Characterizing Spanish labour pathways of young people with vocational lower-secondary education

Abstract

School-to-work transition is no longer considered to take place at one point but it is seen as a sequential process with multiple stages that extend over a relatively long period in which individuals gradually settle down into the labour market. This article constructs and examines the labour pathways followed by Spanish young people who have completed vocational lower secondary education. To do so, we use the Survey on Educational-Training Transition and Labour Integration, a retrospective longitudinal dataset that contains monthly calendar information about the labour states in the 4 years after completing vocational education. A wide range of states is covered: part-time employment, full-time employment, unemployment and several situations of inactivity. To examine young labour pathways we use sequence analysis, an exploratory technique which allows us to obtain a measure of dissimilarity between sequences. Later, a cluster analysis is applied that allow us to identify seven patterns of transition into the labour market. Nevertheless, most trajectories may be classified as linear, with a successful integration into the labour market. A multinomial logit model confirms that demographic and educational variables are relevant to explain the pathways followed by young people.

Keywords:

• school-to-work transition
• sequence analysis
• vocational training

JEL Classification:

• J24
• J64
• I21

Notes

¹ Nowadays, Spanish Vocational Training is ruled by the Act on the General Organization of the Education System (Ley Orgánica de Ordenación General del Sistema Educativo, LOGSE) passed in 1990.

² Access to the lower level cycles can be attained once the certificate of basic education has been obtained. This requirement, which ensures a minimum level of knowledge and abilities, has contributed to increase the quality of these training cycles.

³ In the academic year 2000/2001, there were 304 580 people enrolled in lower vocational training system, of which 130 715 were women. This represents 28% of total students enrolled in upper secondary education and 24% of women. In 2008/2009, the rates have increased and 32% of all students enrolled in upper secondary education opt for vocational studies (28% of women).

⁴ Also, low retention rates and drop-outs could increase the exclusion problem.

⁵ Gangl (2003) distinguishes between three patterns of school-to-work transitions depending on the differences in educational systems. Spain is classified as a country with a low level of educational attainment and a weakly developed vocational training system.

⁶ To know more about several theories explaining the vocational effect on the transition from school to work, see Iannelli and Raffe (2007).

⁷ See Instituto Nacional de Estadística (2007) for a more detailed description of data.

⁸ In any case, employment has always been the prevalent state. For instance, those individuals who are working part-time and studying at the same time are considered to be in the employment state.

⁹ As derived from explanation, QHA relies on two choices: the number of intervals in which the study period must be divided and the amplitude of intervals, because these amplitudes do not necessarily have to be of equal length.

¹⁰ For a review, see Abbott and Tsay (2000), Chan (1995), Halpin and Chan (1998), Brzinsky-Fay et al. (2006), Halpin (2010).

¹¹ Substitution is also interpreted as a combination of an insertion and a deletion. For example, transforming ABA into AAA involves deleting B and then inserting an A.

¹² These two operations used to be treated in the same way, given that deleting an element from a sequence is the same as inserting this element in the other sequence. For example, the difference between sequences ABA and AA can be eliminated either by removing B from ABA or adding B to AA.

¹³ Assigning costs to indel operations is useless when the sequences have equal length.

¹⁴ We use a Stata program created by Lesnard (2006). The program generates several files that contains the series of monthly transition matrices, substitution cost matrices and the distance matrix.

¹⁵ A previous description of sequences based on aggregated measures related to the number of states, the time spent on states and most frequent sequences has discovered that a high percentage of young people share the same trajectory.

¹⁶ Other authors propose to display and visualize results using self-organizing maps (Giret and Rousset, 2007).

¹⁷ To evaluate the robustness of the cluster analysis, we have investigated the presence of more or fewer clusters, but the identification described in this article with seven clusters was the clearest.

¹⁸ To describe clusters, we retrieve the sequences repeated.

¹⁹ Information about the families that are aggregated can be found in Table 2.

²⁰ At this respect, there exists a vast empirical literature that shows that Spain has suffered a problem of over-education, mainly in the services sector, involving many young people with a university degree in jobs which require less qualifications (Dolado et al., 2000; Budria and Moro-Egido, 2008).

²¹ The robustness of the results with respect to different cost matrices was expected. Most of the papers about the school-to-work transition commented previously have also experimented with different variants of substitution costs and arrived at the same results.