Desistance from crime is not just the cessation of criminal activity itself, but a process of acquiring roles, identities, and virtues; of developing new social ties, and of inhabiting new spaces. Such a framing of desistance is undoubtedly affected by structural, cultural and gender issues. This dissertation attempts to fill some of the gaps in our knowledge of desistance in a developing country, and to contribute to an intercultural dialogue with mainstream explanations. The aim of this thesis is to explore the transition from crime to conformity among a sample of Chilean juvenile offenders, and to identify theoretical, policy and practice implications from the findings. For these purposes, the research is focused towards examining three main areas: (1) the process of moving away from crime, its obstacles and challenges; (2) possible factors involved in the transition from crime to conformity; and (3) the particularities of gender in the process of moving away from crime, focusing on young female offenders. The data used in the present study come from the Trajectories Study, a longitudinal research project that uses mixed methods to explore the criminal and life trajectories within a sample panel of young male and female offenders in Chile. The present thesis contributes to the ways in which we understand, explain and research the process of moving away from crime. Perhaps the most crucial finding is that, by researching the process of moving away from crime among late adolescent offenders, the liminal space of ambivalences and inconsistencies that most individuals experience on their path to desistance from crime, acquires a pivotal relevance. This is supported by the first finding of the study, that temporary desistance and persistence categories are far from absolute, and are rather two ends of a continuum that individuals can move along fluidly. Ambivalent desisters and conformist offenders who persisted in order to align themselves with mainstream society emerged as new categories that challenged the traditional ways of understanding the process of moving away from crime. The factors associated with this process among the young participants of the study varied according to the theoretical model and the operationalisation of desistance used. While some factors emerged as significantly associated with desistance in models based on social control and routine activities approaches, for instance, they were not associated with desistance in the final integrated model that included psychosocial factors as well. The integrated model shows definitively that the desistance process is not simply associated with social control factors that appear in life and promote changes without individuals being aware of them. Psychosocial factors might be also relevant throughout the process and they do not operate in isolation from each other: rather, there is an orientation towards pro-sociality in a broad sense. Finally, the findings show that the process of crime abandonment is strongly gendered. In the case of the young men from the study, it was observed through the changing process that they learned new forms of doing masculinities by exerting control over themselves, restraining from violence, pursuing a productive life, assuming adult roles and leaving behind expressive crime (caused by humiliation and the pursuit of identity and status). Temporary desistance among women was related to a transition towards traditional feminine caring roles, and persistence operated as a way both to resist and affirm gender subordination and patriarchy. Although cultural particularities were found in the present study, several results were consistent and confirmed findings from previous research conducted in developed and anglophone countries. This might be due to the fact that, in societies with similar market-based economic systems and liberal welfare regimes, several structural factors operate in similar ways. Perhaps the main differences and the main contributions from a Southern Criminological perspective, are found in the intensity by which some structural constraints operate in participants’ lives. This was certainly the case of the great salience observed of psycho social factors over social control triggers and by the fact that ambivalence, attachment, consumerism and masculinity emerged as pivotal in the process of moving away from crime, affecting participants bilaterally, capable of both pulling them away from crime and pushing them back towards it.These findings lead to interesting research and policy implications to counteract current correctional approaches, characterized by an excessive orientation towards compliance, without considering the complexities of the dynamic changing process of moving away from crime and their particularities in the Latin American context.