the way she talks about Elizabeth. They react directly to her behaviour by saying she is annoying, or rude, or by criticising her role as a mother, etc. Readers also elaborate more on how her opinions denounce aspects of 19th century British society, like “Back in the 1800's people prefer beauty over brains. A concept I can never agree with”. The second trend concerns Mr. Bennet’s wit and sarcasm. Readers love the way he often replies to his wife. For instance, for the paragraph “You mistake me, my dear. I have a high respect for your nerves. They are my old friends. I have heard you mention them with consideration these last twenty years at least”, the most recurrent comments are “savage” and “slay”, referring to the sharpness of Mr. Bennet’s utterance. Something similar also happens in chapter 3. The two most commented paragraphs of the chapter are the ones in which Mr. Darcy is presented for the first time and the one in which—talking to Mr. Bingley—he dismisses Elizabeth as “tolerable, but not handsome enough to tempt me ”. Readers’ comments are mostly reactions to his arrogance and rudeness in saying such things openly.

For TF, Wattpad’s social dimension plays a more important role in eliciting comments than for Classics. The number of comments to the two overall most commented paragraphs is at least four times bigger than the number of comments to other paragraphs. As many stories on Wattpad, Growing Up is a serialized novel, chapter 38 is one of the last of the book and its final paragraphs are an “author’s note” asking readers with whom of two boys they would like the female character to have a relationship. The plot denouement is still undecided and the author is giving agency to their readers by asking them what their desires are. However, beside this moment the other paragraphs of Growing Up did not elicit more than 6,000 comments each. On the contrary, She’s With me is a story able to trigger more than 4,000 comments for 50 different paragraphs (see dataset). The most commented is at the beginning of the story (ch. 1, par. 42): readers just had a glimpse of the main character’s thoughts and in this paragraph it is the first time she speaks in the story. She accidentally bumps into the school’s “bad boy” and, after his rude and offensive words, she utters a ferocious and hilarious reply that amazes all the students in the corridor. The other most commented paragraphs are a mystery about the protagonist’s past (ch.1, par. 7, 9,500 comments), a surprising behaviour of the bad boy (ch. 13, par. 106), the revelation that both boys like her and are jealous (ch. 38, par. 37), the girl’s admission that she is horny (ch. 54, par. 78, 9,400 comments), and the first kiss after a very long suspense (ch. 56, 111, 9,400 comments). Other highly commented paragraphs are My Wattpad Love scene in which one of the two boys finally says to the girl “You’re mine” (ch. 28, par. 174), and the long-awaited kiss of The Hoodie Girl (ch. 44, par. 160).

The Bad Boy’s Girl has 15 paragraphs with more than 4,000 comments, all in different chapters. Chapter 31 has two of them: one is the anticipatory title—“What It Feels Like To Get Your Heart Broken”—where readers are very emotionally expressing their concern for what they are about to read, fearing that Tessa and Cole might break up; the other one is the last paragraph of the chapter, just after the cliffhanging cheating confession, where Tessa thinks “This is what it feels like to get your heart broken and smashed right? If it is, then why do people even bother falling in love?”. It is the most tragic moment in the story and the character’s feelings are expressed with two questions. In the comments, we see readers answering these questions, as if they were addressed directly to them, or verbalizing their emotional embodied distress for the sad event. For instance, more than 500 people say they are crying (dataset), and many others share their similar personal experiences. Other highly commented paragraphs in the novel are surprising story events (ch. 24, par. 89; ch. 26, par. 102), a scene that readers already anticipated in previous chapters (ch. 22, par. 55), honest and hilarious characters’ thoughts or utterances (ch. 4, par. 56; ch. 8, par. 46; ch. 12, par. 151; ch. 30, par. 20), a reference to a very famous song (ch. 13, par. 112), a revelation about Cole’s past (ch. 14, par. 8), and a narrative ellipsis that denies to readers satisfaction after a long perceived suspense (ch. 41, par. 86).

After having browsed the most commented paragraphs, we used sentiment analysis to compare the emotional valence of the story and the comments of each of the 12 books (Figs ​ (Figs13 13 and ​ and14 14 ).

All individual graphs are available as supporting information ( S8 – S13 Figs).

All individual graphs are available as supporting information ( S14 – S19 Figs).

From a first look at the graphs we can spot a major difference between the two genres: the two arcs have much more harmonious trends for TF than for Classics. We also checked (1) correlations between the sentiment values of the two emotional arcs and (2) effects of story’s emotional valences on readers’ emotional responses. We found significant correlations for all 12 titles ( p < .001), with higher values for TF ( Table 7 ). Pride and Prejudice is the title with the lowest correlation (0.11). We also performed linear regressions to check whether the emotional effect of the stories on readers’ response varied between the two genres. All TF stories have a positive effect higher than 0.5, with She’s With Me reaching 0.99 (R 2 = 0.62). Classics have a much lower effect, the lowest being Pride and Prejudice (0.13, R 2 = 0.01) ( Table 8 ). Romeo and Juliet (0.81, R 2 = 0.75) and Jane Eyre (0.72, R 2 = 0.64) are exceptions. We will discuss the reasons behind these values in section 4.5.

To better understand the relation between the emotional valence of story and comments, we focused on Pride and Prejudice and The Bad Boy’s Girl , selecting for close reading the parts with bigger gaps between story and comments or extreme values in either of them. Fig 15 shows the emotional arcs of Pride and Prejudice with highlighted the parts that we analysed.

For Pride and Prejudice , the first notable interval regards chapters 7 to 10 (range 6.3–12.7%), from Jane’s departure to visit Mr. Bingley at Netherfield, where she falls sick, until the moment Mr. Darcy is writing a letter to her sister, under the gaze of Miss Bingley. The paragraphs have an almost neutral average sentiment (+0.3/-0.1), whereas comments peak (+0.5/+1) before a long declining slope begins. Many readers especially “like” and “love” Mr. Bennet’s frank remarks about his younger and “silliest” daughters. Then come the dialogues in chapter 8, Mr. Bingley’s and Mr. Darcy’s replies to Miss Bingley in defence of Elizabeth and Jane: “I thought Miss Elizabeth Bennet looked remarkably well when she came into the room this morning. Her dirty petticoat quite escaped my notice”, “It shows an affection for her sister that is very pleasing”, “they [Elizabeth’s eyes] were brightened by the exercise”, and “If they had uncles enough to fill all Cheapside, … it would not make them one jot less agreeable”. But also Elizabeth’s reply to Miss Bingley’s insinuations: “‘Miss Eliza Bennet’, said Miss Bingley, ‘despises cards. She is a great reader, and has no pleasure in anything else’. ‘I deserve neither such praise nor such censure’, cried Elizabeth; ‘I am not a great reader, and I have pleasure in many things’”. Readers also appreciate Elizabeth’s comment on Mr. Darcy’s too demanding idea of an “accomplished woman”: “I am no longer surprised at your knowing only six accomplished women. I rather wonder now at your knowing any”. In chapters 9 and 10, it is again Elizabeth’s way of talking and Mr. Darcy cold attitude towards the obviously flirting Miss Bingley that amuse readers. They also start mocking her, generating a series of hilarious social interactions about funny comments. About the sentiment attribution process to the comments in this interval, we have to point out that the positive value of “like” in the dictionary is misleading, since in the comments it mostly occurs as either a preposition or a conjunction, not as a verb.

The second notable interval includes the scenes from Mr. Wickham revelation of his dispute with Mr. Darcy, until the dialogue between Elizabeth and Mr. Darcy at the ball at Netherfield (ch. 16, par. 21 –ch. 18, par. 39; range 19.4–23.5%). This is the interval with the biggest sentiment gap between paragraphs’ highest positive peak (+0.5/+1) and comments’ negative slope (+0.2/-0.4). In the novel, Mr. Collins’s exaggerated courtesy, Mr. Bennet teasing him, Elizabeth’s positive impression of the good-looking Mr. Wickham, the invitation to the ball, and the politeness displayed by everybody during the event, all contribute to the positive sentiment. Looking closely at the words used for sentiment attribution and comparing them with the text, we noted that positive words like “flattered”, “civilities”, “vivacity”, “acquaintance” and “grace” are not in the Syuzhet dictionary, but they occur in the paragraphs with the highest positive values (range 21.1–22%). In the comments, “bad” occurs often to describe Mr. Wickham’s way of talking about Mr. Darcy, an act that irritates readers. “Annoying” and “bitch” refer to Mr. Wickham and the excessively eloquent Mr. Collins—who makes his first appearance here—especially when he praises Lady Catherine. “Hate” is all for Mr. Wickham, since he is in the way of the love between Elizabeth and Mr. Darcy. “Prejudice” is mentioned often because this is the part where most readers realize that Elizabeth is prone to judge very quickly, based on a first-impression. “Rude” is for Mr. and Miss Bingley, who bring the invitation to the ball to the Bennets but leave very quickly, avoiding Mrs. Bennet. Interestingly, readers express this while admitting that they would do the same, since they dislike Mrs. Bennet.

The third notable interval is close to the end of the novel, when Mr. Bingley proposes to Jane and a merry time follows in the Bennets’ house (ch. 55; range 88.1–90%). Here paragraphs have a slightly negative sentiment (-0.1/-0.2), while comments reach a negative peak (-0.3/-0.6). We noticed that the negative sentiment of the novel is affected by the occurrence of the word “cried” in many paragraphs, which is considered negative by Syuzhet but it is actually a verb commonly used by Jane Austen to refer characters’ speech. Similarly, “enough” is never used in a negative way in this segment. Moreover, the sentiment dictionary does not include words like “cordiality”, “perfections”, “satisfaction”, “satisfy”, “handsomest”, and the British-spelled “favour” and “favourite”. In the comments, we see that readers do not seem very involved with Jane’s engagement, they rather see it as the awaited step that can eventually allow Elizabeth to marry Mr. Darcy. The negative peak corresponds to Jane’s line: “I am certainly the most fortunate creature that ever existed! … Oh! Lizzy, why am I thus singled from my family, and blessed above them all! If I could but see you as happy! If there were but such another man for you!” Readers comment on this with sentences like “wait” and “don’t worry”, which increase the negative sentiment. However, in the comments to this chapter the highly negative words used by Syuzhet to calculate the mean sentiment of the segment do not occur. “No”, “bitch”, “damn” and “slay” occur abundantly in comments to the beginning of the next chapter, when Lady Catherine appears and talks to Elizabeth. As an effect of the moving window, the comments negative peak is assigned with a 1% shift.

Fig 16 shows the emotional arcs of The Bad Boy’s Girl with highlighted the parts that we analysed. The first notable interval (ch. 8, par. 34—ch. 9; range 16.2–19.8%) corresponds to the events after the party where Tessa has been assaulted. At first she is at home with Cole and the Jay arrives, then she is chatting with her best friends Beth and Megan at school (range 18–19%), also talking about her fight with Cole on the night of the party. In the next scene she is “kidnapped” into the janitor’s closet by a “stranger”. Paragraphs’ sentiment reaches here a negative peak (-1/-0.6), while comments’ sentiment is rising to positive values (-0.3/+0.5). In the novel, Tessa is conflicted: she likes Cole more and more and is trying to avoid falling for him, reminding to herself that he is a “bad guy” and that people can get “hurt” staying around him. This is the part when the conflict between the two brothers—Cole and Jay—manifests openly (“hell”, “mad”, “anger”). Interestingly, the negative peak (range 16.2–17.2%) is actually attributed to the intimate moment between Tessa and Cole, which precedes the argument between the brothers: they are playfully fighting and eventually jump in the swimming pool, where there is physical sexual tension for the first time. As it happened with Pride and Prejudice , the peak is slightly shifted due the effect of the moving window. Comments’ sentiment has a positive peak at 18.6%, the scene of the conversation with Beth and Megan. Tessa thinking about matching Beth and Cole—i.e. the “bad girl” and the “bad boy”—sparks a great discussion, with a lot of mentions of high-score words like “good” and “bad”. Moreover, readers say they like Beth and that she is a good friend, but the positive sentiment is also supported by the context, with incidental mentions of details in the preceding paragraphs that triggers excited responses: readers “love” indie music, Shania Twain, Edgar Allan Poe, and Barbies.

The second notable interval includes chapters 14 to 18 (range 38.3–45%). Paragraphs are very high (+0.6/+0.95) and comments have a peak (+0.7/+1) before they start declining. The results for this interval are quite straightforward. Tessa is becoming aware that she is falling in love with Cole and she is excited by seeing him half-naked. Cole’s “cute” side is shown: he is a “softie” with his grandmother and asks Tessa to go and visit her together. The comments reflect this positive part of the narration. Most of them are literally just the word “love” or “cute”. Many re-readers affirm that they “love” this part of the book. It is worth noting that, as already seen elsewhere, “like” is very often used either as a preposition or a conjunction, especially in the novel. Likewise, “right” is used as an adverb or in an interrogative way, either in Tessa’s interior monologue or when her brother Travis is asking about her recent life.

Even the third notable interval (ch. 32–33; range 75.6–80%) is pretty straightforward to interpret. It is a deep declining slope for both paragraphs (-0.3/-0.7) and comments (-0.7/-1). These are the chapters after Cole’s confession that he cheated on Tessa. She is in deep sorrow and this is reflected in her dramatic thoughts. He is also suffering for what he did and tries to apologize. There are also a few scenes in which there are yells and harsh words. The comments reflect both the pain and the hatred felt by the characters. “Stupid” and “bitch” are for Cole, for the girl with whom he cheated, and for Tessa because she cannot see that Cole still loves her. “Hate” is for Cole but also for Tessa because she cannot forgive him. And some readers also hate other readers because they are “haters” who do not understand how love relationships go. “Crying” has a high negative value in this interval, used both by readers, who describe themselves in such a state, and in the text, because Cole is also crying and showing his remorse.

The last notable interval is towards the end (ch. 41; range 94.4–95.7%), where the sentiment of paragraphs (+0.2/-0.1) and comments (+0.2/+0.5) starts to diverge. In the previous chapter, Tessa’s mother has expressed some doubts about her relationship with Cole. In this chapter, Tessa’s father talks to them about whether their relationship is “bad”, “wrong” or a “problem”, but also about their mother having to “stop” interfering. There is “tension” and both Tessa and her father look “angry”. From the comments we can see that the father’s speech starts a discussion about what it means to “love” someone and readers sadly admit that he is “right” and has a “good” point. On the level of narration, readers “love” the anticipating title “BAM, You’re Naked and It’s Go Time”, and they beg “please” the author for this to be the chapter in which Tessa and Cole have sex. Similar remarks appear also in comments to other chapters’ titles, whose meaning—after actually having read the chapter—is often surprisingly revealed to be different from what was expected.

4. Discussion

Reports about reading often rely on market data [ 83 , 84 ], some surveys collect data from randomized samples of the population asking directly about reading habits [ 85 , 86 ], but only a few of them take into account the changing behaviour of young readers including specific questions about reading non-standard book formats like Wattpad stories [ 87 ]. The results in such reports are consistent worldwide: reading is an activity in strong decline after childhood [ 85 , 88 – 92 ]. The magnitude and the steady increase of the activity happening on Wattpad contradicts these data. This is a sign that more appropriate methodology to report about reading needs to be developed, in order to take into account the changes of societies in which digital technology is widespread and influences reading habits.

With more than 30 million stories, Wattpad is the biggest database of user-generated fiction, surpassing both FanFiction . net and Archive of Our Own [ 93 , 94 ]. English is clearly the dominant language, representing nearly 80% of the corpus ( Table 1 ), but there are important collections of stories written in non-Western languages, as well. There are more Turkish stories than French ones, and more Vietnamese and Indonesian stories than German ones. Such a wide and diversified database of story titles, with links to access the texts and the comments, is an extremely useful resource for the investigation of both literary production worldwide and the communication and social dynamics of teenagers sharing a passion and talking about their lives. The corpus of titles is complete but the metadata subcorpus is still limited to English Classics and Teen Fiction stories and thus requires an extension that will also include other languages and genres/categories.

4.2 Stories

If “world literature” is conceived as an “interactive space” that involves the contribution of both authors and readers [ 95 ], Wattpad is the ideal context where to carry out an analysis of the most recent evolution of world literature. While globalization is gradually disrupting the closed canons of traditional literary criticism [ 96 ], recent studies have demonstrated how, in wide collaborative projects such as Wikipedia, the conception of world literature has already changed significantly [ 97 ]. With Wattpad, the scenario evolved even further: new generations of “wreaders” [ 98 ] are not simply expanding canonical narrative worlds (fanfiction and Classics are only a small portion of Wattpad stories), but are actually taking over as original creators, by sharing the most widespread international language: that of narrative invention. In our study, we have just touched the surface of the phenomenon, by demonstrating already how wide and far-reaching a categorization of Wattpad stories can be, and how much further, in-depth studies will be needed in the immediate future. For instance, to understand why English-speaking authors do not face absence (the word “without” rarely occurs in titles), or why only Western authors are concerned with the creation of new worlds and with the relation between characters and world, whereas Asian authors are focusing more on family relationships. This distinction confirms what claimed by psychological studies about the different construction of the “Self” in Western and Oriental cultures, with more references to family relationships in blogs written by Asians [ 99 , 100 ].

Analysing the most frequent words occurring in titles is a very simple way to explore the themes of stories that proved to be effective in highlighting similarities and differences between literatures across the world (Tables ​ (Tables2, 2 , ​ ,3 3 and ​ and4). 4 ). It was a big surprise to acknowledge that the main cultural influence in Wattpad’s world literature does not come from Western prestigious literary models but from a small Asian country’s music industry, South Korean K-Pop. This is true also for Western literatures, marking a significant change in comparison to past literary traditions. Reading authors’ notes, it is clear that teenagers who publish on Wattpad aspire to become famous authors, but the model they choose to follow is not that of literary elitism. They rather embrace in full the social dimension of Wattpad, choosing popular music references and engaging in continuous conversations with the readers [ 28 ]. Further investigation will need to focus on the evolution of themes year after year and on the relation between frequent themes and most popular stories.

4.3 Readers

Reader response is something that changes according to the reader’s experiential background, her language, country of origin, and culture. By mapping Wattpad’s users we can have an idea about books’ reception by looking at the cultural specificity of the readers we are studying: among the readers of a selection of English language stories, only part of them is natively anglophone ( Fig 1 ). There is a great variety of geographical locations, even though the texts are only in English. The extent of the mismatch between the language of texts and users’ location shows that there are many people reading English stories in original language, although they live in countries in which English is not the main language. Part of them may be native English-speaking migrants living in foreign countries, but there may be also people reading in English as a second language. Filipino readership is particularly big: they not only write and read in their native language ( Table 1 ), they also read a lot in English, confirming that they are one of the most relevant groups among Wattpad readers [ 101 ].

Cultural studies have diversified and relativized the perspectives we take when we investigate reader response, contesting the dominance of white voices in literary studies [ 102 , 103 ]. Wattpad is evidence that in the 21 st century, thanks to digital technology, diversity in readers’ response can also happen simultaneously on the same reading platform. It is important to acknowledge that English-speaking Wattpad readers are spread across the world and they are engaging in a social exchange with people in many different countries. This is an opportunity to meet other perspectives on books—and how they affect people’s lives—by starting a dialogue about differences and similarities in how stories are received by readers with different cultures that share the same passion. The prosocial effect that fiction has [ 104 – 106 ] is complemented by the opportunity of actual social interaction offered by digital social reading practices. Further research will need to assess whether social reading is indeed enhancing the prosocial effect of fiction.

Another important aspect to underline is that Wattpad readers’ construction of identity occurs exclusively through the text of their comments, unlike face to face reading groups, and it is affected by the public social context. Both the fact of taking place on a social-network-like platform and of participating in a reading group strengthen the performative aspect of identity construction [ 9 ]. We did not analyse reader response on Wattpad from this perspective, but comments in which users say whether they are rereaders or first-time readers is an example of this phenomenon. The influence of social space on how readers present themselves is not something happening on 21st century social reading only, it also has a historical antecedent in notes by famous readers. In fact, it has been noted that some of John Keats’s annotations in the margins of books have "the signs of performance or at least of a consciousness of shared experience about it" [ 107 ].

4.4 Engagement

By analysing how Wattpad readers engage with Classics and Teen Fiction, we highlighted that there are different kinds of readers using Wattpad, and that the social aspect of reading and genre is affecting the volume of the commenting activity. Classics have a concentration of many comments in the first chapters, whereas TF stories are widely commented until the end (Figs ​ (Figs2 2 and ​ and3). 3 ). Reading some random comments for different titles, we saw that one of the reasons is the difficulty to understand the language in which Classics are written, but further research is needed to determine how important this factor is. Indeed, the low average of comments to Classics ( Table 5 ), probably means that users abandoned the book after reading a few paragraphs or chapters, without commenting much. This is also confirmed by the network graph of Pride and Prejudice , which shows an isolated cluster of users around the first chapters ( Fig 9 ).

Readers of TF have a sustained engagement throughout the stories and the most commented chapters are sometimes in the middle or towards the end of the books ( Fig 3 ). But to have an increase in the number of comments, the number of readers should likely remain the same for the whole story (see dataset, for the number of readers for each chapter). One of the reasons this is happening for TF but not for Classics is that the former are mostly serialized novels, that is chapters are published online as soon as the author has written them, usually with a weekly pace, but the frequency of update can vary widely. This feuilleton-like publishing style likely affects readers’ engagement, since there is a suspense effect generated by paratextual elements, which cannot be present for Classics, since they were uploaded in their complete form. Interestingly, something similar happened with those Classics that were originally published in a serialised form, appearing weekly as newspaper supplements. For instance, there is written evidence of readers' response in the form of letters to the newspapers that were publishing serial novels in the 19th century in the UK [ 108 ].

Looking at the most commented paragraphs (section 3.5), we saw that peaks in the number of comments are related to story events that have a strong narrative interest, i.e. related to suspense, curiosity, or surprise effects [ 109 ]. This is in line with one of the main functions of highlighting, as shown by Rowberry [ 110 ] on a corpus of Kindle Popular Highlights: users highlight pivotal narrative moments. However, in highly commented paragraphs there is no trace of the commonplaces identified by Rowberry: moral values do not spark many discussions among Wattpad readers, although this might be due to genre-specific limitations of our corpus or to the readers' young age. More generally, highlighting and commenting seem to have two distinct functions, even though both activities are publicly shared and not meant for individual use only. Similarly, Barnett [ 4 ] noted that more prestigious books have fewer Kindle highlights and notes than popular and free books, also prompting different kinds of behaviours: the former have notes reflecting attentive and critical reading, the latter have a lot of chatting.

A representation of the flow of comments of the most active users also confirmed that genre affects the commenting activity on Wattpad. Readers that commented both Classics and TF are more active on TF titles but also read and commented Classics ( Fig 5 ). This suggests that encouraging teenagers to read TF on Wattpad might help them to develop a passion for reading Classic literature as well. Looking at the most active users, we observed their interest in Classics was only focused on Pride and Prejudice , Jane Eyre , and Wuthering Heights , suggesting that these three books have something that makes them close to TF stories. Reading the comments we realized that one of the reasons is that these Classics entered the cultural universe of Wattpad readers being mentioned in popular TF stories. Being a generic famous story, like Alice’s Adventures in Wonderland— which has been adapted and transformed in different audiovisual formats (films, video games, comics, etc.)—is not enough, as attested by the almost absolute disregard by TF readers for this title ( Fig 5 and Table 5 ). There is also another kind of readers that reads mostly Classics but did not seem to be interested in TF ( Fig 4 ). However, we cannot exclude that they engaged with other genres on Wattpad. Further research will be needed to understand what other reading preferences characterise readers of Classics on Wattpad.

Network analysis showed another difference between popular and prestigious literature: TF readers have fewer social interactions than readers of Classics. The difference is 8.9%, but the TF network graph ( Fig 8 ) exaggerates this gap because we set to 250 interactions the lower limit for nodes to be visualised. This means that all users who in total had less than 250 interactions are not visualised. Despite this deformation, the visualisation was useful to highlight that TF is able to prompt intense social interactions—i.e. clustering of users—more than Classics. The TF and Classics network graphs have to be compared considering that we had to reduce the number of nodes visualized by setting a threshold of minimum number of interactions which is different for the two categories (5 for Classics, 250 for TF). There were more user-user interactions for Classics (30.6% vs. 21.7% for TF) but readers of TF generated groups with stronger bonds, that is readers that are able to interact with other readers more than 250 times ( Fig 8 ). This happened both when users were reading the same book and across different books. Indeed, the network graphs show that the novels have a large amount of readers in common and, therefore, are kept closer in the network. Further explorations of the network may focus on the interactions within cliques, i.e. users who form groups within which everybody is connected to everybody else. This will allow to investigate what motivate readers to have intense conversations while reading.

Further investigation is also needed to clarify whether and how textual features contribute to this remarkable difference between genres. So far, this aspect has remained unexplored even by other researchers who use network analysis for the study of social reading, like Chiara Faggiolani and Lorenzo Verna, who analysed the Italian platform aNobii focusing on the relations between books and reading preferences [ 111 , 112 ]. The fact that their data are not accessible does not allow us to compare whether in different social reading platforms there are similar relationships between popular and prestigious literature.

The results of this literary modelling implemented combining different techniques brings about differences and similarities in how people engage with reading two different genres socially, but they can be used to plan further in-depth qualitative analysis of notable users. For instance, to investigate the motivations of the high sociability of USER_2411 in reading TF books ( Fig 8 ) and of USER_277 ( Fig 10 ) with Pride and Prejudice , but also the intense commenting of users E and Q (Figs ​ (Figs4 4 and ​ and5). 5 ). Within the scope of the present work, some insights about the reasons for the difference between the social activity with Classics and TF came from analysing the emotional valence of stories and readers’ response, and from reading the content of the comments and the corresponding paragraphs. We discuss this in the next section, but first we have to underline that the genre-related comparison we wanted to do is actually more complex: Teen Fiction and Classics are different not only because of the popular/prestige distinction they represent, but also for other oppositions: 19 th /21 st century, mature/teenage authors, North-American/British authors, whole-novels/serialized-novels. We considered all these differences in interpreting the results of our analyses and future research will have to be designed taking them into account.

4.5 Emotions

Looking at the most commented paragraphs, we discovered that readers like when characters react to violence, bullies, and rude people. Not really in a physical way, rather with sarcasm and words that can shut up the abuser. Similarly, candid characters trigger positive comments, whether they are outspoken or it is only their thoughts that are reported. In the latter case, it remains to be investigated whether the narratorial voice and the point of view affect readers’ response [ 113 , 114 ]. Appreciation for such characters’ actions occurs both in comments to TF and Classics: utterances by Elizabeth, Mr. Darcy, Mr. Bingley and Mr. Bennet, but also Amelia ( She’s With Me ), Beth, and Tessa’s thoughts ( The Bad Boy’s Girl ). Of course, this kind of positive reactions is possible only if there is a negative character that previously acted or talked in an offensive or annoying way: Mrs. Bennet, Mr. Collins, Miss Bingley, Lady Catherine, but also Mr. Darcy in some dialogues with Elizabeth. And for TF, the “bad boys” and the “queen bees”, Aiden ( She’s With Me ), Jay, Cole, Nicole ( The Bad Boy’s Girl ). For Classics, it is interesting to note that commenting on Wattpad is different than highlighting on a Kindle e-readers. In the latter case, Rowberry [ 110 ] has shown that readers focus on “perceived wisdom” and “affective climaxes of a narrative”. We have shown that the most commented paragraphs on Wattpad display more basic narrative effects—suspense, curiosity and surprise—not necessarily related to emotionally intense scenes. The high number of comments on the first paragraphs also happens to be an implicit sign of emotional engagement, since some readers claim that they are rereading the story, plausibly because they felt involved during their first reading. Accordingly, there is a subset of comments written by rereaders, which we did not analyse in detail. Interestingly, this is a practice similar to the cumulation of annotations written in different times in the margins of printed books [ 107 ], although comments on Wattpad seem to be mainly addressed to other readers, sometimes even warning them when there may be spoilers. We also showed one example of another relevant aspect of TF, the interactive nature of serial publishing on Wattpad—and fanfiction websites in general [ 115 ]—which activates readers’ interest to participate in the story construction, manifesting their emotional engagement with characters and expressing their desired plot development.

Thanks to sentiment analysis we were able to quantify and compare the emotional response of a large number of readers. We confirmed that in many cases the emotional valence of a story has a direct effect on reader response. This result is not trivial because the verbalisation of emotional response to a story can take many forms and it is not obvious that positive words in stories trigger readers’ positive utterances. This is an interesting phenomenon that requires further investigation to understand the relation between textual features and reader response. Previous research using sentiment analysis on online reviews showed that they are characterised by “intimacy”, inasmuch as the expression of emotions is one of the main feature of the verbalisation of reader response online [ 116 , 117 ]. However, the correlation with other kinds of textual features is still unclear.

Based on our analysis, we can claim that, among the texts we considered, Pride and Prejudice has the lowest correlation and effect probably because of the many user-user interactions (30.3% of all interactions): readers engage more in conversation and their comments reflect this shift from a direct emotional response to the text towards a discussion about the book—a more social and cognitive-oriented kind of activity. The results will have to be confirmed analysing the networks and the sentiment of other literary works, but this is already an important insight for educators. The resistance of young readers to reading novels written in an English difficult to understand can be overcome by peer learning and collective intelligence [ 37 , 118 , 119 ], as attested by the reciprocal help we witnessed in the comments, where users explain each other paragraphs that they did not understand. In this regard, one of the most remarkable examples is that of users who paraphrased the most difficult paragraphs of Pride and Prejudice in the comments, receiving the praise and thanks of many other readers. A similar form of collective intelligence also emerges from interactions on fanfiction websites, where readers collaborate in figuring out plot developments but also details about the story setting [ 120 ].

With respect to the innovative methodology that we applied, we gained some valuable feedback. Concerning the attribution of sentiment value to comments, we noted that positive response did not correspond to positive sentiment only. For instance, it is true that many comments reacting to witty characters mention “yes”, “well said”, “I’m enjoying this”, “I love her”, but there are also many enthusiastic expressions like “burn”, “slay”, “savage”, “damn”, which are considered as negative sentiment. Syuzhet default dictionary is not suited to correctly interpret neither this kind of comments nor the lingo used by the Wattpad community (e.g., “lmao” for “laugh my ass out”, “lol” for “laugh out loud”, “cinnamon roll” for “cute, sweet”, etc.). As these examples show, an important thing to remember is that negative sentiment does not necessarily mean that reader response is negative. A specific sentiment dictionary will need to be prepared in order to assign appropriate values to these expressions.

With respect to sentiment attribution to novels, Syuzhet dictionary showed some flaws when applied to a 19th century British English text, since it does not include important words or is not correctly tuned for the historical use of some words. On the other hand, although Syuzhet dictionary only includes a limited sample of words, it nevertheless identified the correct sentiment of paragraphs with the highest positive values without taking many other positive words into account. From a stylistic point of view, this result can be seen as a sign that positive words “keep each other company”. From a methodological point of view, it can be a sign that Syuzhet dictionary is good enough for the purpose of identifying sentiment values in novels, provided that the sentiment dictionary is tuned correctly beforehand. On TF novels the default dictionary works better, probably because it has been created on the basis of six 21 st century American novels, including one that can be considered Teen Fiction [ 74 ]. Further research is also needed to confirm whether the fact that for Romeo and Juliet and Jane Eyre the sentiment of the novel has a strong effect on the sentiment of comments is actually an exception, or rather it is the sentiment of the other Classics that has been misattributed. With respect to this, a comparison with other literary genres may be helpful.

Another methodological remark concerns the use of the moving window, because it can be questioned whether it is a good option. Certainly, it can help to identify trends in extended parts of text, without focusing only on the effect of single paragraphs. This is positive because reading is an activity following a linear progression and the sentiment of previous paragraphs affects how we read the following paragraphs. However, using this method we have to be careful when interpreting the results, as we showed in the case of Pride and Prejudice , where the comments’ most negative sentiment does not actually correspond to the exact part of the text to which it is linked (the R Markdown file explaining how Syuzhet calculates the relevance of single words in generating a section of the sentiment analysis graph can be found in the supplementary information S1 File ). If the aim is to more precisely identify the emotional valence of specific textual or narrative features, it may be more fruitful to interpret the raw sentiment values. In this way, the sentiment value will be calculated only based on the words occurring in the selected part of text. After the computer has precisely identified sentiment values, it will then be easier to employ human expertise to interpret what narrative or stylistic phenomena are related to them. However, in this way it is more difficult to contextualize the selected text in a trend of wider scope (e.g., declining or rising sentiment, harmonious or divergent sentiment between text and comments). It should also be considered that the moving window helps to correct the mistakes that are frequently generated when performing sentiment analysis on short texts, such as Wattpad’s paragraphs and comments. Even the most advanced techniques—combining sentiment dictionaries with syntactic parsing and deep learning—struggle to reach an acceptable efficiency [ 121 ], especially when dealing with historical texts [ 122 ].

Regarding our third hypothesis, we can affirm that comparing the sentiment of stories and comments provided empirical evidence to link textual features to readers’ emotional response. From the analysis of matching and diverging intervals we got confirmation that witty characters are very much appreciated, as already suggested by the most commented paragraphs. Moreover, having secondary characters whose personality create a contrast with that of the protagonists is a strategy that amplifies readers’ reaction. For instance, Miss Bingley with respect to Elizabeth, and the double contrast between the proud but warm-hearted Mr. Darcy and the charming but manipulative Mr. Wickham. A similar process can also work for a single character, like in the case of Tessa’s continuous internal conflict: first, between the attraction for Cole and the resistance she has because of her childhood memories; second, between the love she feels for Cole and the will to keep him at a distance after he hurt her. Comments show that readers mimic a similar conflict, especially after an extreme inversion of sentiment in the story (ch. 32): they feel for Tessa and hate Cole because he made her suffer, but they also hate Tessa for not forgiving him and empathize with Cole.

Another textual feature that can warm up emotional response is the mention of details not directly related to the plot that help to prompt identification and empathy with the characters. For instance, the incidental mention of Beth’s love for indie music or Tessa reading Edgar Allan Poe and growing up playing with Barbies (ch. 9). Lastly, titles that generate expectations which are later surprisingly disattended can be used to spark interest and discussions among readers.

5.Conclusions

We presented Wattpad as a resource for literary studies, claiming that neglecting its role in the contemporary reading landscape will inevitably lead to misleading assumptions about reading behaviours, especially of teenagers. We tested three hypotheses and brought evidence to support our arguments: 1. analysing the themes of around 30 million stories published on Wattpad we highlighted differences and similarities of reading interests across 13 different languages, contributing to the study of world literature; 2. analysing users’ behaviours in reading Classics and Teen Fiction we showed that readers’ engagement on Wattpad changes according to the literary prestige of texts; 3. comparing the sentiment of stories and comments we provided empirical evidence to link textual features to readers’ emotional response.

We need to be careful in generalizing about readers’ behaviours, since the texts and the readers on Wattpad are only a subgroup of all kind of literary texts and readers. The young age of Wattpad users and their still relative small cultural capital may affect the way they talk about books, as it also happens in other kinds of digital platforms [ 8 , 123 , 124 ]. Moreover, it should be kept in mind that we did not focus on the response of individual readers, we rather created a statistical model of a collective response to stories.

By examining reader response at scale we can set the ground for statistically valid claims about how a lot of people read. Namely, we can understand what kind of emotions readers feel towards the characters and how they react to specific narrative strategies like suspense. Some of the comments we presented in this article show that readers’ responses are often contradictory, even for short paragraphs, and there is not just one kind of reader [ 10 , 125 ]. Ultimately, the comments are a resource for different kinds of inquiries, investigating how during reading: values are negotiated; identities are constructed; authors interact with readers; metaphors are interpreted; intertextual connections are built; factors like “experiential background” [ 126 ] and “personal relevance” [ 127 ] affect reader response; etc. It has been debated whether marginalia can actually help to understand reading practices of the past [ 7 , 128 – 132 ]. In this regard, studying Wattpad comments to understand contemporary reading practices has the advantage that the retrieved information can be compared with other kinds of data within more comprehensive reading research programmes [ 58 ]. For instance, the quantitative analysis of comments can be used to identify readers that commented a lot and invite them for interviews about their reading experience.

Social reading is rising to prominence among all the various reading practices, namely thanks to the use of digital media. Looking ahead, we have to acknowledge that Wattpad readers are the generation of new readers. A generation who is nurturing a passion for reading being immersed in the use of digital media. Pessimistic claims about the effect of reading on screen compared to paper [ 133 – 136 ] will have to be reconsidered taking into account the changed social context in which millions of reading acts are taking place.

Analysing reading in the digital age can also be useful to understand social processes of reading in general: there is not one reading habit, but many ways to read. Here we showed the differences between reading Classics and Teen Fiction. Our results can be particularly useful for educators, who can find here some indications about the actual reception of Classics by teenagers and develop new approaches to foster engagement with literature.

Supporting information

Acknowledgments.

The authors would like to thank Erzsébeth Tóth-Czifra from Open Methods for her suggestions regarding text mining and copyright, and Anna Al-Damluji from Flourish for her support during the preparation of data visualisations.

Funding Statement

FP received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 792849, "Reading Literature in a Digital Culture". GL received funding from the Swiss National Science Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability

• PLoS One. 2020; 15(1): e0226708.

Decision Letter 0

26 Jul 2019

PONE-D-19-16904

Wattpad as a resource for literary studies in the 21st century: Quantitative and qualitative examples of the importance of digital social reading and readers’ comments in the margins

Dear Dr. Pianzola,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

Many thanks again for submitting your paper. The reviewers agree, in general, that the article needs to be reworked and revised before publication. Please take a look at the detailed reviewers comments, I hope they will be self-explanatory, and have the article proof-read.

We would appreciate receiving your revised manuscript by Sep 09 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

David Orrego-Carmona, Ph.D.

Academic Editor

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for clarifying that you obtained permission from Wattpad to scrape their data and that you consulted with various bodies about the legal and ethical dimension of your work. Could you please in the revision: a) state in your methods section that you did indeed obtain permission from Wattpad to scrape the data; b) in your ethics statement, explain who you consulted, how you protected users' privacy and what other measures you took to comply with ethical standards. Thank you very much in advance.

3. In order to meet the requirements for the Science of Stories collection, the Guest Editors ask that you please make the code to reproduce your analysis available in a stable, public repository (for example, Zenodo, or GitHub) or a suitable cloud computing service (such as Code Ocean) when submitting your revised manuscript. The code should include a license file and detailed readme so that someone with access to the dataset is able to reproduce your analysis using the code. We ask that you include the DOI for the repository holding your code in an updated Data Availability statement with your revised manuscript.

4. We note that  Figure(s) 1 in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright .

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a) You may seek permission from the original copyright holder of Figure(s) [#] to publish the content specifically under the CC BY 4.0 license.  

We recommend that you contact the original copyright holder with the Content Permission Form ( http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf ) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 ( http://creativecommons.org/licenses/by/4.0/ ). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b)  If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

2. Has the statistical analysis been performed appropriately and rigorously?

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #3: No

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript offers an interesting and innovative study of text and paratext on Wattpad in terms of social reading and sentiment analysis. While there have been previous studies of Wattpad, this is, to my knowledge, the most extensive, and will most certainly provoke further discussion of methods for studying Wattpad, as well as broader debates around digital reading communities and contemporary book history.

The manuscript is well structured and highly readable. There are detailed explanations of techniques used. All relevant datasets and R workbooks have been included in the manuscript. The interactive visualisations enhance the reader's comprehension of the primary argument. The majority of analysis is robust and provides sufficient evidence for the proposed conclusions. There is a good mixture of appropriate data analysis techniques to triangulate the paper's main hypotheses. The demographic data presents the greatest challenge. The underlying point of demonstrating international appeal is clear, but how can you ensure the data is reliable. For example, I note 2 users in North Korea, which is as likely to be soft trolling as genuine users.

The authors integrate a broad range of scholarship from relevant disciplines, and demonstrate they are keeping on top of even the most recent publications. At times, the scholarship is underplayed to make claims for originality unnecessarily ("We show for the first time...", p.1), but otherwise this is clearly building on previous work in an exciting way. There are some instances in the bibliography where references require a bit more information to ensure that the material can be located.

My only major concern with the manuscript comes from the ethics of data mining reader data at scale. To my mind, the challenges of researching social media at scale should be addressed in the methodology or in an ethics statement (which would presumably fall under 'human subject research'). While individuals are offered pseudonyms, this may not be sufficient protection, particularly since the authors acknowledge that "data of the comments corpus cannot be shared publicly because this will expose underage users to possible identification." These are some of the trickiest issues around studying reading in 'live' online environments, and at a bare minimum it would be useful to know that this has been reviewed by the authors' institutional ethics board, but preferably this would also include further discussion in the manuscript itself.

A few further comments that might be addressed in a further revision:

p. 14 - the claim that 'thanks' should be treated as a negative in the corpus would benefit from further evidence.

p. 41-2 the focus of the manuscript is on comments, but do likes/views across chapters confirm that most readers abandon the texts? These would be more useful metrics as they require a lower threshold for engagement.

Despite some of my concerns, the manuscript will be a much needed intervention in studies of reading online.

Reviewer #2: ‘Wattpad as a Resource for Literary Studies in the 21st Century’

Decision: Accept with major revisions

The submission is a study of internet-hosted social reading examining readerly interactions about both digital-born teen fiction and established print classics on the globally-dominant Canadian-based platform Wattpad. The cross-linguistic nature of the sample and the researchers’ proficiency in a range of languages are benefits given Wattpad’s international userbase. The study rightly acknowledges that this userbase skews young, female and English-speaking (although not necessarily native-speaking) and thus that the study results cannot be generalised to all online reading formations.

The chief limitation of the study is the way its near-exclusive focus on texts and reader comments brackets off considerations of production, and even of the wider import of the issues raised. In the submission’s late stages many avenues for future research are flagged but these involve a near-exclusive focus on analysing the existing or additional datasets, rather than broader issues of theoretical or commercial import. In short, the submission needs to look up from the text and data occasionally.

The submission lacks contextualisation of its research in reader response theory since the 1970s (classically Iser, but also landmark 1980s works such as Fish and Radway). N.B. especially recent work emerging from book history and publishing studies on readers’ uses of digital platforms (DeNel Rehberg Sedo and Danielle Fuller, Ann Steiner, Anouk Lang’s edited collection). The first of these explicitly discusses reading groups’ cross-cultural interpretations of Pride and Prejudice and hence seems highly relevant here. Lisa Nakamura’s piece for PMLA on Goodreads is also worth mentioning, especially in light of the below methodological concerns).

Overall, the theoretical coverage feels thin compared to the methods section.

The submission makes innovative use of ‘distant reading’ computer-assisted methods for contemporary text corpora. As the authors note, the methodology is almost exclusively used for historical corpora. The mixed methods approach blending quantitative big-picture and qualitative close-reading analysis is logical and well-justified.

The paper is exhaustive in its social sciences-style detailing of the project’s research methodology, corpus development and data visualization techniques. Some of the material in section 3.4 seems repetitious, adumbrating minor variations in datasets or visualisations. It created a strong impatience in this humanities-trained reader to get to the actual analysis (largely delayed until p. 38’s ‘Discussion’ section). Even then, the paper suffers from a lopsidedness, with its vast methodological apparatus generating insufficient analytical pay-off or theoretical takeaways.

One effect of the social-sciences focus on quantification and method is that methodologies that cast their net wider than the text itself are overlooked. e.g. the political economy of Wattpad and its commercial attempts to distil some archetypal story structure from its vast archive to drive print publication and screen adaptation deals (see https://www.theglobeandmail.com/business/article-what-clicks-with-fiction-readers-wattpad-helps-authors-find-out/ ). If such approaches are not going to be engaged with, they should at a minimum be registered and reasons for not exploring them provided.

Literature review:

The submission evidences a broad and multidisciplinary familiarity with extant research. The ‘lament for reading’ school (Baron et al.) makes a late entrance on p.49 but key texts are mentioned. Sven Birkerts’ is the other name commonly found in this company.

The first Simon Peter Rowberry reference [29] is incomplete. His article about Kindle Highlights would seem highly relevant but is not mentioned until p.45. See also Bronwen Thomas’ articles on fan fiction and textual interactivity, as well as her forthcoming monograph on social media and literature.

The relative paucity of readers responding to other readers (less than a third of all comments) creates the suspicion that Wattpad is a potentially interactive reading space, but that much of the actual reader commenting serves a more performative function (cf. lines1057-58). Lines 737-38 appear to reinforce this suspicion, though it is not examined in any detail. Implications of this should be teased out and tested against social reading theory relating to, for example, Goodreads.

There is no escaping the reality that most of the reader comments cited are notable mostly for their banality (e.g. character judgements about Pride and Prejudice). A later comment that a certain passage in a teen fiction ‘sparks a great discussion’ bases this estimation on the number of high-sentiment words it involves, not on the interpretive rigour or ingenuity of such reader comments. This raises the question of whether amateur readers like Wattpad’s users are more worth studying than expert readers such as literary critics, whose professional skills elucidate ‘hidden’ or underexamined innovative readings of existing texts. What exactly do the researchers aim to elucidate by examining amateur readers’ responses at scale if it is not the content of their comments? Granted, high-school English teachers might take heart that digital platforms create reader engagement with classic texts, even in antiquated forms of English, but would any teacher award high marks for the levels of readerly comprehension and analysis quoted in the submission?

The authors’ point about the ‘feuilleton-like’ serial publishing format driving commenting frequency for teen fiction is legitimate, but overlooks the fact that many 19thC classic Anglo-American works were published serially, if not the Bronte and Austen works most commented upon in Wattpad (cf. Dickens, Thackeray).

Expression:

There are pervasive minor ESL slips (e.g. lines 65-66, 232, 758-59, 927, 956, 1098-99, 1136, 1140, 1148, 1222, 1226, 1269) and lapses in noun-verb agreement (e.g. line 1064, 1100, 1156, 1214), preposition choice (line 381, 569-70, 658, 776) and possessive apostrophes (1027).

Non-grammatical sentences occasionally also occur (lines 338-39).

‘Literary studies’ is usually handled as a singular noun

In summary, the submission would benefit from lifting its gaze from the specifics of the study methodology and the Wattpad texts themselves to analyse more of the production and consumption dimensions of Wattpad, specifically the firm’s monetisation of readerly affect, and the theoretical implications (and limitations) of analysing amateur readers’ digital interactions.

17 July 2019

Reviewer #3: This is an important contribution to the scholarship. The submission performs distant and close reading of the texts in a way that expands knowledge and its methodologies. The findings are very useful and open up new knowledge about reading in the digital age.

The findings are relevant and based on good interpretation of the data.

Although I acknowledge that there is a significant body of literature referred to here, it seems to me not always the right or relevant literature. For example, there is a whole field of study around manuscript annotations that should be referred to beyond Kerby-Fulton et al (2012). Heather Jackson's Marginalia, for example. Bonnie Mak's work could be useful here too.

Similarly, reader response theory is a broad field that requires more than a single reference to Bortolussi and Dixon (2003) - an empirical study - to avoid misrepresenting the field.

I suggest Katherine Bode's work in Reading by Numbers and A World of Fiction. And, hesitatingly, I also point out my own article "Social Reading: The Kindle’s Social Highlighting Function and Emerging Reading Practices" just for interest in particular because it discusses readers commenting on Pride and Prejudice, though within the Kindle platform rather than the Wattpad platform.

Terminology and definitions

- social reading isn't limited to online spaces (though it is facilitated by them) and the two shouldn't be conflated.

- the authors use 'literary stories' but I think they mean fiction (rather than getting into the debate about what is and isn't literary)

I don't know of the use of stop words as a verbal phrase as it is used here (line 219). As far as I am aware, it is a noun phrase that would need another verb to make the sentence grammatically correct.

Further errors in expression:

Line 46 - our established literary studies (perhaps should be material literary collections?)

54-5 - readers' responses

65 - Although Wattpad is in place since more than a decade...

78 already existing research intended

There are many examples like this throughout the article and this is why I indicated expression problems at question 4 but these will be easily fixed by a proofreading.

6. PLOS authors have the option to publish the peer review history of their article ( what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy .

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes: Tully Barnett

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at gro.solp@serugif . Please note that Supporting Information files do not need this step.

Author response to Decision Letter 0

13 Sep 2019

All responses have been included in the attached file "Response to reviewers"

Submitted filename: Response_to_reviewers.docx

Decision Letter 1

PONE-D-19-16904R1

Wattpad as a resource for literary studies. Quantitative and qualitative examples of the importance of digital social reading and readers’ comments in the margins

We would appreciate receiving your revised manuscript by Dec 19 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

2. Is the manuscript technically sound, and do the data support the conclusions?

3. Has the statistical analysis been performed appropriately and rigorously?

4. Have the authors made all data underlying the findings in their manuscript fully available?

5. Is the manuscript presented in an intelligible fashion and written in standard English?

6. Review Comments to the Author

Reviewer #1: The revised manuscript is much stronger and addresses all my initial concerns or the authors' comments provide appropriate justification for not making amendments.

Reviewer #3: This article represents an important contribution to the scholarship on reading in digital environments and the use of quantitative methodologies to illuminate new aspects of the reading process and reader response. The manuscript has been improved through the revision process.

I have two further questions the authors might like to consider:

Where is the evidence that wattpad stories are *written* on mobile phones?

What role might platform recommendations have played (ie if you read this you might like to read this book) in associations between texts?

In addition, the manuscript will require some more editing for English expression. For example, the less/fewer error, reader response vs readers' response (sometimes one or the other is required)

7. PLOS authors have the option to publish the peer review history of their article ( what does this mean? ). If published, this will include your full peer review and any attached files.

Author response to Decision Letter 1

28 Nov 2019

response to the reviewers has been uploaded as a separate file

Decision Letter 2

PONE-D-19-16904R2

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/ , click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at gro.solp@gnillibrohtua .

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact gro.solp@sserpeno .

With kind regards,

Additional Editor Comments (optional):

Acceptance letter

16 Dec 2019

Dear Dr. Pianzola:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact gro.solp@sserpeno .

For any other questions or concerns, please email gro.solp@enosolp .

Thank you for submitting your work to PLOS ONE.

PLOS ONE Editorial Office Staff

on behalf of

Dr. David Orrego-Carmona

Wattpad as a resource for literary studies. Quantitative and qualitative examples of the importance of digital social reading and readers' comments in the margins

Affiliations.

• 1 Department of Human Sciences for Education "R. Massa", University of Milano-Bicocca, Milan, Italy.
• 2 School of Media, Arts and Science, Sogang University, Seoul, South Korea.
• 3 Department of Foreign Languages and Literatures, University of Verona, Verona, Italy.
• 4 Digital Humanities Lab, University of Basel, Basel, Switzerland.
• PMID: 31940372
• PMCID: PMC6961871
• DOI: 10.1371/journal.pone.0226708

The end of deep reading is a commonplace in public debates, whenever societies talk about youth, books, and the digital age. In contrast to this, we show for the first time and in detail, how intensively young readers write and comment literary texts at an unprecedented scale. We present several analyses of how fiction is transmitted through the social reading platform Wattpad, one of the largest platforms for user-generated stories, including novels, fanfiction, humour, classics, and poetry. By mixed quantitative and qualitative methods and scalable reading we scrutinise texts and comments on Wattpad, what themes are preferred in 13 languages, what role does genre play for readers behaviour, and what kind of emotional engagement is prevalent when young readers share stories. Our results point out the rise of a global reading culture in youth reading besides national preferences for certain topics and genres, patterns of reading engagement, aesthetic values and social interaction. When reading Teen Fiction social-bonding (affective interaction) is prevalent, when reading Classics social-cognitive interaction (collective intelligence) is prevalent. An educational outcome suggests that readers who engage in Teen Fiction learn to read Classics and to judge books not only in direct emotional response to character's behaviour, but focusing more on contextualised interpretation of the text.

Publication types

• Research Support, Non-U.S. Gov't
• Electrical Equipment and Supplies*
• Interpersonal Relations*

Grants and funding

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

EFFECT OF READING WATTPAD TO THE ACADEMIC PERFORMANCE CAS STUDENTS OF CARAGA STATE UNIVERSITY

Related Papers

keith legados

This study aims to assess the efficacy of phet simulation on alegria national high school students. The result shows that there is a significant difference between the controlled and experimental group.

National Center for …

Courtney Zmach

International Journal of Education & Literacy Studies [IJELS]

This study aimed to determine the in-school and out-of-school reading and writing experiences of middle school students. This study used phenomenology from qualitative research methods. All 12 participants (8 female, 4 male) were middle school grade 6 students. Various data collection techniques were employed: a survey and a semi-structured interview about the use of Wattpad, a semi-structured interview about the Turkish language subject, a semi-structured interview about the relationship between in-school and out-of-school reading and writing experiences, a diary of the students' Wattpad reading list, a diary of the students' Wattpad writing list, and the texts written by the students on Wattpad. Data collection lasted 16 weeks. The data collected from the survey questionnaire were analyzed descriptively. The data from the semi-structured interviews and diaries were analyzed using content analysis. The results indicated that most of the participants had a negative perception of in-school reading. Similar conclusions were drawn from the results in-school writing because most of the participants felt restricted and under pressure during the in-school writing process. Wattpad was considered a venue where participants can express themselves comfortably. The participants shared positive experiences characterized by willingness, diversity, meaningfulness, and entertainment. By contrast, in-school reading evoked negative experiences, such as feeling obliged, bored, and restricted, and perceiving the activity as meaningless. Out-of-school writing experiences were determined to promote internal motivation, to develop self-expression of one's anxiety and enjoyment alike, and to develop digital writing skills in out-of-school writing experiences. Overall, no relationship was found between in-school and out-of-school reading and writing experiences.

Hanae Lei Florendo

Khadar Gahayr

k'naan ali

ABSTRACT: The usage of Facebook has been a 21st century phenomenon and it is widely used for different purposes and functions. This study looks into the usage and perceived effect implications it has on the academic performance of the tertiary institutions particularly University of Hargeisa. The study was carried out in order to understand if the students’ academic performance are affected due to the time spend on Facebook Data was collected from 98 respondents by using stratified random sampling. In this study self-administered questionnaire is used to collect the required primary data, Descriptive statistical analysis technique is employed to obtain useful summary of responses. Statistical Program of Social Scientists (SPSS) version 16 tools and Excel was also employed to generate necessary information by presented in tables and graphs, regression and correlation analysis. The result showed that the usage of Facebook among student has a very weak positive effect to the academic performance on the students of the University of Hargeisa. Keywords: Facebook, academic performance University of Hargeisa.

Supattra Maraya

This paper presents a logical analysis of the characteristics of indexing and their effects on retrieval performance. It establishes the ability to ask the questions one needs to ask as the foundation of performance evaluation, and recall and discrimination as the basic quantitative performance measures for binary non-interactive retrieval systems. It then defines the characteristics of indexing that affect retrieval — namely, indexing devices, viewpoint-based and importance-based indexing exhaustivity, indexing specificity, indexing correctness, and indexing consistency — and examines in detail their effects on retrieval. It concludes that retrieval performance depends chiefly on the match between indexing and the requirements of the individual query and on the adaptation of the query formulation to the characteristics of the retrieval system, and that the ensuing complexity must be considered in the design and testing of retrieval systems.

RELATED TOPICS

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

Impact of Wattpad Reading on Filipino Junior High School’s Reading Habit and Social Values

• Prof. Rosemarie T. Rabang Palawan State University
• Dr. Teresita D. Tajolosa Palawan State University
• Dr. Leonora A. Beguina Palawan State University

The present study investigated the teenage learner’s perceptions of the impact of Wattpad reading on their lives. A total of 120 Filipino junior high school students participated in a qualitative survey of their perceptions of Wattpad reading influence on their reading habit and values. Data showed that Wattpad as a reading platform, has positively influenced the reading habit and values of the teenage users. Undeniably, a good number of the participants have gained maturity, self-reflection, motivation, creative imagination and love of reading from consistent Wattpad engagement. Still, there are negative effects for those who have developed addiction to reading such as (1) development of an irritable temperament, (2) reduced interest in socializing with loved ones and friends, (3) mismanagement of time, (4) disregard for health  and (5) profanity. Despite the disadvantages, the academe could capitalize on students’ natural interest for Wattpad. This study recommends that Wattpad be used in the classroom in developing the reading habits, reflection and writing skills of students while strengthening their sense of maturity, open-mindedness and empathy; thus be a valuable addition to the teachers’ arsenal of teaching and learning tools especially in these challenging times of the COVID-19 pandemic.

Author Biographies

Prof. rosemarie t. rabang, palawan state university.

Prof. Rosemarie T. Rabang has been in the service of Palawan State University for 31 years. She holds a Master of Arts in English Language Teaching.  She is currently the Quality Assurance Coordinator of the College of Arts and Humanities and a faculty of the Department of Languages and Communication at Palawan State University. Her interest in research focuses on reading and literature, English for Specific Purposes, and digital communication.

Dr. Leonora A. Beguina, Palawan State University

Dr. Leonora Agnes.- Beguina is a teacher in the field of social science at Palawan State Univesity,College of  Arts and Humanities . She was  graduated from Palawan State University of AB - History and also Diploma in teaching . So it is her port to be a teacher. She became a teacher in Research I and II of the Philippine Studies program and currently teacher in Social Science subjects. She  has been in the Service of Palawan State University (PSU) for nineteen years now. She is a holder of Ed. D Education Management from Palawan State University. She is currently  the Adviser of Philippines Student Organization and CAH - NBC Coordinator.

Bal, M. (2018). Reading and writing experiences of middle school students in the digital age: Wattpad sample. International Journal of Education & Literacy Studies. Australian International Academic Center. 89-100. Retrieved from http://dx.doi.org/10.7575/aiac.ijels.v.6n.2p.89

Barzo, J. (2018). The effects of wattpad reading. Retrieved from https://www.coursehero . com/file/20124558/THE-EFFECTS-OF-WATTPAD-READING-2docx-filename-UTF-8THE-EFFECTS-OF-WATTPAD-READING-2/

Bold, M. (2019). The return of the social author: Negotiating authority and influence on Wattpad. Retrieved from https://www.academia.edu/29402678/The_return_of_the_social_author_ Negotiating_authority_and_influence_on_Wattpad?email_work_card=view-paper

Burnes, R. (2015). 13 signs you’re addicted to books. Retrieved from https://www.Book bub.com/blog/2015/04/27/addicted-to-books

Carrell, P. (1984) Evidence of a formal schema in second language comprehension. Retrieved from https://doi.org/10.1111/j.1467-1770.1984.tb01005.x

Cho, K.S.; & Krashen, S. (2015). Establishing a long-term reading habit in English as a foreign language: A case study. TESOL INTERNATIONAL JOURNAL, 10 (2) 138-141.

Coban, Osman (2018). Reading Choices and the Effects of Reading Fiction: The Responses of Adolescent Readers in Turkey to Fiction and E-Fiction. PhD thesis.

Contreras, D., Gonzaga, H., Trovela, B., and Kagaoan, M. (2015). The “wattyfever”: constructs of wattpad readers on wattpad’s role in their lives. LPU Laguna Journal of Arts and Sciences, Communication Research Vol. 2 No.1 September 2015. Retrieved from http://lpulaguna. edu.ph/wp-content/uploads/2016/08/13.THE-%E2%80%9C WATTYFEVER%E2%80%9D-CONSTRUCTS-OF-WATTPAD-READERS.pdf

Cullinan, B. (2000). Independent reading and school achievement. Research Journal of the American Association of School Librarians, 3, 1-24. Retrieved from http://www.ala.org /aasl/sites/ala.org.aasl/files/content/aaslpubsandjournals/slr/vol3/SLMR_Independent Reading_V3.pdf

Dali, K. (2014). The dangers of reading. Retrieved from https://www.ebscohost . com/novelist/novelist-special/on-the-dangers-of-reading

Day, R. (2018). “Extensive reading.” The TESOL Encyclopedia of English Language Teaching.

De Guzman, J. (2014). “Why are teenage girls addicted to wattpad stories.” Retrieved from https://prezi.com/3tlogt-s3m9j/why-does-teenage-girls-addicted-to-wattpad-stories/

Department of Education. (2019). Statement on the Philippines ranking in the 2018 PISA results. Retrieved from www.deped.gov.ph/2019/12/04

Fairbrother, H. (2019). The impacts of daily reading on academic achievement, retrieved from https://ukedchat.com/2014/12/17/the-impacts-of-daily-reading-on-academic-achievement-by-mrshollyenglish

Florendo, H. (2014). Click, Swipe, and Read: Redefining the Wonder of Books, (A Wattpad Case Study). Retrieved from https://www.academia.edu/10928180/ Click_ Swipe_ and_Read_Redefining_the_Wonder_of_Books_A_Wattpad_Case_Study_?email_work_card=view-paper

Freebody, P. and Anderson, R. (1983). Effects of vocabulary difficulty, text cohesion, and schema availability on reading comprehension. Retrieved from https://www.ideals.illinois.edu/bitstream/handle/2142/17997/ctrstreadtechrepv01981i00225_opt.pdf

Fussman, T. ( 2015). 7 Signs your reading addiction has reached the intervention level. Retrieved from https://www.theodysseyonline.com/7-signs-reading-addiction

Gabinete, G. (2014). Wattpad addiction. Retrieved on February 14, 2019 from https:// ggggabinete.blogspot.com/2014/03/wattpad-.

Galda, L. (2001). Response to literature as a cultural activity. Reading Research Quarterly Vol. 36, No. 1 January/February/March 2001 ©2001 International Reading Association (pp. 64–73). Retrieved from https://ila.onlinelibrary.wiley.com/doi/pdf/10.1598/RRQ . 36.1.4

Garro, L. (2014). The importance of reading fiction. Abstract. Master of Science in Library Science and Information Services. University of Central Missour. Retrieved from https://docplayer.net/62870126-The-importance-of-reading-fiction-lisa-garro.html

Gottschall, J. (2018). The storytelling animal: how stories make us human. Retrieved from https://www.psychologytoday.com/us/blog/the-storytelling-animal/201205/fiction-addiction

Harris, J. (2016). Is fiction an addiction? Bookriot. Retrieved from https:/ / bookriot.com/2016/08/09/is-fiction-an-addiction/

Harvey, J. (2019). Giving Up an Addiction to "Trashy Novels". Retrieved from

https://www.rc.org/publication/present_time/pt120/pt120_09_km

Horvath, T., Misra, K., Epner, A. and Cooper, G. (2015). The Psychology of Addiction and Recovery. Retrieved from https://www.cascadementalhealth . org/poc/view_doc.php?type= doc&id=48407&cn=1408

Hsu, Christine. (2012). Psychologists discover how people subconsciously become their favorite fictional characters. Medical Daily. Web. 3

Hughes Hassell, S. and Rodge, P. (2007). The Leisure Reading Habits of Urban Adolescents. Journal of Adolescent & Adult Literacy 51.1: 22-33.

Fussman, T. (2015). 7 Signs your reading addiction has reached the intervention level. Retrieved from https://www.theodysseyonline.com/7-signs-reading-addiction

Jacobs, T. (2011). Study: Reading makes us better thinkers. Salon. 15 June 2013. Web. 3.

Jasareno, R., Podolig, P., and Shemfe, M. (2013). #BalitangNetizen: Constructs of citizen journalists on their role in the society.

Kaya, Z.; Kaya, S.; Sagun, A.; & Koc, K. S. (2019). The analysis of risk behavior tendencies of teenagers according to the use of Wattpad and some socio-demographic variables. International Journal of Progressive Education. 15 (6) INASED.

Kidd, D.C., & Castano, E. (2013). Reading literacy fiction improves theory of mind. Science, 342(6156), 377-380. doi: 10.1126/science.1239918

Krashen, S. (1993). The Power of Reading. Englewood, Col.: Libraries Unlimited, Inc.

Martin, L. (2015). Reading addict: the scientific effects of a damn good book on your brain. Retrieved from https://www.elitedaily.com/life/happens-to-brain-when-reading/9377

McKenna, M., Conradi, K., Lawrence, C., Jang, B., and J. Patrick Meyer. (2012). Reading Attitudes of Middle School Students: Results of a U.S. Survey. Reading Research Quarterly 47.3 (2012): 283-306.

Melinis, A. (2011). The Effects of Electronic Books on the Reading Experience of First Grade Students St. John Fisher College. St. John Fisher College Fisher Digital Publications Education Masters School of Education 12-2011

Mingle, J. and Adams, M. ( 2015 ) Social Media Network Participation and Academic Performance in Senior High Schools in Ghana. Library Philosophy and Practice (e-journal). 1286. Retrieved from https://digitalcommons.unl.edu/libphilprac/1286)

Muchtar, N. (2015). The effects of multi-media technology on electrical engineering students’ reading achievement. The Asian ESP Journal. 86-93.

Murphy, P. ( 2012). Your brain on fiction. Retrieved from https://www. nytimes.com/2012/03/18/opinion/sunday/the-neuroscience-of-your-brain-on-fiction.html?pagewanted=1&_r=3&sq=annie%20murphy%20paul&st=cse&

Murphy, P. A. (2013). Reading literature makes us smarter and nicer. Time.com. Retrieved from http://ideas.time.com/2013/06/03/why-we-should-read-literature/

National Center for Educational Statistics (2006). Comparative indicators of education in the United States and other G-8 countries. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp? pubid=2007006

Quora. (2018). Is wattpad addictive. Retrieved from https:www.quora.com/

Seiter, C. (2019). The surprising power of reading fiction: 9 ways it make us happier and more creative. Retrieved from https://open.buffer.com/reading-fiction/

Robson, D. (May 3, 2018). Our fiction addicition: why we need stories. Retrieved from http://www.bbc.com/culture/story/20180503-our-fiction-addiction-why-humans-need-stories

Sosa, J. and Culp, M. (Dec. 1993). The Influence of Nonfiction on Attitudes, Values and Behavior. The English Journal. Vol. 82, No.8, pp. 60-64. National Council of Teachers of English

Thomas, K. (2015). Brief history of Wattpad. Retrieved from https:///techvibes. com/2015 /10/01/a-brief-history-of-wattpad

Totanes, B.; Amarles, A. M. (2020). Critical scaffolding method: A proposed instructional strategy on critical literacy in reading English literary texts. The Asian EFL Journal. (5) 38-61.

Whitten, C. (2016). The impact of Pleasure Reading on Academic Success. The Journal of Multidisciplinary Graduate Research 2016, Volume 2, Article 4, pp. 48-64.

How to Cite

• Endnote/Zotero/Mendeley (RIS)

Copyright (c) 2022 Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution (CC-BY) 4.0 License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.

This work is licensed under a Creative Commons Attribution 4.0 International License .

Information

• For Readers
• For Authors
• For Librarians

Current Issue

Make a submission.

MJSELT uses Turnitin to check the similarity index, Reciteworks.com for in-text and references checking, and Grammarly for grammar and other surface-level features checking. Turnitin can detect AI-generated content, and we only allow 10% similarity index.

• Publication Ethics
• Terms of Use

(415) 520-5258

1(888)499-5521

1(888)814-4206

Some attractive features that you will get with our write essay service

Grab these brilliant features with the best essay writing service of PenMyPaper. With our service, not the quality but the quantity of the draft will be thoroughly under check, and you will be able to get hold of good grades effortlessly. So, hurry up and connect with the essay writer for me now to write.

The experts well detail out the effect relationship between the two given subjects and underline the importance of such a relationship in your writing. Our cheap essay writer service is a lot helpful in making such a write-up a brilliant one.

Margurite J. Perez

The essay writers who will write an essay for me have been in this domain for years and know the consequences that you will face if the draft is found to have plagiarism. Thus, they take notes and then put the information in their own words for the draft. To be double sure about this entire thing, your final draft is being analyzed through anti-plagiarism software, Turnitin. If any sign of plagiarism is detected, immediately the changes will be made. You can get the Turnitin report from the writer on request along with the final deliverable.

Definitely! It's not a matter of "yes you can", but a matter of "yes, you should". Chatting with professional paper writers through a one-on-one encrypted chat allows them to express their views on how the assignment should turn out and share their feedback. Be on the same page with your writer!

Essay Help Services – Sharing Educational Integrity

Hire an expert from our writing services to learn from and ace your next task. We are your one-stop-shop for academic success.

Gain recognition with the help of my essay writer

Generally, our writers, who will write my essay for me, have the responsibility to show their determination in writing the essay for you, but there is more they can do. They can ease your admission process for higher education and write various personal statements, cover letters, admission write-up, and many more. Brilliant drafts for your business studies course, ranging from market analysis to business proposal, can also be done by them. Be it any kind of a draft- the experts have the potential to dig in deep before writing. Doing ‘my draft’ with the utmost efficiency is what matters to us the most.

Allene W. Leflore

Need an essay writer for me? Connect now!

Feeling tired to write drafts on your own or you do not have ample ideas to write with? Be it anything, our writers are here to assist you with the best essay writing service. With our service, you will save a lot of time and get recognition for the academic assignments you are given to write. This will give you ample time to relax as well. Let our experts write for you. With their years of experience in this domain and the knowledge from higher levels of education, the experts can do brilliant essay writing even with strict deadlines. They will get you remarkable remarks on the standard of the academic draft that you will write with us.

Finished Papers

Frequently Asked Questions

Essay help services – sharing educational integrity.

Hire an expert from our writing services to learn from and ace your next task. We are your one-stop-shop for academic success.

Artikel & Berita

Write my essay for me.

• How it Works
• Top Writers

Why choose us

Essay writing help has this amazing ability to save a student’s evening. For example, instead of sitting at home or in a college library the whole evening through, you can buy an essay instead, which takes less than one minute, and save an evening or more. A top grade for homework will come as a pleasant bonus! Here’s what you have to do to have a new 100% custom essay written for you by an expert.

To get the online essay writing service, you have to first provide us with the details regarding your research paper. So visit the order form and tell us a paper type, academic level, subject, topic, number and names of sources, as well as the deadline. Also, don’t forget to select additional services designed to improve your online customer experience with our essay platform.

Once all the form fields are filled, submit the order form that will redirect you to a secure checkout page. See if all the order details were entered correctly and make a payment. Just as payment is through, your mission is complete. The rest is on us!

Enjoy your time, while an online essay writer will be doing your homework. When the deadline comes, you’ll get a notification that your order is complete. Log in to your Customer Area on our site and download the file with your essay. Simply enter your name on the title page on any text editor and you’re good to hand it in. If you need revisions, activate a free 14-30-day revision period. We’ll revise the work and do our best to meet your requirements this time.

Finished Papers

Our Team of Professional Essay Writers

As we are an honest and well-paying essay writer service, writers come flying our way. Nonetheless, in the writers' community, we are known for our strict selection process. You as a client can be sure that you will be working with the best paper writer in the game no matter your subject or the difficulty of the task as all our writers go through testing and have their degrees checked. Only 3% of all applicants are accepted to work with us and even these 3% have a training program and a two-month trial period ahead. We value our reputation and only hire true experts with years of experience in academic writing behind their backs. Nonetheless, being a professional writers service has its challenges. For example, as our employer expectations are high, not all writers can handle the challenge of creating zero-plagiarism essay writing content in a short time frame, so as leading writing services we must keep everything in control.

We do not tolerate any form of plagiarism and use modern software to detect any form of it

• Paraphrasing
• Research Paper
• Research Proposal
• Scholarship Essay
• Speech Presentation
• Statistics Project
• Thesis Proposal

Who is an essay writer? 3 types of essay writers

Andre Cardoso