Health and educational aspirations in adolescence: a longitudinal study in Finland

The study was conducted as part of the project “Redefining adolescent learning: A multilevel longitudinal cohort study of adolescent learning, health, and well-being in educational transitions in Finland” – Metropolitan Longitudinal Finland (MetLoFin) –. It follows a large cohort of students from the Helsinki Metropolitan Region from the lower secondary education to the end of upper secondary education. In 2011, all 7th graders (12–13 years old) were invited to participate. The recruitment occurred through the educational authorities of all 14 municipalities of the Helsinki Metropolitan Region, each of which gave a permission for the study. A follow-up survey was fielded in 2014 when the students were in the 9th grade (15–16 years old).

The study protocol was approved by the Ethical Committee of the Finnish Institute for Health and Welfare. In line with the instructions of the Finnish National Board on Research Integrity (TENK) in 2009, no parental consent was required when the study was conducted as part of the students’ normal schoolwork. Two of the 14 municipalities had adopted a policy that a written parental consent is always required. These were collected. In the other municipalities, information letters were sent to the parents who had the possibility to withdraw their child from the study. The students were instructed about the purpose of the study and that participation was voluntary and that they can decline to answer any question or withdraw from the survey at any time. This was mentioned at the beginning of the questionnaire at the first pageRegistry data on students’ educational aspirations were obtained from the Finnish National Agency for Education, covering the period from Spring 2014 to Spring 2017. In Finland, students can apply via the Joint Application System to a maximum of five study places in upper secondary schools, ranked in the order in which they wish it to be selected. There are two general application rounds—Spring and Autumn—which are followed by an additional application round in which students can apply for vacant study places. Combining the survey answers with the Joint Application Registry was done by a data manager who does not analyze the data himself.

In total, 13,012 students belong to the baseline sample of the MetLoFin project (for a flow diagram representing the formation of the study population, see Fig. 1). In total, 9.078 students (50.0% female) answered the health questionnaire in the 7th grade (response rate of 69.8%). Of these students, 5.741 also participated in the 9th grade (attrition rate of 36.8%). We excluded from the analyzes, those students who never applied via the Joint Application System (n = 50), who had applied for special education at some point (n = 41) [30], or who came from schools where less than five students gave valid answers [31]. The analyzed sample consists of the remaining 5.614 students from 116 schools who responded to both surveys and fulfilled our inclusion criteria. The numbers in the final analyses differed due to missing information in the predictor variables.

The information available in the Joint Application System was used to construct an objective measure of students’ educational aspirations. The resulting variable had three categories: students who “Applied for the academic track” (58.0%, n = 3.258), “Applied for the vocational track” (19.8%, n = 1.111), or were undecided about their future plans and “Applied for both educational tracks” (22.2%, n = 1.244). We treated the recordings of students’ choices as educational aspirations regardless some of the students when applying for a study place did not know whether their GPA will be good enough to be selected, and some of them did not acquire any place to study. Nevertheless, these were their aspirations.

Multilevel multinomial logistic regression analyses with random effects were estimated with generalized structural equation modeling using Stata Version 15. First, we calculated the variance in educational aspirations attributable to differences between schools which the students attended in the 7th grade. Second, we regressed students’ choices between the educational tracks on their health in the 7th grade (12–13 year-olds) and repeated this analysis with students’ health in the 9th grade (15–16 year-olds). Third, we controlled for the sociodemographic background of the students. Fourth, students’ academic achievement was entered into the models. Finally, we looked at within-person changes in health factors over time. Students’ health factors in the 7th grade were included in this analysis to account for starting levels and potential ceiling effects. The results of this analysis of within-person change, however, should not be interpreted as fixed effects estimates because our outcome variable educational aspirations did not change over time [39]. In all models, we controlled for gender differences. The model parameters were presented as odds ratios (OR) with 95% confidence intervals (CI). Akaike (AIC) and Bayesian (BIC) information criteria were reported for comparing the fit of the models to the data. Interaction effects between gender and health factors were not significant (results not shown).

An independent samples t-test revealed that students that answered the survey in both the 7th and the 9th grade were more likely to have better grades than those who dropped out (p < .001). Chi-squared tests revealed that, in the 7th grade, non-participants were also more likely to have psychosocial problems, long-term illness and medicine prescribed, and more daily health complaints (p < .001). There were also statistically significant differences in frequencies for all sociodemographic variables but gender. Participants were more likely to have highly educated and working parents (p < .001) and to live in a nuclear family (p < .05), and were less likely immigrants (p < .001) than non-participants.

Already in 12–13 year-olds (the 7th grade), all health factors were associated with students’ educational aspirations recorded more than 2 years later (Table 2). Concerning the SDQ, having a slightly raised or a high difficulty score, compared to having normal levels of psychosocial problems, was associated with a proportionally higher likelihood to apply for the vocational rather than the academic track. Students with psychosocial problems were also more likely to be undecided about their plans for upper secondary education (i.e., applied for both tracks). The associations were robust to accounting for both sociodemographic background and academic achievement. There was, however, one exception: The associations became non-significant for the high difficulty score category. Having one psychosomatic symptom nearly every day, compared no daily health complaints, was associated with applying for the vocational track. Also, students who were undecided in their future educational plans were more likely to report one psychosomatic symptom. Both associations remained significant after including sociodemographic background variables and academic achievement in the model. Having medicine prescribed by the doctor, compared to no long-term illness, was positively associated with applying for the vocational track. In the models that controlled for all other predictors, having a long-term illness was no longer significantly associated with educational aspirations. Students who reported average or poor health, compared to good health, were more likely to belong to the group that had not decided yet and had thus applied for both educational tracks and these associations were robust to controlling for sociodemographic background and academic achievement.

We also found significant associations when health was assessed in the same year (in the 9th grade, at the age of 15–16 years) in which Finnish students have to decide about their upper secondary education (Table 3). The more psychosocial problems were reported, the more likely the adolescents were to apply for the vocational track instead of the academic track or the more often they were undecided in their choice between tracks. Similar to the results for the SDQ in the 7th grade, all associations persisted over and above sociodemographic background and academic achievement. There was also a weak association between daily health complaints and educational aspirations in this age group. Having two or more psychosomatic symptoms nearly every day, compared to no symptoms, was related to applying for the vocational track rather than the academic track. Having a long-term illness with and without medicine prescribed was associated with being less undecided. The association for the use of medicine persisted over and above sociodemographic control variables. Educational aspirations were also robustly associated with self-rated health with one exception: reporting average or poor health, compared to good health, was no longer associated with applying for both educational tracks when including sociodemographic background into the model.

Of the sociodemographic variables, all but immigrant background showed associations with students’ future plans for upper secondary education (Tables 2 and 3). Across the models, boys were less likely than girls to apply for academic track only. Applying for the academic track, instead of the vocational track or to both, was associated with students’ parental background in terms of higher education and nuclear family structure, while the association between applying for vocational track and parental employment disappeared after accounting for academic achievement. Unsurprisingly, especially academic achievement was a very strong and significant predictor of educational aspirations.

The models that used the data of both surveys simultaneously to assess within-person change from the 7th to the 9th grade and its association with educational aspirations are presented in Table 4. We found a relationship of worsening of health in regards to the SDQ with applying for the vocational track and applying for both tracks. Improvement in this health factor, however, was only weakly associated with applying for the vocational track. Moreover, getting worse health in regards to long-term illness was associated with a decreased likelihood of applying for both educational tracks. Finally, worsening self-reported health over time increased the likelihood to apply for the vocational track. Remarkably, all associations between educational aspirations and increasingly worse health persisted when the sociodemographic background and academic achievement were controlled for.

We cannot exclude bias in our results due to selective attrition. Without attrition, however, the observed effects of the studied health and social factors on educational aspirations might have been even stronger because those who were less healthy and from a more disadvantaged family background were less likely to participate in the second survey.

Among the considerable strengths of the research is the fact that we used a longitudinal multilevel design to understand how health in adolescence links to the choice between educational tracks that took into account the significant effect of the school attended on educational aspirations. Very few, if any other large adolescent cohorts have covered health and education as comprehensively both in terms of health indicators and the opportunity to follow the same individuals over the transition to further education after the compulsory schooling ends. Assessing health longitudinally enabled us to identify those periods in adolescence which are sensitive for their successful paths into adulthood and to examine the effects of within-person change in health over time. Educational aspirations were assessed objectively by obtaining from the national registry covering all students in the country, the choices they have made when applying to upper secondary education. Using national registry data further reduced measurement error and the amount of missing data due to nonresponse.