Crossing the qualitative- quantitative divide II

I Introduction

In this second report we carry forward the themes of working together despite and because of methodological differences by continuing a focus on methods engaged by different kinds of geographers doing different kinds of work. While particular research may suggest particular methodological ways forward – whether quantitative, qualitative, hybrid, or mashup/remix (Sui and DeLyser, 2012) – insightful scholarship for the digital age, creative research for an increasingly connected world, and both individual and collaborative efforts to contribute to that world through our research all demand careful consideration of issues of methods and methodology. Finding multiple ways to cross the qualitative-quantitative chasm takes on new importance.

Here we focus on methods that are (comparatively) new in geography, and on well-established methods reinvigorated by new approaches or data sources. Many of these methods might be what Lury and Wakeford (2012) call inventive: methods that can both enable the happening of the social world and facilitate research into its open-endedness. Such methods, they advise, cannot be external to the problems they address, but must be answerable to those problems, holding the capacity to change those problems as they unfold. Significantly, Lury and Wakeford argue, the inventiveness of methods cannot be determined in advance: methods are not intrinsically inventive, they become inventive in their engagement. Many of the new and renewed methods we here describe could therefore be termed inventive methods – made novel by their purchase on particular problems of our time.

To continue our focus on areas of convergence and mutual interest despite differences, we here highlight three areas of potential engagement across widely perceived geographical divides: communication between ‘paleo’ and ‘neo’ geography in the emerging digital humanities; mutual interests across quantitative and qualitative methods in research on mobilities; and crossing scales, senses, and the physical-human divide in rhythmanalysis.

II Crossing disciplinary boundaries through the emerging digital humanities

The second decade of the 21st century has gained multiple labels related to the huge quantities of data made available by new technologies: the age of ‘big data’ (CORDIS, 2010; Manyika, 2011; ORT, 2011), the ‘data avalanche’ (Miller, 2010), and the ‘exaflood’ (Swanson, 2007). This is accompanied by what Nielsen (2011) heralds as a paradigm shift in research methodologies, what some call the fourth paradigm – data-intensive inquiry across the physical and social sciences (Hey et al., 2009), but also the arts and humanities (Bartscherer and Coover, 2011; Borgman, 2009). Indeed, the exaflood has inundated the humanities under the umbrella terms ‘humanities 2.0’ and ‘digital humanities’. Digital-humanities research, powered by massive searchable databases, enables scholars – and the connected public – to pose new research questions some suggest could lead to better understandings of our world and ourselves (Dalbello, 2011; Manovich, 2011). The emerging digital humanities, with their ‘alliance of geeks and poets’ (Cohen, 2010), may represent a moment of methodological possibility for geographic research and scholarship, and may lead to broader efforts at crossing the qualitative-quantitative chasm (Sui and DeLyser, 2012).

Human geographers have rich traditions of humanities engagement, often through qualitative approaches. Indeed, nearly all the contributions in two volumes devoted to geographical engagements with the humanities are informed by qualitative research (Daniels et al., 2011; Dear et al., 2011). The growing efforts to digitize human traces may signal an opening in methodological approaches, for much digital-humanities scholarship is quantitatively based, or grounded in artful combinations of both quantitative and qualitative methods (Bodenhamer et al., 2010; Daniels et al., 2011; Dear et al., 2011).

Because digital-humanities data are electronically searchable, scholars gain ‘the ability to read the archive of core texts, together with their residual materiality from previous media contexts in order to produce intensive modes of engagement with particular documents, [and] groups of texts’ (Dalbello, 2011: 497), something practically impossible using the original (print) sources. In this kind of environment it is not hard to imagine that quantitative data-mining methods and GIS may have particular purchase. In fact, the ‘spatial humanities’, a subfield of the digital humanities often tied to the so-called spatially integrated social sciences, aims to leverage GIS to analyze the massive amounts of geocoded data now available (Gibson et al., 2010). Not all of this work involves GIS and/or quantitative methods; the emerging ‘geohumanities’ scholarship (Daniels et al., 2011; Dear et al., 2011) encompasses much broader ongoing intellectual efforts to engage geography and the humanities together.

But these massive databases about both past and present have made big data available to those who may not have previously considered using it – databases include digitized books, newspapers, photos, paintings, unpublished manuscripts, music, audio recordings, transactional data like web searches, sensor data, ¹ cell-phone records, social-media postings (such as Facebook or Twitter) and much more. They have unleashed efforts by scholars in the humanities to study their subjects from both spatial/temporal and geographical/historical perspectives – bringing multiple geographical modes of analysis to humanities-based scholarship (Bodenhamer et al., 2010; Daniels et al., 2011; Dear et al., 2011; see also Berry, 2012).

We share the concerns, expressed by skeptical scholars (Gold, 2012), that the current tide towards digital humanities could submerge traditional interpretative scholarship with superficial number crunching. Making sense of such huge quantities of data requires not only computationally based research methods to process and analyze the data, but also the ability to situate the results (Berry, 2012). As Borgman (2009) advises, digital-humanities scholarship should be led by those in the humanities; thus the age of big data and digital humanities calls for a new level of synthesis and synergy between qualitative and quantitative approaches in geographic research (see also Sieber et al., 2011).

Several new mega projects in this area, funded by both public and private sectors, could have lasting impacts. In the USA, Google (TOGB, 2010) and the National Endowment for the Humanities (‘Digging into Data’ initiative; NEH, 2011) and, in Europe, DARIAH (Digital Research Infrastructure for the Arts and Humanities; http://www.dariah.eu) could form part of a new research infrastructure for humanities scholarship. Google, collaborating with Harvard researchers, aims to digitize all published books. Despite copyright controversies (Adrien, 2009), they have already digitized approximately 4% of all books ever printed (approximately 5.2 million books) making them available to the public for free downloads and online searches (Michel et al., 2010; download at http://ngrams.googlelabs.com/datasets). This 500-billion-word database covers books printed between 1800 and 2000 in multiple languages (Chinese, English, French, German, Russian, and Spanish). Borrowing techniques from genomics, researchers have developed new search strategies that treat each individual word as a sort of cultural gene. This branch of digital humanities research, known as ‘culturomics’ (http://www.culturomics.org), opens, through data mining, new possibilities for research and education (Michel et al., 2010). Culturomics and Google’s ‘n-gram database’ can extend the boundaries of quantitative inquiry to a wide array of research problems in the humanities and social sciences traditionally addressed qualitatively (Lieberman-Aiden and Michel, 2011; Michel et al., 2010). Quantitative analysis of this massive corpus of words/texts can potentially enable geographers to investigate broader disciplinary trends – revealing cultural contexts and disciplinary shifts that were previously practically impossible to probe.

For example, we used the n-gram database to track occurrences of four pairs of key geographical terms between 1800 and 2000 for the books in the database (Figure 1). Note the increase, since the mid-1980s, of the term ‘qualitative’ and the significant decline of ‘quantitative’. Despite the steady decline of ‘place’ and increase of ‘space’, ‘place’ still appears twice as frequently as ‘space’. Although ‘global’ has increased dramatically, ‘local’ still far exceeds ‘global’ in the period covered. Until 1960, ‘men’ appeared far more frequently than ‘women’ in the books in the database, but since 1960 ‘women’ sharply increased with a concomitant decline of ‘men’. By the mid-1980s, ‘women’ surpassed ‘men’ in frequency of appearance, and so it has remained. Massive digital databases like n-gram can be mined to gain greater understanding of the changing context for our discipline, but though compelling, merely mining the data provides neither context, analysis, nor interpretation; here qualitative-quantitative synergies grounded in humanities scholarship may lead to new insights.

OPEN IN VIEWER

Figure 1.

Trends of four geographical themes as revealed by n-gram.

III Spanning space and time through mobile methods

Geographers have long appreciated the importance of movement and today engage with participants ‘on the move’ in invigorated ways (Cressie and Wikle, 2011; Cresswell, 2010, 2011a; Cresswell and Merriman, 2010; Shaw and Hesse, 2010; Shaw and Sidaway, 2010). While transport geography holds a generally positivist and science/social-science-based focus (see, for example, Keeling, 2009), others lend an interpretative and social-theoretical focus, as part of the ‘mobilities turn’ (Sheller and Urry, 2006). Seeking to understand the social meanings and political implications of movement, and not seeing movement as dead time on the way to destination, they foreground movement itself (walking, driving, passengering) as worthy of study (Cresswell and Merriman, 2010). Coupled to these new interests lies an upsurge in mobile research methods – methods for understanding mobilities embodied and digital, real and virtual, of people and things, for researchers in offices or mobile with their subjects (Büscher et al., 2011; D’Andrea et al., 2011; Fincham et al., 2010; Hein et al., 2008).

Some mobile methods, rather than relying on after-the-fact, stabilized retellings, prioritize the researcher’s being there, in motion, engaged in active knowledge production, seeking to understand mobile phenomena first hand. But not all mobile experiences can be shared by researchers, so mobilities researchers seek also to enliven stationary methods with the mobile experiences of others (Büscher et al., 2011; Cresswell, 2011b; D’Andrea, 2011; Fincham et al., 2010; Hein et al., 2008; Urry, 2007). We examine each in the context of one mobile method, the ‘go-along’, before turning to mobile things and the promise of mobile ethnography.

Walk-alongs, mobile walking interviews famously used by Kevin Lynch (1960), can be adapted to changing situations, with lists of questions evolving into more open-ended and spontaneous exchanges as the environment and circumstances indicate (Benwell, 2009; Carpiano, 2009; Kusenbach, 2003; Lynch, 1960; Myers, 2011; Rose et al., 2010). Often a part of mobile ethnographies but not a substitute for them, walk-alongs (and go-alongs more broadly) help reveal some of the place- and practice-based insights of participant observation without the intensity and time commitment ethnography demands. Like participant observation (but unlike sit-down interviews), go-alongs can enable researchers to ‘observe spatial practices in situ’ while facilitating discussion about those practices (Kusenbach, 2003: 463). And they can help in meeting and recruiting new participants – likely leading to a more diverse group than community meetings or snow-balling – as well as in building rapport in the community (Carpiano, 2009).

Other forms of go-alongs include ride-alongs – as with the police (Herbert and Beckett, 2009), or with commuters (by car: Laurier et al., 2008; Laurier and Lorimer, 2012; by train: Bissell, 2009, 2010a; on ferries: Vannini, 2011a, 2011b; or on bicycles: Brown and Spinney, 2010; Spinney, 2009, 2011) (and it is also possible to go-alone, as in the transect walk (Paasche and Sidaway, 2010)). Go-alongs are often used productively in conjunction with more stationary methods. So, for example, Carpiano (2009) joined groups of youth on predetermined walking routes where discussion was recorded, and followed up with sit-down interviews, participant observation, mental mapping, and neighborhood surveys (see also Benwell, 2009; Lashua and Cohen, 2010; Rose et al., 2010; Shubin, 2011; Spinney, 2009). Whether alone or in combination with other methods, go-alongs can vibrantly apprehend mobile worlds.

Evans and Jones (2011) developed a qualitative GIS technique to analyze walk-alongs focused on place understanding. Finding a measureable difference between walking and sedentary techniques in the production of place narratives both in terms of their quantity and spatial specificity to the study area (walking-interview data are profoundly informed by the landscapes they take place in), they acknowledge that a technocentric analysis risks emphasizing locational above human elements, but argue that a qualitative GIS approach offers great potential for engaging planners and policy-makers with the importance of local connections to place.

Mobility as an experience shared by researchers can be problematic. Brown and Spinney (2010) found that cycling along could be dangerous (in urban traffic, on mountain trails), physically too demanding (racing, BMX riding), an experience that could not be truly shared (on a narrow trail or skinny urban bikeway when interaction between rider and riding researcher was impossible and the experience of following was not the same as that of leading), and potentially disruptive of the experience it sought to understand (solo rides). Instead, in part, they used headcams and follow-up interviews (‘talking through’ – p. 150): riders discussed their embodied experiences of mobility at length, while watching the real-time videos which, they argue, enabled deeper engagement in the mobile practices of cycling than had the researchers actually been there. Similarly, dashboard-mounted video cameras enabled Laurier and Lorimer (2012) to follow up on their ride-alongs with groups of automobile commuters – the two methods together showed how the vehicle itself becomes a setting for family- and group-based social experiences, revealing commuting as socially organized by all those in the vehicle. Thus, video can bring researchers closer to the experiences they cannot completely share, offering insights when, for the researcher, going along would be impossible or undesirable (Brown and Spinney, 2010; Fincham et al., 2010: 8; Laurier and Lorimer, 2012; Spinney, 2011; see also Garrett, 2011).

Whether the researcher is mobile or not, Bissell (2010b: 58) cautions against privileging active and intentional aspects of mobilities and mobile bodies over those more fragile, pacific, and stationary, for mobility also involves ‘weariness, tiredness, lethargy, hunger and pain: all hugely significant entanglements of body and world that might be precipitated through movement’ but readily overlooked by mobile methods.

Because mobile phones enable people to be contacted wherever they are, and because mobile phones (along with their capabilities – GPS, location tracking, camera, sound recording, pedometer …) have spread so widely, engaging research through mobile phones can facilitate engagements with people mobile and dispersed (Berry and Hamilton, 2010; Hein et al., 2008; Kwok, 2009; Pfaff, 2010a, 2010b). Mikkelsen and Christiansen (2009) used SMS (text) messaging to conduct a rolling survey of children’s mobilities, querying five times daily about activity, mobility, location, and companionship. And Pelckmans suggested that mobile phones’ abilities to call, calendar, and record make them research assistants (2009). Because mobile phones facilitate communication, they enable ongoing connections to respondents in the field, even when the researcher has returned to the office – something requiring more methodological consideration (Pfaff, 2010a; see also Sui and Goodchild, 2011).

The mobilities turn is not confined to studying people’s mobilities. ‘Follow-the-thing’ methods trace objects’ life-cycle journeys, from production through consumption, engaging cultural meanings and use practices (Appadurai, 1986; Cook, 2004; Marcus, 1995). Vitellone (2010: 869) traced the syringe through its movement to understand how the object itself ‘makes space meaningful’. Pfaff (2010b) followed one mobile phone in Africa, tracing its importance in cultural practices to show how the phone lies embedded in processes of exchange and abandonment that influence its meaning and variable value; any particular phone plays a role in individual expression and identification. Ramsay (2009) followed tourist souvenirs from points of production and purchase in Swaziland to homes in the UK to reveal the complex (spatial) relations between people, souvenirs, and places. Schwanen (2007) traced objects using Hägerstrand’s (1970) time geography and actor-network theory (ANT) to reveal delicate interrelations and interactions of spacing and timing on both humans and artifacts. Soon, with rapid-response code and RFID (radio frequency identification) chips embedded in more objects, and the development of the Internet of Things (Ashton, 2009), tracking the mobility of individual objects throughout their entire life cycles will become easier, with profound implications for human-geographic research on things in motion and, along with that, ethical concerns for the people whose lives these things touch and thereby monitor (Beer, 2010; Dodge and Kitchin, 2007).

As D’Andrea et al. (2011) observe, mobilities methods and scholarship has tended to focus on individual, daily, even local mobilities rather than structures, processes, and systems that hinder and/or facilitate mobilities, or within which mobilities lie embedded. In a world of global flows, mobilities researchers – and ethnographers more broadly – also seek to engage multiple (mobile) methods, mustering the diverse tools of ethnography in ‘mobile’ (Ingold and Vergunst, 2008), ‘multi-sited’ (Marcus, 1995), and even ‘global’ (Burawoy et al., 2000) ethnographies (see also Büscher et al., 2011; D’Andrea et al., 2011; Fincham et al., 2010; Urry, 2007). In what Blok (2010: 509, 523) calls a ‘mobile ethnography of situated globalities’, local and global, rather than existing in dichotomy, can attain the ‘fractal character of situated co-presence’. The task, he argues, for today’s mobile ethnographer, is to ‘understand the very production of … localities and globalities’, while heeding the ‘forging [of] globalities-in-the-making’. Because ‘globality is our never-ending common project’ mobile ethnography can help us better understand – and intervene in – our world (Blok, 2010: 514, 516, 525).

IV Crossing scales, senses, and domains through rhythmanalysis

One research method, rhythmanalysis – linked to mobile methods but an approach of its own – merits its own section because rhythmanalysis can help cross the quantitative-qualitative chasm by connecting multiple scales, senses, and domains in geographic research.

Rhythmanalysis, the study of spatiotemporal rhythms and the dynamic time-spaces such rhythms create (at the bodily, institutional, urban, regional, national, and even global scales), was suggested by Henri Lefebvre (2004: 8) who saw that rhythm could ‘reunite’ quantitative and qualitative analyses. Geographers have been interested in the rhythms of both nature and society for some time (Edensor, 2010a; Edensor and Holloway, 2008; Kärrholm, 2009; Mels, 2004; Simpson, 2008). But only in recent years have we analyzed rhythms as a tool to reveal how they ‘shape human experience in timespace and pervade everyday life and place’ (Edensor, 2010c: 1) to reveal new possibilities for methods in both physical and human geography (Edensor, 2010b). Analyses of various rhythms draw out spatiotemporal tensions: Edensor and Holloway (2008) show how tourism’s rhythms reveal a more thoughtful (not duped) tourist; Hall (2010) shows how life rhythms of homeless persons conflict with rhythms imposed by urban managers; Meadows (2010) shows how insomniacs’ diurnal rhythms may fall out of synch but also lead to heightened watchfulness; Hornsey (2010) and Spinney (2010) show how pedestrians’ and bicyclists’ rhythms run against norms of vehicular traffic; and Conlon (2010) shows how asylum seekers resist the rhythms of consumerism. Others reveal how spatiotemporal rhythms help produce our experiences of place and space across different scales (Edensor, 2006, 2010a; Jiron, 2010; Jones, 2010; Kärrholm, 2009; Pinder, 2011; Vergunst, 2010; Wunderlich, 2008).

Conceptually, rhythmanalysis may enable a shift in geographic focus from one ocularcentric to one more auditory: according to Lefebvre (2004: 87), a rhythmanalyst is ‘capable of listening to a house, street, a town as one listens to a symphony, an opera’. Using rhythmanalysis, Obert (2008) revealed viewers’ experience of TV time as cyclic (contrary to the linear time evoked by print media), reframing the supposedly passive viewer as active listener, implicated in the ‘production of televisual tempo’ (p. 415). Such a shift of metaphor in geographic discourse and method could have profound ontological as well as epistemological implications (Sui, 2000).

Rhythmanalysis can also cross domains to illustrate how various embodied rhythms become practiced second nature, linking humans to music, machines, and animals (DeLyser, 2010; Evans and Franklin, 2010; Hensley, 2010). Reaching across human and physical geography, rhythmanalysis can show how tidal shifts impact human understandings of place (Jones, 2010), and how the rhythms of climate change and impending ecological disaster themselves demand new social and economic rhythms (Evans, 2010).

Interdisciplinary research on the security implications of global climate change aims to link natural rhythms (global climate change) to rhythms of socioeconomic/political dynamics (conflicts and human well-being) (Brace and Geoghegan, 2011). This scholarship embodies the spirit of hybrid geography conceptually, and shows a synergy of qualitative and quantitative approaches. Climate-change conflict scenarios often rely on the resource-scarcity thesis (Shlomi, 2011; Tol and Wagner, 2009). While existing studies indicate that this link is plausible, statistical results are mixed for a general link, in the absence of other factors, that makes armed conflict more likely (Buhaug, 2009; Burke et al., 2009; Hsiang et al., 2011; Mazo, 2010; Salehyan, 2008). A more holistic approach can capitalize on both climate and social/behavioral sciences in order to better understand climate change’s complex security implications. Linking environmental-security research to rhythmanalysis may be an important direction.

Yet being aware of rhythms is not always easy or natural, for rhythms often become taken for granted (Edensor, 2011c). Rhythmanalysis, then, requires particular attention to methods to appreciate rhythms. For example, Brown and Spinney (2010) found that head-cam video with cyclists gave visibility to the ordinary and mundane rhythms of cycling that can so easily be taken for granted – the rhythms of breathing and pedaling became more noticeable on video combined with follow-up interviews than in interviews alone.

Though the literature on rhythmanalysis so far has been predominantly qualitative, the sequence-alignment method, developed by molecular biologists for DNA analysis, may offer a promising quantitative method for rhythmanalysis (Shoval and Isaacson, 2007). Spectral analysis – a quantitative method from geography’s quantitative revolution (Rayner, 1971) – could be another useful method for quantitative rhythmanalysis.

V Summary and conclusions

Sweeping changes in communications technologies characterize our age; much of the world is aflutter with Facebook, SMS messages, and Twitter. But even in a day when politicians break their careers in a single (embodied, ahem) public tweet (Parker, 2011), some proclaim these 140-character messages too brief for significance. Yet the US Library of Congress found otherwise, and committed to archiving them all – already more than 50 million tweets per day (billions since Twitter’s 2006 inception) – in order to better document social, economic, and political trends, and improve stewardship of born-digital materials, enabling scholars of tomorrow to better understand our present (Raymond, 2010). Even important ideas, some argue, can be shrunk to 140 characters (Simon, 2011). Indeed, as the Arab Spring of 2011 revealed, some such messages have the ability to change the world (Howard et al., 2011). That, we suggest, will surely change the future of research.

In fact, it already has. Geographers utilizing new digital data sources and engaging new virtual communities have already transformed what counts as data in our field – and in the process have been able to suggest novel understandings of place and space. As de Freitas (2010) points out, wireless information technology transforms understandings and uses of (urban) public and private space: from bounded and distinct entities to entangled realms where users can sit publicly while engaging in private work or conversation, or sit privately while engaging in public work or conversation. The transformation, argues de Freitas, leaves physical space fixed, and digital space fluid, while both kinds of spaces fundamentally overlap and coexist, each shaped by the other – and their users. In this environment what we understand as public space extends across both the physical and digital realms, and the research challenge is methodological, one presenting new opportunities for imaginative and inventive research techniques in both the online and offline, qualitative and quantitative worlds.

In this report we have reviewed methods novel and/or reinvigorated – methods geared to take advantage of these changes in different ways. Because we view our remit as one of consilience, our focus has been explicitly on three areas where scholars apparently on different sides of a divide (paleo/neo, qualitative/quantitative, physical/human) have endeavored to move their research forward: the digital humanities, mobile methods, and rhythmanalysis. With so much in flux particularly in the realm of computing power and the availability of new data sources, it is not surprising that new methods and new versions of old methods have emerged to propel (geographical) scholarship forward. Just the same, the pressure to use ‘new’ methods may cause over-emphasis on the novel (Travers, 2009) while even new forms of data and an ever-changing world have not rendered all old methods redundant or ineffective – far from it. But that will be the focus of our third and final report.