Date of Award

Winter 12-15-2014

Author's School

Graduate School of Arts and Sciences

Author's Department

Social Work

Additional Affiliations

Brown School of Social Work

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Considering that one out of five youths live in poverty in Hong Kong, and poverty is found to have a negative impact on youths' psychological well-being, this dissertation proposes mentoring as a social capital intervention to improve the psychological health of these young people, both who are locally born and who immigrated from Mainland China. In this dissertation, mentoring is defined as a social capital intervention that changes the social resources available to youths living in poverty, and such changes enhance young people's positive psychological outcomes. Based on the social capital theory, poverty-related and psychological theories, three aims were formulated. Aim 1 and 2 focus on examining the definition of childhood poverty and its consequences. Aim 3 explores whether or not mentoring is a social capital intervention. Aim 1. Operationalize childhood poverty in Hong Kong. Aim 2. Examine the psychological consequences of childhood poverty. Aim 3. Explore how mentoring promotes the psychological health of youths in poverty.

Data and Methods. I used both quantitative and qualitative data in my dissertation. The quantitative longitudinal secondary data set consisted of 750 mentees, aged between 10 and 16, participating in the government-funded Child Development Fund Program (CDF). The qualitative data were collected through interviews from mentors (n= 15) and mentees (n=22; aged between 14 and 18) participating in a university initiated School-based Problem Solving Skills Mentoring Program. I only used quantitative data to achieve Aim 1 and 2. Latent class analysis was adopted in Aim 1, and longitudinal multilevel level modeling was employed in Aim 2. For Aim 3, a mixed methods parallel convergent research design, which consisted of the quantitative and the qualitative stand, was adopted.

Results. A latent binary variable of childhood poverty (1= Welfare receivers and 0= Non-welfare receivers) was created in Aim 1 with the following socio-economic indicators: Receiving family welfare, living in rented housing, having low parental educational level, having at least one unemployed parents and residing in single parent family. This latent variable was used in subsequent analysis of Aim 2 and 3. For aim 2, the psychological status of mentees was found to be decreasing at a slow rate over their three years of their enrollment in CDF. Furthermore, parental perceived economic pressure was negatively associated with the psychological status of youths, but not objective indicators of childhood poverty and having experienced financial difficulty. For aim 3, findings showed that positive mentoring relationship was directly associated with the positive psychological health of mentees. In addition, mentoring relationship exerted its positive influence on mentees' psychological health via improving their parent-child relationships. The social resources generated from mentoring were direct advice, companionship, role modeling and encouragements. Mentoring also increases the family social resources of higher level of sharing between the mentees and their parents.

Discussions. Results of this dissertation provide preliminary evidence that mentoring is a social capital intervention. First, mentoring relationship improves youths' psychological health and parent-child relationships. Second, positive mentoring relationship generates specific types of social resources, namely, direct advice, companionship, role modeling and encouragements, conducive to positive psychological well-being of mentees. Best practices of mentoring are identified. Future research and policy directions are also initiated.


English (en)

Chair and Committee

Ramesh Raghavan

Committee Members

Constantino John, Luke Douglas, Sherraden Michael, Trani Jean-Francois, Yip Paul


Permanent URL:

Included in

Social Work Commons