Publisher: GenderIT.org 30 August 2018
In the preceding three columns of this monthly series we introduced the way the Local Access Network project brings a feminist perspective to advocating for community networks in the global South, and highlighted the low representation of women in the regulatory and policy space and in regional and global movements. One of the Local Access Network project’s four activities is 'case studies and analysis’ and in this, research-focussed component, I study the social and gender impact of community networks in rural places, while Mike Jensen studies the technologies and sustainability models that these networks deploy. Thus, in the fourth column of the series on the Local Access Network project, I highlight some issues encountered in the nitty-gritty of the research on social and gender impact. I begin by summarising my approach to generating insights and then explain two aspects of the challenges we encounter in representing women’s perspectives in the research.
Methods and perspectives
My research aims to provide concrete examples of opportunities, and challenges that must be addressed, to realise the positive potential of the community network paradigm. To do so, it describes the roles and meanings of community networks in people’s everyday lives, by exploring how and why people coordinate, interact with, and are affected by, their local community network. My research does not aim to obtain insights that are statistically generalisable to entire populations of organisers, users and non-users of community networks but, rather, produce insights that can help to:
Identify ways that community networks produce the most positive social impacts for the widest range of people in their specific contexts of management, access and use.
Understand the ways that community networks challenge or reproduce social, economic and cultural inequalities in society.
My research assumes that each community network is differently embedded in the everyday lives of the people who lead, manage and use it, and will have various roles and meanings to people who encounter it in their daily lives, including those who do not use its services directly. To obtain data about these roles and meanings I gain insights from practice-oriented interviews that focus on people’s actual activities related to their community network and what people say about their actions. I bring a postcolonial feminist position to generating and analysing this data, which notices power relations arise around not only gender but also race, and colonialism’s enduring political, economic and cultural trajectory.
My commitment to a postcolonial feminist position arises from my research, teaching and facilitation in the field of Human-Computer Interaction (HCI), which for the past 15 years has focused on relationships between ICTs and the communication practices of people who constitute the majority of the world’s population but are marginalised by dominant ICT approaches, specifically in Africa and amongst indigenous groups in Australia. My research applies HCI’s "3rd paradigm", an orientation that is sensitive to the ways that ICT users’, non-users’ and analysts’ actions and interactions are situated in specific meaning-making contexts. This includes research on setting-up and/or analysing community networks in rural South Africa and Namibia (e.g. Bidwell et al, 2013; Bidwell, 2016; Bidwell et al, 2019).
Between January and July 2018 I studied four community networks, or interrelated set of networks, within the Local Access Networks project, in rural areas of four countries in Asia, Latin America and Africa. A total of 188 people, between the ages of 16 and 78 years of age, participated in my interviews and focus groups, and I observed, and discussed with, participants about how they do, and do not interact, with the community network, and how they link what they do to their own lives and the goals of the community network. I broadly categorise participants into: initiators, leaders or champions of the community network programmes; managers, coordinators, and operators at specific community networks; ordinary users of community networks; and non-users. The category of non-users includes people living in the area of a community network who know of the network but have not used the network directly themselves because they cannot or chose not to, or because other people mediate their interactions. Generating data about non-users yields insight into a community network’s inclusiveness and wider socio-technical impacts in a community.
Women’s voices in the community networks studied
Our research is a partnership between APC and the community networks I studied, and we relied on champions of community network programmes, and managers or coordinators of community networks to allocate interpreters for interviews and focus groups, organise most interviews and recruit participants. The diversity of the community networks meant that participants were recruited in different ways. In smaller networks a champion or coordinator organised interviews and convened focus groups with users and non-users. In large community network programmes, champions chose specific networks for us to visit and local managers recruited participants or broadcast a call for participation, and in two of these I was able to walk around to chat with residents informally and arrange interviews with them directly. I did not meet a single champion or manager who was unenthusiastic about involving women, yet time and again I encountered difficulties in ensuring the data represented women’s perspectives.
In preparing the studies with initiators or managers of community networks and programmes, I discussed the importance of recruiting equal numbers of men and women participants. Thus, the total of 86 interviews with individual people comprised 59% men and 41%; and, the total of 19 focus groups comprised 42% men and 58%. However, ratios of men to women in different categories of participants reflects women’s power in the four community networks I studied. Men and women constitute, respectively, 4% and 2% of initiators or champions; 19% and 9% of operators or coordinators; 18% and 24% of users; and 7% and 18% of non-users. These proportions do not reflect probability sampling, but roughly indicate the relative power of men and women in relation to running community networks and recruiting participants.
In all countries I stressed the importance of separate interviews and discussions with men and women, and men and women interpreters, respectively, to facilitate these spaces, but this was rarely achieved. It is generally acknowledged in research that gender power relations constrain women’s ability to speak freely in mixed groups, which has obvious consequences for detailed insights about gender impacts. Despite my repeated requests, and suggestions, in two countries I was provided only a man interpreter. When I was able to undertake interviews and focus groups with only women because participants spoke English and circumvent involving men who had allocated themselves as interpreters these men sometimes expressed that they felt excluded. Of focus groups with users and/or non-users, 9 were composed of men participants only, 6 were composed of women participants only, and 3 included both men and women.
In one country the male interpreter scheduled focus groups to suit times that men and women returned from their agricultural work, but after focus groups explained that few women were present because the timing meant women cooking in their homes. Further, even when I was provided women interpreters, men were in close proximity to 45% of interviews or focus group sessions with users and non-users. Usually this was because men came to, and loitered around, the places where women had invited us to talk, this happened even in front of private homes where the men would sit themselves within hearing distance, and sometimes even try to contribute to interviews. Thus, my attempts to provide women-only spaces and privacy to the most marginalised women were frequently thwarted by men, either because the women did not speak English, and I was provided only a man interpreter, or because local residents disrupted their privacy.
Ensuring the credibility of data on different perspectives towards community networks is further complicated by the differing types and scales of the networks and the way my access to participants is mediated by champions and managers. For instance, ethical research demands respecting the privacy and anonymity of individual people in interviews, and their right for their views to be represented accurately and credibly. For many participants, however, language, literacy and internet access precludes "member checking", such as by sending interview or focus group transcripts directly to interview participants to validate. Since ethical research also demands the pursuit treating participants equally, the approach we have adopted is to send anonymised transcripts to the community networks’ champions and managers who facilitated my studies.
However, especially, for small community networks, it can be impossible to de-identify the data sufficiently to maintain anonymity of individual people in these transcripts. Thus, ensuring credibility of data and reporting becomes a choice between the collective rights of the community network, as interpreted by champions and managers, and the rights of individual people involved in or affected by the community network. The choice about whether credibility should be determined by collectives or individual people points to another fundamental power relation in the research: who gets to decide and why?
Centre-periphery relations in constructing the field of study
Decisions about who participates in the research and the ways they participate are not only located with the community networks champions or managers or processes in the rural places that Mike and I studied. The "field", when talking about our "field research", of course, includes not only the community networks "out there" in the periphery of the global south, but all the reasoning involved in the Local Access Networks project and in studying social and gender impact.
As Steve and Carlos mentioned in their blog posts, the "field" of community networks is constructed by mostly white men; additionally, while our research explicitly aims to support community networks in the global south, its framing originates in the West, or global north. That is, the reasoning embedded in studying the social and gender impact of community networks is as much "in here", close to the centre of certain powers, as it is "out there" in the four networks I have studied so far.
The "centre", in this sense, is where technologies and the rules and standards that govern them are produced; where investments in technologies and research projects originate; where development agendas are forged; and where research insights are interpreted, consumed and their outputs acted upon. Thus, centre-periphery power relations affect, for instance, who gets to choose, and why they get to choose: what the research questions are; which community networks are included in the studies; how visits are timed and coordinated; how studies are ordered and moved to; what methods are used to generate, record, and interpret data community network; what is written about the research, where it is read, and who it is read by. And so on. To illustrate how challenges in representing women’s voices manifest the effects of centre-periphery power relations and infrastructures that cut across time and place, I describe just a couple, of many possible, examples.
The outline for funding originally proposed case studies of ten community networks initiatives in Africa, Asia and Latin America, however budget restrictions abbreviated the research to ten months so we reduced the number of gender and social impact studies to six of the approximately 35 initiatives that might be described as rural community networks in the three continents. The selection of six case studies was mostly based on a balance between the continents and the community network’s visibility, length of establishment, existing relationships with APC, physical accessibility; and, how well they contributed a range of technologies into the group of ten networks Mike studied. This selection was not based on impacts that the community networks themselves identified, the familiarity of champions with people in community networks, or whether community networks were led by women.
Thus the set of networks varied widely in terms of familiarity between champions and local areas, articulated gender targets, and the involvement of women in different roles. Together with time constraints on research, the selection criteria affected creating women-only spaces. For instance, in total I spent an average of 15 days with each community network initiative but, on average, only 9 days in the specific rural contexts of the community networks themselves because some initiators and champions are based far from the areas the community networks serve. This not only amplified my reliance on champions, managers or coordinators in recruiting participants and generating data, but also that the people I interacted most with were often unfamiliar with the everyday lives of users and non-users of the community networks.
Priorities in the entire Local Access Networks project are integral to the methodology, and thus shape the data generated. The project has a very explicit agenda to improve opportunities for community networks in the global south. However, this advocacy and movement building is situated in the white men dominated discourse of telecommunications technology, policy and lobbies around regulation. For instance, the project team’s interactions with policy, regulation, movement building and technical research involve the four technology-oriented men, while the most direct interactions with women associated with community networks occur amongst two socio-technical or international development oriented women. Moreover, as the telecommunications discourse originates in the global north it has particular perspectives on what makes telecommunications viable and sustainable, such as certain ideas about scale, skill, quality.
From a postcolonial feminist perspective, what’s important here is that the methodology is not just about the ways that I generated and interpreted data, but also why it is generated and interpreted. Or, in other words the knowledge produced in studying social and gender impact and the claims made about women in community networks, are shaped by a discourse that is white, male and western. This foundation makes it is difficult to escape from imposing assumptions that gender, ethnicity and class relations in the global south exist in the same form as those in the north. Thus, in generating and analysing data we tend to look for, and prioritise, familiar relations and, in doing so, construct the identity of women in community networks in ways we construct them in the north. Even in adopting a feminist standpoint approach, the reasoning applied to generating and interpreting data is, at best, a dialogue between the standpoint of a women "out there" and our own "in here", via me, and the stage on which this dialogue is set is western, white and male.
Consequences for the Local Access Networks project
Recognising that power relations, that effect representing women’s perspectives on rural community networks, do not just exist "out there" in the periphery in the global south but as much involve us, "in here", the employees and supporters of the Local Access Networks project, provides a basis for future work.
Firstly, recognising that none of us are passive in the studies, "outside" the research process or somehow "objective", demands reflecting on the ways our various actions construct the identity of women associated with community networks, and what this construction means to undertaking, reporting and responding to studies. At this stage I understand my duty is to explicitly focus on language and text by devoting vast amounts of time transcribing, contextualising and analysing, interviews; and then after that reflecting on why I notice what I notice.
Secondly, acknowledging that centre-periphery relations are manifested through every part of the research and embodied in interactions with people "out there", whether these are direct, remote or mediated through others, demands developing new ways for women’s voices to be speak into and influence the research, the project’s goals and the discourse generally.
This column was originally published by GenderIT.org.
Bidwell, N. J., Shipepe, A., Makawa, T. K, Nhinda, G., Sheetekala, S., Rey-Moreno, C., Limbo, A. and Mutonga, M. (2019). Communication practices and togetherness in setting up a Community Network in rural Namibia. In: The Relational Interface. Ed. Satinder Gill. Springer. Forthcoming.
Bidwell, N.J. (2016). Moving the Centre to Design Social Media for Rural Africa. AI&Soc: Journal of Culture, Communication & Knowledge, 31(1) 51-77. Springer.
Bidwell, NJ., Siya, M., Marsden, G., Tucker, WD., Tshemese, M., Gaven, N., Ntlangano, S., Eglinton, KA Robinson, S. (2013). Walking and the Social Life of Solar Charging in Rural Africa. ACM Transactions on Computer- Human Interaction (TOCHI), 20, 4: 1-33.