tracking experiences among Taiwanese youth and their dynamic associations with job-education match, job satisfaction,

and happiness. The sustainable careers concept (De Vos et al., 2020) views individuals as core actors, emphasizing personal,

situational, and temporal dimensions’ influence on career sustainability, expanding school-to-work transition research (Blokker

et al., 2023). However, existing studies often focus on adults and are cross-sectional, missing key youth changes and early

educational experiences’ impact on careers. While educational tracking’s role in occupational choice has been explored,

understanding of how job-education match affects job satisfaction and happiness, especially under different educational

backgrounds, remains limited. This study’s longitudinal analysis aims to provide a comprehensive perspective for sustainable

career theory and practice.

Research shows job-education match, job satisfaction, and happiness are interrelated. Job-education match includes

educational level, field of study, and skills/abilities matching, positively correlating with job satisfaction (Badillo-Amador &

Vila, 2013; Santis et al., 2021). From strengths-use perspective, knowledge and abilities need an environment for application,

with full utilization increasing work engagement, satisfaction, and happiness (Miglianico et al., 2020). Job satisfaction and

happiness’s positive association is often explained by the spillover hypothesis, including top-down and bottom-up effects.

Longitudinal meta-analyses show life satisfaction and happiness have cross-sectional associations and cross-lagged effects

(Bowling et al., 2010; Karabati et al., 2019; Steel et al., 2019). This study analyzes these relationships using data from Taiwan.

The impact of early educational experiences on later career development is worth investigating. Existing research on

Taiwanese youth indicates that high school students’ post-graduation development is influenced by learning experiences and

employment considerations. Key moments for youth to consider career development are during the educational tracking at high

school levels and when facing graduation from university (Kuo et al., 2021; Wu et al., 2018). In Taiwan’s education system,

the educational tracking at the high school level is a crucial stage in youth career development. Vocational high school students

mostly choose their major subjects upon graduating from junior high school, while students in academic high school choose

their stream (liberal arts or science) after completing their first year, and decide on their specific university major when applying

for higher education. This study employs longitudinal tracking to investigate how regret over high school track/major choices

in the third year influences subsequent career development, addressing limitations of retrospective methods in existing research.

It also focuses on youth with different educational tracking experiences, exploring how their subsequent career development trajectories differ. While previous studies have explored how educational tracking affects various work outcomes over time

(Birkelund et al., 2022; Delaruelle et al., 2020; Verhaest et al., 2018), there remains a gap in research specific to Taiwan´s unique

educational and sociocultural context.

Career exploration during university, especially actions to change or expand learning specialties (changing majors, minors,

double majors, taking programs), is a crucial experience for future career development. Educational level is also an important

factor. Although literature shows mixed results regarding the match between educational level and work, studies focusing on

Taiwanese workers mostly indicate that higher education levels help improve job satisfaction and happiness (Chen et al., 2020).

Therefore, this study uses longitudinal analysis to enhance understanding.

The statistical analysis adopts the random intercept cross-lagged panel model (RI-CLPM) (Mulder & Hamaker, 2021),

using data from the Taiwan Education Panel Survey (TEPS) and its follow-up database (TEPS-B) SH sample, with 1,637

participants. It examines the associations between job-education match, job satisfaction, and happiness at three time points:

ages 25 (T1), 30 (T2), and 33 (T3), incorporating relevant factors from the educational period into the model. In the model, jobeducation

match, job satisfaction, and happiness are time-varying factors, divided into random intercepts at the between-person

level to explore correlations, and within-person level to explore autoregression and cross-lagged effects. Educational tracking

type, regret over chosen major in the third year of high school (data from high school period), educational level, and career

exploration are time-invariant predictors of random intercepts, with gender included as a control variable. Additionally, the

study examines whether different educational tracks in high schools (vocational vs. academic) show different effects on various

paths and further observes by inputting two group samples into the model.

The main findings are as follows:

First, the random intercepts of job-education match, job satisfaction, and happiness are positively correlated among

Taiwanese youth, similar to previous research findings. Those with higher job-education match have higher job satisfaction and

happiness due to work content meeting expectations and knowledge and skills matching job requirements (Ilies et al., 2019;

Kristof-Brown et al., 2005; Unanue et al., 2017).

Second, job-education match and happiness show autoregressive effects across the three waves, while job satisfaction only

shows autoregressive effects in the latter two waves. The autoregressive effect of job-education match reflects the profound

impact of the educational stage on career paths. The autoregressive effect of happiness may indicate that increasing age and

stable living environments can increase the stability of happiness. Job satisfaction shows no autoregressive effect from 25-30 but

does from 30-33, reflecting more career transitions and exploration before 30, with less obvious job satisfaction stability. After

30, higher chances of finding ideal jobs and clearer career positioning lead to more significant job satisfaction stability (Wille et

al., 2012).

Third, except for happiness at age 30 positively predicting job satisfaction at age 33, other cross-lagged effects are not

significant. This positive effect aligns with the top-down effect of the spillover hypothesis, i.e., overall life satisfaction extends

to the work domain (Steel et al., 2019). However, the cross-lagged effects between job-education match and job satisfaction are

not obvious, possibly because after entering the workforce, job content and responsibilities may increase, and expectations for

working conditions become more complex.

Fourth, overall, the moderating effect of educational tracking is not obvious, with only a few paths differing between ages

25 and 30. The autoregression of happiness T1 to happiness T2 is higher for those in vocational tracks than those in academic

track; the influence of job satisfaction T1 on happiness T2 and job satisfaction T1 on job-education match T2 is higher for those

in academic track than those in vocational track. This suggests that in the early career stage, the two educational track have

their own characteristics. However, the overall total path effect difference is not significant, indicating that regardless of the

educational track, the within-person career process experiences similar changes.

Fifth, regarding predictor variables, factors such as educational level, third-year high school major regret, and career

exploration during university have different effects depending on the type of educational track. Educational level has a

significant positive predictive effect on job-education match in both tracks; however, while educational level positively predicts

happiness in the academic track, it has no significant effect in the vocational track. Career exploration has a positive effect on

job-education match for those in the academic track, but no effect for those in the vocational track. This study particularly notes

that regretting the chosen track/major in the third year of high school negatively predicts job satisfaction and happiness, but only for vocational track participants. These students may lack occupational knowledge or career exploration when transitioning

from junior high, yet must make choices determining their professional field at age 15. If they regret their choice, changing

fields or retraining may be difficult, affecting job choices and satisfaction. Academic track participants who regret their chosen

track (liberal arts or science) in high school can still select their desired major when entering university at age 18, impacting

career satisfaction differently. This paper suggests these findings may reflect issues related to early tracking, warranting further

investigation.

Regarding research limitations, this study spans many years, and youth may experience job changes, socioeconomic

conditions, and family status changes that affect their careers. However, for model simplicity, this study did not include these

time-varying factors in the analysis, which future research could further incorporate. Although this is a longitudinal study,

inferring causal relationships still requires more exploration, such as mediating effects of other variables or qualitative data

support. The three waves in this study have unequal time intervals, limiting effect comparisons and making it inappropriate to

constrain the model, leading to insufficiencies in inference and model simplification, so it is recommended that future studies use

longitudinal data with equal intervals. The three main variables in this study all use single-item measures as observation sources,

which cannot reflect complex constructs; future research should consider multi-item studies. This study treats educational level,

career exploration, and educational tracking as time-invariant factors, unable to rule out potential changes for a small number

of people between different waves. Understanding of the types of changes in different variables at different time points for

participants is limited; future research could explore this through latent heterogeneity analysis to benefit practical work.