Once in NEET, always in NEET? Childhood and adolescent risk factors for different NEET patterns Open Access

Abstract

Background

Young adults in Neither in Employment, Education nor Training (NEET) are at risk of adverse labour market outcomes. Earlier studies often measured NEET status at one time point or compared persistent NEETs with non-NEETs, neglecting other patterns of NEET status. Evidence on early life factors associated with NEET patterns is lacking. This study aims to (i) identify patterns of NEET status over time and (ii) examine whether factors in childhood and adolescence are associated with these patterns.

Methods

Data were used from 1499 participants of the TRacking Adolescents’ Individual Lives Survey (TRAILS), a Dutch prospective cohort study with 15-year follow-up. NEET status was assessed at ages 19, 22 and 26. Socioeconomic status of parents (SES), intelligence and negative life events were measured at age 11, educational attainment at age 26 and mental health problems at ages 11, 13.5 and 16. Data were analyzed using multinomial logistic regression analysis.

Results

Four NEET patterns were identified: (i) non-NEETs (85.2%), (ii) early NEETs (4.5%), (iii) late NEETs (5.7%) and (iv) persistent NEETs (4.5%). Reporting internalizing problems at age 11 was a risk factor for early and late NEETs [odds ratio (OR) 2.77, 95% confidence interval (CI) 1.16–6.62; OR 5.00, 95% CI 2.22–11.3, respectively]. Low parental SES, lower intelligence scores and negative life events (≥3) were risk factors for persistent NEETs (OR 4.45, 95% CI 2.00–9.91; OR 0.96, 95% CI 0.94–0.98; OR 4.42, 95% CI 1.62–12.08, respectively).

Conclusions

The results highlight the importance of timing and duration of NEET status and emphasize the need for tailored interventions to prevent specific NEET patterns.

Introduction

Young adults outside both the educational system and the labour market early in life, also referred to as young adults in NEET (Neither in Education, Employment or Training), experience an increased risk of adverse labour market outcomes and eventually social exclusion later in life.^1–3 A recent review showed that being female, having a low educational attainment, a low parental socioeconomic background, several negative life events and mental health problems are the most common risk factors in childhood and adolescence risk factors for being in NEET.^4,5

Not all young adults in NEET are at risk for adverse labour market outcomes or social exclusion. Young adults in NEET are a very heterogeneous group, and some young adults are only temporarily in NEET.^6,7 Although a short period of being in NEET can be a natural part of the school-to-work transition,⁸ knowledge is lacking about the timing and duration of young adults’ NEET status, as NEET status is often measured at only one time point.^1,6,8–11 Studies that do take the duration of NEET status into account often compare persistent NEETs with non-NEETs, neglecting other patterns of NEET status.¹² Two recent studies identified, next to the trajectories of young adults who were never or persistently in NEET, trajectories of young adults who were temporarily in NEET.^13,14 These studies showed that both young adults who are persistently in NEET and those who are temporarily in NEET reported lower levels of educational attainment and a higher risk of late-life drug abuse.^13,14 The temporary NEETs might include young adults who manage to go back to school or work after being in NEET, or young adults who complete their education successfully but are not successful in entering the labour market. Improving our understanding of NEET patterns and identifying early childhood and adolescent factors associated with different NEET patterns may provide valuable input for policy and practice and may eventually help to prevent adolescents and young adults from being in NEET.

The aims of this study are to (i) identify NEET patterns in young adulthood and (ii) examine whether early childhood and adolescent factors are associated with these different NEET patterns. Unique data from the TRAILS study (TRacking Adolescents’ Individual Lives Survey), a large prospective cohort study with repeated measures and 15 years of follow-up, are used.

Methods

Ethics statement

The Dutch Central Committee on Research Involving Human Subjects approved all the protocols of the TRAILS study. All participating children and their parents provided written informed consent.¹⁵

Study design and sample

The study used data from the TRAILS study, an ongoing longitudinal cohort study on the psychological, social and physical development of children towards adulthood. The TRAILS study started in March 2001, and 3145 children born between 1989 and 1991 in the Northern part of the Netherlands, in both urban and rural areas, were approached for participation. In total, 2230 children and their parents provided informed consent and were eligible to be included in the study. Participation rates varied between 96.4% of baseline at the second wave (N = 2149) and 72.6% at the sixth wave (N = 1618). More detailed information about the TRAILS study can be found elsewhere.^15–17 The present study used data of 1708 TRAILS participants (76.6% of baseline), for whom the NEET status was known at the fourth wave (age 19). In the Dutch educational system, adolescents follow compulsory education until the age of 18, meaning that the NEET status could not be determined before the age of 18.

Measures and procedures

Outcome

NEET status was determined at ages 19, 22 and 26 by asking the participants if they were engaged in education or employment. At each measurement wave, participants were categorized into two groups: (i) young adults not in NEET or (ii) young adults in NEET. If the NEET status could not be defined at age 22 or age 26, the NEET status at age 19 or 22 (i.e. the previous measurement wave) was used, i.e. last case forward approach (N = 262). When the educational or employment status at age 19 was unknown or when the NEET status was missing at both the fifth and the sixth measurement waves, cases were excluded (N = 209). In all, the NEET status of 1499 participants (67.2% of baseline) was determined at ages 19, 22 and 26.

Childhood and adolescent risk factors

Sex, socioeconomic status of parents (SES), educational level, physical health, negative life events, intelligence and mental health problems were considered as potential childhood or adolescent risk factors to affect NEET patterns in young adulthood.

Parental SES was assessed at 11 years of age, and a scale score per family was calculated including the educational and occupational levels of the mother and father and household income. SES was categorized into low, medium and high. Low SES refers to the lowest quartile, medium SES to the two middle quartiles and high SES to the highest quartile.

Educational level was measured at age 26 by asking the participants about their highest obtained educational level. Educational level was categorized into low (primary, lower vocational and lower secondary education), medium (intermediate vocational and intermediate secondary education) and high (higher secondary, higher vocational education and university).

Physical health in the past 2 years was measured at age 11 by asking the participants to rate their physical health on a 5-point Likert scale (1 = very poor to 5 = very good), which was recoded into poor health (response categories 1 and 2) and good health (response categories 3–5).

Negative life events (yes/no) were assessed at age 11 with eleven items, mostly answered by one of the parents. Items concerned family poverty (i.e. monthly net family income <€1135), divorce of parents, death of parent(s) or sibling(s), serious illness of parent(s) or sibling(s), not living at home for more than 3 months, hospitalization (more than once), having an injury, serious illness of a friend and being bullied. Being bullied was measured by peer nominations. A sum score was calculated and categorized into no negative life events, one or two negative life events, and three or more negative life events.

Intelligence was estimated at age 11 with the Vocabulary and Block Design subtests of the Revised Wechsler Intelligence Scales for Children (WISC-R),¹⁸ leading to IQ scores.^19,20

Mental health problems were measured at ages 11, 13.5 and 16 years with the Youth Self-Report (YSR).^21,22 Participants were asked to answer the items regarding behavioural and emotional problems in the past 6 months, with 0 (not true), 1 (somewhat or sometimes true) or 2 (very often true). Two scale scores were calculated: externalizing problems and internalizing problems. The scale for externalizing problems contains the subscales aggressive behaviour and delinquent behaviour. The scale internalizing problems comprises the subscales anxious/depressed behaviour, somatic complaints and withdrawn/depressed behaviour. Standardized scores were used, with higher scores indicating more mental health problems. As, on average, the onset of mental health problems is around 14 years of age, mental health was included not only at age 11 but also at age 13.5 and 16.^23,24

Statistical analyses

Based on the NEET status at ages 19–26, eight individual trajectories of NEET could be determined. These individual trajectories were then manually clustered into distinct NEET patterns, according to the following rules: Participants who were never in NEET were referred to as non-NEETs, those who were in NEET at two or three time points were referred to as persistent NEETs. In the Netherlands, the average age of leaving school is around 23 years of age.²⁵ Therefore, participants who were in NEET at age 19 or at age 22, were referred to as early NEETs. Participants who were in NEET at age 26, were referred to as late NEETs. Differences between NEET patterns were tested with Chi-square tests for categorical variables and F-tests in one-way analysis of variance for continuous variables.

To examine whether childhood and adolescent factors at ages 11, 13.5 and 16 are associated with different NEET patterns between ages 19 and 26, multinomial regression analyses were performed. First, univariate associations between childhood and adolescent factors and NEET patterns were examined, followed by multivariate models including all childhood and adolescent factors. Multivariate analyses including all childhood and adolescent factors were performed separately for mental health problems at ages 11, 13.5 and 16.

A comparison on childhood and adolescent factors was made between participants who were excluded from further analysis as their NEET status at age 19 or at age 22 and 26 was unknown (N = 209), and participants in the final study sample (N = 1499). No data were missing for the childhood and adolescent risk factors. Because of the potential overlap between income in the SES measure and family poverty, a sensitivity analyses was performed in which poverty was removed from the list of negative life events. All analyses were conducted with SPSS version 26.

Results

Sample characteristics

The total sample consisted of 1277 participants who had never been in NEET (85.2%) and 222 participants who had been in NEET at least once between age 19 and 26 (14.8%) (Supplementary table S1). The descriptive statistics of the negative life events are presented in Supplementary table S2. Eight unique NEET trajectories were identified. These trajectories were manually clustered into four distinct patterns: (i) non-NEETs (i.e. young adults who were never in NEET, N = 1277, 85.2%), (ii) early NEETs (i.e. in NEET at age 19 or 22, N = 68, 4.5%), (iii) late NEETs (in NEET only at age 26, N = 86, 5.7%) and (iv) persistent NEETs (i.e. in NEET at age 19 and/or at age 22 and/or age 26, N = 68, 4.5%) (Supplementary table S1).

Early NEETs were more often female than male (63.2% vs. 36.8%) (table 1). Of the early NEETs, 26.5% had never experienced any negative life events. The percentage of early NEETs with low parental SES was 18.5% and 12.7% reported a low educational level. Early NEETs had the highest intelligence scores [mean score 102.5, standard deviation (SD) 12.7 vs. 100.7, SD 14.2 for the non-NEETs], but not the highest percentage of a high educational level (36.5% vs. 38.8% for late NEETs).

Late NEETs were comparable to the early NEETs regarding sex (more females than males), low parental SES (19.8%), low educational level (14.1%) and the experience of negative life events (80.2% vs. 73.5%). Persistent NEETs more often had low SES parents (16.6% vs. 50.7%), a low educational level (4.1% vs. 49.0%) and were more often exposed to more than three negative life events (18.6% vs. 44.1%), compared with non-NEETs.

Childhood and adolescent risk factors at ages 11, 13.5 and 16 per NEET pattern at ages 19–26

Internalizing problems at ages 11, 13.5 and 16 increased the risk of being early and late in NEET [odds ratios (ORs) ranging from 2.77, 95% confidence interval (CI) 1.16–6.62; OR 5.00, 95% CI 2.22–11.3] in the univariate models (Supplementary table S3). Furthermore, internalizing problems at ages 13.5 and 16 increased the risk of being persistent in NEET.

In the multivariate analyses, only having internalizing problems at age 11 increased the risk of being early in NEET (OR 3.98, 95% CI 1.26–12.6) (table 2). Internalizing problems at age 13.5 and age 16 increased the risk of being early and late in NEET (OR 6.59, 95% CI 2.22–19.5; OR 4.30, 95% CI 1.61–11.5), but not of being persistently in NEET (tables 3 and 4).

Externalizing problems at age 13.5 were identified as a risk factor for late NEETs and persistent NEETs in the univariate models (OR 3.04, 95% CI 1.01–9.12; OR 5.88, 95% CI 1.88–18.4), as did externalizing problems at age 16 for persistent NEETs (OR 4.18, 95% CI 1.35–12.9) (Supplementary table S3). In the multivariate models (tables 3 and 4), these associations attenuated and were no longer significant. The results of the multivariate models showed that low parental SES (OR 3.09, 95% CI 1.27–7.51), intelligence (OR 0.95, 95% CI 0.94–0.97) and exposure to three or more negative life events (OR 2.72, 95% CI 1.02–7.26) in childhood and adolescence increased the risk of being persistently in NEET (tables 2–4).

Additional analyses

Excluded participants without NEET status data at age 22 and 26 (N = 209) were more often boys, had low SES parents, had a lower educational level and intelligence scores, were more often exposed to three or more negative life events, and higher levels of externalizing problems (see Supplementary table S4). The comparison of the results of the sensitivity analyses showed that estimates were slightly smaller in the analysis without family poverty than in the analyses with family poverty, but yielded similar results.

Discussion

Four different NEET patterns were identified: non-NEETs, early NEETs, late NEETs and persistent NEETs. These patterns, including those of young adults who are temporarily in NEET, are comparable to those found by Giret et al.¹³ and Manhica et al.¹⁴ Internalizing problems at age 11 were identified as risk factors for early and late NEETs, whereas low parental SES, lower intelligence scores and exposure to three or more negative life events were identified as risk factors for persistent NEETs.

According to the OECD, more than 40% of all young adults in OECD countries have ever been in NEET.²⁵ For the early NEETs in our study, being in NEET may not have been too problematic, as they managed to get back in education or found a job, but the scaring effect of being in NEET on future employment and mental health cannot be excluded.^26–28 The group of early NEETs consists of more females than males, and almost half of them had parents with a high SES. This group reported the highest mean intelligence score but not the highest percentage of a high educational level. An explanation may be that they drop out of the educational system due to their internalizing problems. Earlier research clearly showed that young adults, and especially females, who suffer from depressive symptoms or anxious complaints have difficulties at school and are more likely to drop out.^29–31 The late NEETs do not stand out for any of the examined childhood and adolescent characteristics. It may be that this group has other reasons to be in NEET. For example, they may be taking a break, travelling around the world after finishing their school or have care duties. As we do not know the reason why they are in NEET, it is important to follow them even longer to see whether their NEET status has a temporary or persistent character. Further follow-up with young adults who have ever been in NEET may also shed light on the question of whether they are facing the scarring effect of their NEET period.^26–28 Gutiérrez-García et al.³² showed that among Mexican young adults, an important reason for being NEET was that they were unable to return to school or to find a job. This particular reason was also mentioned by homemakers, a NEET group often seen as being voluntary NEET and having a legitimate reason to be NEET.⁷ Further research combining the timing and duration of being NEET with the main reasons for being NEET is needed and may provide valuable anchor points for interventions and policy.

Persistent NEETs can be characterized as those with relatively limited socioeconomic resources (reflected by low parental SES, educational level and intelligence scores) and high levels of externalizing problems, although the latter was not identified as a risk factor for being persistently in NEET. The results regarding socioeconomic resources are in line with findings from previous studies^2,33,34 and suggest that the persistent NEETs can be seen as a different group than the early and late NEETs. Young adults who are persistently in NEET need help and support from parents, teachers or social workers to keep them in the educational system as long as possible, or at least until they have a basic educational level. Achieving a basic educational level may improve their chances on the labour market to find and maintain a job.^35,36 The findings of this study also showed that the experience of three or more negative life events increased the risk of becoming persistent in NEET. To our best knowledge, no previous study has examined the association between negative life events and NEET status. However, we do know that the experience of negative life events increases the risk of high school dropout, low educational attainment and poor labour market participation.^37,38 The finding that internalizing problems at age 13.5 and 16 were only associated with being persistent in NEET in the multivariate models and not in the univariate models is counterintuitive. It may suggest that other factors, such as low parental SES and the experience of multiple negative life events, diminish the effect of internalizing problems. We can only speculate about explanations and further research is needed to better understand the relationship between internalizing problems and being persistent in NEET.

This study shows that negative early life experiences may have long-lasting consequences for inclusion in school and work, thereby emphasizing the importance of adopting a life-course perspective. The finding that childhood and adolescent mental health problems increase the risk of being in NEET in young adulthood emphasizes the long-lasting and drastic impact of early mental health problems. For persistent NEETs, the findings suggest that their adverse educational and employment status in young adulthood may be the result of the accumulation of early life factors, i.e. the exposure to low SES and negative life events. Ensuring a smooth transition and integration into the labour market, especially for vulnerable young adults, is of utmost importance, as we know from previous studies that early labour market marginalization increases the risk of long-term labour market disadvantages and instability.²⁶

In the present study, the reason for being NEET was unknown, and the heterogeneous nature of the NEET groups could therefore not be taken into account when identifying different NEET patterns. Some of the identified young adults in NEET may have been in NEET for a good reason, and not all young adults in NEET are at risk of social marginalization or exclusion.^2,39 As suggested by Carcillo et al.² and Elder,³⁹ a breakdown of the NEET groups into those who are unemployed and those who are inactive might provide valuable information. It is recommended to take this additional information into account in future research. Furthermore, to unravel the dynamics between mental health, life events and NEET status over time, we recommend future studies use bigger sample sizes and more sophisticated approaches.

For practice, the results of this study imply that early detection and treatment of mental health problems is crucial and that practitioners should be aware of the long-term consequences of low parental SES, and experiencing mental health problems and/or multiple negative life events. For those adolescents whose mental health problems are lasting, extra support during the school-to-work transition may be beneficial.

A strength of this study is that follow-up data were used from a large population sample with relatively high retention rates ranging from 72.6% to 96.4%. Another strength is that the NEET status and mental health problems were assessed at different time points, covering a total follow-up period of 15 years from childhood to young adulthood, which enhanced the reliability of their assessment. Furthermore, mental health problems were assessed with the most commonly used self-reported questionnaires in child and adolescent research, i.e. the YSR, a valid and reliable measure.^21,22

The results of this study have to be interpreted in light of some limitations. The reported findings may underestimate the number of participants in NEET. Excluded participants reported lower intelligence scores, lower educational levels, and experienced more negative life events, suggesting potential selection bias and that data was not missing at random. Furthermore, it might be that misclassification has occurred, as the participants were asked about their educational and employment status at a particular moment in time, which may not always represent the real situation. For example, a participant may have had a job at the particular measurement wave but was in NEET the whole year before. More detailed data, i.e. event history data, is needed to tackle this potential problem. However, measuring NEET status at different time points has also shown the importance of identifying different NEET patterns, as we found different risk factors per NEET pattern. Misclassification may also have occurred as a last case forward approach was employed, i.e. if NEET status was unknown at age 22 or 26, the NEET status of the previous measurement wave was used. However, the results from a complete case analysis were similar.

In conclusion, we found four different NEET patterns, and once in NEET does not mean always in NEET. Internalizing problems in childhood and adolescence were identified as risk factors for becoming in NEET, but only for early and late NEETs. A lack of socioeconomic resources and exposure to negative life events were identified as risk factors for persistent NEET status. Monitoring during (early) adolescence towards low parental SES, exposure to negative life events and internalizing problems seems important to smoothen the transition from school to the labour market.

Supplementary data

Supplementary data are available at EURPUB online.

Funding

K.V. was funded by the Netherlands Organization for Scientific Research (NWO) Vici Project (Today’s youth is tomorrow’s workforce: Generation Y at work; under grant number NWO Vici 453-16-007/2735) which was granted to U.B.; TRAILS was financially supported by the Netherlands Organization for Scientific Research (NWO) under grant number GB-MW 940-38-011; ZonMW Brainpower under grant number 100-001-004; ZonMw Risk Behavior and Dependence grant under grant number 60-60600-97-118; ZonMw Culture and Health grant under grant number 261-98-710; Social Sciences Council medium-sized investment grants under grant number GBMaGW 480-01-006 and GB-MaGW 480-07-001; Social Sciences Council project grants under grant number GB-MaGW 452-04-314 and GB-MaGW 452-06-004; NWO large-sized investment grant under grant number 175.010.2003.005; NWO Longitudinal Survey and Panel Funding under grant number 481-08-013, the Dutch Ministry of Justice (WODC) (no grant number); the European Science Foundation under grant number EuroSTRESS project FP-006; Biobanking and Biomolecular Resources Research Infrastructure under grant number BBMRI-NL (CP 32), Gratama Foundation (no grant number); Jan Dekker Foundation (no grant number); the participating universities (no grant number); and Accare Centre for Child and Adolescent Psychiatry (no grant number).

Conflicts of interest: None declared.

Data availability

Data may be obtained from a third party and are not publicly available. TRAILS data of the T1, T2, T3, T4 and T5 measurement waves are deposited in the Data Archiving and Networked Services of the Royal Dutch Academy of Sciences (DANS-KNAW) and access can be requested at http://www.dans.knaw.nl.

References