The Future of Tracking Tourists’ Behaviour and Mobility

The field of tracking tourists’ mobility is a rapidly evolving space. In the eighteen months that it has taken to write this book, many innovations, along with world events such as COVID-19 have emerged, which have required updates to be made to this manuscript. There is no reason to believe that these changes will not continue to be necessary, as technological innovations are likely to occur at a rapid pace and will, no doubt, be utilised by those involved in tourism research. The purpose of this chapter is to attempt to investigate the future of the adaptations that are likely to occur with regards to tourist tracking technology and methods. A near-future gaze is taken as technology and world events are evolving so quickly that it is difficult to predict a future beyond the short term. Techniques such as physiological tracking, emergency management, indoor positioning, machine learning and artificial intelligence are assessed along with the future of ethical research conduct. A summary is also made where the pros and cons of each research method is assessed and finally, future research needs are highlighted.

Tracking via Bespoke Research Apps

In the past ten years, several apps have been developed by research teams with the specific intention of tracking tourists. These apps contain user interfaces that explicitly communicate the function of these apps. They differ to apps, websites and social media sites described in the previous chapters, because the primary function of their user interface is to track and understand the characteristics of tourists. This form of tracking produces highly detailed, tourism-specific data which can be of great value to tourism researchers. But it is not without its challenges. These will be explored in this chapter.

Tracking via Volunteered Geographic Information

Over the past 20 years, the use of location-based tracking has become increasingly popular. The introduction of GPS technology into devices such as phones and watches, and its incorporation into tracking apps, has led to widespread use of apps which track activities, particularly those of a sporting nature. There are now over 318,000 health and fitness apps – called mHealth apps (Byambasuren et al., 2018) – and it is estimated that 75% of runners now use them (Janssen et al., 2017). Many of these apps contain the ability for users to track their movement and share it with fellow app users – Strava alone has 42 million accounts with 1 million users each month (Haden, 2019), but others include MapMyFitness, Adidas Running, and Google Fit. Importantly for this book, the data that is produced from mHealth apps is continuous point geo-referenced data that is visualised for the user as a defined route undertaken during a particular activity. This route, and the temporal and spatial aspects of the activity, can be viewed by the user and then released online for their online network to view. Most commonly, it is referred to as volunteered geographic information (VGI). The data that is generated from mHealth apps can be sourced by researchers; this is often referred to as crowd sourcing. Researchers can gather large amounts of data of entire paths taken by individual users, either via gaining consent from individual users to share their routes, or via APIs provided by the app developer which provide access to large amounts of routes and their associated statistics. VGI provides researchers with great potential to facilitate research that assesses tourists’ movement through space and time (Heikinheimo et al., 2017). However, as is the case with single point geo-referenced data (discussed in the previous chapter), research in this space is disparate and tends to focus on one platform at a time, or one context at a time. The rapid increase in VGI is arguably due to three factors: developments in wearable technology; developments in location based technology that has been integrated into smart phone and watch apps; and an increase in usage of urban spaces for walking, running and biking. The latter is largely due to an increased interest in healthy lifestyles and exercise (Santos et al., 2016; Brown et al., 2014) and presents issues for park managers, including those related to environmental impacts due to overuse and conflicts between different types of users, such as walkers and bike riders (Santos et al., 2016; Norman and Pickering, 2017; Pickering et al., 2011; Rossi et al., 2013). This chapter will explore how VGI data can assist researchers and managers in understanding these issues, along with tourists’ mobility.

Introduction

In the past twenty years we have seen changes in technology that have reconfigured the way in which tourists plan, travel, reflect and share experiences. These changes have caused us to reconsider how tourists travel and how they make decisions, as well as how destinations market themselves. The now ubiquitous use of mobile phones has been documented as being a major influence (Wang, Park and Fesenmaier, 2012). Yet, while large swathes of research have focused on the use of technology and the impact that technology has had upon tourists’ decision making, there is comparatively far less research that concentrates on using technology to understand where tourists travel to, and how they move between destinations and attractions. The tourism industry has been documented as lagging far further behind than other industries in its use of technology, particularly that which delivers research insights (Eccleston, Hardy and Hyslop, 2020). The reasons for this have not yet been explored in great detail, but they are quite possibly due to the fact that the tourism industry is dominated by small to medium sized businesses whose capacity for expenditure on the use of technology and research is limited, relative to other industry sectors such as mining and forestry. A second reason is that tourism is reliant on an element that is often far harder to control – people. Unlike sectors that use biological elements as their key resources, and can place sensors where needed without requiring consent, tourism’s reliance on humans and their interaction with technology makes tracking far more complex. A third reason is that tourism arguably lags behind other sectors because the methods available to the industry to track and understand mobility involve complex technology, and different methods require specialist analytical skills. The Director General of the World Health Organisation, Tedros Adhanom Ghebreyesus, argued that decision makers are facing an ‘infodemic’ as a result of large swathes of data being made available in order to assist understanding the impacts of the COVID-19 pandemic (Zarocostas, 2020). The plethora of options facing the industry in regards to which technology to use and how, is undoubtedly adding to this lag.

Tracking via Geotagged Social Media Data

Over the past twenty years, social media has changed the ways in which we plan, travel and reflect on our travels. Tourists use social media while travelling to stay in touch with friends and family, enhance their social status (Guo et al., 2015); and assist others with decision making (Xiang and Gretzel, 2010; Yoo and Gretzel, 2010). They also use it to report back to their friends and family where they are. This can be done using a geotag function that provides a location for where a post is made. While little is known about why tourists choose to geotag their social media posts, Chung and Lee (2016) suggest that geotags may be used in an altruistic manner by tourists, in order to provide information, and because they elicit a sense of anticipated reward. What is known, however, is that the function offers researchers the ability to understand where tourists travel. There are two types of geotagged social media data. The first of these is discussed in this chapter and may be defined as single point geo-referenced data – geotagged social media posts whose release is chosen by the user. This includes data gathered from social media apps such as Facebook, Instagram, Twitter and WeiChat. The method of obtaining this data involves the collation of large numbers of discrete geotagged updates or photographs. Data can be collated via an application programming interface (API) provided by the app developer to researchers, by automated data scraping via computer programs, perhaps written in Python, or manually by researchers. The second type of data is continuous location-based data from applications that are designed to track movement constantly, such as Strava or MyFitnessPal. Tracking methods using this continuous location-based data are discussed in detail in the following chapter.

Tracking via Bluetooth and Wi Fi

The technique of tracking tourists’ mobility using Bluetooth and Wi-Fi technology has emerged as a reliable and viable option for tourism planners and researchers (Shoval and Ahas, 2016; Musa and Eriksson, 2012). Recent studies have employed Bluetooth to measure the time it takes for people to pass through security (Bullock et al., 2010); assess movement flows at festivals (Versichele et al., 2012); and explore movement through cities (Verischele, 2014). Bluetooth has also been used to track high speed movement, such as car and cyclists, whereas Wi-Fi scanning, which takes a longer time to capture a signal, has been used to assess the flows of slower moving objects, such as tourists on foot, or other pedestrians (Abedi et al., 2013). Tracking using Wi-Fi or Bluetooth offers researchers the ability to track vast amounts of data on movement in a relatively short period of time. Verischele et al., (2012) describes the scanning of Wi-Fi and Bluetooth signals as ‘non-participatory’ research because individuals are not required to sign up and participate to studies of this nature, nor are they aware they are being tracked. The advantage of this approach is that tourists do not change their behaviour because of the knowledge that they are being tracked. This chapter will now review these forms of tracking technology, along with their advantages, limitations and ethical implications.

Mobile Phone Tower Tracking

Tracking tourists using mobile phone data involves collating mobile phone call detail records (CDR), that can determine travel patterns of mobile phone users. The size of the data involved in this style of research is enormous; Xiao, Wang, and Fang (2019) received 600 – 800 million records per day when they used mobile phone data from Shanghai, resulting in over 10 billion mobile phone trajectories. However, mobile phone data does not provide precise travel itineraries. Rather, the data is a series of time-space points, showing where mobile phone users were when they made or received calls or text messages. Inferences are required to determine which mobile phone users are tourists, and when they entered countries or regions. However, the ubiquity of mobile phone use and the size of the data sets available to researchers means that this form of data can be used as a proxy for accommodation and visitation (Xiao, Wang, and Fang, 2019; Ahas et al., 2008; Ahas et al., 2007). Many significant findings regarding travel behaviour have emerged from this technique, including understandings of the impacts of seasonality, the impacts of nationality, and the impacts of events. This chapter will review these findings as well as the challenges that arise from the use of this data.

The Ethics of Tracking

Research that tracks tourists’ movement challenges our perception of ethics, privacy, and consent. The introduction of technology with the capability to track tourists in fine grained detail is viewed by some as a gross invasion of privacy, by others as a personal safety mechanism, and is treated by others with almost complete ambivalence. Importantly, in the past fifteen years we have witnessed a great change in the way in which tracking has been viewed by study participants and the general public, along with many mysterious contradictions in our acceptance or resistance to privacy – possibly fuelled by media attention around this issue. In the early 2000s, apps began emerging that conducted GPS tracking covertly in the background. For example, flash light applications (henceforth referred to as ‘apps’) that many of us had on our mobile phones, appeared to be a useful app. However, the business model of these apps was that they tracked users’ movements in the background of the app and on-sold this data to marketing companies. Similarly, The Weather Channel app was recently exposed for on-selling tracking data that was covertly collected, resulting in a legal case against its owner, IBM. In 2017, it was estimated that 70% of apps track and share user information with third parties (Vallina-Roderigue and Sundaresan, 2017). While there is resistance to some forms of tracking, there appears to be acceptance of other types. Strava is one such example. It is estimated that each week, 8 million activities are uploaded onto the app (Goode, 2017). Every 40 days, the app adds one million users (Craft, 2018). It is used by recreational hikers, bikers and runners, who wish to track and share their activities. It is widely known that the business model of Strava is built upon on-selling this data to cities and councils. This practice seems to be widely accepted by users.

Tracking Using GPS Technology

The use of global positioning system (GPS) technology underpins many different methods of tracking. GPS tracking involves the use of a beacon that sends the location of a device to satellites to determine the precise location of the beacon. In recent years, technological improvements have meant that GPS tracking units have become exponentially smaller in size. Whereas early portable beacons such as the Magellan (launched in 1989) were 22 cm in length and around 700 grams in weight, if not larger than television screens, they can now fit into the back of watches and mobile phones (Shoval and Isaacson, 2010). This chapter will explore the development of GPS technology and its application to tourism research, when utilised with portable GPS loggers.

Tracking Tourists’ Mobility via the Internet

Tracking tourists’ mobility and migratory patterns may be conducted by collating their digital footprints via the web. Data of this sort may be sourced via apps such as Google Maps, or websites that collate IP numbers and their proximity to mobile phone towers. It may also be collected via big datasets such as ticketing websites, via mini programs such as those used by WeChat, and via non-big data sources such as blogs. This form of location-based tracking is a highly efficient and cost- effective means of understanding where consumers are located. The devastating impacts of the COVID-19 pandemic upon the tourism industry have clearly indicated the potential for tracking via the internet to assist the tourism industry. Google’s analytical data that was released publicly in March 2020 provided an excellent example of this – both in terms of the insights that can emerge from data of this type, and consumers’ perceptions of the ethics of this form of data. This chapter will explore the technique, including the types of location-based data that can emerge from websites, the conceptual learnings that have emerged from this technique, and, importantly, the ethical implications of this form of data.