The changing patterns of individual and school effects on educational transitions. Evidence from Catalan data (Spain)

Abstract

Background: This article engages with the tradition of educational transitions research, particularly with its attempt to evaluate the effect of exogenous variables on educational attainment. The study revisits a number of hypotheses that have attempted to explain the changing patterns of such effects throughout students’ educational career, particularly: life course hypothesis, selective attrition, maximally maintained inequality, effectively maintained inequality and path dependence hypothesis. Purpose: The article aims at examining the extent to which individual level variables influence the transition patterns that Catalan students follow, in the frame of the Spanish education system, from the completion of compulsory education up to the transition to higher education. Furthermore, the study evaluates the effects of an exogenous independent variable that educational transition’ studies have tended to neglect, that is, school composition. Particular emphasis is placed on observing the scope of changes in variables’ effects across transitions. Sample: Analyses are carried out on the responses to a questionnaire given by a sample of students in the final grade of lower-secondary education (age 16), upper-secondary academic (Baccalaureate) (age 18), middle vocational (age 17/18) and higher vocational education (age 20 and above). The selection of schools was made through cluster sampling based on the following five strata: public lower-secondary education (1580 students), private lower-secondary education (1373 students), Baccalaureate (964 students), Middle Vocational Courses (566 students) and Higher Vocational Courses (377 students). In each of these five strata, a random selection was made from the secondary schools that offer the educational stage in question. In total, 4860 students were surveyed in 56 different schools. Design and methods: In each of the schools included in the sample, questionnaires were distributed to all students in the last year of the various educational stages that the schools offer. The statistical analyses carried out rely on multinomial hierarchical logit models. The dependent variables relate to the transition options students follow when completing each of the educational stages under consideration. These data were devised by fusing students’ transition decisions taken at the end of the last year of each stage and the data provided by the schools after the end of the year regarding the results obtained by each student. The independent dummy variables used in the models are: gender, student educational capital, student immigrant background, parental professional category; prior grade point average (individual level variables); school type; school educational composition; school immigrant background composition; school professional composition (school level variables). Results: Whereas school level variables, as well as individual variables such as parental professional category or student immigrant background, are only significant on conditioning transitions at the end of lower-secondary education (compulsory), other variables such as gender or educational capital maintain their explanatory capacity on transitions across post-compulsory stages. In addition, the multinomial transition perspective utilised here allows us to evaluate the scope of significant path dependent effects. Conclusions: The study concludes that a clear pattern of ‘waning effects’ of exogenous variables does not occur throughout students’ educational career. That is, the effect of individual characteristics on the probability of students transitioning across consecutive educational stages does not systematically decrease as they progress in their educational career.