Data gathered in the French-speaking and Italian-speaking parts of Switzerland, using the protocol of the Children’s Understandings of Well-being multinational qualitative study (Hunner-Kreisel et al., 2016) raises the issue of language differences and taken-for-granted categories. The chapter focuses on the critical question of how far children’s discourses on their well-being are really “subjective”, when considering that any item exchanged in a conversation is already a situational account framed by power relations. The words chosen by social actors are selected according to their perception of the situation (social position, gender, ethnicity), with auto-censorship stronger as children face an adult researcher they don’t know. Therefore, power imbalance in interviews is as much an ethical issue as its mitigation is a methodological challenge. This issue has been addressed by adding a layer in the CUWB study protocol, namely the introduction of a tool called “kaleidoscope of experience” (Stoecklin 2013), functioning as a mediator in the interview situation. The presentation of self (Mead 1934) is negociated through this tool which has the ethical advantage of a more horizontal communication thanks to the use of five categories – activities, relations, values, images of self and motivations – that are universal enough to embrace experience without being trapped in gaps of communication due to local and generational cultures (Stoecklin 2019a). Latest developments are critical of these substantives, replacing them with verbs and considering them as “transactional horizons” (Stoecklin 2019b) whereby the experience that simultaneously conveys sensory perceptions is translated into hierarchised categories forming an intelligible discourse. The centrality of verbs in the ordening of reality (Foucault 1989) is underlined, with theoretical developments on forms of agency rather than levels. Drawing on interview notes and a vignette on the case of a girl of 11 with a relatively complex migration and family history, it will be shown that children’s subjective feeling of well-being is linked to social and personal factors at play in the co-construction of a balanced self. This enriches the capability approach (Sen, 1999) by highlighting that real freedoms depend on modes of acting that are culturally, socially and economically favored.