Special Issue Editorial: Designs & devices: Towards a genealogy of audience research methods at the BBC World Service
Marie Gillespie, Hugh Mackay & Alban Webb
The BBC World Service (WS), established as the Empire Service in the early 1930s, did not develop a systematic approach to audience research until after the Second World War. For much of its early history, it operated with extremely limited knowledge of its overseas audiences. Many BBC staff at that time were deeply averse to the very idea of audience research, especially the Director General, John Reith. Intent on educating and informing audiences as much as entertaining them, staff feared that data about the tastes and preferences of audiences, in the form of crude ratings, might become the corporation’s master, rather than its servant:
any research that might be undertaken should be so controlled as to secure that it never developed from a servant into a master, to the detriment of the essential qualities of good broadcasting – a responsible but sensitive outlook and a readiness to experiment (G.A.C. 23, memorandum presented by Sir Stephen Tallents to the BBC General Advisory Council, January 1936.) 
Early BBC listener research overseas was very marginal to the work of broadcasters and a strict separation of the spheres of research and programme making was instituted. This changed during World War Two. As the need for ‘intelligence’ about radio audiences intensified, especially for propaganda purposes, research assumed more significance. But audience research did not secure for itself a firm institutional foothold until the post-war period, when sample surveys and interview methods came to dominate audience research . These devices were used to evaluate and measure the size, tastes, needs and behavioural characteristics of audiences in ways that have increasingly influenced the thinking behind and construction of the World Service’s output.
This special issue explores the development of the tools and techniques of audience research at the WS, and the political, cultural, and technological factors that have shaped and been shaped by these. The research designs and devices that have been developed and deployed at the WS are analysed not as culturally neutral tools that deliver more or less objective measures of empirical realities but as change agents in organisational life, implicated in editorial decision-making and in processes of policy formation and change. The articles in this special issue examine a range of issues: which methods become hegemonic in WS audience research, why and with what consequences; what forms of agency are enacted in different methodological repertoires; the transformative and critical potential of audience methods; and the invisibilities and absences engendered by the methods used in audience research.
The BBC World Service has travelled a long way since it began broadcasting seventy-nine years ago on a weekly programme budget of just 10 guineas (Briggs, 1995, p.355) . It has evolved into a global organisation of world-wide renown, with an operating expenditure of £217.7 million a year, speaking to a weekly audience of 180 million on multiple platforms in English and twenty-seven other languages . In the intervening period and under various titles – from the Empire Service, Overseas and European Services, to External Services and, finally, the World Service – it has reported on the shifting political, social, diplomatic and cultural landscapes of the last eight decades. In response to changing institutional, technological and strategic imperatives, audience research has become ever more complex, elaborate and important.
Coexisting with this narrative of transformation is a strong sense of continuity. The BBC, both at home and overseas, has long been associated with an independent editorial style and tone – and for that it has earned and for the most part deserved the respect that it enjoys. This independent identity and a recognisable BBC style have been critical to its credibility with audiences and to the “soft power” exercised by an organisation that has been funded by the UK government’s Foreign and Commonwealth Office since the Second World War . Its rules and tone of engagement have had to adapt to the requirements of changing geopolitical and strategic contexts in which it operates, while remaining recognisably “BBC”. And audience research has been vital to processes of adapting output to meet audiences’ political and cultural sensibilities (Skuse et al, 2011) . Without the extensive research in which the WS has invested, it is inconceivable that it would have managed to earn the reputation of trust that it has enjoyed in many parts of the world for nearly eight decades.
The trust enjoyed by WS may seem surprising given its dependence on state funding. The organisation is required to use its own audience research to report annually to the FCO according to a range of criteria which include trust and impartiality. Audience research in this sense has become an integral part of the institution as a tool of accountability but also advocacy and strategy (Wilding, this issue). These different purposes of audience research and the diverse uses of audience data (for marketing as well as for editorial purposes) complicate analysis of their performative role within the organisation. The WS commissions most of its research and uses mainly market researchers to provide ‘market intelligence’. It sets the parameters of the research in line with FCO performance indicators and targets. Under the Broadcasting Agreement (2005), the WS is required to report annually to the FCO on a variety of criteria and to conduct research which shows how WS performance compares with its main competitors . The competing imperatives of accountability procedures, strategic investment and decision-making, programme production and marketing affect and shape audience research which, in turn, directs outcomes.
The papers in this volume bring into dialogue the perspectives and practices of leading international and BBC audience researchers and those of academic audience researchers. They are the result of a collaboration that took place over three years, and continues. Some of the papers look back on the history of audience research activities at the WS, and some examine the current state of practice – descriptively, analytically, and critically. Some early research methods, and the purposes to which they were put, may now seem haphazard and even misguided while the in depth, quasi-anthropological research into everyday listening habits in the 1930s now seem positively enlightened. Understanding the complexity of what early audience researchers were doing is an important step in developing a genealogy of audience research at the WS and in our understanding of how the WS developed as it did, the challenges it faces today, and its future prospects.
The task of tracking the changing intellectual and methodological currents and the forms of technical expertise that evolved in WS audience research has only just begun. We are very fortunate to have contributions that look back on a lifetime’s work as an audience researcher (Graham Mytton, Colin Wilding and Gene Parta) at a time when the WS is facing devastating financial cuts, new governance arrangements (from 2014) that will result in the most significant changes in its 80 year history. Other papers take a very short time frame and dig deep into institutional thinking, for example at a time when knowing audiences behind the Iron Curtain was virtually impossible (Webb). But all the papers contribute to an emerging genealogy of the designs and devices of audience research and how they imagined and shaped, and continue to shape, the WS’s engagement with overseas audiences.
At the heart of this special issue lies an examination of the politics and performativity of researching audiences across national, linguistic and cultural boundaries. Researching polyglot audiences brings with it immense challenges – not least doing research that may be difficult, dangerous or impractical due to war, poverty, ecological disasters, dictatorship or media censorship. The genealogy of audience research at the WS is important because the designs and devices of audience research have been highly influential beyond the BBC’s own institutional parameters. BBC audience research has set global standards and benchmarks for international audience researchers . The articles offer an excavation of how certain methods became hegemonic at particular moments and had important consequences for how news and information from, by and about Britain circulated around the world. They offer insights into how broadcasting overseas was shaped by the intersecting forces of strategic and national interests and by the interactions and transactions between diasporic and British broadcasters (Gillespie et al, 2008).
Key to the BBC World Service’s engagement with its audiences and its cultural affinity with them has been the transnational, diasporic and cosmopolitan nature of its workforce. The ability of staff to translate and transcend national and linguistic differences, while reflecting and balancing the idioms of both, has been a critical factor in the success of WS operations. Diasporic staff have enabled the BBC’s broadcasts (which for a long time were produced in English by the Central Newsroom staff) to be translated, calibrated and embodied in the human voice. The ‘right kind of diasporic voice’ was thought to be uniquely capable of striking an intimate and influential relationship with overseas audience. Some have analysed the diasporic voice as a form of ventriloquism and domination (Hill, 2010), but it has become an essential element of the success of the WS as an international broadcaster, and vital to the exercise of British soft power for decades (Gillespie et al, 2010; Sreberny et al, 2010). Yet the diasporic nature of the WS and the significance of its translation practices have, until recently, been little researched and largely invisible to those outside its walls.
Diasporas at the World Service
The special issue draws upon a three-year project which analysed diasporic and cosmopolitan contact zones through the prism of the BBC World Service . Set up in 1932 as the Empire Service, the BBC’s overseas broadcasts pursued two diasporic goals at once. Its founder and first Director General, Sir John Reith, saw it as ‘a unique opportunity to foster bonds of understanding and friendship between the peoples of Britain’s scattered dominions and the mother country, and to bring to Britons overseas the benefits already enjoyed by the British public at home’ (Mansell 1973:1). The latter diaspora, that of ‘Britons overseas’, included British-born administrators, soldiers, settlers, experts, and expatriates. The Empire Service would keep them in touch with the motherland. A worldwide radio service for ‘Britons overseas’ made no economic sense, so Sir John Reith set out to target an alternative, and politically more opportune, diaspora: ‘the peoples of Britain’s scattered dominions’. It is primarily through audience letters that the Empire Service’s audiences are now ‘known’ to us. Listeners’ letters can be treated as ‘contact zones’ where overseas audiences and diasporic broadcasters at the imperial centre in London met, communed and communicated. Listeners’ letters constituted an important site for some of our research (Robertson, 2009).
A contact zone perspective locates transnational and diasporic subjects, and their embodied interactions, activities, networks and spaces, in historically rooted institutions and specific places, thus opening up possibilities for empirical social scientific and historical enquiries into institutional practices. Our sociological approach contrasts with characterisations in postcolonial cultural studies where the emphasis is on hyper-mobility, rootlessness and transience. For example, Clifford sees ‘contact zones’ in very general and abstract terms, and eschews the empirical and the particular in favour of a subjectivist, culturalist approach to nomadic subjects and their ephemeral experiences (Clifford 1997, pp.244-79). Our research brings a ‘grounded epistemic optic’ (Smith 2001, p.98) to bear on transnational and diasporic contact zones at the World Service, through an analysis of situated, everyday intercultural encounters and experiences – actual and virtual.
Our research, on which this special issue draws, throws up some very intriguing insights into how the BBC was able to create a model of public broadcasting, par excellence, that was replicated in other parts of the world. It also forged ways of researching audiences that placed the British empirical social science tradition, and its associated knowledge practices, research techniques and forms of expertise, in a position of considerable intellectual and technical power, albeit one that competed with the American academy (Savage, 2010, p.237). The BBC developed into a ‘centre of calculation’ where infrastructural investments in the twin pursuits of programme making and audience research beat off competition in the field and produced a ‘gold standard’ of impartial professional journalistic practice. As an imperial centre of calculation, the WS was able to transmit, circulate and reproduce its technical expertise, organisational policy, journalistic and research practices on an international level . The field of devices used by and in BBC audience research soon became a standard and benchmark for how other international broadcasters also conducted and shared research (see Mytton and Parta, this issue).
Our research on audiences is a key plank of the wider research project. It spans a number of languages and historical periods, and encompasses the shift from radio, to TV to online services (more details of which can be found at the project’s website) . Only a small selection of that wider research is represented in this volume. We have been fortunate enough to work alongside WS researchers in the Department of Marketing, Communications and Audiences (MC&A) who, in the busy hubbub of working life, proved to be remarkably open and accommodating to the academic orientation of our lines of enquiry . We were not engaged in research for the WS, but with WS researchers about WS audiences, and with our own very distinctive goals . Of course, the cultures of research are very different (instrumental and pragmatic versus theory-led, academic research). Yet we managed to forge some common ground, especially around the idea of diaspora audiences which for many at MC&A when our project began, which was quite a new concept. We had many long conversations over several years and spent a good deal of time observing and working alongside staff at MC&A. There was an element of mutual curiosity, while at the same time we maintained a critical and sceptical attitude towards each others’ research. We were allowed privileged access the BBC in-house audience data during the research period and subsequently. During this time, there have been also been some career exchanges between academic and BBC researchers. Several of our researchers were permitted to ‘sit’ in MC&A to conduct their research over long periods (the longest was two years). Since then one of our researchers has ‘gone native’ and is still at MC&A, while a BBC researcher has started a PhD at The Open University. This research collaboration continues via a Public Policy Fellowship  that aims to disseminate our academic research to a variety of audiences inside the BBC, the FCO, and internationally. The social context of this research collaboration has been important too in shaping our focus on and approach to ‘the social life of methods’.
The Social Life of Methods
This special issue draws on a prominent research theme at the joint Open University-Manchester University Centre for Research on Socio-Cultural Change, the ‘Social Life of Methods’ (SLoM) . Two key questions animate this strand of our collaborative research: What do social research methods actually do? And how are they shaped by the social world? We are concerned with how research methods construct the social world, and how the methods we use shape and are shaped by social, political and technological changes around those methods . We are not so much concerned with technical matters or with assessing the relative merits of methodological toolkits in conventional terms of, for example, reliability, veracity, comprehensiveness and so on. Important as these issues are, we are making a more assertive claim and proposing that research methods are active agents in shaping our social worlds and in producing what we know about those worlds. We are also interested in how methods become active agents and how they do what they do,
Social research methods have become a core element of every organisation today. Universities and government research agencies, including the BBC, were once at the forefront of social research. Now the private sector has the most sophisticated research tools and techniques at its disposal and gathers data on a scale impossible for academic researchers. Often in real-time, unimaginable volumes of data are generated, for example, on consumption patterns by supermarkets with their loyalty cards, or much more broadly by Google. These methods and data raise important issues for research which are explored in two of the papers in this special issue (see Mackay and Tong and Ampofo, this issue). Digital data may increase opportunities for surveillance and/or they may be democratising, but they are certainly not neutral or ‘innocent’: they are changing the social world (Savage et al, 2010). Some, with reason, perceive these developments as an imminent crisis for empirical sociology (Savage and Burrows, 2007). Commercial research and the availability of real time digital data have overwhelmed and outstripped academic research in scale and volume. Our response to this has been to work collaboratively with institutional partners and to deploy and sharpen our conceptual and analytical tools as we re-analyse and mine data and take it further than institutional researchers have the time or reason to do.
A good example of how such a SloM-informed intervention worked in practice came about early on in our research. As we were re-analysing in-house WS audience research data, we were able to document, in detail and visually, how a majority of the users of WS online sites were not living within the intended geographical target area but outside it, and could be described as ‘diasporic’. Of course, BBC researchers were aware that WS audiences were being reconfigured around new technologies, but diaspora audiences were not on their radar. On the contrary, the ‘digital diasporas’ produced in the contact zones of the WS were thought of as an inconvenient, unintended consequence of media and migration (Andersson et al, 2010). Writing up this research systematically and presenting it in different forums inside the WS drew attention to this phenomenon within the organisation and had some impact inside the WS. The research produced new knowledge about how Somali, Arabic, Persian and Pakistani diaspora audiences/users were making use of the online sites for different purposes and the findings fed in to editorial decision-making (see Andersson, 2008 and Gillespie, 2009).
If audience research is about knowledge production and the development and application of various forms of technical and methodological expertise geared to ‘knowing audiences’, then it makes sense also to consider areas of uncertainty and issues of knowledge failure. Our SloM approach invites one to consider what we don’t know or cannot (easily) know about audiences and where we lack expertise and the difference this makes (see Mytton, this issue). Academic and BBC audiences researchers alike face ever greater difficulty in understanding and knowing audiences. Institutions like the WS are currently vulnerable to unpredictable and sometimes dramatic shifts in audience behaviour and increasingly operate in a media environment of high uncertainty and risk, and intense awareness of what is not known or knowable about audiences (Nye, 2010). Given that WS funding is dependent on their maintaining audiences in target regions defined by the FCO, poor ratings or negative data about their performance can mean the closure of a particular language service. This was the case of the ten, mainly Eastern European, languages services that were cut in 2005-6 to make way for BBC Arabic and Persian TV  . Equally, a strong strategic imperative combined with good audience and performance data can be used as a weapon to increase funding and expand a service, as occurred recently with increased funding for BBC Arabic TV . ‘Known unknowns’ about audiences and the need to manage uncertainty also shape what research is done and how in important ways (Nye, 2010). Whereas in the past, the WS could depend on the loyalty of audiences across generations, especially in places where media are scarce or censored, this is no longer the case. As a result research becomes implicated in the management of uncertainty and risk and future oriented in the effort to avoid knowledge failure that could damage the organisation.
The use of audience research in making dramas for development purposes by the BBC World Service Trust (the charitable arm of the BBC network) offers a particularly apt illustration of how the SLoM approach can deliver an alternative to conventional or instrumental forms of analysis. In making ‘drama for development’ audience research is conducted before, during and after the production and becomes a much more active agent in the process of editorial decision-making than would be the case in the making of other genres. Serial dramas are broadcast over months or years (18 years in the case of the Afghan version of BBC Radio 4’s ‘The Archers’ which we examine in this volume). They have to satisfy audiences and numerous stakeholders at the same time – donors, development advisors and the dramatists on the multil-lingual production teams. Creating culturally sensitive serial dramas is no easy task especially when behaviour change in areas of life regarded as taboo (e.g. increased use of condoms) is the goal. Formative, continuous and evaluative audience research become, in that process, active agents of cultural translation, tools for development, instruments of accountability and fundraising – but not without consequences for the nature of the research itself (see Skuse and Gillespie, this volume).
A SLoM approach underscores how audiences are produced by the WS, in as much as they tune in and engage with the audio/visual/digital material it makes for them. They are also shaped by the methods employed to research and understand them. It is equally true, however, that the WS itself is defined and recalibrated by the very same feedback loops created by audience research. But of course relations between audiences, broadcasters, WS management and FCO officials are not equal or equivalent, and they never have been. Geopolitical imperatives and shifting power relations pull and push audience research practices in different directions at different moments. Our SloM approach deals with the relations and tensions between historical and political processes at particular moments via the pursuit of a series of comparative and historical empirical case studies in which we use and adapt an often innovative mix of methods. Case studies have been an important feature of our research as have, for example, our witness seminar series. These constitute a special form of oral history which brings WS staff together to reflect on a critical event or issue, and they enable particularly lucid forms of multi-disciplinary, conjunctural analyses of historical processes and contribute to the production of grounded theory .
Case studies allow us to explore how social science methods shape the organization, administration and transformation of the WS. They allow us to reflect theoretically on how ideas from different bodies of knowledge (including science and technology studies, postcolonialism, transnationalism and cosmopolitanism) might shape our efforts to develop a genealogy of audience research methods at WS. We examine contact zones as knowledge spaces and practices that simultaneously and symbiotically shape: (a) the WS as an institution (audiences, broadcasters and researchers as knowledge producers, and the systems of circulating ‘intelligence’ about audiences); (b) media representations of the world; and inflect the (c) social and political realities that are represented in and through the WS which performs as a national institution on a global media stage.
Through a series of inter-related historical and comparative case studies our core research team of 20 researchers (and many others working for short periods on very particular projects) has been able to use different points of entry to WS, and has drawn critically on a wide range of both academic and BBC literature on the history of BBC and WS audience research. In the following sections we first offer a broad historical sketch of some of the key moments and methods used in WS research. Within this context, we examine how social science methods have been used to help define, imagine and project Britain as modern, rational nation on the global media stage (Savage, 2010). Finally, we examine the shift to digital media and the significant challenges that this has posed to researchers.
From Imperial Subjects to World Citizens? Audiences Research at the World Service
From the moment the BBC’s Chairman, John Whitely, inaugurated the Empire Service on 19 December 1932 (and once technical and financial obstacles had been set aside) the BBC became a global broadcaster with a responsibility and a reach that extended far beyond British or, indeed, imperial borders. What to make of the BBC’s new audiences, though, was less clear-cut. At the end of 1929, three years before the Empire Service first crackled into life, a BBC memorandum for the Colonial Office had noted that there was ‘a real listeners’ interest’ in overseas broadcast services from the BBC and a desire to ‘participate in great occasions and exciting events’ . This was a speculative assessment, and one not always corroborated by views expressed across the Empire, particularly in Dominions territories. Nevertheless, the process of understanding overseas audiences, if only to confirm or justify broadcasting intentions, was under way. These speculative assessments based on ‘gentlemanly expertise’ were carried out by colonial officials, missionaries, BBC staff on duty tours and other moral guardians and social entrepreneurs concerned with mapping populations and managing publics in various ways for various purposes (Savage, 2010, p.93).
The degree of interest expressed by the BBC in its overseas audiences and the conscious level of engagement with them has altered over time. As a service for all those who thought of Britain as home, the Empire Service represented an attempt to establish an imperial network of the ether, in English only, with London as its imperial centre, projecting outwards. From the start, the BBC sought the reactions of their listeners – often referred to as ‘the lonely listener in the Bush’ . Audiences were encouraged to write letters offering their reactions to programmes and were sometimes offered a prize for the best submission. In 1936 the Empire Programme Director urged listeners: ‘Please don’t be discouraged by the reflection that your views may be exactly cancelled out by the next man’s, because you know his tastes are unlike yours. … Listeners’ personal experiences are the making of this service .’ Listeners’ letters were read out on air and printed in official BBC publications such as the programme guide, World Radio, which was distributed to audiences overseas as well as in internal reports.
Robertson (2009) points out that, although letters rarely survived intact, they have remained an important method of listener research even if unreliable and haphazard. To overcome this, in her research she combined an analysis of letters and reports based on studies of audience panels, with fascinating results. Although from the 1930s various kinds of audience research were conducted informally and in a rather haphazard way, it wasn’t until the 1950s that formal panels of listeners were set up in various locations across the empire to enable systematic and regular research (Robertson, 2009, p 1). From then on, the panel method, using questionnaires and interviews, became a key instrument of research especially when and where sample surveys were difficult . Robertson traces a shift in responses in listeners’ letters from the colonies and the dominions between the 1930s and the 1960s, ’from a literal and very explicit identification of BBC broadcasts with ‘home’ and ‘family’ in Britain, to a more critical engagement with the BBC as representing the former imperial centre’ (Robertson, 2009, p.16). Her work shows how listeners’ letters combined with other data offer a barometer of changing attitudes to Britain and British imperialism, and intimate insights into domestic and public radio listening practice set in the context of everyday life in the empire.
The age of empire, however, was already starting to wane and it was not long before the BBC’s editorial ambitions extended to foreign language broadcasting in 1938. Wars, whether World, Cold, or on Terror, have continually redefined the remit and purpose of the World Service, as have issues of empire, decolonisation, and development. The interplay between these contextualising influences and the BBC’s overseas audiences is reflected in the range of programming styles and voices that have been built up by the World Service. These external forces also reflect the strategic habitat in which BBC overseas broadcasting historically has had to survive. It is perhaps not surprising, therefore, that the BBC’s growing number of linguistically, politically and culturally diverse audiences quickly moved from the periphery of the World Service’s outlook to a place far closer to the formation of policy and programme making.
The WS adapted and used the proliferating raft of methods that had become hegemonic inside the BBC domestic services. Under the leadership of Robert Silvey, who was appointed the first BBC Listener Research Officer in 1936, the Listener Research Department grew in strength and sophistication until his retirement 30 years later (see Mytton this issue). Nicholas (2008) offers a beautifully lucid and engaging account of the development of the BBC Listener Research Department (1937-1950) as an introduction to the British Library’s online archive . She emphasises the particularly sociological (as opposed to purely statistical) nature of BBC audience research in its formative years, and the influence exercised by such path-breaking initiatives as the Mass Observation Movement (set up in 1937) and its participatory ethnographic methods. Her work also documents how, throughout the 1930s and 1940s, Silvey was experimenting with a range of methods (interview surveys, panel studies, the Listener Thermometer and Listener Barometer), the results of which ‘did things’ that he captured in his book on his research career at the BBC Who’s listening? (Silvey, 1974). For example, he conducted the first ever national survey of listening habits - and as a consequence, living habits - using the sample survey method. It showed, much to the shock of senior BBC management, that most people had finished their evening meal before 7pm. This data re-ordered BBC schedules for good, as did a later survey which showed that the majority of working people had their lunch at 1pm and not noon. These examples neatly capture what we mean by the social life of methods and the performative nature of research designs and devices. Nicholas shows the political and financial pressures that came to bear on Listener research. Our research too has underscored the uses of data as a weapon of defence (against government cuts or attempts to influence output) or of attack (by the Ministry of Information at times of war).
The journey of BBC audience research from being an adjunct to the business of broadcasting to a core position at the World Service is well observed in Graham Mytton’s paper in this issue. This was not a trajectory many would have predicted in the early days of the Empire Service. Rather it is a story of gradual cumulative gains, where information about audiences, as well as the equally significant knowledge gaps, not only helped to create a ‘mind’s eye’ picture of audience characteristics, but also constructed an analytical framework within which modes of analysis began to reshape the way the World Service thought about itself. The size and influence of the BBC’s audience, the reach of its programmes, audibility and reception conditions, and the impacts of outputs, were shown to have the potential to be measured. Thus a set of vital statistics were developed and used by those at Bush House and others such as the UK Foreign Office to measure the success of overseas broadcasting. The implications of methodological innovations in this early period have left an indelible mark on the history of the World Service, from the imperial letter-writer listening to the King in 1932 to today’s social media consumer, and have enduringly influenced its editorial purpose and direction.
The rudimentary methods employed by the BBC’s first overseas ’missionaries’ and ’gentlemanly experts’ reflected the mind-set of interwar imperial ambitions. World War and Cold War established an understanding of the imperatives of broadcasting into different, non-British, cultures, but now the real hunger for audience assessments was sometimes matched by the difficulty or impossibility of obtaining reliable information. The problems associated with broadcasting to closed societies are explored in the papers by Gene Parta and Alban Webb which focus on these challenges in terms of the Cold War. Webb explores how, by the late 1940s, in the absence of access to the most basic data, audience research was already an intellectual exercise engaging a wide range of institutional opinion, including policy officials, programme makers, journalists, technicians and engineers. In similar territory, Parta reveals how the appetite for audience assessments sparked a revolution in research techniques which turned brief encounters with listeners into detailed audience models.
Interaction with audiences, as Mytton points out in his paper, has been a staple and motivating component of international broadcasting from its earliest days. From the 1930s, subscriptions to BBC overseas publications such as London Calling (containing schedules), and participation in listener competitions, gave a rudimentary sense of audience geography and levels of interest in particular territories. Letters and today emails have always been considered a vital source, not just of information about reception conditions and the likes and dislikes of audiences, but also an indicator of the extent to which BBC output motivated people to write (whether in support or in disgust). The BBC has also always had a highly competitive level of engagement with other national and international broadcasters which has spurred on and shaped its own development. In 1929 the Corporation noted that ‘Short-wave broadcasting stations are springing up everywhere, and some are under the direct control of governments’. With the BBC eager to exploit the opportunity of overseas audiences, surely, they argued, Britain was ‘entitled no less than the others to diffuse its ideas and its culture’ . The close historic interaction between the World Service and the British government, primarily though Grant–in-Aid funding, but extending to matters of strategic guidance, has also been a necessary part of the BBC’s engagement with overseas audiences through diplomatic reporting of the reach and impact of BBC output. More recently, the wealth of listener, viewer and user data available, and the sophisticated models needed to manage it, pose problems of comprehension of an entirely new order. From open to closed societies like China and Iran, from a paucity of information to the digital overload, measuring audiences has become a hugely complex business (see Mackay and Tong, this issue and Sreberny, 2008).
Researching Digital Media
The shift to digital has transformed profoundly the shape and practices of all media organisations and forms. More than this, it has changed fundamentally the relationship between broadcasters and their audiences – which has become more complex, fluid, dynamic and interactive. Measurement and representation of ‘the audience’ has shaped broadcasting practices from their earliest days. With increasingly sophisticated approaches and instruments, or designs and devices, broadcasters have measured the numbers of listeners or viewers, their demographics, and their appreciation of particular programmes; and this has formed the bedrock for broadcasting. Whatever the difficulties of ‘knowing the audience’ (Ang, 1991), interactivity has challenged profoundly broadcasters’ capacity to understand their listeners, viewers and, now, users. Today the World Service is far from simply ‘London calling’, in that the audience can talk back. The flow has become two-way, not simply with audience members phoning or sending letters or emails to Bush House, but in that they are generating content. Interactivity adds complexity, requiring new instruments to measure new audiences or users who are engaged in new activities – and in an environment where many more media organisations have global reach.
In the Egyptian uprising in early 2011, Al Jazeera handed out mobile phones and provided a website where content could be uploaded. The World Service showed that it had adapted less to the contemporary environment, focusing more on checking the facts than organising itself around user-generated content (UGC). At the time of writing (May 2011) the only material coming out of Syria, where the state media are discredited and foreign journalists are banned, is UGC. The landscape is changing fast, with broadcasters potentially becoming simply platforms.
Boundaries between producers and consumers are becoming increasingly blurred – hence the coinage of terms such as ‘prosumer’ (Toffler, 1990) and ‘produsage’ (Bruns, 2006). Obviously producers, editors and journalists have never operated in a vacuum: they were always avidly ‘reading’ their cultural milieu. But today the ‘back-flow’ has moved on from phone-ins and the use of video clips that listeners or viewers or users have sent in. Programme content and even ideas for programmes and series are being shaped heavily by user debates. Some of this productive activity – for example regarding the BBC Hausa service – is to be found not on BBC forums, but on Facebook (Andersson, 2010). At the same time, much World Service material finds its way on to third-party sites; such uses considerably enhance the online traffic that the World Service generates, but it makes researching users (and the influence of the World Service) a far more complex task.
Measuring the ‘Global Conversation’
Social media complement but also challenge the role and power of traditional broadcasters. They can also play a significant role in facilitating the ‘global conversation’ which the World Service is committed to fostering (BBC, 2007). The ‘global conversation’ can be seen as a key way in which the World Service fulfils the ‘public diplomacy’ remit for which it is funded – until 2014 – by the Foreign and Commonwealth Office. ‘Public diplomacy’ is a form of ‘soft power’ that is concerned with building long-term relationships with key individuals and civil society organisations, so the focus is on engaging foreign publics and understanding their thinking (Nye, 2010). Measuring and understanding the global conversation is, of course, rather more challenging than measuring the audience of a broadcaster. Somehow the World Service has to capture data on not only who it is talking to, but who is contributing to the debates which it is hopefully fostering (Nye, 2010). It has to explore how the information and analysis that it is providing is not only received but also engaged with and re-circulated (used in discussion and ‘recommended’); and how, in these processes, the World Service might be informing opinions and shaping debates.
In this context the World Service uses the conventional industry measures of IP addresses, Unique Users (UUs) and Page Impressions (PIs). Obviously such data is limited, for example in that circumvention technologies can obscure the location of a user; a user of several devices will be counted as several UUs; and PIs tell us nothing about the significance of the encounter for the user, the sense that they make of a page, or how it might shape their social interactions. Mackay and Tong in their paper examine discussion threads on forums on the BBC Chinese Service by way of exploring how such interactivity can be analysed using discourse analysis. This, however, is labour intensive. What is used in the industry is a growing set of automated tools for quantifying and analysing interaction on the web. Webometrics is concerned with web sites, web pages, parts of (or words in) web pages, hyperlinks and search engine results (Thelwall 2009). The World Service is piloting and deploying a range of new, automated, real-time tools to develop an understanding of the networks that are engaging with BBC content and debate. Social media monitoring tools involve both web analytics programmes (to analyse the performance of specific sites on the basis of the behaviour of their users) and content aggregators (that accumulate content) (see Ampofo, this issue). Developments in natural language processing are automating content and discourse analysis. Sysomos, for example, which is used by the World Service, is a social media monitoring tool which operates in real-time, multi-lingually and globally to identify key ‘conversations’, what there is a ‘buzz’ about, and the core sentiment of posts, all of which are related to the geo-demographics of those involved, and with a remarkable level of granularity. Although in their infancy, and raising a host of issues about ethics and privacy, such tools offer possibilities for making sense of the growing complexity of media uses in the social media environment.
Our approach to these issues, as we have explained, is not that research methods measure with certainty the objective reality of living audiences. Rather, our ongoing research on the World Service examines how audiences are imagined and produced by institutions and their research methods at the same time as audience research findings are shaping organisational policies and practices. New digital designs and devices, like the panel studies, the sample survey and the interviews before them, are shaping and will continue to shape organisational practices. In some domains digital methods and data are displacing traditional research methods, as individuals and groups are mobilised around new methods, measures, forms of expertise and practices, such as citizen journalism (Gillespie, 2010). Yet, the ‘known unknowns’ of future audience configurations and patterns of uses of old and new media far outweigh the ‘known knowns’, as the former Head of Audience Research at the WS recently argued (Nye, 2010). To add to this uncertainty, the WS is facing devastating financial cuts and the biggest changes and challenges in its eighty year history posing a real threat to services to overseas audiences and to audience research at the WS. What is clear, however, is that the hard-earned trust that the WS enjoys with its audiences, the intimate relationship that diasporic broadcasters have cultivated with them, and the respect for the UK that it generates in many parts of the world are precious commodities that currently appear undervalued by the UK government.
Andersson, M. Gillespie, M. and Mackay, H. (2010) Mapping Digital Diasporas at the BBC World Service: Users and Uses of the Persian and Arabic Websites. Middle East Journal of Culture and Communication, 3(2), pp.256-79.
Andersson, M. (2008) Mapping Digital Diasporas, World Agenda, September 2008.
http://www.bbc.co.uk/worldservice/specials/1641_wagus08/page5.shtml accessed 1/6/2011
Andersson, M. (2010) Changing production practices of news and new audience configurations at BBC World Service: a case study of BBC Hausa on Facebook. Masters Research Dissertation, The Open University, Milton Keynes.
Ang, I. (1991) Desperately Seeking the Audience. Routledge, London.
BBC (2007) World Service Operating Agreement.
Briggs, A. (1995) The History of Broadcasting in the United Kingdom, Volume II: The Golden Age of Wireless, Oxford: OUP.
Bruns, A. (2006) Blogs, Wikipedi, Second Life and Beyond. From Production to Produsage. Peter Lang, New York.
Clifford, J. (1997) Routes: Travel and Translation in the Late Twentieth Century, Cambridge, Mass. and London: Harvard University Press.
Gillespie, M. Pinkerton, A. and Baumann, G. (2010) South Asian Diaspora and the BBC World Service: Contacts, Conflicts, and Contestations. South Asian Diaspora, 2(1), pp.3-23.
Gillespie, M. (2010) Real-Time Social Networking, Digital Diplomacy and Mobile Methods. Paper presented at the Conference on International Broadcasting Audience Research (CIBAR). Radio Netherlands Worldwide (RNW), Amsterdam, The Netherlands 14th – 17th November
Gillespie, M. Herbert, D. and Andersson, M. (2010) The Mumbai Attacks and Diasporic Nationalism: BBC World Service Online Forums as Conflict, Contact and Comfort Zones. South Asian Diaspora, 2(1), pp.109-29.
Gillespie, M. (2009) ‘Anytime, Anyplace, Anywhere’: Digital Diasporas and the BBC World Service. Journalism: Theory, Practice, Criticism, 10(3), 322-25.
Gillespie, M. Webb, A. and Baumann, G. (2008) Broadcasting Britishness, Strategic Challenges and the Ecology of Overseas Broadcasting by the BBC. Historical Journal of Film, Radio and Television, 28 (4), pp.453-58
Hill, A. (2010) The BBC Empire Service: the Voice, the Discourse of the Master and Ventriloquism. South Asian Diaspora, 2(1), pp.25-38.
Mansell, G. (1982) Let Truth be Told: 50 Years of BBC External Broadcasting, London: Weidenfeld and Nicolson
Mansell, G. (1973) Broadcasting to the World: Forty years of BBC External Broadcasting, London: BBC.
Nicholas, S. (2006) The Good Servant: the Origins and Development of BBC Listener Research 1936-1950, http://www.britishonlinearchives.co.uk/guides/9781851171248.php.
Nye, J. (2010a) ‘Known Unknowns: Working with Uncertainty in the Digital World. How the BBC Global News Division Copes with Uncertainty.’ Unpublished paper presented at ‘Designs and Devices: Audience Research at the BBC World Service 1932-2010’, CRESC Annual Conference: Culture and Citizenship, St Hugh's College, Oxford, 31 August - 3 September 2010.
Nye, J. (2010b) The New Public Diplomacy.
http://www.project-syndicate.org/commentary/nye79/English, accessed 2 November 2010.
Potter, S. (2008) Who Listened When London Called? Reactions to the BBC Empire Service in Canada, Australia and New Zealand, 1932–1939. Historical Journal of Film, Radio and Television, 28 (4), pp.475-87.
Roberston, E. (2009) ’“It is a real joy to get listening of any kind from the Homeland”: Empire Audience Responses to BBC External Broadcasts, 1932-1967’. Unpublished Paper presented at seminar at Leeds Metropolitan University. 7th October.
Robertson, E. (2008) ‘I get a real kick out of Big Ben’: BBC versions of Britishness on the Empire and General Overseas Service, 1932–1948. Historical Journal of Film, Radio and Television, 28 (4), pp.459-73.
Savage, M. (2010) Identities and Social Change in Britain since 1940: the politics of method, Oxford: Oxford University Press
Savage, M., Ruppert, E. and Law, J. (2010) Digital Devices: nine theses CRESC Working Paper Series No. 86 http://www.cresc.ac.uk/sites/default/files/Paper%20No%2086_0.pdf
Savage, M., and Burrows, R. (2007) ‘The coming crisis of empirical sociology’, Sociology 45
Silvey, R. (1974) Who’s Listening? The Story of BBC Audience Research. London: George Allen and Unwin.
Skuse, A., Gillespie, M., and Power, G. eds., (2011) Drama for Development: Cultural Translation and Social Change. New Delhi: Sage India.
Smith, M. P. (2001) Transnational Urbanism: Locating Globalisation, Oxford: Blackwell
Sreberny, A. Gillespie, M. and Baumann, G., eds. (2010) The BBC World Service and the Greater Middle East: Comparisons, Contrasts, Conflicts. Special Issue, Middle East Journal of Culture and Communication, 3(2).
Sreberny, A. Torfeh, M. (2008) The BBC Persian Service, 1941-1979. Historical Journal of Film, Radio and Television, 28 (4), pp.515-35.
Thelwall, M. (2009) Introduction to Sebometrics. Quantitative Web research for the Social sciences. Morgan & Claypool.
Toffler, A., (1990) Powershift. New York, Bantam.
 Cited in Nicholas, S. (2008) The Good Servant: the Origins and Development of BBC Listener Research 1936-1950, http://www.britishonlinearchives.co.uk/guides/9781851171248.php. Last updated: 27 February 2008.
 Equivalent to £10.50. Briggs, A. (1995) The History of Broadcasting in the United Kingdom, Volume II: The Golden Age of Wireless, Oxford: OUP, p.355.
 As a result of the 2010 Comprehensive Spending Review the operating expenditure budget for 2010/11 of £236.7 million has been reduced by £19 million for 2011/12.
 The government’s Comprehensive Spending Review announcement also means that from 2014 the Parliamentary Grant-in-Aid funding administered by the Foreign Office will come to and end and thereafter the BBC World Service will be funded from the BBC Licence Fee.
 For an analysis of how audience research functions as a tool for cultural translation and social change see, in particular, Skuse, A., Gillespie, M., and Power, G. eds., (2011) Drama for Development: Cultural Translation and Social Change. New Delhi: Sage India.
 See http://www.bbc.co.uk/worldservice/us/ba_2005.pdf, and http://downloads.bbc.co.uk/worldservice/pdf/bbc_world_service_financial_review_0910.pdf. Accessed 1 June 2010.
 See for example the prominent role played by WS audience research at Conference of International Broadcasting Audience Reseachers at http://www.cibar.org/
 The project was entitled ‘Diasporic Contact Zones at the BBC World Service’ and was funded by the UK’s Arts and Humanities Research Council Award Number AH/ES58693/1 For details of the projects in the wider study see: http://www8.open.ac.uk/researchprojects/diasporas/files/diasporas/bbcws_180407_paper.pdf accessed 01/06/2011.
 Latour (1990) developed the notion of ‘centre of calculation’ in order to make sense of the scientific process. Science, like war, he argues is a process of assembling more and better allies than enemies or competitors. Inscriptions (maps, computer printouts, graphs and charts and data of various kinds) are like allies in that they allow evidence to be mobilised with ease to influence and persuade sceptics. Simplifying inscriptions and making data do things takes place in what Latour refers to as “centers of calculation” which include institutions such as museums, laboratories and, as we use the term here, the BBCWS audience research section. (See Latour, B. (1990), “Drawing things together”, in Lynch, M. and Woolgar, S. (Eds), Representation in Scientific Practice, MIT Press, Cambridge, MA, pp. 19-68.
 Audience research, and an examination of methods, is an integral part of most of the theme and cross theme research in the wider project and is usually combined with other data sources. But of particular interest and relevance to this special issue is the work on digital diasporas see http://www8.open.ac.uk/researchprojects/diasporas/cross-research/digital-diasporas. Accessed 1 June 2011.
 Warm gratitude to Alan Booth, Colin Wilding, Jeremy Nye and Marie Dziedzic at the BBCWS for their openness and willingness to engage with us and help us. Graham Mytton also gave freely of his time and energies both in his capacity as a member of the advisory board and as a contributor to project events and publications – as indeed did Andrew Taussig)
 We were invited by the BBCWS to conduct research on media consumption in the Pakistani diaspora in the UK. Details of this project are written up in a report entitled ‘Pakistan Connection’. See http://www8.open.ac.uk/researchprojects/diasporas/projects/pakistan-connection. Accessed 1 June 2011.
 See http://www8.open.ac.uk/researchprojects/diasporas/news/public-policy-fellowship-at-the-bbc-world-service. Working Paper Series, accessed 1 June 2010.
 We are grateful to our colleagues at CRESC for all the insights that they have offered us on‘the social life of methods’ and to the CRESC website and work paper series (see Note xii above) which have informed this section of the article. However, the use and interpretation of this approach is distinct and adapted to the purposes of our WS research.
 See, for example, a US perspective on these closures at http://www.abcmoney.co.uk/news/2720051216.htm. Accessed 1 June 2011.
 http://www.bbc.co.uk/news/entertainment-arts-13873853. Accessed 28 June 2011.
 Details of Witness Seminars and Case Study method can be found at: http://www8.open.ac.uk/researchprojects/diasporas/cross-research/witness-seminars
 Briggs, Golden Age of Wireless, p.346.
 For reference to how Empire Broadcasts were targeted at the ‘lonely listener in the bush’ see http://www.michaelnelsonbooks.com/war_synopsis.htm.
 Cited in Roberston, E., 2009.’“It is a real joy to get listening of any kind from the Homeland”: Empire Audience Responses to BBC External Broadcasts, 1932-1967’. Unpublished Paper presented at seminar at Leeds Metropolitan University. 7th October. p.2.
 See the British Library Listener Research Letter and Reports at http://www.britishonlinearchives.co.uk/collection.php?cid=9781851171248&keywords=
 Briggs, Golden Age of Wireless, p. 347.