Owing to their adverse unemployment characteristics, unemployed social assistance recipients struggle to secure employment, thus prompting the use of activation measures intended to boost their employability. In Croatia, unemployed social assistance recipients are obliged to accept public work jobs and join community service activities on a voluntary basis. Thus, to receive social assistance, jobseekers are obliged to carry out community service which serves to provide them with work experience and skills in exchange for making contributions to the community. The Social Welfare Act regulates only the lower and upper limit of the monthly community service workload, while local self-government has a discretion to design activities that meet the needs of social assistance recipients and provide them with relevant work experience. Presently, social workers at social welfare centres and employees at local self-government (who are predominately not from helping professions) do not have guidelines or training that might help them in tailoring community service to the needs of social assistance recipients. Considering professionals who do not have proper preparation and education with the impact of community service on jobseekers’ well-being and employment prospective still unexplored, it remains to be determined whether community service provides a confining or stimulating environment for social assistance recipients. Since community service reflects some aspects of paid work environments this doctoral research takes Jahoda’s Latent Deprivation Theory (Jahoda, Lazarsfeld i Zeisel, 2002) as its theoretical framework. This doctoral research studies the impact of community service on mental health of jobseekers and their latent functions as well as it investigates whether community service can be regarded as a form of community support. In order to gain insight into the experiences of social assistance recipients, from this general objective, three specific research objectives are derived: (1) to determine the socio-demographic characteristics related to employment prospective of social assistance recipients. (2) to examine the role of hidden economy, community support and latent functions in predicting symptoms of internalizing problems (depression, anxiety, and stress) and mental health of unemployed social assistance recipients. (3) to examine the change in mental health, latent functions, and the perception of community support based on jobseekers’ enrolment in community service. In order to answer the above stated objectives, a quasi-experimental study was conducted using untreated control group design with dependent pretest and posttest. Treatment and comparison group were equally sized, where pretest was carried out before the intervention (community service) and posttest during the intervention. Pretest included 212 social assistance recipients, while the posttest included 209 social assistance recipients. The methods used to address the first research question included descriptive statistics and difference testing, the hierarchical regression analysis was used to address the second research question, while the third research question was addressed using 2x2 mixed ANOVA. Responses that were collected during the pretest were used to address the first research question. An analysis revealed that there is no difference between social assistance recipients and the unemployed population regarding gender structure and previous work experience. However, social assistance recipients are unemployed for significantly longer periods, they are on average middle-aged people and with poor education. Based on these data, it is concluded that social assistance recipients are very hard to employ. Accumulated adverse characteristics illustrate how barriers to employment of social assistance recipients are multidimensional and how they may find it exceedingly hard to escape the clutches of unemployment. Research shows that social assistance recipients require more extensive support to increase their employability. In order to prevent the accumulation of adverse characteristics the support should be provided immediately after becoming eligible for social assistance. This doctoral research shows that a quarter of social assistance recipients have severe symptoms of anxiety and that on the average social assistance recipients have common symptoms of stress and depression as well as poor mental health. Overall, social assistance recipients are deprived of all latent functions with collective purpose and social contacts being at the least fulfilled functions. However, even though they rarely engage in social contacts outside of their own family, social assistance recipients perceive themselves as being well integrated into their community and that the support from community organizations is moderately available. These data suggest that support initiatives to social assistance recipients need to include regular social contacts and activities that will make them feel that they are contributing to their community. Regression analysis shows that socio-demographic variables (sex, age, duration of unemployment), working in the hidden economy, community support and latent functions account for significant amount of mental health (58.4%) and symptoms of depression (39.9%). The same set of predictors accounts for 25.1% variance of stress and 19.1% variance of anxiety. Predictors of internalizing problems are availability of community support and two latent benefits of employment (regular activity and social status). Accordingly, those social assistance recipients who perceive better support from community organizations, who are engaged in more activities and regard themselves to have better social status have less symptoms of depression, anxiety, and stress. Generally, internalizing problems have similar predictors probably due to their similar emotional background. Additional predictor that accounts only for variance of anxiety is collective purpose in a way that unemployed social assistance recipients who perceive themselves as contributing members of society have less symptoms of anxiety. Significant predictors of mental health are community integration, available support of community organizations, regular activity, and collective purpose. The obtained results demonstrate that social assistance recipients who are engaged in regular activities, contribute to their community, feel integrated in community, and perceive more support from community organizations will have better mental health. Several implications can be drawn from these results. Support for unemployed social assistance recipients should reinforce their trust in institutions, promote mutual solidarity and respect, increase the visibility and availability of existing services and programmes, encourage them to use community services, increase their activity and collective purpose, thereby ensuring that they perceive themselves as being contributing members of their communities. This paper does not examine the reasonableness and fairness of community service, but rather investigates its social integrative function and impact on mental health for people living in poverty. Doctoral research demonstrates that community service enables the fulfilment of almost all latent benefits of employment. During their participation in community service, unemployed social assistance recipients had more time structure, they have engaged more in social contact outside of their own family, perceived better social status and stronger sense of collective purpose as community service increased their sense of being contributing members of their communities. Yet, participation in community service did not increase activity of unemployed social assistance recipients. Social assistance recipients who participated in community service also perceived they are more integrated in community, but did not perceive more support from community organizations is available. It is concluded that while community service helped unemployed feel more incorporated into local environment, this specific community service design did not constitute effective support resource. Also, community service did not affect the level of mental health, nor the symptoms of depression, anxiety and stress. The obtained results suggest that community service in its present form do not constitute a measure that can mitigate the negative effects of unemployment and living in poverty. This doctoral research gives contribution to Croatian field of social work being the first study to explore the effects of a national activation measure (community service) on the mental health of social assistance recipients. The present doctoral research also constitutes a contribution to mental health promotion, prevention science and social work since the obtained data highlight the importance of expanding the scope of support for people living in poverty, by providing them and their family mental health support and by presenting evidence that can be used to advocate for an expansion of the scope of support. Along with mentioned above, this doctoral thesis advocates right to health for poor families. Doctoral research illustrates how the effectiveness of state services can be measured by psychosocial indicators and suggests further direction for a science-based organization of community support with the aim of strengthening employability and mental health of social assistance recipients. Thus, the present findings can enhance better cooperation between employment, social welfare, and health policies at local and national level.