One hundred tweets from library land: A case study of RMIT University Library (academic library) and State Library of Victoria (public state library) in Australia

Abstract

Twitter is used by different library types to communicate and engage with their community. This case study focuses on content choices in tweets and the level of engagement generated, in the context of two different types of library. The current study attempts to examine the contents of Library Twitter account of two libraries (academic and public) with the aim of evaluating in a comparative mode, their themes and levels of user engagement. This research used a mixed method research approach. For quantitative approach, this study analyzed 100 Tweets from each library, i.e. RMIT University Library and State Library of Victoria. Also, the study analyzed a number of tweets and levels of engagement by recording numbers of likes, replies and retweets. The qualitative aspect analyzed each Tweet to determine engagement level and type of content shared by each library. The results demonstrate that the public and academic libraries publish different types of content. The varied nature, audience and mandate of each library appear to influence the focus of their tweets. Also, the results show that the level of engagement is a factor of the numbers of tweets in any theme. The limitation of this study is the data set contained only 100 tweets for each library. Moreover, the number of users and registered library accounts had probably increased since the study was conducted.

Keywords

Higher- and lower-order themes level of engagement libraries’ tweets social media in libraries thematic analysis Twitter

Introduction

Many libraries use Twitter to broadcast information to their stakeholders. They update their Twitter accounts with a range of information, both original and retweeted from other Twitter users. The information contained in tweets ranges across different types of content such as events, new resources, resolving user inquiries and opening hours.

Twitter brought lots of challenges and opportunities for libraries to exchange and disseminate information. It has enabled libraries to interact with their users (Kenchakkanavar, 2015). It can be said that it brought new ways and means for libraries to cooperate with their community. As a result, most libraries in Australia have their accounts on Twitter. Al-Daihani and AlAwadhi (2015) define Twitter as a Web 2.0 platform and microblogging tool freely available and accessible online for providing users a space of 140-characters, called ‘tweets’, to share diverse content including links to images and videos.

Studies revealed that both academic and public libraries recognize the power of Twitter as a useful channel and a valuable way of communication and recommend libraries make an effort to produce a tweet at least once per day. Twitter enhances the functions of libraries and attracts their users. Additionally, studies find that Twitter has become imperative for libraries, to enhance their services and improve communications with their users (Aharony, 2010; Shiri and Rathi, 2013).

Despite the large number of studies that have been conducted in the use of Twitter and social media in libraries and tried to explore the content shared through these channels, there are no studies that have explored the use of social media in Australian libraries nor studies that have tried to compare the engagement with these channels in different library types. Hence, this research proposes to study the content of tweets from two libraries, categorize the types of information they contain and analyze the level of engagement. This is interesting because libraries use Twitter as a way of engaging with their audience, in an environment where social media form an increasingly important method of communication. The two library types (academic and public state library) have different mandates, and it is essential to see if this is reflected in their Twitter content.

The research seeks to answer the research questions: What information types are contained in tweets, what level of engagement do they generate, from an academic library and a public library, and how do they compare.

The study will focus on two libraries: public and academic library. The first library is RMIT University. The library has one Twitter account for all RMIT Library sites, with the Twitter username @library_rmit. The date of joining Twitter was 8 February 2010. The second library is the State Library Victoria (SLV). The library has a Twitter presence with the username @Library_Vic. The date of joining Twitter was 23 February 2010. The Twitter accounts of both libraries are publicly open and accessible, they can be freely followed on Twitter and their tweet content is visible. Their tweets are available over time, back to the date they first joined Twitter. Therefore, there are no logistical or access barriers to obtaining the data.

This study will contribute to understanding the use of social media in Australian libraries and will reveal the different approach that academic libraries and public libraries utilize to engage with their users in the social web. Additionally, the study will contribute to identifying the kinds of information that libraries share through the social web.

Literature review

Social media use in general and their use in libraries of various types have been widely discussed in the literature. A wide range of social media platforms has been covered, with the most common being Twitter and Facebook (Kenchakkanavar, 2015). Joosten (2012) states that social media is a broadly used term to describe any number of technological systems related to collaboration, community, interaction, and the creation and sharing of content in real time. According to Kietzmann et al. (2011: 241) ‘Social media employ mobile and web-based technologies to create highly interactive platforms via which individuals and communities share, cocreate, discuss and modify user-generated content.’ Similarly, Kaplan and Haenlein (2010: 61) defined social media as ‘a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content’. The previous definitions are similar in that the essence of social media is that these websites are being used to generate and share content among their users.

Shiri and Rathi (2013) and Taha et al. (2017) argue that the rise of social media, as an integral part of people’s daily lives, led to libraries embracing them in order to engage, connect and communicate with their patrons. While not discounting this argument, some studies have focused on Twitter, proposing that its microblogging nature and sharing features make it well suited to library communications (Kenchakkanavar, 2015; Palmer, 2014). The literature has also revealed that there are differences between social media and social networking sites. Basically, in the light of the previous definitions social networking sites aim to connect people while social media websites aim to share content with people (Boyd and Ellison, 2007).

Holland and Verploeg (2009) discussed that libraries use Twitter as a free advertising tool on a daily basis. Twitter also contributes to libraries being more engaged, productive and building relationships. Research by Vassilakaki and Garoufallou (2015) found that 77.8% cent of the libraries used Twitter for work purposes. Huang et al. (2017) discussed that libraries also use Twitter to create and promote a sense of community among libraries and users. Similarly, Murray (2010) argues that Twitter can assist library administration and librarians by promoting a sense of community. Also, studies found that libraries use Twitter to enhance the library experience for the users, for surveys and collecting opinions from the clients, and as a marketing tool for their services (Huang et al. 2017; Neilson, 2016; Vassilakaki and Garoufallou, 2015).

As a result, studies have found that the use of Twitter in libraries is increasing; for instance, Chen et al. (2012) found that the number of libraries adopting Twitter accounts is increasing. These opinions demonstrate a body of thought that Twitter provides libraries with an innovative and useful way to interact with their users. Studies have also found that social media are being used to save time and attract more users. Chu and Du (2013) argue that Twitter can convey considerable information, without users losing much time and effort. Similarly, in a recent study Neilson (2016) notes that the time and effort invested in a library’s Twitter account is minimal, so all libraries should utilize Twitter. However, the study also revealed that Twitter requires a lot of effort to use it effectively, engage users and promote services. Hence, libraries should consider their efforts on Twitter and focus on posting quality content in ways that impact library activity.

Type of library content on Twitter

Honeycutt and Herring (cited in Stvilia and Gibradze, 2014: 3) identify 11 content categories of tweets in academic libraries, including ‘about addressee, announce/advertise, exhort, information for others, information for self, meta-commentary, media use, express opinion, other’s experience, self-experience, and solicit information’. An earlier study by Naaman et al. (2010: 2–3) documented eight categories of tweet content types shared by academic libraries. These content categories are: ‘information sharing, self-promotion, opinions/complaints, statements and random thoughts, me now, questions to followers, presence maintenance, and anecdote’. On the same topic, Bosque et al. (2012: 16) identified seven categories: ‘campus events, community events, hours, library events, responses to reference questions, links to outside sites, and resources’. Cuddy et al. (2010: 4) identified five topics: ‘news about facilities, resources, downtime, events, and staff’.

Focusing on public libraries, Shiri and Rathi (2013) evaluated the content of tweets from only one public library: Edmonton Public Library in Canada. The tweets are divided into 13 categories. The analysis of tweets revealed that they were mainly used to communicate traditional library/patron interests. The highest number of tweets is observed to be in the category of ‘announcement’. The second highest category was ‘information sharing’ and the third highest category was ‘recommendations’.

Aharony (2010) did undertake research comparing tweets in academic and public libraries, considering frequency, language and content. The study examined data from 15 of each library type, using both descriptive statistical analysis and content analysis. The analysis revealed that public library tweets are divided into six categories: ‘library, information about, miscellaneous, general information, general recommendations, and technology’. The academic libraries’ tweets are divided into five categories: ‘library, information about, miscellaneous, technology and general information’. Interestingly, the research revealed that in both kinds of libraries, the ‘library’ category is the largest category. The ‘library’ category includes ‘announcements about library events, collections, services, reference questions, and general library questions’ (Aharony, 2010).

Users’ engagement on Twitter and the social web

While several studies investigated library tweets content, the question of how to evaluate user engagement levels has been considered by only a very small number of studies. For example, Okazaki et al. (2015) posit that engagement can be conceptualized as the state of being engaged, connected, involved and interested in information shared through Twitter. The study argues that engagement influences users and helps libraries to brand themselves. This is supported by Ibrahim et al. (2017) who believe that users’ engagement on Twitter is one of the most powerful ways libraries can improve their connectivity with followers, reputation and overall user satisfaction with library services.

According to Nguyen (2015) two-way communication leads to greater engagement, but he reports that there is a low level of user engagement in most of the libraries he studied. Similarly, Neilson (2016) found that although most libraries have Twitter accounts, some never use two-way communication with their users. He concludes that for a higher level of user engagement, libraries should encourage two-way communication.

Ibrahim et al. (2017) identified three levels of user engagement: low level, medium level and high level of engagement. In an attempt to evaluate user engagement, the authors suggested that one reply indicates a low level of engagement, two to three replies denote a medium level of user engagement and more than three replies constitute a high level of user engagement.

Measuring users’ engagement did not stop on Twitter. Studies have tried to explore and compare users’ engagement in different platforms available to them. For example a study by Palmer (2014) tried to characterize users’ engagement on Facebook and Twitter. Chen et al. (2012) examined four types of interactions in the social web which are knowledge sharing, information dissemination, communication and knowledge gathering. The study revealed that knowledge sharing attracted a large number of library users. The study suggested that in order to gain more engagement from library users the libraries have to utilize different types of social network channels to benefit from the unique characteristics of each channel. Al-Daihani and AlAwadhi (2015) suggested that there is a need for guidelines for libraries to help them to get more engagement for their users on social media.

Studies have also explored the difference in users’ engagement in different countries. For example, Vaughan and Gao (2016) compared users engagement in USA and China. The study revealed that libraries in the USA have a huge number of followers who interact with their pages in social media. However, the study revealed that there is a rapid growth in the number of Chinees followers of libraries’ accounts on Twitter.

Methodology

This study was conducted using a mixed method of quantitative and qualitative research. To analyze the qualitative data the researcher adopted thematic analysis approach. Braun and Clarke (2006) refer to thematic analysis as a method to identify, analyze and report themes within data. It interprets various aspects of the research topic. Thematic analysis is a foundational method for qualitative analysis. The thematic analysis in this research allowed for the reporting and determination of themes within the data using an inductive approach, without pre-conceived categories.

In order to gather the data for this research a sample of 100 tweets were selected from each library’s account. The tweets were collected from 19 June to 31 July 2017. The researcher selected the most recent 100 tweets from each account as it was thought that recent tweets would reflect the current activity of each library. Equal number of tweets were collected to ensure the reliability of the results. The researcher avoided collecting older tweets because both libraries did not have Twitter accounts in the same period. The researcher planned to collect more tweets after analyzing the data collected in the first phase of the research; however, the during the analysis the researcher reached saturation. Saturation was reached because no new data had emerged from the analysis; as a result the number of collected tweets was deemed to be sufficient for the purpose of the research.

The collection of the data included several steps:

The first step involved capturing Twitter metrics from the sample dataset; during this step the researcher collected the available analytics of engagement with tweets.

The second step included recording the content of the tweets and using thematic analysis to identify and report themes within the data. The themes were identified using inductive approach, without preconceived categories. The data content was repeatedly studied first, and diverse categories then emerged from the data. This allowed for multiple ways to explore the research question and a more comprehensive understanding than either quantitative or qualitative approaches alone.

As a result of using the thematic approach five core categories/themes emerged from the data. The five main themes are:

General Information

Promotion of Services and Events

News and Announcements

Library Collection

Communication with Users.

These themes were classified as high-order themes as all the posts can be categorized under them. However, the analysis revealed many posts that could be gathered under sub-categories. These themes were classified as low-order themes. Lower-order themes were grouped under the higher-order themes in order to understand the relation between them. It was noticed from the results that many of the lower-end themes were not mentioned in the literature unlike the higher-end themes.

While the research focused on the higher-order themes the author could not neglect the importance of the lower-order themes in understanding the engagement of the users with the tweets. The reliability of the results was ensured through peer debriefing. This allowed the researcher to show the results to peers in the departments to check the accuracy and confirm the reliability, the credibility and the validity of the results.

Data collection and data analysis

The dataset consisted of 100 sequential tweets from each library, from a time period with the same end-date. Tweets from the selected libraries were collected from 19 June to 31 July 2017. It is pertinent to mention here that both original tweets and retweets are considered. It is expected that the number of tweets and the number of followers have increased since that time. This sample size gives a snapshot of the Twitter accounts of each library. Observations of levels of engagement are recorded, including numbers of likes, replies and retweets. The types of content are categorized and each category’s engagement levels reported. Comparison between the two libraries was considered in the context of their respective number of Twitter followers. The selected data through this approach helped to answer the research question and satisfies the research objectives. Data were collected by following criteria:

Only 100 tweets of each library were selected during a period (June–July 2017);

Only Australian libraries were included in the sample;

The tweets are in English;

The Twitter account is actively used; that is, on a daily or weekly basis.

The data were analyzed to explore what libraries tweet and the interaction patterns with followers. For analysis of the collected qualitative data, a thematic analysis method was adopted. This analysis aims to build an organized system of categories of Twitter content produced by each library. The process was conducted manually to chart data into a framework matrix. This process resulted in the creation of a conceptual framework that explains the analysis the content type of tweets for each library. Statistical analysis was conducted to analyze engagement level of followers with the content produced by the libraries. Engagement level was measured by analyzing a number of tweets, replies, retweets and likes.

Results and discussion

The analysis of this research was based on tweets from only one academic library (RMIT University Library) and one public library (State Library of Victoria). This part of the research discusses the results of the research, which explored the types of tweet content published by these libraries, and the user engagement level. All data captured was correct as at 23 July 2017.

The numbers of tweets and Twitter followers were tabulated. A summary of the results is listed in Table 1 which shows the summary of Twitter accounts studied during the research. It shows the total number of tweets and number of followers for each library, as at 23 July 2017. This is the overall summary of their Twitter accounts since their dates of joining. It does not represent the sample of 100 tweets which is analyzed in this report.

Table 1.

Australian libraries, numerical summary of Twitter accounts.

Name of library	Twitter handle	Number of tweets	Number of account following	Number of followers	Likes	Date joined Twitter
RMIT University Library	(Library_RMIT)	4154	2000	2907	614	February 2010
State Library of Victoria	(Library_VIC)	11,200	856	27,600	8103	February 2010

Type of content

Understanding the type of information/content posted on Twitter is imperative for the research question. Initially, tweets were categorized by lower-order themes, which were then grouped into higher-order themes. The results show that libraries share various types of information related to themes in the following main categories:

General Information

Promotion of Services and Events

News and Announcements

Library Collection

Communication with Users.

Based on these categories, Table 2 shows, for each library, what percentages of the sample tweets are published in each of the lower-order themes and grouped higher-order theme categories. The total sample number of tweets examined for this purpose was 100 tweets from each library. Because the frequency of tweets is different for the two libraries, the time periods for the sample 100 tweets are:

RMIT University Library: 26 June to 23 July 2017.

State Library of Victoria: 19 June to 23 July 2017.

Table 2.

RMIT University Library and State Library, summary of the percentage of tweets, by lower and higher-order themes.

Lower-order themes	RMIT Library/percentage of tweets	State Library/percentage of tweets	Higher-order themes	RMIT Library/percentage of tweets	State Library/percentage of tweets
Library as place	0.00%	6.00%	General Information	0.00%	6.00%
Promotion of library events	7.00%	30.00%	Promotion of services and events	54.00%	48.00%
Promotion of library services	19.00%	3.00%
Promotion of state/national/international events	8.00%	6.00%
Promotion of external events	20.00%	9.00%
Library spaces/facilities	8.00%	1.00%	News & announcements	18.00%	3.00%
Opening hours	10.00%	0.00%
Job opportunities	0.00%	2.00%
General library collection	0.00%	5.00%	Library collection	24.00%	33.00%
Trial	7.00%	0.00%
Databases	11.00%	0.00%
Blog post	0.00%	9.00%
Special Collections	6.00%	11.00%
Special Collections – historical photo	0.00%	8.00%
Direct communication	4.00%	9.00%	Communication with users	4.00%	10.00%
Questions	0.00%	1.00%	Communication with users	4.00%	10.00%

Table 2 shows a combined results summary of the percentage of tweets published by each selected library, categorized by the higher order themes. Although there is a range of content, this Table indicates that the highest percentage of the libraries’ tweets are about Promotion of Services and Events.

Comparing the content types of the two libraries’ tweets revealed substantial overlap. For example, 48% and 54%, respectively, of the tweets by SLV and RMIT fell under the higher-order theme of Promotion of Services and Events. Similarly, 33% and 24%, respectively, of the tweets by SLV and RMIT fell under the higher-order theme of Library Collection. Interestingly, in the higher-order theme of News and Announcements, there was a marked difference in the percentage of tweets by the two libraries, with 3% and 18%, respectively, of the tweets by SLV and RMIT falling under this higher-order theme as shown in Table 2. Also, the findings uncovered a new theme, which was not found in the literature review: Library as Place (lower-order theme)/General Information (corresponding higher-order theme). Furthermore, tweets under this theme were exclusive to SLV, with no such tweets by RMIT. This may be due to the distinct nature of SLV and its significance in the city of Melbourne. In addition, SLV’s user base extends to the entire population of Victoria, and indeed national and international visitors.

Higher-order themes

The first part of the research question focuses on identifying types of information shared via Twitter accounts of an academic and a public library. Themes have been identified by using a thematic analysis process. The results revealed five higher-orders themes. These themes categorize the types of contents posted in these Tweets. The following themes were identified:

General Information

Promotion of Services and Events

News and Announcements

Library Collection

Communication with Users.

Higher-order theme A: General Information

The higher-order theme of General Information refers to information that the library posted on its Twitter account which is general in nature. The information posted in this theme can be of any type which cannot be categorized under other themes. As mentioned earlier, this theme was not discovered in the literature review. From Table 2, the results demonstrate that SLV shared such information with the users (6 %). However, on the other side, the academic library did not post any tweets that were general in nature. From Table 2, it is noted that RMIT published (0%) of tweets in the higher-order theme of General Information. As previously discussed, SLV has distinct characteristics which may explain this. RMIT, on the other hand, shares information specific to its defined user base of students and academics. Figure 1 provides an example for General Information tweets.

Figure 1.

Example of General Information tweets.

This higher order theme has only one lower-order theme:

Lower-order Theme 1: Library as Place

This lower-order theme refers to the type of information which promotes the library as a preferred place to visit. Table 2 shows that the lower-order theme is found only in SLV’s sample tweets and the number of tweets is six. Based on these results, it can be seen that RMIT Library does not Tweet on ‘library as place’.

Higher-order Theme B: Promotion of Services and Events

The findings showed that the higher-order theme of Promotion of Services and Events covered the highest percentage of content posted on Twitter by both libraries. This higher-order theme covers information shared on Twitter library accounts which is related to the promotion of specific library services and various events.

Table 2 shows the percentages of the libraries’ tweet content under this theme. RMIT Library posted 54%, whereas SLV posted 48% of content, concerned with marketing their services and informing followers about the latest facilities and events updates. These high percentages in both libraries show that Twitter is widely used as a promotional tool, to attract followers to more fully utilize a library’s offerings. These results are validated by AL-Daihani and AL-Awadhi (2015), who noted that libraries adopt Twitter to promote their presence through branding their services and events. Moreover, Aharony (2010) argued that the ‘library’ category is the highest category in both kinds of libraries, which includes ‘announcements about library events and services’. As shown in Table 2, the results of this research align with their findings. The Promotion of Services and Events higher-order theme was divided into the following lower-order themes:

Theme 1: Promotion of Library Events

Theme 2: Promotion of Library Services

Theme 3: Promotion of State/National/ International Events

Theme 4: Promotion of external events.

The findings illustrated in Table 2 revealed that, of the four lower-order themes in this category, Promotion of Library Events had the highest number of tweets by SLV, while the highest number of tweets by RMIT were categorized as Promotion of External Events. The lowest number of tweets by SLV were categorized as Promotion of Library Services, whereas the lowest number of tweets by RMIT were under the theme of Promotion of Library Events.

After examining the nature of tweets in these categories, it is evident that SLV has a focus on promoting its own library events. Examples indicate that this may, in part, be explained by the wide range of SLV’s audience. SLV holds events for children, researchers, the general public and more. Events include topics as diverse as zines, book design, music, indigenous issues and immigration stories. Conversely, RMIT Library has a focus on promoting external events, primarily events held by RMIT University. For example, the library tweeted about an event hosted by the RMIT Centre for Innovative Justice. This may be because RMIT Library does not itself initiate or host many events, but does serve as a conduit for promoting university-wide events. In this way, it is serving as a connection between its own followers and the goals of RMIT as a whole.

Higher-order Theme C: News and Announcements

This theme includes tweets covering any library-related news and announcements. The results suggest that academic libraries have a greater focus on this type of content. The trend of posting under this theme is markedly different in RMIT and SLV. From Table 2, it is evident that 18% of RMIT’s sample tweets are categorized as News and Announcements. However, Table 2 shows that only 3% of SLV’s sample tweets are categorized as News and Announcements.

AlDaihani and AlAwadh (2015) posited that Twitter is also mainly used for news and announcement by libraries. Moreover, Milstein (2009) argued that academic libraries use Twitter for specific announcements, such as upcoming deadlines. RMIT are more focused on sharing tweets on opening hours and library space/facilities, whereas SLV tweeted on job opportunities at the library. This aligns with Huang et al. (2017), who contended that most academic libraries use Twitter to share news specifically related to what is happening in the library itself, such as library opening hours. However, the findings of this research differ from Shiri and Rathi (2013), who indicated that the highest number of tweets by public libraries was in the ‘announcement’ category.

The News and Announcement higher-order theme was divided into the following lower-order themes:

Theme 1: Library Spaces/Facilities

Theme 2: Opening Hours

Theme 3: Job Opportunities.

The findings outlined in Table 2 revealed that, of the three lower-order themes in this category, Job Opportunities had the highest number of tweets by SLV, while the highest number of tweets by RMIT were categorized as Opening Hours. The lowest number of tweets by SLV were categorized as Library Spaces/Facilities, whereas the lowest number of tweets by RMIT were under the theme of Job Opportunities. The only lower-order theme in which both libraries tweeted was Library Spaces/Facilities. It is evident that the opening hours of the library are an important focus for RMIT. This may be because this information is critical for students needing to study, including those who need to work together in groups. The fact that SLV tweeted twice about a new job opportunity at the library indicates that they believe Twitter is a good way to spread news of this role to a wide audience. RMIT Library’s recruitment practices may not allow for this kind of communication.

Higher-order Theme D: Library Collection

Library Collection is the higher-order theme associated with tweets concerning the collections that the libraries hold. These collections include books, journals, databases, historical and special collections, as well as resources on trial at the library. The results show that this is the second highest type of content tweeted by both SLV and RMIT.

This echoes the literature review finding that academic and public libraries frequently post on their resources and collections. This is supported, particularly by Shiri and Rathi (2013), who found that the second highest category was ‘information sharing’ which includes different types of library collection.

Harrison et al. (2017) posited that libraries update followers with information about different items in the library collections. Bosque et al. (2012) found that over a half of academic libraries tweeted ‘about library resources, followed by library hours and events.’

Interestingly, the results show that a higher percentage of SLV’s tweets were in this theme than was the case for RMIT. Table 2 shows that 24% of RMIT’s sample tweets are categorized as Library Collection, while it shows that 33% of SLV’s sample tweets are categorized as Library Collection.

The Library Collection higher-order theme was divided into the following lower-order themes:

Theme 1: General Library Collection

Theme 2: Trial

Theme 3: Databases

Theme 4: Blog post

Theme 5: Special Collections

Theme 6: Special Collections – Historical Photo.

The findings show in Table 2 that, of the six lower-order themes in this category, Special Collections had the highest number of tweets by SLV, while the highest number of tweets by RMIT were categorized as Databases. The lowest number of tweets by SLV were categorized as Databases and Trial, whereas the lowest number of tweets by RMIT were under the themes of Special Collections – Historical photo, Blog post and General Library Collection.

The breakdown of tweets into these lower-order themes reflects the nature and focus of the collections in each library. SLV holds rich special collections, including photos of historical interest. This fits their role as custodian of such material. Conversely, RMIT’s major focus is on databases, which meet the needs of their students and academics.

Higher-order Theme E: Communication with Users

This higher-order theme shows that both libraries are involved in some types of direct communication with their followers. This communication can be in the form of questions, inquiries, or any direct communication. From Table 2, it appears that 4% of RMIT’s sample tweets are categorized as Communication with Users. On the other hand, it is evident that 10% of SLV’s sample tweets are categorized as Communication with Users. The results suggest that there is not a high level of interaction between the libraries and their followers. This is supported by Bosque et al. (2012) who found that less than 10% of libraries used Twitter to answer reference questions. Also, according to Nguyen (2015), there is a low level of direct user engagement in most libraries. The Communication with Users higher-order theme was divided into the following lower-order themes:

Theme 1: Direct communication

Theme 2: Questions.

The findings show in Table 2 that, of the two lower-order themes in this category, Direct Communication had the highest number of tweets by SLV and RMIT. The lowest number of tweets by both SLV and RMIT were under the theme of Questions. It can also be noted that SLV has one tweet concerned with Questions, whereas RMIT Library does not have any Questions under the relevant higher-order theme. The numbers of tweets in these themes may be too small to be significant, although it is interesting to note that SLV has more Direct Communication with its followers overall. This may be linked to SLV having a wider demographic range of people amongst its followers.

Level of engagement

The term engagement refers to demonstrated interaction with followers, which may be considered an indication of success in library activity. This term determines the story of who is listening and talking to the library. Therefore, understanding level of engagement is very important for the libraries. It shows the frequency of engagement between followers and the libraries and how they interact with the libraries, which shows up in their timeline (Simply Measured, 2014). User engagement is measured by the following interactions:

Replies (@): When the follower directly responds by click (@), and talks with the library account directly on the Twitter timeline by using Twitter handle at the beginning of the Tweet.

Retweets (): A retweet is a re-post of the original tweet of the library, sent by different users to share with their audience (Simply Measured, 2014).

Likes (): A Like is when a follower clicks on the heart icon in the button of a Tweet without having to retweet or reply to the tweet (Simply Measured, 2014).

The results demonstrated that SLV has a significantly higher number of tweets and followers than RMIT Library. While these totals are not the subject of this analysis, it should be noted that both libraries joined Twitter in February 2010. However, the size of each library’s potential audience would need to be taken into account. As a result a further analysis needs to be done on the totals.

The results in Tables 3 and 4 show that the followers of both libraries have the highest level of engagement for content in the Promotion of Services and Events higher-order theme. This substantial level of engagement could be explained by the fact that both libraries post most of their content in this theme. In contrast, the table also shows that both libraries have the lowest engagement level for the higher-order theme of Communication with Users. It is also evident that the libraries publish the lowest number of tweets in this theme. So, it can be argued that the level of engagement is a factor of the numbers of tweets in any theme.

Table 3.

Summary of number of level of engagement for SLV, by higher-order theme.

RMIT Library
Higher-order themes	Number of replies	Number of retweet by other(s)	Number of likes
General Information	0	0	0
Promotion of Services and Events	0	35	51
News &Announcements	0	12	39
Library Collection	2	19	34
Communication with Users	3	0	4

Table 4.

Summary of number of level of engagement for RMIT, by higher-order theme.

State Library
Higher-order themes	Number of replies	Number of retweet by other(s)	Number of likes
General Information	4	122	295
Promotion of Services and Events	29	252	639
News &Announcements	4	89	93
Library Collection	9	171	379
Communication with Users	4	42	53

It can be said that if the libraries intend to increase users’ engagement, then they are required to publish content in the theme more frequently. These findings will provide a new foundation for future research as there is a scarcity of research available on the level of engagement, specifically in academic and public libraries.

RMIT Library

The results demonstrated that the RMIT library got the highest number of engagements in the form of likes and retweets − 51 and 35 respectively – for the theme of Promotion of Services and Events. Table 3 also shows that the second highest number of engagements in the form of likes is in the News and Announcement theme (39). However, the count of retweets is only 12 for this higher-order theme. It is observed that the library gets a low count of engagements in the form of replies, i.e. three for Communication with Users and two for Library Collections.

State Library of Victoria

The data showed a very high count of user engagement actions. It shows that the highest number of engagement level is in the form of likes, for all the themes. Retweets is the lowest recorded engagement action for all themes. However, the results also indicate that the library has some engagement in the form of replies in all of the themes.

Although the literature review did not reveal any similar data or defined criteria for evaluating the level of engagement, for the purpose of this research, three interactions are defined that determine the user level of engagement i.e. replies, retweets and likes. The results suggest that both types of libraries receive the highest number of engagement ‘likes’ for all themes, whereas the users are least likely to ‘reply’ for all themes. The second highest Number of the engagement is in the form of ‘retweets’. Although these results suggest that the users have the highest engagement in the form of ‘likes’, this differs somewhat from the research in the literature review.

A study by Ibrahim et al. (2017) focused on evaluating user engagement, by the number of replies. The study stated that if the number of replies are greater, it determines that the content receives higher user engagement. Therefore, by this measure, it can be argued that the users of both SLV and RMIT showed a low level of engagement. This is supported by Neilson (2016) who found that some libraries never use two-way communication with their users. However, these results and discussion also provide a new dimension of evaluating the level of engagement, by analyzing the numbers of ‘likes’. This method can be applied in future research in public and academic libraries.

It is important to focus on the level of the engagement in social media as it is an indicator to measure the success or the failure of library services and even the events organized by the library. Measuring users’ engagement is a way to provide the library with statistics that can help to provide more sophisticated services that fit users’ needs. Social media are considered as tools that help the libraries in this area; however, libraries should be able to maintain their services on social media to guarantee more engagement from the users. The current study has proposed that counting the number of ‘likes’ would serve as indicator of the level of engagement.

The results of the study are consistent with the literature which considered the level of engagement in social media specially Twitter is a key performance indicator for the libraries, e.g. (Chen et al., 2012; Okazaki et al., 2015; Palmer, 2014). Twitter provides metrics that would help the libraries to find about the number of people who are interested in a specific service provided by the library or social activity. Previous studies have measured the engagement by counting the interaction of the users in Twitter; Counting the number of views and retweets is a way to know about the success of the failure of a service. Moreover, Twitter is a platform that can be utilized to receive feedback and suggestions from the library community which is faster than the old methods such as surveys.

Conclusion

In conclusion, although there are some differences in the use of Twitter between RMIT and SLV, (level of engagement and content theme), both libraries share information about their activities, collections and news.

The thematic analysis revealed 16 lower-order themes, which were grouped into five higher-order content themes. One of the key findings of this study was that a significant number of tweets, in both libraries, were associated with the theme of Promotion of Services and Events. It is concluded that both RMIT and SLV utilize Twitter heavily to promote their services, as well as both library and external events. The libraries sometimes linked their own collections and activities to state, national and international events.

Both libraries also post content on databases, eBooks, photos and other significant collections. They use Twitter to let their followers know about the specific collections and resources they have. The nature and number of tweets by content theme is a reflection of the mandate and societal status of each library. Further, this research also explored the engagement of the followers with the libraries’ Twitter content. It was measured by numbers of replies, likes and retweets by the followers.

Content related to the promotion of services and events generated more user engagement than any other type of content. Therefore, it can be argued that the followers are mostly looking for services and events information on the libraries’ Twitter accounts and prefer to engage with such content. It was also observed that the second highest engagement level was with content about libraries’ collections. It is concluded that followers also find content about special collections, databases, general collections and photo collections, to be of interest. This also answers the research questions which aimed to analyze the level of user engagement level with various content.

The timeframe, number of libraries and number of the sample’s tweets in the study may limit its usefulness in generalizing to a wider number of libraries. The issue of data being collected only via the publicly available Twitter interface restricted the depth of analysis possible. Overall, the results of this study contributed to understanding how the libraries’ users engage with their post in social media and the types of the posts. This may lead to improving the quality of the posts to attract more users to participate. It is important to understand that the quality of the posts on the social media attract more users to engage. Additionally, the results showed that it is essential for both libraries to focus on some posts more than others to reach more users. However, because of the previously mentioned limitations. It is also suggested that to increase the engagement on the social media the libraries should focus on promoting their services and accounts on social media among the regular users of the libraries facilities. Additionally, the libraries can train the users on how to use social media and interact with their accounts. Email alerts can serve the same purpose of promoting the library accounts on the social media.

It is important to indicate that libraries should put more emphasis on managing social media accounts to attract more engagement with their accounts on the social web. Libraries should have a clear strategy to utilize social media services to attract users who do not use Twitter as a medium for social engagement. Various subjects that fit all library visitors should be shared through social media. Additionally, it is noticed that libraries can increase the engagement by sharing more tweets that concern the users, promote the social media account within the library, focus on two-way communication and provide more services through the social web such as the reference services. Furthermore, the library should hire staff who have the skills to manage social media accounts which would guarantee that these accounts are regularly updated.

Libraries should also actively and regularly monitor Twitter Analytics, which will allow them to gain deeper insights into the behaviour of their users. Libraries can improve their services by understanding what are the topics that users engage with, which will help to put more focus on these services. Additionally, data analysis of users’ trend in social media would help in informing and evaluating libraries social media strategy.

Further research may be warranted, by both a broader sample of libraries and deeper analysis of data via Twitter Analytics. A more extended study timeline would also give greater insight to trends over time. Broadening the number of sample libraries would enable testing of the applicability of the content themes revealed in this study, across a range of libraries. The similarities and differences between public and academic libraries could also be explored in more depth.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biography

Reham Ali Alsuhaibani is a lecturer at Princess Nora bint Abdulrahman University, Faculty of Arts, Saudi Arabia. She received her Master’s in Information Management, from RMIT University, Melbourne, Australia. Her areas of research interest includes: digital libraries, information retrieval systems, social media, information behaviour, information seeking, information and knowledge management, social informatics.

References

Aharony

(2010) Twitter use in libraries: An exploratory analysis. Journal of Web Librarianship 4(4): 333–350.

Al-Daihani

AlAwadhi

(2015) Exploring academic libraries’ use of Twitter: A content analysis. The Electronic Library 33(6):1002–1015.

Bosque

Leif

Skarl

(2012) Libraries atwitter: Trends in academic library tweeting. Reference Services Review 40(2): 199–213.

Boyd

Ellison

(2007) Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication 13(1): 210–230.

Braun

Clarke

(2006) Using thematic analysis in psychology. Qualitative Research in Psychology 3(2):77–101.

Chen

Chu

(2012) How do libraries use social networking sites to interact with users. Proceedings of the American Society for Information Science and Technology 49(1): 1–10.

Chu

(2013) Social networking tools for academic libraries. Journal of Librarianship and Information Science 45(1): 64–75.

Cuddy

Graham

Morto

(2010) Implementing Twitter in a Health Sciences Library. Medical Reference Services Quarterly 29(4): 320–30.

Harrison

Burress

Velasquez

(2017) Social media use in academic libraries: A phenomenological study. Journal of Academic Librarianship 43(3): 248–256.

10.

Holland

Verploeg

(2009) No easy targets: Six libraries in the economy’s dark days. Public Libraries 48(4): 27–38.

11.

Huang

Kai

Yuyang

, et al. (2017) Understanding user-librarian interaction types in academic library microblogging: A comparison study in Twitter and Weibo. Journal of Academic Librarianship 43(4): 329–336.

12.

Ibrahim

Wang

Bourne

(2017) Exploring the effect of user engagement in online brand communities: evidence from Twitter. Computers in Human Behavior 72(July): 321–338.

13.

Joosten

(2012) Social Media for Educators: Strategies and Best Practices. Hoboken, NJ: John Wiley & Sons.

14.

Kaplan

Haenlein

(2010) Users of the world, unite! The challenges and opportunities of social media. Business Horizons 53(1): 59–68.

15.

Kenchakkanavar

(2015) Facebook and Twitter for academic libraries in the twenty first century. International Research: Journal of Library and Information Science 5(1): 162–173.

16.

Kietzmann

Hermkens

McCarthy

, et al. (2011) Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons 54(3): 241–251.

17.

Milstein

(2009) Twitter for libraries (and librarians). Available at: www.infotoday.com/cilmag/may09/Milstein.shtml (accessed 19 September 2017).

18.

Murray

(2010) A mutable cloud: Fostering community through cloud computing. Library Leadership and Management 24(2): 58–64.

19.

Naaman

Boase

Lai

(2010) Is it really about me? Message content in social awareness streams. In: Proceedings of the 2010 ACM conference on computer supported cooperative work (CSCW’10), Savannah, USA, 6–10 February 2010, pp. 189–192. New York: ACM. Available at: http://infolab.stanford.edu/~mor/research/naamanCSCW10.pdf (accessed 20 September 2017).

20.

Neilson

(2016) What do health libraries tweet about? A content analysis. Canadian Journal of Library and Information Practice and Research 11(1):1–18.

21.

Nguyen

(2015) Establishing a participatory library model: A grounded theory study. Journal of Academic Librarianship 41(4):475–487.

22.

Okazaki

Díaz-Martín

Rozano

, et al. (2015) Using Twitter to engage with customers: A data mining approach. Internet Research 25(3):416–434.

23.

Palmer

(2014) Characterizing university library use of social media: A case study of Twitter and Facebook from Australia. Journal of Academic Librarianship 40(6): 611–619.

24.

Shiri

Rathi

(2013) Twitter content categorization: A public library perspective. Journal of Information & Knowledge Management 12(4):1–8.

25.

Simply Measured (2014) The complete guide to Twitter Analytics: How to analyze the metrics that matter. Available at: https://www-users.cs.york.ac.uk/alistair/research/Social%20media/CompleteGuidetoTwitterAnalyticsSimplyMeasured.pdf (accessed 22 July 2017).

26.

Stvilia

Gibradze

(2014) What do academic libraries tweet about, and what makes a library tweet useful? Library and Information Science Research 36(4):136–141.

27.

Taha

Al-Sayyed

Alqatawna

, et al. (2017) Social Media Shaping E-Publishing and Academia. Cham: Springer International.

28.

Vassilakaki

Garoufallou

(2015) The impact of Twitter on libraries: A critical review of the literature. The Electronic Library 33(4):795–809.

29.

Vaughan

Gao

(2016) The use of microblogs and social networking services: A comparison between academic libraries of the United States and China. Journal of Web Librarianship 10(1):1–13.

30.

Zou

Chen

Dey

(2015) A quantitative analysis of Pinterest: Understanding library user engagement strategies for effective social media use. Journal of Information Technology Management 26(3): 21–32.