Informed, uninformed and participative consent in social media research

The use of online data is becoming increasingly essential for the generation of insight in today’s research environment. This reflects the much wider range of data available online and the key role that social media now plays in interpersonal communication. However, the process of gaining permission to use social media data for research purposes creates a number of significant issues when considering compatibility with professional ethics guidelines. This paper critically explores the application of existing informed consent policies to social media research and compares with the form of consent gained by the social networks themselves, which we label ‘uninformed consent’. We argue that, as currently constructed, informed consent carries assumptions about the nature of privacy that are not consistent with the way that consumers behave in an online environment. On the other hand, uninformed consent relies on asymmetric relationships that are unlikely to succeed in an environment based on co-creation of value. The paper highlights the ethical ambiguity created by current approaches for gaining customer consent, and proposes a new conceptual framework based on participative consent that allows for greater alignment between consumer privacy and ethical concerns.

Institutional framing of CSR and social entrepreneurship interactions in South Africa

Objectives. This paper considers the intersection of Corporate Social Responsibility (CSR) and social entrepreneurship in South Africa through the lens of institutional theories and draws upon a number of illustrative case study examples. In particular it: (1) charts the historically evolving relationship between CSR and social entrepreneurship in South Africa, and how this relationship has been informed by institutional changes since the end of apartheid, particularly over the last few years; (2) identifies different interactional relationship forms between social enterprises and corporates engaging in CSR, with an emphasis on new innovative multi-stakeholder partnerships; and (3) considers internal engagements with social responsibility by SME social enterprises in South Africa. Prior Work. Reflecting South Africa’s history of division, the controversial role of business during apartheid, and the ongoing legacies of that period, the South African government has been particularly pro-active in encouraging companies to contribute to development and societal transformation through CSR and Black Economic Empowerment (BEE). Accordingly a substantial body of work now exists examining and critically reflecting upon CSR and BEE across a range of sectors. In response to perceived problems with BEE, efforts have recently been made to foster broader-based economic empowerment. However the implications of these transitions for the relationship between CSR and social entrepreneurship in South Africa have received scant academic attention. Approach. Analysis is undertaken of legislative and policy changes in South Africa with a bearing on CSR and social entrepreneurship. Data collected during fieldwork in South Africa working with 6 social enterprise case studies is utilised including qualitative data from key informant interviews, focus groups with stakeholders and observational research. Results. The paper considers the historically evolving relationship between CSR and social entrepreneurship in South Africa informed by institutional change. Five different relationship forms are identified and illustrated with reference to case examples. Finally internal engagement with social responsibility concerns by small and medium social enterprises are critically discussed. Implications. This paper sheds light on some of the innovative partnerships emerging between corporates and social enterprises in South Africa. It reflects on some of the strengths and weaknesses of South Africa’s policy and legislative approaches. Value. The paper provides insights useful for academic and practitioner audiences. It also has policy relevance, in particularly for other African countries potentially looking to follow South Africa’s example, in the development of legislative and policy frameworks to promote corporate responsibility, empowerment and transformation.

Weighted alpha-rate dominating sets in social networks

We are looking into variants of a domination set problem in social networks. While randomised algorithms for solving the minimum weighted domination set problem and the minimum alpha and alpha-rate domination problem on simple graphs are already present in the literature, we propose here a randomised algorithm for the minimum weighted alpha-rate domination set problem which is, to the best of our knowledge, the first such algorithm. A theoretical approximation bound based on a simple randomised rounding technique is given. The algorithm is implemented in Python and applied to a UK Twitter mentions networks using a measure of individuals’ influence (klout) as weights. We argue that the weights of vertices could be interpreted as the costs of getting those individuals on board for a campaign or a behaviour change intervention. The minimum weighted alpha-rate dominating set problem can therefore be seen as finding a set that minimises the total cost and each individual in a network has at least alpha percentage of its neighbours in the chosen set. We also test our algorithm on generated graphs with several thousand vertices and edges. Our results on this real-life Twitter networks and generated graphs show that the implementation is reasonably efficient and thus can be used for real-life applications when creating social network based interventions, designing social media campaigns and potentially improving users’ social media experience.

Memory and burstiness in dynamic networks

A discrete-time random process is described, which can generate bursty sequences of events. A Bernoulli process, where the probability of an event occurring at time t is given by a fixed probability x, is modified to include a memory effect where the event probability is increased proportionally to the number of events that occurred within a given amount of time preceding t. For small values of x the interevent time distribution follows a power law with exponent −2−x. We consider a dynamic network where each node forms, and breaks connections according to this process. The value of x for each node depends on the fitness distribution, \rho(x), from which it is drawn; we find exact solutions for the expectation of the degree distribution for a variety of possible fitness distributions, and for both cases where the memory effect either is, or is not present. This work can potentially lead to methods to uncover hidden fitness distributions from fast changing, temporal network data, such as online social communications and fMRI scans.