Migration and education in Amazonia / Helene Barros.

Tese (Ed.D.)--Harvard University, 1978.

This thesis tests several sets of hypotheses in view of verifying the existing inter-state migration theory in the Amazonian context. The study begins by a brief survey of the theories and a summary of some of the empirical evidence on migration in general and particularly on its relationship with education. The second chapter attempts to bring out the relevance of these theories and findings for the Amazonian region and describes the migratory movements taking place in the region as indicated by the 1970 census data. The third chapter presents the purpose and limitations of the regression analysis, the variables and data, the models and estimation techniques applied and a summary of the main results. The fourth and last chapter makes a final assessment and spells out some of the basic conclusions to be drawn. In trying to ascertain the determinants and the impacts of interregional migration, it is speculated that the main determinants consist of origin and destination differentials, specifically the frictional effects of economic conditions, of intervening elements such as distance and of the relative attractive power of the Amazonian region. It is also expected that migration provokes a negative impact as far as the out-migration is concerned and a positive impact with regards to the in-migration flows. As one of the main purposes is to discover whether education has a significant effect on migratory fluxes to and from Amazonia, the study looks at the influence of the educational background of the labour force in origin and destination areas. It also considers the educational characteristics of the school age population as possible factor attracting migrants into Amazonia. It further seeks to establish what may be the impact of migration on the educational level of the labour force of the school age population between 7 and 10 years and on the performance of the educational system in providing adequate services at primary level. Regression analysis in its different techniques was used as the basic Instrument for specifying and estimating the models. Three basic specification forms were tried out (the linear, simple double log and relative double log forms) to test which can better explain and predict the factors influencing migration. The analysis of the determinants indicated that the Amazonian out-migration follows rather closely the usual patterns of the migration process whereas the in-migration flows reveal a somewhat different pattern, most probably due to the peculiarities of the region. The socio-economic variables took on considerable importance among the origin and destination differentials. The analysis confirms the relevance of these differentials in explaining the two directional flows of population movements. As for the relationship between education and migration, education certainly turns out to be a significant factor explaining in- and out-migration, particularly in the case of urban destination moves. Moreover, it did not appear to be an influencing element with regards to migrants seeking rural areas as destination. The analysis of the impacts of migration, indicated that the elasticities between cumulative migration and the economic and education variables are significant for both rural and urban areas, although the elasticity of these Impacts was below unity and thus cannot be considered as high.