Educational Expectation from Adolescence to Young Adulthood: A Latent Growth Modelling Analysis

• In Event: Poster session 2

Proposal

1. Background and Objectives
Over the past 50 years, a number of research studies have shown that educational expectation is an important predictor of educational attainment (Jacob & Wilder, 2011). Yet, rarely have the changes in educational expectation across adolescence and young adulthood been examined (Johnson & Reynolds, 2013). This study investigates the changes in educational expectation during adolescence and young adulthood and the possible contributing factors by analyzing longitudinal data with latent growth modelling.

2. Methods
The data is taken from a sample of 2728 Hong Kong young adults who have participated in PISA 2012 and its follow-up 5-year longitudinal study, namely HKLSA, until 2017. At the age of 15, the participants were assessed on PISA tests of mathematics, reading and science, and background information were collected including gender, father’s and mother’s educational level, highest parental occupational status and four kinds of family resources (i.e., cultural possessions, home educational resources, ICT resources and material resources). Educational expectations were also collected at four time points when they were aged 15 (secondary 4), 16 (secondary 5), 18 (postsecondary year 1) and 20 (postsecondary year 3), which are represented by a dichotomous variable (1 = bachelor or above; 0 = lower than bachelor).

3. Results and Discussion
Descriptive analysis shows that the sample comprises of similar proportions of males (51.2%) and females (48.8%). At age 15, the percentage of participants who expected for a bachelor degree is 54.8%, which is much lower than the respective percentage of many other countries (e.g., 81% for Korea and 70% for Singapore).

This percentage increases steadily to 68.3% (age 16) and 83.0% (age 18), and then levelled off at 83.5% (age 20), indicating that the educational expectation of Hong Kong young people may become more stable at age 18 during the transition to young adulthood.

Three latent growth models are constructed for investigating the effects of time (Model 1), gender and family socio-economic background (Model 2), and cognitive abilities (Model 3) on the change in educational expectation. In Model 1, both linear and quadratic components of the time variable are incorporated to investigate whether there is a nonlinear change in educational expectation over time. Results show a significant linear increase in educational expectation over the four time points [odds ratio = 2.132, p < 0.001], indicating that the participants are 2.132 times more likely to expect for a bachelor degree when they move from one time point to another. The quadratic effect is also significant [odds ratio = 0.910, p < 0.001], which indicates a decelerated growth in educational expectation over time starting from age 18.

For Model 2, results indicate that gender does not have any significant effect on educational expectation or its growth rate. On the other hand, father’s and mother’s educational level, highest parental occupational status, cultural possessions and home educational resources have significant positive effects on educational expectation, with an odds ratio ranging from 1.014 to 1.344. Yet, these family socio-economic background factors do not have any significant effect on the growth rate of educational expectation.

For Model 3, results indicate that after taking into account of gender and family background factors, PISA scores in mathematics and reading have significant positive effects on educational expectation. Yet their odds ratios (from 1.004 to 1.006) are relatively smaller than those of family background factors (from 1.010 to 1.312), which suggests that family background has a stronger impact on young people’s educational expectation. Interestingly, PISA score in mathematics has a small yet significant positive effect on the growth rate of educational expectation [odds ratio = 1.002, p < 0.05], which implies that mathematical literacy may have an important effect on the development of educational expectation.

Overall, the present study has shown a continuous yet decelerated growth in educational expectation from adolescence to young adulthood among Hong Kong young people. Further studies may investigate how this growth pattern is related to the lack of alternative pathways to university education in Hong Kong. Besides, while some of the cognitive abilities and family socio-economic background factors are positively related to educational expectation, the latter appears to exert a greater effect, which may possibly be a source of cultural reproduction.

Authors

• Esther Sui Chu HO, The Chinese Uni of Hong Kong

• Kwok Wing Sum, The Chinese Uni of Hong KOng