Hester Paanakker

things are done around here” (Liebling, 2000) adds another interesting dimension. Clearly, prison management is a significant actor in the value chain, in between policy developers and street level bureaucrats. The friction between top down policy imposition and required room for discretion (both from management and street-level prison officers) introduces an interesting scenario for the examination of convergent or divergent value approaches to street-level craftsmanship and their effects. Finally, at the time of data collection in 2014 and 2015, the Dutch prison sector was going through a process of wide-ranging organizational reform (with significant cutbacks in personnel and the number of correctional facilities in the country) and the implementation of a new policy paradigm called the Modernization Program (aimed at a culture change in Dutch correctional institutions, emphasizing prisoners’ supervision and coaching, embedded in a lifecycle approach to incarceration). This provides a setting of change in which ideas on street-level values and their enactment are challenged and renegotiated (Wright, Christensen, & Isett, 2013; Wynen et al., 2019), bringing the mechanisms of value understandings to the surface (Stewart, 2006). It forces penal officials, from policy development level through organizational management level, to middle management (head of departments) and street level bureaucrats to determine their position on what good work at implementation level looks like. As the role of values in shaping views on the frontline craft becomes particularly visible, it makes this a case study of specific relevance. 1.4 Methods The data for this thesis is collected using a case study design in the prison sector in the Netherlands. The qualitative data collection consists of a two-month period of participatory field observation (spread over 75 hours), in two correctional facilities and across eleven different departments, document analysis, and a total of 55 in-depth semi-structured interviews. Data was collected among four different groups of respondents: prison officers (street level workers, N= 32), prison middle management ( N= 9), prison management (the management team of the facility at large, N= 8 ), and policy officers in the ministerial department that constitutes the prison sector’s headquarters at the Ministry of Justice and Security( N= 6). Using an inductive approach to data interpretation and code creation, data analysis consisted of a systematic content analysis through software-supported (MAXQDA) coding. 23 Introduction