Volume 15, Issue 4 - Special issue: The project P.A.T.H.S. in Hong Kong: A pioneer six-year longitudinal study / Editors: Daniel T.L. Shek, Cecilia M.S. Ma, Janet T.Y. Leung and Joav Merrick

In adolescent developmental research, cross-sectional and longitudinal research designs are commonly used. In this paper, examples of adolescent developmental research utilizing cross-sectional designs are presented and the basic attributes, strengths and weaknesses of cross-sectional designs are outlined. Besides, examples of longitudinal adolescent developmental research as well as its characteristics, advantages and pitfalls are examined. Based on this review, it is argued that longitudinal research design plays an important role in adolescent developmental research. Implications of the present review on the design of longitudinal adolescent research studies in the Chinese context are discussed.

Longitudinal studies are important for understanding adolescent development. In this study, eight examples of longitudinal adolescent research in different fields in Mainland China were examined. The review showed that longitudinal studies in Mainland China were grossly inadequate. There were also few studies on adolescent life satisfaction, sensitive developmental issues (such as sexuality and suicide) and program evaluation. Panel study was the dominant longitudinal design but most of them did not have many time points for data collection. Most studies were based on convenience sampling and some adolescent sub-populations, while adolescents in rural areas and those with disabilities were not adequately covered. Finally, most of the tools used were self-report measures and validated measures are desperately needed.

A literature review based on PsycINFO and MEDLINE was conducted to understand longitudinal adolescent studies in Hong Kong. Based on a review of 23 studies, several observations can be highlighted. First, there are very few positive youth development programs. Second, many intervention studies have not been rigorously evaluated. Third, many survey studies focused primarily on the negative aspects of youth development, neglecting the positive dimensions. Fourth, while the sample size of most studies is respectable, very few studies used random sampling. Fifth, most studies used quantitative research methods, with very few studies using qualitative or mixed research methods. Sixth, only two waves of data were collected in most studies. Seventh, advanced statistical data analysis techniques were seldom employed. Finally, assessment tools in some of the studies were not validated.

This study examined the development of self-efficacy in the high school years and the related socio-demographic and family correlates. A longitudinal research design was used to collect data from students in Grade 7 to Grade 12. At each wave, students responded to measures of socio-demographic characteristics (gender, economic disadvantage and non-intactness), family processes (family functioning and parent-child relational quality), academic and school competence and self-efficacy. Results showed that self-efficacy increased in the adolescent years. Regarding socio-demographic predictors, economic disadvantage and family non-intactness were related to adolescent self-efficacy. Findings also showed that family processes (family functioning and parent-child subsystem quality) and academic and school competence were related to adolescent self-efficacy but the nature of relationships was more complex than expected.

Based on six waves of longitudinal data collected from high school students, the study examined developmental pattern and correlates of resilience in adolescents in Hong Kong. Over 6 years, the students were invited to complete a questionnaire containing measures of psychosocial functioning including family functioning, parent-child subsystem quality and resilience. There was a slight decrease of resilience throughout adolescence. Adolescents having better parent-child subsystem quality and family functioning reported higher levels of resilience initially. However, better mother-child subsystem quality and family functioning were significantly associated with faster decrease in resilience, although adolescents with better parent-child subsystem quality and family functioning always reported higher resilience levels over 6 years. The findings suggest that strengthening family processes can help to promote resilience in adolescents in Hong Kong.

To understand adolescent development in the Chinese context, developmental trajectories of adolescent emotional competence over time and the related socio-demographic and family determinants were examined in a 6-year longitudinal study. Over six waves of data collection, students were invited to complete a questionnaire measuring emotional competence, paternal and maternal control, father-child and mother-child relationship, family functioning and socio-demographic characteristics. Results showed that adolescent emotional competence showed a linear increase across time. At the initial stage, gender, paternal and maternal control, father-child and mother-child relationship and family functioning influenced adolescent emotional competence. Over time, maternal control, mother-child relationship and family functioning predicted changes in the trajectory of adolescent emotional competence. The theoretical and practical implications of the findings are discussed.

The developmental trend and correlates of adolescent cognitive competence were examined in a 6-year longitudinal study in Hong Kong. From Secondary 1 to Secondary 6 years, the same cohort of high school students responded to validated measures of cognitive competence, socio-demographic status (i.e. gender, family intactness and economic disadvantage) and family processes (i.e. family functioning, father-child subsystem quality and mother-child subsystem quality). Results indicated that there was a rising trend of cognitive competence over the high school years. Adolescents with better family functioning or better father-child or mother-child subsystem quality demonstrated higher initial level of cognitive competence. However, adolescents with better family functioning showed slower growth rate on cognitive competence over time. Finally, teenage boys showed faster growth rate on cognitive competence than did teenage girls.

In response to the severe lack of longitudinal data on adolescent development, we conducted a 6-year longitudinal study examining the change and psychosocial correlates of social competence in Chinese adolescents in Hong Kong. During the high school years (Secondary 1 to Secondary 6), participants responded to measures of social competence, socio-demographic characteristics (gender, family intactness and economic disadvantage) and family processes (father-child relational quality, mother-child relational quality and family functioning). Results showed that social competence dropped across the six waves. While gender, father-child subsystem quality, mother-child subsystem quality, and family functioning significantly predicted social competence at the initial level, rate of decrease in social competence was greater for adolescents who came from intact families, had better mother-child subsystem quality, and reported better family functioning. Economic disadvantage was the only factor that did not affect the development of social competence in adolescents.

This six-wave longitudinal study in Project P.A.T.H.S. examined the growth trajectories and predictors of moral competence in Chinese adolescents in Hong Kong. Linear mixed models utilizing individual growth curves were used to analyze the initial status and rate of change of moral competence with reference to gender, economic disadvantage, family intactness, family functioning and parent-child subsystem quality. Results showed that adolescent moral competence increased across 6 years. Female adolescents had higher initial levels of moral competence than that of male adolescents. Adolescents from well-functioning families had higher initial levels of moral competence but slower growth of moral competence than did those from families with lower family functioning. Adolescents with higher mother-child subsystem quality had higher initial levels of moral competence but slower growth of moral competence than did those with relatively lower mother-child subsystem quality. Finally, adolescents with higher father-child subsystem quality had higher initial levels of moral competence than those with relatively lower father-child subsystem quality.

Based on the five waves of longitudinal data (n=3328 at Wave 1), the influence of positive youth development (PYD) attributes and family functioning on academic adjustment and academic stress in adolescents growing up in families with parental separation or divorce at Time 1 was examined. Compared with adolescents without the experience of parental divorce or separation, adolescents with such experiences displayed poorer wellbeing and academic adjustment as well as higher risk behavior. At Wave 5, both PYD and family functioning positively predicted academic adjustment but negatively predicted study stress. Longitudinally, Wave 1 PYD predicted school adjustment at Wave 3 and perception of the new secondary school curriculum at Wave 5. While Wave 4 PYD had an influence on all academic-related measures at Wave 5, Wave 4 family functioning predicted school conduct at Wave 5 only. The protective role of PYD and family functioning in academic resilience in adolescents growing up in non-intact families is discussed.

With the launching of the education reform in Hong Kong, the secondary school structure was changed from 7 years to 6 years in the 2006–07 school year. This paper describes Secondary four students’ views about the new secondary school curriculum, including their confidence and related stress. The students were also asked to assess the importance of life skills and its perceived adequacy in the formal curriculum. Results showed that roughly four-tenths of the students did not feel confident about their study and around six-tenths felt stressed. Although most of the students agreed that life skills were important, around 37% of them thought that the coverage of such knowledge in the formal curriculum was not adequate. Consistent with our hypotheses, academic confidence, academic stress, support provided by the school, and positive youth development were significantly related, with positive youth development predicting academic confidence.