Transições: três aplicações a partir de dados das pesquisas domiciliares no Brasil

Abstract

The central question in the present thesis was to understand how important life transitions occur in Brazil: leave to school, the first job and family formation (the first child). The theme was discussed in three individual manuscripts, although these share an intimate relationship and commingle with the line of thought directing the central issue: typical transitions of children and teenagers. The first manuscript identified individual and dependent effects of Age, Period and Cohort APC on the insertion of children and teenagers between 10 and 18 years into the job market in detriment to school, concomitantly with school or yet inactivity (neither having a job nor attending school). Child cohorts were analyzed by sex in the following occupational and educational status: 1. children that work and study; 2. children solely studying; 3. children solely working; 4. children neither working nor studying. Besides the analysis by sex, cohort subgroups according to sociodemographic characteristics were also evaluated, such as highest level of education achieved of the head of household, skin color or race, and dwelling conditions. This analysis was performed in order to identify peculiar profiles of transition in population segments that, although belonging to the same birth cohort, would be distinguished by individual and dwelling characteristics. The dwelling research from Pesquisa Nacional de Amostra por Domicílios (PNAD, National Research of Dwelling Sampling) provided the input data for the analyses.The assessment of period effects on the occurrence probabilities of social events, as in the present study, may give indications of the efficacy of policies implemented in specific situations. It also may indicate whether such efficacy was restricted to the period or extended to other periods. The methodology of survival analysis was used in the second manuscript to evaluated the relationship among the ages when such transitions occur (leave to school, the first job and family formation). Firstly, a descriptive analysis was carried out using a nonparametric model (Kaplan Meier) to estimate the survival functions for each transition based on sex, birth cohort and place of residence until 15 years of age. A semiparametric approach was then performed (Coxs Proportional Hazards Assessment). Some of the variables used are: highest level of education of each parent, dwelling condition until 15 years, and socio-occupational category of the parents. The database used was Pesquisa Sobre Padrões de Vida (PPV, Research of Life Conditions).In this study, it was used the transition experience of subjects aged between 20 and 49 years during the research (1996-1997), thus the results do not translate necessarily the transition experience of youths that are going through transitions in the current decade. Nevertheless, the value of the findings lies exactly on that characteristic. The results showed clearly that the transition to the first child did not represent the main reason for school absence or drop-out in the experience of these cohorts, since the transition sequences and the age at parenthood reduces only in 0.9% the risk of dropping school.Once the challenges, ages and interfering factors of a transition are identified, they constitute highly relevant information available to public policies for education, qualification of individuals, job creation, and housing and health care policies. The third manuscript tested the hypothesis of the additional worker effect for children and teenagers (offsprings) in Brazilian metropolitan regions. The aim was to evaluate the transition of mandatory entrance in the job market in order to substitute for an unemployed head of the family. It was assumed the existence of such effect on the offsprings and that the effect was greater in magnitude as suggested by Fernandes & Felício (2002), whose estimation methodology was adapted to be used herein.The hypothesis was refuted in the face of the data obtained and the adopted methodology. On the other hand, considerations about some aspects of the methodology were made and these may be assessed in further studies.It has been also concluded that studies of this sort should include institutional measures that contribute to family income and increased time for school completion, in the shapes of the current bolsa família (family scholarship) or other mechanisms that alleviate income loss, such as unemployment insurance programs.