Ethical and critical dimensions in learning and teaching: the case of geodemographics

Abstract

Geographers face a challenge in reconciling the more formal signalling of ethical and moral dimensions in our practice, with the growing emphases on business relevance and graduate employability. This is most strongly felt in the sub-field of geodemographics, a valuable modern skills vehicle with a clear pedigree in quantitative social geography dating back to the 1970s. This paper asks how the students of this sub-field might be enabled to construct their own ‘moral geographies’ after recounting the very different attitudes and conceptions of the GIS ‘cheerleaders’ and their academic critics. It suggests, finally, a simple starting point for discussing and classifying critical and ethical issues.

Introduction

’Managers like information and reports. If the report includes a map they can suddenly become your best friend or start to besiege you with further requests for similar reports. … In Chapter 12 we explain the best ways to maximise the impact of maps and graphics.’

(Chainey and Ratcliffe 2005, p. 25)

Higher Education (HE) tutors face a challenge in scholarship and teaching arising from the more formal signalling of ethical concerns (Kearns et al 1998; Cutchin 2003; Valentine 2005), and the saliency of corporate interests in such matters as graduate employability and curricular relevance (Owen 2001; Brown 2004). It seems to be increasingly difficult to maintain ethical responsibility and critical credibility, at the same time as orienting students’ skills and attitudes toward entrepreneurial and industry-relevant outcomes.

This paper aims to chart the troubled terrains encountered in teaching the subfield of geodemographics at Honours level. Following Valentine (2005, p. 486), the paper goes on to propose a framework for students themselves to negotiate the ethical and critical issues which are largely avoided in the key texts. The result, it is hoped, will be seen as a more complete ‘reading’ of geodemographics, allowing proper scope for ethical and critical evaluation. In so doing, at a fundamental level, learning how can be extended into learning why, a process that can only enhance the position of similarly ‘applied’ studies in the HE curriculum.

Geodemographics has existed as an Honours option at Bath Spa for twelve years. At inception, and since, part of the rationale was to recast urban social geography into a more practical, applied and career-relevant sub-field of geography. A prominent theme linking theory and practice has been the conceptual and methodological legacies of interest in neighbourhood types and patterns of deprivation, carried through the inductive innovations of the 1960s to the modern commercial products and their uses today.

This narrative may seem overly ‘whiggish’ in outline, though the detail has a rougher, less predictable quality owing as much to changes in conceptual and methodological direction and the work of individual facilitators as to the onward march of technology. Census data are easily available, and critiques of the data and their geographies are a staple of student project work and exam answers. Until recently, ethical considerations have been ‘dealt with’ only by reference to regulatory minima, in relation to census anonymity and data protection. The focus on the census as the main data source is now outdated as non-census databases have mushroomed in the UK in the past five years (Longley, 2005).

The twin focus at Bath Spa on ‘public sector’ applications, related to long-standing academic interests in urban structure and social inequality, is also now questionable in a truly balanced account. Students, lately, have been encouraged to think about the implications of the ‘surveillance society’ in one class towards the end of the semester, but the critical and ethical ‘content’ of the syllabus has been limited and formal. There is no obligation on students to include ethical judgements in their projects to date. These are conceived as simulated briefs from companies or organisations with objective GI needs that the students are encouraged to meet in ways that maximise their virtuosity with data summary and visualization techniques, rather than exercise any ethical or critical oversight of the brief itself. In other words, the students have been implicitly placed in subaltern positions, responding as ‘junior employees’ rather than ‘consultants to the board’.

Geodemographics is an important demonstrator of geography’s applicability to the informational needs of a wide range of contemporary organizations, in both the public and private sectors. It belongs to a larger set of GIS techniques the practitioners of which are beginning to assert independent conceptual rationales (e.g. Schuurman, 2004) alongside more conventional ‘how to’ texts (e.g. Harris et al, 2005). The usefulness of GIS and GI-related skills to graduate employability is surely beyond doubt (Brown, 2004), though their attractiveness to students who may lack confidence in more conceptual studies does need exploration. It may be argued by some that sub-fields such as geodemographics, even GIS as a whole, have no place in the HE curriculum if only that students are entitled to a brief period of respite from corporate interests, and such interests can only distort, or undermine, the pursuit of intellectual excellence.

Encouraging undergraduates to take applied options, such as geodemographics, risks compromising their intellectual independence at a critical period in their development. A more practical objection might observe the short-term rise and fall of applied skills as technology continues its headlong rush, and that HE should be about enabling more generic capacities of intellectual development. I agree that students should have an informed choice of optional studies to suit their abilities, interests and career aspirations. It behoves us as tutors and guides to deal with critical and ethical questions more openly, as part of the curriculum. This includes going beyond the standard literatures, which can broadly be divided between ‘cheerleaders’ and ‘critics’, and include consideration of corporate responsibility, human rights and academic discourse.

The ‘cheerleaders’

The mainstream GIS literature has progressed beyond the ‘how to’ stage, though given the complexity of the field it is understandable that technical content should continue to dominate (Harris et al, 2005; Longley, 2003, 2005, Sleight, 2004). However, the critical terrain around the practice of geodemographics is rather narrowly defined. Sleight (2004) acknowledges that the central clustering technique demands subjective judgement by someone with experience: ‘one needs to be able to recognise an ‘acceptable’ solution.’ (91) The labelling of the resulting types is ‘a difficult process’ and can take much longer than the original analysis’. (92) Geodemographics ‘should’ not be about constructing differences where none exist, but the reifying of neighbourhood types is seen largely in the narrow sense of arbitrary boundaries and the modifiable areal unit problem (Harris et al 2005, p. 210).

Crucially, pragmatism is extolled in favour of ‘academic theory’ in orienting the reader towards business attitudes. ‘Does it work for me?’ is the test, and the solution appears to be that it does, though most of the evaluations are taken from the public sector, illustrating starkly the commercial sensitivity of even this most crucial defence for geodemographics, that it is cost effective. Clusters are seen in this rationality as benign. As the regulatory environment has become more hostile, and as more consumers elect to invalidate the use of their valuable lifestyle data, cluster-based neighbourhood profiles can ‘fill the holes in databases’ without recourse to intrusive personal data. This is interesting, as the long foretold demise of Census-based neighbourhood typologies may not now occur. ‘Rooftop marketing’ may be too intrusive even in the US in the long run. The insider dilemma appears to be between using ‘accurate’ but intrusive personal data, or stereotyped but anonymous clusters. The opinion from the cheerleaders is pretty clear!

Schuurman (2004) contrasts the external view of GIS as overwhelmingly positivist in epistemology but suggests that pragmatism as actually dominant. She endorses Foucault’s (1979) idea that governments and businesses operate within specific ‘rationalities’, and designates the post 9/11 developed world as existing under a state of ‘carto-security’ (Crampton, 2002), accompanied by the collection, commodification and cross-matching of large inventories of personal data, much of which is available to both state and corporate surveillance technologies, such as geodemographics. Schuurman (2004) detects important cultural differences between US and European governance and regulatory mechanisms, quoting the chair of Sun Microsystems at one extreme — ‘You already have zero privacy. Get over it.’ — (Schuurman, 2004; p. 130) and Article 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms at the other.

Ethics only make sense in cultural context, ergo the rationality dominant in the USA need not and does not apply in European geodemographics. This is heartening: the basic message is that the use of technology is dependent on the rationality, or variant, locally practised rather than being inevitably subservient to a globalising imperative.

However, rationalities can change: the tightening of UK security after 7/7 is an example. More generally, the boundary between public and private has changed through history, and people will adjust their behaviour to mitigate changes in surveillance. Public participation GIS use is contrasted with ‘rooftop marketing’ to illustrate the broad applicability of the method across a range of public goals and rationalities, though to Goss (1995) non-profit and electoral applications can be seen as special cases of ‘social marketing’ that belie the hidden intent to control (Goss, 1995, p. 172).

The critical terrain as laid out by the mainstream literature is one concerned primarily with better data, better technologies and more timely decision support. An optimistic, progressive ‘firstspace’ (Smith, 2005) narrative enshrines the idea that society produces the spatial patterns that geodemographers have discerned, mapped and made use of for economic gain and social advancement. Indeed, Longley (2003, 2005) goes as far as to maintain that GIS is on the brink of rejuvenating and enriching measures in class, consumption and citizenship (Longley, 2003, 114–5), that the ‘meaningful representation of individual behaviour in something approaching real time is becoming a reality’ (Longley, 2003, p. 118), and that GIS and geodemographics research is in the vanguard of social science research, and that the neighbourhood classifications they spawned ‘work in the real world’ (Longley, 2005, p. 61). Where an ethical dimension is articulated, this has generally been seen to lie in the direction of personal privacy and anonymity within a grand compromise, that the citizen’s contribution to better governance and more market choice is a redefining of the public and private spheres and the ‘commodification’ of personal data.

Critical and ethical dimensions

A more nuanced reading of the challenges geodemographics poses to society can be found in Goss (1995a, 1995b), Curry (1997) and Graham (2005). These take two forms, critical and ethical. The former mode is subtly and carefully composed in the rich language of the cultural turn. To Goss (1995a), for instance, ‘geodemographics displays a strategic intent to control social life’, as it is based on ‘an instrumental rationality that seeks to bring the processes of consumption further under the control of the regime of production’ (Goss, 1995a, p. 172). The critiques do include military and positivist discourses, in stirring prose, establishing for itself a privileged, Archimedian position from which to survey and manage the life of the other, the language of manipulation, omniscience, distancing and control, the use of consumer ’targeting’ and market ‘intelligence’ etc. (Goss, 1995a, p. 182–183).

Market segmentation is ‘ordered and managed’ rather than sensitively portrayed, involving nothing more than a ‘rhetoric of intimacy’ (Goss, 1995a, p. 185). Lifestyles and citizenship is crudely summarised as consumption. The wide flux of social identity is crystallized into this meagre ground, normalizing choice based on individual values and marginalizing income, workplace, race, gender and other solidarities, with the implicit privileging of higher ranking groups. (Goss, 1995a, p. 189) Neighbourhood types become hard, homogenous, reified entities ever more closely linked to ‘constellations of commodities’. ‘The targeting of specific types of neighbourhood may effect a de facto redlining of social life’, shutting out places and people from full participation in consumption.

Curry (1997) adds that geodemographics is redefining the objects and groups that make up social life, no less. Rather than reflecting social realities, classifications can result in residents feeling insulted by tags such as ‘X-Tra Needy’ and ‘Zero Mobility’ (Curry, 1997, p.691), to which one might add individual lifestyle categories such as ‘Staid at Home’ and ‘Put Kettle On’ (Harris et al 2005, p. 244). Further, the socially motivating bonds between individuals and social groups risk being replaced by assigned membership of ‘evanescent’ lifestyle formulations, dismissive of memory, history and continuity. Places constructed around the ‘bare contingency’ of geodemographics deviate markedly from the ways people actually construct places that are laden with other meanings. The result is a ‘digital puppet’ or virtual individual. ‘The filtering of information to what I ‘really want’ means I have lost control of my life.’ (Goss, 1995a, p. 694) the solution is to be able to regulate our ‘virtual selves’ according to European notions of intellectual property rights.

To Graham (2005), the ‘worlds of code’ require urgent excavation to reveal how they ‘continually constitute, structure and facilitate the place-based practices of the material world’ (Graham, 2005, p. 563) after Dodge and Kitchin (2004). This is located within a broad shift from Keynsian welfare states to splintered post-Keynsian regimes of infrastructure, service and space production and consumption. (Dodge and Kitchin, 2004, p. 565) This ‘splintering’ allied to consumer surveillance allows unbundling and customization of niche services. The mobilities and freedoms accorded to privileged groups and places can be reconfigured to exclude the unprofitable.

Further, there is a tendency that virtual placetypes, based on statistical summaries that obscure small-scale variation, become reified as social realities that provoke unequal or ‘area-based’ corporate and state actions. The public availability of these types exerts a powerful influence, crucially enabling ‘strategically inclined social groups’ to find ‘their’ place within increasingly complex and dynamic urban systems. People and communities can be actively matched. Graham foresees a general polarization as privileged groups secede from the remaining collective bonds of public services and universal citizenship, whilst those excluded from online GIS see their places and prospects worsen.

These accounts situate geodemographics in a moral geography, implicated and condemned as regressive in effect, however laudable the insiders’ declared intent. The Joseph Rowntree Foundation has joined the critical chorus, emphasising the segregative agency enabled by internet-based neighbourhood information systems, and some of the new social risks foreseen (Cross, 2005). Journalistic interest in this theme has implicated geodemographics in the mis-selling of credit by targeting a neighbourhood type — ‘Happy Families’ — said to be a ‘culture that is keen to take advantage of easy credit’ (Ronson, 2005, p. 17). The profile of a suicide who died owing £130 000 on 22 credit cards is powerful and moving, and hinges on a simple question: why are some households ‘bombarded’ with credit offers whilst others are not? This line of investigation takes Ronson to meet Richard Webber (the most influential figure in facilitating the corporate adoption of neighbourhood taxa in the UK), and claims to reveal the rapacious, impersonal nature of credit targeting, the lack of effective protection for consumers and subsequent high levels of unsustainable consumer debt.

Is it possible to compile a moral balance sheet for geodemographics? To the cheerleaders, a separation of technology and ideology of use is paramount: ‘it’s not the technology that’s evil, its what people do with it. After all maps have always been used as a military device.’ (Stan Openshaw quoted in Davies, 1995, p. 19). In this conception, it is enough that geodemographics can be demonstrated to work, by implication, and it is up to the ‘watchers’ to join the ‘doers’ in turning the technology to more benign uses.

Ethical objections include routinized ‘consumer espionage’ and loss of personal privacy. The main threats are seen as error, which is cumbersome and difficult to correct under US statutes, and data combination, which can enable covert uses not approved by the subject. Curry (1997) pursues the privacy question in some depth, providing some evidence for Schuurman’s assertions about the cultural relativity of ethics. US courts have taken a fairly consistent and implicit view of ‘autonomous technology’ such as wire tapping, aerial surveillance and tracking devices which have progressively undermined the traditional model of domestic privacy (Curry, 1997; p. 687).

The combination of data and the use of compounded datasets for purposes unforeseen by the subject is not therefore seen as an issue separate from the ‘inevitable and natural’ advances in technology. Software sorting as practised is not transparent, ironically protected by proprietary confidentiality. Transparency is essential in revealing how greater freedom for the privileged has as its corollary the undermining of the prospects for marginalized groups and communities. (Graham, 2005; p. 575) Graham finishes on an optimistic note: the inherent flexibility of systems allows grassroots GIS to create alternatives that are at once empowering and progressive.

In ethical terms, Goss (1995) sees geography challenged by geodemographics, as geography graduates are sought after for their expertise. This places a responsibility on the subject ‘due to its pedagogical role, its familiarity with theory and methods, and its increasingly sophisticated understanding of the spatial constitution of relations of power in contemporary society’ (Goss, 1995; p. 175). Geographers, then, are best placed to challenge the instrumental rationality behind geodemographics, whilst preparing graduates for well-paid work in the sector Openshaw. How is this to be done?

Firstly, it is necessary to distinguish meaningfully between academic criticism of geodemographics, and ethical concerns. Clearly, criticism often raises ethical dilemmas, but how is the student to navigate a personal position in relation to the technique and its products should they be encountered in professional life? It is important that some attempt be made to distinguish critical from ethical challenges, and to evaluate their relative magnitude in a clearly charted moral geography.

It is necessary in my view for students to be able to distinguish meaningfully between academic criticism of geodemographics, and ethical concerns around many of its applications. Clearly, criticism often raises ethical dilemmas, but how is the student to navigate a personal position in relation to the technique and its products should they be encountered in professional life? The grid (below) is suggested as a starting point. It encourages students to separate ‘problems’ with geodemographics into one or both of two columns, those that derive from academic criticism and those which pose more profound ethical challenges in the world of work. In the following discussion of the grids, individually compiled at first, levels of agreement and dispute can then contribute to a more ‘clearly charted moral geography’. The outcome should reassure students discouraged by the tone of the academic criticisms, yet they will remain aware of the ethical dilemmas as yet so little acknowledged by geodemographics practitioners.

Table

Issue:	Critical issue? (tick)	Ethical issue? (tick)
Privacy
Data accuracy
Data combination
Imposed identities
Homogenization
Segregation
Exclusion
Redlining
Secrecy
Exclusivity, priority
Filtering
Misrepresentation
Secession
Imposed values, opinions