Think back on your own time at school, and ask yourself the following questions:

• 1 Did your school encourage you to have a lifelong love of reading, or was it perhaps only concerned with reading for the purposes of assessment?

• 2 If you feel that your school encouraged you to have a lifelong love of reading, what was it about your school that led you to this conclusion?

Given that this book is focused on imagining and creating a reading culture, and fostering a lifelong reader identity in young people, critically reflecting on your own experiences as a student can be a good place to start.

Perhaps you found yourself within a richly supportive environment, where reading was celebrated, encouraged and supported. As explored in this book, and as you may recollect if this was your experience, schools can communicate valuing of reading in many diverse ways.

Alternatively, your school may have been a space where enjoyment of reading was positioned as peripheral and irrelevant. If this was you, and yet you are an avid reader, your love of reading may have been fostered at home, never nurtured within your school.

One of the key purposes of this book is to enable you to identify and harness the approaches, resources, and strategies that contribute to a school reading culture whether you had prior exposure to them or not. Even a thriving school reading culture can benefit from new insights that can make the culture even more effective, and help you to foster a love of reading in all of your students.

School culture is influential

Students spend so much time at school, it is vital that the activities, experiences and school cultures they are exposed to build their capacity as lifelong learners who will continue to develop their knowledge and skills in adulthood. Students in Organisation for Economic Co-operation and Development (OECD) countries spend an average of 7,590 hours of required instruction at primary and secondary school (OECD, 2019a), which is no small chunk of time. They are at a formative stage of their lives, and their schooling experiences can have a significant impact on their current and future trajectories. Unsurprisingly, the kinds of learning environments that students experience in their schools impact on their academic achievement (e.g. MacNeil et al., 2009), but students’ experiences in school can also shape their behaviours and beliefs well beyond their schooling years.

I hope that by the time you reach this stage of the book, you arrive with strengthened confidence in your ability to roll out or rejuvenate a whole school reading culture in your context.

While I have presented a vast array of options and pathways with research support that you might pursue, please use this book as a tool to inform rather than dictate your choices. Please seek to establish a whole school reading culture that reflects the unique needs of your school, following the process that resonates with you and your team, and aligns with your current resourcing constraints.

Literacy proponents and policymakers often fall into the trap of proposing that a single solution can be the answer to all students’ literacy problems. As I have explored extensively elsewhere, this is almost always problematic given that learners have diverse strengths, needs and backgrounds (Merga, 2020a). While building students’ motivation and opportunities to read in a reading-supportive culture can make a big difference to students’ literacy attainment, specific targeted interventions that address students’ literacy gaps, as well as cover the required literacy curriculum, will be needed too.

While not the only essential literacy initiative for schools, increasing reading engagement by fostering a whole school reading culture using research-supported strategies can transform students’ literacy learning, as well as their attitudes towards reading and related literacy-supportive activities. It will probably not independently solve all issues concerned with literacy that your students face, and ‘integration of direct skills instruction and motivation interventions might produce synergistic effects and optimize gains in reading performance over time’ (Toste et al., 2020, p. 446). To this end, you go a long way towards enhancing student literacy when you pay close attention to making gains in motivation and opportunities to read, while also offering a range of additional targeted literacy interventions as explored in my previous book (Merga, 2023b) and through the work of many other researchers. Many of the strategies that underpin a whole school reading culture yield significant gains for SLLs in particular (e.g. reading aloud to SLLs as explored by Westbrook et al., 2019), so a whole school reading culture can play a key role in closing the gap for these students as part of a multifaceted strategy.

As I explore in this chapter, there is a wealth of research exploring the benefits of a range of socially and environmentally mediated practices designed to build reading engagement and a reading culture in a school. Simply, reading engagement cannot continue to be ignored as a crucial support for enhancing students’ literacy performance. We need to foster whole school reading cultures that seek to improve students’ attitudes towards and frequency of engagement in literacy-supportive practices such as reading, and do this alongside the explicit literacy skills instruction that ideally should be occurring across subject areas (e.g. Merga, 2023b).

As previously noted, young people who believe that reading remains important beyond learning to read are nearly twice as likely to read daily than those who did not share this view (Merga & Mat Roni, 2018a), and we cannot take for granted that students understand the importance of an ongoing commitment to reading. A survey of Australian children aged 6–17 found that only 58% of children understood that reading books for fun is highly important, and children's views on the importance of reading may decline dramatically by age (Scholastic, 2016a).

But how can we effectively encourage students to enjoy and value reading according to the research? What are evidence-supported practices when it comes to building reading engagement?

I explore research-supported practices that can feature in a whole school reading culture here, drawing on key research findings, and making reference to current research that has been released since my earlier book in this space (e.g. as reported in Merga, 2018a). While there are some good auditing tools that have been created around fostering a whole school reading culture (e.g. National Library of New Zealand, n.d.-a; n.d.-c), the links to the extant research in such tools can often be cursory, so reading this chapter will equip you to make a strong argument to support the implementation of any of the strategies detailed herein.

Opportunities for regular silent reading for pleasure

If we want young people to be readers, we need to provide opportunities for them to read. With research ‘linking volume of reading to reading achievement and oral reading fluency’, it is surprising that this educational practice is peripheral in many contemporary schools (Allington, 2014, p. 22).

By the time this book is published, it will be over a decade since I first approached Facet Publishing about writing Web Metrics for Library and Information Professionals, and the rapid pace of change in the field means that a second edition is already overdue. Of course, that would not be the case in all fields; books in different fields have different rates of decay. On the same day that I sat down to write this preface I also took delivery of a third edition of The Elements of New Testament Greek (Cambridge University Press, 2005). Although the third edition was almost 20 years old when I bought it, I expect the fundamentals of learning New Testament Greek to mostly remain the same, and even relying on a second edition (1965) or first edition (1914) probably wouldn't have been the greatest impediment to my elementary studies. Web metrics, however, have changed rapidly over the past decade, and a book that is to be of practical use to library and information professionals must necessarily be kept up to date: the web metric landscape of 2023 is very different to that of 2012.

One of the most notable changes has been the increased use of metrics more widely. Of course, not all these metrics are web metrics, or even bibliometrics, but the growth of web metrics is taking place within a wider growth of interest in the measurement of anything that might provide actionable insights in the move towards greater efficiency and return on investment. It means that within a university, for example, web metrics have gone over the last decade from being something that was primarily of interest to researchers, through something that was of interest to library and information professionals, to something that has become of interest to university management. As is explored within the book, this has both advantages and disadvantages for the information professional.

Undoubtedly part of the reason for the rise in interest in web metrics has been the rapid rise in interest in altmetrics. It is a rise that has, inevitably, led to a decline (or at least a stagnation) in interest in other areas of web metrics.

Introduction

This last chapter considers the potential future of web metrics and its relationship with librarians. The future of web metrics as written in the first edition of this book is now, to a certain extent, the past. As was noted at the time, it would be a fool's errand to attempt to predict the emergence of any specific new technology or tool, or the future dominance of any particular website, but it is reasonable to expect certain current trends to continue, such as more data and more web metrics. It was also expected that much of the data would be open, although this prediction was always less certain and now requires a more nuanced discussion. It is also necessary to add two new predictions: ‘More AI’ and ‘More Caution’, the second of which is more in hope than expectation. Each of these trends will have potential ramifications for web metrics and how they can be used by librarians.

Before we hurtle towards the future it is important to pause for a moment and reflect on how far we have come, not only in this book but with web metrics.

How far we have come

As would be expected of a book called Web Metrics for the Library and Information Professional, this book has discussed a wide range of tools and sources. These have included: web crawlers, social network sites, APIs, data aggregators, visualisation software and programming packages. The full list would be a long one, and as we necessarily move rapidly from one technology to another it is easy to overlook the scale of what is being discussed. After all, it is not unreasonable to liken the development of web crawlers and services such as Google Trends to early examples of scientific instruments that provided insights into the physical world.

When Robert Hooke's Micrographia was published in 1665 he created the world's first scientific best-seller. His modified microscope allowed him to create intricate drawings of objects whose finer details had largely been unseen, although the objects themselves were familiar to 17th-century gentlemen. Hooke's drawings included fleas, lice, nettle stings and even crystals of frozen urine, and he is credited with increasing interest in microscopy more widely.

Introduction

Much of the information that is placed online is not on an individual's or an institution's personal server but, rather, makes use of external sites and services: images are uploaded to Instagram, videos to YouTube, presentations to SlideShare and comments to Twitter. Such sites and services not only ease the publishing process for individuals but also enable the delivery of content that has the potential to go viral. There are advantages and disadvantages in this for web metrics. On the one hand, external sites and services can provide additional or more complete metrics: they provide a large amount of data structured in the same manner, and it is theoretically possible for all data that meets a particular criterion to be retrieved. On the other hand, metrics are severely limited to the functionality that the service allows; although the data may be structured, the website may not facilitate access to this data.

With such a wide range of services available, from a web analytic point of view it is necessary for librarians to ask which services are worthwhile (Vucovich et al., 2013) and at what point to stop spending their time and effort on a service. As with the web metrics discussed in the previous chapter, social media metrics offer the potential for a far wider range of insights than just into a library's own web presence. A handful of individual sites and services with hundreds of millions of users can also provide insights into a range of real-world user behaviours, forming the basis of a wide range of webometric investigations.

This chapter starts by considering the types of social network site that are available and the types of social network site content that librarians may be interested in evaluating. This is followed by a closer look at some of the popular social network sites that may form the basis of a web metric investigation, previous investigations that have taken place and some of the tools that are available.

Aspects of social network sites

The web is dominated by a small number of sites, many of which have incorporated some level of social networking functionality. Boyd and Ellison (2007) define a social network site as comprising three parts, allowing users to (1) construct a public profile in a bounded system, (2) articulate connections to other users and (3) navigate the connections they and others make.

Introduction

As we explained in Chapter 2, the term ‘web bibliometrics’ is used within this book to refer to the intersection between bibliometrics and web metrics, and involves the quantitative investigation of traditional publications (e.g., journals and books) through the web.

Traditional bibliometric investigations have not interrogated a universal library containing all bibliographic material that has ever been written but, rather, are operationalised through limited bibliographic resources, most often Scopus or the Web of Science citation index. Such bibliographic resources are restricted to a small proportion of the items that would be included within a universal library, and the information stored about the items is also restricted.

The web has made traditional bibliographic resources more accessible and extended potential bibliometric investigations in three ways. It:

• • potentially provides access to a greater proportion of items than would be housed in a universal library;

• • can provide access to the full text rather than just metadata;

• • enables the analysis of bibliographic content within the context of how it is being used;

and it has introduced new players into the market.

These extensions to traditional bibliometrics can potentially facilitate the provision of better services by librarians, through the provision of more detailed maps of science or the development of altmetrics measuring the impact of traditional publications online.

This chapter investigates each of these three areas of web bibliometrics and looks more closely at some of the research that has taken place and resources available.

More bibliographic items

Within this book use of the term ‘bibliometrics’ is restricted to the spirit of its Greek stem, ‘biblio’, and relates to the quantitative analysis of books, journals and similar types of publication that would traditionally have formed the basis of paper-based publications. By keeping to its more restrictive meaning it may seem as though by definition there are no new objects for investigation, but the vast majority of publications were never included within the bibliographic databases that formed the basis of many previous investigations.

The creation of bibliographic databases such as the Science Citation Index has traditionally been very resource intensive and it was therefore in the interests of abstracting and indexing services to limit the number of publications that were included.

Introduction

Web metrics for librarians combines tools and methodologies from the information science community with some of the goals and applications of the marketing community.

The first part of this chapter explores the relationship between web metrics and associated terms from the information science community (e.g., bibliometrics, altmetrics, scientometrics and webometrics), as well as its relationship with web analytics from the marketing community. Within these different areas web metric investigations may be broadly categorised as either relational or evaluative. The second part of the chapter looks more closely at the theoretical basis and practical investigations that can be considered under these headings, and how such investigations may be applied by librarians. The final part of the chapter considers the validation of web metric findings.

Web metrics

The term ‘web metrics’ is used throughout this book to refer to the quantitative measurement of the creation and use of web content. It has been adopted as a broad term, inclusive of the many different uses to which web metrics have been put and for which varying terminology has emerged over the years. In the style of Björneborn and Ingwersen's (2004) diagram of the metric terminology, Figure 2.1 on the next page shows the overlapping scope of the different metric terminology used within this book, with web metrics, the shaded area in the diagram, comprising web bibliometrics, web scientometrics, altmetrics, webometrics and web analytics. The sizes of the ellipses are not indicative of the size of the research areas but, rather, are for the purpose of clarity. Each of the areas that form web metrics is discussed in more detail below.

Information science metrics

Information science has coined a number of ‘metric’ neologisms over the years. Some have become widely adopted (e.g., bibliometrics), some have been short lived (e.g., internetometrics) and some remain extremely niche areas of investigation (e.g., discometrics). It is important to define the terminology clearly so researchers can be understood by one another and science policy makers (Lazarev, 1996), although in reality the metric terms are often ‘indistinguishable’ (De Bellis, 2009, 2) or subject to rapidly changing meanings. In recognition of this fact Milojević and Leydesdorff (2013) introduced the term iMetrics to encompass bibliometrics, scientometrics, informetrics and webometrics, although the metric language refuses to be pinned down, and by 2020 iMetrics was also being used to explicitly include altmetrics and librametrics as well (Maltseva and Batagelj, 2020).

Introduction

The web provides the opportunity for metrics to be used for a wide range of investigations, most of which may be broadly categorised as either evaluative or relational. This chapter takes a general view of evaluative metrics for websites, blogs and other self-hosted content; Chapter 5 considers evaluative metrics for third-party social media services that a large number of individuals and organisations make use of; and Chapter 6 considers relational web metrics. In many ways the split between self-hosted and third-party hosted content is an artificial one. For example, millions of blogs are hosted on third-party sites such as Blogger.com, while individuals and organisations do not have to rely on a microblogging service such as Twitter, but could host their own microblogging service using open source software such as Mastodon (https://joinmastodon.org). The split reflects a significant difference in the control a user has over the data that is available and an uncertainty in the data that is collected.

When people or organisations host their own content on their own servers the only limitations on the information that they gather about the use of content on their site are the technical problems to be overcome. However, when a website is integrated into the wider ecosystem there is more uncertainty: in the same way that there is no simple answer to how many web pages exist, there is not a simple answer to how many links a website has pointing to it or how many times it is mentioned. This is a problem that is made more difficult as the web is necessarily viewed through the limited view of a particular search engine or other tool, and the limitations of the tool are not always explicit. In comparison, many of the large social network sites provide a restricted environment where only certain behaviour is allowed (because of editorial and technical restrictions) and only certain data may be accessed. These restrictions also reduce the ambiguity in the data available from the services; a Twitter account has a known number of followers, a Facebook user has a known number of friends, a YouTube video has a known number of views. We may disagree with the concept of Facebook's ‘Like’ button or what counts as a view on YouTube, but these are known discernible units.

Introduction

Relational web metrics investigate the relationships between online actors, whether these are individuals, organisations, Twitter accounts or web pages. They may be considered distinct from evaluative web metrics as they are not necessarily designed to provide insights into the impact of content but, rather, provide insights into the structure of the network as a whole, or the position of actors within a network. As has been pointed out already, the distinction between evaluative and relational metrics may merely be one of perception; being central to a network may be considered an indicator of impact, or may merely be an indicator of being central to the network.

This chapter considers some of the social network analysis methodologies that have been adopted by the web metrics community and how these methodologies and visualisation technologies can be used to draw insights that may be of use for librarians.

Social network analysis methods

Social network analysis methodologies were developed in the social sciences, based on the idea that actors should not be considered in isolation but, rather, that their behaviour was best understood within the context of connected actors. The importance of people's social networks for sharing information and changing user behaviours is now widely recognised (Christakis and Fowler, 2009), and the application of social network analysis methodologies has not only been suggested for the information science community for traditional bibliometric investigations (Otte and Rousseau, 2002), but has also formed the basis of webometric investigations analysing the hyperlinks between web pages (e.g., Park, Barnett and Nam, 2002; Holmberg, 2009).

Although social network analysis methodologies may be applied to the web and information on it, it has a far broader set of applications, so the terminology generally used is one of nodes and edges rather than web pages and hyperlinks. A network (or graph) comprises nodes (sometimes referred to as vertices) and edges, which are the connections between nodes. These edges may be undirected or directed (in which case they may be referred to as arcs), depending on the nature of the connection. For example, web pages are connected by directional links, but just because a blog links to a BBC news story this does not mean that the BBC story automatically links back to the blog.

Introduction

Like the library in Ranganathan's (1931) five laws of library science, the web is a growing organism, and the changing nature of the web is one of the themes that will be developed throughout this book. As this chapter will show, the technologies being utilised are changing, as are the sites and services available to investigate the web. Librarians need to understand the changing nature of such sites and services if they are to make use of web metrics for either relational or evaluative purposes.

Since the mid-1990s the web has become a ubiquitous part of the modern world, and the development of associated tools and technologies has had a big impact on the nature of the web metric investigations that have taken place. During this period search engines have indexed large amounts of the web, and provided advanced search functionality that has enabled increasingly complex investigations of web content and hyperlink networks. The Web 2.0 revolution saw the major search engines and social media services provide application programming interfaces (APIs) so that developers could automatically interact with a website's content, enabling larger-scale investigations to take place. The establishing of semantic web standards promised a web where many of people's mundane computing tasks could be completed automatically by computer programs (Berners-Lee, Hendler and Lassila, 2001), while the development of cookies has enabled organisations to track users both within and across websites (Turow, 2011). However, technological progress has not been smooth and uninterrupted. Search engine and API functionality has in many instances been retracted, a semantic web has not emerged in the way some people expected and restrictions have been placed on cookies through data protection legislation, most notably General Data Protection Regulation (GDPR; Information Commissioner's Office, n.d.). All these changes have had an effect on the types of web metrics that can be developed: those metrics that were available yesterday may not be available tomorrow, but new ones will emerge. Today we are seemingly at a pivotal point in the emergence of artificial intelligence. Understanding how the tools have changed in the past can help us to understand how they may change in the future and help us to recognise potential areas of opportunity.

Increasingly every organisation, profession and service is subject to a mounting number of metrics and performance indicators. League tables are no longer reserved for the winning and losing of sports teams, but are used to make comparisons between a host of different bodies, from hospitals and schools to train companies and police forces. As the COVID-19 pandemic swept across the world in 2020, it was not just the number of coronavirus cases or deaths that gained public attention, but also how countries measured up against one another.

The growth and interest in new metrics and indicators has been facilitated by the growth and interest in the web, open data and data science. Firstly, the web facilitates the sharing of established metrics and data. Enter a term of your choice in Google's Dataset Search (http://datasetsearch.research.google.com) and you’re likely to be met with dozens of associated data sets, whether you’re interested in the latest contagious disease or in taxidermy exports. The web also enables the soliciting of new data. Web services have facilitated the ranking of politicians according to their ‘hotness’ and professors according to their helpfulness.

At the same time increasingly sophisticated data analysis and visualisation software has become widely available. At the most user-friendly end of the scale, a service such as Google Public Data (www.google.com/publicdata) provides a simple front end to a wide range of public data, while at the other end of the scale those willing to experiment with programming will discover an increasingly large set of libraries and packages that enable your data to be analysed and visualised however you want.

Most importantly, at least from the perspective of this book, the web provides new media for analysis. In the same way that the traditional media of books and journals provided the basis for new metrics, such as the number of books published in a country or the number of citations a journal receives, so the new media of web pages, blog posts and Twitter updates provide the basis for a host of new metrics. These web metrics can be far richer than those associated with traditional media as a greater variety of data can be collected at increasingly fine levels of granularity.