Conference Agenda

Network analysts have long used personal network visualizations to conduct qualitative interviews with respondents about the impact of their social context on outcome variables. This network ethnographic approach is often combined with quantitative analyses across respondents to generate a more complete and nuanced understanding of ways social context may be used as an intervention. Since the expansion of AI in 2022 there are now many tools for easily translating interview recordings to text for systematically analyzing the transcripts. We use a manual text analysis approach combined with generative AI to analyze 60 personal network interviews of former smokers regarding their efforts to stay quit, and to assist those in their networks to also quit smoking. We compare the concepts generated through manual text analysis versus using AI tools, and demonstrate findings from the interviews that would be difficult or impossible to discover using only network models of quantitative data.

The paper and presentation will discuss a new qualitative method to measure and evaluate the impacts of “impact networks” (D. Ehrlichman). How can we create useful and reliable evidence of the actual positive, negative, intended and unintended changes caused by networks of individuals or organizations?

Impact evaluation research and practice offer quantitative experimental and quasi-experimental methods to measure attributable treatment effects, on the one hand, as well as many approaches that focus on theory-building, participation of respondents and the use of qualitative data, on the other hand. The Qualitative Impact Assessment Protocol (QuIP) developed by the University of Bath attempts to bring together the strengths of the latter type of evaluation methods and offers a much-needed level of standardization. It has been utilized in numerous evaluations in diverse fields such as development, education or social services, sometimes to complement experimental designs, and has much potential for measuring network impacts in the future as well.

The presentation will discuss the author’s ongoing work with QuIP and the Causal Map App to analyze and visualize the causal statements gathered in interviews with network members based on two completed evaluations of networks of organizations. It will focus on methodological aspects such as the role of blindfolding, inter-coder reliability, transparency and representativity. Current attempts to utilize AI for coding and conducting interviews will also be discussed. In all, the presentation will aim to offer valuable insights and inspire discussion of the strengths and limitations of qualitative methods to evaluate and research networks.

Participatory water governance involves collaboration among government agencies, non-governmental organizations, water users, and academic institutions to address water-related issues. This collaboration may manifest in various forms, such as conducting research, managing water resources, and formulating policy. Effective stakeholder coordination and cooperation are typically operationalized through participatory structures, including water research/practice networks. The Arizona Water for All (AW4A) initiative aims to establish a multistakeholder network in the format of a Community of Practice—a group of individuals who share common interests, face similar challenges or are passionate about water security—to support water-insecure communities and facilitate participatory water research and decision-making throughout Arizona. This study introduces a mixed-methods framework for developing participatory initiatives and evaluating the enhancement of linkages (such as network ties) among various stakeholders across different regions—south, central, and north Arizona—within the AW4A Network initiative. We integrated Social Network Analysis (SNA) with ethnographic research methodologies to develop this comprehensive framework. Ethnography, which involves in-depth immersion in the context being studied, provided important insights into the complexity of social interactions that SNA techniques captured. By employing this combined approach, we were able to provide a more nuanced interpretation of the network data by validating the patterns and results obtained through SNA against the insider knowledge gathered from ethnographic interviews. We contributed to the research and methodological design of mixed-methods studies to explore the formation and dynamics of collaborative water networks.

We present an application of a mixed methods social network study in the field of collaborative environmental management. Effective environmental management generally requires coordination and collaboration between diverse heterogenous actors. However, communication and working together towards a common objective are often difficult in these settings. Actors involved need to overcome tacit boundaries and reconcile different viewpoints to reach a shared understanding on the environmental problem and envisioned solution(s). Thus, based on the notion that meaning and structure in social networks are co-produced, we assume that a common narrative can serve as a base for a shared understanding and ask the question what types of relationships influence the emergence of a common narrative. This presentation will showcase a mixed methods network study and underlying conceptual ideas and empirical implementation. We apply the idea of narrative congruence, which relates to the similarity of narrations that actors tell, to investigate the effects of the types of relationships between two actors as well as specific leadership roles using an Exponential Random Graph Model. Furthermore, we highlight how we combined qualitative and quantitative data meaningfully and what findings we got from this approach. Lastly, we would like to reflect on our taken mixed methods approach highlighting opportunities and challenges from which others could learn for the future.

Key words: collaboration, narratives, leadership, ERGM

Scientific literature shows that in Africa, women mostly exchange seeds between themselves. However, few studies have examined these inter-female exchanges and how relationships between women influence them. Yet, these practices play a central role for women, engaging social, economic and food dynamics. To shed light on the dynamics of cowpea circulation and exchange in Senegal, we propose to combine the ethnographic method with social network analysis (SNA).

Our fieldwork was based on ethnography over two cultural seasons, combining semi-structured interviews and participant observation. This approach helps to highlight peripheral exchange dynamics among women during the harvest, practices that are difficult to reveal by conventional quantitative survey methods. At the same time, SNA provides a quantitative perspective by mapping relationships and interactions, making visible the dynamics of reciprocity and centrality in these exchanges. Combined with the ethnographic approach, it is also a level of abstraction that allows for a deeper understanding of how kinship, alliance, residence, but also domestic cycles participate to organizing these exchanges.

The objective of this study is twofold. First, it seeks to highlight the complexity of social dynamics, power relations, and inequalities involved in the circulation of cowpeas, which are often obscured by gender disparities through a male-female opposition. Second, it offers a deeper reflection on the methodological hybridization between ethnography and network analysis as a tool for analyzing female moral economies in an agricultural context.