This paper presents Saharaline, an intervention designed to provide collective social support for teachers in low-income schools. Implemented as a WhatsApp-based helpline, Saharaline enables teachers to reach out for personalized, long-term assistance with a wide range of problems and stressors, including pedagogical, technological, and emotional challenges. Depending on the support needed, teachers' requests are routed to appropriate domain experts--- staff employed by educational non-profit organizations who understand teachers' on-the-ground realities---who offer localized and contextualized assistance. Via a three-month exploratory deployment with 28 teachers in India, we show how Saharaline's design enabled a collective of diverse education experts to craft and deliver localized solutions that teachers could incorporate into their practice. We conclude by reflecting on the efficacy of our intervention in low-resource work contexts and provide recommendations to enhance collective social support interventions similar to Saharaline.
doi.org/10.1145/3613904.3642617
Online peer support provides space for individuals to connect with others and seek support. However, while empathy is critical for effective support, studies have found that highly empathetic support on these platforms can be rare. Using data from online peer support platforms, we conducted a mixed-methods analysis to study the factors that lead to support seekers’ perceived empathy. We found that CBT techniques like active listening and reflective restatements, along with fostering a space for exploration, increase perceived empathy, whereas rigid adherence to structure, misalignment of concerns, and lack of emotional validation can contribute to low perceived empathy. In addition, despite the high levels of empathy reported by most support seekers (85%), computational models reported low averaged empathy (1.69/6). Lastly, we propose that empathy is not a quantifiable metric and that future algorithmic empathy measurements require human perspectives.
doi.org/10.1145/3613904.3642034
US military veterans (USMVs) are a vulnerable population with an elevated risk of mental health issues and suicide. Peer support, especially through mobile technology, has proven effective in addressing mental health related challenges, but ensuring long-term engagement remains a concern. This study explores the opportunity of designing persuasive technology, particularly persuasive reminders, to enhance engagement in peer support interventions for veterans. We followed community-based participatory research with ten veterans to identify specific peer support processes that can benefit from persuasive reminders and to uncover the underlying community values and needs to guide design. The findings emphasize the importance of designing reminders that focus on personalized strategies, effective delivery of success stories, understanding motivation levels, careful language selection, actionable reminders, and mutual accountability. The study advocates context-specific design and highlights the need for a broader user-centered persuasion design perspective to cater to veterans' unique needs.
doi.org/10.1145/3613904.3642962
Politicians and care associations advocate for the use of machine learning (ML) systems to improve the delivery of adult social services. Yet, guidance on how to implement ML systems remains limited and research indicates that future implementation efforts are likely to encounter difficulties. We aim to enhance the understanding of ML system implementations by conducting a longitudinal field study with a team responsible for deploying a ML system within an adult social services department. The ML system implementation represented a cross-organisational effort to facilitate the department’s transition to a proactive practice. Throughout this process, stakeholders adapted to numerous challenges in real-time. This study makes three contributions. First, we provide a description of how ML systems are implemented and highlight practical challenges. Second, we illustrate the utility of HCI knowledge in designing workflows for ML-assisted preventative care programmes. Finally, we provide recommendations for future deployments of ML systems in social care.
doi.org/10.1145/3613904.3642247
The present work explores the potential of leveraging a teamwork agent's identity -- signaled through its gendered voice -- to support marginalized individuals in gender-imbalanced teams. In a mixed design experiment (N = 178), participants were randomly assigned to work with a female and a male voice agent in either a female-dominated or male-dominated team. Results show the presence of a same-gender voice agent is particularly beneficial to the performance of marginalized female members, such that they would contribute more ideas and talk more when a female agent was present. Conversely, marginalized male members became more talkative but were less focused on the teamwork tasks at hand when working with a male-sounding agent. The findings of the present experiment support existing literature on the effect of social presence in gender-imbalanced teams, such that gendered agents serve similar benefits as human teammates of the same gender identities. However, the effect of agents' presence remains limited when participants have experienced severe marginalization in the past. Based on findings from the present study, we discuss relevant design implications and avenues for future research.
doi.org/10.1145/3613904.3642202