Logo: to the web site of Uppsala University

Digitalization has created opportunities for people to have contact with representatives of human service organizations through various text-based channels (e.g., online chat, social media, email). Despite the fact that such text-based interaction is common in social work practice today, there is remarkably little knowledge about how the interaction is structured in practice and what characteristics different forms have. Such knowledge is central for designing appropriate contacts that respond to clients’ needs. This study uses conversation analysis to study two forms of digital text-based interaction between clients and representatives of the Swedish Social Insurance Agency (SSIA). The data consist of circa 900 exchanges by email and on the SSIA’s official Facebook page for parents. Identified patterns are also compared with 300 recorded phone calls to the SSIA to make visible how participants orient to limitations and affordances that the textual arenas provide in practice and which functions they thereby fulfill with regard to client relations. The analysis focuses on how practitioners manage responsiveness in replies to client questions. We show how practitioners’ replies tend to deviate from conversational norms by not being designed in line with the design or content of the questions. For example, practitioners often provide information that is conditioned (e.g., “if…then…”), rather than yes/no answers (even when replying to yes/no questions). Through this, the analysis also demonstrates how practitioners realize policy and organizational goals in digital text-based interaction. The study contributes knowledge about how help and service is provided in digital text-based interaction and highlights the problem of generally held replies that are not adapted to the recipient, as the interaction risks not fulfilling the expected function for the clients. We also problematize how textual interaction on social media platforms poses risks for clients’ privacy from an ethical perspective. 

This article investigates how mental health counsellors on helplines in Sweden deal with racism from callers who self-categorise as non-racialised. Previous studies have identified racism as a problem in health care interactions, but there is limited knowledge about the features of racialised talk and how staff respond. In this study, we use conversation analysis and membership categorisation analysis to examine racialised talk in 17 audiorecorded calls, a subset of 458 calls to suicide preventive helplines. The analysis shows that racialisation functions as a resource for callers to make sense of their mental health difficulties. This speaks to the complexity of responding to racism in a mental health setting, as counsellors must see to callers’ needs, and calling out racialised talk may alienate callers. Call-takers manage this problem in three ways: (1) questioning racialised talk, (2) supporting the callers’ stance in a way that makes it ambiguous if call-takers are coproducing racism or affiliating with callers’ lives being difficult and (3) supporting callers’ problems as mental health issues while resisting a potentially racist trajectory. The study offers direct insight into the workings of racism in health care and how practitioners can balance health care users' needs for support with an antiracist position.

Swedish welfare organisations have a long history of promoting gender equality and inclusivity, involving implementation of more gender-inclusive language (e.g., non-gendered terminology). This study analysed the use of non/gendered expressions at the Swedish Social Insurance Agency, as state policy is realised in conjunction with personalised service in textual interaction in screen-level bureaucracy. We drew on conversation analysis and membership categorisation analysis to analyse 378 email interactions between clients and social insurance officers, focusing on gendered institutional terminology and gendered person references. The findings suggest that social insurance officers avoided a heteronormative language, in line with Swedish official public language policy, but that a gender-neutral language may come at the cost of formality, standardisation, and de-personalisation. The study brings new knowledge on the ways in which street-level bureaucrats manage responsiveness and standardisation in digital text-based interaction with clients.